instruction stringlengths 34 186 | input stringlengths 2.02k 93.8k | output stringlengths 2 418 | meta_questiontype stringclasses 6 values | meta_inputlicense stringclasses 6 values | meta_pmid stringlengths 8 8 | meta_safetyreportid int64 9.51M 21M | meta_articlepubdate stringlengths 4 10 |
|---|---|---|---|---|---|---|---|
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Off label use'. | Single-shot bevacizumab for cerebral radiation injury.
BACKGROUND
Cerebral radiation injury, including subacute radiation reactions and later stage radiation necrosis, is a severe side effect of brain tumor radiotherapy. A protocol of four infusions of the monoclonal antibody bevacizumab has been shown to be a highly effective treatment. However, bevacizumab is costly and can cause severe complications including thrombosis, bleeding and gastrointestinal perforations.
METHODS
We performed a retrospective analysis of patients treated in our clinic for cerebral radiation injury who received only a singular treatment with bevacizumab. Single-shot was defined as a singular administration of bevacizumab without a second administration during an interval of at least 6 weeks.
RESULTS
We identified 11 patients who had received a singular administration of bevacizumab to treat cerebral radiation injury. Prior radiation had been administered to treat gliomas (ten patients) or breast cancer brain metastases (one patient). 9 of 10 patients with available MRIs showed a marked reduction of edema at first follow-up. Discontinuation of Dexamethasone was possible in 6 patients and a significant dose reduction could be achieved in all other patients. One patient developed pulmonary artery embolism 2 months after bevacizumab administration. The median time to treatment failure of any cause was 3 months.
CONCLUSIONS
Single-shot bevacizumab therefore has meaningful activity in cerebral radiation injury, but durable control is rarely achieved. In patients where a complete protocol of four infusions with bevacizumab is not feasible due to medical contraindications or lack of reimbursement, single-shot bevacizumab treatment may be considered.
Background
Radiation necrosis has been reported in approximately 6 % of patients with brain tumors after radiation therapy and can lead to significant morbidity and, if untreated, mortality by progressive necrosis and brain edema [1]. Additionally, the risk of misinterpreting radiation injury (including subacute radiation reactions and later stage radiation necrosis) for tumor progression can prevent adequate therapy [2]. The risk of radiation injury is highest in patients who undergo repeated courses of radiotherapy, even with prolonged intervals between the two treatments. Lee et al. reported a rate of 64 % radiation necrosis for hypofractionated re-irradiation (45 Gy in 15 fractions) in glioma patients at least 12 months post-treatment [3]. Conceptually, radiation-induced injury is thought to result from damage to vascular endothelial and glial cells. Secretion of vascular endothelial growth factor (VEGF)-A appears to be responsible for edema formation via increasing vascular permeability and inducing a pro-inflammatory environment [4].
Bevacizumab is an antibody targeting VEGF-A induced angiogenesis and has been evaluated as a treatment for malignant brain tumors. While several phase III trials of first-line therapy failed to show any effect on overall survival [5–7], there was still a pronounced effect of bevacizumab on the blood brain barrier with reduced gadolinium contrast enhancement and edema reducing the rate of pseudoprogression in MRI scans from 9.3 to 2.2 % in the AVAglio trial [8]. Bevacizumab has also been used in small clinical trials as a treatment for radiation necrosis. Levin et al. reported a randomized, placebo-controlled trial of four infusions of bevacizumab 7.5 mg/kg at 3-week intervals for radiation necrosis of the central nervous system [9]. This trial demonstrated an impressive clinical and radiological improvement in all patients receiving bevacizumab while no patient with placebo treatment improved spontaneously. This treatment efficacy came at the cost of a high rate of adverse events in the bevacizumab group (6 of 11 patients) while no adverse events occurred in the placebo group. Common serious side effects of bevacizumab regimens include pulmonary artery embolism, venous thrombosis and intracranial hemorrhage [10, 11]. Whether a reduced number of bevacizumab cycles could also suffice to adequately treat radiation injury with a potentially reduced side effect profile remains unclear. In tumor treatment, clinical trials comparing standard and low-dose bevacizumab regimens found no significant differences in efficacy [12] but suggested a more favorable toxicity profile [13, 14].
Bevacizumab has not been approved by the European Medicines Agency (EMA), neither for progressive glioblastoma nor for the treatment of radiation reaction (FDA approval for adult patients with progressive glioblastoma) but is often used as an individual, off-label therapy for dexamethasone-refractory radiation necrosis or following steroid discontinuation due to adverse effects. In cases of rapid clinical deterioration, immediate treatment with bevacizumab can be considered to prevent permanent damage to eloquent areas. However, reimbursement by insurance companies can be difficult. If a singular administration (single-shot) of bevacizumab, that is considered financially affordable, was sufficient to treat cerebral radiation injury, this would broaden options for both, patients and physicians due to lower financial as well as potential side effect risks, especially for patients with prior vascular contraindications.
Methods
We performed a retrospective analysis of patients treated in our clinic between 2016 and 2019 to identify patients with cerebral radiation side effects who received a singular treatment with bevacizumab. Diagnosis of acute radiation reaction and radiation necrosis had been made in the interdisciplinary tumor board based on MRI and considering the field of radiation and the time from last radiation therapy (results section). Single-shot bevacizumab was defined as a singular administration of bevacizumab without a second administration during an interval of at least six weeks. The patient collective was evaluated with regard to histology, patient age at diagnosis of radiation injury, duration and maximum dose of dexamethasone, clinical course and possible side-effects, as well as the radiologic response to bevacizumab treatment. MRI scans included at least axial fluid-attenuated inversion recovery (FLAIR), T2-weighted, and T1-weighted images before and after application of gadolinium-based contrast agent. The extent of edema was estimated on the axial FLAIR or T2-weighted sequence. Response to bevacizumab treatment was defined as a reduction of the edema by at least 25 % [9].
Written consent by the individual patient for this retrospective data collection was waived by the ethics committee of the University Hospital Frankfurt; Goethe University which also approved the access to the patients’ data (IRB decision # 4/09, project SNO_01–08). Microsoft Excel was used for data management and analysis. Corel Draw 2019 was used to create figures.
Results
From 2016 until the end of 2019 approximately 400 patients received radiation of the brain for any reason (brain tumor or brain metastasis including primary therapy and re-irradiation therapy) at our cancer center. During this time, about 65 patients were treated with bevacizumab for radiation reaction. Retrospective analysis revealed 11 patients who were initially treated with a single-shot (Table 1). Ten patients had received prior fractionated radiation therapy for gliomas including 2 patients being treated primarily by radiotherapy at initial diagnosis (radiotherapy doses: 54 and 60 Gy) and 8 patients, who underwent re-irradiation for recurrent tumor (radiotherapy doses: 20–36 Gy), whereas one patient received re-irradiation (dose: 30 Gy) for recurrent brain metastasis of breast cancer after an initial radiosurgery.
Table 1 Patient characteristics
Number of Patients 11
Age at treatment with BEV [years]
Median (range) 47 (22 – 73)
Histology
Glioma 91% (10)
Brain metastasis (breast cancer) 9% (1)
Radiation for
Recurrent tumor 82% (9)
Primary therapy 18% (2)
Last radiation therapy prior to BEV [Gy]
5x4 18% (2)
10x3 9% (1)
10x3,5 36% (4)
12x3 9% (1)
15x2,67 9% (1)
30x1,8 9% (1)
30x2 9% (1)
Time from radiation to diagnosis of radiation injury [months]
Median (range) 2 (1-7)
Maximum Dose of dexamethasone [mg]
Before therapy, Median (range)
After therapy, Median (range)
8 (0 – 40)
0 (0 - 4)
Karnofsky-Score [%]
Before therapy, Median (range) 50 (40 – 80)
After therapy, Median (range) 60 (40 – 80)
Dose of BEV single-shot
7,5mg/kg 73% (8)
10mg/kg 27% (3)
Reported benefit by patient
Yes 64% (7)
No 36% (4)
Abbreviation: BEV bevacizumab
As soon as acute radiation reaction / radiation necrosis was diagnosed, therapy with dexamethasone was started or an already established therapy with steroids was intensified following a mean interval of 2 months post-radiation therapy. Median peak dose of dexamethasone was 8 mg/day, with a maximum dose of 40 mg/day in 2 patients. Diagnosis of radiation injury was based on MRI in 10 patients using additional MR-perfusion in 6 patients. In one patient, diagnosis was confirmed by positron emission tomography (F-18-fluroethyltyrosine). In no case had a biopsy been performed to confirm the diagnosis histologically.
When dexamethasone did not improve clinical symptoms or could not be tolerated at the required doses due to side effects, off-label treatment with bevacizumab was recommended at the institutional multidisciplinary tumor board. Four of these patients had a single-shot of bevacizumab treatment because of a high-risk situation for side effects rendering long-term repeat treatment with bevacizumab unfeasible (pulmonary embolism, deep vein thrombosis, fracture of several rips, hemorrhage of the tumor). In another patient there were concerns of possible increased toxicity as the patient received ongoing therapy with lomustin and temozolomide [15], and in a further patient bevacizumab was only administered once because of the ensuing palliative setting aimed at improving aphasia (Patient 1). Moreover, one patient initially received one singular infusion due to personal concerns with regard to side effects (patient 11), and two did not consent to further infusions (Patient 6 and patient 9). Two patients did not receive reimbursement by the insurance company for further treatment after the single-shot of bevacizumab.
Eight patients received 7.5 mg/kg as proposed by Levin et al., three patients received 10 mg/kg as used in the neuro-oncological trials for bevacizumab at that time [6, 9]. The treatment was well-tolerated without any acute side effects during the infusion. One patient with immobility developed deep vein thrombosis with subsequent pulmonary artery embolism two months after bevacizumab.
After a median interval of 55 days following the administration of bevacizumab first MRI showed a marked reduction of brain edema (at least 25 %) in 9/10 evaluable patients. An example is given in Fig. 1.
Fig. 1 MRI scans of a 34 year old patient with IDH-mutated astrocytoma. a MRI revealed a small recurrent tumor adjacent to the dorsal resection cavity with small surrounding edema. b The patient was treated with re-radiation therapy with 35 Gy and concomitant temozolomide. First MRI after the treatment showed an increase of contrast enhancement and edema which was diagnosed as radiation necrosis. c Therapy with 8 mg of dexamethasone did neither improve the MRI nor the clinical symptoms and bevacizumab 7.5 mg/kg was administered as a single-shot. d First scan one month later displayed a marked reduction in contrast enhancement and of the edema. Treatment with dexamethasone could be stopped. The follow up 3 months later was stable (not shown)
After single-shot bevacizumab, patients Karnofsky Performance Score (KPS) improved from a median of 50–60 % and 7 patients reported markedly improved clinical symptoms at the first visit after bevacizumab. Here, we noticed that the only slight improvement of KPS underestimated the clinical benefit in the activity of daily life. Indeed, the ability for an independent transfer from the wheelchair to a bed or toilet has a great impact on the patient’s quality of life that is not accurately reflected in the Karnofsky-Index. Notably, dexamethasone could be stopped in 6 of the patients. In all other patients, the dose of dexamethasone could be gradually reduced, finally reaching doses between 0.5 and 4 mg/day after a median time of 39 days after the single-shot.
Mean time to treatment failure was 3 months (range 1–10 months). Importantly, treatment failure to bevacizumab was due to tumor progression (patient 2, 4 and 6) or death (patient 1) in four patients, therefore, tumor progression should always be taken into account when interpreting clinical deterioration as the latter likely reflects a mixture of tumor progression and radiation necrosis (Fig. 2). One patient (patient 8) had both a marked improvement in clinical symptoms and MRI with a decline in contrast enhancement after single-shot. In this patient, however, treatment with bevacizumab was resumed 8 weeks after the first infusion, since the patient still experienced disabilities in the activities of daily life, and the single-shot had been tolerated well. Treatment failure in the other patients was diagnosed due to recurrent edema in follow-up MRI with or without clinical symptoms (Fig. 3).
Fig. 2 MRI scans of a 33 year old patient with IDH-wildtype glioblastoma. First (a) and second (b) MRI after resection and radiation therapy of recurrent glioblastoma showed not signs of tumor progression. Dexamethasone was started because of clinical deterioration before the third control (c) which showed a substantial increase in contrast enhancement and edema which were interpreted as late radiation necrosis. Bevacizumab 7.5 mg/kg was administered as a single-shot. d First scan 1.5 months later displayed a marked reduction of the edema while there was only a minor reduction of contrast enhancement. Diagnosis was changed from radiation necrosis to recurrent tumor
Fig. 3 Time to treatment failure. The swimmer plot shows the course of the individual patients labeled at the left side. The radiological diagnosis is indicated by color-coded dots (yellow: MRI, orange: MRI and MR-perfusion, purple: PET). The color-coded diamonds indicate the treatment failure of the single-shot bevacizumab (green: treatment of recurrent edema with corticosteroids, blue: treatment of recurrent edema with bevacizumab, blue border: resumed bevacizumab as the symptoms did not completely resolve, red: recurrent tumor, black: death of the patient because of recurrent tumor). Median time from single-shot to time of treatment failure of any cause) was three months
Discussion
Cerebral irradiation is an integral part of the treatment of brain cancer. One of the most severe complications is cerebral necrosis that can occur in patients with primary or metastatic brain tumors especially after a second course of irradiation for recurrent tumors. Despite promising efficacy in the treatment of cerebral radiation necrosis from smaller clinical trials, no application for bevacizumab approval for this indication has been filed. Reimbursement by insurance companies therefore remains difficult and is granted only on a case by case basis limiting the availability of bevacizumab. Assessing the efficacy of a singular bevacizumab treatment with a potentially more favorable side-effect profile and lesser financial burden than the cyclic treatment addresses a clinically important challenge. In the present work we show that a singular dose of bevacizumab resulted in significantly reduced edema on MRI sequences in all evaluable patients with two-thirds of patients reporting a meaningful improvement of clinical symptoms. In this context the very shot interval from radiation therapy to the development of brain lesions has to be noted. The mean time of two month is rather short for radiation necrosis in comparison to the trial by Levin et al. [9]. Therefore it is plausible that some of the patients suffered from a subacute or early delayed radiation reaction rather than manifest necrosis. Despite the small series, this study provides encouraging data, indicating that singular administration of bevacizumab might be a useful option for the treatment of radiation reaction / necrosis, especially in patients where prolonged bevacizumab treatment is not deemed feasible, as for example due to prior thromboembolic events or due to denial of therapy reimbursement. Additionally, even when bevacizumab is available for multiple treatments, a single-shot might be sufficient treatment for some patients. At the present time it is unclear whether a resumption of therapy in cases where a single-shot is not sufficient has disadvantages on the course of cerebral radiation necrosis.
In our analysis, we identified only one potentially severe side effect in one patient who was diagnosed with pulmonary artery embolism two months after bevacizumab. Whether this was instead attributable to immobility of the patient, who was later also diagnosed with deep vein thrombosis, or if this only contributed to the embolism remains unclear.
An interesting variant of the single-shot bevacizumab concept to further reduce the systemic side effects could be a local administration. The ongoing LIBERTI trial (NCT02819479) evaluates the efficacy of a single, intra-arterial dose of only 2.5 mg/kg bevacizumab [16]. Despite the large molecular mass of bevacizumab some penetration of the disrupted blood-brain barrier in regions of radiation injury appears possible [17]. The reduced side effect might come at the cost of a decreased duration of edema control. With a half-life of three weeks, the effect on the blood brain barrier might not be lasting, and the downside could be a rebound phenomenon with the need of bevacizumab re-challenge, as also indicated by the short time to treatment failure in our collective of only three months. The trial of Levin et al. with four administrations of 7.5 mg/ reported a relapse of radiation necrosis in 25 % of patients [9]. Zhuang et al. reported 14 patients with cerebral radiation necrosis who were treated with a lower dose of 5 mg/kg bevacizumab for at least 3 cycles of therapy [18]. MRI showed improvement in 13 of the 14 patients, but 10 of the 13 responsive patients exhibited a rebound phenomenon in the later follow-up. One option would be to further lower the dose of bevacizumab but keep the continuous administration. This approach of lowering the dose to 1 mg/kg bevacizumab every three weeks has revealed promising results in a phase 2 trial [19]. The trial included 21 patients and the grade of the edema index was improved in 19 patients. In contrast to Levin et al., no adverse events above grade 2 were reported. This concept has been further supported by case reports with low-dose bevacizumab and even longer intervals between the administration [20].
Prophylactic administration of a singular dose bevacizumab in high-risk situations (re-irradiation therapy, large irradiation fields or/and already present widespread edema prior to irradiation) could also be an option to consider. Thereby clinical deterioration might be prevented and the need for corticosteroids as well as the risk of a rebound effect after termination of bevacizumab treatment might be reduced. Such an approach has been explored in a phase 1 trial by Clarke et al. This trial included bevacizumab treatment to intensify the radiation dose of hypofractionated stereotactic re-irradiation [21]. This concept could be even more beneficial in cyber knife radiosurgery [22, 23].
While histological confirmation of cerebral radiation necrosis is clearly limited to only the most ambiguous cases, we here report a cohort of 11 patients whose radiological scans and dynamics indicated radiation necrosis. Two patients (patient 2 and 6) had treatment failure shortly after the single-shot due to tumor progress in the MRI and these were also patients where diagnosis was based on conventional MRI without MR-perfusion or MR-spectroscopy. A more selected collective of histologically diagnosed radiation necrosis could have shown more sustained effects of bevacizumab.
Conclusions
In summary, bevacizumab is an effective treatment for patients with cerebral radiation injury. Optimal dosing and intervals still have to be defined but most likely lower doses and longer intervals than investigated in previous trials [9] are sufficient. In the case of patients at high risk for side effects, single-shot of bevacizumab may be used as a test-dose and treatment can be continued when necessary.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Acknowledgements
None.
Authors’ contributions
All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by M.V., K.W. E.F., MT.F. and M.W.R. The first draft of the manuscript was written by M.V., J.P.S. and M.W.R and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Funding
The authors received no specific funding for this work.
Availability of data and materials
Raw data were generated at the Dr. Senckenberg Institute of Neurooncology. The datasets generated are available from the corresponding author on reasonable request.
Ethics approval and consent to participate
Written consent by the individual patient for this retrospective data collection was waived by the ethics committee of the University Hospital Frankfurt; Goethe University which also approved the access to the patients’ data (IRB decision # 4/09, project SNO_01–08).
Consent for publication
No individual person’s data is included in this article.
Competing interests
J.P.S. has received honoraria for lectures or advisory board participation, consulting or travel grants from Abbvie, Roche, Boehringer, Bristol-Myers Squibb, Medac, Mundipharma and UCB. M.W.R. has received a grant from UCB. The other authors report no conflict of interest. | BEVACIZUMAB | DrugsGivenReaction | CC BY | 33596839 | 18,986,888 | 2021-02-17 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Pulmonary embolism'. | Single-shot bevacizumab for cerebral radiation injury.
BACKGROUND
Cerebral radiation injury, including subacute radiation reactions and later stage radiation necrosis, is a severe side effect of brain tumor radiotherapy. A protocol of four infusions of the monoclonal antibody bevacizumab has been shown to be a highly effective treatment. However, bevacizumab is costly and can cause severe complications including thrombosis, bleeding and gastrointestinal perforations.
METHODS
We performed a retrospective analysis of patients treated in our clinic for cerebral radiation injury who received only a singular treatment with bevacizumab. Single-shot was defined as a singular administration of bevacizumab without a second administration during an interval of at least 6 weeks.
RESULTS
We identified 11 patients who had received a singular administration of bevacizumab to treat cerebral radiation injury. Prior radiation had been administered to treat gliomas (ten patients) or breast cancer brain metastases (one patient). 9 of 10 patients with available MRIs showed a marked reduction of edema at first follow-up. Discontinuation of Dexamethasone was possible in 6 patients and a significant dose reduction could be achieved in all other patients. One patient developed pulmonary artery embolism 2 months after bevacizumab administration. The median time to treatment failure of any cause was 3 months.
CONCLUSIONS
Single-shot bevacizumab therefore has meaningful activity in cerebral radiation injury, but durable control is rarely achieved. In patients where a complete protocol of four infusions with bevacizumab is not feasible due to medical contraindications or lack of reimbursement, single-shot bevacizumab treatment may be considered.
Background
Radiation necrosis has been reported in approximately 6 % of patients with brain tumors after radiation therapy and can lead to significant morbidity and, if untreated, mortality by progressive necrosis and brain edema [1]. Additionally, the risk of misinterpreting radiation injury (including subacute radiation reactions and later stage radiation necrosis) for tumor progression can prevent adequate therapy [2]. The risk of radiation injury is highest in patients who undergo repeated courses of radiotherapy, even with prolonged intervals between the two treatments. Lee et al. reported a rate of 64 % radiation necrosis for hypofractionated re-irradiation (45 Gy in 15 fractions) in glioma patients at least 12 months post-treatment [3]. Conceptually, radiation-induced injury is thought to result from damage to vascular endothelial and glial cells. Secretion of vascular endothelial growth factor (VEGF)-A appears to be responsible for edema formation via increasing vascular permeability and inducing a pro-inflammatory environment [4].
Bevacizumab is an antibody targeting VEGF-A induced angiogenesis and has been evaluated as a treatment for malignant brain tumors. While several phase III trials of first-line therapy failed to show any effect on overall survival [5–7], there was still a pronounced effect of bevacizumab on the blood brain barrier with reduced gadolinium contrast enhancement and edema reducing the rate of pseudoprogression in MRI scans from 9.3 to 2.2 % in the AVAglio trial [8]. Bevacizumab has also been used in small clinical trials as a treatment for radiation necrosis. Levin et al. reported a randomized, placebo-controlled trial of four infusions of bevacizumab 7.5 mg/kg at 3-week intervals for radiation necrosis of the central nervous system [9]. This trial demonstrated an impressive clinical and radiological improvement in all patients receiving bevacizumab while no patient with placebo treatment improved spontaneously. This treatment efficacy came at the cost of a high rate of adverse events in the bevacizumab group (6 of 11 patients) while no adverse events occurred in the placebo group. Common serious side effects of bevacizumab regimens include pulmonary artery embolism, venous thrombosis and intracranial hemorrhage [10, 11]. Whether a reduced number of bevacizumab cycles could also suffice to adequately treat radiation injury with a potentially reduced side effect profile remains unclear. In tumor treatment, clinical trials comparing standard and low-dose bevacizumab regimens found no significant differences in efficacy [12] but suggested a more favorable toxicity profile [13, 14].
Bevacizumab has not been approved by the European Medicines Agency (EMA), neither for progressive glioblastoma nor for the treatment of radiation reaction (FDA approval for adult patients with progressive glioblastoma) but is often used as an individual, off-label therapy for dexamethasone-refractory radiation necrosis or following steroid discontinuation due to adverse effects. In cases of rapid clinical deterioration, immediate treatment with bevacizumab can be considered to prevent permanent damage to eloquent areas. However, reimbursement by insurance companies can be difficult. If a singular administration (single-shot) of bevacizumab, that is considered financially affordable, was sufficient to treat cerebral radiation injury, this would broaden options for both, patients and physicians due to lower financial as well as potential side effect risks, especially for patients with prior vascular contraindications.
Methods
We performed a retrospective analysis of patients treated in our clinic between 2016 and 2019 to identify patients with cerebral radiation side effects who received a singular treatment with bevacizumab. Diagnosis of acute radiation reaction and radiation necrosis had been made in the interdisciplinary tumor board based on MRI and considering the field of radiation and the time from last radiation therapy (results section). Single-shot bevacizumab was defined as a singular administration of bevacizumab without a second administration during an interval of at least six weeks. The patient collective was evaluated with regard to histology, patient age at diagnosis of radiation injury, duration and maximum dose of dexamethasone, clinical course and possible side-effects, as well as the radiologic response to bevacizumab treatment. MRI scans included at least axial fluid-attenuated inversion recovery (FLAIR), T2-weighted, and T1-weighted images before and after application of gadolinium-based contrast agent. The extent of edema was estimated on the axial FLAIR or T2-weighted sequence. Response to bevacizumab treatment was defined as a reduction of the edema by at least 25 % [9].
Written consent by the individual patient for this retrospective data collection was waived by the ethics committee of the University Hospital Frankfurt; Goethe University which also approved the access to the patients’ data (IRB decision # 4/09, project SNO_01–08). Microsoft Excel was used for data management and analysis. Corel Draw 2019 was used to create figures.
Results
From 2016 until the end of 2019 approximately 400 patients received radiation of the brain for any reason (brain tumor or brain metastasis including primary therapy and re-irradiation therapy) at our cancer center. During this time, about 65 patients were treated with bevacizumab for radiation reaction. Retrospective analysis revealed 11 patients who were initially treated with a single-shot (Table 1). Ten patients had received prior fractionated radiation therapy for gliomas including 2 patients being treated primarily by radiotherapy at initial diagnosis (radiotherapy doses: 54 and 60 Gy) and 8 patients, who underwent re-irradiation for recurrent tumor (radiotherapy doses: 20–36 Gy), whereas one patient received re-irradiation (dose: 30 Gy) for recurrent brain metastasis of breast cancer after an initial radiosurgery.
Table 1 Patient characteristics
Number of Patients 11
Age at treatment with BEV [years]
Median (range) 47 (22 – 73)
Histology
Glioma 91% (10)
Brain metastasis (breast cancer) 9% (1)
Radiation for
Recurrent tumor 82% (9)
Primary therapy 18% (2)
Last radiation therapy prior to BEV [Gy]
5x4 18% (2)
10x3 9% (1)
10x3,5 36% (4)
12x3 9% (1)
15x2,67 9% (1)
30x1,8 9% (1)
30x2 9% (1)
Time from radiation to diagnosis of radiation injury [months]
Median (range) 2 (1-7)
Maximum Dose of dexamethasone [mg]
Before therapy, Median (range)
After therapy, Median (range)
8 (0 – 40)
0 (0 - 4)
Karnofsky-Score [%]
Before therapy, Median (range) 50 (40 – 80)
After therapy, Median (range) 60 (40 – 80)
Dose of BEV single-shot
7,5mg/kg 73% (8)
10mg/kg 27% (3)
Reported benefit by patient
Yes 64% (7)
No 36% (4)
Abbreviation: BEV bevacizumab
As soon as acute radiation reaction / radiation necrosis was diagnosed, therapy with dexamethasone was started or an already established therapy with steroids was intensified following a mean interval of 2 months post-radiation therapy. Median peak dose of dexamethasone was 8 mg/day, with a maximum dose of 40 mg/day in 2 patients. Diagnosis of radiation injury was based on MRI in 10 patients using additional MR-perfusion in 6 patients. In one patient, diagnosis was confirmed by positron emission tomography (F-18-fluroethyltyrosine). In no case had a biopsy been performed to confirm the diagnosis histologically.
When dexamethasone did not improve clinical symptoms or could not be tolerated at the required doses due to side effects, off-label treatment with bevacizumab was recommended at the institutional multidisciplinary tumor board. Four of these patients had a single-shot of bevacizumab treatment because of a high-risk situation for side effects rendering long-term repeat treatment with bevacizumab unfeasible (pulmonary embolism, deep vein thrombosis, fracture of several rips, hemorrhage of the tumor). In another patient there were concerns of possible increased toxicity as the patient received ongoing therapy with lomustin and temozolomide [15], and in a further patient bevacizumab was only administered once because of the ensuing palliative setting aimed at improving aphasia (Patient 1). Moreover, one patient initially received one singular infusion due to personal concerns with regard to side effects (patient 11), and two did not consent to further infusions (Patient 6 and patient 9). Two patients did not receive reimbursement by the insurance company for further treatment after the single-shot of bevacizumab.
Eight patients received 7.5 mg/kg as proposed by Levin et al., three patients received 10 mg/kg as used in the neuro-oncological trials for bevacizumab at that time [6, 9]. The treatment was well-tolerated without any acute side effects during the infusion. One patient with immobility developed deep vein thrombosis with subsequent pulmonary artery embolism two months after bevacizumab.
After a median interval of 55 days following the administration of bevacizumab first MRI showed a marked reduction of brain edema (at least 25 %) in 9/10 evaluable patients. An example is given in Fig. 1.
Fig. 1 MRI scans of a 34 year old patient with IDH-mutated astrocytoma. a MRI revealed a small recurrent tumor adjacent to the dorsal resection cavity with small surrounding edema. b The patient was treated with re-radiation therapy with 35 Gy and concomitant temozolomide. First MRI after the treatment showed an increase of contrast enhancement and edema which was diagnosed as radiation necrosis. c Therapy with 8 mg of dexamethasone did neither improve the MRI nor the clinical symptoms and bevacizumab 7.5 mg/kg was administered as a single-shot. d First scan one month later displayed a marked reduction in contrast enhancement and of the edema. Treatment with dexamethasone could be stopped. The follow up 3 months later was stable (not shown)
After single-shot bevacizumab, patients Karnofsky Performance Score (KPS) improved from a median of 50–60 % and 7 patients reported markedly improved clinical symptoms at the first visit after bevacizumab. Here, we noticed that the only slight improvement of KPS underestimated the clinical benefit in the activity of daily life. Indeed, the ability for an independent transfer from the wheelchair to a bed or toilet has a great impact on the patient’s quality of life that is not accurately reflected in the Karnofsky-Index. Notably, dexamethasone could be stopped in 6 of the patients. In all other patients, the dose of dexamethasone could be gradually reduced, finally reaching doses between 0.5 and 4 mg/day after a median time of 39 days after the single-shot.
Mean time to treatment failure was 3 months (range 1–10 months). Importantly, treatment failure to bevacizumab was due to tumor progression (patient 2, 4 and 6) or death (patient 1) in four patients, therefore, tumor progression should always be taken into account when interpreting clinical deterioration as the latter likely reflects a mixture of tumor progression and radiation necrosis (Fig. 2). One patient (patient 8) had both a marked improvement in clinical symptoms and MRI with a decline in contrast enhancement after single-shot. In this patient, however, treatment with bevacizumab was resumed 8 weeks after the first infusion, since the patient still experienced disabilities in the activities of daily life, and the single-shot had been tolerated well. Treatment failure in the other patients was diagnosed due to recurrent edema in follow-up MRI with or without clinical symptoms (Fig. 3).
Fig. 2 MRI scans of a 33 year old patient with IDH-wildtype glioblastoma. First (a) and second (b) MRI after resection and radiation therapy of recurrent glioblastoma showed not signs of tumor progression. Dexamethasone was started because of clinical deterioration before the third control (c) which showed a substantial increase in contrast enhancement and edema which were interpreted as late radiation necrosis. Bevacizumab 7.5 mg/kg was administered as a single-shot. d First scan 1.5 months later displayed a marked reduction of the edema while there was only a minor reduction of contrast enhancement. Diagnosis was changed from radiation necrosis to recurrent tumor
Fig. 3 Time to treatment failure. The swimmer plot shows the course of the individual patients labeled at the left side. The radiological diagnosis is indicated by color-coded dots (yellow: MRI, orange: MRI and MR-perfusion, purple: PET). The color-coded diamonds indicate the treatment failure of the single-shot bevacizumab (green: treatment of recurrent edema with corticosteroids, blue: treatment of recurrent edema with bevacizumab, blue border: resumed bevacizumab as the symptoms did not completely resolve, red: recurrent tumor, black: death of the patient because of recurrent tumor). Median time from single-shot to time of treatment failure of any cause) was three months
Discussion
Cerebral irradiation is an integral part of the treatment of brain cancer. One of the most severe complications is cerebral necrosis that can occur in patients with primary or metastatic brain tumors especially after a second course of irradiation for recurrent tumors. Despite promising efficacy in the treatment of cerebral radiation necrosis from smaller clinical trials, no application for bevacizumab approval for this indication has been filed. Reimbursement by insurance companies therefore remains difficult and is granted only on a case by case basis limiting the availability of bevacizumab. Assessing the efficacy of a singular bevacizumab treatment with a potentially more favorable side-effect profile and lesser financial burden than the cyclic treatment addresses a clinically important challenge. In the present work we show that a singular dose of bevacizumab resulted in significantly reduced edema on MRI sequences in all evaluable patients with two-thirds of patients reporting a meaningful improvement of clinical symptoms. In this context the very shot interval from radiation therapy to the development of brain lesions has to be noted. The mean time of two month is rather short for radiation necrosis in comparison to the trial by Levin et al. [9]. Therefore it is plausible that some of the patients suffered from a subacute or early delayed radiation reaction rather than manifest necrosis. Despite the small series, this study provides encouraging data, indicating that singular administration of bevacizumab might be a useful option for the treatment of radiation reaction / necrosis, especially in patients where prolonged bevacizumab treatment is not deemed feasible, as for example due to prior thromboembolic events or due to denial of therapy reimbursement. Additionally, even when bevacizumab is available for multiple treatments, a single-shot might be sufficient treatment for some patients. At the present time it is unclear whether a resumption of therapy in cases where a single-shot is not sufficient has disadvantages on the course of cerebral radiation necrosis.
In our analysis, we identified only one potentially severe side effect in one patient who was diagnosed with pulmonary artery embolism two months after bevacizumab. Whether this was instead attributable to immobility of the patient, who was later also diagnosed with deep vein thrombosis, or if this only contributed to the embolism remains unclear.
An interesting variant of the single-shot bevacizumab concept to further reduce the systemic side effects could be a local administration. The ongoing LIBERTI trial (NCT02819479) evaluates the efficacy of a single, intra-arterial dose of only 2.5 mg/kg bevacizumab [16]. Despite the large molecular mass of bevacizumab some penetration of the disrupted blood-brain barrier in regions of radiation injury appears possible [17]. The reduced side effect might come at the cost of a decreased duration of edema control. With a half-life of three weeks, the effect on the blood brain barrier might not be lasting, and the downside could be a rebound phenomenon with the need of bevacizumab re-challenge, as also indicated by the short time to treatment failure in our collective of only three months. The trial of Levin et al. with four administrations of 7.5 mg/ reported a relapse of radiation necrosis in 25 % of patients [9]. Zhuang et al. reported 14 patients with cerebral radiation necrosis who were treated with a lower dose of 5 mg/kg bevacizumab for at least 3 cycles of therapy [18]. MRI showed improvement in 13 of the 14 patients, but 10 of the 13 responsive patients exhibited a rebound phenomenon in the later follow-up. One option would be to further lower the dose of bevacizumab but keep the continuous administration. This approach of lowering the dose to 1 mg/kg bevacizumab every three weeks has revealed promising results in a phase 2 trial [19]. The trial included 21 patients and the grade of the edema index was improved in 19 patients. In contrast to Levin et al., no adverse events above grade 2 were reported. This concept has been further supported by case reports with low-dose bevacizumab and even longer intervals between the administration [20].
Prophylactic administration of a singular dose bevacizumab in high-risk situations (re-irradiation therapy, large irradiation fields or/and already present widespread edema prior to irradiation) could also be an option to consider. Thereby clinical deterioration might be prevented and the need for corticosteroids as well as the risk of a rebound effect after termination of bevacizumab treatment might be reduced. Such an approach has been explored in a phase 1 trial by Clarke et al. This trial included bevacizumab treatment to intensify the radiation dose of hypofractionated stereotactic re-irradiation [21]. This concept could be even more beneficial in cyber knife radiosurgery [22, 23].
While histological confirmation of cerebral radiation necrosis is clearly limited to only the most ambiguous cases, we here report a cohort of 11 patients whose radiological scans and dynamics indicated radiation necrosis. Two patients (patient 2 and 6) had treatment failure shortly after the single-shot due to tumor progress in the MRI and these were also patients where diagnosis was based on conventional MRI without MR-perfusion or MR-spectroscopy. A more selected collective of histologically diagnosed radiation necrosis could have shown more sustained effects of bevacizumab.
Conclusions
In summary, bevacizumab is an effective treatment for patients with cerebral radiation injury. Optimal dosing and intervals still have to be defined but most likely lower doses and longer intervals than investigated in previous trials [9] are sufficient. In the case of patients at high risk for side effects, single-shot of bevacizumab may be used as a test-dose and treatment can be continued when necessary.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Acknowledgements
None.
Authors’ contributions
All authors contributed to the study conception and design. Material preparation, data collection and analysis were performed by M.V., K.W. E.F., MT.F. and M.W.R. The first draft of the manuscript was written by M.V., J.P.S. and M.W.R and all authors commented on previous versions of the manuscript. All authors read and approved the final manuscript.
Funding
The authors received no specific funding for this work.
Availability of data and materials
Raw data were generated at the Dr. Senckenberg Institute of Neurooncology. The datasets generated are available from the corresponding author on reasonable request.
Ethics approval and consent to participate
Written consent by the individual patient for this retrospective data collection was waived by the ethics committee of the University Hospital Frankfurt; Goethe University which also approved the access to the patients’ data (IRB decision # 4/09, project SNO_01–08).
Consent for publication
No individual person’s data is included in this article.
Competing interests
J.P.S. has received honoraria for lectures or advisory board participation, consulting or travel grants from Abbvie, Roche, Boehringer, Bristol-Myers Squibb, Medac, Mundipharma and UCB. M.W.R. has received a grant from UCB. The other authors report no conflict of interest. | BEVACIZUMAB | DrugsGivenReaction | CC BY | 33596839 | 18,986,888 | 2021-02-17 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Off label use'. | Efficacy and safety of intravitreal aflibercept in ranibizumab-refractory patients with neovascular age-related macular degeneration.
BACKGROUND
Anti-vascular endothelial growth factor (anti-VEGF) agents have become the standard of care in neovascular age-related macular degeneration (nAMD). Despite generally excellent response rates to anti-VEGF therapy, some patients do not respond or may respond suboptimally. In the case of refractory or rapidly recurring fluid in nAMD, clinicians may switch to another anti-VEGF agent. TITAN was an observational study that assessed the effectiveness and safety of intravitreal aflibercept (IVT-AFL) in patients with nAMD refractory to ranibizumab who switched to IVT-AFL after less than 12 months of ranibizumab treatment in routine clinical practice in France.
METHODS
TITAN was an observational, retrospective and prospective 12-month study conducted at 28 centres in France. Patients with nAMD refractory to ranibizumab were enrolled. Patients who were switched from ranibizumab to IVT-AFL were followed for 12 months. Data were obtained from medical records for retrospectively included patients, and at routine follow-up visits for those included prospectively. The main outcome measure was percentage of patients who achieved treatment success (gain of ≥1 Early Treatment Diabetic Retinopathy Study letters in best-corrected visual acuity [BCVA] and/or any reduction in central retinal thickness [CRT]) from baseline to 12 months after switching. A sample size of 225 patients was determined based on a 2-sided 95% confidence interval with a width equal to 0.12 when the sample proportion was 0.70.
RESULTS
We analysed safety data (N = 217) and clinical outcomes from patients in the per-protocol population (n = 125). The mean (standard deviation) number of IVT-AFL injections was 7.5 (2.6). Treatment success was achieved in 68.8% of patients. Mean BCVA change from baseline to Month 12 was + 1.5 letters (P = 0.105) and the mean CRT change was - 45.0 μm (P < 0.001). In a subgroup analysis, in patients who received three initial monthly IVT-AFL injections, mean BCVA gain was 3.3 letters at Month 12 (P = 0.015). Excluding lack of efficacy and inappropriate scheduling of drug administration, the most common adverse event was eye pain (2.3%).
CONCLUSIONS
Switching ranibizumab-refractory patients with nAMD to IVT-AFL may improve visual outcomes in some patients, particularly those who receive three initial monthly injections.
BACKGROUND
ClinicalTrials.gov , NCT02321241 . First posted: December 22, 2014; Last update posted: July 2, 2018.
Background
Neovascular age-related macular degeneration (nAMD) is the most severe form of AMD and is the most common cause of legal blindness [1, 2]. Anti–vascular endothelial growth factor (anti-VEGF) agents, including intravitreal aflibercept (IVT-AFL) and ranibizumab, have become standard of care in nAMD [3–8]. Despite generally excellent response rates to anti-VEGF therapy, some patients do not respond or may respond suboptimally [3–8]. Some studies have shown reduced anatomic response over time with ranibizumab treatment in patients with nAMD, and there have been reports of loss of bioefficacy after repeated ranibizumab treatment [9–11].
In the case of refractory or rapidly recurring fluid in nAMD, clinicians may switch from the current anti-VEGF agent to another anti-VEGF agent [12]. The VIEW 1 and VIEW 2 studies [13] assessed the efficacy and safety of IVT-AFL in patients with nAMD and demonstrated non-inferiority of IVT-AFL 2 mg, given every 8 weeks after three initial monthly doses, versus ranibizumab 0.5 mg, every 4 weeks, in maintaining vision (loss of < 15 Early Treatment Diabetic Retinopathy [ETDRS] letters in best-corrected visual acuity [BCVA]) in treatment-naïve patients over a 12-month period [14–17].
Retrospective studies have shown that individualised IVT-AFL treatment can significantly reduce retinal fluid and preserve vision in patients with nAMD who are resistant to anti-VEGF agents [17–20]. However, prospective studies examining a switch to IVT-AFL in patients refractory to ranibizumab treatment in a real-world setting are scarce [21].
TITAN was an observational study that assessed the effectiveness and safety of IVT-AFL in patients with nAMD refractory to ranibizumab (persistence of intraretinal [IRF] and/or subretinal fluid [SRF]) who switched to IVT-AFL after less than 12 months of ranibizumab treatment in routine clinical practice in France.
Methods
Study design
TITAN (NCT02321241) was an observational 12-month study to assess the effectiveness and safety of IVT-AFL in patients with nAMD refractory to ranibizumab. The study was conducted in 28 centres in France and enrolled patients both retrospectively and prospectively. Data were analysed from patients who received IVT-AFL treatment between January 1, 2014 and December 31, 2015. The date of the first IVT-AFL injection was considered the baseline visit, and data collection continued for a maximum of 12 months after the first injection. For retrospectively enrolled patients, the number of office visits, eye exams and treatments were collected from medical records; for prospectively enrolled patients, this information was recorded at routine follow-up visits. The study protocol was approved by a French data privacy committee (Comité Consultatif sur le Traitement de l’Information en Matière de Recherche dans le Domaine de la Santé and Commission Nationale de l’Informatique et des Libertés). All patients provided written informed consent to participate.
Participants
Patients diagnosed with nAMD who had been treated with ranibizumab for > 3 but < 12 months and switched to prescribed IVT-AFL by their physician were included. Eligible patients must have been refractory to ranibizumab, defined as persistent IRF and/or SRF on optical coherence tomography despite treatment with ranibizumab, in accordance with the Haute Autorité de Santé recommendations of at least three injections of ranibizumab.
Patients excluded from the study were those with any ranibizumab-treated eyes for nAMD previously switched to IVT-AFL, absence of treatment criteria for IVT-AFL, eyes previously treated with photodynamic therapy, another retinal disease (diabetic retinopathy, diabetic macular oedema, myopia or angioid streaks) or participation in any interventional study.
The safety analysis set (SAS) included all patients who received ≥1 IVT-AFL treatment in any eye. The per-protocol (PP) population was defined as all patients in the full analysis set (FAS) (i.e., BCVA ≥35 letters at baseline, or delay of ≤380 days between first and last ranibizumab injection, or ≥ 3 ranibizumab injections), with BCVA and central retinal thickness (CRT) assessments prior to any treatment (including ranibizumab), at baseline, and at Month 12. The FAS included patients who received ≥1 IVT-AFL treatment and had BCVA and CRT assessments in the study eye at baseline and during follow-up. Patients from the FAS who received a BCVA/CRT assessment before any treatment (including ranibizumab) at enrolment and at Month 12 were included in the subgroup analysis. This subset of patients was stratified by whether or not they received three initial monthly IVT-AFL injections following the switch.
Outcomes
The primary endpoint was treatment success rate at 12 months (defined as a gain of ≥1 letter in BCVA and/or any decrease in CRT [in μm] between initial visit [first injection of IVT-AFL] and 12-month follow-up visit). BCVA was measured using ETDRS letters (preferentially) or any other visual scale. For data analysis, we transformed any other visual acuity score to ETDRS letter score.
Secondary outcomes included change in BCVA between baseline and final study visits (12 months after the first injection of IVT-AFL or study discontinuation), mean duration of ranibizumab treatment before initiation of IVT-AFL, and frequency and mean number of IVT-AFL injections over the study period. No BCVA data were collected at the end of ranibizumab treatment; however, switching to IVT-AFL occurred relatively soon after the last ranibizumab injection (median of 44.0 days). Therefore, the change in BCVA during ranibizumab treatment was estimated based on the change between BCVA values before any treatment and before first injection of IVT-AFL.
All adverse events (AEs) occurring after the first IVT-AFL injection were documented in the electronic case report form. All patient medical records were evaluated for demographic as well as clinical characteristics, and AEs were summarised using the Medical Dictionary for Regulatory Activities coding system. The event rates for single AEs were calculated based on the total number of documented patients and AEs were categorised according to connection with medication, seriousness, discontinuation of therapy and outcome.
Statistical analyses
A sample size of 225 patients was determined based on a 2-sided 95% confidence interval (CI) with a width equal to 0.12 when the sample proportion was 0.70. Primary analysis criteria (success rate) considered patients who discontinued IVT-AFL treatment prematurely to be treatment failures. We expressed the incidence of treatment success at 12 months as number and percentage of patients (n [%]) and provided a 2-sided 95% CI.
The main analysis of secondary criteria (BCVA and CRT) was performed without replacing missing values at study end. Change in BCVA (ETDRS letters) was expressed as mean (standard deviation [SD]) and provided a 2-sided 95% CI. We performed two sensitivity analyses with two imputation methods for missing data: imputation of missing value by the last observation carried forward (LOCF) method and imputation of missing data with the median value of population. Statistical analyses were conducted with SAS software release 9.4 (SAS Institute Inc., Cary, NC, USA).
Results
Participants
A total of 236 patients were screened. Of these, 217 were included in the SAS and 125 were included in the PP population (Fig. 1). Demographic characteristics are shown in Table 1.
Fig. 1 Patient disposition. *Three initial monthly injections (− 1/+ 2 weeks). The per-protocol population was defined as patients without protocol deviation and with BCVA/CRT assessments prior to any treatment (ranibizumab) at baseline and at Month 12 (n = 125). The FAS included patients who received ≥1 IVT-AFL treatment and had BCVA and CRT assessments in the study eye at baseline and during follow-up (n = 185). BCVA = best-corrected visual acuity; CRT = central retinal thickness; FAS = full analysis set; IVT-AFL = intravitreal aflibercept; SAS = safety analysis set
Table 1 Demographic Characteristics
Full Analysis Set
(n = 185) Per-Protocol Set
(n = 125)
Characteristic at baseline
Age, years 77.6 (8.5) 78.2 (8.0)
Female, n (%) 109 (58.9) 79 (63.2)
Duration of nAMD, months 8.5 (7.3) 7.4 (3.3)
Characteristic before any treatment, including ranibizumab
BCVA, ETDRS letters 62.4 (16.6)a 64.0 (15.6)
CRT, μm 368.3 (124.4)b 362.4 (115.2)c
Subretinal fluid, n (%) 133 (76.9)b 92 (74.2)c
Intraretinal fluid, n (%) 76 (43.9)b 55 (44.4)c
Pigment epithelium detachment, n (%) 125 (72.3)b 88 (71.0)c
Mean (standard deviation) unless otherwise stated
an = 163; bn = 161; cn = 116; values correspond to assessments before any treatment, including ranibizumab, had been administered
BCVA Best-corrected visual acuity, CRT Central retinal thickness, ETDRS Early Treatment Diabetic Retinopathy Study, nAMD Neovascular age-related macular degeneration
Ranibizumab treatment and outcomes prior to switching
Treatment with ranibizumab was initiated soon after diagnosis of nAMD, with a mean lapse of < 1 month (Table 2). Approximately 75% of patients had their first injection of ranibizumab < 0.3 months after diagnosis of nAMD. Patients received a mean (SD) of 5.1 (2.5) injections over 5.6 (4.3) months [22]. Approximately 50% of patients received ≥4 ranibizumab injections over a median duration of 4.6 months. Mean (SD) BCVA improved significantly with ranibizumab treatment prior to the switch (+ 2.2 [12.0] ETDRS letters, P = 0.046) (Table 2). Distribution of patients with SRF, IRF and subretinal pigment epithelium (sub-RPE) at baseline according to the absence or presence of SRF, IRF and sub-RPE before treatment (including ranibizumab), is shown in Fig. 2.
Table 2 Ranibizumab Treatment and Outcomes Before Switch to Intravitreal Aflibercept
Patients
(n = 125)
Delay between diagnosis of nAMD and first injection of ranibizumab, months 0.5 (1.2)
Number of ranibizumab injections 4.8 (1.9)
Duration of ranibizumab treatment, months 4.9 (2.8)
BCVA before any treatment (including ranibizumab), ETDRS letters 64.0 (15.6)
BCVA at baseline (switch to IVT-AFL), ETDRS letters 66.2 (12.1)
Change in BCVA from before the start of any treatment (including ranibizumab) and baseline (switch to IVT-AFL) 2.2 (12.0)
P value 0.046
All values are expressed as mean (standard deviation); per-protocol population
P value is for paired sample t test
BCVA Best-corrected visual acuity, ETDRS Early Treatment Diabetic Retinopathy Study, IVT-AFL Intravitreal aflibercept, nAMD Neovascular age-related macular degeneration
Fig. 2 Proportion of patients with SRF, IRF and sub-RPE. At baseline (prior to switch) (a) and at Month 12 (following switch) (b). IRF = intraretinal fluid; SRF = subretinal fluid; sub-RPE = sub-retinal pigment epithelium. Any treatment includes ranibizumab. *Twelve patients without SRF at baseline were found with SRF at least once over the follow-up and three of them switched from IVT-AFL to ranibizumab. †Twenty-three patients without IRF at baseline were found with IRF at least once over the follow-up and one of them switched from IVT-AFL to ranibizumab. ‡Thirteen patients without sub-RPE fluid at baseline were found with sub-RPE fluid at least once over the follow-up and two of them switched from IVT-AFL to ranibizumab
Intravitreal aflibercept treatment and outcomes following the switch
The success rate (proportion of patients with a gain of ≥1 letter in BCVA and/or any decrease in CRT [in μm] between baseline [the date of the first IVT-AFL injection] and 12-month follow-up visit) was 68.8%. The mean BCVA improvement from baseline to Month 12 was 1.5 letters (P = 0.105) (Table 3). At Month 12, 55.2% of patients (n = 69/125) were able to read ≥70 letters.
Table 3 Intravitreal Aflibercept Treatment and Outcomes Following Switch to Intravitreal Aflibercept
Patients
(n = 125)
Delay between last injection of ranibizumab and first IVT-AFL, days 61.2 (46.1)
Reasons for switch to IVT-AFL, n (%)
Refractory 122 (97.6%)
AE/SAE 3 (2.4%)
Other 0 (0%)
Time between diagnosis of nAMD and first IVT-AFL, months 7.4 (3.3)
Duration of IVT-AFL treatment, months 11.3 (3.1)
Follow-up duration, months 12.7 (2.0)
Number of IVT-AFL injections 7.5 (2.6)
BCVA, ETDRS letters
Baseline (switch to IVT-AFL) 66.2 (12.1)
Month 12 67.7 (13.6)
Change in BCVA score between baseline and 12 months 1.5 (10.3)
P value 0.105
CRT, μm
Baseline (switch to IVT-AFL) 331.2 (103.3)
Month 12 286.2 (84.7)
Change in CRT between baseline and 12 months −45.0 (101.1)
P value < 0.001
Patients with SRF,
n (%)
Baseline 87 (70.7)
Month 12 58 (48.3)
Patients with IRF,
n (%)
Baseline 44 (35.8)
Month 12 27 (22.5)
All values are reported as mean (standard deviation) unless otherwise indicated; per-protocol population
P values are for the paired samples t test
AE Adverse event, BCVA Best-corrected visual acuity, CRT Central retinal thickness, ETDRS Early Treatment Diabetic Retinopathy Study, IRF Intraretinal fluid;IVT-AFL Intravitreal aflibercept, nAMD Neovascular age-related macular degeneration, SAE Serious adverse event, SRF Subretinal fluid
Data on the timing of first injection, treatment duration and number of IVT-AFL injections are shown in Table 3. Given the presence of extreme values, the median data is presented instead of the mean. The delay between the last injection of ranibizumab and the first IVT-AFL injection ranged from 9 to 314 days (mean [SD], 61.2 [46.1]; median 43 days). The main reason for switching from ranibizumab to IVT-AFL was that the patient was considered refractory to ranibizumab.
The range of IVT-AFL treatments received over the 12-month study duration is shown in Fig. 3. More than half of patients (52.8%) received three initial monthly IVT-AFL injections. Overall, 17.6% (n = 22) of patients switched at least once from IVT-AFL to ranibizumab, and 5.6% (n = 7) switched back to IVT-AFL.
Fig. 3 Distribution of number of intravitreal aflibercept injections over the 12-month study. Per-protocol population. IVT-AFL = intravitreal aflibercept
Anatomic outcomes following the switch
Anatomic outcomes improved in patients who were refractory to ranibizumab and switched to IVT-AFL (Table 3). Mean CRT reduction from baseline to Month 12 was 45.0 μm (P < 0.001). The proportion of patients with SRF at baseline was 70.7% (n = 87/123) and at Month 12 was 48.3% (n = 58/120). The proportion of patients with IRF was 35.8% at baseline (n = 44/123) and 22.5% (n = 27/120) at Month 12, respectively. Differences between the proportions of patients with and without fluid at baseline and Month 12 were significant for SRF and IRF (McNemar test P < 0.001 and P = 0.014, respectively), but not sub-RPE. Distribution of patients with SRF, IRF and sub-RPE at Month 12 according to the absence or presence of SRF, IRF and sub-RPE at baseline are shown in Fig. 2.
Approximately one-quarter of patients (24.0%; n = 30/125) gained 0 to 4 letters; 16.8% (n = 21/125) gained 5 to 9 letters; 10.4% (n = 13/125) gained 10 to 14 letters; and 11.2% (n = 14/125) gained ≥15 letters from baseline to Month 12. Among patients gaining ≥15 letters, mean BCVA change was 20.6 (8.0) letters. Conversely, 8.0% (n = 10/125) lost ≥15 letters. Figure 4 shows the distribution of patients by final absolute BCVA subgroups (< 50; 50 to 55; 55 to 70; and ≥ 70 letters).
Fig. 4 Distribution of patients by absolute final BCVA subgroups (< 50, 50–55, 55–70 and ≥ 70 letters). *n = 125. Per-protocol population. BCVA = best–corrected visual acuity
Exploratory and subgroup analyses following switch
In an exploratory analysis conducted in the PP population, patients with < 3 months of ranibizumab treatment prior to baseline gained a mean (SD) of 4.0 (11.1) letters from baseline to Month 12 (P = 0.025). This gain was greater than that observed in patients with 3 to 5 months, 6 to 8 months, and ≥ 9 months of ranibizumab treatment prior to baseline (+ 0.1 [10.8], − 0.6 [8.2] and + 2.1 [7.8] letters, respectively) (Supplementary Fig. 1).
In a subgroup analysis in the FAS population (patients with BCVA/CRT assessment before any treatment [including ranibizumab], at baseline and at Month 12), the overall success rate was 66.7%, with higher rates in patients who received three initial monthly IVT-AFL injections compared with those who did not (69.0% vs 64.1%, respectively; Table 4). No statistical test was performed, but 95% CIs largely overlap.
Table 4 Success Rate by Use or Non-use of 3 Initial Monthly Injections (FAS Population)
Overall
(n = 135) With 3 initial monthly injections
(n = 71) Without 3 initial monthly injections
(n = 64)
Success rate at 12 months,a
n (%) [95% CI]
90 (66.7)
[58.0–74.5]
49 (69.0)
[56.9–79.5]
41 (64.1)
[51.1–75.7]
BCVA, ETDRS letters
Baseline 65.1 (13.7) 63.6 (12.3) 66.9 (15.1)
Month 12 66.9 (15.0) 66.8 (12.1) 66.9 (17.8)
Change in BCVA between initial visit and 12 months 1.7 (10.5) 3.3 (11.0) 0.0 (9.6)
P value 0.056 0.015 NS
CRT, μm
Baseline 328.0 (100.9) 338.6 (114.6) 316.3 (82.5)
Month 12 285.3 (84.3) 296.1 (88.0) 272.6 (78.6)
Change in CRT between initial visit and 12 months −47.0 (104.1) −45.5 (106.0) −48.8 (102.6)
P value < 0.001 < 0.001 < 0.001
All values are reported as mean (standard deviation) unless otherwise indicated
P values are for the paired samples t test
aPatients who gained ≥1 ETDRS letter (BCVA) and/or a reduction in CRT from baseline (prior to switch) to 12 months after the first intravitreal aflibercept injection
BCVA Best–corrected visual acuity, CI Confidence interval, CRT Central retinal thickness, ETDRS Early Treatment Diabetic Retinopathy Study, FAS Full analysis set, NS Not significant
At Month 12, in patients who received three initial monthly IVT-AFL injections, mean BCVA gain was 3.3 letters (P = 0.015). Mean BCVA was stable in patients who did not receive three initial monthly injections (Table 4). There was a trend in favour of higher BCVA gains among patients with lower initial BCVA.
Sensitivity analyses
Results of the sensitivity analyses were consistent with the results of the main analyses. Specifically, the success rate in the overall population was 68.8% when we replaced missing data by the median of the population, or by LOCF. Mean (SD) gain in BCVA was 1.5 (10.3) letters (P = 0.105) when missing data were replaced by the median of the population, or by LOCF. At Month 12, the proportion of patients who could read ≥70 letters was 55.2% for median value replacement and for LOCF value replacement.
Safety outcomes
Excluding lack of efficacy and inappropriate scheduling of drug administration, which were also considered AEs according to the protocol and occurred in 12.4 and 5.1% of patients, respectively, the most common AE was eye pain (2.3%) (Table 5). Treatment-emergent serious AEs were reported in 1.8% of patients, none associated with IVT-AFL treatment. The most common non-ocular AE was falls, occurring in 1.4% of patients. Twenty-seven patients discontinued IVT-AFL due to lack of efficacy and a further seven discontinued IVT-AFL due to other treatment-emergent AEs; in two patients, the AEs (both PED, one with retinal exudate) were considered related to IVT-AFL.
Discussion
Results from the TITAN study indicate that approximately two-thirds (67.7%) of patients with nAMD who switched from ranibizumab to IVT-AFL treatment (after ≤12 months of treatment with ranibizumab) achieved success, defined as a gain of ≥1 letter in BCVA and/or any decrease in CRT between baseline and Month 12. Patients considered refractory to first-line treatment with ranibizumab had an additional gain (beyond the mean 2.2 letters initially achieved with ranibizumab) of 1.5 letters at Month 12 after IVT-AFL treatment was initiated, indicating that switching from ranibizumab to IVT-AFL can be considered in ranibizumab-refractory nAMD. There was an inverse relationship between baseline BCVA and visual acuity gain at Month 12. Generally, patients with lower baseline BCVA had greater visual acuity gains at Month 12. Patients who had previously received ranibizumab for < 3 months prior to switch had a greater improvement in BCVA. In addition to decreases in CRT after switching treatment, an enhanced anatomic response was observed, with reductions in fluid accumulation observed in the SRF, IRF and sub-RPE compartments. Overall, these results suggest that considering a switch to IVT-AFL from ranibizumab after < 3 months may be warranted; however, these findings should be interpreted with caution as the TITAN study was not specifically designed to answer this question, and other confounding variables such as duration of disease are likely to have had an effect.
Findings from other studies of patients refractory to ranibizumab, who switched to IVT-AFL after varying durations of treatment, suggest that there may be a benefit to switching, particularly with anatomical benefit after the switch in terms of improvements in central retinal thickness and pigment epithelium detachment [23]. In fact, the effect of switching on functional outcomes has been shown to be variable [23] and, in our analysis, the overall mean change in BCVA was relatively small. However, over 20% of the ranibizumab-refractory patients achieved BCVA gains of ≥10 letters after receiving IVT-AFL, suggesting there may be specific subgroups of patients who could respond particularly well to switching to IVT-AFL, despite an initially poor response to anti-VEGF therapy with ranibizumab. Further randomised, controlled studies with appropriate control groups are, however, required.
The present study highlights the importance of three initial monthly injections at the start of IVT-AFL treatment. Greater visual improvements were observed in patients who received three initial monthly injections than in those who did not, which is consistent with previous studies of patients with nAMD who were switched from ranibizumab to IVT-AFL [21, 24]. It is notable that the proportion of patients who received three initial monthly injections in TITAN was lower (52.8%) than expected given the recommended dosing regimen in France [25]. Findings from this study were consistent with the known safety profile of IVT-AFL in nAMD [13, 26].
A few limitations inherent in the observational study design should be noted, including the use of different charts to evaluate visual acuity. Here, ETDRS letter charts or any other visual scale were used to evaluate visual acuity; if the latter, results were converted to ETDRS letters, potentially introducing a bias, especially when measuring the number of letters gained or lost after treatment. Also, our study design includes both retrospective and prospective components, and lacks a control group, which is a fundamental aspect of an observational study. In addition, findings are from a single European country, which may not be representative of other countries. Some data were imputed due to variability in data collection; a feature common in observational studies. Given that success was not achieved if patients discontinued IVT-AFL prematurely, even if they switched back to IVT-AFL at a later point, the proportion of patients achieving success may have been underestimated. The success rate was 70.3% (FAS) using median data. Patients who switched from IVT-AFL prior to receiving 12 months of treatment were considered failures; therefore, replacement of missing data by median value of population only slightly affected the success rate. Finally, factors that could influence differences in visual and anatomic outcomes between groups (e.g., disease severity at baseline, frequency of injections and physician monitoring) were not explored in the TITAN study.
Conclusions
The TITAN study demonstrated the effectiveness and safety of IVT-AFL in patients with ranibizumab-refractory nAMD in routine clinical practice in France. These findings suggest that an early switch (< 12 months) from ranibizumab to IVT-AFL may improve visual outcomes over 12 months for some patients in this difficult-to-treat population. The study further highlighted that initiating IVT-AFL treatment with three initial monthly injections after switching from ranibizumab may improve visual outcomes. Further studies with appropriate control groups are required to understand how to best identify those patients most likely to benefit from switching to IVT-AFL.
Supplementary Information
Additional file 1: Supplementary Figure 1. Visual gains in patients treated with ranibizumab prior to switching to IVT-AFL. Per-protocol population. BCVA = best–corrected visual acuity; IVT-AFL, intravitreal aflibercept.
Abbreviations
AEAdverse event
anti-VEGFAnti-vascular endothelial growth factor
BCVABest-corrected visual acuity
CRTCentral retinal thickness
ETDRSEarly Treatment Diabetic Retinopathy
FASFull analysis set
IRFIntraretinal fluid
IVT-AFLIntravitreal aflibercept
LOCFLast observation carried forward
nAMDNeovascular age-related macular degeneration
PPPer-protocol
SASSafety analysis set
SDStandard deviation
SRFSubretinal fluid
sub-RPESubretinal pigment epithelium
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Acknowledgements
The authors wish to thank the TITAN study investigators: Dr. Aouizerate, Dr. Averous, Professor Baillif, Dr. Benouaich, Dr. Benyelles, Professor Berrod, Dr. Cantaloube-Bessière, Dr. Chahed, Professor Chiambaretta, Dr. Cohen, Dr. Coscas, Dr. De Bats, Dr. Deudon-Combe, Dr. Dominguez, Dr. Donati, Professor Dot, Dr. Dumas, Dr. Giocanti-Auregan, Dr. Guigui, Professor Kodjikian, Dr. Oubraham, Dr. Rothschild, Dr. Rumen, Dr. Sampo, Dr. Scholtès, Dr. Sibille-Dabadie, Professor Souied, Dr. Stanescu, Dr. Tran, Dr. Uzzan, Dr. Wolff and Dr. Zerbib.
Medical writing assistance was provided by Apothecom, UK, and was funded by Bayer Consumer Care AG, Basel, Switzerland.
Authors’ contributions
All authors have read and approved the manuscript. SR contributed to the design; data acquisition, analysis and interpretation; and preparation and final review of the manuscript. LK contributed to the design; data acquisition, analysis and interpretation; and preparation and final review of the manuscript. AGA contributed to the design; data acquisition, analysis and interpretation; and preparation and final review of the manuscript. ID contributed to the design; data acquisition, analysis and interpretation; and preparation and final review of the manuscript. ES contributed to data acquisition, analysis and interpretation; and preparation and final review of the manuscript.
Funding
The TITAN study was sponsored by Bayer HealthCare SAS, France. The sponsor participated in the design and conduct of the study, analysis of the data, and preparation of the manuscript.
Availability of data and materials
Availability of the data underlying this publication will be determined according to Bayer’s commitment to the EFPIA/PhRMA “Principles for responsible clinical trial data sharing”. This pertains to scope, time point and process of data access.
As such, Bayer commits to sharing, upon request from qualified scientific and medical researchers, patient-level clinical trial data, study-level clinical trial data, and protocols from clinical trials in patients for medicines and indications approved in the United States (US) and European Union (EU) as necessary for conducting legitimate research. This applies to data on new medicines and indications that have been approved by the EU and US regulatory agencies on or after January 1, 2014.
Interested researchers can use www.clinicalstudydatarequest.com to request access to anonymised patient-level data and supporting documents from clinical studies to conduct further research that can help advance medical science or improve patient care. Information on the Bayer criteria for listing studies and other relevant information is provided in the Study sponsors section of the portal.
Ethics approval and consent to participate
Bayer France received a positive statement from the CCTIRS (Comité Consultatif sur le Traitement de l’Information en Matière de Recherche dans le Domaine de la Santé: Advisory Committee on Information Processing in Material Research in the Field of Health) on 26 November 2014 and an authorization from the CNIL (Commission Nationale de l’Informatique et des Libertés: National Commission on Computer Technology and Freedom) on 24 December 2015 concerning the TITAN study. These gave Bayer the possibility to collect, analyse and use anonymised data of patients included in this study. All patients provided written informed consent to participate.
Consent for publication
Not applicable.
Competing interests
SR: Consulting fees from Allergan, Bayer and Novartis; LK: Financial support from Novartis, Allergan, Bayer, Thea and Alcon; consulting fees from Alcon, Alimera, Allergan, Bayer, Roche and Novartis; research funding from Alcon, Alimera, Allergan, Bayer, Horus, Novartis and Thea Pharmaceuticals; AGA: Consulting fees from Alimera, Allergan, Bayer and Novartis; ID: Employee of Bayer HealthCare SAS; ES: Consulting fees and financial support from Allergan, Bayer, Novartis, Roche and Thea Pharmaceuticals. | AFLIBERCEPT | DrugsGivenReaction | CC BY | 33596867 | 19,020,429 | 2021-02-17 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Retinal exudates'. | Efficacy and safety of intravitreal aflibercept in ranibizumab-refractory patients with neovascular age-related macular degeneration.
BACKGROUND
Anti-vascular endothelial growth factor (anti-VEGF) agents have become the standard of care in neovascular age-related macular degeneration (nAMD). Despite generally excellent response rates to anti-VEGF therapy, some patients do not respond or may respond suboptimally. In the case of refractory or rapidly recurring fluid in nAMD, clinicians may switch to another anti-VEGF agent. TITAN was an observational study that assessed the effectiveness and safety of intravitreal aflibercept (IVT-AFL) in patients with nAMD refractory to ranibizumab who switched to IVT-AFL after less than 12 months of ranibizumab treatment in routine clinical practice in France.
METHODS
TITAN was an observational, retrospective and prospective 12-month study conducted at 28 centres in France. Patients with nAMD refractory to ranibizumab were enrolled. Patients who were switched from ranibizumab to IVT-AFL were followed for 12 months. Data were obtained from medical records for retrospectively included patients, and at routine follow-up visits for those included prospectively. The main outcome measure was percentage of patients who achieved treatment success (gain of ≥1 Early Treatment Diabetic Retinopathy Study letters in best-corrected visual acuity [BCVA] and/or any reduction in central retinal thickness [CRT]) from baseline to 12 months after switching. A sample size of 225 patients was determined based on a 2-sided 95% confidence interval with a width equal to 0.12 when the sample proportion was 0.70.
RESULTS
We analysed safety data (N = 217) and clinical outcomes from patients in the per-protocol population (n = 125). The mean (standard deviation) number of IVT-AFL injections was 7.5 (2.6). Treatment success was achieved in 68.8% of patients. Mean BCVA change from baseline to Month 12 was + 1.5 letters (P = 0.105) and the mean CRT change was - 45.0 μm (P < 0.001). In a subgroup analysis, in patients who received three initial monthly IVT-AFL injections, mean BCVA gain was 3.3 letters at Month 12 (P = 0.015). Excluding lack of efficacy and inappropriate scheduling of drug administration, the most common adverse event was eye pain (2.3%).
CONCLUSIONS
Switching ranibizumab-refractory patients with nAMD to IVT-AFL may improve visual outcomes in some patients, particularly those who receive three initial monthly injections.
BACKGROUND
ClinicalTrials.gov , NCT02321241 . First posted: December 22, 2014; Last update posted: July 2, 2018.
Background
Neovascular age-related macular degeneration (nAMD) is the most severe form of AMD and is the most common cause of legal blindness [1, 2]. Anti–vascular endothelial growth factor (anti-VEGF) agents, including intravitreal aflibercept (IVT-AFL) and ranibizumab, have become standard of care in nAMD [3–8]. Despite generally excellent response rates to anti-VEGF therapy, some patients do not respond or may respond suboptimally [3–8]. Some studies have shown reduced anatomic response over time with ranibizumab treatment in patients with nAMD, and there have been reports of loss of bioefficacy after repeated ranibizumab treatment [9–11].
In the case of refractory or rapidly recurring fluid in nAMD, clinicians may switch from the current anti-VEGF agent to another anti-VEGF agent [12]. The VIEW 1 and VIEW 2 studies [13] assessed the efficacy and safety of IVT-AFL in patients with nAMD and demonstrated non-inferiority of IVT-AFL 2 mg, given every 8 weeks after three initial monthly doses, versus ranibizumab 0.5 mg, every 4 weeks, in maintaining vision (loss of < 15 Early Treatment Diabetic Retinopathy [ETDRS] letters in best-corrected visual acuity [BCVA]) in treatment-naïve patients over a 12-month period [14–17].
Retrospective studies have shown that individualised IVT-AFL treatment can significantly reduce retinal fluid and preserve vision in patients with nAMD who are resistant to anti-VEGF agents [17–20]. However, prospective studies examining a switch to IVT-AFL in patients refractory to ranibizumab treatment in a real-world setting are scarce [21].
TITAN was an observational study that assessed the effectiveness and safety of IVT-AFL in patients with nAMD refractory to ranibizumab (persistence of intraretinal [IRF] and/or subretinal fluid [SRF]) who switched to IVT-AFL after less than 12 months of ranibizumab treatment in routine clinical practice in France.
Methods
Study design
TITAN (NCT02321241) was an observational 12-month study to assess the effectiveness and safety of IVT-AFL in patients with nAMD refractory to ranibizumab. The study was conducted in 28 centres in France and enrolled patients both retrospectively and prospectively. Data were analysed from patients who received IVT-AFL treatment between January 1, 2014 and December 31, 2015. The date of the first IVT-AFL injection was considered the baseline visit, and data collection continued for a maximum of 12 months after the first injection. For retrospectively enrolled patients, the number of office visits, eye exams and treatments were collected from medical records; for prospectively enrolled patients, this information was recorded at routine follow-up visits. The study protocol was approved by a French data privacy committee (Comité Consultatif sur le Traitement de l’Information en Matière de Recherche dans le Domaine de la Santé and Commission Nationale de l’Informatique et des Libertés). All patients provided written informed consent to participate.
Participants
Patients diagnosed with nAMD who had been treated with ranibizumab for > 3 but < 12 months and switched to prescribed IVT-AFL by their physician were included. Eligible patients must have been refractory to ranibizumab, defined as persistent IRF and/or SRF on optical coherence tomography despite treatment with ranibizumab, in accordance with the Haute Autorité de Santé recommendations of at least three injections of ranibizumab.
Patients excluded from the study were those with any ranibizumab-treated eyes for nAMD previously switched to IVT-AFL, absence of treatment criteria for IVT-AFL, eyes previously treated with photodynamic therapy, another retinal disease (diabetic retinopathy, diabetic macular oedema, myopia or angioid streaks) or participation in any interventional study.
The safety analysis set (SAS) included all patients who received ≥1 IVT-AFL treatment in any eye. The per-protocol (PP) population was defined as all patients in the full analysis set (FAS) (i.e., BCVA ≥35 letters at baseline, or delay of ≤380 days between first and last ranibizumab injection, or ≥ 3 ranibizumab injections), with BCVA and central retinal thickness (CRT) assessments prior to any treatment (including ranibizumab), at baseline, and at Month 12. The FAS included patients who received ≥1 IVT-AFL treatment and had BCVA and CRT assessments in the study eye at baseline and during follow-up. Patients from the FAS who received a BCVA/CRT assessment before any treatment (including ranibizumab) at enrolment and at Month 12 were included in the subgroup analysis. This subset of patients was stratified by whether or not they received three initial monthly IVT-AFL injections following the switch.
Outcomes
The primary endpoint was treatment success rate at 12 months (defined as a gain of ≥1 letter in BCVA and/or any decrease in CRT [in μm] between initial visit [first injection of IVT-AFL] and 12-month follow-up visit). BCVA was measured using ETDRS letters (preferentially) or any other visual scale. For data analysis, we transformed any other visual acuity score to ETDRS letter score.
Secondary outcomes included change in BCVA between baseline and final study visits (12 months after the first injection of IVT-AFL or study discontinuation), mean duration of ranibizumab treatment before initiation of IVT-AFL, and frequency and mean number of IVT-AFL injections over the study period. No BCVA data were collected at the end of ranibizumab treatment; however, switching to IVT-AFL occurred relatively soon after the last ranibizumab injection (median of 44.0 days). Therefore, the change in BCVA during ranibizumab treatment was estimated based on the change between BCVA values before any treatment and before first injection of IVT-AFL.
All adverse events (AEs) occurring after the first IVT-AFL injection were documented in the electronic case report form. All patient medical records were evaluated for demographic as well as clinical characteristics, and AEs were summarised using the Medical Dictionary for Regulatory Activities coding system. The event rates for single AEs were calculated based on the total number of documented patients and AEs were categorised according to connection with medication, seriousness, discontinuation of therapy and outcome.
Statistical analyses
A sample size of 225 patients was determined based on a 2-sided 95% confidence interval (CI) with a width equal to 0.12 when the sample proportion was 0.70. Primary analysis criteria (success rate) considered patients who discontinued IVT-AFL treatment prematurely to be treatment failures. We expressed the incidence of treatment success at 12 months as number and percentage of patients (n [%]) and provided a 2-sided 95% CI.
The main analysis of secondary criteria (BCVA and CRT) was performed without replacing missing values at study end. Change in BCVA (ETDRS letters) was expressed as mean (standard deviation [SD]) and provided a 2-sided 95% CI. We performed two sensitivity analyses with two imputation methods for missing data: imputation of missing value by the last observation carried forward (LOCF) method and imputation of missing data with the median value of population. Statistical analyses were conducted with SAS software release 9.4 (SAS Institute Inc., Cary, NC, USA).
Results
Participants
A total of 236 patients were screened. Of these, 217 were included in the SAS and 125 were included in the PP population (Fig. 1). Demographic characteristics are shown in Table 1.
Fig. 1 Patient disposition. *Three initial monthly injections (− 1/+ 2 weeks). The per-protocol population was defined as patients without protocol deviation and with BCVA/CRT assessments prior to any treatment (ranibizumab) at baseline and at Month 12 (n = 125). The FAS included patients who received ≥1 IVT-AFL treatment and had BCVA and CRT assessments in the study eye at baseline and during follow-up (n = 185). BCVA = best-corrected visual acuity; CRT = central retinal thickness; FAS = full analysis set; IVT-AFL = intravitreal aflibercept; SAS = safety analysis set
Table 1 Demographic Characteristics
Full Analysis Set
(n = 185) Per-Protocol Set
(n = 125)
Characteristic at baseline
Age, years 77.6 (8.5) 78.2 (8.0)
Female, n (%) 109 (58.9) 79 (63.2)
Duration of nAMD, months 8.5 (7.3) 7.4 (3.3)
Characteristic before any treatment, including ranibizumab
BCVA, ETDRS letters 62.4 (16.6)a 64.0 (15.6)
CRT, μm 368.3 (124.4)b 362.4 (115.2)c
Subretinal fluid, n (%) 133 (76.9)b 92 (74.2)c
Intraretinal fluid, n (%) 76 (43.9)b 55 (44.4)c
Pigment epithelium detachment, n (%) 125 (72.3)b 88 (71.0)c
Mean (standard deviation) unless otherwise stated
an = 163; bn = 161; cn = 116; values correspond to assessments before any treatment, including ranibizumab, had been administered
BCVA Best-corrected visual acuity, CRT Central retinal thickness, ETDRS Early Treatment Diabetic Retinopathy Study, nAMD Neovascular age-related macular degeneration
Ranibizumab treatment and outcomes prior to switching
Treatment with ranibizumab was initiated soon after diagnosis of nAMD, with a mean lapse of < 1 month (Table 2). Approximately 75% of patients had their first injection of ranibizumab < 0.3 months after diagnosis of nAMD. Patients received a mean (SD) of 5.1 (2.5) injections over 5.6 (4.3) months [22]. Approximately 50% of patients received ≥4 ranibizumab injections over a median duration of 4.6 months. Mean (SD) BCVA improved significantly with ranibizumab treatment prior to the switch (+ 2.2 [12.0] ETDRS letters, P = 0.046) (Table 2). Distribution of patients with SRF, IRF and subretinal pigment epithelium (sub-RPE) at baseline according to the absence or presence of SRF, IRF and sub-RPE before treatment (including ranibizumab), is shown in Fig. 2.
Table 2 Ranibizumab Treatment and Outcomes Before Switch to Intravitreal Aflibercept
Patients
(n = 125)
Delay between diagnosis of nAMD and first injection of ranibizumab, months 0.5 (1.2)
Number of ranibizumab injections 4.8 (1.9)
Duration of ranibizumab treatment, months 4.9 (2.8)
BCVA before any treatment (including ranibizumab), ETDRS letters 64.0 (15.6)
BCVA at baseline (switch to IVT-AFL), ETDRS letters 66.2 (12.1)
Change in BCVA from before the start of any treatment (including ranibizumab) and baseline (switch to IVT-AFL) 2.2 (12.0)
P value 0.046
All values are expressed as mean (standard deviation); per-protocol population
P value is for paired sample t test
BCVA Best-corrected visual acuity, ETDRS Early Treatment Diabetic Retinopathy Study, IVT-AFL Intravitreal aflibercept, nAMD Neovascular age-related macular degeneration
Fig. 2 Proportion of patients with SRF, IRF and sub-RPE. At baseline (prior to switch) (a) and at Month 12 (following switch) (b). IRF = intraretinal fluid; SRF = subretinal fluid; sub-RPE = sub-retinal pigment epithelium. Any treatment includes ranibizumab. *Twelve patients without SRF at baseline were found with SRF at least once over the follow-up and three of them switched from IVT-AFL to ranibizumab. †Twenty-three patients without IRF at baseline were found with IRF at least once over the follow-up and one of them switched from IVT-AFL to ranibizumab. ‡Thirteen patients without sub-RPE fluid at baseline were found with sub-RPE fluid at least once over the follow-up and two of them switched from IVT-AFL to ranibizumab
Intravitreal aflibercept treatment and outcomes following the switch
The success rate (proportion of patients with a gain of ≥1 letter in BCVA and/or any decrease in CRT [in μm] between baseline [the date of the first IVT-AFL injection] and 12-month follow-up visit) was 68.8%. The mean BCVA improvement from baseline to Month 12 was 1.5 letters (P = 0.105) (Table 3). At Month 12, 55.2% of patients (n = 69/125) were able to read ≥70 letters.
Table 3 Intravitreal Aflibercept Treatment and Outcomes Following Switch to Intravitreal Aflibercept
Patients
(n = 125)
Delay between last injection of ranibizumab and first IVT-AFL, days 61.2 (46.1)
Reasons for switch to IVT-AFL, n (%)
Refractory 122 (97.6%)
AE/SAE 3 (2.4%)
Other 0 (0%)
Time between diagnosis of nAMD and first IVT-AFL, months 7.4 (3.3)
Duration of IVT-AFL treatment, months 11.3 (3.1)
Follow-up duration, months 12.7 (2.0)
Number of IVT-AFL injections 7.5 (2.6)
BCVA, ETDRS letters
Baseline (switch to IVT-AFL) 66.2 (12.1)
Month 12 67.7 (13.6)
Change in BCVA score between baseline and 12 months 1.5 (10.3)
P value 0.105
CRT, μm
Baseline (switch to IVT-AFL) 331.2 (103.3)
Month 12 286.2 (84.7)
Change in CRT between baseline and 12 months −45.0 (101.1)
P value < 0.001
Patients with SRF,
n (%)
Baseline 87 (70.7)
Month 12 58 (48.3)
Patients with IRF,
n (%)
Baseline 44 (35.8)
Month 12 27 (22.5)
All values are reported as mean (standard deviation) unless otherwise indicated; per-protocol population
P values are for the paired samples t test
AE Adverse event, BCVA Best-corrected visual acuity, CRT Central retinal thickness, ETDRS Early Treatment Diabetic Retinopathy Study, IRF Intraretinal fluid;IVT-AFL Intravitreal aflibercept, nAMD Neovascular age-related macular degeneration, SAE Serious adverse event, SRF Subretinal fluid
Data on the timing of first injection, treatment duration and number of IVT-AFL injections are shown in Table 3. Given the presence of extreme values, the median data is presented instead of the mean. The delay between the last injection of ranibizumab and the first IVT-AFL injection ranged from 9 to 314 days (mean [SD], 61.2 [46.1]; median 43 days). The main reason for switching from ranibizumab to IVT-AFL was that the patient was considered refractory to ranibizumab.
The range of IVT-AFL treatments received over the 12-month study duration is shown in Fig. 3. More than half of patients (52.8%) received three initial monthly IVT-AFL injections. Overall, 17.6% (n = 22) of patients switched at least once from IVT-AFL to ranibizumab, and 5.6% (n = 7) switched back to IVT-AFL.
Fig. 3 Distribution of number of intravitreal aflibercept injections over the 12-month study. Per-protocol population. IVT-AFL = intravitreal aflibercept
Anatomic outcomes following the switch
Anatomic outcomes improved in patients who were refractory to ranibizumab and switched to IVT-AFL (Table 3). Mean CRT reduction from baseline to Month 12 was 45.0 μm (P < 0.001). The proportion of patients with SRF at baseline was 70.7% (n = 87/123) and at Month 12 was 48.3% (n = 58/120). The proportion of patients with IRF was 35.8% at baseline (n = 44/123) and 22.5% (n = 27/120) at Month 12, respectively. Differences between the proportions of patients with and without fluid at baseline and Month 12 were significant for SRF and IRF (McNemar test P < 0.001 and P = 0.014, respectively), but not sub-RPE. Distribution of patients with SRF, IRF and sub-RPE at Month 12 according to the absence or presence of SRF, IRF and sub-RPE at baseline are shown in Fig. 2.
Approximately one-quarter of patients (24.0%; n = 30/125) gained 0 to 4 letters; 16.8% (n = 21/125) gained 5 to 9 letters; 10.4% (n = 13/125) gained 10 to 14 letters; and 11.2% (n = 14/125) gained ≥15 letters from baseline to Month 12. Among patients gaining ≥15 letters, mean BCVA change was 20.6 (8.0) letters. Conversely, 8.0% (n = 10/125) lost ≥15 letters. Figure 4 shows the distribution of patients by final absolute BCVA subgroups (< 50; 50 to 55; 55 to 70; and ≥ 70 letters).
Fig. 4 Distribution of patients by absolute final BCVA subgroups (< 50, 50–55, 55–70 and ≥ 70 letters). *n = 125. Per-protocol population. BCVA = best–corrected visual acuity
Exploratory and subgroup analyses following switch
In an exploratory analysis conducted in the PP population, patients with < 3 months of ranibizumab treatment prior to baseline gained a mean (SD) of 4.0 (11.1) letters from baseline to Month 12 (P = 0.025). This gain was greater than that observed in patients with 3 to 5 months, 6 to 8 months, and ≥ 9 months of ranibizumab treatment prior to baseline (+ 0.1 [10.8], − 0.6 [8.2] and + 2.1 [7.8] letters, respectively) (Supplementary Fig. 1).
In a subgroup analysis in the FAS population (patients with BCVA/CRT assessment before any treatment [including ranibizumab], at baseline and at Month 12), the overall success rate was 66.7%, with higher rates in patients who received three initial monthly IVT-AFL injections compared with those who did not (69.0% vs 64.1%, respectively; Table 4). No statistical test was performed, but 95% CIs largely overlap.
Table 4 Success Rate by Use or Non-use of 3 Initial Monthly Injections (FAS Population)
Overall
(n = 135) With 3 initial monthly injections
(n = 71) Without 3 initial monthly injections
(n = 64)
Success rate at 12 months,a
n (%) [95% CI]
90 (66.7)
[58.0–74.5]
49 (69.0)
[56.9–79.5]
41 (64.1)
[51.1–75.7]
BCVA, ETDRS letters
Baseline 65.1 (13.7) 63.6 (12.3) 66.9 (15.1)
Month 12 66.9 (15.0) 66.8 (12.1) 66.9 (17.8)
Change in BCVA between initial visit and 12 months 1.7 (10.5) 3.3 (11.0) 0.0 (9.6)
P value 0.056 0.015 NS
CRT, μm
Baseline 328.0 (100.9) 338.6 (114.6) 316.3 (82.5)
Month 12 285.3 (84.3) 296.1 (88.0) 272.6 (78.6)
Change in CRT between initial visit and 12 months −47.0 (104.1) −45.5 (106.0) −48.8 (102.6)
P value < 0.001 < 0.001 < 0.001
All values are reported as mean (standard deviation) unless otherwise indicated
P values are for the paired samples t test
aPatients who gained ≥1 ETDRS letter (BCVA) and/or a reduction in CRT from baseline (prior to switch) to 12 months after the first intravitreal aflibercept injection
BCVA Best–corrected visual acuity, CI Confidence interval, CRT Central retinal thickness, ETDRS Early Treatment Diabetic Retinopathy Study, FAS Full analysis set, NS Not significant
At Month 12, in patients who received three initial monthly IVT-AFL injections, mean BCVA gain was 3.3 letters (P = 0.015). Mean BCVA was stable in patients who did not receive three initial monthly injections (Table 4). There was a trend in favour of higher BCVA gains among patients with lower initial BCVA.
Sensitivity analyses
Results of the sensitivity analyses were consistent with the results of the main analyses. Specifically, the success rate in the overall population was 68.8% when we replaced missing data by the median of the population, or by LOCF. Mean (SD) gain in BCVA was 1.5 (10.3) letters (P = 0.105) when missing data were replaced by the median of the population, or by LOCF. At Month 12, the proportion of patients who could read ≥70 letters was 55.2% for median value replacement and for LOCF value replacement.
Safety outcomes
Excluding lack of efficacy and inappropriate scheduling of drug administration, which were also considered AEs according to the protocol and occurred in 12.4 and 5.1% of patients, respectively, the most common AE was eye pain (2.3%) (Table 5). Treatment-emergent serious AEs were reported in 1.8% of patients, none associated with IVT-AFL treatment. The most common non-ocular AE was falls, occurring in 1.4% of patients. Twenty-seven patients discontinued IVT-AFL due to lack of efficacy and a further seven discontinued IVT-AFL due to other treatment-emergent AEs; in two patients, the AEs (both PED, one with retinal exudate) were considered related to IVT-AFL.
Discussion
Results from the TITAN study indicate that approximately two-thirds (67.7%) of patients with nAMD who switched from ranibizumab to IVT-AFL treatment (after ≤12 months of treatment with ranibizumab) achieved success, defined as a gain of ≥1 letter in BCVA and/or any decrease in CRT between baseline and Month 12. Patients considered refractory to first-line treatment with ranibizumab had an additional gain (beyond the mean 2.2 letters initially achieved with ranibizumab) of 1.5 letters at Month 12 after IVT-AFL treatment was initiated, indicating that switching from ranibizumab to IVT-AFL can be considered in ranibizumab-refractory nAMD. There was an inverse relationship between baseline BCVA and visual acuity gain at Month 12. Generally, patients with lower baseline BCVA had greater visual acuity gains at Month 12. Patients who had previously received ranibizumab for < 3 months prior to switch had a greater improvement in BCVA. In addition to decreases in CRT after switching treatment, an enhanced anatomic response was observed, with reductions in fluid accumulation observed in the SRF, IRF and sub-RPE compartments. Overall, these results suggest that considering a switch to IVT-AFL from ranibizumab after < 3 months may be warranted; however, these findings should be interpreted with caution as the TITAN study was not specifically designed to answer this question, and other confounding variables such as duration of disease are likely to have had an effect.
Findings from other studies of patients refractory to ranibizumab, who switched to IVT-AFL after varying durations of treatment, suggest that there may be a benefit to switching, particularly with anatomical benefit after the switch in terms of improvements in central retinal thickness and pigment epithelium detachment [23]. In fact, the effect of switching on functional outcomes has been shown to be variable [23] and, in our analysis, the overall mean change in BCVA was relatively small. However, over 20% of the ranibizumab-refractory patients achieved BCVA gains of ≥10 letters after receiving IVT-AFL, suggesting there may be specific subgroups of patients who could respond particularly well to switching to IVT-AFL, despite an initially poor response to anti-VEGF therapy with ranibizumab. Further randomised, controlled studies with appropriate control groups are, however, required.
The present study highlights the importance of three initial monthly injections at the start of IVT-AFL treatment. Greater visual improvements were observed in patients who received three initial monthly injections than in those who did not, which is consistent with previous studies of patients with nAMD who were switched from ranibizumab to IVT-AFL [21, 24]. It is notable that the proportion of patients who received three initial monthly injections in TITAN was lower (52.8%) than expected given the recommended dosing regimen in France [25]. Findings from this study were consistent with the known safety profile of IVT-AFL in nAMD [13, 26].
A few limitations inherent in the observational study design should be noted, including the use of different charts to evaluate visual acuity. Here, ETDRS letter charts or any other visual scale were used to evaluate visual acuity; if the latter, results were converted to ETDRS letters, potentially introducing a bias, especially when measuring the number of letters gained or lost after treatment. Also, our study design includes both retrospective and prospective components, and lacks a control group, which is a fundamental aspect of an observational study. In addition, findings are from a single European country, which may not be representative of other countries. Some data were imputed due to variability in data collection; a feature common in observational studies. Given that success was not achieved if patients discontinued IVT-AFL prematurely, even if they switched back to IVT-AFL at a later point, the proportion of patients achieving success may have been underestimated. The success rate was 70.3% (FAS) using median data. Patients who switched from IVT-AFL prior to receiving 12 months of treatment were considered failures; therefore, replacement of missing data by median value of population only slightly affected the success rate. Finally, factors that could influence differences in visual and anatomic outcomes between groups (e.g., disease severity at baseline, frequency of injections and physician monitoring) were not explored in the TITAN study.
Conclusions
The TITAN study demonstrated the effectiveness and safety of IVT-AFL in patients with ranibizumab-refractory nAMD in routine clinical practice in France. These findings suggest that an early switch (< 12 months) from ranibizumab to IVT-AFL may improve visual outcomes over 12 months for some patients in this difficult-to-treat population. The study further highlighted that initiating IVT-AFL treatment with three initial monthly injections after switching from ranibizumab may improve visual outcomes. Further studies with appropriate control groups are required to understand how to best identify those patients most likely to benefit from switching to IVT-AFL.
Supplementary Information
Additional file 1: Supplementary Figure 1. Visual gains in patients treated with ranibizumab prior to switching to IVT-AFL. Per-protocol population. BCVA = best–corrected visual acuity; IVT-AFL, intravitreal aflibercept.
Abbreviations
AEAdverse event
anti-VEGFAnti-vascular endothelial growth factor
BCVABest-corrected visual acuity
CRTCentral retinal thickness
ETDRSEarly Treatment Diabetic Retinopathy
FASFull analysis set
IRFIntraretinal fluid
IVT-AFLIntravitreal aflibercept
LOCFLast observation carried forward
nAMDNeovascular age-related macular degeneration
PPPer-protocol
SASSafety analysis set
SDStandard deviation
SRFSubretinal fluid
sub-RPESubretinal pigment epithelium
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Acknowledgements
The authors wish to thank the TITAN study investigators: Dr. Aouizerate, Dr. Averous, Professor Baillif, Dr. Benouaich, Dr. Benyelles, Professor Berrod, Dr. Cantaloube-Bessière, Dr. Chahed, Professor Chiambaretta, Dr. Cohen, Dr. Coscas, Dr. De Bats, Dr. Deudon-Combe, Dr. Dominguez, Dr. Donati, Professor Dot, Dr. Dumas, Dr. Giocanti-Auregan, Dr. Guigui, Professor Kodjikian, Dr. Oubraham, Dr. Rothschild, Dr. Rumen, Dr. Sampo, Dr. Scholtès, Dr. Sibille-Dabadie, Professor Souied, Dr. Stanescu, Dr. Tran, Dr. Uzzan, Dr. Wolff and Dr. Zerbib.
Medical writing assistance was provided by Apothecom, UK, and was funded by Bayer Consumer Care AG, Basel, Switzerland.
Authors’ contributions
All authors have read and approved the manuscript. SR contributed to the design; data acquisition, analysis and interpretation; and preparation and final review of the manuscript. LK contributed to the design; data acquisition, analysis and interpretation; and preparation and final review of the manuscript. AGA contributed to the design; data acquisition, analysis and interpretation; and preparation and final review of the manuscript. ID contributed to the design; data acquisition, analysis and interpretation; and preparation and final review of the manuscript. ES contributed to data acquisition, analysis and interpretation; and preparation and final review of the manuscript.
Funding
The TITAN study was sponsored by Bayer HealthCare SAS, France. The sponsor participated in the design and conduct of the study, analysis of the data, and preparation of the manuscript.
Availability of data and materials
Availability of the data underlying this publication will be determined according to Bayer’s commitment to the EFPIA/PhRMA “Principles for responsible clinical trial data sharing”. This pertains to scope, time point and process of data access.
As such, Bayer commits to sharing, upon request from qualified scientific and medical researchers, patient-level clinical trial data, study-level clinical trial data, and protocols from clinical trials in patients for medicines and indications approved in the United States (US) and European Union (EU) as necessary for conducting legitimate research. This applies to data on new medicines and indications that have been approved by the EU and US regulatory agencies on or after January 1, 2014.
Interested researchers can use www.clinicalstudydatarequest.com to request access to anonymised patient-level data and supporting documents from clinical studies to conduct further research that can help advance medical science or improve patient care. Information on the Bayer criteria for listing studies and other relevant information is provided in the Study sponsors section of the portal.
Ethics approval and consent to participate
Bayer France received a positive statement from the CCTIRS (Comité Consultatif sur le Traitement de l’Information en Matière de Recherche dans le Domaine de la Santé: Advisory Committee on Information Processing in Material Research in the Field of Health) on 26 November 2014 and an authorization from the CNIL (Commission Nationale de l’Informatique et des Libertés: National Commission on Computer Technology and Freedom) on 24 December 2015 concerning the TITAN study. These gave Bayer the possibility to collect, analyse and use anonymised data of patients included in this study. All patients provided written informed consent to participate.
Consent for publication
Not applicable.
Competing interests
SR: Consulting fees from Allergan, Bayer and Novartis; LK: Financial support from Novartis, Allergan, Bayer, Thea and Alcon; consulting fees from Alcon, Alimera, Allergan, Bayer, Roche and Novartis; research funding from Alcon, Alimera, Allergan, Bayer, Horus, Novartis and Thea Pharmaceuticals; AGA: Consulting fees from Alimera, Allergan, Bayer and Novartis; ID: Employee of Bayer HealthCare SAS; ES: Consulting fees and financial support from Allergan, Bayer, Novartis, Roche and Thea Pharmaceuticals. | AFLIBERCEPT | DrugsGivenReaction | CC BY | 33596867 | 19,020,429 | 2021-02-17 |
What was the administration route of drug 'AFLIBERCEPT'? | Efficacy and safety of intravitreal aflibercept in ranibizumab-refractory patients with neovascular age-related macular degeneration.
BACKGROUND
Anti-vascular endothelial growth factor (anti-VEGF) agents have become the standard of care in neovascular age-related macular degeneration (nAMD). Despite generally excellent response rates to anti-VEGF therapy, some patients do not respond or may respond suboptimally. In the case of refractory or rapidly recurring fluid in nAMD, clinicians may switch to another anti-VEGF agent. TITAN was an observational study that assessed the effectiveness and safety of intravitreal aflibercept (IVT-AFL) in patients with nAMD refractory to ranibizumab who switched to IVT-AFL after less than 12 months of ranibizumab treatment in routine clinical practice in France.
METHODS
TITAN was an observational, retrospective and prospective 12-month study conducted at 28 centres in France. Patients with nAMD refractory to ranibizumab were enrolled. Patients who were switched from ranibizumab to IVT-AFL were followed for 12 months. Data were obtained from medical records for retrospectively included patients, and at routine follow-up visits for those included prospectively. The main outcome measure was percentage of patients who achieved treatment success (gain of ≥1 Early Treatment Diabetic Retinopathy Study letters in best-corrected visual acuity [BCVA] and/or any reduction in central retinal thickness [CRT]) from baseline to 12 months after switching. A sample size of 225 patients was determined based on a 2-sided 95% confidence interval with a width equal to 0.12 when the sample proportion was 0.70.
RESULTS
We analysed safety data (N = 217) and clinical outcomes from patients in the per-protocol population (n = 125). The mean (standard deviation) number of IVT-AFL injections was 7.5 (2.6). Treatment success was achieved in 68.8% of patients. Mean BCVA change from baseline to Month 12 was + 1.5 letters (P = 0.105) and the mean CRT change was - 45.0 μm (P < 0.001). In a subgroup analysis, in patients who received three initial monthly IVT-AFL injections, mean BCVA gain was 3.3 letters at Month 12 (P = 0.015). Excluding lack of efficacy and inappropriate scheduling of drug administration, the most common adverse event was eye pain (2.3%).
CONCLUSIONS
Switching ranibizumab-refractory patients with nAMD to IVT-AFL may improve visual outcomes in some patients, particularly those who receive three initial monthly injections.
BACKGROUND
ClinicalTrials.gov , NCT02321241 . First posted: December 22, 2014; Last update posted: July 2, 2018.
Background
Neovascular age-related macular degeneration (nAMD) is the most severe form of AMD and is the most common cause of legal blindness [1, 2]. Anti–vascular endothelial growth factor (anti-VEGF) agents, including intravitreal aflibercept (IVT-AFL) and ranibizumab, have become standard of care in nAMD [3–8]. Despite generally excellent response rates to anti-VEGF therapy, some patients do not respond or may respond suboptimally [3–8]. Some studies have shown reduced anatomic response over time with ranibizumab treatment in patients with nAMD, and there have been reports of loss of bioefficacy after repeated ranibizumab treatment [9–11].
In the case of refractory or rapidly recurring fluid in nAMD, clinicians may switch from the current anti-VEGF agent to another anti-VEGF agent [12]. The VIEW 1 and VIEW 2 studies [13] assessed the efficacy and safety of IVT-AFL in patients with nAMD and demonstrated non-inferiority of IVT-AFL 2 mg, given every 8 weeks after three initial monthly doses, versus ranibizumab 0.5 mg, every 4 weeks, in maintaining vision (loss of < 15 Early Treatment Diabetic Retinopathy [ETDRS] letters in best-corrected visual acuity [BCVA]) in treatment-naïve patients over a 12-month period [14–17].
Retrospective studies have shown that individualised IVT-AFL treatment can significantly reduce retinal fluid and preserve vision in patients with nAMD who are resistant to anti-VEGF agents [17–20]. However, prospective studies examining a switch to IVT-AFL in patients refractory to ranibizumab treatment in a real-world setting are scarce [21].
TITAN was an observational study that assessed the effectiveness and safety of IVT-AFL in patients with nAMD refractory to ranibizumab (persistence of intraretinal [IRF] and/or subretinal fluid [SRF]) who switched to IVT-AFL after less than 12 months of ranibizumab treatment in routine clinical practice in France.
Methods
Study design
TITAN (NCT02321241) was an observational 12-month study to assess the effectiveness and safety of IVT-AFL in patients with nAMD refractory to ranibizumab. The study was conducted in 28 centres in France and enrolled patients both retrospectively and prospectively. Data were analysed from patients who received IVT-AFL treatment between January 1, 2014 and December 31, 2015. The date of the first IVT-AFL injection was considered the baseline visit, and data collection continued for a maximum of 12 months after the first injection. For retrospectively enrolled patients, the number of office visits, eye exams and treatments were collected from medical records; for prospectively enrolled patients, this information was recorded at routine follow-up visits. The study protocol was approved by a French data privacy committee (Comité Consultatif sur le Traitement de l’Information en Matière de Recherche dans le Domaine de la Santé and Commission Nationale de l’Informatique et des Libertés). All patients provided written informed consent to participate.
Participants
Patients diagnosed with nAMD who had been treated with ranibizumab for > 3 but < 12 months and switched to prescribed IVT-AFL by their physician were included. Eligible patients must have been refractory to ranibizumab, defined as persistent IRF and/or SRF on optical coherence tomography despite treatment with ranibizumab, in accordance with the Haute Autorité de Santé recommendations of at least three injections of ranibizumab.
Patients excluded from the study were those with any ranibizumab-treated eyes for nAMD previously switched to IVT-AFL, absence of treatment criteria for IVT-AFL, eyes previously treated with photodynamic therapy, another retinal disease (diabetic retinopathy, diabetic macular oedema, myopia or angioid streaks) or participation in any interventional study.
The safety analysis set (SAS) included all patients who received ≥1 IVT-AFL treatment in any eye. The per-protocol (PP) population was defined as all patients in the full analysis set (FAS) (i.e., BCVA ≥35 letters at baseline, or delay of ≤380 days between first and last ranibizumab injection, or ≥ 3 ranibizumab injections), with BCVA and central retinal thickness (CRT) assessments prior to any treatment (including ranibizumab), at baseline, and at Month 12. The FAS included patients who received ≥1 IVT-AFL treatment and had BCVA and CRT assessments in the study eye at baseline and during follow-up. Patients from the FAS who received a BCVA/CRT assessment before any treatment (including ranibizumab) at enrolment and at Month 12 were included in the subgroup analysis. This subset of patients was stratified by whether or not they received three initial monthly IVT-AFL injections following the switch.
Outcomes
The primary endpoint was treatment success rate at 12 months (defined as a gain of ≥1 letter in BCVA and/or any decrease in CRT [in μm] between initial visit [first injection of IVT-AFL] and 12-month follow-up visit). BCVA was measured using ETDRS letters (preferentially) or any other visual scale. For data analysis, we transformed any other visual acuity score to ETDRS letter score.
Secondary outcomes included change in BCVA between baseline and final study visits (12 months after the first injection of IVT-AFL or study discontinuation), mean duration of ranibizumab treatment before initiation of IVT-AFL, and frequency and mean number of IVT-AFL injections over the study period. No BCVA data were collected at the end of ranibizumab treatment; however, switching to IVT-AFL occurred relatively soon after the last ranibizumab injection (median of 44.0 days). Therefore, the change in BCVA during ranibizumab treatment was estimated based on the change between BCVA values before any treatment and before first injection of IVT-AFL.
All adverse events (AEs) occurring after the first IVT-AFL injection were documented in the electronic case report form. All patient medical records were evaluated for demographic as well as clinical characteristics, and AEs were summarised using the Medical Dictionary for Regulatory Activities coding system. The event rates for single AEs were calculated based on the total number of documented patients and AEs were categorised according to connection with medication, seriousness, discontinuation of therapy and outcome.
Statistical analyses
A sample size of 225 patients was determined based on a 2-sided 95% confidence interval (CI) with a width equal to 0.12 when the sample proportion was 0.70. Primary analysis criteria (success rate) considered patients who discontinued IVT-AFL treatment prematurely to be treatment failures. We expressed the incidence of treatment success at 12 months as number and percentage of patients (n [%]) and provided a 2-sided 95% CI.
The main analysis of secondary criteria (BCVA and CRT) was performed without replacing missing values at study end. Change in BCVA (ETDRS letters) was expressed as mean (standard deviation [SD]) and provided a 2-sided 95% CI. We performed two sensitivity analyses with two imputation methods for missing data: imputation of missing value by the last observation carried forward (LOCF) method and imputation of missing data with the median value of population. Statistical analyses were conducted with SAS software release 9.4 (SAS Institute Inc., Cary, NC, USA).
Results
Participants
A total of 236 patients were screened. Of these, 217 were included in the SAS and 125 were included in the PP population (Fig. 1). Demographic characteristics are shown in Table 1.
Fig. 1 Patient disposition. *Three initial monthly injections (− 1/+ 2 weeks). The per-protocol population was defined as patients without protocol deviation and with BCVA/CRT assessments prior to any treatment (ranibizumab) at baseline and at Month 12 (n = 125). The FAS included patients who received ≥1 IVT-AFL treatment and had BCVA and CRT assessments in the study eye at baseline and during follow-up (n = 185). BCVA = best-corrected visual acuity; CRT = central retinal thickness; FAS = full analysis set; IVT-AFL = intravitreal aflibercept; SAS = safety analysis set
Table 1 Demographic Characteristics
Full Analysis Set
(n = 185) Per-Protocol Set
(n = 125)
Characteristic at baseline
Age, years 77.6 (8.5) 78.2 (8.0)
Female, n (%) 109 (58.9) 79 (63.2)
Duration of nAMD, months 8.5 (7.3) 7.4 (3.3)
Characteristic before any treatment, including ranibizumab
BCVA, ETDRS letters 62.4 (16.6)a 64.0 (15.6)
CRT, μm 368.3 (124.4)b 362.4 (115.2)c
Subretinal fluid, n (%) 133 (76.9)b 92 (74.2)c
Intraretinal fluid, n (%) 76 (43.9)b 55 (44.4)c
Pigment epithelium detachment, n (%) 125 (72.3)b 88 (71.0)c
Mean (standard deviation) unless otherwise stated
an = 163; bn = 161; cn = 116; values correspond to assessments before any treatment, including ranibizumab, had been administered
BCVA Best-corrected visual acuity, CRT Central retinal thickness, ETDRS Early Treatment Diabetic Retinopathy Study, nAMD Neovascular age-related macular degeneration
Ranibizumab treatment and outcomes prior to switching
Treatment with ranibizumab was initiated soon after diagnosis of nAMD, with a mean lapse of < 1 month (Table 2). Approximately 75% of patients had their first injection of ranibizumab < 0.3 months after diagnosis of nAMD. Patients received a mean (SD) of 5.1 (2.5) injections over 5.6 (4.3) months [22]. Approximately 50% of patients received ≥4 ranibizumab injections over a median duration of 4.6 months. Mean (SD) BCVA improved significantly with ranibizumab treatment prior to the switch (+ 2.2 [12.0] ETDRS letters, P = 0.046) (Table 2). Distribution of patients with SRF, IRF and subretinal pigment epithelium (sub-RPE) at baseline according to the absence or presence of SRF, IRF and sub-RPE before treatment (including ranibizumab), is shown in Fig. 2.
Table 2 Ranibizumab Treatment and Outcomes Before Switch to Intravitreal Aflibercept
Patients
(n = 125)
Delay between diagnosis of nAMD and first injection of ranibizumab, months 0.5 (1.2)
Number of ranibizumab injections 4.8 (1.9)
Duration of ranibizumab treatment, months 4.9 (2.8)
BCVA before any treatment (including ranibizumab), ETDRS letters 64.0 (15.6)
BCVA at baseline (switch to IVT-AFL), ETDRS letters 66.2 (12.1)
Change in BCVA from before the start of any treatment (including ranibizumab) and baseline (switch to IVT-AFL) 2.2 (12.0)
P value 0.046
All values are expressed as mean (standard deviation); per-protocol population
P value is for paired sample t test
BCVA Best-corrected visual acuity, ETDRS Early Treatment Diabetic Retinopathy Study, IVT-AFL Intravitreal aflibercept, nAMD Neovascular age-related macular degeneration
Fig. 2 Proportion of patients with SRF, IRF and sub-RPE. At baseline (prior to switch) (a) and at Month 12 (following switch) (b). IRF = intraretinal fluid; SRF = subretinal fluid; sub-RPE = sub-retinal pigment epithelium. Any treatment includes ranibizumab. *Twelve patients without SRF at baseline were found with SRF at least once over the follow-up and three of them switched from IVT-AFL to ranibizumab. †Twenty-three patients without IRF at baseline were found with IRF at least once over the follow-up and one of them switched from IVT-AFL to ranibizumab. ‡Thirteen patients without sub-RPE fluid at baseline were found with sub-RPE fluid at least once over the follow-up and two of them switched from IVT-AFL to ranibizumab
Intravitreal aflibercept treatment and outcomes following the switch
The success rate (proportion of patients with a gain of ≥1 letter in BCVA and/or any decrease in CRT [in μm] between baseline [the date of the first IVT-AFL injection] and 12-month follow-up visit) was 68.8%. The mean BCVA improvement from baseline to Month 12 was 1.5 letters (P = 0.105) (Table 3). At Month 12, 55.2% of patients (n = 69/125) were able to read ≥70 letters.
Table 3 Intravitreal Aflibercept Treatment and Outcomes Following Switch to Intravitreal Aflibercept
Patients
(n = 125)
Delay between last injection of ranibizumab and first IVT-AFL, days 61.2 (46.1)
Reasons for switch to IVT-AFL, n (%)
Refractory 122 (97.6%)
AE/SAE 3 (2.4%)
Other 0 (0%)
Time between diagnosis of nAMD and first IVT-AFL, months 7.4 (3.3)
Duration of IVT-AFL treatment, months 11.3 (3.1)
Follow-up duration, months 12.7 (2.0)
Number of IVT-AFL injections 7.5 (2.6)
BCVA, ETDRS letters
Baseline (switch to IVT-AFL) 66.2 (12.1)
Month 12 67.7 (13.6)
Change in BCVA score between baseline and 12 months 1.5 (10.3)
P value 0.105
CRT, μm
Baseline (switch to IVT-AFL) 331.2 (103.3)
Month 12 286.2 (84.7)
Change in CRT between baseline and 12 months −45.0 (101.1)
P value < 0.001
Patients with SRF,
n (%)
Baseline 87 (70.7)
Month 12 58 (48.3)
Patients with IRF,
n (%)
Baseline 44 (35.8)
Month 12 27 (22.5)
All values are reported as mean (standard deviation) unless otherwise indicated; per-protocol population
P values are for the paired samples t test
AE Adverse event, BCVA Best-corrected visual acuity, CRT Central retinal thickness, ETDRS Early Treatment Diabetic Retinopathy Study, IRF Intraretinal fluid;IVT-AFL Intravitreal aflibercept, nAMD Neovascular age-related macular degeneration, SAE Serious adverse event, SRF Subretinal fluid
Data on the timing of first injection, treatment duration and number of IVT-AFL injections are shown in Table 3. Given the presence of extreme values, the median data is presented instead of the mean. The delay between the last injection of ranibizumab and the first IVT-AFL injection ranged from 9 to 314 days (mean [SD], 61.2 [46.1]; median 43 days). The main reason for switching from ranibizumab to IVT-AFL was that the patient was considered refractory to ranibizumab.
The range of IVT-AFL treatments received over the 12-month study duration is shown in Fig. 3. More than half of patients (52.8%) received three initial monthly IVT-AFL injections. Overall, 17.6% (n = 22) of patients switched at least once from IVT-AFL to ranibizumab, and 5.6% (n = 7) switched back to IVT-AFL.
Fig. 3 Distribution of number of intravitreal aflibercept injections over the 12-month study. Per-protocol population. IVT-AFL = intravitreal aflibercept
Anatomic outcomes following the switch
Anatomic outcomes improved in patients who were refractory to ranibizumab and switched to IVT-AFL (Table 3). Mean CRT reduction from baseline to Month 12 was 45.0 μm (P < 0.001). The proportion of patients with SRF at baseline was 70.7% (n = 87/123) and at Month 12 was 48.3% (n = 58/120). The proportion of patients with IRF was 35.8% at baseline (n = 44/123) and 22.5% (n = 27/120) at Month 12, respectively. Differences between the proportions of patients with and without fluid at baseline and Month 12 were significant for SRF and IRF (McNemar test P < 0.001 and P = 0.014, respectively), but not sub-RPE. Distribution of patients with SRF, IRF and sub-RPE at Month 12 according to the absence or presence of SRF, IRF and sub-RPE at baseline are shown in Fig. 2.
Approximately one-quarter of patients (24.0%; n = 30/125) gained 0 to 4 letters; 16.8% (n = 21/125) gained 5 to 9 letters; 10.4% (n = 13/125) gained 10 to 14 letters; and 11.2% (n = 14/125) gained ≥15 letters from baseline to Month 12. Among patients gaining ≥15 letters, mean BCVA change was 20.6 (8.0) letters. Conversely, 8.0% (n = 10/125) lost ≥15 letters. Figure 4 shows the distribution of patients by final absolute BCVA subgroups (< 50; 50 to 55; 55 to 70; and ≥ 70 letters).
Fig. 4 Distribution of patients by absolute final BCVA subgroups (< 50, 50–55, 55–70 and ≥ 70 letters). *n = 125. Per-protocol population. BCVA = best–corrected visual acuity
Exploratory and subgroup analyses following switch
In an exploratory analysis conducted in the PP population, patients with < 3 months of ranibizumab treatment prior to baseline gained a mean (SD) of 4.0 (11.1) letters from baseline to Month 12 (P = 0.025). This gain was greater than that observed in patients with 3 to 5 months, 6 to 8 months, and ≥ 9 months of ranibizumab treatment prior to baseline (+ 0.1 [10.8], − 0.6 [8.2] and + 2.1 [7.8] letters, respectively) (Supplementary Fig. 1).
In a subgroup analysis in the FAS population (patients with BCVA/CRT assessment before any treatment [including ranibizumab], at baseline and at Month 12), the overall success rate was 66.7%, with higher rates in patients who received three initial monthly IVT-AFL injections compared with those who did not (69.0% vs 64.1%, respectively; Table 4). No statistical test was performed, but 95% CIs largely overlap.
Table 4 Success Rate by Use or Non-use of 3 Initial Monthly Injections (FAS Population)
Overall
(n = 135) With 3 initial monthly injections
(n = 71) Without 3 initial monthly injections
(n = 64)
Success rate at 12 months,a
n (%) [95% CI]
90 (66.7)
[58.0–74.5]
49 (69.0)
[56.9–79.5]
41 (64.1)
[51.1–75.7]
BCVA, ETDRS letters
Baseline 65.1 (13.7) 63.6 (12.3) 66.9 (15.1)
Month 12 66.9 (15.0) 66.8 (12.1) 66.9 (17.8)
Change in BCVA between initial visit and 12 months 1.7 (10.5) 3.3 (11.0) 0.0 (9.6)
P value 0.056 0.015 NS
CRT, μm
Baseline 328.0 (100.9) 338.6 (114.6) 316.3 (82.5)
Month 12 285.3 (84.3) 296.1 (88.0) 272.6 (78.6)
Change in CRT between initial visit and 12 months −47.0 (104.1) −45.5 (106.0) −48.8 (102.6)
P value < 0.001 < 0.001 < 0.001
All values are reported as mean (standard deviation) unless otherwise indicated
P values are for the paired samples t test
aPatients who gained ≥1 ETDRS letter (BCVA) and/or a reduction in CRT from baseline (prior to switch) to 12 months after the first intravitreal aflibercept injection
BCVA Best–corrected visual acuity, CI Confidence interval, CRT Central retinal thickness, ETDRS Early Treatment Diabetic Retinopathy Study, FAS Full analysis set, NS Not significant
At Month 12, in patients who received three initial monthly IVT-AFL injections, mean BCVA gain was 3.3 letters (P = 0.015). Mean BCVA was stable in patients who did not receive three initial monthly injections (Table 4). There was a trend in favour of higher BCVA gains among patients with lower initial BCVA.
Sensitivity analyses
Results of the sensitivity analyses were consistent with the results of the main analyses. Specifically, the success rate in the overall population was 68.8% when we replaced missing data by the median of the population, or by LOCF. Mean (SD) gain in BCVA was 1.5 (10.3) letters (P = 0.105) when missing data were replaced by the median of the population, or by LOCF. At Month 12, the proportion of patients who could read ≥70 letters was 55.2% for median value replacement and for LOCF value replacement.
Safety outcomes
Excluding lack of efficacy and inappropriate scheduling of drug administration, which were also considered AEs according to the protocol and occurred in 12.4 and 5.1% of patients, respectively, the most common AE was eye pain (2.3%) (Table 5). Treatment-emergent serious AEs were reported in 1.8% of patients, none associated with IVT-AFL treatment. The most common non-ocular AE was falls, occurring in 1.4% of patients. Twenty-seven patients discontinued IVT-AFL due to lack of efficacy and a further seven discontinued IVT-AFL due to other treatment-emergent AEs; in two patients, the AEs (both PED, one with retinal exudate) were considered related to IVT-AFL.
Discussion
Results from the TITAN study indicate that approximately two-thirds (67.7%) of patients with nAMD who switched from ranibizumab to IVT-AFL treatment (after ≤12 months of treatment with ranibizumab) achieved success, defined as a gain of ≥1 letter in BCVA and/or any decrease in CRT between baseline and Month 12. Patients considered refractory to first-line treatment with ranibizumab had an additional gain (beyond the mean 2.2 letters initially achieved with ranibizumab) of 1.5 letters at Month 12 after IVT-AFL treatment was initiated, indicating that switching from ranibizumab to IVT-AFL can be considered in ranibizumab-refractory nAMD. There was an inverse relationship between baseline BCVA and visual acuity gain at Month 12. Generally, patients with lower baseline BCVA had greater visual acuity gains at Month 12. Patients who had previously received ranibizumab for < 3 months prior to switch had a greater improvement in BCVA. In addition to decreases in CRT after switching treatment, an enhanced anatomic response was observed, with reductions in fluid accumulation observed in the SRF, IRF and sub-RPE compartments. Overall, these results suggest that considering a switch to IVT-AFL from ranibizumab after < 3 months may be warranted; however, these findings should be interpreted with caution as the TITAN study was not specifically designed to answer this question, and other confounding variables such as duration of disease are likely to have had an effect.
Findings from other studies of patients refractory to ranibizumab, who switched to IVT-AFL after varying durations of treatment, suggest that there may be a benefit to switching, particularly with anatomical benefit after the switch in terms of improvements in central retinal thickness and pigment epithelium detachment [23]. In fact, the effect of switching on functional outcomes has been shown to be variable [23] and, in our analysis, the overall mean change in BCVA was relatively small. However, over 20% of the ranibizumab-refractory patients achieved BCVA gains of ≥10 letters after receiving IVT-AFL, suggesting there may be specific subgroups of patients who could respond particularly well to switching to IVT-AFL, despite an initially poor response to anti-VEGF therapy with ranibizumab. Further randomised, controlled studies with appropriate control groups are, however, required.
The present study highlights the importance of three initial monthly injections at the start of IVT-AFL treatment. Greater visual improvements were observed in patients who received three initial monthly injections than in those who did not, which is consistent with previous studies of patients with nAMD who were switched from ranibizumab to IVT-AFL [21, 24]. It is notable that the proportion of patients who received three initial monthly injections in TITAN was lower (52.8%) than expected given the recommended dosing regimen in France [25]. Findings from this study were consistent with the known safety profile of IVT-AFL in nAMD [13, 26].
A few limitations inherent in the observational study design should be noted, including the use of different charts to evaluate visual acuity. Here, ETDRS letter charts or any other visual scale were used to evaluate visual acuity; if the latter, results were converted to ETDRS letters, potentially introducing a bias, especially when measuring the number of letters gained or lost after treatment. Also, our study design includes both retrospective and prospective components, and lacks a control group, which is a fundamental aspect of an observational study. In addition, findings are from a single European country, which may not be representative of other countries. Some data were imputed due to variability in data collection; a feature common in observational studies. Given that success was not achieved if patients discontinued IVT-AFL prematurely, even if they switched back to IVT-AFL at a later point, the proportion of patients achieving success may have been underestimated. The success rate was 70.3% (FAS) using median data. Patients who switched from IVT-AFL prior to receiving 12 months of treatment were considered failures; therefore, replacement of missing data by median value of population only slightly affected the success rate. Finally, factors that could influence differences in visual and anatomic outcomes between groups (e.g., disease severity at baseline, frequency of injections and physician monitoring) were not explored in the TITAN study.
Conclusions
The TITAN study demonstrated the effectiveness and safety of IVT-AFL in patients with ranibizumab-refractory nAMD in routine clinical practice in France. These findings suggest that an early switch (< 12 months) from ranibizumab to IVT-AFL may improve visual outcomes over 12 months for some patients in this difficult-to-treat population. The study further highlighted that initiating IVT-AFL treatment with three initial monthly injections after switching from ranibizumab may improve visual outcomes. Further studies with appropriate control groups are required to understand how to best identify those patients most likely to benefit from switching to IVT-AFL.
Supplementary Information
Additional file 1: Supplementary Figure 1. Visual gains in patients treated with ranibizumab prior to switching to IVT-AFL. Per-protocol population. BCVA = best–corrected visual acuity; IVT-AFL, intravitreal aflibercept.
Abbreviations
AEAdverse event
anti-VEGFAnti-vascular endothelial growth factor
BCVABest-corrected visual acuity
CRTCentral retinal thickness
ETDRSEarly Treatment Diabetic Retinopathy
FASFull analysis set
IRFIntraretinal fluid
IVT-AFLIntravitreal aflibercept
LOCFLast observation carried forward
nAMDNeovascular age-related macular degeneration
PPPer-protocol
SASSafety analysis set
SDStandard deviation
SRFSubretinal fluid
sub-RPESubretinal pigment epithelium
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Acknowledgements
The authors wish to thank the TITAN study investigators: Dr. Aouizerate, Dr. Averous, Professor Baillif, Dr. Benouaich, Dr. Benyelles, Professor Berrod, Dr. Cantaloube-Bessière, Dr. Chahed, Professor Chiambaretta, Dr. Cohen, Dr. Coscas, Dr. De Bats, Dr. Deudon-Combe, Dr. Dominguez, Dr. Donati, Professor Dot, Dr. Dumas, Dr. Giocanti-Auregan, Dr. Guigui, Professor Kodjikian, Dr. Oubraham, Dr. Rothschild, Dr. Rumen, Dr. Sampo, Dr. Scholtès, Dr. Sibille-Dabadie, Professor Souied, Dr. Stanescu, Dr. Tran, Dr. Uzzan, Dr. Wolff and Dr. Zerbib.
Medical writing assistance was provided by Apothecom, UK, and was funded by Bayer Consumer Care AG, Basel, Switzerland.
Authors’ contributions
All authors have read and approved the manuscript. SR contributed to the design; data acquisition, analysis and interpretation; and preparation and final review of the manuscript. LK contributed to the design; data acquisition, analysis and interpretation; and preparation and final review of the manuscript. AGA contributed to the design; data acquisition, analysis and interpretation; and preparation and final review of the manuscript. ID contributed to the design; data acquisition, analysis and interpretation; and preparation and final review of the manuscript. ES contributed to data acquisition, analysis and interpretation; and preparation and final review of the manuscript.
Funding
The TITAN study was sponsored by Bayer HealthCare SAS, France. The sponsor participated in the design and conduct of the study, analysis of the data, and preparation of the manuscript.
Availability of data and materials
Availability of the data underlying this publication will be determined according to Bayer’s commitment to the EFPIA/PhRMA “Principles for responsible clinical trial data sharing”. This pertains to scope, time point and process of data access.
As such, Bayer commits to sharing, upon request from qualified scientific and medical researchers, patient-level clinical trial data, study-level clinical trial data, and protocols from clinical trials in patients for medicines and indications approved in the United States (US) and European Union (EU) as necessary for conducting legitimate research. This applies to data on new medicines and indications that have been approved by the EU and US regulatory agencies on or after January 1, 2014.
Interested researchers can use www.clinicalstudydatarequest.com to request access to anonymised patient-level data and supporting documents from clinical studies to conduct further research that can help advance medical science or improve patient care. Information on the Bayer criteria for listing studies and other relevant information is provided in the Study sponsors section of the portal.
Ethics approval and consent to participate
Bayer France received a positive statement from the CCTIRS (Comité Consultatif sur le Traitement de l’Information en Matière de Recherche dans le Domaine de la Santé: Advisory Committee on Information Processing in Material Research in the Field of Health) on 26 November 2014 and an authorization from the CNIL (Commission Nationale de l’Informatique et des Libertés: National Commission on Computer Technology and Freedom) on 24 December 2015 concerning the TITAN study. These gave Bayer the possibility to collect, analyse and use anonymised data of patients included in this study. All patients provided written informed consent to participate.
Consent for publication
Not applicable.
Competing interests
SR: Consulting fees from Allergan, Bayer and Novartis; LK: Financial support from Novartis, Allergan, Bayer, Thea and Alcon; consulting fees from Alcon, Alimera, Allergan, Bayer, Roche and Novartis; research funding from Alcon, Alimera, Allergan, Bayer, Horus, Novartis and Thea Pharmaceuticals; AGA: Consulting fees from Alimera, Allergan, Bayer and Novartis; ID: Employee of Bayer HealthCare SAS; ES: Consulting fees and financial support from Allergan, Bayer, Novartis, Roche and Thea Pharmaceuticals. | Intraocular | DrugAdministrationRoute | CC BY | 33596867 | 19,020,429 | 2021-02-17 |
What was the outcome of reaction 'Off label use'? | Efficacy and safety of intravitreal aflibercept in ranibizumab-refractory patients with neovascular age-related macular degeneration.
BACKGROUND
Anti-vascular endothelial growth factor (anti-VEGF) agents have become the standard of care in neovascular age-related macular degeneration (nAMD). Despite generally excellent response rates to anti-VEGF therapy, some patients do not respond or may respond suboptimally. In the case of refractory or rapidly recurring fluid in nAMD, clinicians may switch to another anti-VEGF agent. TITAN was an observational study that assessed the effectiveness and safety of intravitreal aflibercept (IVT-AFL) in patients with nAMD refractory to ranibizumab who switched to IVT-AFL after less than 12 months of ranibizumab treatment in routine clinical practice in France.
METHODS
TITAN was an observational, retrospective and prospective 12-month study conducted at 28 centres in France. Patients with nAMD refractory to ranibizumab were enrolled. Patients who were switched from ranibizumab to IVT-AFL were followed for 12 months. Data were obtained from medical records for retrospectively included patients, and at routine follow-up visits for those included prospectively. The main outcome measure was percentage of patients who achieved treatment success (gain of ≥1 Early Treatment Diabetic Retinopathy Study letters in best-corrected visual acuity [BCVA] and/or any reduction in central retinal thickness [CRT]) from baseline to 12 months after switching. A sample size of 225 patients was determined based on a 2-sided 95% confidence interval with a width equal to 0.12 when the sample proportion was 0.70.
RESULTS
We analysed safety data (N = 217) and clinical outcomes from patients in the per-protocol population (n = 125). The mean (standard deviation) number of IVT-AFL injections was 7.5 (2.6). Treatment success was achieved in 68.8% of patients. Mean BCVA change from baseline to Month 12 was + 1.5 letters (P = 0.105) and the mean CRT change was - 45.0 μm (P < 0.001). In a subgroup analysis, in patients who received three initial monthly IVT-AFL injections, mean BCVA gain was 3.3 letters at Month 12 (P = 0.015). Excluding lack of efficacy and inappropriate scheduling of drug administration, the most common adverse event was eye pain (2.3%).
CONCLUSIONS
Switching ranibizumab-refractory patients with nAMD to IVT-AFL may improve visual outcomes in some patients, particularly those who receive three initial monthly injections.
BACKGROUND
ClinicalTrials.gov , NCT02321241 . First posted: December 22, 2014; Last update posted: July 2, 2018.
Background
Neovascular age-related macular degeneration (nAMD) is the most severe form of AMD and is the most common cause of legal blindness [1, 2]. Anti–vascular endothelial growth factor (anti-VEGF) agents, including intravitreal aflibercept (IVT-AFL) and ranibizumab, have become standard of care in nAMD [3–8]. Despite generally excellent response rates to anti-VEGF therapy, some patients do not respond or may respond suboptimally [3–8]. Some studies have shown reduced anatomic response over time with ranibizumab treatment in patients with nAMD, and there have been reports of loss of bioefficacy after repeated ranibizumab treatment [9–11].
In the case of refractory or rapidly recurring fluid in nAMD, clinicians may switch from the current anti-VEGF agent to another anti-VEGF agent [12]. The VIEW 1 and VIEW 2 studies [13] assessed the efficacy and safety of IVT-AFL in patients with nAMD and demonstrated non-inferiority of IVT-AFL 2 mg, given every 8 weeks after three initial monthly doses, versus ranibizumab 0.5 mg, every 4 weeks, in maintaining vision (loss of < 15 Early Treatment Diabetic Retinopathy [ETDRS] letters in best-corrected visual acuity [BCVA]) in treatment-naïve patients over a 12-month period [14–17].
Retrospective studies have shown that individualised IVT-AFL treatment can significantly reduce retinal fluid and preserve vision in patients with nAMD who are resistant to anti-VEGF agents [17–20]. However, prospective studies examining a switch to IVT-AFL in patients refractory to ranibizumab treatment in a real-world setting are scarce [21].
TITAN was an observational study that assessed the effectiveness and safety of IVT-AFL in patients with nAMD refractory to ranibizumab (persistence of intraretinal [IRF] and/or subretinal fluid [SRF]) who switched to IVT-AFL after less than 12 months of ranibizumab treatment in routine clinical practice in France.
Methods
Study design
TITAN (NCT02321241) was an observational 12-month study to assess the effectiveness and safety of IVT-AFL in patients with nAMD refractory to ranibizumab. The study was conducted in 28 centres in France and enrolled patients both retrospectively and prospectively. Data were analysed from patients who received IVT-AFL treatment between January 1, 2014 and December 31, 2015. The date of the first IVT-AFL injection was considered the baseline visit, and data collection continued for a maximum of 12 months after the first injection. For retrospectively enrolled patients, the number of office visits, eye exams and treatments were collected from medical records; for prospectively enrolled patients, this information was recorded at routine follow-up visits. The study protocol was approved by a French data privacy committee (Comité Consultatif sur le Traitement de l’Information en Matière de Recherche dans le Domaine de la Santé and Commission Nationale de l’Informatique et des Libertés). All patients provided written informed consent to participate.
Participants
Patients diagnosed with nAMD who had been treated with ranibizumab for > 3 but < 12 months and switched to prescribed IVT-AFL by their physician were included. Eligible patients must have been refractory to ranibizumab, defined as persistent IRF and/or SRF on optical coherence tomography despite treatment with ranibizumab, in accordance with the Haute Autorité de Santé recommendations of at least three injections of ranibizumab.
Patients excluded from the study were those with any ranibizumab-treated eyes for nAMD previously switched to IVT-AFL, absence of treatment criteria for IVT-AFL, eyes previously treated with photodynamic therapy, another retinal disease (diabetic retinopathy, diabetic macular oedema, myopia or angioid streaks) or participation in any interventional study.
The safety analysis set (SAS) included all patients who received ≥1 IVT-AFL treatment in any eye. The per-protocol (PP) population was defined as all patients in the full analysis set (FAS) (i.e., BCVA ≥35 letters at baseline, or delay of ≤380 days between first and last ranibizumab injection, or ≥ 3 ranibizumab injections), with BCVA and central retinal thickness (CRT) assessments prior to any treatment (including ranibizumab), at baseline, and at Month 12. The FAS included patients who received ≥1 IVT-AFL treatment and had BCVA and CRT assessments in the study eye at baseline and during follow-up. Patients from the FAS who received a BCVA/CRT assessment before any treatment (including ranibizumab) at enrolment and at Month 12 were included in the subgroup analysis. This subset of patients was stratified by whether or not they received three initial monthly IVT-AFL injections following the switch.
Outcomes
The primary endpoint was treatment success rate at 12 months (defined as a gain of ≥1 letter in BCVA and/or any decrease in CRT [in μm] between initial visit [first injection of IVT-AFL] and 12-month follow-up visit). BCVA was measured using ETDRS letters (preferentially) or any other visual scale. For data analysis, we transformed any other visual acuity score to ETDRS letter score.
Secondary outcomes included change in BCVA between baseline and final study visits (12 months after the first injection of IVT-AFL or study discontinuation), mean duration of ranibizumab treatment before initiation of IVT-AFL, and frequency and mean number of IVT-AFL injections over the study period. No BCVA data were collected at the end of ranibizumab treatment; however, switching to IVT-AFL occurred relatively soon after the last ranibizumab injection (median of 44.0 days). Therefore, the change in BCVA during ranibizumab treatment was estimated based on the change between BCVA values before any treatment and before first injection of IVT-AFL.
All adverse events (AEs) occurring after the first IVT-AFL injection were documented in the electronic case report form. All patient medical records were evaluated for demographic as well as clinical characteristics, and AEs were summarised using the Medical Dictionary for Regulatory Activities coding system. The event rates for single AEs were calculated based on the total number of documented patients and AEs were categorised according to connection with medication, seriousness, discontinuation of therapy and outcome.
Statistical analyses
A sample size of 225 patients was determined based on a 2-sided 95% confidence interval (CI) with a width equal to 0.12 when the sample proportion was 0.70. Primary analysis criteria (success rate) considered patients who discontinued IVT-AFL treatment prematurely to be treatment failures. We expressed the incidence of treatment success at 12 months as number and percentage of patients (n [%]) and provided a 2-sided 95% CI.
The main analysis of secondary criteria (BCVA and CRT) was performed without replacing missing values at study end. Change in BCVA (ETDRS letters) was expressed as mean (standard deviation [SD]) and provided a 2-sided 95% CI. We performed two sensitivity analyses with two imputation methods for missing data: imputation of missing value by the last observation carried forward (LOCF) method and imputation of missing data with the median value of population. Statistical analyses were conducted with SAS software release 9.4 (SAS Institute Inc., Cary, NC, USA).
Results
Participants
A total of 236 patients were screened. Of these, 217 were included in the SAS and 125 were included in the PP population (Fig. 1). Demographic characteristics are shown in Table 1.
Fig. 1 Patient disposition. *Three initial monthly injections (− 1/+ 2 weeks). The per-protocol population was defined as patients without protocol deviation and with BCVA/CRT assessments prior to any treatment (ranibizumab) at baseline and at Month 12 (n = 125). The FAS included patients who received ≥1 IVT-AFL treatment and had BCVA and CRT assessments in the study eye at baseline and during follow-up (n = 185). BCVA = best-corrected visual acuity; CRT = central retinal thickness; FAS = full analysis set; IVT-AFL = intravitreal aflibercept; SAS = safety analysis set
Table 1 Demographic Characteristics
Full Analysis Set
(n = 185) Per-Protocol Set
(n = 125)
Characteristic at baseline
Age, years 77.6 (8.5) 78.2 (8.0)
Female, n (%) 109 (58.9) 79 (63.2)
Duration of nAMD, months 8.5 (7.3) 7.4 (3.3)
Characteristic before any treatment, including ranibizumab
BCVA, ETDRS letters 62.4 (16.6)a 64.0 (15.6)
CRT, μm 368.3 (124.4)b 362.4 (115.2)c
Subretinal fluid, n (%) 133 (76.9)b 92 (74.2)c
Intraretinal fluid, n (%) 76 (43.9)b 55 (44.4)c
Pigment epithelium detachment, n (%) 125 (72.3)b 88 (71.0)c
Mean (standard deviation) unless otherwise stated
an = 163; bn = 161; cn = 116; values correspond to assessments before any treatment, including ranibizumab, had been administered
BCVA Best-corrected visual acuity, CRT Central retinal thickness, ETDRS Early Treatment Diabetic Retinopathy Study, nAMD Neovascular age-related macular degeneration
Ranibizumab treatment and outcomes prior to switching
Treatment with ranibizumab was initiated soon after diagnosis of nAMD, with a mean lapse of < 1 month (Table 2). Approximately 75% of patients had their first injection of ranibizumab < 0.3 months after diagnosis of nAMD. Patients received a mean (SD) of 5.1 (2.5) injections over 5.6 (4.3) months [22]. Approximately 50% of patients received ≥4 ranibizumab injections over a median duration of 4.6 months. Mean (SD) BCVA improved significantly with ranibizumab treatment prior to the switch (+ 2.2 [12.0] ETDRS letters, P = 0.046) (Table 2). Distribution of patients with SRF, IRF and subretinal pigment epithelium (sub-RPE) at baseline according to the absence or presence of SRF, IRF and sub-RPE before treatment (including ranibizumab), is shown in Fig. 2.
Table 2 Ranibizumab Treatment and Outcomes Before Switch to Intravitreal Aflibercept
Patients
(n = 125)
Delay between diagnosis of nAMD and first injection of ranibizumab, months 0.5 (1.2)
Number of ranibizumab injections 4.8 (1.9)
Duration of ranibizumab treatment, months 4.9 (2.8)
BCVA before any treatment (including ranibizumab), ETDRS letters 64.0 (15.6)
BCVA at baseline (switch to IVT-AFL), ETDRS letters 66.2 (12.1)
Change in BCVA from before the start of any treatment (including ranibizumab) and baseline (switch to IVT-AFL) 2.2 (12.0)
P value 0.046
All values are expressed as mean (standard deviation); per-protocol population
P value is for paired sample t test
BCVA Best-corrected visual acuity, ETDRS Early Treatment Diabetic Retinopathy Study, IVT-AFL Intravitreal aflibercept, nAMD Neovascular age-related macular degeneration
Fig. 2 Proportion of patients with SRF, IRF and sub-RPE. At baseline (prior to switch) (a) and at Month 12 (following switch) (b). IRF = intraretinal fluid; SRF = subretinal fluid; sub-RPE = sub-retinal pigment epithelium. Any treatment includes ranibizumab. *Twelve patients without SRF at baseline were found with SRF at least once over the follow-up and three of them switched from IVT-AFL to ranibizumab. †Twenty-three patients without IRF at baseline were found with IRF at least once over the follow-up and one of them switched from IVT-AFL to ranibizumab. ‡Thirteen patients without sub-RPE fluid at baseline were found with sub-RPE fluid at least once over the follow-up and two of them switched from IVT-AFL to ranibizumab
Intravitreal aflibercept treatment and outcomes following the switch
The success rate (proportion of patients with a gain of ≥1 letter in BCVA and/or any decrease in CRT [in μm] between baseline [the date of the first IVT-AFL injection] and 12-month follow-up visit) was 68.8%. The mean BCVA improvement from baseline to Month 12 was 1.5 letters (P = 0.105) (Table 3). At Month 12, 55.2% of patients (n = 69/125) were able to read ≥70 letters.
Table 3 Intravitreal Aflibercept Treatment and Outcomes Following Switch to Intravitreal Aflibercept
Patients
(n = 125)
Delay between last injection of ranibizumab and first IVT-AFL, days 61.2 (46.1)
Reasons for switch to IVT-AFL, n (%)
Refractory 122 (97.6%)
AE/SAE 3 (2.4%)
Other 0 (0%)
Time between diagnosis of nAMD and first IVT-AFL, months 7.4 (3.3)
Duration of IVT-AFL treatment, months 11.3 (3.1)
Follow-up duration, months 12.7 (2.0)
Number of IVT-AFL injections 7.5 (2.6)
BCVA, ETDRS letters
Baseline (switch to IVT-AFL) 66.2 (12.1)
Month 12 67.7 (13.6)
Change in BCVA score between baseline and 12 months 1.5 (10.3)
P value 0.105
CRT, μm
Baseline (switch to IVT-AFL) 331.2 (103.3)
Month 12 286.2 (84.7)
Change in CRT between baseline and 12 months −45.0 (101.1)
P value < 0.001
Patients with SRF,
n (%)
Baseline 87 (70.7)
Month 12 58 (48.3)
Patients with IRF,
n (%)
Baseline 44 (35.8)
Month 12 27 (22.5)
All values are reported as mean (standard deviation) unless otherwise indicated; per-protocol population
P values are for the paired samples t test
AE Adverse event, BCVA Best-corrected visual acuity, CRT Central retinal thickness, ETDRS Early Treatment Diabetic Retinopathy Study, IRF Intraretinal fluid;IVT-AFL Intravitreal aflibercept, nAMD Neovascular age-related macular degeneration, SAE Serious adverse event, SRF Subretinal fluid
Data on the timing of first injection, treatment duration and number of IVT-AFL injections are shown in Table 3. Given the presence of extreme values, the median data is presented instead of the mean. The delay between the last injection of ranibizumab and the first IVT-AFL injection ranged from 9 to 314 days (mean [SD], 61.2 [46.1]; median 43 days). The main reason for switching from ranibizumab to IVT-AFL was that the patient was considered refractory to ranibizumab.
The range of IVT-AFL treatments received over the 12-month study duration is shown in Fig. 3. More than half of patients (52.8%) received three initial monthly IVT-AFL injections. Overall, 17.6% (n = 22) of patients switched at least once from IVT-AFL to ranibizumab, and 5.6% (n = 7) switched back to IVT-AFL.
Fig. 3 Distribution of number of intravitreal aflibercept injections over the 12-month study. Per-protocol population. IVT-AFL = intravitreal aflibercept
Anatomic outcomes following the switch
Anatomic outcomes improved in patients who were refractory to ranibizumab and switched to IVT-AFL (Table 3). Mean CRT reduction from baseline to Month 12 was 45.0 μm (P < 0.001). The proportion of patients with SRF at baseline was 70.7% (n = 87/123) and at Month 12 was 48.3% (n = 58/120). The proportion of patients with IRF was 35.8% at baseline (n = 44/123) and 22.5% (n = 27/120) at Month 12, respectively. Differences between the proportions of patients with and without fluid at baseline and Month 12 were significant for SRF and IRF (McNemar test P < 0.001 and P = 0.014, respectively), but not sub-RPE. Distribution of patients with SRF, IRF and sub-RPE at Month 12 according to the absence or presence of SRF, IRF and sub-RPE at baseline are shown in Fig. 2.
Approximately one-quarter of patients (24.0%; n = 30/125) gained 0 to 4 letters; 16.8% (n = 21/125) gained 5 to 9 letters; 10.4% (n = 13/125) gained 10 to 14 letters; and 11.2% (n = 14/125) gained ≥15 letters from baseline to Month 12. Among patients gaining ≥15 letters, mean BCVA change was 20.6 (8.0) letters. Conversely, 8.0% (n = 10/125) lost ≥15 letters. Figure 4 shows the distribution of patients by final absolute BCVA subgroups (< 50; 50 to 55; 55 to 70; and ≥ 70 letters).
Fig. 4 Distribution of patients by absolute final BCVA subgroups (< 50, 50–55, 55–70 and ≥ 70 letters). *n = 125. Per-protocol population. BCVA = best–corrected visual acuity
Exploratory and subgroup analyses following switch
In an exploratory analysis conducted in the PP population, patients with < 3 months of ranibizumab treatment prior to baseline gained a mean (SD) of 4.0 (11.1) letters from baseline to Month 12 (P = 0.025). This gain was greater than that observed in patients with 3 to 5 months, 6 to 8 months, and ≥ 9 months of ranibizumab treatment prior to baseline (+ 0.1 [10.8], − 0.6 [8.2] and + 2.1 [7.8] letters, respectively) (Supplementary Fig. 1).
In a subgroup analysis in the FAS population (patients with BCVA/CRT assessment before any treatment [including ranibizumab], at baseline and at Month 12), the overall success rate was 66.7%, with higher rates in patients who received three initial monthly IVT-AFL injections compared with those who did not (69.0% vs 64.1%, respectively; Table 4). No statistical test was performed, but 95% CIs largely overlap.
Table 4 Success Rate by Use or Non-use of 3 Initial Monthly Injections (FAS Population)
Overall
(n = 135) With 3 initial monthly injections
(n = 71) Without 3 initial monthly injections
(n = 64)
Success rate at 12 months,a
n (%) [95% CI]
90 (66.7)
[58.0–74.5]
49 (69.0)
[56.9–79.5]
41 (64.1)
[51.1–75.7]
BCVA, ETDRS letters
Baseline 65.1 (13.7) 63.6 (12.3) 66.9 (15.1)
Month 12 66.9 (15.0) 66.8 (12.1) 66.9 (17.8)
Change in BCVA between initial visit and 12 months 1.7 (10.5) 3.3 (11.0) 0.0 (9.6)
P value 0.056 0.015 NS
CRT, μm
Baseline 328.0 (100.9) 338.6 (114.6) 316.3 (82.5)
Month 12 285.3 (84.3) 296.1 (88.0) 272.6 (78.6)
Change in CRT between initial visit and 12 months −47.0 (104.1) −45.5 (106.0) −48.8 (102.6)
P value < 0.001 < 0.001 < 0.001
All values are reported as mean (standard deviation) unless otherwise indicated
P values are for the paired samples t test
aPatients who gained ≥1 ETDRS letter (BCVA) and/or a reduction in CRT from baseline (prior to switch) to 12 months after the first intravitreal aflibercept injection
BCVA Best–corrected visual acuity, CI Confidence interval, CRT Central retinal thickness, ETDRS Early Treatment Diabetic Retinopathy Study, FAS Full analysis set, NS Not significant
At Month 12, in patients who received three initial monthly IVT-AFL injections, mean BCVA gain was 3.3 letters (P = 0.015). Mean BCVA was stable in patients who did not receive three initial monthly injections (Table 4). There was a trend in favour of higher BCVA gains among patients with lower initial BCVA.
Sensitivity analyses
Results of the sensitivity analyses were consistent with the results of the main analyses. Specifically, the success rate in the overall population was 68.8% when we replaced missing data by the median of the population, or by LOCF. Mean (SD) gain in BCVA was 1.5 (10.3) letters (P = 0.105) when missing data were replaced by the median of the population, or by LOCF. At Month 12, the proportion of patients who could read ≥70 letters was 55.2% for median value replacement and for LOCF value replacement.
Safety outcomes
Excluding lack of efficacy and inappropriate scheduling of drug administration, which were also considered AEs according to the protocol and occurred in 12.4 and 5.1% of patients, respectively, the most common AE was eye pain (2.3%) (Table 5). Treatment-emergent serious AEs were reported in 1.8% of patients, none associated with IVT-AFL treatment. The most common non-ocular AE was falls, occurring in 1.4% of patients. Twenty-seven patients discontinued IVT-AFL due to lack of efficacy and a further seven discontinued IVT-AFL due to other treatment-emergent AEs; in two patients, the AEs (both PED, one with retinal exudate) were considered related to IVT-AFL.
Discussion
Results from the TITAN study indicate that approximately two-thirds (67.7%) of patients with nAMD who switched from ranibizumab to IVT-AFL treatment (after ≤12 months of treatment with ranibizumab) achieved success, defined as a gain of ≥1 letter in BCVA and/or any decrease in CRT between baseline and Month 12. Patients considered refractory to first-line treatment with ranibizumab had an additional gain (beyond the mean 2.2 letters initially achieved with ranibizumab) of 1.5 letters at Month 12 after IVT-AFL treatment was initiated, indicating that switching from ranibizumab to IVT-AFL can be considered in ranibizumab-refractory nAMD. There was an inverse relationship between baseline BCVA and visual acuity gain at Month 12. Generally, patients with lower baseline BCVA had greater visual acuity gains at Month 12. Patients who had previously received ranibizumab for < 3 months prior to switch had a greater improvement in BCVA. In addition to decreases in CRT after switching treatment, an enhanced anatomic response was observed, with reductions in fluid accumulation observed in the SRF, IRF and sub-RPE compartments. Overall, these results suggest that considering a switch to IVT-AFL from ranibizumab after < 3 months may be warranted; however, these findings should be interpreted with caution as the TITAN study was not specifically designed to answer this question, and other confounding variables such as duration of disease are likely to have had an effect.
Findings from other studies of patients refractory to ranibizumab, who switched to IVT-AFL after varying durations of treatment, suggest that there may be a benefit to switching, particularly with anatomical benefit after the switch in terms of improvements in central retinal thickness and pigment epithelium detachment [23]. In fact, the effect of switching on functional outcomes has been shown to be variable [23] and, in our analysis, the overall mean change in BCVA was relatively small. However, over 20% of the ranibizumab-refractory patients achieved BCVA gains of ≥10 letters after receiving IVT-AFL, suggesting there may be specific subgroups of patients who could respond particularly well to switching to IVT-AFL, despite an initially poor response to anti-VEGF therapy with ranibizumab. Further randomised, controlled studies with appropriate control groups are, however, required.
The present study highlights the importance of three initial monthly injections at the start of IVT-AFL treatment. Greater visual improvements were observed in patients who received three initial monthly injections than in those who did not, which is consistent with previous studies of patients with nAMD who were switched from ranibizumab to IVT-AFL [21, 24]. It is notable that the proportion of patients who received three initial monthly injections in TITAN was lower (52.8%) than expected given the recommended dosing regimen in France [25]. Findings from this study were consistent with the known safety profile of IVT-AFL in nAMD [13, 26].
A few limitations inherent in the observational study design should be noted, including the use of different charts to evaluate visual acuity. Here, ETDRS letter charts or any other visual scale were used to evaluate visual acuity; if the latter, results were converted to ETDRS letters, potentially introducing a bias, especially when measuring the number of letters gained or lost after treatment. Also, our study design includes both retrospective and prospective components, and lacks a control group, which is a fundamental aspect of an observational study. In addition, findings are from a single European country, which may not be representative of other countries. Some data were imputed due to variability in data collection; a feature common in observational studies. Given that success was not achieved if patients discontinued IVT-AFL prematurely, even if they switched back to IVT-AFL at a later point, the proportion of patients achieving success may have been underestimated. The success rate was 70.3% (FAS) using median data. Patients who switched from IVT-AFL prior to receiving 12 months of treatment were considered failures; therefore, replacement of missing data by median value of population only slightly affected the success rate. Finally, factors that could influence differences in visual and anatomic outcomes between groups (e.g., disease severity at baseline, frequency of injections and physician monitoring) were not explored in the TITAN study.
Conclusions
The TITAN study demonstrated the effectiveness and safety of IVT-AFL in patients with ranibizumab-refractory nAMD in routine clinical practice in France. These findings suggest that an early switch (< 12 months) from ranibizumab to IVT-AFL may improve visual outcomes over 12 months for some patients in this difficult-to-treat population. The study further highlighted that initiating IVT-AFL treatment with three initial monthly injections after switching from ranibizumab may improve visual outcomes. Further studies with appropriate control groups are required to understand how to best identify those patients most likely to benefit from switching to IVT-AFL.
Supplementary Information
Additional file 1: Supplementary Figure 1. Visual gains in patients treated with ranibizumab prior to switching to IVT-AFL. Per-protocol population. BCVA = best–corrected visual acuity; IVT-AFL, intravitreal aflibercept.
Abbreviations
AEAdverse event
anti-VEGFAnti-vascular endothelial growth factor
BCVABest-corrected visual acuity
CRTCentral retinal thickness
ETDRSEarly Treatment Diabetic Retinopathy
FASFull analysis set
IRFIntraretinal fluid
IVT-AFLIntravitreal aflibercept
LOCFLast observation carried forward
nAMDNeovascular age-related macular degeneration
PPPer-protocol
SASSafety analysis set
SDStandard deviation
SRFSubretinal fluid
sub-RPESubretinal pigment epithelium
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Acknowledgements
The authors wish to thank the TITAN study investigators: Dr. Aouizerate, Dr. Averous, Professor Baillif, Dr. Benouaich, Dr. Benyelles, Professor Berrod, Dr. Cantaloube-Bessière, Dr. Chahed, Professor Chiambaretta, Dr. Cohen, Dr. Coscas, Dr. De Bats, Dr. Deudon-Combe, Dr. Dominguez, Dr. Donati, Professor Dot, Dr. Dumas, Dr. Giocanti-Auregan, Dr. Guigui, Professor Kodjikian, Dr. Oubraham, Dr. Rothschild, Dr. Rumen, Dr. Sampo, Dr. Scholtès, Dr. Sibille-Dabadie, Professor Souied, Dr. Stanescu, Dr. Tran, Dr. Uzzan, Dr. Wolff and Dr. Zerbib.
Medical writing assistance was provided by Apothecom, UK, and was funded by Bayer Consumer Care AG, Basel, Switzerland.
Authors’ contributions
All authors have read and approved the manuscript. SR contributed to the design; data acquisition, analysis and interpretation; and preparation and final review of the manuscript. LK contributed to the design; data acquisition, analysis and interpretation; and preparation and final review of the manuscript. AGA contributed to the design; data acquisition, analysis and interpretation; and preparation and final review of the manuscript. ID contributed to the design; data acquisition, analysis and interpretation; and preparation and final review of the manuscript. ES contributed to data acquisition, analysis and interpretation; and preparation and final review of the manuscript.
Funding
The TITAN study was sponsored by Bayer HealthCare SAS, France. The sponsor participated in the design and conduct of the study, analysis of the data, and preparation of the manuscript.
Availability of data and materials
Availability of the data underlying this publication will be determined according to Bayer’s commitment to the EFPIA/PhRMA “Principles for responsible clinical trial data sharing”. This pertains to scope, time point and process of data access.
As such, Bayer commits to sharing, upon request from qualified scientific and medical researchers, patient-level clinical trial data, study-level clinical trial data, and protocols from clinical trials in patients for medicines and indications approved in the United States (US) and European Union (EU) as necessary for conducting legitimate research. This applies to data on new medicines and indications that have been approved by the EU and US regulatory agencies on or after January 1, 2014.
Interested researchers can use www.clinicalstudydatarequest.com to request access to anonymised patient-level data and supporting documents from clinical studies to conduct further research that can help advance medical science or improve patient care. Information on the Bayer criteria for listing studies and other relevant information is provided in the Study sponsors section of the portal.
Ethics approval and consent to participate
Bayer France received a positive statement from the CCTIRS (Comité Consultatif sur le Traitement de l’Information en Matière de Recherche dans le Domaine de la Santé: Advisory Committee on Information Processing in Material Research in the Field of Health) on 26 November 2014 and an authorization from the CNIL (Commission Nationale de l’Informatique et des Libertés: National Commission on Computer Technology and Freedom) on 24 December 2015 concerning the TITAN study. These gave Bayer the possibility to collect, analyse and use anonymised data of patients included in this study. All patients provided written informed consent to participate.
Consent for publication
Not applicable.
Competing interests
SR: Consulting fees from Allergan, Bayer and Novartis; LK: Financial support from Novartis, Allergan, Bayer, Thea and Alcon; consulting fees from Alcon, Alimera, Allergan, Bayer, Roche and Novartis; research funding from Alcon, Alimera, Allergan, Bayer, Horus, Novartis and Thea Pharmaceuticals; AGA: Consulting fees from Alimera, Allergan, Bayer and Novartis; ID: Employee of Bayer HealthCare SAS; ES: Consulting fees and financial support from Allergan, Bayer, Novartis, Roche and Thea Pharmaceuticals. | Not recovered | ReactionOutcome | CC BY | 33596867 | 19,020,429 | 2021-02-17 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Blood bilirubin increased'. | Hemochromatosis, alcoholism and unhealthy dietary fat: a case report.
BACKGROUND
Hereditary hemochromatosis is an autosomal recessive disorder where the clinical phenotype of skin pigmentation and organ damage occurs only in homozygotes. Simple heterozygotes, that is, just C282Y, typically do not develop iron overload. Here we present a case where a simple heterozygote in combination with alcoholism developed high ferritin and high transferrin saturation levels indicative of iron overload. Though alcoholism alone could explain her presentation, we hypothesize that an inflammatory cocktail of iron and alcohol probably caused our patient to succumb to acute liver failure at a very young age.
METHODS
A 29-year-old Caucasian woman presented to the hospital with progressively worsening yellowish discoloration of her eyes and skin associated with anorexia, nausea, vomiting, diffuse abdominal discomfort, increasing abdominal girth, dark urine and pale stools for about 2 weeks. Family history was significant for hereditary hemochromatosis. Her father was a simple heterozygote and her grandmother was homozygous for C282Y. Physical examination showed scleral icterus, distended abdomen with hepatomegaly and mild generalized tenderness. Lab test results showed an elevated white blood cell count, ferritin 539 ng/dL, transferrin saturation 58.23%, elevated liver enzymes, elevated international normalized ratio (INR), low albumin, Alcoholic Liver Disease/Nonalcoholic Fatty Liver Disease (ALD/NAFLD) Index (ANI) of 2.6, suggesting a 93.2% probability of alcoholic liver disease, and phosphatidyl ethanol level of 537ng/ml. Genetic testing showed that the patient was heterozygous for human homeostatic iron regulator protein (HFE) C282Y mutation and the normal allele. Computed tomography (CT) of the abdomen revealed hepatomegaly, portal hypertension and generalized anasarca. Magnetic resonance cholangiopancreatography (MRCP) showed negative results for bile duct pathology. Workup for other causes of liver disease was negative. A diagnosis of acute alcoholic hepatitis was made, with Maddrey's discriminant function of > 32, so prednisolone was started. Her bilirubin and INR continued to increase despite steroids, and the patient unfortunately died.
CONCLUSIONS
Our case highlights the importance of considering hemochromatosis in the differential diagnosis of young patients presenting with liver failure, including cases suggestive of alcoholism as the likely etiology. Larger studies are needed to investigate the role of non-iron factors like alcohol and viral hepatitis in the progression of liver disease in simple heterozygotes with hereditary hemochromatosis, given the high prevalence of this mutation in persons of Northern European descent.
Background
Hereditary hemochromatosis (HH) (Fig. 1) is an autosomal recessive disorder where the clinical phenotype of skin pigmentation and organ damage occurs only in homozygotes; even in homozygotes, the phenotype has a broad spectrum depending on sex and penetrance which is age-related [1, 2]. With regard to heterozygotes, it is mostly the compound heterozygotes—C282Y and the H63D or S65C variant allele—that develop iron overload [3, 4]. Simple heterozygotes—that is, just C282Y—almost never develop iron overload or organ damage [5]. Although the role of non-iron-related factors like alcohol in modulating the iron threshold required to induce liver damage is well known, the strength of their association in each of the HH phenotypes remains an area that is largely unexplored.Fig. 1 HFE gene mutations causing Hereditary Hemochromatosis
We present a case where a simple heterozygote with alcoholism developed high ferritin and high transferrin saturation indicative of iron overload. This is very rare considering the young age, female sex and the genotype of the patient. The iron overload coupled with probable unhealthy dietary habits in the setting of alcoholism (more fat, less essential nutrients as reported in studies) [6] resulted in an inflammatory cocktail and caused our patient to succumb to acute liver failure at a young age.
Case presentation
A 29-year-old Caucasian woman presented to the hospital with 2 weeks of progressively worsening yellowish discoloration of her eyes and skin associated with anorexia, nausea, vomiting, diffuse abdominal discomfort, increasing abdominal girth, dark urine and pale stools. Past medical history was significant for prior episodes of hospitalization for acute alcoholic intoxication including an episode a few months prior. Imaging at that time showed hepatic steatosis but no features suggestive of hepatic cirrhosis or portal hypertension. Family history was significant for hereditary hemochromatosis. The patient’s father was heterozygous for C282Y and the paternal grandmother was homozygous for C282Y. The patient reported drinking about 1–2 glasses of wine every day, and denied smoking and illicit drug use. Vitals signs were as follows: pulse rate 94 beats per minute, respiratory rate 20 per minute, blood pressure 112/78 mmHg, temperature 36.9 °C and oxygen saturation 100% on room air. Physical examination showed scleral icterus, distended abdomen with hepatomegaly and mild generalized tenderness.
Laboratory results were as follows (Table 1): white blood cell count 31,600/μL, hemoglobin 10.1 g/dL, platelets 172/μL, ferritin 539 ng/dL, transferrin saturation 58.23%, peripheral blood smear showing stomatocytosis (Fig. 2), total bilirubin 8.7 mg/dL, direct bilirubin 7.4 mg/dL, aspartate aminotransferase (AST) 90 U/L, alanine aminotransferase (ALT) 30 U/L, alkaline phosphatase (ALKP) 420 U/L, prothrombin time (PT) 18.6 seconds, international normalized ratio (INR) 1.5, blood urea nitrogen 1.0 mg/dL, serum creatinine 0.5 mg/dL, Alcoholic Liver Disease/Nonalcoholic Fatty Liver Disease (ALD/NAFLD) Index (ANI) 2.6, suggesting a 93.2% probability of alcoholic liver disease, and phosphatidyl ethanol level 537 ng/ml (levels > 20 ng/ml indicate moderate–heavy ethanol consumption). Genetic testing results revealed that the patient was heterozygous for the HFE C282Y mutation and the normal allele, as well as negative for H63D and S65C. Imaging showed features suggestive of parenchymal liver disease, portal hypertension and generalized anasarca (Fig. 3). Magnetic resonance cholangiopancreatography (MRCP) was negative for bile duct pathology. Workup for other causes of liver disease including autoimmune hepatitis, Wilson’s disease, alpha-1 antitrypsin deficiency, celiac disease, primary biliary cirrhosis (PBC), Epstein–Barr virus (EBV), cytomegalovirus (CMV), viral hepatitis, tick-borne illnesses and leptospirosis all returned negative results. Biopsy of the liver was considered but was held due to the patient’s worsening general medical condition.Table 1 Trend of laboratory values over the course of hospitalization
Laboratory test Day 1 Day 8 Day 16 Day 23 Reference range with units
WBC 31.6 50.5 84 48.6 4.5–10.0 10 × 3/μL
RBC 3.08 2.95 2.64 2.10 4.00–5.00 10 × 6/μL
Platelets 172 370 309 176 140–400 10 × /μL
Hemoglobin 10.1 9.6 8.7 7.0 12.0–16.0 g/dL
Hematocrit 27.9 29.3 27.2 20.3 39.0–54.0%
MCV 90.6 99.3 103 96.7 80–99 fl
MCH 32.8 32.5 33 33.3 27–34 pg
MCHC 36.2 32.8 32 34.5 32–36 g/dL
RDW 18.1 23.2 24.6 22.8 39.0–54.0%
INR 1.5 2.2 1.3 1.9 0.1–1.4
BUN 1 5 38 104 7–17 mg/dL
Creatinine 0.55 0.53 0.73 4.28 0.55–1.02 mg/dL
Total bilirubin 8.7 10.3 17.4 24.7 0.2–1.3 mg/dL
Direct bilirubin 7.4 11.3 0.0–0.2 mg/dL
AST 90 64 45 71 15.00–37.00 U/L
ALT 30 23 27 12 12–78 U/L
ALKP 420 383 245 168 46–116 U/L
WBC white blood cells, RBC red blood cells, MCV cytomegalovirus, MCHC mean corpuscular hemoglobin concentration, RDW red cell distribution width, BUN blood urea nitrogen, AST aspartate aminotransferase, ALT alanine aminotransferase, ALKP alkaline phosphatase
Fig. 2 Peripheral smear showing stomatocytosis (orange arrows pointing to the stomatocytes)
Fig. 3 Magnetic Resonance Cholangiopancreatography (MRCP) showing enlarged liver on the left side of the image (shown using yellow array of lines on the left), enlarged spleen on the right side of the image (shown using yellow array of lines on the right) and generalized anasarca (pointed using orange arrows)
The patient was diagnosed with acute alcoholic hepatitis, and Gastroenterology was consulted. The patient’s Maddrey’s discriminant function was 46.4 (a score > 32 indicates poor prognosis and that the patient might benefit from glucocorticoid therapy), so orally administered prednisolone 40 mg/day was started. Although the patient’s total bilirubin (TBIL) and INR initially improved after initiating steroidal therapy, a rebound increase was noted (Table 1), raising concerns for impending liver failure. Also, her creatinine increased from 0.5 to 4 mg/dL. Nephrology was consulted, and a diagnosis of hepatorenal syndrome type 1 was favored. In addition to worsening TBIL, INR and creatinine, the patient developed encephalopathy, succumbed to the disease and died.
Discussion
In this article we focus on the mechanisms of liver injury from the effects of iron, alcohol and dietary habits, and it may not be surprising to see that some of these mechanisms overlap.
Mechanism of liver injury with iron
Excess iron in the hepatocytes and Kupffer cells results in the Fenton reaction and reactive oxygen species production. The free radicals induce lipid peroxidation, which damages the mitochondria, resulting in release of cytochrome c and liver cell apoptosis. Iron overload also stimulates the production of proinflammatory and pro-fibrogenic cytokines including transforming growth factor beta (TGF-β). TGF-β leads to the activation of hepatic stellate cells and excess collagen production. Excess collagen and cross-linking coupled with iron inhibit activation of the liver progenitor cells required for the regeneration of liver cells, resulting in fibrosis [2, 7].
Mechanism of liver injury with alcohol
Alcohol is metabolized to acetaldehyde. Acetaldehyde results in the generation of reactive oxygen species, which causes lipid peroxidation and cell membrane and DNA damage. Damaged hepatocytes express antigens which are otherwise hidden from the immune system, resulting in immune stimulation. Chronically heightened immune activity results in immune exhaustion, overwhelming bacterial infection, multi-organ damage and death. Also, chronic alcohol abuse results in overgrowth of gut bacteria, and this along with alcohol-induced leaky gut results in increased delivery of endotoxins to the liver and liver damage [8].
Mechanism of liver injury with non-healthy dietary habits
Excess dietary fat increases insulin resistance and hyperinsulinemia, which leads to accumulation of fatty acids. Accumulated fatty acids result in the generation of lipotoxic species, hepatocellular oxidant stress and cell death. The dying hepatocytes release signals and express antigens which are otherwise hidden from the immune system. turning on the immunogenic and fibrogenesis cascade [9]. Individual susceptibility to fatty acid-induced oxidant stress depends on other factors including iron overload states such as HFE and alcoholism [2].
Thus, many of the mechanisms of liver injury from iron, alcohol and unhealthy dietary habits overlap (Fig. 4). Although non-HH factors like alcoholism, NAFLD and nonalcoholic steatohepatitis (NASH) are associated with hyperferritinemia from chronic inflammation, the patient’s elevated transferrin saturation [(serum iron/total iron binding capacity) × 100] can only be explained by her HH status. While heavy alcohol consumption alone could cause severe liver damage, we hypothesize that her HFE status and possible unhealthy dietary fat in the setting of alcoholism accelerated the progression of liver disease. Studies have shown that alcoholics consume a higher amount of fatty food and carbohydrates along with lower consumption of vegetables and dairy products, which could have a detrimental effect on health [6].Fig. 4 Mechanisms of liver injury from the effects of iron, alcohol and dietary habits
Clinicians must continually probe for factors like personal or family history of hemochromatosis, dietary habits and alcoholism using different strategies and reformatting questions. This is especially important because early recognition followed by referral to specialized centers for treatment for hereditary hemochromatosis and detoxification would be pivotal in the prognosis of these patients. Our patient persistently denied any unhealthy alcohol use until later in the disease course. This, coupled with her blood alcohol level of < 3 at admission, very high white blood cell counts, young age and female sex, pointed more towards other differentials like autoimmune hepatitis and infectious etiologies. Although we were fortunate enough to be redirected towards alcohol as the etiology from reports of stomatocytosis in the peripheral blood, high ANI and very high phosphatidyl ethanol level, the patient unfortunately succumbed to her acute liver failure.
Conclusion
Considering the high prevalence of HH and the rising mortality from alcoholic liver disease among young adults [10, 11], more studies exploring the role of alcohol in the development of liver damage in simple heterozygotes and vice versa are essential to determine whether all alcoholics have to be screened for hereditary hemochromatosis. This is because more than one factor may often be involved in the pathogenesis and progression of liver dysfunction.
Abbreviations
WBCWhite blood cells
TSTransferrin saturation [(serum iron/total iron binding capacity) × 100]
ALTAlanine transaminase
ASTAspartate transaminase
ALKPAlkaline phosphatase
PTProthrombin time
INRInternational normalized ratio
ALDAlcoholic liver disease
NAFLDNonalcoholic fatty liver disease
ANIALD/NAFLD Index
CTComputed tomography
MRCPMagnetic resonance cholangiopancreatography
HFEHuman homeostatic iron regulator protein
HHHereditary hemochromatosis
TGFTransforming growth factor
TBILTotal bilirubin
GIGastroenterology
NASHNonalcoholic steatohepatitis
PBCPrimary biliary cirrhosis
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Acknowledgements
Not applicable.
Authors' contributions
SV—first author, performed literature review, prepared the manuscript. AA—second author, contributed to figures and images in the manuscript. BB—pathologist for our case report. MK—senior author, edited the manuscript. All authors read and approved the final manuscript.
Funding
No funding was received or used for this case report.
Ethics approval and consent to participate
Nashville General Hospital Research Oversight Committee (ROC) has determined that a case report does not produce generalizable knowledge, nor is it an investigation of an FDA regulated production.
Consent for publication
Written informed consent was obtained from the patient for publication of this case report and any accompanying images. A copy of the written consent form is available for review by the Editor-in-Chief of this journal.
Availability of data and material
Our patient’s health record is available in our electronic medical record system, and information can be verified if required by the reviewer.
Competing interests
The authors declare that they have no competing interests. | PREDNISOLONE | DrugsGivenReaction | CC BY | 33596964 | 19,428,372 | 2021-02-17 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'International normalised ratio increased'. | Hemochromatosis, alcoholism and unhealthy dietary fat: a case report.
BACKGROUND
Hereditary hemochromatosis is an autosomal recessive disorder where the clinical phenotype of skin pigmentation and organ damage occurs only in homozygotes. Simple heterozygotes, that is, just C282Y, typically do not develop iron overload. Here we present a case where a simple heterozygote in combination with alcoholism developed high ferritin and high transferrin saturation levels indicative of iron overload. Though alcoholism alone could explain her presentation, we hypothesize that an inflammatory cocktail of iron and alcohol probably caused our patient to succumb to acute liver failure at a very young age.
METHODS
A 29-year-old Caucasian woman presented to the hospital with progressively worsening yellowish discoloration of her eyes and skin associated with anorexia, nausea, vomiting, diffuse abdominal discomfort, increasing abdominal girth, dark urine and pale stools for about 2 weeks. Family history was significant for hereditary hemochromatosis. Her father was a simple heterozygote and her grandmother was homozygous for C282Y. Physical examination showed scleral icterus, distended abdomen with hepatomegaly and mild generalized tenderness. Lab test results showed an elevated white blood cell count, ferritin 539 ng/dL, transferrin saturation 58.23%, elevated liver enzymes, elevated international normalized ratio (INR), low albumin, Alcoholic Liver Disease/Nonalcoholic Fatty Liver Disease (ALD/NAFLD) Index (ANI) of 2.6, suggesting a 93.2% probability of alcoholic liver disease, and phosphatidyl ethanol level of 537ng/ml. Genetic testing showed that the patient was heterozygous for human homeostatic iron regulator protein (HFE) C282Y mutation and the normal allele. Computed tomography (CT) of the abdomen revealed hepatomegaly, portal hypertension and generalized anasarca. Magnetic resonance cholangiopancreatography (MRCP) showed negative results for bile duct pathology. Workup for other causes of liver disease was negative. A diagnosis of acute alcoholic hepatitis was made, with Maddrey's discriminant function of > 32, so prednisolone was started. Her bilirubin and INR continued to increase despite steroids, and the patient unfortunately died.
CONCLUSIONS
Our case highlights the importance of considering hemochromatosis in the differential diagnosis of young patients presenting with liver failure, including cases suggestive of alcoholism as the likely etiology. Larger studies are needed to investigate the role of non-iron factors like alcohol and viral hepatitis in the progression of liver disease in simple heterozygotes with hereditary hemochromatosis, given the high prevalence of this mutation in persons of Northern European descent.
Background
Hereditary hemochromatosis (HH) (Fig. 1) is an autosomal recessive disorder where the clinical phenotype of skin pigmentation and organ damage occurs only in homozygotes; even in homozygotes, the phenotype has a broad spectrum depending on sex and penetrance which is age-related [1, 2]. With regard to heterozygotes, it is mostly the compound heterozygotes—C282Y and the H63D or S65C variant allele—that develop iron overload [3, 4]. Simple heterozygotes—that is, just C282Y—almost never develop iron overload or organ damage [5]. Although the role of non-iron-related factors like alcohol in modulating the iron threshold required to induce liver damage is well known, the strength of their association in each of the HH phenotypes remains an area that is largely unexplored.Fig. 1 HFE gene mutations causing Hereditary Hemochromatosis
We present a case where a simple heterozygote with alcoholism developed high ferritin and high transferrin saturation indicative of iron overload. This is very rare considering the young age, female sex and the genotype of the patient. The iron overload coupled with probable unhealthy dietary habits in the setting of alcoholism (more fat, less essential nutrients as reported in studies) [6] resulted in an inflammatory cocktail and caused our patient to succumb to acute liver failure at a young age.
Case presentation
A 29-year-old Caucasian woman presented to the hospital with 2 weeks of progressively worsening yellowish discoloration of her eyes and skin associated with anorexia, nausea, vomiting, diffuse abdominal discomfort, increasing abdominal girth, dark urine and pale stools. Past medical history was significant for prior episodes of hospitalization for acute alcoholic intoxication including an episode a few months prior. Imaging at that time showed hepatic steatosis but no features suggestive of hepatic cirrhosis or portal hypertension. Family history was significant for hereditary hemochromatosis. The patient’s father was heterozygous for C282Y and the paternal grandmother was homozygous for C282Y. The patient reported drinking about 1–2 glasses of wine every day, and denied smoking and illicit drug use. Vitals signs were as follows: pulse rate 94 beats per minute, respiratory rate 20 per minute, blood pressure 112/78 mmHg, temperature 36.9 °C and oxygen saturation 100% on room air. Physical examination showed scleral icterus, distended abdomen with hepatomegaly and mild generalized tenderness.
Laboratory results were as follows (Table 1): white blood cell count 31,600/μL, hemoglobin 10.1 g/dL, platelets 172/μL, ferritin 539 ng/dL, transferrin saturation 58.23%, peripheral blood smear showing stomatocytosis (Fig. 2), total bilirubin 8.7 mg/dL, direct bilirubin 7.4 mg/dL, aspartate aminotransferase (AST) 90 U/L, alanine aminotransferase (ALT) 30 U/L, alkaline phosphatase (ALKP) 420 U/L, prothrombin time (PT) 18.6 seconds, international normalized ratio (INR) 1.5, blood urea nitrogen 1.0 mg/dL, serum creatinine 0.5 mg/dL, Alcoholic Liver Disease/Nonalcoholic Fatty Liver Disease (ALD/NAFLD) Index (ANI) 2.6, suggesting a 93.2% probability of alcoholic liver disease, and phosphatidyl ethanol level 537 ng/ml (levels > 20 ng/ml indicate moderate–heavy ethanol consumption). Genetic testing results revealed that the patient was heterozygous for the HFE C282Y mutation and the normal allele, as well as negative for H63D and S65C. Imaging showed features suggestive of parenchymal liver disease, portal hypertension and generalized anasarca (Fig. 3). Magnetic resonance cholangiopancreatography (MRCP) was negative for bile duct pathology. Workup for other causes of liver disease including autoimmune hepatitis, Wilson’s disease, alpha-1 antitrypsin deficiency, celiac disease, primary biliary cirrhosis (PBC), Epstein–Barr virus (EBV), cytomegalovirus (CMV), viral hepatitis, tick-borne illnesses and leptospirosis all returned negative results. Biopsy of the liver was considered but was held due to the patient’s worsening general medical condition.Table 1 Trend of laboratory values over the course of hospitalization
Laboratory test Day 1 Day 8 Day 16 Day 23 Reference range with units
WBC 31.6 50.5 84 48.6 4.5–10.0 10 × 3/μL
RBC 3.08 2.95 2.64 2.10 4.00–5.00 10 × 6/μL
Platelets 172 370 309 176 140–400 10 × /μL
Hemoglobin 10.1 9.6 8.7 7.0 12.0–16.0 g/dL
Hematocrit 27.9 29.3 27.2 20.3 39.0–54.0%
MCV 90.6 99.3 103 96.7 80–99 fl
MCH 32.8 32.5 33 33.3 27–34 pg
MCHC 36.2 32.8 32 34.5 32–36 g/dL
RDW 18.1 23.2 24.6 22.8 39.0–54.0%
INR 1.5 2.2 1.3 1.9 0.1–1.4
BUN 1 5 38 104 7–17 mg/dL
Creatinine 0.55 0.53 0.73 4.28 0.55–1.02 mg/dL
Total bilirubin 8.7 10.3 17.4 24.7 0.2–1.3 mg/dL
Direct bilirubin 7.4 11.3 0.0–0.2 mg/dL
AST 90 64 45 71 15.00–37.00 U/L
ALT 30 23 27 12 12–78 U/L
ALKP 420 383 245 168 46–116 U/L
WBC white blood cells, RBC red blood cells, MCV cytomegalovirus, MCHC mean corpuscular hemoglobin concentration, RDW red cell distribution width, BUN blood urea nitrogen, AST aspartate aminotransferase, ALT alanine aminotransferase, ALKP alkaline phosphatase
Fig. 2 Peripheral smear showing stomatocytosis (orange arrows pointing to the stomatocytes)
Fig. 3 Magnetic Resonance Cholangiopancreatography (MRCP) showing enlarged liver on the left side of the image (shown using yellow array of lines on the left), enlarged spleen on the right side of the image (shown using yellow array of lines on the right) and generalized anasarca (pointed using orange arrows)
The patient was diagnosed with acute alcoholic hepatitis, and Gastroenterology was consulted. The patient’s Maddrey’s discriminant function was 46.4 (a score > 32 indicates poor prognosis and that the patient might benefit from glucocorticoid therapy), so orally administered prednisolone 40 mg/day was started. Although the patient’s total bilirubin (TBIL) and INR initially improved after initiating steroidal therapy, a rebound increase was noted (Table 1), raising concerns for impending liver failure. Also, her creatinine increased from 0.5 to 4 mg/dL. Nephrology was consulted, and a diagnosis of hepatorenal syndrome type 1 was favored. In addition to worsening TBIL, INR and creatinine, the patient developed encephalopathy, succumbed to the disease and died.
Discussion
In this article we focus on the mechanisms of liver injury from the effects of iron, alcohol and dietary habits, and it may not be surprising to see that some of these mechanisms overlap.
Mechanism of liver injury with iron
Excess iron in the hepatocytes and Kupffer cells results in the Fenton reaction and reactive oxygen species production. The free radicals induce lipid peroxidation, which damages the mitochondria, resulting in release of cytochrome c and liver cell apoptosis. Iron overload also stimulates the production of proinflammatory and pro-fibrogenic cytokines including transforming growth factor beta (TGF-β). TGF-β leads to the activation of hepatic stellate cells and excess collagen production. Excess collagen and cross-linking coupled with iron inhibit activation of the liver progenitor cells required for the regeneration of liver cells, resulting in fibrosis [2, 7].
Mechanism of liver injury with alcohol
Alcohol is metabolized to acetaldehyde. Acetaldehyde results in the generation of reactive oxygen species, which causes lipid peroxidation and cell membrane and DNA damage. Damaged hepatocytes express antigens which are otherwise hidden from the immune system, resulting in immune stimulation. Chronically heightened immune activity results in immune exhaustion, overwhelming bacterial infection, multi-organ damage and death. Also, chronic alcohol abuse results in overgrowth of gut bacteria, and this along with alcohol-induced leaky gut results in increased delivery of endotoxins to the liver and liver damage [8].
Mechanism of liver injury with non-healthy dietary habits
Excess dietary fat increases insulin resistance and hyperinsulinemia, which leads to accumulation of fatty acids. Accumulated fatty acids result in the generation of lipotoxic species, hepatocellular oxidant stress and cell death. The dying hepatocytes release signals and express antigens which are otherwise hidden from the immune system. turning on the immunogenic and fibrogenesis cascade [9]. Individual susceptibility to fatty acid-induced oxidant stress depends on other factors including iron overload states such as HFE and alcoholism [2].
Thus, many of the mechanisms of liver injury from iron, alcohol and unhealthy dietary habits overlap (Fig. 4). Although non-HH factors like alcoholism, NAFLD and nonalcoholic steatohepatitis (NASH) are associated with hyperferritinemia from chronic inflammation, the patient’s elevated transferrin saturation [(serum iron/total iron binding capacity) × 100] can only be explained by her HH status. While heavy alcohol consumption alone could cause severe liver damage, we hypothesize that her HFE status and possible unhealthy dietary fat in the setting of alcoholism accelerated the progression of liver disease. Studies have shown that alcoholics consume a higher amount of fatty food and carbohydrates along with lower consumption of vegetables and dairy products, which could have a detrimental effect on health [6].Fig. 4 Mechanisms of liver injury from the effects of iron, alcohol and dietary habits
Clinicians must continually probe for factors like personal or family history of hemochromatosis, dietary habits and alcoholism using different strategies and reformatting questions. This is especially important because early recognition followed by referral to specialized centers for treatment for hereditary hemochromatosis and detoxification would be pivotal in the prognosis of these patients. Our patient persistently denied any unhealthy alcohol use until later in the disease course. This, coupled with her blood alcohol level of < 3 at admission, very high white blood cell counts, young age and female sex, pointed more towards other differentials like autoimmune hepatitis and infectious etiologies. Although we were fortunate enough to be redirected towards alcohol as the etiology from reports of stomatocytosis in the peripheral blood, high ANI and very high phosphatidyl ethanol level, the patient unfortunately succumbed to her acute liver failure.
Conclusion
Considering the high prevalence of HH and the rising mortality from alcoholic liver disease among young adults [10, 11], more studies exploring the role of alcohol in the development of liver damage in simple heterozygotes and vice versa are essential to determine whether all alcoholics have to be screened for hereditary hemochromatosis. This is because more than one factor may often be involved in the pathogenesis and progression of liver dysfunction.
Abbreviations
WBCWhite blood cells
TSTransferrin saturation [(serum iron/total iron binding capacity) × 100]
ALTAlanine transaminase
ASTAspartate transaminase
ALKPAlkaline phosphatase
PTProthrombin time
INRInternational normalized ratio
ALDAlcoholic liver disease
NAFLDNonalcoholic fatty liver disease
ANIALD/NAFLD Index
CTComputed tomography
MRCPMagnetic resonance cholangiopancreatography
HFEHuman homeostatic iron regulator protein
HHHereditary hemochromatosis
TGFTransforming growth factor
TBILTotal bilirubin
GIGastroenterology
NASHNonalcoholic steatohepatitis
PBCPrimary biliary cirrhosis
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Acknowledgements
Not applicable.
Authors' contributions
SV—first author, performed literature review, prepared the manuscript. AA—second author, contributed to figures and images in the manuscript. BB—pathologist for our case report. MK—senior author, edited the manuscript. All authors read and approved the final manuscript.
Funding
No funding was received or used for this case report.
Ethics approval and consent to participate
Nashville General Hospital Research Oversight Committee (ROC) has determined that a case report does not produce generalizable knowledge, nor is it an investigation of an FDA regulated production.
Consent for publication
Written informed consent was obtained from the patient for publication of this case report and any accompanying images. A copy of the written consent form is available for review by the Editor-in-Chief of this journal.
Availability of data and material
Our patient’s health record is available in our electronic medical record system, and information can be verified if required by the reviewer.
Competing interests
The authors declare that they have no competing interests. | PREDNISOLONE | DrugsGivenReaction | CC BY | 33596964 | 19,428,372 | 2021-02-17 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Rebound effect'. | Hemochromatosis, alcoholism and unhealthy dietary fat: a case report.
BACKGROUND
Hereditary hemochromatosis is an autosomal recessive disorder where the clinical phenotype of skin pigmentation and organ damage occurs only in homozygotes. Simple heterozygotes, that is, just C282Y, typically do not develop iron overload. Here we present a case where a simple heterozygote in combination with alcoholism developed high ferritin and high transferrin saturation levels indicative of iron overload. Though alcoholism alone could explain her presentation, we hypothesize that an inflammatory cocktail of iron and alcohol probably caused our patient to succumb to acute liver failure at a very young age.
METHODS
A 29-year-old Caucasian woman presented to the hospital with progressively worsening yellowish discoloration of her eyes and skin associated with anorexia, nausea, vomiting, diffuse abdominal discomfort, increasing abdominal girth, dark urine and pale stools for about 2 weeks. Family history was significant for hereditary hemochromatosis. Her father was a simple heterozygote and her grandmother was homozygous for C282Y. Physical examination showed scleral icterus, distended abdomen with hepatomegaly and mild generalized tenderness. Lab test results showed an elevated white blood cell count, ferritin 539 ng/dL, transferrin saturation 58.23%, elevated liver enzymes, elevated international normalized ratio (INR), low albumin, Alcoholic Liver Disease/Nonalcoholic Fatty Liver Disease (ALD/NAFLD) Index (ANI) of 2.6, suggesting a 93.2% probability of alcoholic liver disease, and phosphatidyl ethanol level of 537ng/ml. Genetic testing showed that the patient was heterozygous for human homeostatic iron regulator protein (HFE) C282Y mutation and the normal allele. Computed tomography (CT) of the abdomen revealed hepatomegaly, portal hypertension and generalized anasarca. Magnetic resonance cholangiopancreatography (MRCP) showed negative results for bile duct pathology. Workup for other causes of liver disease was negative. A diagnosis of acute alcoholic hepatitis was made, with Maddrey's discriminant function of > 32, so prednisolone was started. Her bilirubin and INR continued to increase despite steroids, and the patient unfortunately died.
CONCLUSIONS
Our case highlights the importance of considering hemochromatosis in the differential diagnosis of young patients presenting with liver failure, including cases suggestive of alcoholism as the likely etiology. Larger studies are needed to investigate the role of non-iron factors like alcohol and viral hepatitis in the progression of liver disease in simple heterozygotes with hereditary hemochromatosis, given the high prevalence of this mutation in persons of Northern European descent.
Background
Hereditary hemochromatosis (HH) (Fig. 1) is an autosomal recessive disorder where the clinical phenotype of skin pigmentation and organ damage occurs only in homozygotes; even in homozygotes, the phenotype has a broad spectrum depending on sex and penetrance which is age-related [1, 2]. With regard to heterozygotes, it is mostly the compound heterozygotes—C282Y and the H63D or S65C variant allele—that develop iron overload [3, 4]. Simple heterozygotes—that is, just C282Y—almost never develop iron overload or organ damage [5]. Although the role of non-iron-related factors like alcohol in modulating the iron threshold required to induce liver damage is well known, the strength of their association in each of the HH phenotypes remains an area that is largely unexplored.Fig. 1 HFE gene mutations causing Hereditary Hemochromatosis
We present a case where a simple heterozygote with alcoholism developed high ferritin and high transferrin saturation indicative of iron overload. This is very rare considering the young age, female sex and the genotype of the patient. The iron overload coupled with probable unhealthy dietary habits in the setting of alcoholism (more fat, less essential nutrients as reported in studies) [6] resulted in an inflammatory cocktail and caused our patient to succumb to acute liver failure at a young age.
Case presentation
A 29-year-old Caucasian woman presented to the hospital with 2 weeks of progressively worsening yellowish discoloration of her eyes and skin associated with anorexia, nausea, vomiting, diffuse abdominal discomfort, increasing abdominal girth, dark urine and pale stools. Past medical history was significant for prior episodes of hospitalization for acute alcoholic intoxication including an episode a few months prior. Imaging at that time showed hepatic steatosis but no features suggestive of hepatic cirrhosis or portal hypertension. Family history was significant for hereditary hemochromatosis. The patient’s father was heterozygous for C282Y and the paternal grandmother was homozygous for C282Y. The patient reported drinking about 1–2 glasses of wine every day, and denied smoking and illicit drug use. Vitals signs were as follows: pulse rate 94 beats per minute, respiratory rate 20 per minute, blood pressure 112/78 mmHg, temperature 36.9 °C and oxygen saturation 100% on room air. Physical examination showed scleral icterus, distended abdomen with hepatomegaly and mild generalized tenderness.
Laboratory results were as follows (Table 1): white blood cell count 31,600/μL, hemoglobin 10.1 g/dL, platelets 172/μL, ferritin 539 ng/dL, transferrin saturation 58.23%, peripheral blood smear showing stomatocytosis (Fig. 2), total bilirubin 8.7 mg/dL, direct bilirubin 7.4 mg/dL, aspartate aminotransferase (AST) 90 U/L, alanine aminotransferase (ALT) 30 U/L, alkaline phosphatase (ALKP) 420 U/L, prothrombin time (PT) 18.6 seconds, international normalized ratio (INR) 1.5, blood urea nitrogen 1.0 mg/dL, serum creatinine 0.5 mg/dL, Alcoholic Liver Disease/Nonalcoholic Fatty Liver Disease (ALD/NAFLD) Index (ANI) 2.6, suggesting a 93.2% probability of alcoholic liver disease, and phosphatidyl ethanol level 537 ng/ml (levels > 20 ng/ml indicate moderate–heavy ethanol consumption). Genetic testing results revealed that the patient was heterozygous for the HFE C282Y mutation and the normal allele, as well as negative for H63D and S65C. Imaging showed features suggestive of parenchymal liver disease, portal hypertension and generalized anasarca (Fig. 3). Magnetic resonance cholangiopancreatography (MRCP) was negative for bile duct pathology. Workup for other causes of liver disease including autoimmune hepatitis, Wilson’s disease, alpha-1 antitrypsin deficiency, celiac disease, primary biliary cirrhosis (PBC), Epstein–Barr virus (EBV), cytomegalovirus (CMV), viral hepatitis, tick-borne illnesses and leptospirosis all returned negative results. Biopsy of the liver was considered but was held due to the patient’s worsening general medical condition.Table 1 Trend of laboratory values over the course of hospitalization
Laboratory test Day 1 Day 8 Day 16 Day 23 Reference range with units
WBC 31.6 50.5 84 48.6 4.5–10.0 10 × 3/μL
RBC 3.08 2.95 2.64 2.10 4.00–5.00 10 × 6/μL
Platelets 172 370 309 176 140–400 10 × /μL
Hemoglobin 10.1 9.6 8.7 7.0 12.0–16.0 g/dL
Hematocrit 27.9 29.3 27.2 20.3 39.0–54.0%
MCV 90.6 99.3 103 96.7 80–99 fl
MCH 32.8 32.5 33 33.3 27–34 pg
MCHC 36.2 32.8 32 34.5 32–36 g/dL
RDW 18.1 23.2 24.6 22.8 39.0–54.0%
INR 1.5 2.2 1.3 1.9 0.1–1.4
BUN 1 5 38 104 7–17 mg/dL
Creatinine 0.55 0.53 0.73 4.28 0.55–1.02 mg/dL
Total bilirubin 8.7 10.3 17.4 24.7 0.2–1.3 mg/dL
Direct bilirubin 7.4 11.3 0.0–0.2 mg/dL
AST 90 64 45 71 15.00–37.00 U/L
ALT 30 23 27 12 12–78 U/L
ALKP 420 383 245 168 46–116 U/L
WBC white blood cells, RBC red blood cells, MCV cytomegalovirus, MCHC mean corpuscular hemoglobin concentration, RDW red cell distribution width, BUN blood urea nitrogen, AST aspartate aminotransferase, ALT alanine aminotransferase, ALKP alkaline phosphatase
Fig. 2 Peripheral smear showing stomatocytosis (orange arrows pointing to the stomatocytes)
Fig. 3 Magnetic Resonance Cholangiopancreatography (MRCP) showing enlarged liver on the left side of the image (shown using yellow array of lines on the left), enlarged spleen on the right side of the image (shown using yellow array of lines on the right) and generalized anasarca (pointed using orange arrows)
The patient was diagnosed with acute alcoholic hepatitis, and Gastroenterology was consulted. The patient’s Maddrey’s discriminant function was 46.4 (a score > 32 indicates poor prognosis and that the patient might benefit from glucocorticoid therapy), so orally administered prednisolone 40 mg/day was started. Although the patient’s total bilirubin (TBIL) and INR initially improved after initiating steroidal therapy, a rebound increase was noted (Table 1), raising concerns for impending liver failure. Also, her creatinine increased from 0.5 to 4 mg/dL. Nephrology was consulted, and a diagnosis of hepatorenal syndrome type 1 was favored. In addition to worsening TBIL, INR and creatinine, the patient developed encephalopathy, succumbed to the disease and died.
Discussion
In this article we focus on the mechanisms of liver injury from the effects of iron, alcohol and dietary habits, and it may not be surprising to see that some of these mechanisms overlap.
Mechanism of liver injury with iron
Excess iron in the hepatocytes and Kupffer cells results in the Fenton reaction and reactive oxygen species production. The free radicals induce lipid peroxidation, which damages the mitochondria, resulting in release of cytochrome c and liver cell apoptosis. Iron overload also stimulates the production of proinflammatory and pro-fibrogenic cytokines including transforming growth factor beta (TGF-β). TGF-β leads to the activation of hepatic stellate cells and excess collagen production. Excess collagen and cross-linking coupled with iron inhibit activation of the liver progenitor cells required for the regeneration of liver cells, resulting in fibrosis [2, 7].
Mechanism of liver injury with alcohol
Alcohol is metabolized to acetaldehyde. Acetaldehyde results in the generation of reactive oxygen species, which causes lipid peroxidation and cell membrane and DNA damage. Damaged hepatocytes express antigens which are otherwise hidden from the immune system, resulting in immune stimulation. Chronically heightened immune activity results in immune exhaustion, overwhelming bacterial infection, multi-organ damage and death. Also, chronic alcohol abuse results in overgrowth of gut bacteria, and this along with alcohol-induced leaky gut results in increased delivery of endotoxins to the liver and liver damage [8].
Mechanism of liver injury with non-healthy dietary habits
Excess dietary fat increases insulin resistance and hyperinsulinemia, which leads to accumulation of fatty acids. Accumulated fatty acids result in the generation of lipotoxic species, hepatocellular oxidant stress and cell death. The dying hepatocytes release signals and express antigens which are otherwise hidden from the immune system. turning on the immunogenic and fibrogenesis cascade [9]. Individual susceptibility to fatty acid-induced oxidant stress depends on other factors including iron overload states such as HFE and alcoholism [2].
Thus, many of the mechanisms of liver injury from iron, alcohol and unhealthy dietary habits overlap (Fig. 4). Although non-HH factors like alcoholism, NAFLD and nonalcoholic steatohepatitis (NASH) are associated with hyperferritinemia from chronic inflammation, the patient’s elevated transferrin saturation [(serum iron/total iron binding capacity) × 100] can only be explained by her HH status. While heavy alcohol consumption alone could cause severe liver damage, we hypothesize that her HFE status and possible unhealthy dietary fat in the setting of alcoholism accelerated the progression of liver disease. Studies have shown that alcoholics consume a higher amount of fatty food and carbohydrates along with lower consumption of vegetables and dairy products, which could have a detrimental effect on health [6].Fig. 4 Mechanisms of liver injury from the effects of iron, alcohol and dietary habits
Clinicians must continually probe for factors like personal or family history of hemochromatosis, dietary habits and alcoholism using different strategies and reformatting questions. This is especially important because early recognition followed by referral to specialized centers for treatment for hereditary hemochromatosis and detoxification would be pivotal in the prognosis of these patients. Our patient persistently denied any unhealthy alcohol use until later in the disease course. This, coupled with her blood alcohol level of < 3 at admission, very high white blood cell counts, young age and female sex, pointed more towards other differentials like autoimmune hepatitis and infectious etiologies. Although we were fortunate enough to be redirected towards alcohol as the etiology from reports of stomatocytosis in the peripheral blood, high ANI and very high phosphatidyl ethanol level, the patient unfortunately succumbed to her acute liver failure.
Conclusion
Considering the high prevalence of HH and the rising mortality from alcoholic liver disease among young adults [10, 11], more studies exploring the role of alcohol in the development of liver damage in simple heterozygotes and vice versa are essential to determine whether all alcoholics have to be screened for hereditary hemochromatosis. This is because more than one factor may often be involved in the pathogenesis and progression of liver dysfunction.
Abbreviations
WBCWhite blood cells
TSTransferrin saturation [(serum iron/total iron binding capacity) × 100]
ALTAlanine transaminase
ASTAspartate transaminase
ALKPAlkaline phosphatase
PTProthrombin time
INRInternational normalized ratio
ALDAlcoholic liver disease
NAFLDNonalcoholic fatty liver disease
ANIALD/NAFLD Index
CTComputed tomography
MRCPMagnetic resonance cholangiopancreatography
HFEHuman homeostatic iron regulator protein
HHHereditary hemochromatosis
TGFTransforming growth factor
TBILTotal bilirubin
GIGastroenterology
NASHNonalcoholic steatohepatitis
PBCPrimary biliary cirrhosis
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Acknowledgements
Not applicable.
Authors' contributions
SV—first author, performed literature review, prepared the manuscript. AA—second author, contributed to figures and images in the manuscript. BB—pathologist for our case report. MK—senior author, edited the manuscript. All authors read and approved the final manuscript.
Funding
No funding was received or used for this case report.
Ethics approval and consent to participate
Nashville General Hospital Research Oversight Committee (ROC) has determined that a case report does not produce generalizable knowledge, nor is it an investigation of an FDA regulated production.
Consent for publication
Written informed consent was obtained from the patient for publication of this case report and any accompanying images. A copy of the written consent form is available for review by the Editor-in-Chief of this journal.
Availability of data and material
Our patient’s health record is available in our electronic medical record system, and information can be verified if required by the reviewer.
Competing interests
The authors declare that they have no competing interests. | PREDNISOLONE | DrugsGivenReaction | CC BY | 33596964 | 19,428,372 | 2021-02-17 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Therapy partial responder'. | Hemochromatosis, alcoholism and unhealthy dietary fat: a case report.
BACKGROUND
Hereditary hemochromatosis is an autosomal recessive disorder where the clinical phenotype of skin pigmentation and organ damage occurs only in homozygotes. Simple heterozygotes, that is, just C282Y, typically do not develop iron overload. Here we present a case where a simple heterozygote in combination with alcoholism developed high ferritin and high transferrin saturation levels indicative of iron overload. Though alcoholism alone could explain her presentation, we hypothesize that an inflammatory cocktail of iron and alcohol probably caused our patient to succumb to acute liver failure at a very young age.
METHODS
A 29-year-old Caucasian woman presented to the hospital with progressively worsening yellowish discoloration of her eyes and skin associated with anorexia, nausea, vomiting, diffuse abdominal discomfort, increasing abdominal girth, dark urine and pale stools for about 2 weeks. Family history was significant for hereditary hemochromatosis. Her father was a simple heterozygote and her grandmother was homozygous for C282Y. Physical examination showed scleral icterus, distended abdomen with hepatomegaly and mild generalized tenderness. Lab test results showed an elevated white blood cell count, ferritin 539 ng/dL, transferrin saturation 58.23%, elevated liver enzymes, elevated international normalized ratio (INR), low albumin, Alcoholic Liver Disease/Nonalcoholic Fatty Liver Disease (ALD/NAFLD) Index (ANI) of 2.6, suggesting a 93.2% probability of alcoholic liver disease, and phosphatidyl ethanol level of 537ng/ml. Genetic testing showed that the patient was heterozygous for human homeostatic iron regulator protein (HFE) C282Y mutation and the normal allele. Computed tomography (CT) of the abdomen revealed hepatomegaly, portal hypertension and generalized anasarca. Magnetic resonance cholangiopancreatography (MRCP) showed negative results for bile duct pathology. Workup for other causes of liver disease was negative. A diagnosis of acute alcoholic hepatitis was made, with Maddrey's discriminant function of > 32, so prednisolone was started. Her bilirubin and INR continued to increase despite steroids, and the patient unfortunately died.
CONCLUSIONS
Our case highlights the importance of considering hemochromatosis in the differential diagnosis of young patients presenting with liver failure, including cases suggestive of alcoholism as the likely etiology. Larger studies are needed to investigate the role of non-iron factors like alcohol and viral hepatitis in the progression of liver disease in simple heterozygotes with hereditary hemochromatosis, given the high prevalence of this mutation in persons of Northern European descent.
Background
Hereditary hemochromatosis (HH) (Fig. 1) is an autosomal recessive disorder where the clinical phenotype of skin pigmentation and organ damage occurs only in homozygotes; even in homozygotes, the phenotype has a broad spectrum depending on sex and penetrance which is age-related [1, 2]. With regard to heterozygotes, it is mostly the compound heterozygotes—C282Y and the H63D or S65C variant allele—that develop iron overload [3, 4]. Simple heterozygotes—that is, just C282Y—almost never develop iron overload or organ damage [5]. Although the role of non-iron-related factors like alcohol in modulating the iron threshold required to induce liver damage is well known, the strength of their association in each of the HH phenotypes remains an area that is largely unexplored.Fig. 1 HFE gene mutations causing Hereditary Hemochromatosis
We present a case where a simple heterozygote with alcoholism developed high ferritin and high transferrin saturation indicative of iron overload. This is very rare considering the young age, female sex and the genotype of the patient. The iron overload coupled with probable unhealthy dietary habits in the setting of alcoholism (more fat, less essential nutrients as reported in studies) [6] resulted in an inflammatory cocktail and caused our patient to succumb to acute liver failure at a young age.
Case presentation
A 29-year-old Caucasian woman presented to the hospital with 2 weeks of progressively worsening yellowish discoloration of her eyes and skin associated with anorexia, nausea, vomiting, diffuse abdominal discomfort, increasing abdominal girth, dark urine and pale stools. Past medical history was significant for prior episodes of hospitalization for acute alcoholic intoxication including an episode a few months prior. Imaging at that time showed hepatic steatosis but no features suggestive of hepatic cirrhosis or portal hypertension. Family history was significant for hereditary hemochromatosis. The patient’s father was heterozygous for C282Y and the paternal grandmother was homozygous for C282Y. The patient reported drinking about 1–2 glasses of wine every day, and denied smoking and illicit drug use. Vitals signs were as follows: pulse rate 94 beats per minute, respiratory rate 20 per minute, blood pressure 112/78 mmHg, temperature 36.9 °C and oxygen saturation 100% on room air. Physical examination showed scleral icterus, distended abdomen with hepatomegaly and mild generalized tenderness.
Laboratory results were as follows (Table 1): white blood cell count 31,600/μL, hemoglobin 10.1 g/dL, platelets 172/μL, ferritin 539 ng/dL, transferrin saturation 58.23%, peripheral blood smear showing stomatocytosis (Fig. 2), total bilirubin 8.7 mg/dL, direct bilirubin 7.4 mg/dL, aspartate aminotransferase (AST) 90 U/L, alanine aminotransferase (ALT) 30 U/L, alkaline phosphatase (ALKP) 420 U/L, prothrombin time (PT) 18.6 seconds, international normalized ratio (INR) 1.5, blood urea nitrogen 1.0 mg/dL, serum creatinine 0.5 mg/dL, Alcoholic Liver Disease/Nonalcoholic Fatty Liver Disease (ALD/NAFLD) Index (ANI) 2.6, suggesting a 93.2% probability of alcoholic liver disease, and phosphatidyl ethanol level 537 ng/ml (levels > 20 ng/ml indicate moderate–heavy ethanol consumption). Genetic testing results revealed that the patient was heterozygous for the HFE C282Y mutation and the normal allele, as well as negative for H63D and S65C. Imaging showed features suggestive of parenchymal liver disease, portal hypertension and generalized anasarca (Fig. 3). Magnetic resonance cholangiopancreatography (MRCP) was negative for bile duct pathology. Workup for other causes of liver disease including autoimmune hepatitis, Wilson’s disease, alpha-1 antitrypsin deficiency, celiac disease, primary biliary cirrhosis (PBC), Epstein–Barr virus (EBV), cytomegalovirus (CMV), viral hepatitis, tick-borne illnesses and leptospirosis all returned negative results. Biopsy of the liver was considered but was held due to the patient’s worsening general medical condition.Table 1 Trend of laboratory values over the course of hospitalization
Laboratory test Day 1 Day 8 Day 16 Day 23 Reference range with units
WBC 31.6 50.5 84 48.6 4.5–10.0 10 × 3/μL
RBC 3.08 2.95 2.64 2.10 4.00–5.00 10 × 6/μL
Platelets 172 370 309 176 140–400 10 × /μL
Hemoglobin 10.1 9.6 8.7 7.0 12.0–16.0 g/dL
Hematocrit 27.9 29.3 27.2 20.3 39.0–54.0%
MCV 90.6 99.3 103 96.7 80–99 fl
MCH 32.8 32.5 33 33.3 27–34 pg
MCHC 36.2 32.8 32 34.5 32–36 g/dL
RDW 18.1 23.2 24.6 22.8 39.0–54.0%
INR 1.5 2.2 1.3 1.9 0.1–1.4
BUN 1 5 38 104 7–17 mg/dL
Creatinine 0.55 0.53 0.73 4.28 0.55–1.02 mg/dL
Total bilirubin 8.7 10.3 17.4 24.7 0.2–1.3 mg/dL
Direct bilirubin 7.4 11.3 0.0–0.2 mg/dL
AST 90 64 45 71 15.00–37.00 U/L
ALT 30 23 27 12 12–78 U/L
ALKP 420 383 245 168 46–116 U/L
WBC white blood cells, RBC red blood cells, MCV cytomegalovirus, MCHC mean corpuscular hemoglobin concentration, RDW red cell distribution width, BUN blood urea nitrogen, AST aspartate aminotransferase, ALT alanine aminotransferase, ALKP alkaline phosphatase
Fig. 2 Peripheral smear showing stomatocytosis (orange arrows pointing to the stomatocytes)
Fig. 3 Magnetic Resonance Cholangiopancreatography (MRCP) showing enlarged liver on the left side of the image (shown using yellow array of lines on the left), enlarged spleen on the right side of the image (shown using yellow array of lines on the right) and generalized anasarca (pointed using orange arrows)
The patient was diagnosed with acute alcoholic hepatitis, and Gastroenterology was consulted. The patient’s Maddrey’s discriminant function was 46.4 (a score > 32 indicates poor prognosis and that the patient might benefit from glucocorticoid therapy), so orally administered prednisolone 40 mg/day was started. Although the patient’s total bilirubin (TBIL) and INR initially improved after initiating steroidal therapy, a rebound increase was noted (Table 1), raising concerns for impending liver failure. Also, her creatinine increased from 0.5 to 4 mg/dL. Nephrology was consulted, and a diagnosis of hepatorenal syndrome type 1 was favored. In addition to worsening TBIL, INR and creatinine, the patient developed encephalopathy, succumbed to the disease and died.
Discussion
In this article we focus on the mechanisms of liver injury from the effects of iron, alcohol and dietary habits, and it may not be surprising to see that some of these mechanisms overlap.
Mechanism of liver injury with iron
Excess iron in the hepatocytes and Kupffer cells results in the Fenton reaction and reactive oxygen species production. The free radicals induce lipid peroxidation, which damages the mitochondria, resulting in release of cytochrome c and liver cell apoptosis. Iron overload also stimulates the production of proinflammatory and pro-fibrogenic cytokines including transforming growth factor beta (TGF-β). TGF-β leads to the activation of hepatic stellate cells and excess collagen production. Excess collagen and cross-linking coupled with iron inhibit activation of the liver progenitor cells required for the regeneration of liver cells, resulting in fibrosis [2, 7].
Mechanism of liver injury with alcohol
Alcohol is metabolized to acetaldehyde. Acetaldehyde results in the generation of reactive oxygen species, which causes lipid peroxidation and cell membrane and DNA damage. Damaged hepatocytes express antigens which are otherwise hidden from the immune system, resulting in immune stimulation. Chronically heightened immune activity results in immune exhaustion, overwhelming bacterial infection, multi-organ damage and death. Also, chronic alcohol abuse results in overgrowth of gut bacteria, and this along with alcohol-induced leaky gut results in increased delivery of endotoxins to the liver and liver damage [8].
Mechanism of liver injury with non-healthy dietary habits
Excess dietary fat increases insulin resistance and hyperinsulinemia, which leads to accumulation of fatty acids. Accumulated fatty acids result in the generation of lipotoxic species, hepatocellular oxidant stress and cell death. The dying hepatocytes release signals and express antigens which are otherwise hidden from the immune system. turning on the immunogenic and fibrogenesis cascade [9]. Individual susceptibility to fatty acid-induced oxidant stress depends on other factors including iron overload states such as HFE and alcoholism [2].
Thus, many of the mechanisms of liver injury from iron, alcohol and unhealthy dietary habits overlap (Fig. 4). Although non-HH factors like alcoholism, NAFLD and nonalcoholic steatohepatitis (NASH) are associated with hyperferritinemia from chronic inflammation, the patient’s elevated transferrin saturation [(serum iron/total iron binding capacity) × 100] can only be explained by her HH status. While heavy alcohol consumption alone could cause severe liver damage, we hypothesize that her HFE status and possible unhealthy dietary fat in the setting of alcoholism accelerated the progression of liver disease. Studies have shown that alcoholics consume a higher amount of fatty food and carbohydrates along with lower consumption of vegetables and dairy products, which could have a detrimental effect on health [6].Fig. 4 Mechanisms of liver injury from the effects of iron, alcohol and dietary habits
Clinicians must continually probe for factors like personal or family history of hemochromatosis, dietary habits and alcoholism using different strategies and reformatting questions. This is especially important because early recognition followed by referral to specialized centers for treatment for hereditary hemochromatosis and detoxification would be pivotal in the prognosis of these patients. Our patient persistently denied any unhealthy alcohol use until later in the disease course. This, coupled with her blood alcohol level of < 3 at admission, very high white blood cell counts, young age and female sex, pointed more towards other differentials like autoimmune hepatitis and infectious etiologies. Although we were fortunate enough to be redirected towards alcohol as the etiology from reports of stomatocytosis in the peripheral blood, high ANI and very high phosphatidyl ethanol level, the patient unfortunately succumbed to her acute liver failure.
Conclusion
Considering the high prevalence of HH and the rising mortality from alcoholic liver disease among young adults [10, 11], more studies exploring the role of alcohol in the development of liver damage in simple heterozygotes and vice versa are essential to determine whether all alcoholics have to be screened for hereditary hemochromatosis. This is because more than one factor may often be involved in the pathogenesis and progression of liver dysfunction.
Abbreviations
WBCWhite blood cells
TSTransferrin saturation [(serum iron/total iron binding capacity) × 100]
ALTAlanine transaminase
ASTAspartate transaminase
ALKPAlkaline phosphatase
PTProthrombin time
INRInternational normalized ratio
ALDAlcoholic liver disease
NAFLDNonalcoholic fatty liver disease
ANIALD/NAFLD Index
CTComputed tomography
MRCPMagnetic resonance cholangiopancreatography
HFEHuman homeostatic iron regulator protein
HHHereditary hemochromatosis
TGFTransforming growth factor
TBILTotal bilirubin
GIGastroenterology
NASHNonalcoholic steatohepatitis
PBCPrimary biliary cirrhosis
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Acknowledgements
Not applicable.
Authors' contributions
SV—first author, performed literature review, prepared the manuscript. AA—second author, contributed to figures and images in the manuscript. BB—pathologist for our case report. MK—senior author, edited the manuscript. All authors read and approved the final manuscript.
Funding
No funding was received or used for this case report.
Ethics approval and consent to participate
Nashville General Hospital Research Oversight Committee (ROC) has determined that a case report does not produce generalizable knowledge, nor is it an investigation of an FDA regulated production.
Consent for publication
Written informed consent was obtained from the patient for publication of this case report and any accompanying images. A copy of the written consent form is available for review by the Editor-in-Chief of this journal.
Availability of data and material
Our patient’s health record is available in our electronic medical record system, and information can be verified if required by the reviewer.
Competing interests
The authors declare that they have no competing interests. | PREDNISOLONE | DrugsGivenReaction | CC BY | 33596964 | 19,456,434 | 2021-02-17 |
What was the administration route of drug 'PREDNISOLONE'? | Hemochromatosis, alcoholism and unhealthy dietary fat: a case report.
BACKGROUND
Hereditary hemochromatosis is an autosomal recessive disorder where the clinical phenotype of skin pigmentation and organ damage occurs only in homozygotes. Simple heterozygotes, that is, just C282Y, typically do not develop iron overload. Here we present a case where a simple heterozygote in combination with alcoholism developed high ferritin and high transferrin saturation levels indicative of iron overload. Though alcoholism alone could explain her presentation, we hypothesize that an inflammatory cocktail of iron and alcohol probably caused our patient to succumb to acute liver failure at a very young age.
METHODS
A 29-year-old Caucasian woman presented to the hospital with progressively worsening yellowish discoloration of her eyes and skin associated with anorexia, nausea, vomiting, diffuse abdominal discomfort, increasing abdominal girth, dark urine and pale stools for about 2 weeks. Family history was significant for hereditary hemochromatosis. Her father was a simple heterozygote and her grandmother was homozygous for C282Y. Physical examination showed scleral icterus, distended abdomen with hepatomegaly and mild generalized tenderness. Lab test results showed an elevated white blood cell count, ferritin 539 ng/dL, transferrin saturation 58.23%, elevated liver enzymes, elevated international normalized ratio (INR), low albumin, Alcoholic Liver Disease/Nonalcoholic Fatty Liver Disease (ALD/NAFLD) Index (ANI) of 2.6, suggesting a 93.2% probability of alcoholic liver disease, and phosphatidyl ethanol level of 537ng/ml. Genetic testing showed that the patient was heterozygous for human homeostatic iron regulator protein (HFE) C282Y mutation and the normal allele. Computed tomography (CT) of the abdomen revealed hepatomegaly, portal hypertension and generalized anasarca. Magnetic resonance cholangiopancreatography (MRCP) showed negative results for bile duct pathology. Workup for other causes of liver disease was negative. A diagnosis of acute alcoholic hepatitis was made, with Maddrey's discriminant function of > 32, so prednisolone was started. Her bilirubin and INR continued to increase despite steroids, and the patient unfortunately died.
CONCLUSIONS
Our case highlights the importance of considering hemochromatosis in the differential diagnosis of young patients presenting with liver failure, including cases suggestive of alcoholism as the likely etiology. Larger studies are needed to investigate the role of non-iron factors like alcohol and viral hepatitis in the progression of liver disease in simple heterozygotes with hereditary hemochromatosis, given the high prevalence of this mutation in persons of Northern European descent.
Background
Hereditary hemochromatosis (HH) (Fig. 1) is an autosomal recessive disorder where the clinical phenotype of skin pigmentation and organ damage occurs only in homozygotes; even in homozygotes, the phenotype has a broad spectrum depending on sex and penetrance which is age-related [1, 2]. With regard to heterozygotes, it is mostly the compound heterozygotes—C282Y and the H63D or S65C variant allele—that develop iron overload [3, 4]. Simple heterozygotes—that is, just C282Y—almost never develop iron overload or organ damage [5]. Although the role of non-iron-related factors like alcohol in modulating the iron threshold required to induce liver damage is well known, the strength of their association in each of the HH phenotypes remains an area that is largely unexplored.Fig. 1 HFE gene mutations causing Hereditary Hemochromatosis
We present a case where a simple heterozygote with alcoholism developed high ferritin and high transferrin saturation indicative of iron overload. This is very rare considering the young age, female sex and the genotype of the patient. The iron overload coupled with probable unhealthy dietary habits in the setting of alcoholism (more fat, less essential nutrients as reported in studies) [6] resulted in an inflammatory cocktail and caused our patient to succumb to acute liver failure at a young age.
Case presentation
A 29-year-old Caucasian woman presented to the hospital with 2 weeks of progressively worsening yellowish discoloration of her eyes and skin associated with anorexia, nausea, vomiting, diffuse abdominal discomfort, increasing abdominal girth, dark urine and pale stools. Past medical history was significant for prior episodes of hospitalization for acute alcoholic intoxication including an episode a few months prior. Imaging at that time showed hepatic steatosis but no features suggestive of hepatic cirrhosis or portal hypertension. Family history was significant for hereditary hemochromatosis. The patient’s father was heterozygous for C282Y and the paternal grandmother was homozygous for C282Y. The patient reported drinking about 1–2 glasses of wine every day, and denied smoking and illicit drug use. Vitals signs were as follows: pulse rate 94 beats per minute, respiratory rate 20 per minute, blood pressure 112/78 mmHg, temperature 36.9 °C and oxygen saturation 100% on room air. Physical examination showed scleral icterus, distended abdomen with hepatomegaly and mild generalized tenderness.
Laboratory results were as follows (Table 1): white blood cell count 31,600/μL, hemoglobin 10.1 g/dL, platelets 172/μL, ferritin 539 ng/dL, transferrin saturation 58.23%, peripheral blood smear showing stomatocytosis (Fig. 2), total bilirubin 8.7 mg/dL, direct bilirubin 7.4 mg/dL, aspartate aminotransferase (AST) 90 U/L, alanine aminotransferase (ALT) 30 U/L, alkaline phosphatase (ALKP) 420 U/L, prothrombin time (PT) 18.6 seconds, international normalized ratio (INR) 1.5, blood urea nitrogen 1.0 mg/dL, serum creatinine 0.5 mg/dL, Alcoholic Liver Disease/Nonalcoholic Fatty Liver Disease (ALD/NAFLD) Index (ANI) 2.6, suggesting a 93.2% probability of alcoholic liver disease, and phosphatidyl ethanol level 537 ng/ml (levels > 20 ng/ml indicate moderate–heavy ethanol consumption). Genetic testing results revealed that the patient was heterozygous for the HFE C282Y mutation and the normal allele, as well as negative for H63D and S65C. Imaging showed features suggestive of parenchymal liver disease, portal hypertension and generalized anasarca (Fig. 3). Magnetic resonance cholangiopancreatography (MRCP) was negative for bile duct pathology. Workup for other causes of liver disease including autoimmune hepatitis, Wilson’s disease, alpha-1 antitrypsin deficiency, celiac disease, primary biliary cirrhosis (PBC), Epstein–Barr virus (EBV), cytomegalovirus (CMV), viral hepatitis, tick-borne illnesses and leptospirosis all returned negative results. Biopsy of the liver was considered but was held due to the patient’s worsening general medical condition.Table 1 Trend of laboratory values over the course of hospitalization
Laboratory test Day 1 Day 8 Day 16 Day 23 Reference range with units
WBC 31.6 50.5 84 48.6 4.5–10.0 10 × 3/μL
RBC 3.08 2.95 2.64 2.10 4.00–5.00 10 × 6/μL
Platelets 172 370 309 176 140–400 10 × /μL
Hemoglobin 10.1 9.6 8.7 7.0 12.0–16.0 g/dL
Hematocrit 27.9 29.3 27.2 20.3 39.0–54.0%
MCV 90.6 99.3 103 96.7 80–99 fl
MCH 32.8 32.5 33 33.3 27–34 pg
MCHC 36.2 32.8 32 34.5 32–36 g/dL
RDW 18.1 23.2 24.6 22.8 39.0–54.0%
INR 1.5 2.2 1.3 1.9 0.1–1.4
BUN 1 5 38 104 7–17 mg/dL
Creatinine 0.55 0.53 0.73 4.28 0.55–1.02 mg/dL
Total bilirubin 8.7 10.3 17.4 24.7 0.2–1.3 mg/dL
Direct bilirubin 7.4 11.3 0.0–0.2 mg/dL
AST 90 64 45 71 15.00–37.00 U/L
ALT 30 23 27 12 12–78 U/L
ALKP 420 383 245 168 46–116 U/L
WBC white blood cells, RBC red blood cells, MCV cytomegalovirus, MCHC mean corpuscular hemoglobin concentration, RDW red cell distribution width, BUN blood urea nitrogen, AST aspartate aminotransferase, ALT alanine aminotransferase, ALKP alkaline phosphatase
Fig. 2 Peripheral smear showing stomatocytosis (orange arrows pointing to the stomatocytes)
Fig. 3 Magnetic Resonance Cholangiopancreatography (MRCP) showing enlarged liver on the left side of the image (shown using yellow array of lines on the left), enlarged spleen on the right side of the image (shown using yellow array of lines on the right) and generalized anasarca (pointed using orange arrows)
The patient was diagnosed with acute alcoholic hepatitis, and Gastroenterology was consulted. The patient’s Maddrey’s discriminant function was 46.4 (a score > 32 indicates poor prognosis and that the patient might benefit from glucocorticoid therapy), so orally administered prednisolone 40 mg/day was started. Although the patient’s total bilirubin (TBIL) and INR initially improved after initiating steroidal therapy, a rebound increase was noted (Table 1), raising concerns for impending liver failure. Also, her creatinine increased from 0.5 to 4 mg/dL. Nephrology was consulted, and a diagnosis of hepatorenal syndrome type 1 was favored. In addition to worsening TBIL, INR and creatinine, the patient developed encephalopathy, succumbed to the disease and died.
Discussion
In this article we focus on the mechanisms of liver injury from the effects of iron, alcohol and dietary habits, and it may not be surprising to see that some of these mechanisms overlap.
Mechanism of liver injury with iron
Excess iron in the hepatocytes and Kupffer cells results in the Fenton reaction and reactive oxygen species production. The free radicals induce lipid peroxidation, which damages the mitochondria, resulting in release of cytochrome c and liver cell apoptosis. Iron overload also stimulates the production of proinflammatory and pro-fibrogenic cytokines including transforming growth factor beta (TGF-β). TGF-β leads to the activation of hepatic stellate cells and excess collagen production. Excess collagen and cross-linking coupled with iron inhibit activation of the liver progenitor cells required for the regeneration of liver cells, resulting in fibrosis [2, 7].
Mechanism of liver injury with alcohol
Alcohol is metabolized to acetaldehyde. Acetaldehyde results in the generation of reactive oxygen species, which causes lipid peroxidation and cell membrane and DNA damage. Damaged hepatocytes express antigens which are otherwise hidden from the immune system, resulting in immune stimulation. Chronically heightened immune activity results in immune exhaustion, overwhelming bacterial infection, multi-organ damage and death. Also, chronic alcohol abuse results in overgrowth of gut bacteria, and this along with alcohol-induced leaky gut results in increased delivery of endotoxins to the liver and liver damage [8].
Mechanism of liver injury with non-healthy dietary habits
Excess dietary fat increases insulin resistance and hyperinsulinemia, which leads to accumulation of fatty acids. Accumulated fatty acids result in the generation of lipotoxic species, hepatocellular oxidant stress and cell death. The dying hepatocytes release signals and express antigens which are otherwise hidden from the immune system. turning on the immunogenic and fibrogenesis cascade [9]. Individual susceptibility to fatty acid-induced oxidant stress depends on other factors including iron overload states such as HFE and alcoholism [2].
Thus, many of the mechanisms of liver injury from iron, alcohol and unhealthy dietary habits overlap (Fig. 4). Although non-HH factors like alcoholism, NAFLD and nonalcoholic steatohepatitis (NASH) are associated with hyperferritinemia from chronic inflammation, the patient’s elevated transferrin saturation [(serum iron/total iron binding capacity) × 100] can only be explained by her HH status. While heavy alcohol consumption alone could cause severe liver damage, we hypothesize that her HFE status and possible unhealthy dietary fat in the setting of alcoholism accelerated the progression of liver disease. Studies have shown that alcoholics consume a higher amount of fatty food and carbohydrates along with lower consumption of vegetables and dairy products, which could have a detrimental effect on health [6].Fig. 4 Mechanisms of liver injury from the effects of iron, alcohol and dietary habits
Clinicians must continually probe for factors like personal or family history of hemochromatosis, dietary habits and alcoholism using different strategies and reformatting questions. This is especially important because early recognition followed by referral to specialized centers for treatment for hereditary hemochromatosis and detoxification would be pivotal in the prognosis of these patients. Our patient persistently denied any unhealthy alcohol use until later in the disease course. This, coupled with her blood alcohol level of < 3 at admission, very high white blood cell counts, young age and female sex, pointed more towards other differentials like autoimmune hepatitis and infectious etiologies. Although we were fortunate enough to be redirected towards alcohol as the etiology from reports of stomatocytosis in the peripheral blood, high ANI and very high phosphatidyl ethanol level, the patient unfortunately succumbed to her acute liver failure.
Conclusion
Considering the high prevalence of HH and the rising mortality from alcoholic liver disease among young adults [10, 11], more studies exploring the role of alcohol in the development of liver damage in simple heterozygotes and vice versa are essential to determine whether all alcoholics have to be screened for hereditary hemochromatosis. This is because more than one factor may often be involved in the pathogenesis and progression of liver dysfunction.
Abbreviations
WBCWhite blood cells
TSTransferrin saturation [(serum iron/total iron binding capacity) × 100]
ALTAlanine transaminase
ASTAspartate transaminase
ALKPAlkaline phosphatase
PTProthrombin time
INRInternational normalized ratio
ALDAlcoholic liver disease
NAFLDNonalcoholic fatty liver disease
ANIALD/NAFLD Index
CTComputed tomography
MRCPMagnetic resonance cholangiopancreatography
HFEHuman homeostatic iron regulator protein
HHHereditary hemochromatosis
TGFTransforming growth factor
TBILTotal bilirubin
GIGastroenterology
NASHNonalcoholic steatohepatitis
PBCPrimary biliary cirrhosis
Publisher's Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Acknowledgements
Not applicable.
Authors' contributions
SV—first author, performed literature review, prepared the manuscript. AA—second author, contributed to figures and images in the manuscript. BB—pathologist for our case report. MK—senior author, edited the manuscript. All authors read and approved the final manuscript.
Funding
No funding was received or used for this case report.
Ethics approval and consent to participate
Nashville General Hospital Research Oversight Committee (ROC) has determined that a case report does not produce generalizable knowledge, nor is it an investigation of an FDA regulated production.
Consent for publication
Written informed consent was obtained from the patient for publication of this case report and any accompanying images. A copy of the written consent form is available for review by the Editor-in-Chief of this journal.
Availability of data and material
Our patient’s health record is available in our electronic medical record system, and information can be verified if required by the reviewer.
Competing interests
The authors declare that they have no competing interests. | Oral | DrugAdministrationRoute | CC BY | 33596964 | 19,456,434 | 2021-02-17 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Blood sodium abnormal'. | Arginine-vasopressin infusion in a child with cranial diabetes insipidus during hyperhydration therapy with chemotherapy: a therapeutic challenge.
An 11-year-old girl presented with acute lower limb weakness, dehydration, hypernatraemia and secondary rhabdomyolysis on a background of an 8-month history of polyuria. Radiological investigations revealed a suprasellar tumour which was diagnosed on biopsy as a non-metastatic germinoma. Further endocrinological investigations confirmed panhypopituitarism and she commenced desmopressin, hydrocortisone and thyroxine. Her chemotherapeutic regime consisted of etoposide, carboplatin and ifosfamide, the latter of which required 4 litres of hyperhydration therapy daily. During the first course of ifosfamide, titration of oral desmopressin was trialled but this resulted in erratic sodium control leading to disorientation. Based on limited literature, we then trialled an arginine-vasopressin (AVP) infusion. A sliding scale was developed to adjust the AVP dose, with an aim to achieve a urine output of 3-4 mL/kg/h. During the second course of ifosamide, AVP infusion was commenced at the outset and tighter control of urine output and sodium levels was achieved. In conclusion, we found that an AVP infusion during hyperhydration therapy was required to achieve eunatraemia in a patient with cranial diabetes insipidus. Developing an AVP sliding scale requires individual variation; further reports/case series are required to underpin practice.
Certain chemotherapeutic regimens require large fluid volumes of hyperhydration therapy which can result in significant complications secondary to rapid serum sodium shifts in patients with diabetes insipidus. The use of a continuous AVP infusion and titrating with a sliding scale is more effective than oral desmopressin in regulating plasma sodium and fluid balance during hyperhydration therapy. No adverse effects were found in our patient using a continuous AVP infusion. Adjustment of the AVP infusion rate depends on urine output, fluid balance, plasma sodium levels and sensitivity/response of the child to titrated AVP doses.
Background
Central diabetes insipidus (CDI) is a disease caused by arginine-vasopressin (AVP) deficiency leading to polyuria and polydipsia (1). The destruction of neurons in the paraventricular and supraoptic nuclei of the posterior pituitary gland leads to AVP deficiency (1). Causes of CDI include germinomas, craniopharyngiomas, trauma, and 20-50% of cases are idiopathic (1).
The initial presentation of a child with a germinoma is often a protracted history of evolving endocrine deficiencies (2). Most initially develop CDI followed by other endocrine deficiencies such as hypocortisolism, hypothyroidism and growth failure. CDI is treated with desmopressin, which is a synthetic analogue of AVP (1). Doses required to manage diuresis vary greatly between patients. The outcome with treatment is favourable with more than 85% overall 5-year survival (2). However, the endocrine deficiencies do not resolve completely with tumour resolution and the children are often on lifelong hormone replacement therapy.
Certain chemotherapeutic agent protocols used in the treatment of suprasellar tumours require large fluid volumes (hyperhydration therapy) to reduce the risk of nephrotoxicity and haemorrhagic cystitis (3). Patients receiving hyperhydration therapy require strict monitoring of fluid balance and electrolytes to avoid complications secondary to sodium and potassium imbalance. Having a diagnosis of CDI while undergoing hyperhydration therapy presents an additional unique challenge. A study by Afzal et al. (4) showed that in children where cisplatin and/or ifosfamide chemotherapy were used (and hyperhydration therapy was therefore required), having CDI was a risk factor for prolonged admissions and complications including seizures, transient encephalopathy, and hyperreflexia with tremor. All children with CDI in their study also required daily changes in dosage and schedule of desmopressin.
In our patient, titration of oral desmopressin was insufficient to manage her fluid and electrolyte balance. We, therefore, used a continuous AVP infusion on a personalised sliding scale to successfully control her fluid balance and thus serum sodium concentration.
Case presentation
We report on an 11-year-old girl who had been previously well. Eight months prior to presentation, she developed polyuria, polydipsia and nocturia. Six months later, she developed frequent headaches. She then presented acutely to the paediatric team at a District General Hospital with generalised weakness and an inability to mobilise. Polyuria and polydipsia had resolved in the 24 h prior to presentation.
Investigation
She was found to have a serum sodium of 173 mmol/L (133–144 mmol/L), serum osmolality of 365 mosm/kg (275–295 mosm/kg), urine osmolality of 356 mosm/kg and creatinine kinase of 1825 units/L (30–170 units/L). A CT Head showed a large suprasellar mass.
She was transferred to our tertiary endocrine unit the following day under the care of the neurosurgery and endocrinology teams. MRI of brain showed a 3 × 3 cm mass involving the optic chiasm, optic nerves and hypothalamus with perilesional oedema (Fig. 1A and B). Biopsy of the mass revealed this to be a non-metastatic germinoma.
Figure 1 A coronal (A) and sagittal (B) MRI image of the germinoma.
Endocrinological bloods demonstrated the following results: cortisol <22 nmol/L at 08:00 h (80–580 nmol/L), ACTH <1.0 ng/L, TSH 3.09 mU/L (0.5–3.6 mU/L) and free T4 6.6 pmol/L (10–16.9 pmol/L) at 18:00 h and IGF-1 82 μg/L at 11:00 h (160–581 μg/L) thus leading to the diagnosis of panhypopituitarism.
An ophthalmology review revealed a severe restriction in visual acuity (right eye 6/85, left eye 6/38), reduced colour vision (right eye 15/17, left eye 16/17), and temporal optic disc pallor.
Treatment
Our patient was started on 0.9% saline at maintenance plus 5% deficit to support rehydration, with urine output also replaced mL for mL with 0.9% saline. A brief course of dexamethasone was commenced to reduce intracranial oedema surrounding the tumour. At this point, the polyuria reoccurred, thereby showing that the ‘initial resolution’ of polyuria was due to cortisol deficiency. She was started on a titrating dose of desmopressin (0–75 µg BD), 50 µg thyroxine and 10 mg/m2/day hydrocortisone in three divided doses. With this treatment, her sodium levels gradually improved over the next 7 days. She was found clinically to have a degree of hypodipsia which persisted after the resolution of the intracranial oedema. Hence her parents were advised to ensure that she took her maintenance fluid requirement daily.
The tumour was not amenable to surgery and she was started on alternating cycles of carboplatin/etoposide (cycles 1 and 3) and ifosfamide/etoposide (cycles 2 and 4). The two ifosfamide/etoposide cycles required 4 L of hyperhydration fluids daily using 0.45% saline/2.5% dextrose with 20 mmol/L potassium chloride. In the first cycle, due to concerns about water intoxication we omitted desmopressin while allowing her to drink freely in addition to hyperhydration therapy. However, this led to a diuresis of 7 L on the first day resulting in lethargy due to polyuria preventing sleep, and she required additional intravenous fluid to replace the fluid deficit. The following day 25–50 µg doses of oral desmopressin given at the beginning of diuresis were trialled, which led to significant swings in her plasma sodium (Fig. 2) causing disorientation.
Figure 2 Serum sodium (mmol/L) levels during first chemotherapy cycle involving hyperhydration. The normal serum sodium ranges are 135–145 mmol/L. The timing of blood samples was rounded to the nearest hour.
Based on a single publication reporting the management of 2 patients with CDI requiring hyperhydration therapy (5) we elected to trial an AVP infusion. A 0.04 units/mL AVP solution was prepared by adding 2 units of AVP (0.1 mL from 20 unit/mL ampoule) to 49.9 mL of 0.9% saline.
A starting dose of 0.0001units/kg/h was used and we created an ‘AVP sliding scale’ in order to titrate the AVP infusion based on urine output (initially aiming for 3–4 mL/kg/h to compensate for the large amount of infused fluid). The outline of the initial sliding scale we used is attached (Table 1). She was catheterised and her urine output monitored hourly.
Table 1 The AVP (arginine vasopressin) infusion sliding scale initially used.
Urine output, mL/kg/h Action taken
<2 Stop infusion till urine output is 3–4 mL/kg/h
2–3 Reduce infusion rate by 50%
3–4 Maintain infusion rate
4–5 Increase infusion rate by 50%
>5 Increase infusion rate by 100%
Urine output of 3–4 mL/kg/h was targeted.
The original protocol required further development in response to her clinical condition. Initially, we adjusted the AVP infusion rate by 50% depending on the urine output. However, we found that sizeable dose changes caused large swings in her serum sodium and therefore the rate change was reduced to 25%. Furthermore, the target urine output required adjustment based on fluid balance. When the AVP infusion was initially commenced during the first ifosfamide/etoposide cycle, she was hypernatraemic with a large negative fluid balance. Therefore the target urine output was reduced to 2–3 mL/kg/h until sodium levels normalised. With the second ifosamide/etoposide cycle, target urine output of 3–4 mL/kg/h led to rapid decline in sodium levels due to large intravenous fluid intake. Therefore target urine output was increased to 5–6 mL/kg/h to counteract this and maintain a neutral fluid balance.
Outcome and follow-up
During the first ifosfamide cycle, our patient’s plasma sodium showed significant fluctuation (Fig. 2). During the second cycle, AVP infusion was started at the onset and thus tighter control of plasma sodium was achieved (Fig. 3) and she did not experience any related symptoms.
Figure 3 Serum sodium (mmol/L) levels during second chemotherapy cycle involving hyperhydration. The normal serum sodium ranges are 135–145 mmol/L. The timing of blood samples was rounded to the nearest hour.
Following three cycles of chemotherapy, intracranial MRI demonstrated a tiny suprasellar residuum. Visual acuity was 6/45 in the right eye and 6/15 in the left, an improvement from 6/85 and 6/38 prior to starting chemotherapy.
She received proton therapy at University Hospital Essen in Germany, and following this there was further regression of the focal abnormality on her cranial MRI, and further improvement in her visual acuity (6/30 in right eye, 6/15 in left eye).
Both growth hormone (GH) stimulation with clonidine (200 µg) and LHRH testing showed suboptimal responses. She was started on GH and will soon commence pubertal induction with transdermal oestrogen. She continues to be monitored by the endocrinology and oncology teams, and remains well on desmopressin 75 µg/75 µg/100 µg daily, levothyroxine 75 µg daily, hydrocortisone 10 mg/m2 daily in three divided doses and GH 0.7 mg/m2 daily.
Discussion
Our patient had symptoms suggestive of diabetes insipidus for several months prior to admission. She had also developed hypodipsia which potentially precipitated her decompensated diabetes insipidus. The polyuria had appeared to ‘resolve’ just before she presented but its resumption once hydrocortisone was commenced demonstrated that progression to cortisol deficiency had masked her diabetes insipidus. The ‘masking’ of diabetes insipidus is due to impact of cortisol deficiency on AVP dependent and AVP independent mechanisms in the distal convoluting tubules and collecting ducts of the kidney, leading to reduced water clearance (6).
Hyperhydration presents a challenge when patients requiring chemotherapy also have CDI. This challenge was further compounded in our patient due to impaired thirst, which leads to a higher risk of decompensation. We found that by using an AVP infusion during hyperhydration therapy, we achieved better control of our patient’s serum sodium and fluid balance. She also felt less lethargic and was better able to cope with the symptoms relating to chemotherapy and she suffered no adverse effects.
There is a limited literature of case series on the use of AVP infusions in paediatric and adult patients. Wise-Faberowski et al. (7) used a continuous AVP infusion in 18 children with pre- and post-operative DI and compared their findings with 19 historical controls. They were able to maintain the children’s plasma sodium levels within range much more effectively with a continuous AVP infusion. No adverse effects were detected. However, none of these patients required hyperhydration therapy. Levine et al. (8) used an AVP infusion in adult patients with cranial lymphoma who did receive hyperhydration therapy with methotrexate. They initially trialled oral and subcutaneous desmopressin in one patient, and similar to us they found variable fluctuation in plasma sodium levels. Their protocol used a more dilute infusion (0.005 units/mL vs 0.04 units/mL) and a different titration method (titration by 50–100%) compared to ours. They also found it more effective to maintain plasma sodium within range with a continuous AVP infusion and neither of their two patients had side effects. Most pertinent to our case, Bryant et al. (5) trialled an AVP infusion in two children with DI and suprasellar germinoma, and compared with one child in whom desmopressin was withheld, all of whom required the same chemotherapy and hyperhydration regime. Our protocol for preparing an AVP infusion and the starting dose was adapted from theirs. They found that by using an AVP infusion, better control of the fluid and electrolyte balance was achieved. The children also required less fluid (the child who did not receive an AVP infusion required 20 L/m2/day of fluids compared to the other two children who received 3.8 L/m2/day).
Currently, there are no clinical trials in literature looking at continuous AVP infusions in patients with CDI requiring hyperhydration, probably due to rarity of this occurrence. Therefore it seems likely that management of these patients will depend on individual case reports or small case series. Based on our experience and the current literature, in the absence of other effective alternatives we recommend the use of continuous AVP infusions in the management of children with CDI requiring chemotherapy that necessitates hyperhydration. Using a continuous AVP infusion enables a tighter control of plasma sodium levels to be achieved through regulating urine output, and thus avoids the clinical effects that are seen with sharp swings in serum sodium. AVP infusions also have a rapid onset and termination profile and thus rate titrations lead to a rapid clinical effect. AVP infusions have a good safety profile with no adverse effects. Levine et al. (7) recognised hypertension as a possible adverse effect, however, the doses used to manage DI are significantly smaller (<0.5%) than the doses used to manage hypotension in adults.
Based on our experience, we recommend the following steps to be taken when starting a child on continuous AVP infusion for hyperhydration therapy:
Prepare AVP infusion by adding 2 units (0.1 mL from 20 unit/mL ampoule) to 49.9 mL 0.9% saline or 5% dextrose to make up 50 mL syringe. Solution is therefore 0.04 units/mL. If a more concentrated infusion is required, then five units of AVP (0.25 mL) added to 49.75 mL solution to give 0.1 units/mL could be used (5).
Start AVP infusion at the onset of hyperhydration therapy. Omit the usual oral desmopressin doses for the duration of AVP infusion.
HDU (High Dependency Unit) admission is recommended.
Some children may need a separate cannula inserted if the AVP infusion is incompatible with their chemotherapy drugs.
Insert a urinary catheter in all patients to enable hourly monitoring of urine output and fluid balance. Urea and electrolytes should be measured 6 hourly and the child should be weighed daily if possible.
Aim for a urine output of 3–4 mL/kg/h. However, if there is a significantly negative fluid balance as a result, aim for a lower urine output during AVP titrations. Conversely, if there is significantly positive fluid balance, target a higher urine output.
Start the AVP infusion at 0.0001 units/kg/h and alter the rate according to our recommended protocol (Table 2). The size of increment/decrement may need to be bigger or smaller depending on the sensitivity of the patient to desmopressin. We recommend that titration is performed hourly.
Table 2 Recommended protocol using the AVP (arginine vasopressin) Infusion Sliding Scale.
Urine output, mL/kg/h Action taken
<2 Stop infusion till urine output is 3–4 mL/kg/h
2–3 Reduce infusion rate by 25%
3–4 Maintain infusion rate
4–5 Increase infusion rate by 25%
>5 Increase infusion rate by 50%
Aim for urine output 3–4 mL/kg/h, however, this target will be dependent on fluid balance. Calculate hourly urine output to enable titration of AVP to be done hourly. Size of increment/decrement may need to be altered depending on patient’s sensitivity to desmopressin.
Once therapy is complete and serum sodium is within the normal range for the patient, oral desmopression should be recommenced 1 h prior to discontinuing the AVP infusion.
We recognise that admission to a specialist HDU may be challenging in some clinical settings; however, with the intensity of observations and hourly titrations needed, 1:1 nursing care is recommended. We also acknowledge the limitation of extrapolating an AVP infusion sliding scale based on a single patient and limited experience in published literature. Thus we recommend that further experience in using this protocol or alternatives published in the literature would aid in refining the sliding scale and management of patients with CDI undergoing chemotherapeutic regimens requiring hyperhydration.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
Patient consent
A written consent has been obtained from the patient’s parents and are attached to the submission.
Author contribution statement
Dr V Lee is the patient’s Paediatric Oncology Consultant and Prof P Dimitri is her Paediatric Endocrinology Consultant and managed this patient throughout the course of her therapy. J Devaraja is a Paediatric Registrar who had been involved with the endocrinological care of the patient when she was initially admitted, and during her chemotherapy regimens. J Devaraja and S Sloan wrote the manuscript. P Dimitri and V Lee reviewed and contributed to the content of the manuscript. P Dimitri is the senior author. | CISPLATIN, DESMOPRESSIN, DEXAMETHASONE, DEXTROSE, ETOPOSIDE, HYDROCORTISONE, IFOSFAMIDE, SODIUM CHLORIDE, VASOPRESSIN | DrugsGivenReaction | CC BY-NC-ND | 33597311 | 19,607,817 | 2021-02-17 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Glucocorticoid deficiency'. | Arginine-vasopressin infusion in a child with cranial diabetes insipidus during hyperhydration therapy with chemotherapy: a therapeutic challenge.
An 11-year-old girl presented with acute lower limb weakness, dehydration, hypernatraemia and secondary rhabdomyolysis on a background of an 8-month history of polyuria. Radiological investigations revealed a suprasellar tumour which was diagnosed on biopsy as a non-metastatic germinoma. Further endocrinological investigations confirmed panhypopituitarism and she commenced desmopressin, hydrocortisone and thyroxine. Her chemotherapeutic regime consisted of etoposide, carboplatin and ifosfamide, the latter of which required 4 litres of hyperhydration therapy daily. During the first course of ifosfamide, titration of oral desmopressin was trialled but this resulted in erratic sodium control leading to disorientation. Based on limited literature, we then trialled an arginine-vasopressin (AVP) infusion. A sliding scale was developed to adjust the AVP dose, with an aim to achieve a urine output of 3-4 mL/kg/h. During the second course of ifosamide, AVP infusion was commenced at the outset and tighter control of urine output and sodium levels was achieved. In conclusion, we found that an AVP infusion during hyperhydration therapy was required to achieve eunatraemia in a patient with cranial diabetes insipidus. Developing an AVP sliding scale requires individual variation; further reports/case series are required to underpin practice.
Certain chemotherapeutic regimens require large fluid volumes of hyperhydration therapy which can result in significant complications secondary to rapid serum sodium shifts in patients with diabetes insipidus. The use of a continuous AVP infusion and titrating with a sliding scale is more effective than oral desmopressin in regulating plasma sodium and fluid balance during hyperhydration therapy. No adverse effects were found in our patient using a continuous AVP infusion. Adjustment of the AVP infusion rate depends on urine output, fluid balance, plasma sodium levels and sensitivity/response of the child to titrated AVP doses.
Background
Central diabetes insipidus (CDI) is a disease caused by arginine-vasopressin (AVP) deficiency leading to polyuria and polydipsia (1). The destruction of neurons in the paraventricular and supraoptic nuclei of the posterior pituitary gland leads to AVP deficiency (1). Causes of CDI include germinomas, craniopharyngiomas, trauma, and 20-50% of cases are idiopathic (1).
The initial presentation of a child with a germinoma is often a protracted history of evolving endocrine deficiencies (2). Most initially develop CDI followed by other endocrine deficiencies such as hypocortisolism, hypothyroidism and growth failure. CDI is treated with desmopressin, which is a synthetic analogue of AVP (1). Doses required to manage diuresis vary greatly between patients. The outcome with treatment is favourable with more than 85% overall 5-year survival (2). However, the endocrine deficiencies do not resolve completely with tumour resolution and the children are often on lifelong hormone replacement therapy.
Certain chemotherapeutic agent protocols used in the treatment of suprasellar tumours require large fluid volumes (hyperhydration therapy) to reduce the risk of nephrotoxicity and haemorrhagic cystitis (3). Patients receiving hyperhydration therapy require strict monitoring of fluid balance and electrolytes to avoid complications secondary to sodium and potassium imbalance. Having a diagnosis of CDI while undergoing hyperhydration therapy presents an additional unique challenge. A study by Afzal et al. (4) showed that in children where cisplatin and/or ifosfamide chemotherapy were used (and hyperhydration therapy was therefore required), having CDI was a risk factor for prolonged admissions and complications including seizures, transient encephalopathy, and hyperreflexia with tremor. All children with CDI in their study also required daily changes in dosage and schedule of desmopressin.
In our patient, titration of oral desmopressin was insufficient to manage her fluid and electrolyte balance. We, therefore, used a continuous AVP infusion on a personalised sliding scale to successfully control her fluid balance and thus serum sodium concentration.
Case presentation
We report on an 11-year-old girl who had been previously well. Eight months prior to presentation, she developed polyuria, polydipsia and nocturia. Six months later, she developed frequent headaches. She then presented acutely to the paediatric team at a District General Hospital with generalised weakness and an inability to mobilise. Polyuria and polydipsia had resolved in the 24 h prior to presentation.
Investigation
She was found to have a serum sodium of 173 mmol/L (133–144 mmol/L), serum osmolality of 365 mosm/kg (275–295 mosm/kg), urine osmolality of 356 mosm/kg and creatinine kinase of 1825 units/L (30–170 units/L). A CT Head showed a large suprasellar mass.
She was transferred to our tertiary endocrine unit the following day under the care of the neurosurgery and endocrinology teams. MRI of brain showed a 3 × 3 cm mass involving the optic chiasm, optic nerves and hypothalamus with perilesional oedema (Fig. 1A and B). Biopsy of the mass revealed this to be a non-metastatic germinoma.
Figure 1 A coronal (A) and sagittal (B) MRI image of the germinoma.
Endocrinological bloods demonstrated the following results: cortisol <22 nmol/L at 08:00 h (80–580 nmol/L), ACTH <1.0 ng/L, TSH 3.09 mU/L (0.5–3.6 mU/L) and free T4 6.6 pmol/L (10–16.9 pmol/L) at 18:00 h and IGF-1 82 μg/L at 11:00 h (160–581 μg/L) thus leading to the diagnosis of panhypopituitarism.
An ophthalmology review revealed a severe restriction in visual acuity (right eye 6/85, left eye 6/38), reduced colour vision (right eye 15/17, left eye 16/17), and temporal optic disc pallor.
Treatment
Our patient was started on 0.9% saline at maintenance plus 5% deficit to support rehydration, with urine output also replaced mL for mL with 0.9% saline. A brief course of dexamethasone was commenced to reduce intracranial oedema surrounding the tumour. At this point, the polyuria reoccurred, thereby showing that the ‘initial resolution’ of polyuria was due to cortisol deficiency. She was started on a titrating dose of desmopressin (0–75 µg BD), 50 µg thyroxine and 10 mg/m2/day hydrocortisone in three divided doses. With this treatment, her sodium levels gradually improved over the next 7 days. She was found clinically to have a degree of hypodipsia which persisted after the resolution of the intracranial oedema. Hence her parents were advised to ensure that she took her maintenance fluid requirement daily.
The tumour was not amenable to surgery and she was started on alternating cycles of carboplatin/etoposide (cycles 1 and 3) and ifosfamide/etoposide (cycles 2 and 4). The two ifosfamide/etoposide cycles required 4 L of hyperhydration fluids daily using 0.45% saline/2.5% dextrose with 20 mmol/L potassium chloride. In the first cycle, due to concerns about water intoxication we omitted desmopressin while allowing her to drink freely in addition to hyperhydration therapy. However, this led to a diuresis of 7 L on the first day resulting in lethargy due to polyuria preventing sleep, and she required additional intravenous fluid to replace the fluid deficit. The following day 25–50 µg doses of oral desmopressin given at the beginning of diuresis were trialled, which led to significant swings in her plasma sodium (Fig. 2) causing disorientation.
Figure 2 Serum sodium (mmol/L) levels during first chemotherapy cycle involving hyperhydration. The normal serum sodium ranges are 135–145 mmol/L. The timing of blood samples was rounded to the nearest hour.
Based on a single publication reporting the management of 2 patients with CDI requiring hyperhydration therapy (5) we elected to trial an AVP infusion. A 0.04 units/mL AVP solution was prepared by adding 2 units of AVP (0.1 mL from 20 unit/mL ampoule) to 49.9 mL of 0.9% saline.
A starting dose of 0.0001units/kg/h was used and we created an ‘AVP sliding scale’ in order to titrate the AVP infusion based on urine output (initially aiming for 3–4 mL/kg/h to compensate for the large amount of infused fluid). The outline of the initial sliding scale we used is attached (Table 1). She was catheterised and her urine output monitored hourly.
Table 1 The AVP (arginine vasopressin) infusion sliding scale initially used.
Urine output, mL/kg/h Action taken
<2 Stop infusion till urine output is 3–4 mL/kg/h
2–3 Reduce infusion rate by 50%
3–4 Maintain infusion rate
4–5 Increase infusion rate by 50%
>5 Increase infusion rate by 100%
Urine output of 3–4 mL/kg/h was targeted.
The original protocol required further development in response to her clinical condition. Initially, we adjusted the AVP infusion rate by 50% depending on the urine output. However, we found that sizeable dose changes caused large swings in her serum sodium and therefore the rate change was reduced to 25%. Furthermore, the target urine output required adjustment based on fluid balance. When the AVP infusion was initially commenced during the first ifosfamide/etoposide cycle, she was hypernatraemic with a large negative fluid balance. Therefore the target urine output was reduced to 2–3 mL/kg/h until sodium levels normalised. With the second ifosamide/etoposide cycle, target urine output of 3–4 mL/kg/h led to rapid decline in sodium levels due to large intravenous fluid intake. Therefore target urine output was increased to 5–6 mL/kg/h to counteract this and maintain a neutral fluid balance.
Outcome and follow-up
During the first ifosfamide cycle, our patient’s plasma sodium showed significant fluctuation (Fig. 2). During the second cycle, AVP infusion was started at the onset and thus tighter control of plasma sodium was achieved (Fig. 3) and she did not experience any related symptoms.
Figure 3 Serum sodium (mmol/L) levels during second chemotherapy cycle involving hyperhydration. The normal serum sodium ranges are 135–145 mmol/L. The timing of blood samples was rounded to the nearest hour.
Following three cycles of chemotherapy, intracranial MRI demonstrated a tiny suprasellar residuum. Visual acuity was 6/45 in the right eye and 6/15 in the left, an improvement from 6/85 and 6/38 prior to starting chemotherapy.
She received proton therapy at University Hospital Essen in Germany, and following this there was further regression of the focal abnormality on her cranial MRI, and further improvement in her visual acuity (6/30 in right eye, 6/15 in left eye).
Both growth hormone (GH) stimulation with clonidine (200 µg) and LHRH testing showed suboptimal responses. She was started on GH and will soon commence pubertal induction with transdermal oestrogen. She continues to be monitored by the endocrinology and oncology teams, and remains well on desmopressin 75 µg/75 µg/100 µg daily, levothyroxine 75 µg daily, hydrocortisone 10 mg/m2 daily in three divided doses and GH 0.7 mg/m2 daily.
Discussion
Our patient had symptoms suggestive of diabetes insipidus for several months prior to admission. She had also developed hypodipsia which potentially precipitated her decompensated diabetes insipidus. The polyuria had appeared to ‘resolve’ just before she presented but its resumption once hydrocortisone was commenced demonstrated that progression to cortisol deficiency had masked her diabetes insipidus. The ‘masking’ of diabetes insipidus is due to impact of cortisol deficiency on AVP dependent and AVP independent mechanisms in the distal convoluting tubules and collecting ducts of the kidney, leading to reduced water clearance (6).
Hyperhydration presents a challenge when patients requiring chemotherapy also have CDI. This challenge was further compounded in our patient due to impaired thirst, which leads to a higher risk of decompensation. We found that by using an AVP infusion during hyperhydration therapy, we achieved better control of our patient’s serum sodium and fluid balance. She also felt less lethargic and was better able to cope with the symptoms relating to chemotherapy and she suffered no adverse effects.
There is a limited literature of case series on the use of AVP infusions in paediatric and adult patients. Wise-Faberowski et al. (7) used a continuous AVP infusion in 18 children with pre- and post-operative DI and compared their findings with 19 historical controls. They were able to maintain the children’s plasma sodium levels within range much more effectively with a continuous AVP infusion. No adverse effects were detected. However, none of these patients required hyperhydration therapy. Levine et al. (8) used an AVP infusion in adult patients with cranial lymphoma who did receive hyperhydration therapy with methotrexate. They initially trialled oral and subcutaneous desmopressin in one patient, and similar to us they found variable fluctuation in plasma sodium levels. Their protocol used a more dilute infusion (0.005 units/mL vs 0.04 units/mL) and a different titration method (titration by 50–100%) compared to ours. They also found it more effective to maintain plasma sodium within range with a continuous AVP infusion and neither of their two patients had side effects. Most pertinent to our case, Bryant et al. (5) trialled an AVP infusion in two children with DI and suprasellar germinoma, and compared with one child in whom desmopressin was withheld, all of whom required the same chemotherapy and hyperhydration regime. Our protocol for preparing an AVP infusion and the starting dose was adapted from theirs. They found that by using an AVP infusion, better control of the fluid and electrolyte balance was achieved. The children also required less fluid (the child who did not receive an AVP infusion required 20 L/m2/day of fluids compared to the other two children who received 3.8 L/m2/day).
Currently, there are no clinical trials in literature looking at continuous AVP infusions in patients with CDI requiring hyperhydration, probably due to rarity of this occurrence. Therefore it seems likely that management of these patients will depend on individual case reports or small case series. Based on our experience and the current literature, in the absence of other effective alternatives we recommend the use of continuous AVP infusions in the management of children with CDI requiring chemotherapy that necessitates hyperhydration. Using a continuous AVP infusion enables a tighter control of plasma sodium levels to be achieved through regulating urine output, and thus avoids the clinical effects that are seen with sharp swings in serum sodium. AVP infusions also have a rapid onset and termination profile and thus rate titrations lead to a rapid clinical effect. AVP infusions have a good safety profile with no adverse effects. Levine et al. (7) recognised hypertension as a possible adverse effect, however, the doses used to manage DI are significantly smaller (<0.5%) than the doses used to manage hypotension in adults.
Based on our experience, we recommend the following steps to be taken when starting a child on continuous AVP infusion for hyperhydration therapy:
Prepare AVP infusion by adding 2 units (0.1 mL from 20 unit/mL ampoule) to 49.9 mL 0.9% saline or 5% dextrose to make up 50 mL syringe. Solution is therefore 0.04 units/mL. If a more concentrated infusion is required, then five units of AVP (0.25 mL) added to 49.75 mL solution to give 0.1 units/mL could be used (5).
Start AVP infusion at the onset of hyperhydration therapy. Omit the usual oral desmopressin doses for the duration of AVP infusion.
HDU (High Dependency Unit) admission is recommended.
Some children may need a separate cannula inserted if the AVP infusion is incompatible with their chemotherapy drugs.
Insert a urinary catheter in all patients to enable hourly monitoring of urine output and fluid balance. Urea and electrolytes should be measured 6 hourly and the child should be weighed daily if possible.
Aim for a urine output of 3–4 mL/kg/h. However, if there is a significantly negative fluid balance as a result, aim for a lower urine output during AVP titrations. Conversely, if there is significantly positive fluid balance, target a higher urine output.
Start the AVP infusion at 0.0001 units/kg/h and alter the rate according to our recommended protocol (Table 2). The size of increment/decrement may need to be bigger or smaller depending on the sensitivity of the patient to desmopressin. We recommend that titration is performed hourly.
Table 2 Recommended protocol using the AVP (arginine vasopressin) Infusion Sliding Scale.
Urine output, mL/kg/h Action taken
<2 Stop infusion till urine output is 3–4 mL/kg/h
2–3 Reduce infusion rate by 25%
3–4 Maintain infusion rate
4–5 Increase infusion rate by 25%
>5 Increase infusion rate by 50%
Aim for urine output 3–4 mL/kg/h, however, this target will be dependent on fluid balance. Calculate hourly urine output to enable titration of AVP to be done hourly. Size of increment/decrement may need to be altered depending on patient’s sensitivity to desmopressin.
Once therapy is complete and serum sodium is within the normal range for the patient, oral desmopression should be recommenced 1 h prior to discontinuing the AVP infusion.
We recognise that admission to a specialist HDU may be challenging in some clinical settings; however, with the intensity of observations and hourly titrations needed, 1:1 nursing care is recommended. We also acknowledge the limitation of extrapolating an AVP infusion sliding scale based on a single patient and limited experience in published literature. Thus we recommend that further experience in using this protocol or alternatives published in the literature would aid in refining the sliding scale and management of patients with CDI undergoing chemotherapeutic regimens requiring hyperhydration.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
Patient consent
A written consent has been obtained from the patient’s parents and are attached to the submission.
Author contribution statement
Dr V Lee is the patient’s Paediatric Oncology Consultant and Prof P Dimitri is her Paediatric Endocrinology Consultant and managed this patient throughout the course of her therapy. J Devaraja is a Paediatric Registrar who had been involved with the endocrinological care of the patient when she was initially admitted, and during her chemotherapy regimens. J Devaraja and S Sloan wrote the manuscript. P Dimitri and V Lee reviewed and contributed to the content of the manuscript. P Dimitri is the senior author. | CISPLATIN, DESMOPRESSIN, DEXAMETHASONE, DEXTROSE, ETOPOSIDE, HYDROCORTISONE, IFOSFAMIDE, SODIUM CHLORIDE, VASOPRESSIN | DrugsGivenReaction | CC BY-NC-ND | 33597311 | 19,607,817 | 2021-02-17 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Hypernatraemia'. | Arginine-vasopressin infusion in a child with cranial diabetes insipidus during hyperhydration therapy with chemotherapy: a therapeutic challenge.
An 11-year-old girl presented with acute lower limb weakness, dehydration, hypernatraemia and secondary rhabdomyolysis on a background of an 8-month history of polyuria. Radiological investigations revealed a suprasellar tumour which was diagnosed on biopsy as a non-metastatic germinoma. Further endocrinological investigations confirmed panhypopituitarism and she commenced desmopressin, hydrocortisone and thyroxine. Her chemotherapeutic regime consisted of etoposide, carboplatin and ifosfamide, the latter of which required 4 litres of hyperhydration therapy daily. During the first course of ifosfamide, titration of oral desmopressin was trialled but this resulted in erratic sodium control leading to disorientation. Based on limited literature, we then trialled an arginine-vasopressin (AVP) infusion. A sliding scale was developed to adjust the AVP dose, with an aim to achieve a urine output of 3-4 mL/kg/h. During the second course of ifosamide, AVP infusion was commenced at the outset and tighter control of urine output and sodium levels was achieved. In conclusion, we found that an AVP infusion during hyperhydration therapy was required to achieve eunatraemia in a patient with cranial diabetes insipidus. Developing an AVP sliding scale requires individual variation; further reports/case series are required to underpin practice.
Certain chemotherapeutic regimens require large fluid volumes of hyperhydration therapy which can result in significant complications secondary to rapid serum sodium shifts in patients with diabetes insipidus. The use of a continuous AVP infusion and titrating with a sliding scale is more effective than oral desmopressin in regulating plasma sodium and fluid balance during hyperhydration therapy. No adverse effects were found in our patient using a continuous AVP infusion. Adjustment of the AVP infusion rate depends on urine output, fluid balance, plasma sodium levels and sensitivity/response of the child to titrated AVP doses.
Background
Central diabetes insipidus (CDI) is a disease caused by arginine-vasopressin (AVP) deficiency leading to polyuria and polydipsia (1). The destruction of neurons in the paraventricular and supraoptic nuclei of the posterior pituitary gland leads to AVP deficiency (1). Causes of CDI include germinomas, craniopharyngiomas, trauma, and 20-50% of cases are idiopathic (1).
The initial presentation of a child with a germinoma is often a protracted history of evolving endocrine deficiencies (2). Most initially develop CDI followed by other endocrine deficiencies such as hypocortisolism, hypothyroidism and growth failure. CDI is treated with desmopressin, which is a synthetic analogue of AVP (1). Doses required to manage diuresis vary greatly between patients. The outcome with treatment is favourable with more than 85% overall 5-year survival (2). However, the endocrine deficiencies do not resolve completely with tumour resolution and the children are often on lifelong hormone replacement therapy.
Certain chemotherapeutic agent protocols used in the treatment of suprasellar tumours require large fluid volumes (hyperhydration therapy) to reduce the risk of nephrotoxicity and haemorrhagic cystitis (3). Patients receiving hyperhydration therapy require strict monitoring of fluid balance and electrolytes to avoid complications secondary to sodium and potassium imbalance. Having a diagnosis of CDI while undergoing hyperhydration therapy presents an additional unique challenge. A study by Afzal et al. (4) showed that in children where cisplatin and/or ifosfamide chemotherapy were used (and hyperhydration therapy was therefore required), having CDI was a risk factor for prolonged admissions and complications including seizures, transient encephalopathy, and hyperreflexia with tremor. All children with CDI in their study also required daily changes in dosage and schedule of desmopressin.
In our patient, titration of oral desmopressin was insufficient to manage her fluid and electrolyte balance. We, therefore, used a continuous AVP infusion on a personalised sliding scale to successfully control her fluid balance and thus serum sodium concentration.
Case presentation
We report on an 11-year-old girl who had been previously well. Eight months prior to presentation, she developed polyuria, polydipsia and nocturia. Six months later, she developed frequent headaches. She then presented acutely to the paediatric team at a District General Hospital with generalised weakness and an inability to mobilise. Polyuria and polydipsia had resolved in the 24 h prior to presentation.
Investigation
She was found to have a serum sodium of 173 mmol/L (133–144 mmol/L), serum osmolality of 365 mosm/kg (275–295 mosm/kg), urine osmolality of 356 mosm/kg and creatinine kinase of 1825 units/L (30–170 units/L). A CT Head showed a large suprasellar mass.
She was transferred to our tertiary endocrine unit the following day under the care of the neurosurgery and endocrinology teams. MRI of brain showed a 3 × 3 cm mass involving the optic chiasm, optic nerves and hypothalamus with perilesional oedema (Fig. 1A and B). Biopsy of the mass revealed this to be a non-metastatic germinoma.
Figure 1 A coronal (A) and sagittal (B) MRI image of the germinoma.
Endocrinological bloods demonstrated the following results: cortisol <22 nmol/L at 08:00 h (80–580 nmol/L), ACTH <1.0 ng/L, TSH 3.09 mU/L (0.5–3.6 mU/L) and free T4 6.6 pmol/L (10–16.9 pmol/L) at 18:00 h and IGF-1 82 μg/L at 11:00 h (160–581 μg/L) thus leading to the diagnosis of panhypopituitarism.
An ophthalmology review revealed a severe restriction in visual acuity (right eye 6/85, left eye 6/38), reduced colour vision (right eye 15/17, left eye 16/17), and temporal optic disc pallor.
Treatment
Our patient was started on 0.9% saline at maintenance plus 5% deficit to support rehydration, with urine output also replaced mL for mL with 0.9% saline. A brief course of dexamethasone was commenced to reduce intracranial oedema surrounding the tumour. At this point, the polyuria reoccurred, thereby showing that the ‘initial resolution’ of polyuria was due to cortisol deficiency. She was started on a titrating dose of desmopressin (0–75 µg BD), 50 µg thyroxine and 10 mg/m2/day hydrocortisone in three divided doses. With this treatment, her sodium levels gradually improved over the next 7 days. She was found clinically to have a degree of hypodipsia which persisted after the resolution of the intracranial oedema. Hence her parents were advised to ensure that she took her maintenance fluid requirement daily.
The tumour was not amenable to surgery and she was started on alternating cycles of carboplatin/etoposide (cycles 1 and 3) and ifosfamide/etoposide (cycles 2 and 4). The two ifosfamide/etoposide cycles required 4 L of hyperhydration fluids daily using 0.45% saline/2.5% dextrose with 20 mmol/L potassium chloride. In the first cycle, due to concerns about water intoxication we omitted desmopressin while allowing her to drink freely in addition to hyperhydration therapy. However, this led to a diuresis of 7 L on the first day resulting in lethargy due to polyuria preventing sleep, and she required additional intravenous fluid to replace the fluid deficit. The following day 25–50 µg doses of oral desmopressin given at the beginning of diuresis were trialled, which led to significant swings in her plasma sodium (Fig. 2) causing disorientation.
Figure 2 Serum sodium (mmol/L) levels during first chemotherapy cycle involving hyperhydration. The normal serum sodium ranges are 135–145 mmol/L. The timing of blood samples was rounded to the nearest hour.
Based on a single publication reporting the management of 2 patients with CDI requiring hyperhydration therapy (5) we elected to trial an AVP infusion. A 0.04 units/mL AVP solution was prepared by adding 2 units of AVP (0.1 mL from 20 unit/mL ampoule) to 49.9 mL of 0.9% saline.
A starting dose of 0.0001units/kg/h was used and we created an ‘AVP sliding scale’ in order to titrate the AVP infusion based on urine output (initially aiming for 3–4 mL/kg/h to compensate for the large amount of infused fluid). The outline of the initial sliding scale we used is attached (Table 1). She was catheterised and her urine output monitored hourly.
Table 1 The AVP (arginine vasopressin) infusion sliding scale initially used.
Urine output, mL/kg/h Action taken
<2 Stop infusion till urine output is 3–4 mL/kg/h
2–3 Reduce infusion rate by 50%
3–4 Maintain infusion rate
4–5 Increase infusion rate by 50%
>5 Increase infusion rate by 100%
Urine output of 3–4 mL/kg/h was targeted.
The original protocol required further development in response to her clinical condition. Initially, we adjusted the AVP infusion rate by 50% depending on the urine output. However, we found that sizeable dose changes caused large swings in her serum sodium and therefore the rate change was reduced to 25%. Furthermore, the target urine output required adjustment based on fluid balance. When the AVP infusion was initially commenced during the first ifosfamide/etoposide cycle, she was hypernatraemic with a large negative fluid balance. Therefore the target urine output was reduced to 2–3 mL/kg/h until sodium levels normalised. With the second ifosamide/etoposide cycle, target urine output of 3–4 mL/kg/h led to rapid decline in sodium levels due to large intravenous fluid intake. Therefore target urine output was increased to 5–6 mL/kg/h to counteract this and maintain a neutral fluid balance.
Outcome and follow-up
During the first ifosfamide cycle, our patient’s plasma sodium showed significant fluctuation (Fig. 2). During the second cycle, AVP infusion was started at the onset and thus tighter control of plasma sodium was achieved (Fig. 3) and she did not experience any related symptoms.
Figure 3 Serum sodium (mmol/L) levels during second chemotherapy cycle involving hyperhydration. The normal serum sodium ranges are 135–145 mmol/L. The timing of blood samples was rounded to the nearest hour.
Following three cycles of chemotherapy, intracranial MRI demonstrated a tiny suprasellar residuum. Visual acuity was 6/45 in the right eye and 6/15 in the left, an improvement from 6/85 and 6/38 prior to starting chemotherapy.
She received proton therapy at University Hospital Essen in Germany, and following this there was further regression of the focal abnormality on her cranial MRI, and further improvement in her visual acuity (6/30 in right eye, 6/15 in left eye).
Both growth hormone (GH) stimulation with clonidine (200 µg) and LHRH testing showed suboptimal responses. She was started on GH and will soon commence pubertal induction with transdermal oestrogen. She continues to be monitored by the endocrinology and oncology teams, and remains well on desmopressin 75 µg/75 µg/100 µg daily, levothyroxine 75 µg daily, hydrocortisone 10 mg/m2 daily in three divided doses and GH 0.7 mg/m2 daily.
Discussion
Our patient had symptoms suggestive of diabetes insipidus for several months prior to admission. She had also developed hypodipsia which potentially precipitated her decompensated diabetes insipidus. The polyuria had appeared to ‘resolve’ just before she presented but its resumption once hydrocortisone was commenced demonstrated that progression to cortisol deficiency had masked her diabetes insipidus. The ‘masking’ of diabetes insipidus is due to impact of cortisol deficiency on AVP dependent and AVP independent mechanisms in the distal convoluting tubules and collecting ducts of the kidney, leading to reduced water clearance (6).
Hyperhydration presents a challenge when patients requiring chemotherapy also have CDI. This challenge was further compounded in our patient due to impaired thirst, which leads to a higher risk of decompensation. We found that by using an AVP infusion during hyperhydration therapy, we achieved better control of our patient’s serum sodium and fluid balance. She also felt less lethargic and was better able to cope with the symptoms relating to chemotherapy and she suffered no adverse effects.
There is a limited literature of case series on the use of AVP infusions in paediatric and adult patients. Wise-Faberowski et al. (7) used a continuous AVP infusion in 18 children with pre- and post-operative DI and compared their findings with 19 historical controls. They were able to maintain the children’s plasma sodium levels within range much more effectively with a continuous AVP infusion. No adverse effects were detected. However, none of these patients required hyperhydration therapy. Levine et al. (8) used an AVP infusion in adult patients with cranial lymphoma who did receive hyperhydration therapy with methotrexate. They initially trialled oral and subcutaneous desmopressin in one patient, and similar to us they found variable fluctuation in plasma sodium levels. Their protocol used a more dilute infusion (0.005 units/mL vs 0.04 units/mL) and a different titration method (titration by 50–100%) compared to ours. They also found it more effective to maintain plasma sodium within range with a continuous AVP infusion and neither of their two patients had side effects. Most pertinent to our case, Bryant et al. (5) trialled an AVP infusion in two children with DI and suprasellar germinoma, and compared with one child in whom desmopressin was withheld, all of whom required the same chemotherapy and hyperhydration regime. Our protocol for preparing an AVP infusion and the starting dose was adapted from theirs. They found that by using an AVP infusion, better control of the fluid and electrolyte balance was achieved. The children also required less fluid (the child who did not receive an AVP infusion required 20 L/m2/day of fluids compared to the other two children who received 3.8 L/m2/day).
Currently, there are no clinical trials in literature looking at continuous AVP infusions in patients with CDI requiring hyperhydration, probably due to rarity of this occurrence. Therefore it seems likely that management of these patients will depend on individual case reports or small case series. Based on our experience and the current literature, in the absence of other effective alternatives we recommend the use of continuous AVP infusions in the management of children with CDI requiring chemotherapy that necessitates hyperhydration. Using a continuous AVP infusion enables a tighter control of plasma sodium levels to be achieved through regulating urine output, and thus avoids the clinical effects that are seen with sharp swings in serum sodium. AVP infusions also have a rapid onset and termination profile and thus rate titrations lead to a rapid clinical effect. AVP infusions have a good safety profile with no adverse effects. Levine et al. (7) recognised hypertension as a possible adverse effect, however, the doses used to manage DI are significantly smaller (<0.5%) than the doses used to manage hypotension in adults.
Based on our experience, we recommend the following steps to be taken when starting a child on continuous AVP infusion for hyperhydration therapy:
Prepare AVP infusion by adding 2 units (0.1 mL from 20 unit/mL ampoule) to 49.9 mL 0.9% saline or 5% dextrose to make up 50 mL syringe. Solution is therefore 0.04 units/mL. If a more concentrated infusion is required, then five units of AVP (0.25 mL) added to 49.75 mL solution to give 0.1 units/mL could be used (5).
Start AVP infusion at the onset of hyperhydration therapy. Omit the usual oral desmopressin doses for the duration of AVP infusion.
HDU (High Dependency Unit) admission is recommended.
Some children may need a separate cannula inserted if the AVP infusion is incompatible with their chemotherapy drugs.
Insert a urinary catheter in all patients to enable hourly monitoring of urine output and fluid balance. Urea and electrolytes should be measured 6 hourly and the child should be weighed daily if possible.
Aim for a urine output of 3–4 mL/kg/h. However, if there is a significantly negative fluid balance as a result, aim for a lower urine output during AVP titrations. Conversely, if there is significantly positive fluid balance, target a higher urine output.
Start the AVP infusion at 0.0001 units/kg/h and alter the rate according to our recommended protocol (Table 2). The size of increment/decrement may need to be bigger or smaller depending on the sensitivity of the patient to desmopressin. We recommend that titration is performed hourly.
Table 2 Recommended protocol using the AVP (arginine vasopressin) Infusion Sliding Scale.
Urine output, mL/kg/h Action taken
<2 Stop infusion till urine output is 3–4 mL/kg/h
2–3 Reduce infusion rate by 25%
3–4 Maintain infusion rate
4–5 Increase infusion rate by 25%
>5 Increase infusion rate by 50%
Aim for urine output 3–4 mL/kg/h, however, this target will be dependent on fluid balance. Calculate hourly urine output to enable titration of AVP to be done hourly. Size of increment/decrement may need to be altered depending on patient’s sensitivity to desmopressin.
Once therapy is complete and serum sodium is within the normal range for the patient, oral desmopression should be recommenced 1 h prior to discontinuing the AVP infusion.
We recognise that admission to a specialist HDU may be challenging in some clinical settings; however, with the intensity of observations and hourly titrations needed, 1:1 nursing care is recommended. We also acknowledge the limitation of extrapolating an AVP infusion sliding scale based on a single patient and limited experience in published literature. Thus we recommend that further experience in using this protocol or alternatives published in the literature would aid in refining the sliding scale and management of patients with CDI undergoing chemotherapeutic regimens requiring hyperhydration.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
Patient consent
A written consent has been obtained from the patient’s parents and are attached to the submission.
Author contribution statement
Dr V Lee is the patient’s Paediatric Oncology Consultant and Prof P Dimitri is her Paediatric Endocrinology Consultant and managed this patient throughout the course of her therapy. J Devaraja is a Paediatric Registrar who had been involved with the endocrinological care of the patient when she was initially admitted, and during her chemotherapy regimens. J Devaraja and S Sloan wrote the manuscript. P Dimitri and V Lee reviewed and contributed to the content of the manuscript. P Dimitri is the senior author. | CISPLATIN, DESMOPRESSIN, DEXAMETHASONE, DEXTROSE, ETOPOSIDE, HYDROCORTISONE, IFOSFAMIDE, SODIUM CHLORIDE, VASOPRESSIN | DrugsGivenReaction | CC BY-NC-ND | 33597311 | 19,607,817 | 2021-02-17 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Unmasking of previously unidentified disease'. | Arginine-vasopressin infusion in a child with cranial diabetes insipidus during hyperhydration therapy with chemotherapy: a therapeutic challenge.
An 11-year-old girl presented with acute lower limb weakness, dehydration, hypernatraemia and secondary rhabdomyolysis on a background of an 8-month history of polyuria. Radiological investigations revealed a suprasellar tumour which was diagnosed on biopsy as a non-metastatic germinoma. Further endocrinological investigations confirmed panhypopituitarism and she commenced desmopressin, hydrocortisone and thyroxine. Her chemotherapeutic regime consisted of etoposide, carboplatin and ifosfamide, the latter of which required 4 litres of hyperhydration therapy daily. During the first course of ifosfamide, titration of oral desmopressin was trialled but this resulted in erratic sodium control leading to disorientation. Based on limited literature, we then trialled an arginine-vasopressin (AVP) infusion. A sliding scale was developed to adjust the AVP dose, with an aim to achieve a urine output of 3-4 mL/kg/h. During the second course of ifosamide, AVP infusion was commenced at the outset and tighter control of urine output and sodium levels was achieved. In conclusion, we found that an AVP infusion during hyperhydration therapy was required to achieve eunatraemia in a patient with cranial diabetes insipidus. Developing an AVP sliding scale requires individual variation; further reports/case series are required to underpin practice.
Certain chemotherapeutic regimens require large fluid volumes of hyperhydration therapy which can result in significant complications secondary to rapid serum sodium shifts in patients with diabetes insipidus. The use of a continuous AVP infusion and titrating with a sliding scale is more effective than oral desmopressin in regulating plasma sodium and fluid balance during hyperhydration therapy. No adverse effects were found in our patient using a continuous AVP infusion. Adjustment of the AVP infusion rate depends on urine output, fluid balance, plasma sodium levels and sensitivity/response of the child to titrated AVP doses.
Background
Central diabetes insipidus (CDI) is a disease caused by arginine-vasopressin (AVP) deficiency leading to polyuria and polydipsia (1). The destruction of neurons in the paraventricular and supraoptic nuclei of the posterior pituitary gland leads to AVP deficiency (1). Causes of CDI include germinomas, craniopharyngiomas, trauma, and 20-50% of cases are idiopathic (1).
The initial presentation of a child with a germinoma is often a protracted history of evolving endocrine deficiencies (2). Most initially develop CDI followed by other endocrine deficiencies such as hypocortisolism, hypothyroidism and growth failure. CDI is treated with desmopressin, which is a synthetic analogue of AVP (1). Doses required to manage diuresis vary greatly between patients. The outcome with treatment is favourable with more than 85% overall 5-year survival (2). However, the endocrine deficiencies do not resolve completely with tumour resolution and the children are often on lifelong hormone replacement therapy.
Certain chemotherapeutic agent protocols used in the treatment of suprasellar tumours require large fluid volumes (hyperhydration therapy) to reduce the risk of nephrotoxicity and haemorrhagic cystitis (3). Patients receiving hyperhydration therapy require strict monitoring of fluid balance and electrolytes to avoid complications secondary to sodium and potassium imbalance. Having a diagnosis of CDI while undergoing hyperhydration therapy presents an additional unique challenge. A study by Afzal et al. (4) showed that in children where cisplatin and/or ifosfamide chemotherapy were used (and hyperhydration therapy was therefore required), having CDI was a risk factor for prolonged admissions and complications including seizures, transient encephalopathy, and hyperreflexia with tremor. All children with CDI in their study also required daily changes in dosage and schedule of desmopressin.
In our patient, titration of oral desmopressin was insufficient to manage her fluid and electrolyte balance. We, therefore, used a continuous AVP infusion on a personalised sliding scale to successfully control her fluid balance and thus serum sodium concentration.
Case presentation
We report on an 11-year-old girl who had been previously well. Eight months prior to presentation, she developed polyuria, polydipsia and nocturia. Six months later, she developed frequent headaches. She then presented acutely to the paediatric team at a District General Hospital with generalised weakness and an inability to mobilise. Polyuria and polydipsia had resolved in the 24 h prior to presentation.
Investigation
She was found to have a serum sodium of 173 mmol/L (133–144 mmol/L), serum osmolality of 365 mosm/kg (275–295 mosm/kg), urine osmolality of 356 mosm/kg and creatinine kinase of 1825 units/L (30–170 units/L). A CT Head showed a large suprasellar mass.
She was transferred to our tertiary endocrine unit the following day under the care of the neurosurgery and endocrinology teams. MRI of brain showed a 3 × 3 cm mass involving the optic chiasm, optic nerves and hypothalamus with perilesional oedema (Fig. 1A and B). Biopsy of the mass revealed this to be a non-metastatic germinoma.
Figure 1 A coronal (A) and sagittal (B) MRI image of the germinoma.
Endocrinological bloods demonstrated the following results: cortisol <22 nmol/L at 08:00 h (80–580 nmol/L), ACTH <1.0 ng/L, TSH 3.09 mU/L (0.5–3.6 mU/L) and free T4 6.6 pmol/L (10–16.9 pmol/L) at 18:00 h and IGF-1 82 μg/L at 11:00 h (160–581 μg/L) thus leading to the diagnosis of panhypopituitarism.
An ophthalmology review revealed a severe restriction in visual acuity (right eye 6/85, left eye 6/38), reduced colour vision (right eye 15/17, left eye 16/17), and temporal optic disc pallor.
Treatment
Our patient was started on 0.9% saline at maintenance plus 5% deficit to support rehydration, with urine output also replaced mL for mL with 0.9% saline. A brief course of dexamethasone was commenced to reduce intracranial oedema surrounding the tumour. At this point, the polyuria reoccurred, thereby showing that the ‘initial resolution’ of polyuria was due to cortisol deficiency. She was started on a titrating dose of desmopressin (0–75 µg BD), 50 µg thyroxine and 10 mg/m2/day hydrocortisone in three divided doses. With this treatment, her sodium levels gradually improved over the next 7 days. She was found clinically to have a degree of hypodipsia which persisted after the resolution of the intracranial oedema. Hence her parents were advised to ensure that she took her maintenance fluid requirement daily.
The tumour was not amenable to surgery and she was started on alternating cycles of carboplatin/etoposide (cycles 1 and 3) and ifosfamide/etoposide (cycles 2 and 4). The two ifosfamide/etoposide cycles required 4 L of hyperhydration fluids daily using 0.45% saline/2.5% dextrose with 20 mmol/L potassium chloride. In the first cycle, due to concerns about water intoxication we omitted desmopressin while allowing her to drink freely in addition to hyperhydration therapy. However, this led to a diuresis of 7 L on the first day resulting in lethargy due to polyuria preventing sleep, and she required additional intravenous fluid to replace the fluid deficit. The following day 25–50 µg doses of oral desmopressin given at the beginning of diuresis were trialled, which led to significant swings in her plasma sodium (Fig. 2) causing disorientation.
Figure 2 Serum sodium (mmol/L) levels during first chemotherapy cycle involving hyperhydration. The normal serum sodium ranges are 135–145 mmol/L. The timing of blood samples was rounded to the nearest hour.
Based on a single publication reporting the management of 2 patients with CDI requiring hyperhydration therapy (5) we elected to trial an AVP infusion. A 0.04 units/mL AVP solution was prepared by adding 2 units of AVP (0.1 mL from 20 unit/mL ampoule) to 49.9 mL of 0.9% saline.
A starting dose of 0.0001units/kg/h was used and we created an ‘AVP sliding scale’ in order to titrate the AVP infusion based on urine output (initially aiming for 3–4 mL/kg/h to compensate for the large amount of infused fluid). The outline of the initial sliding scale we used is attached (Table 1). She was catheterised and her urine output monitored hourly.
Table 1 The AVP (arginine vasopressin) infusion sliding scale initially used.
Urine output, mL/kg/h Action taken
<2 Stop infusion till urine output is 3–4 mL/kg/h
2–3 Reduce infusion rate by 50%
3–4 Maintain infusion rate
4–5 Increase infusion rate by 50%
>5 Increase infusion rate by 100%
Urine output of 3–4 mL/kg/h was targeted.
The original protocol required further development in response to her clinical condition. Initially, we adjusted the AVP infusion rate by 50% depending on the urine output. However, we found that sizeable dose changes caused large swings in her serum sodium and therefore the rate change was reduced to 25%. Furthermore, the target urine output required adjustment based on fluid balance. When the AVP infusion was initially commenced during the first ifosfamide/etoposide cycle, she was hypernatraemic with a large negative fluid balance. Therefore the target urine output was reduced to 2–3 mL/kg/h until sodium levels normalised. With the second ifosamide/etoposide cycle, target urine output of 3–4 mL/kg/h led to rapid decline in sodium levels due to large intravenous fluid intake. Therefore target urine output was increased to 5–6 mL/kg/h to counteract this and maintain a neutral fluid balance.
Outcome and follow-up
During the first ifosfamide cycle, our patient’s plasma sodium showed significant fluctuation (Fig. 2). During the second cycle, AVP infusion was started at the onset and thus tighter control of plasma sodium was achieved (Fig. 3) and she did not experience any related symptoms.
Figure 3 Serum sodium (mmol/L) levels during second chemotherapy cycle involving hyperhydration. The normal serum sodium ranges are 135–145 mmol/L. The timing of blood samples was rounded to the nearest hour.
Following three cycles of chemotherapy, intracranial MRI demonstrated a tiny suprasellar residuum. Visual acuity was 6/45 in the right eye and 6/15 in the left, an improvement from 6/85 and 6/38 prior to starting chemotherapy.
She received proton therapy at University Hospital Essen in Germany, and following this there was further regression of the focal abnormality on her cranial MRI, and further improvement in her visual acuity (6/30 in right eye, 6/15 in left eye).
Both growth hormone (GH) stimulation with clonidine (200 µg) and LHRH testing showed suboptimal responses. She was started on GH and will soon commence pubertal induction with transdermal oestrogen. She continues to be monitored by the endocrinology and oncology teams, and remains well on desmopressin 75 µg/75 µg/100 µg daily, levothyroxine 75 µg daily, hydrocortisone 10 mg/m2 daily in three divided doses and GH 0.7 mg/m2 daily.
Discussion
Our patient had symptoms suggestive of diabetes insipidus for several months prior to admission. She had also developed hypodipsia which potentially precipitated her decompensated diabetes insipidus. The polyuria had appeared to ‘resolve’ just before she presented but its resumption once hydrocortisone was commenced demonstrated that progression to cortisol deficiency had masked her diabetes insipidus. The ‘masking’ of diabetes insipidus is due to impact of cortisol deficiency on AVP dependent and AVP independent mechanisms in the distal convoluting tubules and collecting ducts of the kidney, leading to reduced water clearance (6).
Hyperhydration presents a challenge when patients requiring chemotherapy also have CDI. This challenge was further compounded in our patient due to impaired thirst, which leads to a higher risk of decompensation. We found that by using an AVP infusion during hyperhydration therapy, we achieved better control of our patient’s serum sodium and fluid balance. She also felt less lethargic and was better able to cope with the symptoms relating to chemotherapy and she suffered no adverse effects.
There is a limited literature of case series on the use of AVP infusions in paediatric and adult patients. Wise-Faberowski et al. (7) used a continuous AVP infusion in 18 children with pre- and post-operative DI and compared their findings with 19 historical controls. They were able to maintain the children’s plasma sodium levels within range much more effectively with a continuous AVP infusion. No adverse effects were detected. However, none of these patients required hyperhydration therapy. Levine et al. (8) used an AVP infusion in adult patients with cranial lymphoma who did receive hyperhydration therapy with methotrexate. They initially trialled oral and subcutaneous desmopressin in one patient, and similar to us they found variable fluctuation in plasma sodium levels. Their protocol used a more dilute infusion (0.005 units/mL vs 0.04 units/mL) and a different titration method (titration by 50–100%) compared to ours. They also found it more effective to maintain plasma sodium within range with a continuous AVP infusion and neither of their two patients had side effects. Most pertinent to our case, Bryant et al. (5) trialled an AVP infusion in two children with DI and suprasellar germinoma, and compared with one child in whom desmopressin was withheld, all of whom required the same chemotherapy and hyperhydration regime. Our protocol for preparing an AVP infusion and the starting dose was adapted from theirs. They found that by using an AVP infusion, better control of the fluid and electrolyte balance was achieved. The children also required less fluid (the child who did not receive an AVP infusion required 20 L/m2/day of fluids compared to the other two children who received 3.8 L/m2/day).
Currently, there are no clinical trials in literature looking at continuous AVP infusions in patients with CDI requiring hyperhydration, probably due to rarity of this occurrence. Therefore it seems likely that management of these patients will depend on individual case reports or small case series. Based on our experience and the current literature, in the absence of other effective alternatives we recommend the use of continuous AVP infusions in the management of children with CDI requiring chemotherapy that necessitates hyperhydration. Using a continuous AVP infusion enables a tighter control of plasma sodium levels to be achieved through regulating urine output, and thus avoids the clinical effects that are seen with sharp swings in serum sodium. AVP infusions also have a rapid onset and termination profile and thus rate titrations lead to a rapid clinical effect. AVP infusions have a good safety profile with no adverse effects. Levine et al. (7) recognised hypertension as a possible adverse effect, however, the doses used to manage DI are significantly smaller (<0.5%) than the doses used to manage hypotension in adults.
Based on our experience, we recommend the following steps to be taken when starting a child on continuous AVP infusion for hyperhydration therapy:
Prepare AVP infusion by adding 2 units (0.1 mL from 20 unit/mL ampoule) to 49.9 mL 0.9% saline or 5% dextrose to make up 50 mL syringe. Solution is therefore 0.04 units/mL. If a more concentrated infusion is required, then five units of AVP (0.25 mL) added to 49.75 mL solution to give 0.1 units/mL could be used (5).
Start AVP infusion at the onset of hyperhydration therapy. Omit the usual oral desmopressin doses for the duration of AVP infusion.
HDU (High Dependency Unit) admission is recommended.
Some children may need a separate cannula inserted if the AVP infusion is incompatible with their chemotherapy drugs.
Insert a urinary catheter in all patients to enable hourly monitoring of urine output and fluid balance. Urea and electrolytes should be measured 6 hourly and the child should be weighed daily if possible.
Aim for a urine output of 3–4 mL/kg/h. However, if there is a significantly negative fluid balance as a result, aim for a lower urine output during AVP titrations. Conversely, if there is significantly positive fluid balance, target a higher urine output.
Start the AVP infusion at 0.0001 units/kg/h and alter the rate according to our recommended protocol (Table 2). The size of increment/decrement may need to be bigger or smaller depending on the sensitivity of the patient to desmopressin. We recommend that titration is performed hourly.
Table 2 Recommended protocol using the AVP (arginine vasopressin) Infusion Sliding Scale.
Urine output, mL/kg/h Action taken
<2 Stop infusion till urine output is 3–4 mL/kg/h
2–3 Reduce infusion rate by 25%
3–4 Maintain infusion rate
4–5 Increase infusion rate by 25%
>5 Increase infusion rate by 50%
Aim for urine output 3–4 mL/kg/h, however, this target will be dependent on fluid balance. Calculate hourly urine output to enable titration of AVP to be done hourly. Size of increment/decrement may need to be altered depending on patient’s sensitivity to desmopressin.
Once therapy is complete and serum sodium is within the normal range for the patient, oral desmopression should be recommenced 1 h prior to discontinuing the AVP infusion.
We recognise that admission to a specialist HDU may be challenging in some clinical settings; however, with the intensity of observations and hourly titrations needed, 1:1 nursing care is recommended. We also acknowledge the limitation of extrapolating an AVP infusion sliding scale based on a single patient and limited experience in published literature. Thus we recommend that further experience in using this protocol or alternatives published in the literature would aid in refining the sliding scale and management of patients with CDI undergoing chemotherapeutic regimens requiring hyperhydration.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
Patient consent
A written consent has been obtained from the patient’s parents and are attached to the submission.
Author contribution statement
Dr V Lee is the patient’s Paediatric Oncology Consultant and Prof P Dimitri is her Paediatric Endocrinology Consultant and managed this patient throughout the course of her therapy. J Devaraja is a Paediatric Registrar who had been involved with the endocrinological care of the patient when she was initially admitted, and during her chemotherapy regimens. J Devaraja and S Sloan wrote the manuscript. P Dimitri and V Lee reviewed and contributed to the content of the manuscript. P Dimitri is the senior author. | CISPLATIN, DESMOPRESSIN, DEXAMETHASONE, DEXTROSE, ETOPOSIDE, HYDROCORTISONE, IFOSFAMIDE, SODIUM CHLORIDE, VASOPRESSIN | DrugsGivenReaction | CC BY-NC-ND | 33597311 | 19,607,817 | 2021-02-17 |
What was the administration route of drug 'DESMOPRESSIN'? | Arginine-vasopressin infusion in a child with cranial diabetes insipidus during hyperhydration therapy with chemotherapy: a therapeutic challenge.
An 11-year-old girl presented with acute lower limb weakness, dehydration, hypernatraemia and secondary rhabdomyolysis on a background of an 8-month history of polyuria. Radiological investigations revealed a suprasellar tumour which was diagnosed on biopsy as a non-metastatic germinoma. Further endocrinological investigations confirmed panhypopituitarism and she commenced desmopressin, hydrocortisone and thyroxine. Her chemotherapeutic regime consisted of etoposide, carboplatin and ifosfamide, the latter of which required 4 litres of hyperhydration therapy daily. During the first course of ifosfamide, titration of oral desmopressin was trialled but this resulted in erratic sodium control leading to disorientation. Based on limited literature, we then trialled an arginine-vasopressin (AVP) infusion. A sliding scale was developed to adjust the AVP dose, with an aim to achieve a urine output of 3-4 mL/kg/h. During the second course of ifosamide, AVP infusion was commenced at the outset and tighter control of urine output and sodium levels was achieved. In conclusion, we found that an AVP infusion during hyperhydration therapy was required to achieve eunatraemia in a patient with cranial diabetes insipidus. Developing an AVP sliding scale requires individual variation; further reports/case series are required to underpin practice.
Certain chemotherapeutic regimens require large fluid volumes of hyperhydration therapy which can result in significant complications secondary to rapid serum sodium shifts in patients with diabetes insipidus. The use of a continuous AVP infusion and titrating with a sliding scale is more effective than oral desmopressin in regulating plasma sodium and fluid balance during hyperhydration therapy. No adverse effects were found in our patient using a continuous AVP infusion. Adjustment of the AVP infusion rate depends on urine output, fluid balance, plasma sodium levels and sensitivity/response of the child to titrated AVP doses.
Background
Central diabetes insipidus (CDI) is a disease caused by arginine-vasopressin (AVP) deficiency leading to polyuria and polydipsia (1). The destruction of neurons in the paraventricular and supraoptic nuclei of the posterior pituitary gland leads to AVP deficiency (1). Causes of CDI include germinomas, craniopharyngiomas, trauma, and 20-50% of cases are idiopathic (1).
The initial presentation of a child with a germinoma is often a protracted history of evolving endocrine deficiencies (2). Most initially develop CDI followed by other endocrine deficiencies such as hypocortisolism, hypothyroidism and growth failure. CDI is treated with desmopressin, which is a synthetic analogue of AVP (1). Doses required to manage diuresis vary greatly between patients. The outcome with treatment is favourable with more than 85% overall 5-year survival (2). However, the endocrine deficiencies do not resolve completely with tumour resolution and the children are often on lifelong hormone replacement therapy.
Certain chemotherapeutic agent protocols used in the treatment of suprasellar tumours require large fluid volumes (hyperhydration therapy) to reduce the risk of nephrotoxicity and haemorrhagic cystitis (3). Patients receiving hyperhydration therapy require strict monitoring of fluid balance and electrolytes to avoid complications secondary to sodium and potassium imbalance. Having a diagnosis of CDI while undergoing hyperhydration therapy presents an additional unique challenge. A study by Afzal et al. (4) showed that in children where cisplatin and/or ifosfamide chemotherapy were used (and hyperhydration therapy was therefore required), having CDI was a risk factor for prolonged admissions and complications including seizures, transient encephalopathy, and hyperreflexia with tremor. All children with CDI in their study also required daily changes in dosage and schedule of desmopressin.
In our patient, titration of oral desmopressin was insufficient to manage her fluid and electrolyte balance. We, therefore, used a continuous AVP infusion on a personalised sliding scale to successfully control her fluid balance and thus serum sodium concentration.
Case presentation
We report on an 11-year-old girl who had been previously well. Eight months prior to presentation, she developed polyuria, polydipsia and nocturia. Six months later, she developed frequent headaches. She then presented acutely to the paediatric team at a District General Hospital with generalised weakness and an inability to mobilise. Polyuria and polydipsia had resolved in the 24 h prior to presentation.
Investigation
She was found to have a serum sodium of 173 mmol/L (133–144 mmol/L), serum osmolality of 365 mosm/kg (275–295 mosm/kg), urine osmolality of 356 mosm/kg and creatinine kinase of 1825 units/L (30–170 units/L). A CT Head showed a large suprasellar mass.
She was transferred to our tertiary endocrine unit the following day under the care of the neurosurgery and endocrinology teams. MRI of brain showed a 3 × 3 cm mass involving the optic chiasm, optic nerves and hypothalamus with perilesional oedema (Fig. 1A and B). Biopsy of the mass revealed this to be a non-metastatic germinoma.
Figure 1 A coronal (A) and sagittal (B) MRI image of the germinoma.
Endocrinological bloods demonstrated the following results: cortisol <22 nmol/L at 08:00 h (80–580 nmol/L), ACTH <1.0 ng/L, TSH 3.09 mU/L (0.5–3.6 mU/L) and free T4 6.6 pmol/L (10–16.9 pmol/L) at 18:00 h and IGF-1 82 μg/L at 11:00 h (160–581 μg/L) thus leading to the diagnosis of panhypopituitarism.
An ophthalmology review revealed a severe restriction in visual acuity (right eye 6/85, left eye 6/38), reduced colour vision (right eye 15/17, left eye 16/17), and temporal optic disc pallor.
Treatment
Our patient was started on 0.9% saline at maintenance plus 5% deficit to support rehydration, with urine output also replaced mL for mL with 0.9% saline. A brief course of dexamethasone was commenced to reduce intracranial oedema surrounding the tumour. At this point, the polyuria reoccurred, thereby showing that the ‘initial resolution’ of polyuria was due to cortisol deficiency. She was started on a titrating dose of desmopressin (0–75 µg BD), 50 µg thyroxine and 10 mg/m2/day hydrocortisone in three divided doses. With this treatment, her sodium levels gradually improved over the next 7 days. She was found clinically to have a degree of hypodipsia which persisted after the resolution of the intracranial oedema. Hence her parents were advised to ensure that she took her maintenance fluid requirement daily.
The tumour was not amenable to surgery and she was started on alternating cycles of carboplatin/etoposide (cycles 1 and 3) and ifosfamide/etoposide (cycles 2 and 4). The two ifosfamide/etoposide cycles required 4 L of hyperhydration fluids daily using 0.45% saline/2.5% dextrose with 20 mmol/L potassium chloride. In the first cycle, due to concerns about water intoxication we omitted desmopressin while allowing her to drink freely in addition to hyperhydration therapy. However, this led to a diuresis of 7 L on the first day resulting in lethargy due to polyuria preventing sleep, and she required additional intravenous fluid to replace the fluid deficit. The following day 25–50 µg doses of oral desmopressin given at the beginning of diuresis were trialled, which led to significant swings in her plasma sodium (Fig. 2) causing disorientation.
Figure 2 Serum sodium (mmol/L) levels during first chemotherapy cycle involving hyperhydration. The normal serum sodium ranges are 135–145 mmol/L. The timing of blood samples was rounded to the nearest hour.
Based on a single publication reporting the management of 2 patients with CDI requiring hyperhydration therapy (5) we elected to trial an AVP infusion. A 0.04 units/mL AVP solution was prepared by adding 2 units of AVP (0.1 mL from 20 unit/mL ampoule) to 49.9 mL of 0.9% saline.
A starting dose of 0.0001units/kg/h was used and we created an ‘AVP sliding scale’ in order to titrate the AVP infusion based on urine output (initially aiming for 3–4 mL/kg/h to compensate for the large amount of infused fluid). The outline of the initial sliding scale we used is attached (Table 1). She was catheterised and her urine output monitored hourly.
Table 1 The AVP (arginine vasopressin) infusion sliding scale initially used.
Urine output, mL/kg/h Action taken
<2 Stop infusion till urine output is 3–4 mL/kg/h
2–3 Reduce infusion rate by 50%
3–4 Maintain infusion rate
4–5 Increase infusion rate by 50%
>5 Increase infusion rate by 100%
Urine output of 3–4 mL/kg/h was targeted.
The original protocol required further development in response to her clinical condition. Initially, we adjusted the AVP infusion rate by 50% depending on the urine output. However, we found that sizeable dose changes caused large swings in her serum sodium and therefore the rate change was reduced to 25%. Furthermore, the target urine output required adjustment based on fluid balance. When the AVP infusion was initially commenced during the first ifosfamide/etoposide cycle, she was hypernatraemic with a large negative fluid balance. Therefore the target urine output was reduced to 2–3 mL/kg/h until sodium levels normalised. With the second ifosamide/etoposide cycle, target urine output of 3–4 mL/kg/h led to rapid decline in sodium levels due to large intravenous fluid intake. Therefore target urine output was increased to 5–6 mL/kg/h to counteract this and maintain a neutral fluid balance.
Outcome and follow-up
During the first ifosfamide cycle, our patient’s plasma sodium showed significant fluctuation (Fig. 2). During the second cycle, AVP infusion was started at the onset and thus tighter control of plasma sodium was achieved (Fig. 3) and she did not experience any related symptoms.
Figure 3 Serum sodium (mmol/L) levels during second chemotherapy cycle involving hyperhydration. The normal serum sodium ranges are 135–145 mmol/L. The timing of blood samples was rounded to the nearest hour.
Following three cycles of chemotherapy, intracranial MRI demonstrated a tiny suprasellar residuum. Visual acuity was 6/45 in the right eye and 6/15 in the left, an improvement from 6/85 and 6/38 prior to starting chemotherapy.
She received proton therapy at University Hospital Essen in Germany, and following this there was further regression of the focal abnormality on her cranial MRI, and further improvement in her visual acuity (6/30 in right eye, 6/15 in left eye).
Both growth hormone (GH) stimulation with clonidine (200 µg) and LHRH testing showed suboptimal responses. She was started on GH and will soon commence pubertal induction with transdermal oestrogen. She continues to be monitored by the endocrinology and oncology teams, and remains well on desmopressin 75 µg/75 µg/100 µg daily, levothyroxine 75 µg daily, hydrocortisone 10 mg/m2 daily in three divided doses and GH 0.7 mg/m2 daily.
Discussion
Our patient had symptoms suggestive of diabetes insipidus for several months prior to admission. She had also developed hypodipsia which potentially precipitated her decompensated diabetes insipidus. The polyuria had appeared to ‘resolve’ just before she presented but its resumption once hydrocortisone was commenced demonstrated that progression to cortisol deficiency had masked her diabetes insipidus. The ‘masking’ of diabetes insipidus is due to impact of cortisol deficiency on AVP dependent and AVP independent mechanisms in the distal convoluting tubules and collecting ducts of the kidney, leading to reduced water clearance (6).
Hyperhydration presents a challenge when patients requiring chemotherapy also have CDI. This challenge was further compounded in our patient due to impaired thirst, which leads to a higher risk of decompensation. We found that by using an AVP infusion during hyperhydration therapy, we achieved better control of our patient’s serum sodium and fluid balance. She also felt less lethargic and was better able to cope with the symptoms relating to chemotherapy and she suffered no adverse effects.
There is a limited literature of case series on the use of AVP infusions in paediatric and adult patients. Wise-Faberowski et al. (7) used a continuous AVP infusion in 18 children with pre- and post-operative DI and compared their findings with 19 historical controls. They were able to maintain the children’s plasma sodium levels within range much more effectively with a continuous AVP infusion. No adverse effects were detected. However, none of these patients required hyperhydration therapy. Levine et al. (8) used an AVP infusion in adult patients with cranial lymphoma who did receive hyperhydration therapy with methotrexate. They initially trialled oral and subcutaneous desmopressin in one patient, and similar to us they found variable fluctuation in plasma sodium levels. Their protocol used a more dilute infusion (0.005 units/mL vs 0.04 units/mL) and a different titration method (titration by 50–100%) compared to ours. They also found it more effective to maintain plasma sodium within range with a continuous AVP infusion and neither of their two patients had side effects. Most pertinent to our case, Bryant et al. (5) trialled an AVP infusion in two children with DI and suprasellar germinoma, and compared with one child in whom desmopressin was withheld, all of whom required the same chemotherapy and hyperhydration regime. Our protocol for preparing an AVP infusion and the starting dose was adapted from theirs. They found that by using an AVP infusion, better control of the fluid and electrolyte balance was achieved. The children also required less fluid (the child who did not receive an AVP infusion required 20 L/m2/day of fluids compared to the other two children who received 3.8 L/m2/day).
Currently, there are no clinical trials in literature looking at continuous AVP infusions in patients with CDI requiring hyperhydration, probably due to rarity of this occurrence. Therefore it seems likely that management of these patients will depend on individual case reports or small case series. Based on our experience and the current literature, in the absence of other effective alternatives we recommend the use of continuous AVP infusions in the management of children with CDI requiring chemotherapy that necessitates hyperhydration. Using a continuous AVP infusion enables a tighter control of plasma sodium levels to be achieved through regulating urine output, and thus avoids the clinical effects that are seen with sharp swings in serum sodium. AVP infusions also have a rapid onset and termination profile and thus rate titrations lead to a rapid clinical effect. AVP infusions have a good safety profile with no adverse effects. Levine et al. (7) recognised hypertension as a possible adverse effect, however, the doses used to manage DI are significantly smaller (<0.5%) than the doses used to manage hypotension in adults.
Based on our experience, we recommend the following steps to be taken when starting a child on continuous AVP infusion for hyperhydration therapy:
Prepare AVP infusion by adding 2 units (0.1 mL from 20 unit/mL ampoule) to 49.9 mL 0.9% saline or 5% dextrose to make up 50 mL syringe. Solution is therefore 0.04 units/mL. If a more concentrated infusion is required, then five units of AVP (0.25 mL) added to 49.75 mL solution to give 0.1 units/mL could be used (5).
Start AVP infusion at the onset of hyperhydration therapy. Omit the usual oral desmopressin doses for the duration of AVP infusion.
HDU (High Dependency Unit) admission is recommended.
Some children may need a separate cannula inserted if the AVP infusion is incompatible with their chemotherapy drugs.
Insert a urinary catheter in all patients to enable hourly monitoring of urine output and fluid balance. Urea and electrolytes should be measured 6 hourly and the child should be weighed daily if possible.
Aim for a urine output of 3–4 mL/kg/h. However, if there is a significantly negative fluid balance as a result, aim for a lower urine output during AVP titrations. Conversely, if there is significantly positive fluid balance, target a higher urine output.
Start the AVP infusion at 0.0001 units/kg/h and alter the rate according to our recommended protocol (Table 2). The size of increment/decrement may need to be bigger or smaller depending on the sensitivity of the patient to desmopressin. We recommend that titration is performed hourly.
Table 2 Recommended protocol using the AVP (arginine vasopressin) Infusion Sliding Scale.
Urine output, mL/kg/h Action taken
<2 Stop infusion till urine output is 3–4 mL/kg/h
2–3 Reduce infusion rate by 25%
3–4 Maintain infusion rate
4–5 Increase infusion rate by 25%
>5 Increase infusion rate by 50%
Aim for urine output 3–4 mL/kg/h, however, this target will be dependent on fluid balance. Calculate hourly urine output to enable titration of AVP to be done hourly. Size of increment/decrement may need to be altered depending on patient’s sensitivity to desmopressin.
Once therapy is complete and serum sodium is within the normal range for the patient, oral desmopression should be recommenced 1 h prior to discontinuing the AVP infusion.
We recognise that admission to a specialist HDU may be challenging in some clinical settings; however, with the intensity of observations and hourly titrations needed, 1:1 nursing care is recommended. We also acknowledge the limitation of extrapolating an AVP infusion sliding scale based on a single patient and limited experience in published literature. Thus we recommend that further experience in using this protocol or alternatives published in the literature would aid in refining the sliding scale and management of patients with CDI undergoing chemotherapeutic regimens requiring hyperhydration.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
Patient consent
A written consent has been obtained from the patient’s parents and are attached to the submission.
Author contribution statement
Dr V Lee is the patient’s Paediatric Oncology Consultant and Prof P Dimitri is her Paediatric Endocrinology Consultant and managed this patient throughout the course of her therapy. J Devaraja is a Paediatric Registrar who had been involved with the endocrinological care of the patient when she was initially admitted, and during her chemotherapy regimens. J Devaraja and S Sloan wrote the manuscript. P Dimitri and V Lee reviewed and contributed to the content of the manuscript. P Dimitri is the senior author. | Oral | DrugAdministrationRoute | CC BY-NC-ND | 33597311 | 19,607,817 | 2021-02-17 |
What was the administration route of drug 'VASOPRESSIN'? | Arginine-vasopressin infusion in a child with cranial diabetes insipidus during hyperhydration therapy with chemotherapy: a therapeutic challenge.
An 11-year-old girl presented with acute lower limb weakness, dehydration, hypernatraemia and secondary rhabdomyolysis on a background of an 8-month history of polyuria. Radiological investigations revealed a suprasellar tumour which was diagnosed on biopsy as a non-metastatic germinoma. Further endocrinological investigations confirmed panhypopituitarism and she commenced desmopressin, hydrocortisone and thyroxine. Her chemotherapeutic regime consisted of etoposide, carboplatin and ifosfamide, the latter of which required 4 litres of hyperhydration therapy daily. During the first course of ifosfamide, titration of oral desmopressin was trialled but this resulted in erratic sodium control leading to disorientation. Based on limited literature, we then trialled an arginine-vasopressin (AVP) infusion. A sliding scale was developed to adjust the AVP dose, with an aim to achieve a urine output of 3-4 mL/kg/h. During the second course of ifosamide, AVP infusion was commenced at the outset and tighter control of urine output and sodium levels was achieved. In conclusion, we found that an AVP infusion during hyperhydration therapy was required to achieve eunatraemia in a patient with cranial diabetes insipidus. Developing an AVP sliding scale requires individual variation; further reports/case series are required to underpin practice.
Certain chemotherapeutic regimens require large fluid volumes of hyperhydration therapy which can result in significant complications secondary to rapid serum sodium shifts in patients with diabetes insipidus. The use of a continuous AVP infusion and titrating with a sliding scale is more effective than oral desmopressin in regulating plasma sodium and fluid balance during hyperhydration therapy. No adverse effects were found in our patient using a continuous AVP infusion. Adjustment of the AVP infusion rate depends on urine output, fluid balance, plasma sodium levels and sensitivity/response of the child to titrated AVP doses.
Background
Central diabetes insipidus (CDI) is a disease caused by arginine-vasopressin (AVP) deficiency leading to polyuria and polydipsia (1). The destruction of neurons in the paraventricular and supraoptic nuclei of the posterior pituitary gland leads to AVP deficiency (1). Causes of CDI include germinomas, craniopharyngiomas, trauma, and 20-50% of cases are idiopathic (1).
The initial presentation of a child with a germinoma is often a protracted history of evolving endocrine deficiencies (2). Most initially develop CDI followed by other endocrine deficiencies such as hypocortisolism, hypothyroidism and growth failure. CDI is treated with desmopressin, which is a synthetic analogue of AVP (1). Doses required to manage diuresis vary greatly between patients. The outcome with treatment is favourable with more than 85% overall 5-year survival (2). However, the endocrine deficiencies do not resolve completely with tumour resolution and the children are often on lifelong hormone replacement therapy.
Certain chemotherapeutic agent protocols used in the treatment of suprasellar tumours require large fluid volumes (hyperhydration therapy) to reduce the risk of nephrotoxicity and haemorrhagic cystitis (3). Patients receiving hyperhydration therapy require strict monitoring of fluid balance and electrolytes to avoid complications secondary to sodium and potassium imbalance. Having a diagnosis of CDI while undergoing hyperhydration therapy presents an additional unique challenge. A study by Afzal et al. (4) showed that in children where cisplatin and/or ifosfamide chemotherapy were used (and hyperhydration therapy was therefore required), having CDI was a risk factor for prolonged admissions and complications including seizures, transient encephalopathy, and hyperreflexia with tremor. All children with CDI in their study also required daily changes in dosage and schedule of desmopressin.
In our patient, titration of oral desmopressin was insufficient to manage her fluid and electrolyte balance. We, therefore, used a continuous AVP infusion on a personalised sliding scale to successfully control her fluid balance and thus serum sodium concentration.
Case presentation
We report on an 11-year-old girl who had been previously well. Eight months prior to presentation, she developed polyuria, polydipsia and nocturia. Six months later, she developed frequent headaches. She then presented acutely to the paediatric team at a District General Hospital with generalised weakness and an inability to mobilise. Polyuria and polydipsia had resolved in the 24 h prior to presentation.
Investigation
She was found to have a serum sodium of 173 mmol/L (133–144 mmol/L), serum osmolality of 365 mosm/kg (275–295 mosm/kg), urine osmolality of 356 mosm/kg and creatinine kinase of 1825 units/L (30–170 units/L). A CT Head showed a large suprasellar mass.
She was transferred to our tertiary endocrine unit the following day under the care of the neurosurgery and endocrinology teams. MRI of brain showed a 3 × 3 cm mass involving the optic chiasm, optic nerves and hypothalamus with perilesional oedema (Fig. 1A and B). Biopsy of the mass revealed this to be a non-metastatic germinoma.
Figure 1 A coronal (A) and sagittal (B) MRI image of the germinoma.
Endocrinological bloods demonstrated the following results: cortisol <22 nmol/L at 08:00 h (80–580 nmol/L), ACTH <1.0 ng/L, TSH 3.09 mU/L (0.5–3.6 mU/L) and free T4 6.6 pmol/L (10–16.9 pmol/L) at 18:00 h and IGF-1 82 μg/L at 11:00 h (160–581 μg/L) thus leading to the diagnosis of panhypopituitarism.
An ophthalmology review revealed a severe restriction in visual acuity (right eye 6/85, left eye 6/38), reduced colour vision (right eye 15/17, left eye 16/17), and temporal optic disc pallor.
Treatment
Our patient was started on 0.9% saline at maintenance plus 5% deficit to support rehydration, with urine output also replaced mL for mL with 0.9% saline. A brief course of dexamethasone was commenced to reduce intracranial oedema surrounding the tumour. At this point, the polyuria reoccurred, thereby showing that the ‘initial resolution’ of polyuria was due to cortisol deficiency. She was started on a titrating dose of desmopressin (0–75 µg BD), 50 µg thyroxine and 10 mg/m2/day hydrocortisone in three divided doses. With this treatment, her sodium levels gradually improved over the next 7 days. She was found clinically to have a degree of hypodipsia which persisted after the resolution of the intracranial oedema. Hence her parents were advised to ensure that she took her maintenance fluid requirement daily.
The tumour was not amenable to surgery and she was started on alternating cycles of carboplatin/etoposide (cycles 1 and 3) and ifosfamide/etoposide (cycles 2 and 4). The two ifosfamide/etoposide cycles required 4 L of hyperhydration fluids daily using 0.45% saline/2.5% dextrose with 20 mmol/L potassium chloride. In the first cycle, due to concerns about water intoxication we omitted desmopressin while allowing her to drink freely in addition to hyperhydration therapy. However, this led to a diuresis of 7 L on the first day resulting in lethargy due to polyuria preventing sleep, and she required additional intravenous fluid to replace the fluid deficit. The following day 25–50 µg doses of oral desmopressin given at the beginning of diuresis were trialled, which led to significant swings in her plasma sodium (Fig. 2) causing disorientation.
Figure 2 Serum sodium (mmol/L) levels during first chemotherapy cycle involving hyperhydration. The normal serum sodium ranges are 135–145 mmol/L. The timing of blood samples was rounded to the nearest hour.
Based on a single publication reporting the management of 2 patients with CDI requiring hyperhydration therapy (5) we elected to trial an AVP infusion. A 0.04 units/mL AVP solution was prepared by adding 2 units of AVP (0.1 mL from 20 unit/mL ampoule) to 49.9 mL of 0.9% saline.
A starting dose of 0.0001units/kg/h was used and we created an ‘AVP sliding scale’ in order to titrate the AVP infusion based on urine output (initially aiming for 3–4 mL/kg/h to compensate for the large amount of infused fluid). The outline of the initial sliding scale we used is attached (Table 1). She was catheterised and her urine output monitored hourly.
Table 1 The AVP (arginine vasopressin) infusion sliding scale initially used.
Urine output, mL/kg/h Action taken
<2 Stop infusion till urine output is 3–4 mL/kg/h
2–3 Reduce infusion rate by 50%
3–4 Maintain infusion rate
4–5 Increase infusion rate by 50%
>5 Increase infusion rate by 100%
Urine output of 3–4 mL/kg/h was targeted.
The original protocol required further development in response to her clinical condition. Initially, we adjusted the AVP infusion rate by 50% depending on the urine output. However, we found that sizeable dose changes caused large swings in her serum sodium and therefore the rate change was reduced to 25%. Furthermore, the target urine output required adjustment based on fluid balance. When the AVP infusion was initially commenced during the first ifosfamide/etoposide cycle, she was hypernatraemic with a large negative fluid balance. Therefore the target urine output was reduced to 2–3 mL/kg/h until sodium levels normalised. With the second ifosamide/etoposide cycle, target urine output of 3–4 mL/kg/h led to rapid decline in sodium levels due to large intravenous fluid intake. Therefore target urine output was increased to 5–6 mL/kg/h to counteract this and maintain a neutral fluid balance.
Outcome and follow-up
During the first ifosfamide cycle, our patient’s plasma sodium showed significant fluctuation (Fig. 2). During the second cycle, AVP infusion was started at the onset and thus tighter control of plasma sodium was achieved (Fig. 3) and she did not experience any related symptoms.
Figure 3 Serum sodium (mmol/L) levels during second chemotherapy cycle involving hyperhydration. The normal serum sodium ranges are 135–145 mmol/L. The timing of blood samples was rounded to the nearest hour.
Following three cycles of chemotherapy, intracranial MRI demonstrated a tiny suprasellar residuum. Visual acuity was 6/45 in the right eye and 6/15 in the left, an improvement from 6/85 and 6/38 prior to starting chemotherapy.
She received proton therapy at University Hospital Essen in Germany, and following this there was further regression of the focal abnormality on her cranial MRI, and further improvement in her visual acuity (6/30 in right eye, 6/15 in left eye).
Both growth hormone (GH) stimulation with clonidine (200 µg) and LHRH testing showed suboptimal responses. She was started on GH and will soon commence pubertal induction with transdermal oestrogen. She continues to be monitored by the endocrinology and oncology teams, and remains well on desmopressin 75 µg/75 µg/100 µg daily, levothyroxine 75 µg daily, hydrocortisone 10 mg/m2 daily in three divided doses and GH 0.7 mg/m2 daily.
Discussion
Our patient had symptoms suggestive of diabetes insipidus for several months prior to admission. She had also developed hypodipsia which potentially precipitated her decompensated diabetes insipidus. The polyuria had appeared to ‘resolve’ just before she presented but its resumption once hydrocortisone was commenced demonstrated that progression to cortisol deficiency had masked her diabetes insipidus. The ‘masking’ of diabetes insipidus is due to impact of cortisol deficiency on AVP dependent and AVP independent mechanisms in the distal convoluting tubules and collecting ducts of the kidney, leading to reduced water clearance (6).
Hyperhydration presents a challenge when patients requiring chemotherapy also have CDI. This challenge was further compounded in our patient due to impaired thirst, which leads to a higher risk of decompensation. We found that by using an AVP infusion during hyperhydration therapy, we achieved better control of our patient’s serum sodium and fluid balance. She also felt less lethargic and was better able to cope with the symptoms relating to chemotherapy and she suffered no adverse effects.
There is a limited literature of case series on the use of AVP infusions in paediatric and adult patients. Wise-Faberowski et al. (7) used a continuous AVP infusion in 18 children with pre- and post-operative DI and compared their findings with 19 historical controls. They were able to maintain the children’s plasma sodium levels within range much more effectively with a continuous AVP infusion. No adverse effects were detected. However, none of these patients required hyperhydration therapy. Levine et al. (8) used an AVP infusion in adult patients with cranial lymphoma who did receive hyperhydration therapy with methotrexate. They initially trialled oral and subcutaneous desmopressin in one patient, and similar to us they found variable fluctuation in plasma sodium levels. Their protocol used a more dilute infusion (0.005 units/mL vs 0.04 units/mL) and a different titration method (titration by 50–100%) compared to ours. They also found it more effective to maintain plasma sodium within range with a continuous AVP infusion and neither of their two patients had side effects. Most pertinent to our case, Bryant et al. (5) trialled an AVP infusion in two children with DI and suprasellar germinoma, and compared with one child in whom desmopressin was withheld, all of whom required the same chemotherapy and hyperhydration regime. Our protocol for preparing an AVP infusion and the starting dose was adapted from theirs. They found that by using an AVP infusion, better control of the fluid and electrolyte balance was achieved. The children also required less fluid (the child who did not receive an AVP infusion required 20 L/m2/day of fluids compared to the other two children who received 3.8 L/m2/day).
Currently, there are no clinical trials in literature looking at continuous AVP infusions in patients with CDI requiring hyperhydration, probably due to rarity of this occurrence. Therefore it seems likely that management of these patients will depend on individual case reports or small case series. Based on our experience and the current literature, in the absence of other effective alternatives we recommend the use of continuous AVP infusions in the management of children with CDI requiring chemotherapy that necessitates hyperhydration. Using a continuous AVP infusion enables a tighter control of plasma sodium levels to be achieved through regulating urine output, and thus avoids the clinical effects that are seen with sharp swings in serum sodium. AVP infusions also have a rapid onset and termination profile and thus rate titrations lead to a rapid clinical effect. AVP infusions have a good safety profile with no adverse effects. Levine et al. (7) recognised hypertension as a possible adverse effect, however, the doses used to manage DI are significantly smaller (<0.5%) than the doses used to manage hypotension in adults.
Based on our experience, we recommend the following steps to be taken when starting a child on continuous AVP infusion for hyperhydration therapy:
Prepare AVP infusion by adding 2 units (0.1 mL from 20 unit/mL ampoule) to 49.9 mL 0.9% saline or 5% dextrose to make up 50 mL syringe. Solution is therefore 0.04 units/mL. If a more concentrated infusion is required, then five units of AVP (0.25 mL) added to 49.75 mL solution to give 0.1 units/mL could be used (5).
Start AVP infusion at the onset of hyperhydration therapy. Omit the usual oral desmopressin doses for the duration of AVP infusion.
HDU (High Dependency Unit) admission is recommended.
Some children may need a separate cannula inserted if the AVP infusion is incompatible with their chemotherapy drugs.
Insert a urinary catheter in all patients to enable hourly monitoring of urine output and fluid balance. Urea and electrolytes should be measured 6 hourly and the child should be weighed daily if possible.
Aim for a urine output of 3–4 mL/kg/h. However, if there is a significantly negative fluid balance as a result, aim for a lower urine output during AVP titrations. Conversely, if there is significantly positive fluid balance, target a higher urine output.
Start the AVP infusion at 0.0001 units/kg/h and alter the rate according to our recommended protocol (Table 2). The size of increment/decrement may need to be bigger or smaller depending on the sensitivity of the patient to desmopressin. We recommend that titration is performed hourly.
Table 2 Recommended protocol using the AVP (arginine vasopressin) Infusion Sliding Scale.
Urine output, mL/kg/h Action taken
<2 Stop infusion till urine output is 3–4 mL/kg/h
2–3 Reduce infusion rate by 25%
3–4 Maintain infusion rate
4–5 Increase infusion rate by 25%
>5 Increase infusion rate by 50%
Aim for urine output 3–4 mL/kg/h, however, this target will be dependent on fluid balance. Calculate hourly urine output to enable titration of AVP to be done hourly. Size of increment/decrement may need to be altered depending on patient’s sensitivity to desmopressin.
Once therapy is complete and serum sodium is within the normal range for the patient, oral desmopression should be recommenced 1 h prior to discontinuing the AVP infusion.
We recognise that admission to a specialist HDU may be challenging in some clinical settings; however, with the intensity of observations and hourly titrations needed, 1:1 nursing care is recommended. We also acknowledge the limitation of extrapolating an AVP infusion sliding scale based on a single patient and limited experience in published literature. Thus we recommend that further experience in using this protocol or alternatives published in the literature would aid in refining the sliding scale and management of patients with CDI undergoing chemotherapeutic regimens requiring hyperhydration.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
Patient consent
A written consent has been obtained from the patient’s parents and are attached to the submission.
Author contribution statement
Dr V Lee is the patient’s Paediatric Oncology Consultant and Prof P Dimitri is her Paediatric Endocrinology Consultant and managed this patient throughout the course of her therapy. J Devaraja is a Paediatric Registrar who had been involved with the endocrinological care of the patient when she was initially admitted, and during her chemotherapy regimens. J Devaraja and S Sloan wrote the manuscript. P Dimitri and V Lee reviewed and contributed to the content of the manuscript. P Dimitri is the senior author. | Other | DrugAdministrationRoute | CC BY-NC-ND | 33597311 | 19,607,817 | 2021-02-17 |
What was the dosage of drug 'VASOPRESSIN'? | Arginine-vasopressin infusion in a child with cranial diabetes insipidus during hyperhydration therapy with chemotherapy: a therapeutic challenge.
An 11-year-old girl presented with acute lower limb weakness, dehydration, hypernatraemia and secondary rhabdomyolysis on a background of an 8-month history of polyuria. Radiological investigations revealed a suprasellar tumour which was diagnosed on biopsy as a non-metastatic germinoma. Further endocrinological investigations confirmed panhypopituitarism and she commenced desmopressin, hydrocortisone and thyroxine. Her chemotherapeutic regime consisted of etoposide, carboplatin and ifosfamide, the latter of which required 4 litres of hyperhydration therapy daily. During the first course of ifosfamide, titration of oral desmopressin was trialled but this resulted in erratic sodium control leading to disorientation. Based on limited literature, we then trialled an arginine-vasopressin (AVP) infusion. A sliding scale was developed to adjust the AVP dose, with an aim to achieve a urine output of 3-4 mL/kg/h. During the second course of ifosamide, AVP infusion was commenced at the outset and tighter control of urine output and sodium levels was achieved. In conclusion, we found that an AVP infusion during hyperhydration therapy was required to achieve eunatraemia in a patient with cranial diabetes insipidus. Developing an AVP sliding scale requires individual variation; further reports/case series are required to underpin practice.
Certain chemotherapeutic regimens require large fluid volumes of hyperhydration therapy which can result in significant complications secondary to rapid serum sodium shifts in patients with diabetes insipidus. The use of a continuous AVP infusion and titrating with a sliding scale is more effective than oral desmopressin in regulating plasma sodium and fluid balance during hyperhydration therapy. No adverse effects were found in our patient using a continuous AVP infusion. Adjustment of the AVP infusion rate depends on urine output, fluid balance, plasma sodium levels and sensitivity/response of the child to titrated AVP doses.
Background
Central diabetes insipidus (CDI) is a disease caused by arginine-vasopressin (AVP) deficiency leading to polyuria and polydipsia (1). The destruction of neurons in the paraventricular and supraoptic nuclei of the posterior pituitary gland leads to AVP deficiency (1). Causes of CDI include germinomas, craniopharyngiomas, trauma, and 20-50% of cases are idiopathic (1).
The initial presentation of a child with a germinoma is often a protracted history of evolving endocrine deficiencies (2). Most initially develop CDI followed by other endocrine deficiencies such as hypocortisolism, hypothyroidism and growth failure. CDI is treated with desmopressin, which is a synthetic analogue of AVP (1). Doses required to manage diuresis vary greatly between patients. The outcome with treatment is favourable with more than 85% overall 5-year survival (2). However, the endocrine deficiencies do not resolve completely with tumour resolution and the children are often on lifelong hormone replacement therapy.
Certain chemotherapeutic agent protocols used in the treatment of suprasellar tumours require large fluid volumes (hyperhydration therapy) to reduce the risk of nephrotoxicity and haemorrhagic cystitis (3). Patients receiving hyperhydration therapy require strict monitoring of fluid balance and electrolytes to avoid complications secondary to sodium and potassium imbalance. Having a diagnosis of CDI while undergoing hyperhydration therapy presents an additional unique challenge. A study by Afzal et al. (4) showed that in children where cisplatin and/or ifosfamide chemotherapy were used (and hyperhydration therapy was therefore required), having CDI was a risk factor for prolonged admissions and complications including seizures, transient encephalopathy, and hyperreflexia with tremor. All children with CDI in their study also required daily changes in dosage and schedule of desmopressin.
In our patient, titration of oral desmopressin was insufficient to manage her fluid and electrolyte balance. We, therefore, used a continuous AVP infusion on a personalised sliding scale to successfully control her fluid balance and thus serum sodium concentration.
Case presentation
We report on an 11-year-old girl who had been previously well. Eight months prior to presentation, she developed polyuria, polydipsia and nocturia. Six months later, she developed frequent headaches. She then presented acutely to the paediatric team at a District General Hospital with generalised weakness and an inability to mobilise. Polyuria and polydipsia had resolved in the 24 h prior to presentation.
Investigation
She was found to have a serum sodium of 173 mmol/L (133–144 mmol/L), serum osmolality of 365 mosm/kg (275–295 mosm/kg), urine osmolality of 356 mosm/kg and creatinine kinase of 1825 units/L (30–170 units/L). A CT Head showed a large suprasellar mass.
She was transferred to our tertiary endocrine unit the following day under the care of the neurosurgery and endocrinology teams. MRI of brain showed a 3 × 3 cm mass involving the optic chiasm, optic nerves and hypothalamus with perilesional oedema (Fig. 1A and B). Biopsy of the mass revealed this to be a non-metastatic germinoma.
Figure 1 A coronal (A) and sagittal (B) MRI image of the germinoma.
Endocrinological bloods demonstrated the following results: cortisol <22 nmol/L at 08:00 h (80–580 nmol/L), ACTH <1.0 ng/L, TSH 3.09 mU/L (0.5–3.6 mU/L) and free T4 6.6 pmol/L (10–16.9 pmol/L) at 18:00 h and IGF-1 82 μg/L at 11:00 h (160–581 μg/L) thus leading to the diagnosis of panhypopituitarism.
An ophthalmology review revealed a severe restriction in visual acuity (right eye 6/85, left eye 6/38), reduced colour vision (right eye 15/17, left eye 16/17), and temporal optic disc pallor.
Treatment
Our patient was started on 0.9% saline at maintenance plus 5% deficit to support rehydration, with urine output also replaced mL for mL with 0.9% saline. A brief course of dexamethasone was commenced to reduce intracranial oedema surrounding the tumour. At this point, the polyuria reoccurred, thereby showing that the ‘initial resolution’ of polyuria was due to cortisol deficiency. She was started on a titrating dose of desmopressin (0–75 µg BD), 50 µg thyroxine and 10 mg/m2/day hydrocortisone in three divided doses. With this treatment, her sodium levels gradually improved over the next 7 days. She was found clinically to have a degree of hypodipsia which persisted after the resolution of the intracranial oedema. Hence her parents were advised to ensure that she took her maintenance fluid requirement daily.
The tumour was not amenable to surgery and she was started on alternating cycles of carboplatin/etoposide (cycles 1 and 3) and ifosfamide/etoposide (cycles 2 and 4). The two ifosfamide/etoposide cycles required 4 L of hyperhydration fluids daily using 0.45% saline/2.5% dextrose with 20 mmol/L potassium chloride. In the first cycle, due to concerns about water intoxication we omitted desmopressin while allowing her to drink freely in addition to hyperhydration therapy. However, this led to a diuresis of 7 L on the first day resulting in lethargy due to polyuria preventing sleep, and she required additional intravenous fluid to replace the fluid deficit. The following day 25–50 µg doses of oral desmopressin given at the beginning of diuresis were trialled, which led to significant swings in her plasma sodium (Fig. 2) causing disorientation.
Figure 2 Serum sodium (mmol/L) levels during first chemotherapy cycle involving hyperhydration. The normal serum sodium ranges are 135–145 mmol/L. The timing of blood samples was rounded to the nearest hour.
Based on a single publication reporting the management of 2 patients with CDI requiring hyperhydration therapy (5) we elected to trial an AVP infusion. A 0.04 units/mL AVP solution was prepared by adding 2 units of AVP (0.1 mL from 20 unit/mL ampoule) to 49.9 mL of 0.9% saline.
A starting dose of 0.0001units/kg/h was used and we created an ‘AVP sliding scale’ in order to titrate the AVP infusion based on urine output (initially aiming for 3–4 mL/kg/h to compensate for the large amount of infused fluid). The outline of the initial sliding scale we used is attached (Table 1). She was catheterised and her urine output monitored hourly.
Table 1 The AVP (arginine vasopressin) infusion sliding scale initially used.
Urine output, mL/kg/h Action taken
<2 Stop infusion till urine output is 3–4 mL/kg/h
2–3 Reduce infusion rate by 50%
3–4 Maintain infusion rate
4–5 Increase infusion rate by 50%
>5 Increase infusion rate by 100%
Urine output of 3–4 mL/kg/h was targeted.
The original protocol required further development in response to her clinical condition. Initially, we adjusted the AVP infusion rate by 50% depending on the urine output. However, we found that sizeable dose changes caused large swings in her serum sodium and therefore the rate change was reduced to 25%. Furthermore, the target urine output required adjustment based on fluid balance. When the AVP infusion was initially commenced during the first ifosfamide/etoposide cycle, she was hypernatraemic with a large negative fluid balance. Therefore the target urine output was reduced to 2–3 mL/kg/h until sodium levels normalised. With the second ifosamide/etoposide cycle, target urine output of 3–4 mL/kg/h led to rapid decline in sodium levels due to large intravenous fluid intake. Therefore target urine output was increased to 5–6 mL/kg/h to counteract this and maintain a neutral fluid balance.
Outcome and follow-up
During the first ifosfamide cycle, our patient’s plasma sodium showed significant fluctuation (Fig. 2). During the second cycle, AVP infusion was started at the onset and thus tighter control of plasma sodium was achieved (Fig. 3) and she did not experience any related symptoms.
Figure 3 Serum sodium (mmol/L) levels during second chemotherapy cycle involving hyperhydration. The normal serum sodium ranges are 135–145 mmol/L. The timing of blood samples was rounded to the nearest hour.
Following three cycles of chemotherapy, intracranial MRI demonstrated a tiny suprasellar residuum. Visual acuity was 6/45 in the right eye and 6/15 in the left, an improvement from 6/85 and 6/38 prior to starting chemotherapy.
She received proton therapy at University Hospital Essen in Germany, and following this there was further regression of the focal abnormality on her cranial MRI, and further improvement in her visual acuity (6/30 in right eye, 6/15 in left eye).
Both growth hormone (GH) stimulation with clonidine (200 µg) and LHRH testing showed suboptimal responses. She was started on GH and will soon commence pubertal induction with transdermal oestrogen. She continues to be monitored by the endocrinology and oncology teams, and remains well on desmopressin 75 µg/75 µg/100 µg daily, levothyroxine 75 µg daily, hydrocortisone 10 mg/m2 daily in three divided doses and GH 0.7 mg/m2 daily.
Discussion
Our patient had symptoms suggestive of diabetes insipidus for several months prior to admission. She had also developed hypodipsia which potentially precipitated her decompensated diabetes insipidus. The polyuria had appeared to ‘resolve’ just before she presented but its resumption once hydrocortisone was commenced demonstrated that progression to cortisol deficiency had masked her diabetes insipidus. The ‘masking’ of diabetes insipidus is due to impact of cortisol deficiency on AVP dependent and AVP independent mechanisms in the distal convoluting tubules and collecting ducts of the kidney, leading to reduced water clearance (6).
Hyperhydration presents a challenge when patients requiring chemotherapy also have CDI. This challenge was further compounded in our patient due to impaired thirst, which leads to a higher risk of decompensation. We found that by using an AVP infusion during hyperhydration therapy, we achieved better control of our patient’s serum sodium and fluid balance. She also felt less lethargic and was better able to cope with the symptoms relating to chemotherapy and she suffered no adverse effects.
There is a limited literature of case series on the use of AVP infusions in paediatric and adult patients. Wise-Faberowski et al. (7) used a continuous AVP infusion in 18 children with pre- and post-operative DI and compared their findings with 19 historical controls. They were able to maintain the children’s plasma sodium levels within range much more effectively with a continuous AVP infusion. No adverse effects were detected. However, none of these patients required hyperhydration therapy. Levine et al. (8) used an AVP infusion in adult patients with cranial lymphoma who did receive hyperhydration therapy with methotrexate. They initially trialled oral and subcutaneous desmopressin in one patient, and similar to us they found variable fluctuation in plasma sodium levels. Their protocol used a more dilute infusion (0.005 units/mL vs 0.04 units/mL) and a different titration method (titration by 50–100%) compared to ours. They also found it more effective to maintain plasma sodium within range with a continuous AVP infusion and neither of their two patients had side effects. Most pertinent to our case, Bryant et al. (5) trialled an AVP infusion in two children with DI and suprasellar germinoma, and compared with one child in whom desmopressin was withheld, all of whom required the same chemotherapy and hyperhydration regime. Our protocol for preparing an AVP infusion and the starting dose was adapted from theirs. They found that by using an AVP infusion, better control of the fluid and electrolyte balance was achieved. The children also required less fluid (the child who did not receive an AVP infusion required 20 L/m2/day of fluids compared to the other two children who received 3.8 L/m2/day).
Currently, there are no clinical trials in literature looking at continuous AVP infusions in patients with CDI requiring hyperhydration, probably due to rarity of this occurrence. Therefore it seems likely that management of these patients will depend on individual case reports or small case series. Based on our experience and the current literature, in the absence of other effective alternatives we recommend the use of continuous AVP infusions in the management of children with CDI requiring chemotherapy that necessitates hyperhydration. Using a continuous AVP infusion enables a tighter control of plasma sodium levels to be achieved through regulating urine output, and thus avoids the clinical effects that are seen with sharp swings in serum sodium. AVP infusions also have a rapid onset and termination profile and thus rate titrations lead to a rapid clinical effect. AVP infusions have a good safety profile with no adverse effects. Levine et al. (7) recognised hypertension as a possible adverse effect, however, the doses used to manage DI are significantly smaller (<0.5%) than the doses used to manage hypotension in adults.
Based on our experience, we recommend the following steps to be taken when starting a child on continuous AVP infusion for hyperhydration therapy:
Prepare AVP infusion by adding 2 units (0.1 mL from 20 unit/mL ampoule) to 49.9 mL 0.9% saline or 5% dextrose to make up 50 mL syringe. Solution is therefore 0.04 units/mL. If a more concentrated infusion is required, then five units of AVP (0.25 mL) added to 49.75 mL solution to give 0.1 units/mL could be used (5).
Start AVP infusion at the onset of hyperhydration therapy. Omit the usual oral desmopressin doses for the duration of AVP infusion.
HDU (High Dependency Unit) admission is recommended.
Some children may need a separate cannula inserted if the AVP infusion is incompatible with their chemotherapy drugs.
Insert a urinary catheter in all patients to enable hourly monitoring of urine output and fluid balance. Urea and electrolytes should be measured 6 hourly and the child should be weighed daily if possible.
Aim for a urine output of 3–4 mL/kg/h. However, if there is a significantly negative fluid balance as a result, aim for a lower urine output during AVP titrations. Conversely, if there is significantly positive fluid balance, target a higher urine output.
Start the AVP infusion at 0.0001 units/kg/h and alter the rate according to our recommended protocol (Table 2). The size of increment/decrement may need to be bigger or smaller depending on the sensitivity of the patient to desmopressin. We recommend that titration is performed hourly.
Table 2 Recommended protocol using the AVP (arginine vasopressin) Infusion Sliding Scale.
Urine output, mL/kg/h Action taken
<2 Stop infusion till urine output is 3–4 mL/kg/h
2–3 Reduce infusion rate by 25%
3–4 Maintain infusion rate
4–5 Increase infusion rate by 25%
>5 Increase infusion rate by 50%
Aim for urine output 3–4 mL/kg/h, however, this target will be dependent on fluid balance. Calculate hourly urine output to enable titration of AVP to be done hourly. Size of increment/decrement may need to be altered depending on patient’s sensitivity to desmopressin.
Once therapy is complete and serum sodium is within the normal range for the patient, oral desmopression should be recommenced 1 h prior to discontinuing the AVP infusion.
We recognise that admission to a specialist HDU may be challenging in some clinical settings; however, with the intensity of observations and hourly titrations needed, 1:1 nursing care is recommended. We also acknowledge the limitation of extrapolating an AVP infusion sliding scale based on a single patient and limited experience in published literature. Thus we recommend that further experience in using this protocol or alternatives published in the literature would aid in refining the sliding scale and management of patients with CDI undergoing chemotherapeutic regimens requiring hyperhydration.
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of the research reported.
Funding
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
Patient consent
A written consent has been obtained from the patient’s parents and are attached to the submission.
Author contribution statement
Dr V Lee is the patient’s Paediatric Oncology Consultant and Prof P Dimitri is her Paediatric Endocrinology Consultant and managed this patient throughout the course of her therapy. J Devaraja is a Paediatric Registrar who had been involved with the endocrinological care of the patient when she was initially admitted, and during her chemotherapy regimens. J Devaraja and S Sloan wrote the manuscript. P Dimitri and V Lee reviewed and contributed to the content of the manuscript. P Dimitri is the senior author. | A STARTING DOSE OF 0.0001 UNITS/KG/H | DrugDosageText | CC BY-NC-ND | 33597311 | 19,607,817 | 2021-02-17 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Achlorhydria'. | Primary hyperparathyroidism and Zollinger Ellison syndrome during pregnancy: a case report.
Multiple endocrine neoplasia type 1 (MEN1) is a rare inherited endocrine disorder with a high rate of penetrance. The incidence of MEN1 is 1/30,000 in the general population; however, it is quite rare for a patient to present for medical attention with MEN1 for the first time in pregnancy. Primary hyperparathyroidism (PHPT) is one of the most common features of MEN1. The incidence of PHPT occurring in pregnancy is 1%. Despite advances in the medical, surgical and obstetric care over the years, management of this condition during pregnancy may be challenging. It can be difficult to identify pregnant women with PHPT requiring intervention and to monitor safely. Hypercalcemia can result in significant maternal and fetal adverse outcomes including: miscarriage, intrauterine growth restriction, preterm delivery, neonatal hypocalcaemia, pre-eclampsia and maternal nephrolithiasis. Herein, we present a case study of a lady with a strong family history of MEN1, who was biochemically proven to have PHPT and evidence of Zollinger Ellison Syndrome (ZE) on endoscopy. This patient delayed her assisted pregnancy plans for in vitro fertilization (IVF) until completion of the MEN1 workup; nevertheless, she spontaneously achieved an unplanned pregnancy. As a result, she required intervention with parathyroidectomy in the second trimester of her pregnancy as her calcium level continued to rise. This case study highlights the workup, follow up and management of MEN1 presenting with PHPT and ZE in pregnancy.
Women of childbearing age who are suspected to have a diagnosis of primary hyperparathyroidism ideally should have genetic testing and avoid pregnancy until definitive plans are in place. Zollinger Ellison syndrome in pregnancy means off-label use of high dose of proton pump inhibitors (PPI). Use of PPI in pregnancy is considered to be safe based on retrospective studies. Omeprazole, however, is FDA class C drug because of lack of large prospective studies or large case series during pregnancy. Calcium supplements in the form of calcium carbonate must be converted to calcium chloride by gastric acid in order to be absorbed, however, patients rendered achlorhydric as a result of PPI use will have impaired absorption of calcium. Therefore, use of calcium citrate might be considered a better option in this case.
Background
MEN1 is a rare genetic endocrine disorder. It is inherited in an autosomal dominant fashion. There is no gender or ethnic predilection; however, a significant proportion of these cases present in women of childbearing age. This leads to rare and difficult clinical challenges in pregnant women. The incidence of sporadic PHPT occurring in pregnancy is reported as 1% (1, 2). It is well reported that the most common form of PHPT will present as a solitary parathyroid adenoma and less than 2% will have 2 or more lesions (2). The latter is often seen in patients with an underlying genetic mutation. The greatest difficulty is to control maternal calcium level during pregnancy as calcimimetics have not been approved for use in pregnancy (3). It is reasonable to manage mild cases of hypercalcemia conservatively, but surgery may be indicated for moderate to severe cases. In non-pregnant patients, localization of the adenoma is performed by a combination of ultrasound with ionising radiation (99mTc Sestamibi scan (gold standard)/CT scan). Ultrasound is the only radiological modality that can be used in pregnancy to localize the site/sites of the disease. This is a unique case of MEN1 manifested as PHPT, with hypoparathyroidism post-operatively, and ZE syndrome in pregnancy all of which were challenging and were managed with a multidisciplinary (MDT) approach.
Case presentation
Medical history and demographics
We report the case of a 34-year-old lwoman who was referred to our service from a fertility clinic. As part of her workup for primary infertility, it was noted that her adjusted calcium was elevated at 2.84 mmol/L (normal range: 2.20–2.60 mmol/L) and PTH was elevated at 84.4 ng/L (normal range: 15–65 ng/L) (Fig. 1). She had no significant past medical history and was not on regular medications. Of note, she had no renal or bone disease. Her family history was strongly positive for MEN 1 with a genetically proven case in her brother and her mother and five out of six maternal aunts and uncles with clinical diagnosis of MEN 1. The patient was evaluated at a neuroendocrine tumour (NET) clinic and was investigated for MEN 1, with goal to correct any endocrine abnormalities prior to undergoing further treatment for fertility.
Figure 1 Pre- and post-operative ionised Ca+2, PTH, Vitamin D and phosphate.
Figure 2 Endoscopic ultrasound image revealing multiple shallow lesions in the duodenum.
Figure 3 Ultrasound neck showinghypervascular ovoid nodule.
Investigation
As part of the workup, it was noted that Gastrin level was elevated at 208 pg/mL (normal range: 29.4–121 pg/mL) and chromogranin A was 376 mg/mL (45–150 mg/mL). IGF-1, Insulin and glucagon were all within the normal range. The patient underwent endoscopy and endoscopic ultrasound (EUS); features were consistent with ZE syndrome; two to three 4–5 mm hypoechoic lesions consistent with small pNETs in the body and tail of the pancreas, and multiple shallow ulcers in the duodenum (Fig. 2). No obvious tumour in the head of the pancreas was identified and there were no suspicious nodes. She had a CT abdomen which showed a 1.2 cm nodule in the left adrenal gland. Adrenal endocrine functional workup was normal. Additionally, she had an octreotide scan which showed no site of active disease. She had a normal DXA scan and MRI of the pituitary. 99mTc Sestamibi scan demonstrated no focal tracer uptake in superior mediastinum or neck, ultrasound neck demonstrated a hypervascular ovoid nodule measuring 1.5 cm at lower pole right lobe of thyroid suspicious for a parathyroid adenoma (Fig. 3) and so the patient was referred to the endocrine surgeons for further management.
Unfortunately, when the patient presented to the surgical clinic; 12 weeks following previous investigations, she was found to be 7 weeks’ pregnant. Her calcium levels were monitored closely throughout her pregnancy and her ionised calcium remained elevated at 1.52 nmol/L (normal range: 1.19–1.35 nmol/L) with elevated parathyroid hormone (PTH) at 74 ng/L (normal range: 15–65 ng/L) and low vitamin D of 14.8 nmol/L (desired greater than 50 nmol/L). Given the concerns with regards to the use of cinacalcet in pregnancy, the patient was monitored with weekly calcium and PTH levels. She was also closely monitored for hypertension and any signs of foetal growth restriction throughout the pregnancy. Her genetic screening confirmed c831del mutation in MEN gene.
Treatment
The patient was initially commenced on high dose PPI (omeprazole 40 mg twice daily) following her endoscopic findings. The NET MDT was contacted by the maternal foetal medicine team at 24 weeks’ gestation with concerns re the continued rise in calcium levels. After reviewing the patient with an MDT approach, she was admitted electively for parathyroid exploration and subtotal parathyroidectomy. Fluid rehydration alone failed to suppress her calcium level. The procedure went well and three glands were removed, leaving the right upper gland in situ.
Outcome and follow-up
Post-operatively, the patient was transferred to ICU and commenced on IV calcium gluconate infusion. At this time, she was also commenced on oral calcium carbonate and alfacalciferol. The patient was transferred to the wards, with mother and fetus doing well. Despite increasing doses of calcium carbonate, ionised calcium levels were not maintained off infusion with the patient frequently becoming symptomatic with peri-oral paraesthesia and positive Chvostek’s sign. Magnesium replacement was also commenced. The patient remained dependent on IV calcium infusion for the first 12 postoperative days, fetus viability was closely monitored during this time. It was suggested that the patient was not absorbing calcium carbonate due to PPI use, therefore, she was switched to calcium citrate, a non-standard form of oral calcium in Ireland. IV calcium replacement was ceased 2 days later and the patient delivered at 37 weeks a healthy baby with no adverse events. Parathyroid axis recovered and the patient did not require calcium/vitamin D supplements for maintenance. She has since had a second uncomplicated pregnancy.
Discussion
Diagnosis of PHPT in pregnancy is rare, as is MEN1. There are only 155 cases reported in the literature of PHPT diagnosed in pregnancy (5). The majority of cases of PHPT in pregnancy can be managed with surveillance. It is important to establish a threshold for intervention with maternal foetal medicine. Untreated moderate to severe hypercalcemia can cause potential adverse outcomes including foetal and maternal deaths, pre-eclampsia and pregnancy-induced hypertension (6, 7). In our patient, the need for surgical intervention was due to failure to optimise calcium levels with conservative measures. Genetic screening informed the decision for three gland-parathyroidectomy. As our patient remained dependent on IV calcium for a few days post-operatively and failed to switch to oral calcium successfully, we changed the calcium salt to achieve successful result. For the absorption of calcium carbonate to happen, calcium carbonate must be converted to calcium chloride by gastric acid. In patients rendered achlorhydric, as with our patient secondary to high dose PPI, gastric acid levels are greatly reduced meaning that little calcium carbonate is converted into its absorbable form. In these instances, the bioavailability of the calcium carbonate drops from 30% to approximately 4%. Calcium citrate does not require gastric acid to be absorbed and this maintains its absorption rates in this patient population. In fact, calcium carbonate is the standard form of oral calcium replacement in Ireland, hence, it was one of the challenges we faced when switching the patient to calcium citrate while pregnant. Another challenge was the use of high dose PPI and its safety profile in pregnancy. This was thoroughly reviewed in the literature and PPI remain classified as class B/C in pregnancy due to lack of prospective studies but the overall profile is described as safe (8).
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this case report.
Funding
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
Patient consent
A written informed consent has been obtained from the patient for publication of the submitted article and accompanying images.
Author contribution statement
D A and P D wrote the first draft of this manuscript and were involved in collection of clinical data. All authors were involved in the writing of this manuscript and final review of the manuscript. R K C, R P, D O and D O were involved in the clinical care of these patients and collection of data. | CALCIUM CARBONATE, CALCIUM GLUCONATE, ERGOCALCIFEROL, OMEPRAZOLE | DrugsGivenReaction | CC BY-NC-ND | 33597314 | 19,725,934 | 2021-02-17 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Calcium ionised decreased'. | Primary hyperparathyroidism and Zollinger Ellison syndrome during pregnancy: a case report.
Multiple endocrine neoplasia type 1 (MEN1) is a rare inherited endocrine disorder with a high rate of penetrance. The incidence of MEN1 is 1/30,000 in the general population; however, it is quite rare for a patient to present for medical attention with MEN1 for the first time in pregnancy. Primary hyperparathyroidism (PHPT) is one of the most common features of MEN1. The incidence of PHPT occurring in pregnancy is 1%. Despite advances in the medical, surgical and obstetric care over the years, management of this condition during pregnancy may be challenging. It can be difficult to identify pregnant women with PHPT requiring intervention and to monitor safely. Hypercalcemia can result in significant maternal and fetal adverse outcomes including: miscarriage, intrauterine growth restriction, preterm delivery, neonatal hypocalcaemia, pre-eclampsia and maternal nephrolithiasis. Herein, we present a case study of a lady with a strong family history of MEN1, who was biochemically proven to have PHPT and evidence of Zollinger Ellison Syndrome (ZE) on endoscopy. This patient delayed her assisted pregnancy plans for in vitro fertilization (IVF) until completion of the MEN1 workup; nevertheless, she spontaneously achieved an unplanned pregnancy. As a result, she required intervention with parathyroidectomy in the second trimester of her pregnancy as her calcium level continued to rise. This case study highlights the workup, follow up and management of MEN1 presenting with PHPT and ZE in pregnancy.
Women of childbearing age who are suspected to have a diagnosis of primary hyperparathyroidism ideally should have genetic testing and avoid pregnancy until definitive plans are in place. Zollinger Ellison syndrome in pregnancy means off-label use of high dose of proton pump inhibitors (PPI). Use of PPI in pregnancy is considered to be safe based on retrospective studies. Omeprazole, however, is FDA class C drug because of lack of large prospective studies or large case series during pregnancy. Calcium supplements in the form of calcium carbonate must be converted to calcium chloride by gastric acid in order to be absorbed, however, patients rendered achlorhydric as a result of PPI use will have impaired absorption of calcium. Therefore, use of calcium citrate might be considered a better option in this case.
Background
MEN1 is a rare genetic endocrine disorder. It is inherited in an autosomal dominant fashion. There is no gender or ethnic predilection; however, a significant proportion of these cases present in women of childbearing age. This leads to rare and difficult clinical challenges in pregnant women. The incidence of sporadic PHPT occurring in pregnancy is reported as 1% (1, 2). It is well reported that the most common form of PHPT will present as a solitary parathyroid adenoma and less than 2% will have 2 or more lesions (2). The latter is often seen in patients with an underlying genetic mutation. The greatest difficulty is to control maternal calcium level during pregnancy as calcimimetics have not been approved for use in pregnancy (3). It is reasonable to manage mild cases of hypercalcemia conservatively, but surgery may be indicated for moderate to severe cases. In non-pregnant patients, localization of the adenoma is performed by a combination of ultrasound with ionising radiation (99mTc Sestamibi scan (gold standard)/CT scan). Ultrasound is the only radiological modality that can be used in pregnancy to localize the site/sites of the disease. This is a unique case of MEN1 manifested as PHPT, with hypoparathyroidism post-operatively, and ZE syndrome in pregnancy all of which were challenging and were managed with a multidisciplinary (MDT) approach.
Case presentation
Medical history and demographics
We report the case of a 34-year-old lwoman who was referred to our service from a fertility clinic. As part of her workup for primary infertility, it was noted that her adjusted calcium was elevated at 2.84 mmol/L (normal range: 2.20–2.60 mmol/L) and PTH was elevated at 84.4 ng/L (normal range: 15–65 ng/L) (Fig. 1). She had no significant past medical history and was not on regular medications. Of note, she had no renal or bone disease. Her family history was strongly positive for MEN 1 with a genetically proven case in her brother and her mother and five out of six maternal aunts and uncles with clinical diagnosis of MEN 1. The patient was evaluated at a neuroendocrine tumour (NET) clinic and was investigated for MEN 1, with goal to correct any endocrine abnormalities prior to undergoing further treatment for fertility.
Figure 1 Pre- and post-operative ionised Ca+2, PTH, Vitamin D and phosphate.
Figure 2 Endoscopic ultrasound image revealing multiple shallow lesions in the duodenum.
Figure 3 Ultrasound neck showinghypervascular ovoid nodule.
Investigation
As part of the workup, it was noted that Gastrin level was elevated at 208 pg/mL (normal range: 29.4–121 pg/mL) and chromogranin A was 376 mg/mL (45–150 mg/mL). IGF-1, Insulin and glucagon were all within the normal range. The patient underwent endoscopy and endoscopic ultrasound (EUS); features were consistent with ZE syndrome; two to three 4–5 mm hypoechoic lesions consistent with small pNETs in the body and tail of the pancreas, and multiple shallow ulcers in the duodenum (Fig. 2). No obvious tumour in the head of the pancreas was identified and there were no suspicious nodes. She had a CT abdomen which showed a 1.2 cm nodule in the left adrenal gland. Adrenal endocrine functional workup was normal. Additionally, she had an octreotide scan which showed no site of active disease. She had a normal DXA scan and MRI of the pituitary. 99mTc Sestamibi scan demonstrated no focal tracer uptake in superior mediastinum or neck, ultrasound neck demonstrated a hypervascular ovoid nodule measuring 1.5 cm at lower pole right lobe of thyroid suspicious for a parathyroid adenoma (Fig. 3) and so the patient was referred to the endocrine surgeons for further management.
Unfortunately, when the patient presented to the surgical clinic; 12 weeks following previous investigations, she was found to be 7 weeks’ pregnant. Her calcium levels were monitored closely throughout her pregnancy and her ionised calcium remained elevated at 1.52 nmol/L (normal range: 1.19–1.35 nmol/L) with elevated parathyroid hormone (PTH) at 74 ng/L (normal range: 15–65 ng/L) and low vitamin D of 14.8 nmol/L (desired greater than 50 nmol/L). Given the concerns with regards to the use of cinacalcet in pregnancy, the patient was monitored with weekly calcium and PTH levels. She was also closely monitored for hypertension and any signs of foetal growth restriction throughout the pregnancy. Her genetic screening confirmed c831del mutation in MEN gene.
Treatment
The patient was initially commenced on high dose PPI (omeprazole 40 mg twice daily) following her endoscopic findings. The NET MDT was contacted by the maternal foetal medicine team at 24 weeks’ gestation with concerns re the continued rise in calcium levels. After reviewing the patient with an MDT approach, she was admitted electively for parathyroid exploration and subtotal parathyroidectomy. Fluid rehydration alone failed to suppress her calcium level. The procedure went well and three glands were removed, leaving the right upper gland in situ.
Outcome and follow-up
Post-operatively, the patient was transferred to ICU and commenced on IV calcium gluconate infusion. At this time, she was also commenced on oral calcium carbonate and alfacalciferol. The patient was transferred to the wards, with mother and fetus doing well. Despite increasing doses of calcium carbonate, ionised calcium levels were not maintained off infusion with the patient frequently becoming symptomatic with peri-oral paraesthesia and positive Chvostek’s sign. Magnesium replacement was also commenced. The patient remained dependent on IV calcium infusion for the first 12 postoperative days, fetus viability was closely monitored during this time. It was suggested that the patient was not absorbing calcium carbonate due to PPI use, therefore, she was switched to calcium citrate, a non-standard form of oral calcium in Ireland. IV calcium replacement was ceased 2 days later and the patient delivered at 37 weeks a healthy baby with no adverse events. Parathyroid axis recovered and the patient did not require calcium/vitamin D supplements for maintenance. She has since had a second uncomplicated pregnancy.
Discussion
Diagnosis of PHPT in pregnancy is rare, as is MEN1. There are only 155 cases reported in the literature of PHPT diagnosed in pregnancy (5). The majority of cases of PHPT in pregnancy can be managed with surveillance. It is important to establish a threshold for intervention with maternal foetal medicine. Untreated moderate to severe hypercalcemia can cause potential adverse outcomes including foetal and maternal deaths, pre-eclampsia and pregnancy-induced hypertension (6, 7). In our patient, the need for surgical intervention was due to failure to optimise calcium levels with conservative measures. Genetic screening informed the decision for three gland-parathyroidectomy. As our patient remained dependent on IV calcium for a few days post-operatively and failed to switch to oral calcium successfully, we changed the calcium salt to achieve successful result. For the absorption of calcium carbonate to happen, calcium carbonate must be converted to calcium chloride by gastric acid. In patients rendered achlorhydric, as with our patient secondary to high dose PPI, gastric acid levels are greatly reduced meaning that little calcium carbonate is converted into its absorbable form. In these instances, the bioavailability of the calcium carbonate drops from 30% to approximately 4%. Calcium citrate does not require gastric acid to be absorbed and this maintains its absorption rates in this patient population. In fact, calcium carbonate is the standard form of oral calcium replacement in Ireland, hence, it was one of the challenges we faced when switching the patient to calcium citrate while pregnant. Another challenge was the use of high dose PPI and its safety profile in pregnancy. This was thoroughly reviewed in the literature and PPI remain classified as class B/C in pregnancy due to lack of prospective studies but the overall profile is described as safe (8).
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this case report.
Funding
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
Patient consent
A written informed consent has been obtained from the patient for publication of the submitted article and accompanying images.
Author contribution statement
D A and P D wrote the first draft of this manuscript and were involved in collection of clinical data. All authors were involved in the writing of this manuscript and final review of the manuscript. R K C, R P, D O and D O were involved in the clinical care of these patients and collection of data. | CALCIUM CARBONATE, CALCIUM GLUCONATE, ERGOCALCIFEROL, OMEPRAZOLE | DrugsGivenReaction | CC BY-NC-ND | 33597314 | 19,725,934 | 2021-02-17 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Calcium metabolism disorder'. | Primary hyperparathyroidism and Zollinger Ellison syndrome during pregnancy: a case report.
Multiple endocrine neoplasia type 1 (MEN1) is a rare inherited endocrine disorder with a high rate of penetrance. The incidence of MEN1 is 1/30,000 in the general population; however, it is quite rare for a patient to present for medical attention with MEN1 for the first time in pregnancy. Primary hyperparathyroidism (PHPT) is one of the most common features of MEN1. The incidence of PHPT occurring in pregnancy is 1%. Despite advances in the medical, surgical and obstetric care over the years, management of this condition during pregnancy may be challenging. It can be difficult to identify pregnant women with PHPT requiring intervention and to monitor safely. Hypercalcemia can result in significant maternal and fetal adverse outcomes including: miscarriage, intrauterine growth restriction, preterm delivery, neonatal hypocalcaemia, pre-eclampsia and maternal nephrolithiasis. Herein, we present a case study of a lady with a strong family history of MEN1, who was biochemically proven to have PHPT and evidence of Zollinger Ellison Syndrome (ZE) on endoscopy. This patient delayed her assisted pregnancy plans for in vitro fertilization (IVF) until completion of the MEN1 workup; nevertheless, she spontaneously achieved an unplanned pregnancy. As a result, she required intervention with parathyroidectomy in the second trimester of her pregnancy as her calcium level continued to rise. This case study highlights the workup, follow up and management of MEN1 presenting with PHPT and ZE in pregnancy.
Women of childbearing age who are suspected to have a diagnosis of primary hyperparathyroidism ideally should have genetic testing and avoid pregnancy until definitive plans are in place. Zollinger Ellison syndrome in pregnancy means off-label use of high dose of proton pump inhibitors (PPI). Use of PPI in pregnancy is considered to be safe based on retrospective studies. Omeprazole, however, is FDA class C drug because of lack of large prospective studies or large case series during pregnancy. Calcium supplements in the form of calcium carbonate must be converted to calcium chloride by gastric acid in order to be absorbed, however, patients rendered achlorhydric as a result of PPI use will have impaired absorption of calcium. Therefore, use of calcium citrate might be considered a better option in this case.
Background
MEN1 is a rare genetic endocrine disorder. It is inherited in an autosomal dominant fashion. There is no gender or ethnic predilection; however, a significant proportion of these cases present in women of childbearing age. This leads to rare and difficult clinical challenges in pregnant women. The incidence of sporadic PHPT occurring in pregnancy is reported as 1% (1, 2). It is well reported that the most common form of PHPT will present as a solitary parathyroid adenoma and less than 2% will have 2 or more lesions (2). The latter is often seen in patients with an underlying genetic mutation. The greatest difficulty is to control maternal calcium level during pregnancy as calcimimetics have not been approved for use in pregnancy (3). It is reasonable to manage mild cases of hypercalcemia conservatively, but surgery may be indicated for moderate to severe cases. In non-pregnant patients, localization of the adenoma is performed by a combination of ultrasound with ionising radiation (99mTc Sestamibi scan (gold standard)/CT scan). Ultrasound is the only radiological modality that can be used in pregnancy to localize the site/sites of the disease. This is a unique case of MEN1 manifested as PHPT, with hypoparathyroidism post-operatively, and ZE syndrome in pregnancy all of which were challenging and were managed with a multidisciplinary (MDT) approach.
Case presentation
Medical history and demographics
We report the case of a 34-year-old lwoman who was referred to our service from a fertility clinic. As part of her workup for primary infertility, it was noted that her adjusted calcium was elevated at 2.84 mmol/L (normal range: 2.20–2.60 mmol/L) and PTH was elevated at 84.4 ng/L (normal range: 15–65 ng/L) (Fig. 1). She had no significant past medical history and was not on regular medications. Of note, she had no renal or bone disease. Her family history was strongly positive for MEN 1 with a genetically proven case in her brother and her mother and five out of six maternal aunts and uncles with clinical diagnosis of MEN 1. The patient was evaluated at a neuroendocrine tumour (NET) clinic and was investigated for MEN 1, with goal to correct any endocrine abnormalities prior to undergoing further treatment for fertility.
Figure 1 Pre- and post-operative ionised Ca+2, PTH, Vitamin D and phosphate.
Figure 2 Endoscopic ultrasound image revealing multiple shallow lesions in the duodenum.
Figure 3 Ultrasound neck showinghypervascular ovoid nodule.
Investigation
As part of the workup, it was noted that Gastrin level was elevated at 208 pg/mL (normal range: 29.4–121 pg/mL) and chromogranin A was 376 mg/mL (45–150 mg/mL). IGF-1, Insulin and glucagon were all within the normal range. The patient underwent endoscopy and endoscopic ultrasound (EUS); features were consistent with ZE syndrome; two to three 4–5 mm hypoechoic lesions consistent with small pNETs in the body and tail of the pancreas, and multiple shallow ulcers in the duodenum (Fig. 2). No obvious tumour in the head of the pancreas was identified and there were no suspicious nodes. She had a CT abdomen which showed a 1.2 cm nodule in the left adrenal gland. Adrenal endocrine functional workup was normal. Additionally, she had an octreotide scan which showed no site of active disease. She had a normal DXA scan and MRI of the pituitary. 99mTc Sestamibi scan demonstrated no focal tracer uptake in superior mediastinum or neck, ultrasound neck demonstrated a hypervascular ovoid nodule measuring 1.5 cm at lower pole right lobe of thyroid suspicious for a parathyroid adenoma (Fig. 3) and so the patient was referred to the endocrine surgeons for further management.
Unfortunately, when the patient presented to the surgical clinic; 12 weeks following previous investigations, she was found to be 7 weeks’ pregnant. Her calcium levels were monitored closely throughout her pregnancy and her ionised calcium remained elevated at 1.52 nmol/L (normal range: 1.19–1.35 nmol/L) with elevated parathyroid hormone (PTH) at 74 ng/L (normal range: 15–65 ng/L) and low vitamin D of 14.8 nmol/L (desired greater than 50 nmol/L). Given the concerns with regards to the use of cinacalcet in pregnancy, the patient was monitored with weekly calcium and PTH levels. She was also closely monitored for hypertension and any signs of foetal growth restriction throughout the pregnancy. Her genetic screening confirmed c831del mutation in MEN gene.
Treatment
The patient was initially commenced on high dose PPI (omeprazole 40 mg twice daily) following her endoscopic findings. The NET MDT was contacted by the maternal foetal medicine team at 24 weeks’ gestation with concerns re the continued rise in calcium levels. After reviewing the patient with an MDT approach, she was admitted electively for parathyroid exploration and subtotal parathyroidectomy. Fluid rehydration alone failed to suppress her calcium level. The procedure went well and three glands were removed, leaving the right upper gland in situ.
Outcome and follow-up
Post-operatively, the patient was transferred to ICU and commenced on IV calcium gluconate infusion. At this time, she was also commenced on oral calcium carbonate and alfacalciferol. The patient was transferred to the wards, with mother and fetus doing well. Despite increasing doses of calcium carbonate, ionised calcium levels were not maintained off infusion with the patient frequently becoming symptomatic with peri-oral paraesthesia and positive Chvostek’s sign. Magnesium replacement was also commenced. The patient remained dependent on IV calcium infusion for the first 12 postoperative days, fetus viability was closely monitored during this time. It was suggested that the patient was not absorbing calcium carbonate due to PPI use, therefore, she was switched to calcium citrate, a non-standard form of oral calcium in Ireland. IV calcium replacement was ceased 2 days later and the patient delivered at 37 weeks a healthy baby with no adverse events. Parathyroid axis recovered and the patient did not require calcium/vitamin D supplements for maintenance. She has since had a second uncomplicated pregnancy.
Discussion
Diagnosis of PHPT in pregnancy is rare, as is MEN1. There are only 155 cases reported in the literature of PHPT diagnosed in pregnancy (5). The majority of cases of PHPT in pregnancy can be managed with surveillance. It is important to establish a threshold for intervention with maternal foetal medicine. Untreated moderate to severe hypercalcemia can cause potential adverse outcomes including foetal and maternal deaths, pre-eclampsia and pregnancy-induced hypertension (6, 7). In our patient, the need for surgical intervention was due to failure to optimise calcium levels with conservative measures. Genetic screening informed the decision for three gland-parathyroidectomy. As our patient remained dependent on IV calcium for a few days post-operatively and failed to switch to oral calcium successfully, we changed the calcium salt to achieve successful result. For the absorption of calcium carbonate to happen, calcium carbonate must be converted to calcium chloride by gastric acid. In patients rendered achlorhydric, as with our patient secondary to high dose PPI, gastric acid levels are greatly reduced meaning that little calcium carbonate is converted into its absorbable form. In these instances, the bioavailability of the calcium carbonate drops from 30% to approximately 4%. Calcium citrate does not require gastric acid to be absorbed and this maintains its absorption rates in this patient population. In fact, calcium carbonate is the standard form of oral calcium replacement in Ireland, hence, it was one of the challenges we faced when switching the patient to calcium citrate while pregnant. Another challenge was the use of high dose PPI and its safety profile in pregnancy. This was thoroughly reviewed in the literature and PPI remain classified as class B/C in pregnancy due to lack of prospective studies but the overall profile is described as safe (8).
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this case report.
Funding
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
Patient consent
A written informed consent has been obtained from the patient for publication of the submitted article and accompanying images.
Author contribution statement
D A and P D wrote the first draft of this manuscript and were involved in collection of clinical data. All authors were involved in the writing of this manuscript and final review of the manuscript. R K C, R P, D O and D O were involved in the clinical care of these patients and collection of data. | ALFACALCIDOL, CALCIUM, CALCIUM CARBONATE, CALCIUM CITRATE, CALCIUM GLUCONATE, MAGNESIUM, OMEPRAZOLE, VITAMIN D NOS | DrugsGivenReaction | CC BY-NC-ND | 33597314 | 19,895,614 | 2021-02-17 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Chvostek^s sign'. | Primary hyperparathyroidism and Zollinger Ellison syndrome during pregnancy: a case report.
Multiple endocrine neoplasia type 1 (MEN1) is a rare inherited endocrine disorder with a high rate of penetrance. The incidence of MEN1 is 1/30,000 in the general population; however, it is quite rare for a patient to present for medical attention with MEN1 for the first time in pregnancy. Primary hyperparathyroidism (PHPT) is one of the most common features of MEN1. The incidence of PHPT occurring in pregnancy is 1%. Despite advances in the medical, surgical and obstetric care over the years, management of this condition during pregnancy may be challenging. It can be difficult to identify pregnant women with PHPT requiring intervention and to monitor safely. Hypercalcemia can result in significant maternal and fetal adverse outcomes including: miscarriage, intrauterine growth restriction, preterm delivery, neonatal hypocalcaemia, pre-eclampsia and maternal nephrolithiasis. Herein, we present a case study of a lady with a strong family history of MEN1, who was biochemically proven to have PHPT and evidence of Zollinger Ellison Syndrome (ZE) on endoscopy. This patient delayed her assisted pregnancy plans for in vitro fertilization (IVF) until completion of the MEN1 workup; nevertheless, she spontaneously achieved an unplanned pregnancy. As a result, she required intervention with parathyroidectomy in the second trimester of her pregnancy as her calcium level continued to rise. This case study highlights the workup, follow up and management of MEN1 presenting with PHPT and ZE in pregnancy.
Women of childbearing age who are suspected to have a diagnosis of primary hyperparathyroidism ideally should have genetic testing and avoid pregnancy until definitive plans are in place. Zollinger Ellison syndrome in pregnancy means off-label use of high dose of proton pump inhibitors (PPI). Use of PPI in pregnancy is considered to be safe based on retrospective studies. Omeprazole, however, is FDA class C drug because of lack of large prospective studies or large case series during pregnancy. Calcium supplements in the form of calcium carbonate must be converted to calcium chloride by gastric acid in order to be absorbed, however, patients rendered achlorhydric as a result of PPI use will have impaired absorption of calcium. Therefore, use of calcium citrate might be considered a better option in this case.
Background
MEN1 is a rare genetic endocrine disorder. It is inherited in an autosomal dominant fashion. There is no gender or ethnic predilection; however, a significant proportion of these cases present in women of childbearing age. This leads to rare and difficult clinical challenges in pregnant women. The incidence of sporadic PHPT occurring in pregnancy is reported as 1% (1, 2). It is well reported that the most common form of PHPT will present as a solitary parathyroid adenoma and less than 2% will have 2 or more lesions (2). The latter is often seen in patients with an underlying genetic mutation. The greatest difficulty is to control maternal calcium level during pregnancy as calcimimetics have not been approved for use in pregnancy (3). It is reasonable to manage mild cases of hypercalcemia conservatively, but surgery may be indicated for moderate to severe cases. In non-pregnant patients, localization of the adenoma is performed by a combination of ultrasound with ionising radiation (99mTc Sestamibi scan (gold standard)/CT scan). Ultrasound is the only radiological modality that can be used in pregnancy to localize the site/sites of the disease. This is a unique case of MEN1 manifested as PHPT, with hypoparathyroidism post-operatively, and ZE syndrome in pregnancy all of which were challenging and were managed with a multidisciplinary (MDT) approach.
Case presentation
Medical history and demographics
We report the case of a 34-year-old lwoman who was referred to our service from a fertility clinic. As part of her workup for primary infertility, it was noted that her adjusted calcium was elevated at 2.84 mmol/L (normal range: 2.20–2.60 mmol/L) and PTH was elevated at 84.4 ng/L (normal range: 15–65 ng/L) (Fig. 1). She had no significant past medical history and was not on regular medications. Of note, she had no renal or bone disease. Her family history was strongly positive for MEN 1 with a genetically proven case in her brother and her mother and five out of six maternal aunts and uncles with clinical diagnosis of MEN 1. The patient was evaluated at a neuroendocrine tumour (NET) clinic and was investigated for MEN 1, with goal to correct any endocrine abnormalities prior to undergoing further treatment for fertility.
Figure 1 Pre- and post-operative ionised Ca+2, PTH, Vitamin D and phosphate.
Figure 2 Endoscopic ultrasound image revealing multiple shallow lesions in the duodenum.
Figure 3 Ultrasound neck showinghypervascular ovoid nodule.
Investigation
As part of the workup, it was noted that Gastrin level was elevated at 208 pg/mL (normal range: 29.4–121 pg/mL) and chromogranin A was 376 mg/mL (45–150 mg/mL). IGF-1, Insulin and glucagon were all within the normal range. The patient underwent endoscopy and endoscopic ultrasound (EUS); features were consistent with ZE syndrome; two to three 4–5 mm hypoechoic lesions consistent with small pNETs in the body and tail of the pancreas, and multiple shallow ulcers in the duodenum (Fig. 2). No obvious tumour in the head of the pancreas was identified and there were no suspicious nodes. She had a CT abdomen which showed a 1.2 cm nodule in the left adrenal gland. Adrenal endocrine functional workup was normal. Additionally, she had an octreotide scan which showed no site of active disease. She had a normal DXA scan and MRI of the pituitary. 99mTc Sestamibi scan demonstrated no focal tracer uptake in superior mediastinum or neck, ultrasound neck demonstrated a hypervascular ovoid nodule measuring 1.5 cm at lower pole right lobe of thyroid suspicious for a parathyroid adenoma (Fig. 3) and so the patient was referred to the endocrine surgeons for further management.
Unfortunately, when the patient presented to the surgical clinic; 12 weeks following previous investigations, she was found to be 7 weeks’ pregnant. Her calcium levels were monitored closely throughout her pregnancy and her ionised calcium remained elevated at 1.52 nmol/L (normal range: 1.19–1.35 nmol/L) with elevated parathyroid hormone (PTH) at 74 ng/L (normal range: 15–65 ng/L) and low vitamin D of 14.8 nmol/L (desired greater than 50 nmol/L). Given the concerns with regards to the use of cinacalcet in pregnancy, the patient was monitored with weekly calcium and PTH levels. She was also closely monitored for hypertension and any signs of foetal growth restriction throughout the pregnancy. Her genetic screening confirmed c831del mutation in MEN gene.
Treatment
The patient was initially commenced on high dose PPI (omeprazole 40 mg twice daily) following her endoscopic findings. The NET MDT was contacted by the maternal foetal medicine team at 24 weeks’ gestation with concerns re the continued rise in calcium levels. After reviewing the patient with an MDT approach, she was admitted electively for parathyroid exploration and subtotal parathyroidectomy. Fluid rehydration alone failed to suppress her calcium level. The procedure went well and three glands were removed, leaving the right upper gland in situ.
Outcome and follow-up
Post-operatively, the patient was transferred to ICU and commenced on IV calcium gluconate infusion. At this time, she was also commenced on oral calcium carbonate and alfacalciferol. The patient was transferred to the wards, with mother and fetus doing well. Despite increasing doses of calcium carbonate, ionised calcium levels were not maintained off infusion with the patient frequently becoming symptomatic with peri-oral paraesthesia and positive Chvostek’s sign. Magnesium replacement was also commenced. The patient remained dependent on IV calcium infusion for the first 12 postoperative days, fetus viability was closely monitored during this time. It was suggested that the patient was not absorbing calcium carbonate due to PPI use, therefore, she was switched to calcium citrate, a non-standard form of oral calcium in Ireland. IV calcium replacement was ceased 2 days later and the patient delivered at 37 weeks a healthy baby with no adverse events. Parathyroid axis recovered and the patient did not require calcium/vitamin D supplements for maintenance. She has since had a second uncomplicated pregnancy.
Discussion
Diagnosis of PHPT in pregnancy is rare, as is MEN1. There are only 155 cases reported in the literature of PHPT diagnosed in pregnancy (5). The majority of cases of PHPT in pregnancy can be managed with surveillance. It is important to establish a threshold for intervention with maternal foetal medicine. Untreated moderate to severe hypercalcemia can cause potential adverse outcomes including foetal and maternal deaths, pre-eclampsia and pregnancy-induced hypertension (6, 7). In our patient, the need for surgical intervention was due to failure to optimise calcium levels with conservative measures. Genetic screening informed the decision for three gland-parathyroidectomy. As our patient remained dependent on IV calcium for a few days post-operatively and failed to switch to oral calcium successfully, we changed the calcium salt to achieve successful result. For the absorption of calcium carbonate to happen, calcium carbonate must be converted to calcium chloride by gastric acid. In patients rendered achlorhydric, as with our patient secondary to high dose PPI, gastric acid levels are greatly reduced meaning that little calcium carbonate is converted into its absorbable form. In these instances, the bioavailability of the calcium carbonate drops from 30% to approximately 4%. Calcium citrate does not require gastric acid to be absorbed and this maintains its absorption rates in this patient population. In fact, calcium carbonate is the standard form of oral calcium replacement in Ireland, hence, it was one of the challenges we faced when switching the patient to calcium citrate while pregnant. Another challenge was the use of high dose PPI and its safety profile in pregnancy. This was thoroughly reviewed in the literature and PPI remain classified as class B/C in pregnancy due to lack of prospective studies but the overall profile is described as safe (8).
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this case report.
Funding
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
Patient consent
A written informed consent has been obtained from the patient for publication of the submitted article and accompanying images.
Author contribution statement
D A and P D wrote the first draft of this manuscript and were involved in collection of clinical data. All authors were involved in the writing of this manuscript and final review of the manuscript. R K C, R P, D O and D O were involved in the clinical care of these patients and collection of data. | CALCIUM CARBONATE, CALCIUM GLUCONATE, ERGOCALCIFEROL, OMEPRAZOLE | DrugsGivenReaction | CC BY-NC-ND | 33597314 | 19,725,934 | 2021-02-17 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Malabsorption'. | Primary hyperparathyroidism and Zollinger Ellison syndrome during pregnancy: a case report.
Multiple endocrine neoplasia type 1 (MEN1) is a rare inherited endocrine disorder with a high rate of penetrance. The incidence of MEN1 is 1/30,000 in the general population; however, it is quite rare for a patient to present for medical attention with MEN1 for the first time in pregnancy. Primary hyperparathyroidism (PHPT) is one of the most common features of MEN1. The incidence of PHPT occurring in pregnancy is 1%. Despite advances in the medical, surgical and obstetric care over the years, management of this condition during pregnancy may be challenging. It can be difficult to identify pregnant women with PHPT requiring intervention and to monitor safely. Hypercalcemia can result in significant maternal and fetal adverse outcomes including: miscarriage, intrauterine growth restriction, preterm delivery, neonatal hypocalcaemia, pre-eclampsia and maternal nephrolithiasis. Herein, we present a case study of a lady with a strong family history of MEN1, who was biochemically proven to have PHPT and evidence of Zollinger Ellison Syndrome (ZE) on endoscopy. This patient delayed her assisted pregnancy plans for in vitro fertilization (IVF) until completion of the MEN1 workup; nevertheless, she spontaneously achieved an unplanned pregnancy. As a result, she required intervention with parathyroidectomy in the second trimester of her pregnancy as her calcium level continued to rise. This case study highlights the workup, follow up and management of MEN1 presenting with PHPT and ZE in pregnancy.
Women of childbearing age who are suspected to have a diagnosis of primary hyperparathyroidism ideally should have genetic testing and avoid pregnancy until definitive plans are in place. Zollinger Ellison syndrome in pregnancy means off-label use of high dose of proton pump inhibitors (PPI). Use of PPI in pregnancy is considered to be safe based on retrospective studies. Omeprazole, however, is FDA class C drug because of lack of large prospective studies or large case series during pregnancy. Calcium supplements in the form of calcium carbonate must be converted to calcium chloride by gastric acid in order to be absorbed, however, patients rendered achlorhydric as a result of PPI use will have impaired absorption of calcium. Therefore, use of calcium citrate might be considered a better option in this case.
Background
MEN1 is a rare genetic endocrine disorder. It is inherited in an autosomal dominant fashion. There is no gender or ethnic predilection; however, a significant proportion of these cases present in women of childbearing age. This leads to rare and difficult clinical challenges in pregnant women. The incidence of sporadic PHPT occurring in pregnancy is reported as 1% (1, 2). It is well reported that the most common form of PHPT will present as a solitary parathyroid adenoma and less than 2% will have 2 or more lesions (2). The latter is often seen in patients with an underlying genetic mutation. The greatest difficulty is to control maternal calcium level during pregnancy as calcimimetics have not been approved for use in pregnancy (3). It is reasonable to manage mild cases of hypercalcemia conservatively, but surgery may be indicated for moderate to severe cases. In non-pregnant patients, localization of the adenoma is performed by a combination of ultrasound with ionising radiation (99mTc Sestamibi scan (gold standard)/CT scan). Ultrasound is the only radiological modality that can be used in pregnancy to localize the site/sites of the disease. This is a unique case of MEN1 manifested as PHPT, with hypoparathyroidism post-operatively, and ZE syndrome in pregnancy all of which were challenging and were managed with a multidisciplinary (MDT) approach.
Case presentation
Medical history and demographics
We report the case of a 34-year-old lwoman who was referred to our service from a fertility clinic. As part of her workup for primary infertility, it was noted that her adjusted calcium was elevated at 2.84 mmol/L (normal range: 2.20–2.60 mmol/L) and PTH was elevated at 84.4 ng/L (normal range: 15–65 ng/L) (Fig. 1). She had no significant past medical history and was not on regular medications. Of note, she had no renal or bone disease. Her family history was strongly positive for MEN 1 with a genetically proven case in her brother and her mother and five out of six maternal aunts and uncles with clinical diagnosis of MEN 1. The patient was evaluated at a neuroendocrine tumour (NET) clinic and was investigated for MEN 1, with goal to correct any endocrine abnormalities prior to undergoing further treatment for fertility.
Figure 1 Pre- and post-operative ionised Ca+2, PTH, Vitamin D and phosphate.
Figure 2 Endoscopic ultrasound image revealing multiple shallow lesions in the duodenum.
Figure 3 Ultrasound neck showinghypervascular ovoid nodule.
Investigation
As part of the workup, it was noted that Gastrin level was elevated at 208 pg/mL (normal range: 29.4–121 pg/mL) and chromogranin A was 376 mg/mL (45–150 mg/mL). IGF-1, Insulin and glucagon were all within the normal range. The patient underwent endoscopy and endoscopic ultrasound (EUS); features were consistent with ZE syndrome; two to three 4–5 mm hypoechoic lesions consistent with small pNETs in the body and tail of the pancreas, and multiple shallow ulcers in the duodenum (Fig. 2). No obvious tumour in the head of the pancreas was identified and there were no suspicious nodes. She had a CT abdomen which showed a 1.2 cm nodule in the left adrenal gland. Adrenal endocrine functional workup was normal. Additionally, she had an octreotide scan which showed no site of active disease. She had a normal DXA scan and MRI of the pituitary. 99mTc Sestamibi scan demonstrated no focal tracer uptake in superior mediastinum or neck, ultrasound neck demonstrated a hypervascular ovoid nodule measuring 1.5 cm at lower pole right lobe of thyroid suspicious for a parathyroid adenoma (Fig. 3) and so the patient was referred to the endocrine surgeons for further management.
Unfortunately, when the patient presented to the surgical clinic; 12 weeks following previous investigations, she was found to be 7 weeks’ pregnant. Her calcium levels were monitored closely throughout her pregnancy and her ionised calcium remained elevated at 1.52 nmol/L (normal range: 1.19–1.35 nmol/L) with elevated parathyroid hormone (PTH) at 74 ng/L (normal range: 15–65 ng/L) and low vitamin D of 14.8 nmol/L (desired greater than 50 nmol/L). Given the concerns with regards to the use of cinacalcet in pregnancy, the patient was monitored with weekly calcium and PTH levels. She was also closely monitored for hypertension and any signs of foetal growth restriction throughout the pregnancy. Her genetic screening confirmed c831del mutation in MEN gene.
Treatment
The patient was initially commenced on high dose PPI (omeprazole 40 mg twice daily) following her endoscopic findings. The NET MDT was contacted by the maternal foetal medicine team at 24 weeks’ gestation with concerns re the continued rise in calcium levels. After reviewing the patient with an MDT approach, she was admitted electively for parathyroid exploration and subtotal parathyroidectomy. Fluid rehydration alone failed to suppress her calcium level. The procedure went well and three glands were removed, leaving the right upper gland in situ.
Outcome and follow-up
Post-operatively, the patient was transferred to ICU and commenced on IV calcium gluconate infusion. At this time, she was also commenced on oral calcium carbonate and alfacalciferol. The patient was transferred to the wards, with mother and fetus doing well. Despite increasing doses of calcium carbonate, ionised calcium levels were not maintained off infusion with the patient frequently becoming symptomatic with peri-oral paraesthesia and positive Chvostek’s sign. Magnesium replacement was also commenced. The patient remained dependent on IV calcium infusion for the first 12 postoperative days, fetus viability was closely monitored during this time. It was suggested that the patient was not absorbing calcium carbonate due to PPI use, therefore, she was switched to calcium citrate, a non-standard form of oral calcium in Ireland. IV calcium replacement was ceased 2 days later and the patient delivered at 37 weeks a healthy baby with no adverse events. Parathyroid axis recovered and the patient did not require calcium/vitamin D supplements for maintenance. She has since had a second uncomplicated pregnancy.
Discussion
Diagnosis of PHPT in pregnancy is rare, as is MEN1. There are only 155 cases reported in the literature of PHPT diagnosed in pregnancy (5). The majority of cases of PHPT in pregnancy can be managed with surveillance. It is important to establish a threshold for intervention with maternal foetal medicine. Untreated moderate to severe hypercalcemia can cause potential adverse outcomes including foetal and maternal deaths, pre-eclampsia and pregnancy-induced hypertension (6, 7). In our patient, the need for surgical intervention was due to failure to optimise calcium levels with conservative measures. Genetic screening informed the decision for three gland-parathyroidectomy. As our patient remained dependent on IV calcium for a few days post-operatively and failed to switch to oral calcium successfully, we changed the calcium salt to achieve successful result. For the absorption of calcium carbonate to happen, calcium carbonate must be converted to calcium chloride by gastric acid. In patients rendered achlorhydric, as with our patient secondary to high dose PPI, gastric acid levels are greatly reduced meaning that little calcium carbonate is converted into its absorbable form. In these instances, the bioavailability of the calcium carbonate drops from 30% to approximately 4%. Calcium citrate does not require gastric acid to be absorbed and this maintains its absorption rates in this patient population. In fact, calcium carbonate is the standard form of oral calcium replacement in Ireland, hence, it was one of the challenges we faced when switching the patient to calcium citrate while pregnant. Another challenge was the use of high dose PPI and its safety profile in pregnancy. This was thoroughly reviewed in the literature and PPI remain classified as class B/C in pregnancy due to lack of prospective studies but the overall profile is described as safe (8).
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this case report.
Funding
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
Patient consent
A written informed consent has been obtained from the patient for publication of the submitted article and accompanying images.
Author contribution statement
D A and P D wrote the first draft of this manuscript and were involved in collection of clinical data. All authors were involved in the writing of this manuscript and final review of the manuscript. R K C, R P, D O and D O were involved in the clinical care of these patients and collection of data. | CALCIUM CARBONATE, CALCIUM CITRATE, CALCIUM GLUCONATE, OMEPRAZOLE | DrugsGivenReaction | CC BY-NC-ND | 33597314 | 19,930,939 | 2021-02-17 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Normal newborn'. | Primary hyperparathyroidism and Zollinger Ellison syndrome during pregnancy: a case report.
Multiple endocrine neoplasia type 1 (MEN1) is a rare inherited endocrine disorder with a high rate of penetrance. The incidence of MEN1 is 1/30,000 in the general population; however, it is quite rare for a patient to present for medical attention with MEN1 for the first time in pregnancy. Primary hyperparathyroidism (PHPT) is one of the most common features of MEN1. The incidence of PHPT occurring in pregnancy is 1%. Despite advances in the medical, surgical and obstetric care over the years, management of this condition during pregnancy may be challenging. It can be difficult to identify pregnant women with PHPT requiring intervention and to monitor safely. Hypercalcemia can result in significant maternal and fetal adverse outcomes including: miscarriage, intrauterine growth restriction, preterm delivery, neonatal hypocalcaemia, pre-eclampsia and maternal nephrolithiasis. Herein, we present a case study of a lady with a strong family history of MEN1, who was biochemically proven to have PHPT and evidence of Zollinger Ellison Syndrome (ZE) on endoscopy. This patient delayed her assisted pregnancy plans for in vitro fertilization (IVF) until completion of the MEN1 workup; nevertheless, she spontaneously achieved an unplanned pregnancy. As a result, she required intervention with parathyroidectomy in the second trimester of her pregnancy as her calcium level continued to rise. This case study highlights the workup, follow up and management of MEN1 presenting with PHPT and ZE in pregnancy.
Women of childbearing age who are suspected to have a diagnosis of primary hyperparathyroidism ideally should have genetic testing and avoid pregnancy until definitive plans are in place. Zollinger Ellison syndrome in pregnancy means off-label use of high dose of proton pump inhibitors (PPI). Use of PPI in pregnancy is considered to be safe based on retrospective studies. Omeprazole, however, is FDA class C drug because of lack of large prospective studies or large case series during pregnancy. Calcium supplements in the form of calcium carbonate must be converted to calcium chloride by gastric acid in order to be absorbed, however, patients rendered achlorhydric as a result of PPI use will have impaired absorption of calcium. Therefore, use of calcium citrate might be considered a better option in this case.
Background
MEN1 is a rare genetic endocrine disorder. It is inherited in an autosomal dominant fashion. There is no gender or ethnic predilection; however, a significant proportion of these cases present in women of childbearing age. This leads to rare and difficult clinical challenges in pregnant women. The incidence of sporadic PHPT occurring in pregnancy is reported as 1% (1, 2). It is well reported that the most common form of PHPT will present as a solitary parathyroid adenoma and less than 2% will have 2 or more lesions (2). The latter is often seen in patients with an underlying genetic mutation. The greatest difficulty is to control maternal calcium level during pregnancy as calcimimetics have not been approved for use in pregnancy (3). It is reasonable to manage mild cases of hypercalcemia conservatively, but surgery may be indicated for moderate to severe cases. In non-pregnant patients, localization of the adenoma is performed by a combination of ultrasound with ionising radiation (99mTc Sestamibi scan (gold standard)/CT scan). Ultrasound is the only radiological modality that can be used in pregnancy to localize the site/sites of the disease. This is a unique case of MEN1 manifested as PHPT, with hypoparathyroidism post-operatively, and ZE syndrome in pregnancy all of which were challenging and were managed with a multidisciplinary (MDT) approach.
Case presentation
Medical history and demographics
We report the case of a 34-year-old lwoman who was referred to our service from a fertility clinic. As part of her workup for primary infertility, it was noted that her adjusted calcium was elevated at 2.84 mmol/L (normal range: 2.20–2.60 mmol/L) and PTH was elevated at 84.4 ng/L (normal range: 15–65 ng/L) (Fig. 1). She had no significant past medical history and was not on regular medications. Of note, she had no renal or bone disease. Her family history was strongly positive for MEN 1 with a genetically proven case in her brother and her mother and five out of six maternal aunts and uncles with clinical diagnosis of MEN 1. The patient was evaluated at a neuroendocrine tumour (NET) clinic and was investigated for MEN 1, with goal to correct any endocrine abnormalities prior to undergoing further treatment for fertility.
Figure 1 Pre- and post-operative ionised Ca+2, PTH, Vitamin D and phosphate.
Figure 2 Endoscopic ultrasound image revealing multiple shallow lesions in the duodenum.
Figure 3 Ultrasound neck showinghypervascular ovoid nodule.
Investigation
As part of the workup, it was noted that Gastrin level was elevated at 208 pg/mL (normal range: 29.4–121 pg/mL) and chromogranin A was 376 mg/mL (45–150 mg/mL). IGF-1, Insulin and glucagon were all within the normal range. The patient underwent endoscopy and endoscopic ultrasound (EUS); features were consistent with ZE syndrome; two to three 4–5 mm hypoechoic lesions consistent with small pNETs in the body and tail of the pancreas, and multiple shallow ulcers in the duodenum (Fig. 2). No obvious tumour in the head of the pancreas was identified and there were no suspicious nodes. She had a CT abdomen which showed a 1.2 cm nodule in the left adrenal gland. Adrenal endocrine functional workup was normal. Additionally, she had an octreotide scan which showed no site of active disease. She had a normal DXA scan and MRI of the pituitary. 99mTc Sestamibi scan demonstrated no focal tracer uptake in superior mediastinum or neck, ultrasound neck demonstrated a hypervascular ovoid nodule measuring 1.5 cm at lower pole right lobe of thyroid suspicious for a parathyroid adenoma (Fig. 3) and so the patient was referred to the endocrine surgeons for further management.
Unfortunately, when the patient presented to the surgical clinic; 12 weeks following previous investigations, she was found to be 7 weeks’ pregnant. Her calcium levels were monitored closely throughout her pregnancy and her ionised calcium remained elevated at 1.52 nmol/L (normal range: 1.19–1.35 nmol/L) with elevated parathyroid hormone (PTH) at 74 ng/L (normal range: 15–65 ng/L) and low vitamin D of 14.8 nmol/L (desired greater than 50 nmol/L). Given the concerns with regards to the use of cinacalcet in pregnancy, the patient was monitored with weekly calcium and PTH levels. She was also closely monitored for hypertension and any signs of foetal growth restriction throughout the pregnancy. Her genetic screening confirmed c831del mutation in MEN gene.
Treatment
The patient was initially commenced on high dose PPI (omeprazole 40 mg twice daily) following her endoscopic findings. The NET MDT was contacted by the maternal foetal medicine team at 24 weeks’ gestation with concerns re the continued rise in calcium levels. After reviewing the patient with an MDT approach, she was admitted electively for parathyroid exploration and subtotal parathyroidectomy. Fluid rehydration alone failed to suppress her calcium level. The procedure went well and three glands were removed, leaving the right upper gland in situ.
Outcome and follow-up
Post-operatively, the patient was transferred to ICU and commenced on IV calcium gluconate infusion. At this time, she was also commenced on oral calcium carbonate and alfacalciferol. The patient was transferred to the wards, with mother and fetus doing well. Despite increasing doses of calcium carbonate, ionised calcium levels were not maintained off infusion with the patient frequently becoming symptomatic with peri-oral paraesthesia and positive Chvostek’s sign. Magnesium replacement was also commenced. The patient remained dependent on IV calcium infusion for the first 12 postoperative days, fetus viability was closely monitored during this time. It was suggested that the patient was not absorbing calcium carbonate due to PPI use, therefore, she was switched to calcium citrate, a non-standard form of oral calcium in Ireland. IV calcium replacement was ceased 2 days later and the patient delivered at 37 weeks a healthy baby with no adverse events. Parathyroid axis recovered and the patient did not require calcium/vitamin D supplements for maintenance. She has since had a second uncomplicated pregnancy.
Discussion
Diagnosis of PHPT in pregnancy is rare, as is MEN1. There are only 155 cases reported in the literature of PHPT diagnosed in pregnancy (5). The majority of cases of PHPT in pregnancy can be managed with surveillance. It is important to establish a threshold for intervention with maternal foetal medicine. Untreated moderate to severe hypercalcemia can cause potential adverse outcomes including foetal and maternal deaths, pre-eclampsia and pregnancy-induced hypertension (6, 7). In our patient, the need for surgical intervention was due to failure to optimise calcium levels with conservative measures. Genetic screening informed the decision for three gland-parathyroidectomy. As our patient remained dependent on IV calcium for a few days post-operatively and failed to switch to oral calcium successfully, we changed the calcium salt to achieve successful result. For the absorption of calcium carbonate to happen, calcium carbonate must be converted to calcium chloride by gastric acid. In patients rendered achlorhydric, as with our patient secondary to high dose PPI, gastric acid levels are greatly reduced meaning that little calcium carbonate is converted into its absorbable form. In these instances, the bioavailability of the calcium carbonate drops from 30% to approximately 4%. Calcium citrate does not require gastric acid to be absorbed and this maintains its absorption rates in this patient population. In fact, calcium carbonate is the standard form of oral calcium replacement in Ireland, hence, it was one of the challenges we faced when switching the patient to calcium citrate while pregnant. Another challenge was the use of high dose PPI and its safety profile in pregnancy. This was thoroughly reviewed in the literature and PPI remain classified as class B/C in pregnancy due to lack of prospective studies but the overall profile is described as safe (8).
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this case report.
Funding
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
Patient consent
A written informed consent has been obtained from the patient for publication of the submitted article and accompanying images.
Author contribution statement
D A and P D wrote the first draft of this manuscript and were involved in collection of clinical data. All authors were involved in the writing of this manuscript and final review of the manuscript. R K C, R P, D O and D O were involved in the clinical care of these patients and collection of data. | ALFACALCIDOL, CALCIUM, CALCIUM CARBONATE, CALCIUM CITRATE, CALCIUM GLUCONATE, MAGNESIUM, OMEPRAZOLE, VITAMIN D NOS | DrugsGivenReaction | CC BY-NC-ND | 33597314 | 19,895,614 | 2021-02-17 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Off label use'. | Primary hyperparathyroidism and Zollinger Ellison syndrome during pregnancy: a case report.
Multiple endocrine neoplasia type 1 (MEN1) is a rare inherited endocrine disorder with a high rate of penetrance. The incidence of MEN1 is 1/30,000 in the general population; however, it is quite rare for a patient to present for medical attention with MEN1 for the first time in pregnancy. Primary hyperparathyroidism (PHPT) is one of the most common features of MEN1. The incidence of PHPT occurring in pregnancy is 1%. Despite advances in the medical, surgical and obstetric care over the years, management of this condition during pregnancy may be challenging. It can be difficult to identify pregnant women with PHPT requiring intervention and to monitor safely. Hypercalcemia can result in significant maternal and fetal adverse outcomes including: miscarriage, intrauterine growth restriction, preterm delivery, neonatal hypocalcaemia, pre-eclampsia and maternal nephrolithiasis. Herein, we present a case study of a lady with a strong family history of MEN1, who was biochemically proven to have PHPT and evidence of Zollinger Ellison Syndrome (ZE) on endoscopy. This patient delayed her assisted pregnancy plans for in vitro fertilization (IVF) until completion of the MEN1 workup; nevertheless, she spontaneously achieved an unplanned pregnancy. As a result, she required intervention with parathyroidectomy in the second trimester of her pregnancy as her calcium level continued to rise. This case study highlights the workup, follow up and management of MEN1 presenting with PHPT and ZE in pregnancy.
Women of childbearing age who are suspected to have a diagnosis of primary hyperparathyroidism ideally should have genetic testing and avoid pregnancy until definitive plans are in place. Zollinger Ellison syndrome in pregnancy means off-label use of high dose of proton pump inhibitors (PPI). Use of PPI in pregnancy is considered to be safe based on retrospective studies. Omeprazole, however, is FDA class C drug because of lack of large prospective studies or large case series during pregnancy. Calcium supplements in the form of calcium carbonate must be converted to calcium chloride by gastric acid in order to be absorbed, however, patients rendered achlorhydric as a result of PPI use will have impaired absorption of calcium. Therefore, use of calcium citrate might be considered a better option in this case.
Background
MEN1 is a rare genetic endocrine disorder. It is inherited in an autosomal dominant fashion. There is no gender or ethnic predilection; however, a significant proportion of these cases present in women of childbearing age. This leads to rare and difficult clinical challenges in pregnant women. The incidence of sporadic PHPT occurring in pregnancy is reported as 1% (1, 2). It is well reported that the most common form of PHPT will present as a solitary parathyroid adenoma and less than 2% will have 2 or more lesions (2). The latter is often seen in patients with an underlying genetic mutation. The greatest difficulty is to control maternal calcium level during pregnancy as calcimimetics have not been approved for use in pregnancy (3). It is reasonable to manage mild cases of hypercalcemia conservatively, but surgery may be indicated for moderate to severe cases. In non-pregnant patients, localization of the adenoma is performed by a combination of ultrasound with ionising radiation (99mTc Sestamibi scan (gold standard)/CT scan). Ultrasound is the only radiological modality that can be used in pregnancy to localize the site/sites of the disease. This is a unique case of MEN1 manifested as PHPT, with hypoparathyroidism post-operatively, and ZE syndrome in pregnancy all of which were challenging and were managed with a multidisciplinary (MDT) approach.
Case presentation
Medical history and demographics
We report the case of a 34-year-old lwoman who was referred to our service from a fertility clinic. As part of her workup for primary infertility, it was noted that her adjusted calcium was elevated at 2.84 mmol/L (normal range: 2.20–2.60 mmol/L) and PTH was elevated at 84.4 ng/L (normal range: 15–65 ng/L) (Fig. 1). She had no significant past medical history and was not on regular medications. Of note, she had no renal or bone disease. Her family history was strongly positive for MEN 1 with a genetically proven case in her brother and her mother and five out of six maternal aunts and uncles with clinical diagnosis of MEN 1. The patient was evaluated at a neuroendocrine tumour (NET) clinic and was investigated for MEN 1, with goal to correct any endocrine abnormalities prior to undergoing further treatment for fertility.
Figure 1 Pre- and post-operative ionised Ca+2, PTH, Vitamin D and phosphate.
Figure 2 Endoscopic ultrasound image revealing multiple shallow lesions in the duodenum.
Figure 3 Ultrasound neck showinghypervascular ovoid nodule.
Investigation
As part of the workup, it was noted that Gastrin level was elevated at 208 pg/mL (normal range: 29.4–121 pg/mL) and chromogranin A was 376 mg/mL (45–150 mg/mL). IGF-1, Insulin and glucagon were all within the normal range. The patient underwent endoscopy and endoscopic ultrasound (EUS); features were consistent with ZE syndrome; two to three 4–5 mm hypoechoic lesions consistent with small pNETs in the body and tail of the pancreas, and multiple shallow ulcers in the duodenum (Fig. 2). No obvious tumour in the head of the pancreas was identified and there were no suspicious nodes. She had a CT abdomen which showed a 1.2 cm nodule in the left adrenal gland. Adrenal endocrine functional workup was normal. Additionally, she had an octreotide scan which showed no site of active disease. She had a normal DXA scan and MRI of the pituitary. 99mTc Sestamibi scan demonstrated no focal tracer uptake in superior mediastinum or neck, ultrasound neck demonstrated a hypervascular ovoid nodule measuring 1.5 cm at lower pole right lobe of thyroid suspicious for a parathyroid adenoma (Fig. 3) and so the patient was referred to the endocrine surgeons for further management.
Unfortunately, when the patient presented to the surgical clinic; 12 weeks following previous investigations, she was found to be 7 weeks’ pregnant. Her calcium levels were monitored closely throughout her pregnancy and her ionised calcium remained elevated at 1.52 nmol/L (normal range: 1.19–1.35 nmol/L) with elevated parathyroid hormone (PTH) at 74 ng/L (normal range: 15–65 ng/L) and low vitamin D of 14.8 nmol/L (desired greater than 50 nmol/L). Given the concerns with regards to the use of cinacalcet in pregnancy, the patient was monitored with weekly calcium and PTH levels. She was also closely monitored for hypertension and any signs of foetal growth restriction throughout the pregnancy. Her genetic screening confirmed c831del mutation in MEN gene.
Treatment
The patient was initially commenced on high dose PPI (omeprazole 40 mg twice daily) following her endoscopic findings. The NET MDT was contacted by the maternal foetal medicine team at 24 weeks’ gestation with concerns re the continued rise in calcium levels. After reviewing the patient with an MDT approach, she was admitted electively for parathyroid exploration and subtotal parathyroidectomy. Fluid rehydration alone failed to suppress her calcium level. The procedure went well and three glands were removed, leaving the right upper gland in situ.
Outcome and follow-up
Post-operatively, the patient was transferred to ICU and commenced on IV calcium gluconate infusion. At this time, she was also commenced on oral calcium carbonate and alfacalciferol. The patient was transferred to the wards, with mother and fetus doing well. Despite increasing doses of calcium carbonate, ionised calcium levels were not maintained off infusion with the patient frequently becoming symptomatic with peri-oral paraesthesia and positive Chvostek’s sign. Magnesium replacement was also commenced. The patient remained dependent on IV calcium infusion for the first 12 postoperative days, fetus viability was closely monitored during this time. It was suggested that the patient was not absorbing calcium carbonate due to PPI use, therefore, she was switched to calcium citrate, a non-standard form of oral calcium in Ireland. IV calcium replacement was ceased 2 days later and the patient delivered at 37 weeks a healthy baby with no adverse events. Parathyroid axis recovered and the patient did not require calcium/vitamin D supplements for maintenance. She has since had a second uncomplicated pregnancy.
Discussion
Diagnosis of PHPT in pregnancy is rare, as is MEN1. There are only 155 cases reported in the literature of PHPT diagnosed in pregnancy (5). The majority of cases of PHPT in pregnancy can be managed with surveillance. It is important to establish a threshold for intervention with maternal foetal medicine. Untreated moderate to severe hypercalcemia can cause potential adverse outcomes including foetal and maternal deaths, pre-eclampsia and pregnancy-induced hypertension (6, 7). In our patient, the need for surgical intervention was due to failure to optimise calcium levels with conservative measures. Genetic screening informed the decision for three gland-parathyroidectomy. As our patient remained dependent on IV calcium for a few days post-operatively and failed to switch to oral calcium successfully, we changed the calcium salt to achieve successful result. For the absorption of calcium carbonate to happen, calcium carbonate must be converted to calcium chloride by gastric acid. In patients rendered achlorhydric, as with our patient secondary to high dose PPI, gastric acid levels are greatly reduced meaning that little calcium carbonate is converted into its absorbable form. In these instances, the bioavailability of the calcium carbonate drops from 30% to approximately 4%. Calcium citrate does not require gastric acid to be absorbed and this maintains its absorption rates in this patient population. In fact, calcium carbonate is the standard form of oral calcium replacement in Ireland, hence, it was one of the challenges we faced when switching the patient to calcium citrate while pregnant. Another challenge was the use of high dose PPI and its safety profile in pregnancy. This was thoroughly reviewed in the literature and PPI remain classified as class B/C in pregnancy due to lack of prospective studies but the overall profile is described as safe (8).
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this case report.
Funding
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
Patient consent
A written informed consent has been obtained from the patient for publication of the submitted article and accompanying images.
Author contribution statement
D A and P D wrote the first draft of this manuscript and were involved in collection of clinical data. All authors were involved in the writing of this manuscript and final review of the manuscript. R K C, R P, D O and D O were involved in the clinical care of these patients and collection of data. | ALFACALCIDOL, CALCIUM, CALCIUM CARBONATE, CALCIUM CITRATE, CALCIUM GLUCONATE, MAGNESIUM, OMEPRAZOLE, VITAMIN D NOS | DrugsGivenReaction | CC BY-NC-ND | 33597314 | 19,895,614 | 2021-02-17 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Paraesthesia oral'. | Primary hyperparathyroidism and Zollinger Ellison syndrome during pregnancy: a case report.
Multiple endocrine neoplasia type 1 (MEN1) is a rare inherited endocrine disorder with a high rate of penetrance. The incidence of MEN1 is 1/30,000 in the general population; however, it is quite rare for a patient to present for medical attention with MEN1 for the first time in pregnancy. Primary hyperparathyroidism (PHPT) is one of the most common features of MEN1. The incidence of PHPT occurring in pregnancy is 1%. Despite advances in the medical, surgical and obstetric care over the years, management of this condition during pregnancy may be challenging. It can be difficult to identify pregnant women with PHPT requiring intervention and to monitor safely. Hypercalcemia can result in significant maternal and fetal adverse outcomes including: miscarriage, intrauterine growth restriction, preterm delivery, neonatal hypocalcaemia, pre-eclampsia and maternal nephrolithiasis. Herein, we present a case study of a lady with a strong family history of MEN1, who was biochemically proven to have PHPT and evidence of Zollinger Ellison Syndrome (ZE) on endoscopy. This patient delayed her assisted pregnancy plans for in vitro fertilization (IVF) until completion of the MEN1 workup; nevertheless, she spontaneously achieved an unplanned pregnancy. As a result, she required intervention with parathyroidectomy in the second trimester of her pregnancy as her calcium level continued to rise. This case study highlights the workup, follow up and management of MEN1 presenting with PHPT and ZE in pregnancy.
Women of childbearing age who are suspected to have a diagnosis of primary hyperparathyroidism ideally should have genetic testing and avoid pregnancy until definitive plans are in place. Zollinger Ellison syndrome in pregnancy means off-label use of high dose of proton pump inhibitors (PPI). Use of PPI in pregnancy is considered to be safe based on retrospective studies. Omeprazole, however, is FDA class C drug because of lack of large prospective studies or large case series during pregnancy. Calcium supplements in the form of calcium carbonate must be converted to calcium chloride by gastric acid in order to be absorbed, however, patients rendered achlorhydric as a result of PPI use will have impaired absorption of calcium. Therefore, use of calcium citrate might be considered a better option in this case.
Background
MEN1 is a rare genetic endocrine disorder. It is inherited in an autosomal dominant fashion. There is no gender or ethnic predilection; however, a significant proportion of these cases present in women of childbearing age. This leads to rare and difficult clinical challenges in pregnant women. The incidence of sporadic PHPT occurring in pregnancy is reported as 1% (1, 2). It is well reported that the most common form of PHPT will present as a solitary parathyroid adenoma and less than 2% will have 2 or more lesions (2). The latter is often seen in patients with an underlying genetic mutation. The greatest difficulty is to control maternal calcium level during pregnancy as calcimimetics have not been approved for use in pregnancy (3). It is reasonable to manage mild cases of hypercalcemia conservatively, but surgery may be indicated for moderate to severe cases. In non-pregnant patients, localization of the adenoma is performed by a combination of ultrasound with ionising radiation (99mTc Sestamibi scan (gold standard)/CT scan). Ultrasound is the only radiological modality that can be used in pregnancy to localize the site/sites of the disease. This is a unique case of MEN1 manifested as PHPT, with hypoparathyroidism post-operatively, and ZE syndrome in pregnancy all of which were challenging and were managed with a multidisciplinary (MDT) approach.
Case presentation
Medical history and demographics
We report the case of a 34-year-old lwoman who was referred to our service from a fertility clinic. As part of her workup for primary infertility, it was noted that her adjusted calcium was elevated at 2.84 mmol/L (normal range: 2.20–2.60 mmol/L) and PTH was elevated at 84.4 ng/L (normal range: 15–65 ng/L) (Fig. 1). She had no significant past medical history and was not on regular medications. Of note, she had no renal or bone disease. Her family history was strongly positive for MEN 1 with a genetically proven case in her brother and her mother and five out of six maternal aunts and uncles with clinical diagnosis of MEN 1. The patient was evaluated at a neuroendocrine tumour (NET) clinic and was investigated for MEN 1, with goal to correct any endocrine abnormalities prior to undergoing further treatment for fertility.
Figure 1 Pre- and post-operative ionised Ca+2, PTH, Vitamin D and phosphate.
Figure 2 Endoscopic ultrasound image revealing multiple shallow lesions in the duodenum.
Figure 3 Ultrasound neck showinghypervascular ovoid nodule.
Investigation
As part of the workup, it was noted that Gastrin level was elevated at 208 pg/mL (normal range: 29.4–121 pg/mL) and chromogranin A was 376 mg/mL (45–150 mg/mL). IGF-1, Insulin and glucagon were all within the normal range. The patient underwent endoscopy and endoscopic ultrasound (EUS); features were consistent with ZE syndrome; two to three 4–5 mm hypoechoic lesions consistent with small pNETs in the body and tail of the pancreas, and multiple shallow ulcers in the duodenum (Fig. 2). No obvious tumour in the head of the pancreas was identified and there were no suspicious nodes. She had a CT abdomen which showed a 1.2 cm nodule in the left adrenal gland. Adrenal endocrine functional workup was normal. Additionally, she had an octreotide scan which showed no site of active disease. She had a normal DXA scan and MRI of the pituitary. 99mTc Sestamibi scan demonstrated no focal tracer uptake in superior mediastinum or neck, ultrasound neck demonstrated a hypervascular ovoid nodule measuring 1.5 cm at lower pole right lobe of thyroid suspicious for a parathyroid adenoma (Fig. 3) and so the patient was referred to the endocrine surgeons for further management.
Unfortunately, when the patient presented to the surgical clinic; 12 weeks following previous investigations, she was found to be 7 weeks’ pregnant. Her calcium levels were monitored closely throughout her pregnancy and her ionised calcium remained elevated at 1.52 nmol/L (normal range: 1.19–1.35 nmol/L) with elevated parathyroid hormone (PTH) at 74 ng/L (normal range: 15–65 ng/L) and low vitamin D of 14.8 nmol/L (desired greater than 50 nmol/L). Given the concerns with regards to the use of cinacalcet in pregnancy, the patient was monitored with weekly calcium and PTH levels. She was also closely monitored for hypertension and any signs of foetal growth restriction throughout the pregnancy. Her genetic screening confirmed c831del mutation in MEN gene.
Treatment
The patient was initially commenced on high dose PPI (omeprazole 40 mg twice daily) following her endoscopic findings. The NET MDT was contacted by the maternal foetal medicine team at 24 weeks’ gestation with concerns re the continued rise in calcium levels. After reviewing the patient with an MDT approach, she was admitted electively for parathyroid exploration and subtotal parathyroidectomy. Fluid rehydration alone failed to suppress her calcium level. The procedure went well and three glands were removed, leaving the right upper gland in situ.
Outcome and follow-up
Post-operatively, the patient was transferred to ICU and commenced on IV calcium gluconate infusion. At this time, she was also commenced on oral calcium carbonate and alfacalciferol. The patient was transferred to the wards, with mother and fetus doing well. Despite increasing doses of calcium carbonate, ionised calcium levels were not maintained off infusion with the patient frequently becoming symptomatic with peri-oral paraesthesia and positive Chvostek’s sign. Magnesium replacement was also commenced. The patient remained dependent on IV calcium infusion for the first 12 postoperative days, fetus viability was closely monitored during this time. It was suggested that the patient was not absorbing calcium carbonate due to PPI use, therefore, she was switched to calcium citrate, a non-standard form of oral calcium in Ireland. IV calcium replacement was ceased 2 days later and the patient delivered at 37 weeks a healthy baby with no adverse events. Parathyroid axis recovered and the patient did not require calcium/vitamin D supplements for maintenance. She has since had a second uncomplicated pregnancy.
Discussion
Diagnosis of PHPT in pregnancy is rare, as is MEN1. There are only 155 cases reported in the literature of PHPT diagnosed in pregnancy (5). The majority of cases of PHPT in pregnancy can be managed with surveillance. It is important to establish a threshold for intervention with maternal foetal medicine. Untreated moderate to severe hypercalcemia can cause potential adverse outcomes including foetal and maternal deaths, pre-eclampsia and pregnancy-induced hypertension (6, 7). In our patient, the need for surgical intervention was due to failure to optimise calcium levels with conservative measures. Genetic screening informed the decision for three gland-parathyroidectomy. As our patient remained dependent on IV calcium for a few days post-operatively and failed to switch to oral calcium successfully, we changed the calcium salt to achieve successful result. For the absorption of calcium carbonate to happen, calcium carbonate must be converted to calcium chloride by gastric acid. In patients rendered achlorhydric, as with our patient secondary to high dose PPI, gastric acid levels are greatly reduced meaning that little calcium carbonate is converted into its absorbable form. In these instances, the bioavailability of the calcium carbonate drops from 30% to approximately 4%. Calcium citrate does not require gastric acid to be absorbed and this maintains its absorption rates in this patient population. In fact, calcium carbonate is the standard form of oral calcium replacement in Ireland, hence, it was one of the challenges we faced when switching the patient to calcium citrate while pregnant. Another challenge was the use of high dose PPI and its safety profile in pregnancy. This was thoroughly reviewed in the literature and PPI remain classified as class B/C in pregnancy due to lack of prospective studies but the overall profile is described as safe (8).
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this case report.
Funding
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
Patient consent
A written informed consent has been obtained from the patient for publication of the submitted article and accompanying images.
Author contribution statement
D A and P D wrote the first draft of this manuscript and were involved in collection of clinical data. All authors were involved in the writing of this manuscript and final review of the manuscript. R K C, R P, D O and D O were involved in the clinical care of these patients and collection of data. | CALCIUM CARBONATE, CALCIUM GLUCONATE, ERGOCALCIFEROL, OMEPRAZOLE | DrugsGivenReaction | CC BY-NC-ND | 33597314 | 19,725,934 | 2021-02-17 |
What was the administration route of drug 'ALFACALCIDOL'? | Primary hyperparathyroidism and Zollinger Ellison syndrome during pregnancy: a case report.
Multiple endocrine neoplasia type 1 (MEN1) is a rare inherited endocrine disorder with a high rate of penetrance. The incidence of MEN1 is 1/30,000 in the general population; however, it is quite rare for a patient to present for medical attention with MEN1 for the first time in pregnancy. Primary hyperparathyroidism (PHPT) is one of the most common features of MEN1. The incidence of PHPT occurring in pregnancy is 1%. Despite advances in the medical, surgical and obstetric care over the years, management of this condition during pregnancy may be challenging. It can be difficult to identify pregnant women with PHPT requiring intervention and to monitor safely. Hypercalcemia can result in significant maternal and fetal adverse outcomes including: miscarriage, intrauterine growth restriction, preterm delivery, neonatal hypocalcaemia, pre-eclampsia and maternal nephrolithiasis. Herein, we present a case study of a lady with a strong family history of MEN1, who was biochemically proven to have PHPT and evidence of Zollinger Ellison Syndrome (ZE) on endoscopy. This patient delayed her assisted pregnancy plans for in vitro fertilization (IVF) until completion of the MEN1 workup; nevertheless, she spontaneously achieved an unplanned pregnancy. As a result, she required intervention with parathyroidectomy in the second trimester of her pregnancy as her calcium level continued to rise. This case study highlights the workup, follow up and management of MEN1 presenting with PHPT and ZE in pregnancy.
Women of childbearing age who are suspected to have a diagnosis of primary hyperparathyroidism ideally should have genetic testing and avoid pregnancy until definitive plans are in place. Zollinger Ellison syndrome in pregnancy means off-label use of high dose of proton pump inhibitors (PPI). Use of PPI in pregnancy is considered to be safe based on retrospective studies. Omeprazole, however, is FDA class C drug because of lack of large prospective studies or large case series during pregnancy. Calcium supplements in the form of calcium carbonate must be converted to calcium chloride by gastric acid in order to be absorbed, however, patients rendered achlorhydric as a result of PPI use will have impaired absorption of calcium. Therefore, use of calcium citrate might be considered a better option in this case.
Background
MEN1 is a rare genetic endocrine disorder. It is inherited in an autosomal dominant fashion. There is no gender or ethnic predilection; however, a significant proportion of these cases present in women of childbearing age. This leads to rare and difficult clinical challenges in pregnant women. The incidence of sporadic PHPT occurring in pregnancy is reported as 1% (1, 2). It is well reported that the most common form of PHPT will present as a solitary parathyroid adenoma and less than 2% will have 2 or more lesions (2). The latter is often seen in patients with an underlying genetic mutation. The greatest difficulty is to control maternal calcium level during pregnancy as calcimimetics have not been approved for use in pregnancy (3). It is reasonable to manage mild cases of hypercalcemia conservatively, but surgery may be indicated for moderate to severe cases. In non-pregnant patients, localization of the adenoma is performed by a combination of ultrasound with ionising radiation (99mTc Sestamibi scan (gold standard)/CT scan). Ultrasound is the only radiological modality that can be used in pregnancy to localize the site/sites of the disease. This is a unique case of MEN1 manifested as PHPT, with hypoparathyroidism post-operatively, and ZE syndrome in pregnancy all of which were challenging and were managed with a multidisciplinary (MDT) approach.
Case presentation
Medical history and demographics
We report the case of a 34-year-old lwoman who was referred to our service from a fertility clinic. As part of her workup for primary infertility, it was noted that her adjusted calcium was elevated at 2.84 mmol/L (normal range: 2.20–2.60 mmol/L) and PTH was elevated at 84.4 ng/L (normal range: 15–65 ng/L) (Fig. 1). She had no significant past medical history and was not on regular medications. Of note, she had no renal or bone disease. Her family history was strongly positive for MEN 1 with a genetically proven case in her brother and her mother and five out of six maternal aunts and uncles with clinical diagnosis of MEN 1. The patient was evaluated at a neuroendocrine tumour (NET) clinic and was investigated for MEN 1, with goal to correct any endocrine abnormalities prior to undergoing further treatment for fertility.
Figure 1 Pre- and post-operative ionised Ca+2, PTH, Vitamin D and phosphate.
Figure 2 Endoscopic ultrasound image revealing multiple shallow lesions in the duodenum.
Figure 3 Ultrasound neck showinghypervascular ovoid nodule.
Investigation
As part of the workup, it was noted that Gastrin level was elevated at 208 pg/mL (normal range: 29.4–121 pg/mL) and chromogranin A was 376 mg/mL (45–150 mg/mL). IGF-1, Insulin and glucagon were all within the normal range. The patient underwent endoscopy and endoscopic ultrasound (EUS); features were consistent with ZE syndrome; two to three 4–5 mm hypoechoic lesions consistent with small pNETs in the body and tail of the pancreas, and multiple shallow ulcers in the duodenum (Fig. 2). No obvious tumour in the head of the pancreas was identified and there were no suspicious nodes. She had a CT abdomen which showed a 1.2 cm nodule in the left adrenal gland. Adrenal endocrine functional workup was normal. Additionally, she had an octreotide scan which showed no site of active disease. She had a normal DXA scan and MRI of the pituitary. 99mTc Sestamibi scan demonstrated no focal tracer uptake in superior mediastinum or neck, ultrasound neck demonstrated a hypervascular ovoid nodule measuring 1.5 cm at lower pole right lobe of thyroid suspicious for a parathyroid adenoma (Fig. 3) and so the patient was referred to the endocrine surgeons for further management.
Unfortunately, when the patient presented to the surgical clinic; 12 weeks following previous investigations, she was found to be 7 weeks’ pregnant. Her calcium levels were monitored closely throughout her pregnancy and her ionised calcium remained elevated at 1.52 nmol/L (normal range: 1.19–1.35 nmol/L) with elevated parathyroid hormone (PTH) at 74 ng/L (normal range: 15–65 ng/L) and low vitamin D of 14.8 nmol/L (desired greater than 50 nmol/L). Given the concerns with regards to the use of cinacalcet in pregnancy, the patient was monitored with weekly calcium and PTH levels. She was also closely monitored for hypertension and any signs of foetal growth restriction throughout the pregnancy. Her genetic screening confirmed c831del mutation in MEN gene.
Treatment
The patient was initially commenced on high dose PPI (omeprazole 40 mg twice daily) following her endoscopic findings. The NET MDT was contacted by the maternal foetal medicine team at 24 weeks’ gestation with concerns re the continued rise in calcium levels. After reviewing the patient with an MDT approach, she was admitted electively for parathyroid exploration and subtotal parathyroidectomy. Fluid rehydration alone failed to suppress her calcium level. The procedure went well and three glands were removed, leaving the right upper gland in situ.
Outcome and follow-up
Post-operatively, the patient was transferred to ICU and commenced on IV calcium gluconate infusion. At this time, she was also commenced on oral calcium carbonate and alfacalciferol. The patient was transferred to the wards, with mother and fetus doing well. Despite increasing doses of calcium carbonate, ionised calcium levels were not maintained off infusion with the patient frequently becoming symptomatic with peri-oral paraesthesia and positive Chvostek’s sign. Magnesium replacement was also commenced. The patient remained dependent on IV calcium infusion for the first 12 postoperative days, fetus viability was closely monitored during this time. It was suggested that the patient was not absorbing calcium carbonate due to PPI use, therefore, she was switched to calcium citrate, a non-standard form of oral calcium in Ireland. IV calcium replacement was ceased 2 days later and the patient delivered at 37 weeks a healthy baby with no adverse events. Parathyroid axis recovered and the patient did not require calcium/vitamin D supplements for maintenance. She has since had a second uncomplicated pregnancy.
Discussion
Diagnosis of PHPT in pregnancy is rare, as is MEN1. There are only 155 cases reported in the literature of PHPT diagnosed in pregnancy (5). The majority of cases of PHPT in pregnancy can be managed with surveillance. It is important to establish a threshold for intervention with maternal foetal medicine. Untreated moderate to severe hypercalcemia can cause potential adverse outcomes including foetal and maternal deaths, pre-eclampsia and pregnancy-induced hypertension (6, 7). In our patient, the need for surgical intervention was due to failure to optimise calcium levels with conservative measures. Genetic screening informed the decision for three gland-parathyroidectomy. As our patient remained dependent on IV calcium for a few days post-operatively and failed to switch to oral calcium successfully, we changed the calcium salt to achieve successful result. For the absorption of calcium carbonate to happen, calcium carbonate must be converted to calcium chloride by gastric acid. In patients rendered achlorhydric, as with our patient secondary to high dose PPI, gastric acid levels are greatly reduced meaning that little calcium carbonate is converted into its absorbable form. In these instances, the bioavailability of the calcium carbonate drops from 30% to approximately 4%. Calcium citrate does not require gastric acid to be absorbed and this maintains its absorption rates in this patient population. In fact, calcium carbonate is the standard form of oral calcium replacement in Ireland, hence, it was one of the challenges we faced when switching the patient to calcium citrate while pregnant. Another challenge was the use of high dose PPI and its safety profile in pregnancy. This was thoroughly reviewed in the literature and PPI remain classified as class B/C in pregnancy due to lack of prospective studies but the overall profile is described as safe (8).
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this case report.
Funding
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
Patient consent
A written informed consent has been obtained from the patient for publication of the submitted article and accompanying images.
Author contribution statement
D A and P D wrote the first draft of this manuscript and were involved in collection of clinical data. All authors were involved in the writing of this manuscript and final review of the manuscript. R K C, R P, D O and D O were involved in the clinical care of these patients and collection of data. | Oral | DrugAdministrationRoute | CC BY-NC-ND | 33597314 | 19,895,614 | 2021-02-17 |
What was the administration route of drug 'ERGOCALCIFEROL'? | Primary hyperparathyroidism and Zollinger Ellison syndrome during pregnancy: a case report.
Multiple endocrine neoplasia type 1 (MEN1) is a rare inherited endocrine disorder with a high rate of penetrance. The incidence of MEN1 is 1/30,000 in the general population; however, it is quite rare for a patient to present for medical attention with MEN1 for the first time in pregnancy. Primary hyperparathyroidism (PHPT) is one of the most common features of MEN1. The incidence of PHPT occurring in pregnancy is 1%. Despite advances in the medical, surgical and obstetric care over the years, management of this condition during pregnancy may be challenging. It can be difficult to identify pregnant women with PHPT requiring intervention and to monitor safely. Hypercalcemia can result in significant maternal and fetal adverse outcomes including: miscarriage, intrauterine growth restriction, preterm delivery, neonatal hypocalcaemia, pre-eclampsia and maternal nephrolithiasis. Herein, we present a case study of a lady with a strong family history of MEN1, who was biochemically proven to have PHPT and evidence of Zollinger Ellison Syndrome (ZE) on endoscopy. This patient delayed her assisted pregnancy plans for in vitro fertilization (IVF) until completion of the MEN1 workup; nevertheless, she spontaneously achieved an unplanned pregnancy. As a result, she required intervention with parathyroidectomy in the second trimester of her pregnancy as her calcium level continued to rise. This case study highlights the workup, follow up and management of MEN1 presenting with PHPT and ZE in pregnancy.
Women of childbearing age who are suspected to have a diagnosis of primary hyperparathyroidism ideally should have genetic testing and avoid pregnancy until definitive plans are in place. Zollinger Ellison syndrome in pregnancy means off-label use of high dose of proton pump inhibitors (PPI). Use of PPI in pregnancy is considered to be safe based on retrospective studies. Omeprazole, however, is FDA class C drug because of lack of large prospective studies or large case series during pregnancy. Calcium supplements in the form of calcium carbonate must be converted to calcium chloride by gastric acid in order to be absorbed, however, patients rendered achlorhydric as a result of PPI use will have impaired absorption of calcium. Therefore, use of calcium citrate might be considered a better option in this case.
Background
MEN1 is a rare genetic endocrine disorder. It is inherited in an autosomal dominant fashion. There is no gender or ethnic predilection; however, a significant proportion of these cases present in women of childbearing age. This leads to rare and difficult clinical challenges in pregnant women. The incidence of sporadic PHPT occurring in pregnancy is reported as 1% (1, 2). It is well reported that the most common form of PHPT will present as a solitary parathyroid adenoma and less than 2% will have 2 or more lesions (2). The latter is often seen in patients with an underlying genetic mutation. The greatest difficulty is to control maternal calcium level during pregnancy as calcimimetics have not been approved for use in pregnancy (3). It is reasonable to manage mild cases of hypercalcemia conservatively, but surgery may be indicated for moderate to severe cases. In non-pregnant patients, localization of the adenoma is performed by a combination of ultrasound with ionising radiation (99mTc Sestamibi scan (gold standard)/CT scan). Ultrasound is the only radiological modality that can be used in pregnancy to localize the site/sites of the disease. This is a unique case of MEN1 manifested as PHPT, with hypoparathyroidism post-operatively, and ZE syndrome in pregnancy all of which were challenging and were managed with a multidisciplinary (MDT) approach.
Case presentation
Medical history and demographics
We report the case of a 34-year-old lwoman who was referred to our service from a fertility clinic. As part of her workup for primary infertility, it was noted that her adjusted calcium was elevated at 2.84 mmol/L (normal range: 2.20–2.60 mmol/L) and PTH was elevated at 84.4 ng/L (normal range: 15–65 ng/L) (Fig. 1). She had no significant past medical history and was not on regular medications. Of note, she had no renal or bone disease. Her family history was strongly positive for MEN 1 with a genetically proven case in her brother and her mother and five out of six maternal aunts and uncles with clinical diagnosis of MEN 1. The patient was evaluated at a neuroendocrine tumour (NET) clinic and was investigated for MEN 1, with goal to correct any endocrine abnormalities prior to undergoing further treatment for fertility.
Figure 1 Pre- and post-operative ionised Ca+2, PTH, Vitamin D and phosphate.
Figure 2 Endoscopic ultrasound image revealing multiple shallow lesions in the duodenum.
Figure 3 Ultrasound neck showinghypervascular ovoid nodule.
Investigation
As part of the workup, it was noted that Gastrin level was elevated at 208 pg/mL (normal range: 29.4–121 pg/mL) and chromogranin A was 376 mg/mL (45–150 mg/mL). IGF-1, Insulin and glucagon were all within the normal range. The patient underwent endoscopy and endoscopic ultrasound (EUS); features were consistent with ZE syndrome; two to three 4–5 mm hypoechoic lesions consistent with small pNETs in the body and tail of the pancreas, and multiple shallow ulcers in the duodenum (Fig. 2). No obvious tumour in the head of the pancreas was identified and there were no suspicious nodes. She had a CT abdomen which showed a 1.2 cm nodule in the left adrenal gland. Adrenal endocrine functional workup was normal. Additionally, she had an octreotide scan which showed no site of active disease. She had a normal DXA scan and MRI of the pituitary. 99mTc Sestamibi scan demonstrated no focal tracer uptake in superior mediastinum or neck, ultrasound neck demonstrated a hypervascular ovoid nodule measuring 1.5 cm at lower pole right lobe of thyroid suspicious for a parathyroid adenoma (Fig. 3) and so the patient was referred to the endocrine surgeons for further management.
Unfortunately, when the patient presented to the surgical clinic; 12 weeks following previous investigations, she was found to be 7 weeks’ pregnant. Her calcium levels were monitored closely throughout her pregnancy and her ionised calcium remained elevated at 1.52 nmol/L (normal range: 1.19–1.35 nmol/L) with elevated parathyroid hormone (PTH) at 74 ng/L (normal range: 15–65 ng/L) and low vitamin D of 14.8 nmol/L (desired greater than 50 nmol/L). Given the concerns with regards to the use of cinacalcet in pregnancy, the patient was monitored with weekly calcium and PTH levels. She was also closely monitored for hypertension and any signs of foetal growth restriction throughout the pregnancy. Her genetic screening confirmed c831del mutation in MEN gene.
Treatment
The patient was initially commenced on high dose PPI (omeprazole 40 mg twice daily) following her endoscopic findings. The NET MDT was contacted by the maternal foetal medicine team at 24 weeks’ gestation with concerns re the continued rise in calcium levels. After reviewing the patient with an MDT approach, she was admitted electively for parathyroid exploration and subtotal parathyroidectomy. Fluid rehydration alone failed to suppress her calcium level. The procedure went well and three glands were removed, leaving the right upper gland in situ.
Outcome and follow-up
Post-operatively, the patient was transferred to ICU and commenced on IV calcium gluconate infusion. At this time, she was also commenced on oral calcium carbonate and alfacalciferol. The patient was transferred to the wards, with mother and fetus doing well. Despite increasing doses of calcium carbonate, ionised calcium levels were not maintained off infusion with the patient frequently becoming symptomatic with peri-oral paraesthesia and positive Chvostek’s sign. Magnesium replacement was also commenced. The patient remained dependent on IV calcium infusion for the first 12 postoperative days, fetus viability was closely monitored during this time. It was suggested that the patient was not absorbing calcium carbonate due to PPI use, therefore, she was switched to calcium citrate, a non-standard form of oral calcium in Ireland. IV calcium replacement was ceased 2 days later and the patient delivered at 37 weeks a healthy baby with no adverse events. Parathyroid axis recovered and the patient did not require calcium/vitamin D supplements for maintenance. She has since had a second uncomplicated pregnancy.
Discussion
Diagnosis of PHPT in pregnancy is rare, as is MEN1. There are only 155 cases reported in the literature of PHPT diagnosed in pregnancy (5). The majority of cases of PHPT in pregnancy can be managed with surveillance. It is important to establish a threshold for intervention with maternal foetal medicine. Untreated moderate to severe hypercalcemia can cause potential adverse outcomes including foetal and maternal deaths, pre-eclampsia and pregnancy-induced hypertension (6, 7). In our patient, the need for surgical intervention was due to failure to optimise calcium levels with conservative measures. Genetic screening informed the decision for three gland-parathyroidectomy. As our patient remained dependent on IV calcium for a few days post-operatively and failed to switch to oral calcium successfully, we changed the calcium salt to achieve successful result. For the absorption of calcium carbonate to happen, calcium carbonate must be converted to calcium chloride by gastric acid. In patients rendered achlorhydric, as with our patient secondary to high dose PPI, gastric acid levels are greatly reduced meaning that little calcium carbonate is converted into its absorbable form. In these instances, the bioavailability of the calcium carbonate drops from 30% to approximately 4%. Calcium citrate does not require gastric acid to be absorbed and this maintains its absorption rates in this patient population. In fact, calcium carbonate is the standard form of oral calcium replacement in Ireland, hence, it was one of the challenges we faced when switching the patient to calcium citrate while pregnant. Another challenge was the use of high dose PPI and its safety profile in pregnancy. This was thoroughly reviewed in the literature and PPI remain classified as class B/C in pregnancy due to lack of prospective studies but the overall profile is described as safe (8).
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this case report.
Funding
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
Patient consent
A written informed consent has been obtained from the patient for publication of the submitted article and accompanying images.
Author contribution statement
D A and P D wrote the first draft of this manuscript and were involved in collection of clinical data. All authors were involved in the writing of this manuscript and final review of the manuscript. R K C, R P, D O and D O were involved in the clinical care of these patients and collection of data. | Oral | DrugAdministrationRoute | CC BY-NC-ND | 33597314 | 19,725,934 | 2021-02-17 |
What was the dosage of drug 'CALCIUM CARBONATE'? | Primary hyperparathyroidism and Zollinger Ellison syndrome during pregnancy: a case report.
Multiple endocrine neoplasia type 1 (MEN1) is a rare inherited endocrine disorder with a high rate of penetrance. The incidence of MEN1 is 1/30,000 in the general population; however, it is quite rare for a patient to present for medical attention with MEN1 for the first time in pregnancy. Primary hyperparathyroidism (PHPT) is one of the most common features of MEN1. The incidence of PHPT occurring in pregnancy is 1%. Despite advances in the medical, surgical and obstetric care over the years, management of this condition during pregnancy may be challenging. It can be difficult to identify pregnant women with PHPT requiring intervention and to monitor safely. Hypercalcemia can result in significant maternal and fetal adverse outcomes including: miscarriage, intrauterine growth restriction, preterm delivery, neonatal hypocalcaemia, pre-eclampsia and maternal nephrolithiasis. Herein, we present a case study of a lady with a strong family history of MEN1, who was biochemically proven to have PHPT and evidence of Zollinger Ellison Syndrome (ZE) on endoscopy. This patient delayed her assisted pregnancy plans for in vitro fertilization (IVF) until completion of the MEN1 workup; nevertheless, she spontaneously achieved an unplanned pregnancy. As a result, she required intervention with parathyroidectomy in the second trimester of her pregnancy as her calcium level continued to rise. This case study highlights the workup, follow up and management of MEN1 presenting with PHPT and ZE in pregnancy.
Women of childbearing age who are suspected to have a diagnosis of primary hyperparathyroidism ideally should have genetic testing and avoid pregnancy until definitive plans are in place. Zollinger Ellison syndrome in pregnancy means off-label use of high dose of proton pump inhibitors (PPI). Use of PPI in pregnancy is considered to be safe based on retrospective studies. Omeprazole, however, is FDA class C drug because of lack of large prospective studies or large case series during pregnancy. Calcium supplements in the form of calcium carbonate must be converted to calcium chloride by gastric acid in order to be absorbed, however, patients rendered achlorhydric as a result of PPI use will have impaired absorption of calcium. Therefore, use of calcium citrate might be considered a better option in this case.
Background
MEN1 is a rare genetic endocrine disorder. It is inherited in an autosomal dominant fashion. There is no gender or ethnic predilection; however, a significant proportion of these cases present in women of childbearing age. This leads to rare and difficult clinical challenges in pregnant women. The incidence of sporadic PHPT occurring in pregnancy is reported as 1% (1, 2). It is well reported that the most common form of PHPT will present as a solitary parathyroid adenoma and less than 2% will have 2 or more lesions (2). The latter is often seen in patients with an underlying genetic mutation. The greatest difficulty is to control maternal calcium level during pregnancy as calcimimetics have not been approved for use in pregnancy (3). It is reasonable to manage mild cases of hypercalcemia conservatively, but surgery may be indicated for moderate to severe cases. In non-pregnant patients, localization of the adenoma is performed by a combination of ultrasound with ionising radiation (99mTc Sestamibi scan (gold standard)/CT scan). Ultrasound is the only radiological modality that can be used in pregnancy to localize the site/sites of the disease. This is a unique case of MEN1 manifested as PHPT, with hypoparathyroidism post-operatively, and ZE syndrome in pregnancy all of which were challenging and were managed with a multidisciplinary (MDT) approach.
Case presentation
Medical history and demographics
We report the case of a 34-year-old lwoman who was referred to our service from a fertility clinic. As part of her workup for primary infertility, it was noted that her adjusted calcium was elevated at 2.84 mmol/L (normal range: 2.20–2.60 mmol/L) and PTH was elevated at 84.4 ng/L (normal range: 15–65 ng/L) (Fig. 1). She had no significant past medical history and was not on regular medications. Of note, she had no renal or bone disease. Her family history was strongly positive for MEN 1 with a genetically proven case in her brother and her mother and five out of six maternal aunts and uncles with clinical diagnosis of MEN 1. The patient was evaluated at a neuroendocrine tumour (NET) clinic and was investigated for MEN 1, with goal to correct any endocrine abnormalities prior to undergoing further treatment for fertility.
Figure 1 Pre- and post-operative ionised Ca+2, PTH, Vitamin D and phosphate.
Figure 2 Endoscopic ultrasound image revealing multiple shallow lesions in the duodenum.
Figure 3 Ultrasound neck showinghypervascular ovoid nodule.
Investigation
As part of the workup, it was noted that Gastrin level was elevated at 208 pg/mL (normal range: 29.4–121 pg/mL) and chromogranin A was 376 mg/mL (45–150 mg/mL). IGF-1, Insulin and glucagon were all within the normal range. The patient underwent endoscopy and endoscopic ultrasound (EUS); features were consistent with ZE syndrome; two to three 4–5 mm hypoechoic lesions consistent with small pNETs in the body and tail of the pancreas, and multiple shallow ulcers in the duodenum (Fig. 2). No obvious tumour in the head of the pancreas was identified and there were no suspicious nodes. She had a CT abdomen which showed a 1.2 cm nodule in the left adrenal gland. Adrenal endocrine functional workup was normal. Additionally, she had an octreotide scan which showed no site of active disease. She had a normal DXA scan and MRI of the pituitary. 99mTc Sestamibi scan demonstrated no focal tracer uptake in superior mediastinum or neck, ultrasound neck demonstrated a hypervascular ovoid nodule measuring 1.5 cm at lower pole right lobe of thyroid suspicious for a parathyroid adenoma (Fig. 3) and so the patient was referred to the endocrine surgeons for further management.
Unfortunately, when the patient presented to the surgical clinic; 12 weeks following previous investigations, she was found to be 7 weeks’ pregnant. Her calcium levels were monitored closely throughout her pregnancy and her ionised calcium remained elevated at 1.52 nmol/L (normal range: 1.19–1.35 nmol/L) with elevated parathyroid hormone (PTH) at 74 ng/L (normal range: 15–65 ng/L) and low vitamin D of 14.8 nmol/L (desired greater than 50 nmol/L). Given the concerns with regards to the use of cinacalcet in pregnancy, the patient was monitored with weekly calcium and PTH levels. She was also closely monitored for hypertension and any signs of foetal growth restriction throughout the pregnancy. Her genetic screening confirmed c831del mutation in MEN gene.
Treatment
The patient was initially commenced on high dose PPI (omeprazole 40 mg twice daily) following her endoscopic findings. The NET MDT was contacted by the maternal foetal medicine team at 24 weeks’ gestation with concerns re the continued rise in calcium levels. After reviewing the patient with an MDT approach, she was admitted electively for parathyroid exploration and subtotal parathyroidectomy. Fluid rehydration alone failed to suppress her calcium level. The procedure went well and three glands were removed, leaving the right upper gland in situ.
Outcome and follow-up
Post-operatively, the patient was transferred to ICU and commenced on IV calcium gluconate infusion. At this time, she was also commenced on oral calcium carbonate and alfacalciferol. The patient was transferred to the wards, with mother and fetus doing well. Despite increasing doses of calcium carbonate, ionised calcium levels were not maintained off infusion with the patient frequently becoming symptomatic with peri-oral paraesthesia and positive Chvostek’s sign. Magnesium replacement was also commenced. The patient remained dependent on IV calcium infusion for the first 12 postoperative days, fetus viability was closely monitored during this time. It was suggested that the patient was not absorbing calcium carbonate due to PPI use, therefore, she was switched to calcium citrate, a non-standard form of oral calcium in Ireland. IV calcium replacement was ceased 2 days later and the patient delivered at 37 weeks a healthy baby with no adverse events. Parathyroid axis recovered and the patient did not require calcium/vitamin D supplements for maintenance. She has since had a second uncomplicated pregnancy.
Discussion
Diagnosis of PHPT in pregnancy is rare, as is MEN1. There are only 155 cases reported in the literature of PHPT diagnosed in pregnancy (5). The majority of cases of PHPT in pregnancy can be managed with surveillance. It is important to establish a threshold for intervention with maternal foetal medicine. Untreated moderate to severe hypercalcemia can cause potential adverse outcomes including foetal and maternal deaths, pre-eclampsia and pregnancy-induced hypertension (6, 7). In our patient, the need for surgical intervention was due to failure to optimise calcium levels with conservative measures. Genetic screening informed the decision for three gland-parathyroidectomy. As our patient remained dependent on IV calcium for a few days post-operatively and failed to switch to oral calcium successfully, we changed the calcium salt to achieve successful result. For the absorption of calcium carbonate to happen, calcium carbonate must be converted to calcium chloride by gastric acid. In patients rendered achlorhydric, as with our patient secondary to high dose PPI, gastric acid levels are greatly reduced meaning that little calcium carbonate is converted into its absorbable form. In these instances, the bioavailability of the calcium carbonate drops from 30% to approximately 4%. Calcium citrate does not require gastric acid to be absorbed and this maintains its absorption rates in this patient population. In fact, calcium carbonate is the standard form of oral calcium replacement in Ireland, hence, it was one of the challenges we faced when switching the patient to calcium citrate while pregnant. Another challenge was the use of high dose PPI and its safety profile in pregnancy. This was thoroughly reviewed in the literature and PPI remain classified as class B/C in pregnancy due to lack of prospective studies but the overall profile is described as safe (8).
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this case report.
Funding
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
Patient consent
A written informed consent has been obtained from the patient for publication of the submitted article and accompanying images.
Author contribution statement
D A and P D wrote the first draft of this manuscript and were involved in collection of clinical data. All authors were involved in the writing of this manuscript and final review of the manuscript. R K C, R P, D O and D O were involved in the clinical care of these patients and collection of data. | INCREASING | DrugDosageText | CC BY-NC-ND | 33597314 | 19,725,934 | 2021-02-17 |
What was the outcome of reaction 'Achlorhydria'? | Primary hyperparathyroidism and Zollinger Ellison syndrome during pregnancy: a case report.
Multiple endocrine neoplasia type 1 (MEN1) is a rare inherited endocrine disorder with a high rate of penetrance. The incidence of MEN1 is 1/30,000 in the general population; however, it is quite rare for a patient to present for medical attention with MEN1 for the first time in pregnancy. Primary hyperparathyroidism (PHPT) is one of the most common features of MEN1. The incidence of PHPT occurring in pregnancy is 1%. Despite advances in the medical, surgical and obstetric care over the years, management of this condition during pregnancy may be challenging. It can be difficult to identify pregnant women with PHPT requiring intervention and to monitor safely. Hypercalcemia can result in significant maternal and fetal adverse outcomes including: miscarriage, intrauterine growth restriction, preterm delivery, neonatal hypocalcaemia, pre-eclampsia and maternal nephrolithiasis. Herein, we present a case study of a lady with a strong family history of MEN1, who was biochemically proven to have PHPT and evidence of Zollinger Ellison Syndrome (ZE) on endoscopy. This patient delayed her assisted pregnancy plans for in vitro fertilization (IVF) until completion of the MEN1 workup; nevertheless, she spontaneously achieved an unplanned pregnancy. As a result, she required intervention with parathyroidectomy in the second trimester of her pregnancy as her calcium level continued to rise. This case study highlights the workup, follow up and management of MEN1 presenting with PHPT and ZE in pregnancy.
Women of childbearing age who are suspected to have a diagnosis of primary hyperparathyroidism ideally should have genetic testing and avoid pregnancy until definitive plans are in place. Zollinger Ellison syndrome in pregnancy means off-label use of high dose of proton pump inhibitors (PPI). Use of PPI in pregnancy is considered to be safe based on retrospective studies. Omeprazole, however, is FDA class C drug because of lack of large prospective studies or large case series during pregnancy. Calcium supplements in the form of calcium carbonate must be converted to calcium chloride by gastric acid in order to be absorbed, however, patients rendered achlorhydric as a result of PPI use will have impaired absorption of calcium. Therefore, use of calcium citrate might be considered a better option in this case.
Background
MEN1 is a rare genetic endocrine disorder. It is inherited in an autosomal dominant fashion. There is no gender or ethnic predilection; however, a significant proportion of these cases present in women of childbearing age. This leads to rare and difficult clinical challenges in pregnant women. The incidence of sporadic PHPT occurring in pregnancy is reported as 1% (1, 2). It is well reported that the most common form of PHPT will present as a solitary parathyroid adenoma and less than 2% will have 2 or more lesions (2). The latter is often seen in patients with an underlying genetic mutation. The greatest difficulty is to control maternal calcium level during pregnancy as calcimimetics have not been approved for use in pregnancy (3). It is reasonable to manage mild cases of hypercalcemia conservatively, but surgery may be indicated for moderate to severe cases. In non-pregnant patients, localization of the adenoma is performed by a combination of ultrasound with ionising radiation (99mTc Sestamibi scan (gold standard)/CT scan). Ultrasound is the only radiological modality that can be used in pregnancy to localize the site/sites of the disease. This is a unique case of MEN1 manifested as PHPT, with hypoparathyroidism post-operatively, and ZE syndrome in pregnancy all of which were challenging and were managed with a multidisciplinary (MDT) approach.
Case presentation
Medical history and demographics
We report the case of a 34-year-old lwoman who was referred to our service from a fertility clinic. As part of her workup for primary infertility, it was noted that her adjusted calcium was elevated at 2.84 mmol/L (normal range: 2.20–2.60 mmol/L) and PTH was elevated at 84.4 ng/L (normal range: 15–65 ng/L) (Fig. 1). She had no significant past medical history and was not on regular medications. Of note, she had no renal or bone disease. Her family history was strongly positive for MEN 1 with a genetically proven case in her brother and her mother and five out of six maternal aunts and uncles with clinical diagnosis of MEN 1. The patient was evaluated at a neuroendocrine tumour (NET) clinic and was investigated for MEN 1, with goal to correct any endocrine abnormalities prior to undergoing further treatment for fertility.
Figure 1 Pre- and post-operative ionised Ca+2, PTH, Vitamin D and phosphate.
Figure 2 Endoscopic ultrasound image revealing multiple shallow lesions in the duodenum.
Figure 3 Ultrasound neck showinghypervascular ovoid nodule.
Investigation
As part of the workup, it was noted that Gastrin level was elevated at 208 pg/mL (normal range: 29.4–121 pg/mL) and chromogranin A was 376 mg/mL (45–150 mg/mL). IGF-1, Insulin and glucagon were all within the normal range. The patient underwent endoscopy and endoscopic ultrasound (EUS); features were consistent with ZE syndrome; two to three 4–5 mm hypoechoic lesions consistent with small pNETs in the body and tail of the pancreas, and multiple shallow ulcers in the duodenum (Fig. 2). No obvious tumour in the head of the pancreas was identified and there were no suspicious nodes. She had a CT abdomen which showed a 1.2 cm nodule in the left adrenal gland. Adrenal endocrine functional workup was normal. Additionally, she had an octreotide scan which showed no site of active disease. She had a normal DXA scan and MRI of the pituitary. 99mTc Sestamibi scan demonstrated no focal tracer uptake in superior mediastinum or neck, ultrasound neck demonstrated a hypervascular ovoid nodule measuring 1.5 cm at lower pole right lobe of thyroid suspicious for a parathyroid adenoma (Fig. 3) and so the patient was referred to the endocrine surgeons for further management.
Unfortunately, when the patient presented to the surgical clinic; 12 weeks following previous investigations, she was found to be 7 weeks’ pregnant. Her calcium levels were monitored closely throughout her pregnancy and her ionised calcium remained elevated at 1.52 nmol/L (normal range: 1.19–1.35 nmol/L) with elevated parathyroid hormone (PTH) at 74 ng/L (normal range: 15–65 ng/L) and low vitamin D of 14.8 nmol/L (desired greater than 50 nmol/L). Given the concerns with regards to the use of cinacalcet in pregnancy, the patient was monitored with weekly calcium and PTH levels. She was also closely monitored for hypertension and any signs of foetal growth restriction throughout the pregnancy. Her genetic screening confirmed c831del mutation in MEN gene.
Treatment
The patient was initially commenced on high dose PPI (omeprazole 40 mg twice daily) following her endoscopic findings. The NET MDT was contacted by the maternal foetal medicine team at 24 weeks’ gestation with concerns re the continued rise in calcium levels. After reviewing the patient with an MDT approach, she was admitted electively for parathyroid exploration and subtotal parathyroidectomy. Fluid rehydration alone failed to suppress her calcium level. The procedure went well and three glands were removed, leaving the right upper gland in situ.
Outcome and follow-up
Post-operatively, the patient was transferred to ICU and commenced on IV calcium gluconate infusion. At this time, she was also commenced on oral calcium carbonate and alfacalciferol. The patient was transferred to the wards, with mother and fetus doing well. Despite increasing doses of calcium carbonate, ionised calcium levels were not maintained off infusion with the patient frequently becoming symptomatic with peri-oral paraesthesia and positive Chvostek’s sign. Magnesium replacement was also commenced. The patient remained dependent on IV calcium infusion for the first 12 postoperative days, fetus viability was closely monitored during this time. It was suggested that the patient was not absorbing calcium carbonate due to PPI use, therefore, she was switched to calcium citrate, a non-standard form of oral calcium in Ireland. IV calcium replacement was ceased 2 days later and the patient delivered at 37 weeks a healthy baby with no adverse events. Parathyroid axis recovered and the patient did not require calcium/vitamin D supplements for maintenance. She has since had a second uncomplicated pregnancy.
Discussion
Diagnosis of PHPT in pregnancy is rare, as is MEN1. There are only 155 cases reported in the literature of PHPT diagnosed in pregnancy (5). The majority of cases of PHPT in pregnancy can be managed with surveillance. It is important to establish a threshold for intervention with maternal foetal medicine. Untreated moderate to severe hypercalcemia can cause potential adverse outcomes including foetal and maternal deaths, pre-eclampsia and pregnancy-induced hypertension (6, 7). In our patient, the need for surgical intervention was due to failure to optimise calcium levels with conservative measures. Genetic screening informed the decision for three gland-parathyroidectomy. As our patient remained dependent on IV calcium for a few days post-operatively and failed to switch to oral calcium successfully, we changed the calcium salt to achieve successful result. For the absorption of calcium carbonate to happen, calcium carbonate must be converted to calcium chloride by gastric acid. In patients rendered achlorhydric, as with our patient secondary to high dose PPI, gastric acid levels are greatly reduced meaning that little calcium carbonate is converted into its absorbable form. In these instances, the bioavailability of the calcium carbonate drops from 30% to approximately 4%. Calcium citrate does not require gastric acid to be absorbed and this maintains its absorption rates in this patient population. In fact, calcium carbonate is the standard form of oral calcium replacement in Ireland, hence, it was one of the challenges we faced when switching the patient to calcium citrate while pregnant. Another challenge was the use of high dose PPI and its safety profile in pregnancy. This was thoroughly reviewed in the literature and PPI remain classified as class B/C in pregnancy due to lack of prospective studies but the overall profile is described as safe (8).
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this case report.
Funding
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
Patient consent
A written informed consent has been obtained from the patient for publication of the submitted article and accompanying images.
Author contribution statement
D A and P D wrote the first draft of this manuscript and were involved in collection of clinical data. All authors were involved in the writing of this manuscript and final review of the manuscript. R K C, R P, D O and D O were involved in the clinical care of these patients and collection of data. | Recovering | ReactionOutcome | CC BY-NC-ND | 33597314 | 19,725,934 | 2021-02-17 |
What was the outcome of reaction 'Calcium ionised decreased'? | Primary hyperparathyroidism and Zollinger Ellison syndrome during pregnancy: a case report.
Multiple endocrine neoplasia type 1 (MEN1) is a rare inherited endocrine disorder with a high rate of penetrance. The incidence of MEN1 is 1/30,000 in the general population; however, it is quite rare for a patient to present for medical attention with MEN1 for the first time in pregnancy. Primary hyperparathyroidism (PHPT) is one of the most common features of MEN1. The incidence of PHPT occurring in pregnancy is 1%. Despite advances in the medical, surgical and obstetric care over the years, management of this condition during pregnancy may be challenging. It can be difficult to identify pregnant women with PHPT requiring intervention and to monitor safely. Hypercalcemia can result in significant maternal and fetal adverse outcomes including: miscarriage, intrauterine growth restriction, preterm delivery, neonatal hypocalcaemia, pre-eclampsia and maternal nephrolithiasis. Herein, we present a case study of a lady with a strong family history of MEN1, who was biochemically proven to have PHPT and evidence of Zollinger Ellison Syndrome (ZE) on endoscopy. This patient delayed her assisted pregnancy plans for in vitro fertilization (IVF) until completion of the MEN1 workup; nevertheless, she spontaneously achieved an unplanned pregnancy. As a result, she required intervention with parathyroidectomy in the second trimester of her pregnancy as her calcium level continued to rise. This case study highlights the workup, follow up and management of MEN1 presenting with PHPT and ZE in pregnancy.
Women of childbearing age who are suspected to have a diagnosis of primary hyperparathyroidism ideally should have genetic testing and avoid pregnancy until definitive plans are in place. Zollinger Ellison syndrome in pregnancy means off-label use of high dose of proton pump inhibitors (PPI). Use of PPI in pregnancy is considered to be safe based on retrospective studies. Omeprazole, however, is FDA class C drug because of lack of large prospective studies or large case series during pregnancy. Calcium supplements in the form of calcium carbonate must be converted to calcium chloride by gastric acid in order to be absorbed, however, patients rendered achlorhydric as a result of PPI use will have impaired absorption of calcium. Therefore, use of calcium citrate might be considered a better option in this case.
Background
MEN1 is a rare genetic endocrine disorder. It is inherited in an autosomal dominant fashion. There is no gender or ethnic predilection; however, a significant proportion of these cases present in women of childbearing age. This leads to rare and difficult clinical challenges in pregnant women. The incidence of sporadic PHPT occurring in pregnancy is reported as 1% (1, 2). It is well reported that the most common form of PHPT will present as a solitary parathyroid adenoma and less than 2% will have 2 or more lesions (2). The latter is often seen in patients with an underlying genetic mutation. The greatest difficulty is to control maternal calcium level during pregnancy as calcimimetics have not been approved for use in pregnancy (3). It is reasonable to manage mild cases of hypercalcemia conservatively, but surgery may be indicated for moderate to severe cases. In non-pregnant patients, localization of the adenoma is performed by a combination of ultrasound with ionising radiation (99mTc Sestamibi scan (gold standard)/CT scan). Ultrasound is the only radiological modality that can be used in pregnancy to localize the site/sites of the disease. This is a unique case of MEN1 manifested as PHPT, with hypoparathyroidism post-operatively, and ZE syndrome in pregnancy all of which were challenging and were managed with a multidisciplinary (MDT) approach.
Case presentation
Medical history and demographics
We report the case of a 34-year-old lwoman who was referred to our service from a fertility clinic. As part of her workup for primary infertility, it was noted that her adjusted calcium was elevated at 2.84 mmol/L (normal range: 2.20–2.60 mmol/L) and PTH was elevated at 84.4 ng/L (normal range: 15–65 ng/L) (Fig. 1). She had no significant past medical history and was not on regular medications. Of note, she had no renal or bone disease. Her family history was strongly positive for MEN 1 with a genetically proven case in her brother and her mother and five out of six maternal aunts and uncles with clinical diagnosis of MEN 1. The patient was evaluated at a neuroendocrine tumour (NET) clinic and was investigated for MEN 1, with goal to correct any endocrine abnormalities prior to undergoing further treatment for fertility.
Figure 1 Pre- and post-operative ionised Ca+2, PTH, Vitamin D and phosphate.
Figure 2 Endoscopic ultrasound image revealing multiple shallow lesions in the duodenum.
Figure 3 Ultrasound neck showinghypervascular ovoid nodule.
Investigation
As part of the workup, it was noted that Gastrin level was elevated at 208 pg/mL (normal range: 29.4–121 pg/mL) and chromogranin A was 376 mg/mL (45–150 mg/mL). IGF-1, Insulin and glucagon were all within the normal range. The patient underwent endoscopy and endoscopic ultrasound (EUS); features were consistent with ZE syndrome; two to three 4–5 mm hypoechoic lesions consistent with small pNETs in the body and tail of the pancreas, and multiple shallow ulcers in the duodenum (Fig. 2). No obvious tumour in the head of the pancreas was identified and there were no suspicious nodes. She had a CT abdomen which showed a 1.2 cm nodule in the left adrenal gland. Adrenal endocrine functional workup was normal. Additionally, she had an octreotide scan which showed no site of active disease. She had a normal DXA scan and MRI of the pituitary. 99mTc Sestamibi scan demonstrated no focal tracer uptake in superior mediastinum or neck, ultrasound neck demonstrated a hypervascular ovoid nodule measuring 1.5 cm at lower pole right lobe of thyroid suspicious for a parathyroid adenoma (Fig. 3) and so the patient was referred to the endocrine surgeons for further management.
Unfortunately, when the patient presented to the surgical clinic; 12 weeks following previous investigations, she was found to be 7 weeks’ pregnant. Her calcium levels were monitored closely throughout her pregnancy and her ionised calcium remained elevated at 1.52 nmol/L (normal range: 1.19–1.35 nmol/L) with elevated parathyroid hormone (PTH) at 74 ng/L (normal range: 15–65 ng/L) and low vitamin D of 14.8 nmol/L (desired greater than 50 nmol/L). Given the concerns with regards to the use of cinacalcet in pregnancy, the patient was monitored with weekly calcium and PTH levels. She was also closely monitored for hypertension and any signs of foetal growth restriction throughout the pregnancy. Her genetic screening confirmed c831del mutation in MEN gene.
Treatment
The patient was initially commenced on high dose PPI (omeprazole 40 mg twice daily) following her endoscopic findings. The NET MDT was contacted by the maternal foetal medicine team at 24 weeks’ gestation with concerns re the continued rise in calcium levels. After reviewing the patient with an MDT approach, she was admitted electively for parathyroid exploration and subtotal parathyroidectomy. Fluid rehydration alone failed to suppress her calcium level. The procedure went well and three glands were removed, leaving the right upper gland in situ.
Outcome and follow-up
Post-operatively, the patient was transferred to ICU and commenced on IV calcium gluconate infusion. At this time, she was also commenced on oral calcium carbonate and alfacalciferol. The patient was transferred to the wards, with mother and fetus doing well. Despite increasing doses of calcium carbonate, ionised calcium levels were not maintained off infusion with the patient frequently becoming symptomatic with peri-oral paraesthesia and positive Chvostek’s sign. Magnesium replacement was also commenced. The patient remained dependent on IV calcium infusion for the first 12 postoperative days, fetus viability was closely monitored during this time. It was suggested that the patient was not absorbing calcium carbonate due to PPI use, therefore, she was switched to calcium citrate, a non-standard form of oral calcium in Ireland. IV calcium replacement was ceased 2 days later and the patient delivered at 37 weeks a healthy baby with no adverse events. Parathyroid axis recovered and the patient did not require calcium/vitamin D supplements for maintenance. She has since had a second uncomplicated pregnancy.
Discussion
Diagnosis of PHPT in pregnancy is rare, as is MEN1. There are only 155 cases reported in the literature of PHPT diagnosed in pregnancy (5). The majority of cases of PHPT in pregnancy can be managed with surveillance. It is important to establish a threshold for intervention with maternal foetal medicine. Untreated moderate to severe hypercalcemia can cause potential adverse outcomes including foetal and maternal deaths, pre-eclampsia and pregnancy-induced hypertension (6, 7). In our patient, the need for surgical intervention was due to failure to optimise calcium levels with conservative measures. Genetic screening informed the decision for three gland-parathyroidectomy. As our patient remained dependent on IV calcium for a few days post-operatively and failed to switch to oral calcium successfully, we changed the calcium salt to achieve successful result. For the absorption of calcium carbonate to happen, calcium carbonate must be converted to calcium chloride by gastric acid. In patients rendered achlorhydric, as with our patient secondary to high dose PPI, gastric acid levels are greatly reduced meaning that little calcium carbonate is converted into its absorbable form. In these instances, the bioavailability of the calcium carbonate drops from 30% to approximately 4%. Calcium citrate does not require gastric acid to be absorbed and this maintains its absorption rates in this patient population. In fact, calcium carbonate is the standard form of oral calcium replacement in Ireland, hence, it was one of the challenges we faced when switching the patient to calcium citrate while pregnant. Another challenge was the use of high dose PPI and its safety profile in pregnancy. This was thoroughly reviewed in the literature and PPI remain classified as class B/C in pregnancy due to lack of prospective studies but the overall profile is described as safe (8).
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this case report.
Funding
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
Patient consent
A written informed consent has been obtained from the patient for publication of the submitted article and accompanying images.
Author contribution statement
D A and P D wrote the first draft of this manuscript and were involved in collection of clinical data. All authors were involved in the writing of this manuscript and final review of the manuscript. R K C, R P, D O and D O were involved in the clinical care of these patients and collection of data. | Recovering | ReactionOutcome | CC BY-NC-ND | 33597314 | 19,725,934 | 2021-02-17 |
What was the outcome of reaction 'Calcium metabolism disorder'? | Primary hyperparathyroidism and Zollinger Ellison syndrome during pregnancy: a case report.
Multiple endocrine neoplasia type 1 (MEN1) is a rare inherited endocrine disorder with a high rate of penetrance. The incidence of MEN1 is 1/30,000 in the general population; however, it is quite rare for a patient to present for medical attention with MEN1 for the first time in pregnancy. Primary hyperparathyroidism (PHPT) is one of the most common features of MEN1. The incidence of PHPT occurring in pregnancy is 1%. Despite advances in the medical, surgical and obstetric care over the years, management of this condition during pregnancy may be challenging. It can be difficult to identify pregnant women with PHPT requiring intervention and to monitor safely. Hypercalcemia can result in significant maternal and fetal adverse outcomes including: miscarriage, intrauterine growth restriction, preterm delivery, neonatal hypocalcaemia, pre-eclampsia and maternal nephrolithiasis. Herein, we present a case study of a lady with a strong family history of MEN1, who was biochemically proven to have PHPT and evidence of Zollinger Ellison Syndrome (ZE) on endoscopy. This patient delayed her assisted pregnancy plans for in vitro fertilization (IVF) until completion of the MEN1 workup; nevertheless, she spontaneously achieved an unplanned pregnancy. As a result, she required intervention with parathyroidectomy in the second trimester of her pregnancy as her calcium level continued to rise. This case study highlights the workup, follow up and management of MEN1 presenting with PHPT and ZE in pregnancy.
Women of childbearing age who are suspected to have a diagnosis of primary hyperparathyroidism ideally should have genetic testing and avoid pregnancy until definitive plans are in place. Zollinger Ellison syndrome in pregnancy means off-label use of high dose of proton pump inhibitors (PPI). Use of PPI in pregnancy is considered to be safe based on retrospective studies. Omeprazole, however, is FDA class C drug because of lack of large prospective studies or large case series during pregnancy. Calcium supplements in the form of calcium carbonate must be converted to calcium chloride by gastric acid in order to be absorbed, however, patients rendered achlorhydric as a result of PPI use will have impaired absorption of calcium. Therefore, use of calcium citrate might be considered a better option in this case.
Background
MEN1 is a rare genetic endocrine disorder. It is inherited in an autosomal dominant fashion. There is no gender or ethnic predilection; however, a significant proportion of these cases present in women of childbearing age. This leads to rare and difficult clinical challenges in pregnant women. The incidence of sporadic PHPT occurring in pregnancy is reported as 1% (1, 2). It is well reported that the most common form of PHPT will present as a solitary parathyroid adenoma and less than 2% will have 2 or more lesions (2). The latter is often seen in patients with an underlying genetic mutation. The greatest difficulty is to control maternal calcium level during pregnancy as calcimimetics have not been approved for use in pregnancy (3). It is reasonable to manage mild cases of hypercalcemia conservatively, but surgery may be indicated for moderate to severe cases. In non-pregnant patients, localization of the adenoma is performed by a combination of ultrasound with ionising radiation (99mTc Sestamibi scan (gold standard)/CT scan). Ultrasound is the only radiological modality that can be used in pregnancy to localize the site/sites of the disease. This is a unique case of MEN1 manifested as PHPT, with hypoparathyroidism post-operatively, and ZE syndrome in pregnancy all of which were challenging and were managed with a multidisciplinary (MDT) approach.
Case presentation
Medical history and demographics
We report the case of a 34-year-old lwoman who was referred to our service from a fertility clinic. As part of her workup for primary infertility, it was noted that her adjusted calcium was elevated at 2.84 mmol/L (normal range: 2.20–2.60 mmol/L) and PTH was elevated at 84.4 ng/L (normal range: 15–65 ng/L) (Fig. 1). She had no significant past medical history and was not on regular medications. Of note, she had no renal or bone disease. Her family history was strongly positive for MEN 1 with a genetically proven case in her brother and her mother and five out of six maternal aunts and uncles with clinical diagnosis of MEN 1. The patient was evaluated at a neuroendocrine tumour (NET) clinic and was investigated for MEN 1, with goal to correct any endocrine abnormalities prior to undergoing further treatment for fertility.
Figure 1 Pre- and post-operative ionised Ca+2, PTH, Vitamin D and phosphate.
Figure 2 Endoscopic ultrasound image revealing multiple shallow lesions in the duodenum.
Figure 3 Ultrasound neck showinghypervascular ovoid nodule.
Investigation
As part of the workup, it was noted that Gastrin level was elevated at 208 pg/mL (normal range: 29.4–121 pg/mL) and chromogranin A was 376 mg/mL (45–150 mg/mL). IGF-1, Insulin and glucagon were all within the normal range. The patient underwent endoscopy and endoscopic ultrasound (EUS); features were consistent with ZE syndrome; two to three 4–5 mm hypoechoic lesions consistent with small pNETs in the body and tail of the pancreas, and multiple shallow ulcers in the duodenum (Fig. 2). No obvious tumour in the head of the pancreas was identified and there were no suspicious nodes. She had a CT abdomen which showed a 1.2 cm nodule in the left adrenal gland. Adrenal endocrine functional workup was normal. Additionally, she had an octreotide scan which showed no site of active disease. She had a normal DXA scan and MRI of the pituitary. 99mTc Sestamibi scan demonstrated no focal tracer uptake in superior mediastinum or neck, ultrasound neck demonstrated a hypervascular ovoid nodule measuring 1.5 cm at lower pole right lobe of thyroid suspicious for a parathyroid adenoma (Fig. 3) and so the patient was referred to the endocrine surgeons for further management.
Unfortunately, when the patient presented to the surgical clinic; 12 weeks following previous investigations, she was found to be 7 weeks’ pregnant. Her calcium levels were monitored closely throughout her pregnancy and her ionised calcium remained elevated at 1.52 nmol/L (normal range: 1.19–1.35 nmol/L) with elevated parathyroid hormone (PTH) at 74 ng/L (normal range: 15–65 ng/L) and low vitamin D of 14.8 nmol/L (desired greater than 50 nmol/L). Given the concerns with regards to the use of cinacalcet in pregnancy, the patient was monitored with weekly calcium and PTH levels. She was also closely monitored for hypertension and any signs of foetal growth restriction throughout the pregnancy. Her genetic screening confirmed c831del mutation in MEN gene.
Treatment
The patient was initially commenced on high dose PPI (omeprazole 40 mg twice daily) following her endoscopic findings. The NET MDT was contacted by the maternal foetal medicine team at 24 weeks’ gestation with concerns re the continued rise in calcium levels. After reviewing the patient with an MDT approach, she was admitted electively for parathyroid exploration and subtotal parathyroidectomy. Fluid rehydration alone failed to suppress her calcium level. The procedure went well and three glands were removed, leaving the right upper gland in situ.
Outcome and follow-up
Post-operatively, the patient was transferred to ICU and commenced on IV calcium gluconate infusion. At this time, she was also commenced on oral calcium carbonate and alfacalciferol. The patient was transferred to the wards, with mother and fetus doing well. Despite increasing doses of calcium carbonate, ionised calcium levels were not maintained off infusion with the patient frequently becoming symptomatic with peri-oral paraesthesia and positive Chvostek’s sign. Magnesium replacement was also commenced. The patient remained dependent on IV calcium infusion for the first 12 postoperative days, fetus viability was closely monitored during this time. It was suggested that the patient was not absorbing calcium carbonate due to PPI use, therefore, she was switched to calcium citrate, a non-standard form of oral calcium in Ireland. IV calcium replacement was ceased 2 days later and the patient delivered at 37 weeks a healthy baby with no adverse events. Parathyroid axis recovered and the patient did not require calcium/vitamin D supplements for maintenance. She has since had a second uncomplicated pregnancy.
Discussion
Diagnosis of PHPT in pregnancy is rare, as is MEN1. There are only 155 cases reported in the literature of PHPT diagnosed in pregnancy (5). The majority of cases of PHPT in pregnancy can be managed with surveillance. It is important to establish a threshold for intervention with maternal foetal medicine. Untreated moderate to severe hypercalcemia can cause potential adverse outcomes including foetal and maternal deaths, pre-eclampsia and pregnancy-induced hypertension (6, 7). In our patient, the need for surgical intervention was due to failure to optimise calcium levels with conservative measures. Genetic screening informed the decision for three gland-parathyroidectomy. As our patient remained dependent on IV calcium for a few days post-operatively and failed to switch to oral calcium successfully, we changed the calcium salt to achieve successful result. For the absorption of calcium carbonate to happen, calcium carbonate must be converted to calcium chloride by gastric acid. In patients rendered achlorhydric, as with our patient secondary to high dose PPI, gastric acid levels are greatly reduced meaning that little calcium carbonate is converted into its absorbable form. In these instances, the bioavailability of the calcium carbonate drops from 30% to approximately 4%. Calcium citrate does not require gastric acid to be absorbed and this maintains its absorption rates in this patient population. In fact, calcium carbonate is the standard form of oral calcium replacement in Ireland, hence, it was one of the challenges we faced when switching the patient to calcium citrate while pregnant. Another challenge was the use of high dose PPI and its safety profile in pregnancy. This was thoroughly reviewed in the literature and PPI remain classified as class B/C in pregnancy due to lack of prospective studies but the overall profile is described as safe (8).
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this case report.
Funding
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
Patient consent
A written informed consent has been obtained from the patient for publication of the submitted article and accompanying images.
Author contribution statement
D A and P D wrote the first draft of this manuscript and were involved in collection of clinical data. All authors were involved in the writing of this manuscript and final review of the manuscript. R K C, R P, D O and D O were involved in the clinical care of these patients and collection of data. | Recovered | ReactionOutcome | CC BY-NC-ND | 33597314 | 19,895,614 | 2021-02-17 |
What was the outcome of reaction 'Chvostek^s sign'? | Primary hyperparathyroidism and Zollinger Ellison syndrome during pregnancy: a case report.
Multiple endocrine neoplasia type 1 (MEN1) is a rare inherited endocrine disorder with a high rate of penetrance. The incidence of MEN1 is 1/30,000 in the general population; however, it is quite rare for a patient to present for medical attention with MEN1 for the first time in pregnancy. Primary hyperparathyroidism (PHPT) is one of the most common features of MEN1. The incidence of PHPT occurring in pregnancy is 1%. Despite advances in the medical, surgical and obstetric care over the years, management of this condition during pregnancy may be challenging. It can be difficult to identify pregnant women with PHPT requiring intervention and to monitor safely. Hypercalcemia can result in significant maternal and fetal adverse outcomes including: miscarriage, intrauterine growth restriction, preterm delivery, neonatal hypocalcaemia, pre-eclampsia and maternal nephrolithiasis. Herein, we present a case study of a lady with a strong family history of MEN1, who was biochemically proven to have PHPT and evidence of Zollinger Ellison Syndrome (ZE) on endoscopy. This patient delayed her assisted pregnancy plans for in vitro fertilization (IVF) until completion of the MEN1 workup; nevertheless, she spontaneously achieved an unplanned pregnancy. As a result, she required intervention with parathyroidectomy in the second trimester of her pregnancy as her calcium level continued to rise. This case study highlights the workup, follow up and management of MEN1 presenting with PHPT and ZE in pregnancy.
Women of childbearing age who are suspected to have a diagnosis of primary hyperparathyroidism ideally should have genetic testing and avoid pregnancy until definitive plans are in place. Zollinger Ellison syndrome in pregnancy means off-label use of high dose of proton pump inhibitors (PPI). Use of PPI in pregnancy is considered to be safe based on retrospective studies. Omeprazole, however, is FDA class C drug because of lack of large prospective studies or large case series during pregnancy. Calcium supplements in the form of calcium carbonate must be converted to calcium chloride by gastric acid in order to be absorbed, however, patients rendered achlorhydric as a result of PPI use will have impaired absorption of calcium. Therefore, use of calcium citrate might be considered a better option in this case.
Background
MEN1 is a rare genetic endocrine disorder. It is inherited in an autosomal dominant fashion. There is no gender or ethnic predilection; however, a significant proportion of these cases present in women of childbearing age. This leads to rare and difficult clinical challenges in pregnant women. The incidence of sporadic PHPT occurring in pregnancy is reported as 1% (1, 2). It is well reported that the most common form of PHPT will present as a solitary parathyroid adenoma and less than 2% will have 2 or more lesions (2). The latter is often seen in patients with an underlying genetic mutation. The greatest difficulty is to control maternal calcium level during pregnancy as calcimimetics have not been approved for use in pregnancy (3). It is reasonable to manage mild cases of hypercalcemia conservatively, but surgery may be indicated for moderate to severe cases. In non-pregnant patients, localization of the adenoma is performed by a combination of ultrasound with ionising radiation (99mTc Sestamibi scan (gold standard)/CT scan). Ultrasound is the only radiological modality that can be used in pregnancy to localize the site/sites of the disease. This is a unique case of MEN1 manifested as PHPT, with hypoparathyroidism post-operatively, and ZE syndrome in pregnancy all of which were challenging and were managed with a multidisciplinary (MDT) approach.
Case presentation
Medical history and demographics
We report the case of a 34-year-old lwoman who was referred to our service from a fertility clinic. As part of her workup for primary infertility, it was noted that her adjusted calcium was elevated at 2.84 mmol/L (normal range: 2.20–2.60 mmol/L) and PTH was elevated at 84.4 ng/L (normal range: 15–65 ng/L) (Fig. 1). She had no significant past medical history and was not on regular medications. Of note, she had no renal or bone disease. Her family history was strongly positive for MEN 1 with a genetically proven case in her brother and her mother and five out of six maternal aunts and uncles with clinical diagnosis of MEN 1. The patient was evaluated at a neuroendocrine tumour (NET) clinic and was investigated for MEN 1, with goal to correct any endocrine abnormalities prior to undergoing further treatment for fertility.
Figure 1 Pre- and post-operative ionised Ca+2, PTH, Vitamin D and phosphate.
Figure 2 Endoscopic ultrasound image revealing multiple shallow lesions in the duodenum.
Figure 3 Ultrasound neck showinghypervascular ovoid nodule.
Investigation
As part of the workup, it was noted that Gastrin level was elevated at 208 pg/mL (normal range: 29.4–121 pg/mL) and chromogranin A was 376 mg/mL (45–150 mg/mL). IGF-1, Insulin and glucagon were all within the normal range. The patient underwent endoscopy and endoscopic ultrasound (EUS); features were consistent with ZE syndrome; two to three 4–5 mm hypoechoic lesions consistent with small pNETs in the body and tail of the pancreas, and multiple shallow ulcers in the duodenum (Fig. 2). No obvious tumour in the head of the pancreas was identified and there were no suspicious nodes. She had a CT abdomen which showed a 1.2 cm nodule in the left adrenal gland. Adrenal endocrine functional workup was normal. Additionally, she had an octreotide scan which showed no site of active disease. She had a normal DXA scan and MRI of the pituitary. 99mTc Sestamibi scan demonstrated no focal tracer uptake in superior mediastinum or neck, ultrasound neck demonstrated a hypervascular ovoid nodule measuring 1.5 cm at lower pole right lobe of thyroid suspicious for a parathyroid adenoma (Fig. 3) and so the patient was referred to the endocrine surgeons for further management.
Unfortunately, when the patient presented to the surgical clinic; 12 weeks following previous investigations, she was found to be 7 weeks’ pregnant. Her calcium levels were monitored closely throughout her pregnancy and her ionised calcium remained elevated at 1.52 nmol/L (normal range: 1.19–1.35 nmol/L) with elevated parathyroid hormone (PTH) at 74 ng/L (normal range: 15–65 ng/L) and low vitamin D of 14.8 nmol/L (desired greater than 50 nmol/L). Given the concerns with regards to the use of cinacalcet in pregnancy, the patient was monitored with weekly calcium and PTH levels. She was also closely monitored for hypertension and any signs of foetal growth restriction throughout the pregnancy. Her genetic screening confirmed c831del mutation in MEN gene.
Treatment
The patient was initially commenced on high dose PPI (omeprazole 40 mg twice daily) following her endoscopic findings. The NET MDT was contacted by the maternal foetal medicine team at 24 weeks’ gestation with concerns re the continued rise in calcium levels. After reviewing the patient with an MDT approach, she was admitted electively for parathyroid exploration and subtotal parathyroidectomy. Fluid rehydration alone failed to suppress her calcium level. The procedure went well and three glands were removed, leaving the right upper gland in situ.
Outcome and follow-up
Post-operatively, the patient was transferred to ICU and commenced on IV calcium gluconate infusion. At this time, she was also commenced on oral calcium carbonate and alfacalciferol. The patient was transferred to the wards, with mother and fetus doing well. Despite increasing doses of calcium carbonate, ionised calcium levels were not maintained off infusion with the patient frequently becoming symptomatic with peri-oral paraesthesia and positive Chvostek’s sign. Magnesium replacement was also commenced. The patient remained dependent on IV calcium infusion for the first 12 postoperative days, fetus viability was closely monitored during this time. It was suggested that the patient was not absorbing calcium carbonate due to PPI use, therefore, she was switched to calcium citrate, a non-standard form of oral calcium in Ireland. IV calcium replacement was ceased 2 days later and the patient delivered at 37 weeks a healthy baby with no adverse events. Parathyroid axis recovered and the patient did not require calcium/vitamin D supplements for maintenance. She has since had a second uncomplicated pregnancy.
Discussion
Diagnosis of PHPT in pregnancy is rare, as is MEN1. There are only 155 cases reported in the literature of PHPT diagnosed in pregnancy (5). The majority of cases of PHPT in pregnancy can be managed with surveillance. It is important to establish a threshold for intervention with maternal foetal medicine. Untreated moderate to severe hypercalcemia can cause potential adverse outcomes including foetal and maternal deaths, pre-eclampsia and pregnancy-induced hypertension (6, 7). In our patient, the need for surgical intervention was due to failure to optimise calcium levels with conservative measures. Genetic screening informed the decision for three gland-parathyroidectomy. As our patient remained dependent on IV calcium for a few days post-operatively and failed to switch to oral calcium successfully, we changed the calcium salt to achieve successful result. For the absorption of calcium carbonate to happen, calcium carbonate must be converted to calcium chloride by gastric acid. In patients rendered achlorhydric, as with our patient secondary to high dose PPI, gastric acid levels are greatly reduced meaning that little calcium carbonate is converted into its absorbable form. In these instances, the bioavailability of the calcium carbonate drops from 30% to approximately 4%. Calcium citrate does not require gastric acid to be absorbed and this maintains its absorption rates in this patient population. In fact, calcium carbonate is the standard form of oral calcium replacement in Ireland, hence, it was one of the challenges we faced when switching the patient to calcium citrate while pregnant. Another challenge was the use of high dose PPI and its safety profile in pregnancy. This was thoroughly reviewed in the literature and PPI remain classified as class B/C in pregnancy due to lack of prospective studies but the overall profile is described as safe (8).
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this case report.
Funding
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
Patient consent
A written informed consent has been obtained from the patient for publication of the submitted article and accompanying images.
Author contribution statement
D A and P D wrote the first draft of this manuscript and were involved in collection of clinical data. All authors were involved in the writing of this manuscript and final review of the manuscript. R K C, R P, D O and D O were involved in the clinical care of these patients and collection of data. | Recovering | ReactionOutcome | CC BY-NC-ND | 33597314 | 19,725,934 | 2021-02-17 |
What was the outcome of reaction 'Malabsorption'? | Primary hyperparathyroidism and Zollinger Ellison syndrome during pregnancy: a case report.
Multiple endocrine neoplasia type 1 (MEN1) is a rare inherited endocrine disorder with a high rate of penetrance. The incidence of MEN1 is 1/30,000 in the general population; however, it is quite rare for a patient to present for medical attention with MEN1 for the first time in pregnancy. Primary hyperparathyroidism (PHPT) is one of the most common features of MEN1. The incidence of PHPT occurring in pregnancy is 1%. Despite advances in the medical, surgical and obstetric care over the years, management of this condition during pregnancy may be challenging. It can be difficult to identify pregnant women with PHPT requiring intervention and to monitor safely. Hypercalcemia can result in significant maternal and fetal adverse outcomes including: miscarriage, intrauterine growth restriction, preterm delivery, neonatal hypocalcaemia, pre-eclampsia and maternal nephrolithiasis. Herein, we present a case study of a lady with a strong family history of MEN1, who was biochemically proven to have PHPT and evidence of Zollinger Ellison Syndrome (ZE) on endoscopy. This patient delayed her assisted pregnancy plans for in vitro fertilization (IVF) until completion of the MEN1 workup; nevertheless, she spontaneously achieved an unplanned pregnancy. As a result, she required intervention with parathyroidectomy in the second trimester of her pregnancy as her calcium level continued to rise. This case study highlights the workup, follow up and management of MEN1 presenting with PHPT and ZE in pregnancy.
Women of childbearing age who are suspected to have a diagnosis of primary hyperparathyroidism ideally should have genetic testing and avoid pregnancy until definitive plans are in place. Zollinger Ellison syndrome in pregnancy means off-label use of high dose of proton pump inhibitors (PPI). Use of PPI in pregnancy is considered to be safe based on retrospective studies. Omeprazole, however, is FDA class C drug because of lack of large prospective studies or large case series during pregnancy. Calcium supplements in the form of calcium carbonate must be converted to calcium chloride by gastric acid in order to be absorbed, however, patients rendered achlorhydric as a result of PPI use will have impaired absorption of calcium. Therefore, use of calcium citrate might be considered a better option in this case.
Background
MEN1 is a rare genetic endocrine disorder. It is inherited in an autosomal dominant fashion. There is no gender or ethnic predilection; however, a significant proportion of these cases present in women of childbearing age. This leads to rare and difficult clinical challenges in pregnant women. The incidence of sporadic PHPT occurring in pregnancy is reported as 1% (1, 2). It is well reported that the most common form of PHPT will present as a solitary parathyroid adenoma and less than 2% will have 2 or more lesions (2). The latter is often seen in patients with an underlying genetic mutation. The greatest difficulty is to control maternal calcium level during pregnancy as calcimimetics have not been approved for use in pregnancy (3). It is reasonable to manage mild cases of hypercalcemia conservatively, but surgery may be indicated for moderate to severe cases. In non-pregnant patients, localization of the adenoma is performed by a combination of ultrasound with ionising radiation (99mTc Sestamibi scan (gold standard)/CT scan). Ultrasound is the only radiological modality that can be used in pregnancy to localize the site/sites of the disease. This is a unique case of MEN1 manifested as PHPT, with hypoparathyroidism post-operatively, and ZE syndrome in pregnancy all of which were challenging and were managed with a multidisciplinary (MDT) approach.
Case presentation
Medical history and demographics
We report the case of a 34-year-old lwoman who was referred to our service from a fertility clinic. As part of her workup for primary infertility, it was noted that her adjusted calcium was elevated at 2.84 mmol/L (normal range: 2.20–2.60 mmol/L) and PTH was elevated at 84.4 ng/L (normal range: 15–65 ng/L) (Fig. 1). She had no significant past medical history and was not on regular medications. Of note, she had no renal or bone disease. Her family history was strongly positive for MEN 1 with a genetically proven case in her brother and her mother and five out of six maternal aunts and uncles with clinical diagnosis of MEN 1. The patient was evaluated at a neuroendocrine tumour (NET) clinic and was investigated for MEN 1, with goal to correct any endocrine abnormalities prior to undergoing further treatment for fertility.
Figure 1 Pre- and post-operative ionised Ca+2, PTH, Vitamin D and phosphate.
Figure 2 Endoscopic ultrasound image revealing multiple shallow lesions in the duodenum.
Figure 3 Ultrasound neck showinghypervascular ovoid nodule.
Investigation
As part of the workup, it was noted that Gastrin level was elevated at 208 pg/mL (normal range: 29.4–121 pg/mL) and chromogranin A was 376 mg/mL (45–150 mg/mL). IGF-1, Insulin and glucagon were all within the normal range. The patient underwent endoscopy and endoscopic ultrasound (EUS); features were consistent with ZE syndrome; two to three 4–5 mm hypoechoic lesions consistent with small pNETs in the body and tail of the pancreas, and multiple shallow ulcers in the duodenum (Fig. 2). No obvious tumour in the head of the pancreas was identified and there were no suspicious nodes. She had a CT abdomen which showed a 1.2 cm nodule in the left adrenal gland. Adrenal endocrine functional workup was normal. Additionally, she had an octreotide scan which showed no site of active disease. She had a normal DXA scan and MRI of the pituitary. 99mTc Sestamibi scan demonstrated no focal tracer uptake in superior mediastinum or neck, ultrasound neck demonstrated a hypervascular ovoid nodule measuring 1.5 cm at lower pole right lobe of thyroid suspicious for a parathyroid adenoma (Fig. 3) and so the patient was referred to the endocrine surgeons for further management.
Unfortunately, when the patient presented to the surgical clinic; 12 weeks following previous investigations, she was found to be 7 weeks’ pregnant. Her calcium levels were monitored closely throughout her pregnancy and her ionised calcium remained elevated at 1.52 nmol/L (normal range: 1.19–1.35 nmol/L) with elevated parathyroid hormone (PTH) at 74 ng/L (normal range: 15–65 ng/L) and low vitamin D of 14.8 nmol/L (desired greater than 50 nmol/L). Given the concerns with regards to the use of cinacalcet in pregnancy, the patient was monitored with weekly calcium and PTH levels. She was also closely monitored for hypertension and any signs of foetal growth restriction throughout the pregnancy. Her genetic screening confirmed c831del mutation in MEN gene.
Treatment
The patient was initially commenced on high dose PPI (omeprazole 40 mg twice daily) following her endoscopic findings. The NET MDT was contacted by the maternal foetal medicine team at 24 weeks’ gestation with concerns re the continued rise in calcium levels. After reviewing the patient with an MDT approach, she was admitted electively for parathyroid exploration and subtotal parathyroidectomy. Fluid rehydration alone failed to suppress her calcium level. The procedure went well and three glands were removed, leaving the right upper gland in situ.
Outcome and follow-up
Post-operatively, the patient was transferred to ICU and commenced on IV calcium gluconate infusion. At this time, she was also commenced on oral calcium carbonate and alfacalciferol. The patient was transferred to the wards, with mother and fetus doing well. Despite increasing doses of calcium carbonate, ionised calcium levels were not maintained off infusion with the patient frequently becoming symptomatic with peri-oral paraesthesia and positive Chvostek’s sign. Magnesium replacement was also commenced. The patient remained dependent on IV calcium infusion for the first 12 postoperative days, fetus viability was closely monitored during this time. It was suggested that the patient was not absorbing calcium carbonate due to PPI use, therefore, she was switched to calcium citrate, a non-standard form of oral calcium in Ireland. IV calcium replacement was ceased 2 days later and the patient delivered at 37 weeks a healthy baby with no adverse events. Parathyroid axis recovered and the patient did not require calcium/vitamin D supplements for maintenance. She has since had a second uncomplicated pregnancy.
Discussion
Diagnosis of PHPT in pregnancy is rare, as is MEN1. There are only 155 cases reported in the literature of PHPT diagnosed in pregnancy (5). The majority of cases of PHPT in pregnancy can be managed with surveillance. It is important to establish a threshold for intervention with maternal foetal medicine. Untreated moderate to severe hypercalcemia can cause potential adverse outcomes including foetal and maternal deaths, pre-eclampsia and pregnancy-induced hypertension (6, 7). In our patient, the need for surgical intervention was due to failure to optimise calcium levels with conservative measures. Genetic screening informed the decision for three gland-parathyroidectomy. As our patient remained dependent on IV calcium for a few days post-operatively and failed to switch to oral calcium successfully, we changed the calcium salt to achieve successful result. For the absorption of calcium carbonate to happen, calcium carbonate must be converted to calcium chloride by gastric acid. In patients rendered achlorhydric, as with our patient secondary to high dose PPI, gastric acid levels are greatly reduced meaning that little calcium carbonate is converted into its absorbable form. In these instances, the bioavailability of the calcium carbonate drops from 30% to approximately 4%. Calcium citrate does not require gastric acid to be absorbed and this maintains its absorption rates in this patient population. In fact, calcium carbonate is the standard form of oral calcium replacement in Ireland, hence, it was one of the challenges we faced when switching the patient to calcium citrate while pregnant. Another challenge was the use of high dose PPI and its safety profile in pregnancy. This was thoroughly reviewed in the literature and PPI remain classified as class B/C in pregnancy due to lack of prospective studies but the overall profile is described as safe (8).
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this case report.
Funding
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
Patient consent
A written informed consent has been obtained from the patient for publication of the submitted article and accompanying images.
Author contribution statement
D A and P D wrote the first draft of this manuscript and were involved in collection of clinical data. All authors were involved in the writing of this manuscript and final review of the manuscript. R K C, R P, D O and D O were involved in the clinical care of these patients and collection of data. | Recovered | ReactionOutcome | CC BY-NC-ND | 33597314 | 19,930,939 | 2021-02-17 |
What was the outcome of reaction 'Normal newborn'? | Primary hyperparathyroidism and Zollinger Ellison syndrome during pregnancy: a case report.
Multiple endocrine neoplasia type 1 (MEN1) is a rare inherited endocrine disorder with a high rate of penetrance. The incidence of MEN1 is 1/30,000 in the general population; however, it is quite rare for a patient to present for medical attention with MEN1 for the first time in pregnancy. Primary hyperparathyroidism (PHPT) is one of the most common features of MEN1. The incidence of PHPT occurring in pregnancy is 1%. Despite advances in the medical, surgical and obstetric care over the years, management of this condition during pregnancy may be challenging. It can be difficult to identify pregnant women with PHPT requiring intervention and to monitor safely. Hypercalcemia can result in significant maternal and fetal adverse outcomes including: miscarriage, intrauterine growth restriction, preterm delivery, neonatal hypocalcaemia, pre-eclampsia and maternal nephrolithiasis. Herein, we present a case study of a lady with a strong family history of MEN1, who was biochemically proven to have PHPT and evidence of Zollinger Ellison Syndrome (ZE) on endoscopy. This patient delayed her assisted pregnancy plans for in vitro fertilization (IVF) until completion of the MEN1 workup; nevertheless, she spontaneously achieved an unplanned pregnancy. As a result, she required intervention with parathyroidectomy in the second trimester of her pregnancy as her calcium level continued to rise. This case study highlights the workup, follow up and management of MEN1 presenting with PHPT and ZE in pregnancy.
Women of childbearing age who are suspected to have a diagnosis of primary hyperparathyroidism ideally should have genetic testing and avoid pregnancy until definitive plans are in place. Zollinger Ellison syndrome in pregnancy means off-label use of high dose of proton pump inhibitors (PPI). Use of PPI in pregnancy is considered to be safe based on retrospective studies. Omeprazole, however, is FDA class C drug because of lack of large prospective studies or large case series during pregnancy. Calcium supplements in the form of calcium carbonate must be converted to calcium chloride by gastric acid in order to be absorbed, however, patients rendered achlorhydric as a result of PPI use will have impaired absorption of calcium. Therefore, use of calcium citrate might be considered a better option in this case.
Background
MEN1 is a rare genetic endocrine disorder. It is inherited in an autosomal dominant fashion. There is no gender or ethnic predilection; however, a significant proportion of these cases present in women of childbearing age. This leads to rare and difficult clinical challenges in pregnant women. The incidence of sporadic PHPT occurring in pregnancy is reported as 1% (1, 2). It is well reported that the most common form of PHPT will present as a solitary parathyroid adenoma and less than 2% will have 2 or more lesions (2). The latter is often seen in patients with an underlying genetic mutation. The greatest difficulty is to control maternal calcium level during pregnancy as calcimimetics have not been approved for use in pregnancy (3). It is reasonable to manage mild cases of hypercalcemia conservatively, but surgery may be indicated for moderate to severe cases. In non-pregnant patients, localization of the adenoma is performed by a combination of ultrasound with ionising radiation (99mTc Sestamibi scan (gold standard)/CT scan). Ultrasound is the only radiological modality that can be used in pregnancy to localize the site/sites of the disease. This is a unique case of MEN1 manifested as PHPT, with hypoparathyroidism post-operatively, and ZE syndrome in pregnancy all of which were challenging and were managed with a multidisciplinary (MDT) approach.
Case presentation
Medical history and demographics
We report the case of a 34-year-old lwoman who was referred to our service from a fertility clinic. As part of her workup for primary infertility, it was noted that her adjusted calcium was elevated at 2.84 mmol/L (normal range: 2.20–2.60 mmol/L) and PTH was elevated at 84.4 ng/L (normal range: 15–65 ng/L) (Fig. 1). She had no significant past medical history and was not on regular medications. Of note, she had no renal or bone disease. Her family history was strongly positive for MEN 1 with a genetically proven case in her brother and her mother and five out of six maternal aunts and uncles with clinical diagnosis of MEN 1. The patient was evaluated at a neuroendocrine tumour (NET) clinic and was investigated for MEN 1, with goal to correct any endocrine abnormalities prior to undergoing further treatment for fertility.
Figure 1 Pre- and post-operative ionised Ca+2, PTH, Vitamin D and phosphate.
Figure 2 Endoscopic ultrasound image revealing multiple shallow lesions in the duodenum.
Figure 3 Ultrasound neck showinghypervascular ovoid nodule.
Investigation
As part of the workup, it was noted that Gastrin level was elevated at 208 pg/mL (normal range: 29.4–121 pg/mL) and chromogranin A was 376 mg/mL (45–150 mg/mL). IGF-1, Insulin and glucagon were all within the normal range. The patient underwent endoscopy and endoscopic ultrasound (EUS); features were consistent with ZE syndrome; two to three 4–5 mm hypoechoic lesions consistent with small pNETs in the body and tail of the pancreas, and multiple shallow ulcers in the duodenum (Fig. 2). No obvious tumour in the head of the pancreas was identified and there were no suspicious nodes. She had a CT abdomen which showed a 1.2 cm nodule in the left adrenal gland. Adrenal endocrine functional workup was normal. Additionally, she had an octreotide scan which showed no site of active disease. She had a normal DXA scan and MRI of the pituitary. 99mTc Sestamibi scan demonstrated no focal tracer uptake in superior mediastinum or neck, ultrasound neck demonstrated a hypervascular ovoid nodule measuring 1.5 cm at lower pole right lobe of thyroid suspicious for a parathyroid adenoma (Fig. 3) and so the patient was referred to the endocrine surgeons for further management.
Unfortunately, when the patient presented to the surgical clinic; 12 weeks following previous investigations, she was found to be 7 weeks’ pregnant. Her calcium levels were monitored closely throughout her pregnancy and her ionised calcium remained elevated at 1.52 nmol/L (normal range: 1.19–1.35 nmol/L) with elevated parathyroid hormone (PTH) at 74 ng/L (normal range: 15–65 ng/L) and low vitamin D of 14.8 nmol/L (desired greater than 50 nmol/L). Given the concerns with regards to the use of cinacalcet in pregnancy, the patient was monitored with weekly calcium and PTH levels. She was also closely monitored for hypertension and any signs of foetal growth restriction throughout the pregnancy. Her genetic screening confirmed c831del mutation in MEN gene.
Treatment
The patient was initially commenced on high dose PPI (omeprazole 40 mg twice daily) following her endoscopic findings. The NET MDT was contacted by the maternal foetal medicine team at 24 weeks’ gestation with concerns re the continued rise in calcium levels. After reviewing the patient with an MDT approach, she was admitted electively for parathyroid exploration and subtotal parathyroidectomy. Fluid rehydration alone failed to suppress her calcium level. The procedure went well and three glands were removed, leaving the right upper gland in situ.
Outcome and follow-up
Post-operatively, the patient was transferred to ICU and commenced on IV calcium gluconate infusion. At this time, she was also commenced on oral calcium carbonate and alfacalciferol. The patient was transferred to the wards, with mother and fetus doing well. Despite increasing doses of calcium carbonate, ionised calcium levels were not maintained off infusion with the patient frequently becoming symptomatic with peri-oral paraesthesia and positive Chvostek’s sign. Magnesium replacement was also commenced. The patient remained dependent on IV calcium infusion for the first 12 postoperative days, fetus viability was closely monitored during this time. It was suggested that the patient was not absorbing calcium carbonate due to PPI use, therefore, she was switched to calcium citrate, a non-standard form of oral calcium in Ireland. IV calcium replacement was ceased 2 days later and the patient delivered at 37 weeks a healthy baby with no adverse events. Parathyroid axis recovered and the patient did not require calcium/vitamin D supplements for maintenance. She has since had a second uncomplicated pregnancy.
Discussion
Diagnosis of PHPT in pregnancy is rare, as is MEN1. There are only 155 cases reported in the literature of PHPT diagnosed in pregnancy (5). The majority of cases of PHPT in pregnancy can be managed with surveillance. It is important to establish a threshold for intervention with maternal foetal medicine. Untreated moderate to severe hypercalcemia can cause potential adverse outcomes including foetal and maternal deaths, pre-eclampsia and pregnancy-induced hypertension (6, 7). In our patient, the need for surgical intervention was due to failure to optimise calcium levels with conservative measures. Genetic screening informed the decision for three gland-parathyroidectomy. As our patient remained dependent on IV calcium for a few days post-operatively and failed to switch to oral calcium successfully, we changed the calcium salt to achieve successful result. For the absorption of calcium carbonate to happen, calcium carbonate must be converted to calcium chloride by gastric acid. In patients rendered achlorhydric, as with our patient secondary to high dose PPI, gastric acid levels are greatly reduced meaning that little calcium carbonate is converted into its absorbable form. In these instances, the bioavailability of the calcium carbonate drops from 30% to approximately 4%. Calcium citrate does not require gastric acid to be absorbed and this maintains its absorption rates in this patient population. In fact, calcium carbonate is the standard form of oral calcium replacement in Ireland, hence, it was one of the challenges we faced when switching the patient to calcium citrate while pregnant. Another challenge was the use of high dose PPI and its safety profile in pregnancy. This was thoroughly reviewed in the literature and PPI remain classified as class B/C in pregnancy due to lack of prospective studies but the overall profile is described as safe (8).
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this case report.
Funding
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
Patient consent
A written informed consent has been obtained from the patient for publication of the submitted article and accompanying images.
Author contribution statement
D A and P D wrote the first draft of this manuscript and were involved in collection of clinical data. All authors were involved in the writing of this manuscript and final review of the manuscript. R K C, R P, D O and D O were involved in the clinical care of these patients and collection of data. | Recovered | ReactionOutcome | CC BY-NC-ND | 33597314 | 19,846,274 | 2021-02-17 |
What was the outcome of reaction 'Paraesthesia oral'? | Primary hyperparathyroidism and Zollinger Ellison syndrome during pregnancy: a case report.
Multiple endocrine neoplasia type 1 (MEN1) is a rare inherited endocrine disorder with a high rate of penetrance. The incidence of MEN1 is 1/30,000 in the general population; however, it is quite rare for a patient to present for medical attention with MEN1 for the first time in pregnancy. Primary hyperparathyroidism (PHPT) is one of the most common features of MEN1. The incidence of PHPT occurring in pregnancy is 1%. Despite advances in the medical, surgical and obstetric care over the years, management of this condition during pregnancy may be challenging. It can be difficult to identify pregnant women with PHPT requiring intervention and to monitor safely. Hypercalcemia can result in significant maternal and fetal adverse outcomes including: miscarriage, intrauterine growth restriction, preterm delivery, neonatal hypocalcaemia, pre-eclampsia and maternal nephrolithiasis. Herein, we present a case study of a lady with a strong family history of MEN1, who was biochemically proven to have PHPT and evidence of Zollinger Ellison Syndrome (ZE) on endoscopy. This patient delayed her assisted pregnancy plans for in vitro fertilization (IVF) until completion of the MEN1 workup; nevertheless, she spontaneously achieved an unplanned pregnancy. As a result, she required intervention with parathyroidectomy in the second trimester of her pregnancy as her calcium level continued to rise. This case study highlights the workup, follow up and management of MEN1 presenting with PHPT and ZE in pregnancy.
Women of childbearing age who are suspected to have a diagnosis of primary hyperparathyroidism ideally should have genetic testing and avoid pregnancy until definitive plans are in place. Zollinger Ellison syndrome in pregnancy means off-label use of high dose of proton pump inhibitors (PPI). Use of PPI in pregnancy is considered to be safe based on retrospective studies. Omeprazole, however, is FDA class C drug because of lack of large prospective studies or large case series during pregnancy. Calcium supplements in the form of calcium carbonate must be converted to calcium chloride by gastric acid in order to be absorbed, however, patients rendered achlorhydric as a result of PPI use will have impaired absorption of calcium. Therefore, use of calcium citrate might be considered a better option in this case.
Background
MEN1 is a rare genetic endocrine disorder. It is inherited in an autosomal dominant fashion. There is no gender or ethnic predilection; however, a significant proportion of these cases present in women of childbearing age. This leads to rare and difficult clinical challenges in pregnant women. The incidence of sporadic PHPT occurring in pregnancy is reported as 1% (1, 2). It is well reported that the most common form of PHPT will present as a solitary parathyroid adenoma and less than 2% will have 2 or more lesions (2). The latter is often seen in patients with an underlying genetic mutation. The greatest difficulty is to control maternal calcium level during pregnancy as calcimimetics have not been approved for use in pregnancy (3). It is reasonable to manage mild cases of hypercalcemia conservatively, but surgery may be indicated for moderate to severe cases. In non-pregnant patients, localization of the adenoma is performed by a combination of ultrasound with ionising radiation (99mTc Sestamibi scan (gold standard)/CT scan). Ultrasound is the only radiological modality that can be used in pregnancy to localize the site/sites of the disease. This is a unique case of MEN1 manifested as PHPT, with hypoparathyroidism post-operatively, and ZE syndrome in pregnancy all of which were challenging and were managed with a multidisciplinary (MDT) approach.
Case presentation
Medical history and demographics
We report the case of a 34-year-old lwoman who was referred to our service from a fertility clinic. As part of her workup for primary infertility, it was noted that her adjusted calcium was elevated at 2.84 mmol/L (normal range: 2.20–2.60 mmol/L) and PTH was elevated at 84.4 ng/L (normal range: 15–65 ng/L) (Fig. 1). She had no significant past medical history and was not on regular medications. Of note, she had no renal or bone disease. Her family history was strongly positive for MEN 1 with a genetically proven case in her brother and her mother and five out of six maternal aunts and uncles with clinical diagnosis of MEN 1. The patient was evaluated at a neuroendocrine tumour (NET) clinic and was investigated for MEN 1, with goal to correct any endocrine abnormalities prior to undergoing further treatment for fertility.
Figure 1 Pre- and post-operative ionised Ca+2, PTH, Vitamin D and phosphate.
Figure 2 Endoscopic ultrasound image revealing multiple shallow lesions in the duodenum.
Figure 3 Ultrasound neck showinghypervascular ovoid nodule.
Investigation
As part of the workup, it was noted that Gastrin level was elevated at 208 pg/mL (normal range: 29.4–121 pg/mL) and chromogranin A was 376 mg/mL (45–150 mg/mL). IGF-1, Insulin and glucagon were all within the normal range. The patient underwent endoscopy and endoscopic ultrasound (EUS); features were consistent with ZE syndrome; two to three 4–5 mm hypoechoic lesions consistent with small pNETs in the body and tail of the pancreas, and multiple shallow ulcers in the duodenum (Fig. 2). No obvious tumour in the head of the pancreas was identified and there were no suspicious nodes. She had a CT abdomen which showed a 1.2 cm nodule in the left adrenal gland. Adrenal endocrine functional workup was normal. Additionally, she had an octreotide scan which showed no site of active disease. She had a normal DXA scan and MRI of the pituitary. 99mTc Sestamibi scan demonstrated no focal tracer uptake in superior mediastinum or neck, ultrasound neck demonstrated a hypervascular ovoid nodule measuring 1.5 cm at lower pole right lobe of thyroid suspicious for a parathyroid adenoma (Fig. 3) and so the patient was referred to the endocrine surgeons for further management.
Unfortunately, when the patient presented to the surgical clinic; 12 weeks following previous investigations, she was found to be 7 weeks’ pregnant. Her calcium levels were monitored closely throughout her pregnancy and her ionised calcium remained elevated at 1.52 nmol/L (normal range: 1.19–1.35 nmol/L) with elevated parathyroid hormone (PTH) at 74 ng/L (normal range: 15–65 ng/L) and low vitamin D of 14.8 nmol/L (desired greater than 50 nmol/L). Given the concerns with regards to the use of cinacalcet in pregnancy, the patient was monitored with weekly calcium and PTH levels. She was also closely monitored for hypertension and any signs of foetal growth restriction throughout the pregnancy. Her genetic screening confirmed c831del mutation in MEN gene.
Treatment
The patient was initially commenced on high dose PPI (omeprazole 40 mg twice daily) following her endoscopic findings. The NET MDT was contacted by the maternal foetal medicine team at 24 weeks’ gestation with concerns re the continued rise in calcium levels. After reviewing the patient with an MDT approach, she was admitted electively for parathyroid exploration and subtotal parathyroidectomy. Fluid rehydration alone failed to suppress her calcium level. The procedure went well and three glands were removed, leaving the right upper gland in situ.
Outcome and follow-up
Post-operatively, the patient was transferred to ICU and commenced on IV calcium gluconate infusion. At this time, she was also commenced on oral calcium carbonate and alfacalciferol. The patient was transferred to the wards, with mother and fetus doing well. Despite increasing doses of calcium carbonate, ionised calcium levels were not maintained off infusion with the patient frequently becoming symptomatic with peri-oral paraesthesia and positive Chvostek’s sign. Magnesium replacement was also commenced. The patient remained dependent on IV calcium infusion for the first 12 postoperative days, fetus viability was closely monitored during this time. It was suggested that the patient was not absorbing calcium carbonate due to PPI use, therefore, she was switched to calcium citrate, a non-standard form of oral calcium in Ireland. IV calcium replacement was ceased 2 days later and the patient delivered at 37 weeks a healthy baby with no adverse events. Parathyroid axis recovered and the patient did not require calcium/vitamin D supplements for maintenance. She has since had a second uncomplicated pregnancy.
Discussion
Diagnosis of PHPT in pregnancy is rare, as is MEN1. There are only 155 cases reported in the literature of PHPT diagnosed in pregnancy (5). The majority of cases of PHPT in pregnancy can be managed with surveillance. It is important to establish a threshold for intervention with maternal foetal medicine. Untreated moderate to severe hypercalcemia can cause potential adverse outcomes including foetal and maternal deaths, pre-eclampsia and pregnancy-induced hypertension (6, 7). In our patient, the need for surgical intervention was due to failure to optimise calcium levels with conservative measures. Genetic screening informed the decision for three gland-parathyroidectomy. As our patient remained dependent on IV calcium for a few days post-operatively and failed to switch to oral calcium successfully, we changed the calcium salt to achieve successful result. For the absorption of calcium carbonate to happen, calcium carbonate must be converted to calcium chloride by gastric acid. In patients rendered achlorhydric, as with our patient secondary to high dose PPI, gastric acid levels are greatly reduced meaning that little calcium carbonate is converted into its absorbable form. In these instances, the bioavailability of the calcium carbonate drops from 30% to approximately 4%. Calcium citrate does not require gastric acid to be absorbed and this maintains its absorption rates in this patient population. In fact, calcium carbonate is the standard form of oral calcium replacement in Ireland, hence, it was one of the challenges we faced when switching the patient to calcium citrate while pregnant. Another challenge was the use of high dose PPI and its safety profile in pregnancy. This was thoroughly reviewed in the literature and PPI remain classified as class B/C in pregnancy due to lack of prospective studies but the overall profile is described as safe (8).
Declaration of interest
The authors declare that there is no conflict of interest that could be perceived as prejudicing the impartiality of this case report.
Funding
This research did not receive any specific grant from any funding agency in the public, commercial or not-for-profit sector.
Patient consent
A written informed consent has been obtained from the patient for publication of the submitted article and accompanying images.
Author contribution statement
D A and P D wrote the first draft of this manuscript and were involved in collection of clinical data. All authors were involved in the writing of this manuscript and final review of the manuscript. R K C, R P, D O and D O were involved in the clinical care of these patients and collection of data. | Recovering | ReactionOutcome | CC BY-NC-ND | 33597314 | 19,725,934 | 2021-02-17 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Drug ineffective'. | Ramucirumab-Induced Hepatocellular Carcinoma Rupture and Gastrointestinal Perforation.
BACKGROUND Hepatocellular carcinoma (HCC) is a primary liver malignant tumor that typically but not always develops in the setting of chronic liver disease, particularly in patients with cirrhosis or chronic hepatitis B virus infection. Advanced HCC portends a poor prognosis; however, recent advances in first-line and second-line treatment options yield significant survival improvements. Ruptured HCC is an uncommon presentation that occurs in approximately 3-26% of patients. CASE REPORT We present a case of a patient with HCC who was undergoing treatment with the antiangiogenic monoclonal antibody ramucirumab. Subsequently, he presented with signs and symptoms of acute abdomen. The abdominal imaging revealed pneumoperitoneum with multiple abdominal and pelvic collections. The patient underwent exploratory laparotomy and was found to have necrotic liver parenchyma, which appeared to be perforated. Also, a microperforation was noted in the proximal duodenum. The pathology report from liver specimens showed fragments of hepatocellular cancer with extensive necrosis. CONCLUSIONS The mechanism of tumor rupture in HCC is poorly understood. The so-called vascular injury hypothesis states that collagen expansion and elastin proliferation in the arterial wall supplying the tumor could be the leading cause of HCC rupture. We believe that the process mentioned above was accelerated in our patient using the antiangiogenic factor ramucirumab. A similar antiangiogenic mechanism is also implicated in gastrointestinal hemorrhage and perforation related to this drug.
Background
Hepatocellular carcinoma (HCC) is the fourth most common cause of cancer-related mortality worldwide, with over 780,000 deaths in 2018. With a 5-year survival of 18%, liver cancer is the second most lethal tumor after pancreatic cancer [1]. Ramucirumab is a recombinant human immunoglobulin G1 monoclonal antibody that binds to the extracellular binding domain of vascular endothelial growth factor receptor 2 (VEGFR-2) and prevents the binding of VEGFR ligands, thus inhibiting the angiogenesis pathways involved in the development and progression of cancer [2]. It is one of the treatment modalities in treating HCC and is known to be associated with hypertension, proteinuria, hemorrhage, and gastrointestinal (GI) perforation. We present a rare case where ramucirumab was associated with tumor rupture in a known case of HCC and duodenal perforation.
Case Report
We report a case of a 66-year-old man with medical comorbidities of stage 3b HCC, liver cirrhosis, chronic hepatitis C (Child-Turcotte-Pugh B), and hypothyroidism who presented to the emergency department with worsening altered mental status. The patient was diagnosed with HCC 2 years before the presentation. The mass was 3 cm in size and was located in segments V and VI in the right lobe of the liver (Figure 1). He was initially started on treatment with pembrolizumab and Lenvatinib (as part of study treatment). Later on, because of treatment failure, therapy was changed to ramucirumab, an approved second-line treatment in the management of HCC. Four months after starting ramucirumab, he presented to our hospital. His initial vital signs showed blood pressure of 145/90 mmHg, pulse of 92 beats/min, respiratory rate of 17 breaths/min, and oxygen saturation of 98% on room air. Physical examination showed no focal neurological deficits, widespread abdominal tenderness, and absent bowel sounds. Laboratory parameters were as follows: hemoglobin 11.8 g/dL, leukocytes 14×10/μL, alanine aminotransferase 17 U/L, aspartate aminotransferase 54 U/L, total bilirubin 2.0 mg/dL, direct bilirubin 1.0 mg/dL, ammonia 91 μmol/L, and lactate 4.7 mmol/L. The other parameters were within the normal range. His baseline liver function tests (1 month before presentation) were as follows: alanine aminotransferase 15 U/L, aspartate aminotransferase 50 U/L, total bilirubin 1.7 mg/dl, international normalized ratio 1.6, and albumin 2.8 g/L. Computed tomography (CT) of the head and chest X-ray were negative for any significant acute pathology. CT of the abdomen and pelvis with contrast material revealed pneumoperitoneum with multiple abdominal and pelvic collections (Figure 2).
In the setting of leukocytosis, lactic acidosis, peritoneal signs on physical examination, and CT scan with multiple fluid collection areas, the initial impression was suspected gastric or duodenal perforation. The patient was started on broad-spectrum antibiotics and underwent exploratory laparotomy and was found to have a nodular shrunken liver with necrotic parenchyma in segments V and VI of the liver, which appeared to be perforated. Also, a microperforation was noted in the proximal duodenum. Necrotic tissue was removed, the omentum was packed in the liver bed, and 6 L of ascitic fluid was drained. The pathology sample (taken from necrotic liver tissue) showed fragments of hepatocellular carcinoma with extensive necrosis (Figure 3). The peritoneal fluid analysis was consistent with secondary bacterial peritonitis (white blood cell count of 6200 cells/mm3 with 92% neutrophils), and cytology was negative for malignant cells.
Despite antibiotic therapy with piperacillin-tazobactam, the patient developed multiorgan dysfunction. Therefore, we implemented a palliative therapy approach, and the patient was discharged to hospice care.
Discussion
In 85–95% of cases, HCC develops in patients with cirrhotic liver. Studies have shown that once cirrhosis has developed, HCC will occur at a rate of 2–7% per year [3]. The incidence/mortality ratio in HCC is almost 1: 1, which shows that most of the patients with HCC ultimately die [4]. The incidence of spontaneous rupture is <3% in western countries. The overall mortality is almost 50% in Asian countries, with an incidence of 12–14% [5]. Ramucirumab is a recombinant monoclonal antibody that inhibits VEGFR-2. There is an increasing amount of evidence that this antiangiogenic drug is associated with an increased risk of hemorrhage and GI perforation, which may be severe or sometimes fatal. A meta-analysis study published in 2016 including 4963 patients with a variety of solid tumors from 11 different studies demonstrated that overall incidences of all-grade and high-grade hemorrhagic events in cancer patients were 27.6% and 2.3%, respectively [6]. The incidence of all-grade GI perforation with ramucirumab was 1.5%, with 29.8% mortality in another study [7]. Another case report of gastric cancer treated with ramucirumab was complicated by small intestinal metastasis and, ultimately, perforation [8].
Spontaneous rupture is the third leading cause of mortality in the patients with HCC after carcinoma progression and liver failure [9]. In a study, Zhu et al concluded that the presence of liver cirrhosis, hypertension, tumor size >5 cm, and extrahepatic invasion increase the risk of spontaneous rupture [10]. The mechanism of tumor rupture in HCC is poorly understood. The VEGF signaling pathway is broadly involved in HCC angiogenesis and lymphangiogenesis and seems to play a crucial role in disease pathogenesis [11]. The so-called vascular injury hypothesis states that certain changes in the arterial walls supplying the tumor could lead to HCC rupture. These vascular changes may include collagenase expansion leading to degradation of type IV collagen and elastin proliferation [9,12]. We believe that the process mentioned above was accelerated in our patient using the antiangiogenic factor ramucirumab (an inhibitor of VEGF). It is difficult to prove this association; therefore, large-scale studies are needed to explore this association. Also, spontaneous rupture is more common in tumors of size >5 cm. Although possible, it is rare to see a spontaneous rupture in a tumor <5 cm in size. Because of its antiangiogenic activity, there is a concern for serious antiangiogenic adverse effects, including impaired wound healing, GI perforations, and hemorrhage. The gross specimen recovered from surgery in our patient showed necrotic liver tissue with clots, whereas the biopsy specimen showed HCC with extensive necrosis.
Another important finding in our patient was the presence of a microperforation in the proximal duodenum leading to pneumoperitoneum. To our knowledge, this patient had no known risk factors (eg, alcohol abuse, trauma, foreign body ingestion, peptic ulcer disease, primary or metastatic intestinal tumor, nonsteroidal anti-inflammatory drug abuse) for GI perforation. A study done by Wang et al concluded that ramucirumab was associated with GI perforation, with a relative risk of 2.56 [13]. In another study, the incidence rate of all-grade GI perforation with ramucirumab was 1.5%, with a mortality rate of 29.8% [7]. The Food and Drug Administration recommends discontinuing ramucirumab permanently in patients who experience GI perforation. Although unclear, the proposed mechanism for GI perforation is similar to that for HCC rupture (antiangiogenesis) [8].
The best treatment for a ruptured HCC is still debated, the primary goal being the correction of hypovolemic shock. The mortality rate is 85–100% in patients who are managed conservatively [9]. Therapeutic options must be individualized after the initial resuscitation on the basis of tumor staging and resection feasibility. Transarterial embolization effectively controls bleeding from ruptured HCC in the acute phase, with serum bilirubin levels, shock on hospital admission, and prerupture disease state being the critical prognostic factors. Elective liver resection after achieving initial hemostasis via transarterial embolization is preferred over emergency liver resection because, in later cases, the tumor stage and liver function reserve are unclear [14].
Conclusions
Ramucirumab is a new immunotherapy that has been found to be effective in advanced HCC treatment, but there is a concern for serious antiangiogenic adverse effects, including impaired wound healing, GI perforations, and hemorrhage. We presented a rare case of HCC rupture and GI perforation in a patient undergoing treatment with ramucirumab.
Figure 1. Computerized tomography scan of abdomen and pelvis showing the tumor in the right lobe of the liver.
Figure 2. Computerized tomography scan of abdomen and pelvis with contrast material consistent with pneumoperitoneum and cirrhotic liver.
Figure 3. Histopathology showing a nest of tumor cells invading the liver parenchyma (hematoxylin and eosin stain, magnification ×40). | LENVATINIB, PEMBROLIZUMAB, PIPERACILLIN SODIUM\TAZOBACTAM SODIUM, RAMUCIRUMAB | DrugsGivenReaction | CC BY-NC-ND | 33597391 | 19,685,958 | 2021-02-18 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Treatment failure'. | Ramucirumab-Induced Hepatocellular Carcinoma Rupture and Gastrointestinal Perforation.
BACKGROUND Hepatocellular carcinoma (HCC) is a primary liver malignant tumor that typically but not always develops in the setting of chronic liver disease, particularly in patients with cirrhosis or chronic hepatitis B virus infection. Advanced HCC portends a poor prognosis; however, recent advances in first-line and second-line treatment options yield significant survival improvements. Ruptured HCC is an uncommon presentation that occurs in approximately 3-26% of patients. CASE REPORT We present a case of a patient with HCC who was undergoing treatment with the antiangiogenic monoclonal antibody ramucirumab. Subsequently, he presented with signs and symptoms of acute abdomen. The abdominal imaging revealed pneumoperitoneum with multiple abdominal and pelvic collections. The patient underwent exploratory laparotomy and was found to have necrotic liver parenchyma, which appeared to be perforated. Also, a microperforation was noted in the proximal duodenum. The pathology report from liver specimens showed fragments of hepatocellular cancer with extensive necrosis. CONCLUSIONS The mechanism of tumor rupture in HCC is poorly understood. The so-called vascular injury hypothesis states that collagen expansion and elastin proliferation in the arterial wall supplying the tumor could be the leading cause of HCC rupture. We believe that the process mentioned above was accelerated in our patient using the antiangiogenic factor ramucirumab. A similar antiangiogenic mechanism is also implicated in gastrointestinal hemorrhage and perforation related to this drug.
Background
Hepatocellular carcinoma (HCC) is the fourth most common cause of cancer-related mortality worldwide, with over 780,000 deaths in 2018. With a 5-year survival of 18%, liver cancer is the second most lethal tumor after pancreatic cancer [1]. Ramucirumab is a recombinant human immunoglobulin G1 monoclonal antibody that binds to the extracellular binding domain of vascular endothelial growth factor receptor 2 (VEGFR-2) and prevents the binding of VEGFR ligands, thus inhibiting the angiogenesis pathways involved in the development and progression of cancer [2]. It is one of the treatment modalities in treating HCC and is known to be associated with hypertension, proteinuria, hemorrhage, and gastrointestinal (GI) perforation. We present a rare case where ramucirumab was associated with tumor rupture in a known case of HCC and duodenal perforation.
Case Report
We report a case of a 66-year-old man with medical comorbidities of stage 3b HCC, liver cirrhosis, chronic hepatitis C (Child-Turcotte-Pugh B), and hypothyroidism who presented to the emergency department with worsening altered mental status. The patient was diagnosed with HCC 2 years before the presentation. The mass was 3 cm in size and was located in segments V and VI in the right lobe of the liver (Figure 1). He was initially started on treatment with pembrolizumab and Lenvatinib (as part of study treatment). Later on, because of treatment failure, therapy was changed to ramucirumab, an approved second-line treatment in the management of HCC. Four months after starting ramucirumab, he presented to our hospital. His initial vital signs showed blood pressure of 145/90 mmHg, pulse of 92 beats/min, respiratory rate of 17 breaths/min, and oxygen saturation of 98% on room air. Physical examination showed no focal neurological deficits, widespread abdominal tenderness, and absent bowel sounds. Laboratory parameters were as follows: hemoglobin 11.8 g/dL, leukocytes 14×10/μL, alanine aminotransferase 17 U/L, aspartate aminotransferase 54 U/L, total bilirubin 2.0 mg/dL, direct bilirubin 1.0 mg/dL, ammonia 91 μmol/L, and lactate 4.7 mmol/L. The other parameters were within the normal range. His baseline liver function tests (1 month before presentation) were as follows: alanine aminotransferase 15 U/L, aspartate aminotransferase 50 U/L, total bilirubin 1.7 mg/dl, international normalized ratio 1.6, and albumin 2.8 g/L. Computed tomography (CT) of the head and chest X-ray were negative for any significant acute pathology. CT of the abdomen and pelvis with contrast material revealed pneumoperitoneum with multiple abdominal and pelvic collections (Figure 2).
In the setting of leukocytosis, lactic acidosis, peritoneal signs on physical examination, and CT scan with multiple fluid collection areas, the initial impression was suspected gastric or duodenal perforation. The patient was started on broad-spectrum antibiotics and underwent exploratory laparotomy and was found to have a nodular shrunken liver with necrotic parenchyma in segments V and VI of the liver, which appeared to be perforated. Also, a microperforation was noted in the proximal duodenum. Necrotic tissue was removed, the omentum was packed in the liver bed, and 6 L of ascitic fluid was drained. The pathology sample (taken from necrotic liver tissue) showed fragments of hepatocellular carcinoma with extensive necrosis (Figure 3). The peritoneal fluid analysis was consistent with secondary bacterial peritonitis (white blood cell count of 6200 cells/mm3 with 92% neutrophils), and cytology was negative for malignant cells.
Despite antibiotic therapy with piperacillin-tazobactam, the patient developed multiorgan dysfunction. Therefore, we implemented a palliative therapy approach, and the patient was discharged to hospice care.
Discussion
In 85–95% of cases, HCC develops in patients with cirrhotic liver. Studies have shown that once cirrhosis has developed, HCC will occur at a rate of 2–7% per year [3]. The incidence/mortality ratio in HCC is almost 1: 1, which shows that most of the patients with HCC ultimately die [4]. The incidence of spontaneous rupture is <3% in western countries. The overall mortality is almost 50% in Asian countries, with an incidence of 12–14% [5]. Ramucirumab is a recombinant monoclonal antibody that inhibits VEGFR-2. There is an increasing amount of evidence that this antiangiogenic drug is associated with an increased risk of hemorrhage and GI perforation, which may be severe or sometimes fatal. A meta-analysis study published in 2016 including 4963 patients with a variety of solid tumors from 11 different studies demonstrated that overall incidences of all-grade and high-grade hemorrhagic events in cancer patients were 27.6% and 2.3%, respectively [6]. The incidence of all-grade GI perforation with ramucirumab was 1.5%, with 29.8% mortality in another study [7]. Another case report of gastric cancer treated with ramucirumab was complicated by small intestinal metastasis and, ultimately, perforation [8].
Spontaneous rupture is the third leading cause of mortality in the patients with HCC after carcinoma progression and liver failure [9]. In a study, Zhu et al concluded that the presence of liver cirrhosis, hypertension, tumor size >5 cm, and extrahepatic invasion increase the risk of spontaneous rupture [10]. The mechanism of tumor rupture in HCC is poorly understood. The VEGF signaling pathway is broadly involved in HCC angiogenesis and lymphangiogenesis and seems to play a crucial role in disease pathogenesis [11]. The so-called vascular injury hypothesis states that certain changes in the arterial walls supplying the tumor could lead to HCC rupture. These vascular changes may include collagenase expansion leading to degradation of type IV collagen and elastin proliferation [9,12]. We believe that the process mentioned above was accelerated in our patient using the antiangiogenic factor ramucirumab (an inhibitor of VEGF). It is difficult to prove this association; therefore, large-scale studies are needed to explore this association. Also, spontaneous rupture is more common in tumors of size >5 cm. Although possible, it is rare to see a spontaneous rupture in a tumor <5 cm in size. Because of its antiangiogenic activity, there is a concern for serious antiangiogenic adverse effects, including impaired wound healing, GI perforations, and hemorrhage. The gross specimen recovered from surgery in our patient showed necrotic liver tissue with clots, whereas the biopsy specimen showed HCC with extensive necrosis.
Another important finding in our patient was the presence of a microperforation in the proximal duodenum leading to pneumoperitoneum. To our knowledge, this patient had no known risk factors (eg, alcohol abuse, trauma, foreign body ingestion, peptic ulcer disease, primary or metastatic intestinal tumor, nonsteroidal anti-inflammatory drug abuse) for GI perforation. A study done by Wang et al concluded that ramucirumab was associated with GI perforation, with a relative risk of 2.56 [13]. In another study, the incidence rate of all-grade GI perforation with ramucirumab was 1.5%, with a mortality rate of 29.8% [7]. The Food and Drug Administration recommends discontinuing ramucirumab permanently in patients who experience GI perforation. Although unclear, the proposed mechanism for GI perforation is similar to that for HCC rupture (antiangiogenesis) [8].
The best treatment for a ruptured HCC is still debated, the primary goal being the correction of hypovolemic shock. The mortality rate is 85–100% in patients who are managed conservatively [9]. Therapeutic options must be individualized after the initial resuscitation on the basis of tumor staging and resection feasibility. Transarterial embolization effectively controls bleeding from ruptured HCC in the acute phase, with serum bilirubin levels, shock on hospital admission, and prerupture disease state being the critical prognostic factors. Elective liver resection after achieving initial hemostasis via transarterial embolization is preferred over emergency liver resection because, in later cases, the tumor stage and liver function reserve are unclear [14].
Conclusions
Ramucirumab is a new immunotherapy that has been found to be effective in advanced HCC treatment, but there is a concern for serious antiangiogenic adverse effects, including impaired wound healing, GI perforations, and hemorrhage. We presented a rare case of HCC rupture and GI perforation in a patient undergoing treatment with ramucirumab.
Figure 1. Computerized tomography scan of abdomen and pelvis showing the tumor in the right lobe of the liver.
Figure 2. Computerized tomography scan of abdomen and pelvis with contrast material consistent with pneumoperitoneum and cirrhotic liver.
Figure 3. Histopathology showing a nest of tumor cells invading the liver parenchyma (hematoxylin and eosin stain, magnification ×40). | LENVATINIB MESYLATE, PEMBROLIZUMAB | DrugsGivenReaction | CC BY-NC-ND | 33597391 | 19,702,435 | 2021-02-18 |
What was the administration route of drug 'LENVATINIB MESYLATE'? | Ramucirumab-Induced Hepatocellular Carcinoma Rupture and Gastrointestinal Perforation.
BACKGROUND Hepatocellular carcinoma (HCC) is a primary liver malignant tumor that typically but not always develops in the setting of chronic liver disease, particularly in patients with cirrhosis or chronic hepatitis B virus infection. Advanced HCC portends a poor prognosis; however, recent advances in first-line and second-line treatment options yield significant survival improvements. Ruptured HCC is an uncommon presentation that occurs in approximately 3-26% of patients. CASE REPORT We present a case of a patient with HCC who was undergoing treatment with the antiangiogenic monoclonal antibody ramucirumab. Subsequently, he presented with signs and symptoms of acute abdomen. The abdominal imaging revealed pneumoperitoneum with multiple abdominal and pelvic collections. The patient underwent exploratory laparotomy and was found to have necrotic liver parenchyma, which appeared to be perforated. Also, a microperforation was noted in the proximal duodenum. The pathology report from liver specimens showed fragments of hepatocellular cancer with extensive necrosis. CONCLUSIONS The mechanism of tumor rupture in HCC is poorly understood. The so-called vascular injury hypothesis states that collagen expansion and elastin proliferation in the arterial wall supplying the tumor could be the leading cause of HCC rupture. We believe that the process mentioned above was accelerated in our patient using the antiangiogenic factor ramucirumab. A similar antiangiogenic mechanism is also implicated in gastrointestinal hemorrhage and perforation related to this drug.
Background
Hepatocellular carcinoma (HCC) is the fourth most common cause of cancer-related mortality worldwide, with over 780,000 deaths in 2018. With a 5-year survival of 18%, liver cancer is the second most lethal tumor after pancreatic cancer [1]. Ramucirumab is a recombinant human immunoglobulin G1 monoclonal antibody that binds to the extracellular binding domain of vascular endothelial growth factor receptor 2 (VEGFR-2) and prevents the binding of VEGFR ligands, thus inhibiting the angiogenesis pathways involved in the development and progression of cancer [2]. It is one of the treatment modalities in treating HCC and is known to be associated with hypertension, proteinuria, hemorrhage, and gastrointestinal (GI) perforation. We present a rare case where ramucirumab was associated with tumor rupture in a known case of HCC and duodenal perforation.
Case Report
We report a case of a 66-year-old man with medical comorbidities of stage 3b HCC, liver cirrhosis, chronic hepatitis C (Child-Turcotte-Pugh B), and hypothyroidism who presented to the emergency department with worsening altered mental status. The patient was diagnosed with HCC 2 years before the presentation. The mass was 3 cm in size and was located in segments V and VI in the right lobe of the liver (Figure 1). He was initially started on treatment with pembrolizumab and Lenvatinib (as part of study treatment). Later on, because of treatment failure, therapy was changed to ramucirumab, an approved second-line treatment in the management of HCC. Four months after starting ramucirumab, he presented to our hospital. His initial vital signs showed blood pressure of 145/90 mmHg, pulse of 92 beats/min, respiratory rate of 17 breaths/min, and oxygen saturation of 98% on room air. Physical examination showed no focal neurological deficits, widespread abdominal tenderness, and absent bowel sounds. Laboratory parameters were as follows: hemoglobin 11.8 g/dL, leukocytes 14×10/μL, alanine aminotransferase 17 U/L, aspartate aminotransferase 54 U/L, total bilirubin 2.0 mg/dL, direct bilirubin 1.0 mg/dL, ammonia 91 μmol/L, and lactate 4.7 mmol/L. The other parameters were within the normal range. His baseline liver function tests (1 month before presentation) were as follows: alanine aminotransferase 15 U/L, aspartate aminotransferase 50 U/L, total bilirubin 1.7 mg/dl, international normalized ratio 1.6, and albumin 2.8 g/L. Computed tomography (CT) of the head and chest X-ray were negative for any significant acute pathology. CT of the abdomen and pelvis with contrast material revealed pneumoperitoneum with multiple abdominal and pelvic collections (Figure 2).
In the setting of leukocytosis, lactic acidosis, peritoneal signs on physical examination, and CT scan with multiple fluid collection areas, the initial impression was suspected gastric or duodenal perforation. The patient was started on broad-spectrum antibiotics and underwent exploratory laparotomy and was found to have a nodular shrunken liver with necrotic parenchyma in segments V and VI of the liver, which appeared to be perforated. Also, a microperforation was noted in the proximal duodenum. Necrotic tissue was removed, the omentum was packed in the liver bed, and 6 L of ascitic fluid was drained. The pathology sample (taken from necrotic liver tissue) showed fragments of hepatocellular carcinoma with extensive necrosis (Figure 3). The peritoneal fluid analysis was consistent with secondary bacterial peritonitis (white blood cell count of 6200 cells/mm3 with 92% neutrophils), and cytology was negative for malignant cells.
Despite antibiotic therapy with piperacillin-tazobactam, the patient developed multiorgan dysfunction. Therefore, we implemented a palliative therapy approach, and the patient was discharged to hospice care.
Discussion
In 85–95% of cases, HCC develops in patients with cirrhotic liver. Studies have shown that once cirrhosis has developed, HCC will occur at a rate of 2–7% per year [3]. The incidence/mortality ratio in HCC is almost 1: 1, which shows that most of the patients with HCC ultimately die [4]. The incidence of spontaneous rupture is <3% in western countries. The overall mortality is almost 50% in Asian countries, with an incidence of 12–14% [5]. Ramucirumab is a recombinant monoclonal antibody that inhibits VEGFR-2. There is an increasing amount of evidence that this antiangiogenic drug is associated with an increased risk of hemorrhage and GI perforation, which may be severe or sometimes fatal. A meta-analysis study published in 2016 including 4963 patients with a variety of solid tumors from 11 different studies demonstrated that overall incidences of all-grade and high-grade hemorrhagic events in cancer patients were 27.6% and 2.3%, respectively [6]. The incidence of all-grade GI perforation with ramucirumab was 1.5%, with 29.8% mortality in another study [7]. Another case report of gastric cancer treated with ramucirumab was complicated by small intestinal metastasis and, ultimately, perforation [8].
Spontaneous rupture is the third leading cause of mortality in the patients with HCC after carcinoma progression and liver failure [9]. In a study, Zhu et al concluded that the presence of liver cirrhosis, hypertension, tumor size >5 cm, and extrahepatic invasion increase the risk of spontaneous rupture [10]. The mechanism of tumor rupture in HCC is poorly understood. The VEGF signaling pathway is broadly involved in HCC angiogenesis and lymphangiogenesis and seems to play a crucial role in disease pathogenesis [11]. The so-called vascular injury hypothesis states that certain changes in the arterial walls supplying the tumor could lead to HCC rupture. These vascular changes may include collagenase expansion leading to degradation of type IV collagen and elastin proliferation [9,12]. We believe that the process mentioned above was accelerated in our patient using the antiangiogenic factor ramucirumab (an inhibitor of VEGF). It is difficult to prove this association; therefore, large-scale studies are needed to explore this association. Also, spontaneous rupture is more common in tumors of size >5 cm. Although possible, it is rare to see a spontaneous rupture in a tumor <5 cm in size. Because of its antiangiogenic activity, there is a concern for serious antiangiogenic adverse effects, including impaired wound healing, GI perforations, and hemorrhage. The gross specimen recovered from surgery in our patient showed necrotic liver tissue with clots, whereas the biopsy specimen showed HCC with extensive necrosis.
Another important finding in our patient was the presence of a microperforation in the proximal duodenum leading to pneumoperitoneum. To our knowledge, this patient had no known risk factors (eg, alcohol abuse, trauma, foreign body ingestion, peptic ulcer disease, primary or metastatic intestinal tumor, nonsteroidal anti-inflammatory drug abuse) for GI perforation. A study done by Wang et al concluded that ramucirumab was associated with GI perforation, with a relative risk of 2.56 [13]. In another study, the incidence rate of all-grade GI perforation with ramucirumab was 1.5%, with a mortality rate of 29.8% [7]. The Food and Drug Administration recommends discontinuing ramucirumab permanently in patients who experience GI perforation. Although unclear, the proposed mechanism for GI perforation is similar to that for HCC rupture (antiangiogenesis) [8].
The best treatment for a ruptured HCC is still debated, the primary goal being the correction of hypovolemic shock. The mortality rate is 85–100% in patients who are managed conservatively [9]. Therapeutic options must be individualized after the initial resuscitation on the basis of tumor staging and resection feasibility. Transarterial embolization effectively controls bleeding from ruptured HCC in the acute phase, with serum bilirubin levels, shock on hospital admission, and prerupture disease state being the critical prognostic factors. Elective liver resection after achieving initial hemostasis via transarterial embolization is preferred over emergency liver resection because, in later cases, the tumor stage and liver function reserve are unclear [14].
Conclusions
Ramucirumab is a new immunotherapy that has been found to be effective in advanced HCC treatment, but there is a concern for serious antiangiogenic adverse effects, including impaired wound healing, GI perforations, and hemorrhage. We presented a rare case of HCC rupture and GI perforation in a patient undergoing treatment with ramucirumab.
Figure 1. Computerized tomography scan of abdomen and pelvis showing the tumor in the right lobe of the liver.
Figure 2. Computerized tomography scan of abdomen and pelvis with contrast material consistent with pneumoperitoneum and cirrhotic liver.
Figure 3. Histopathology showing a nest of tumor cells invading the liver parenchyma (hematoxylin and eosin stain, magnification ×40). | Oral | DrugAdministrationRoute | CC BY-NC-ND | 33597391 | 19,702,435 | 2021-02-18 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Intentional product use issue'. | A digital single-molecule nanopillar SERS platform for predicting and monitoring immune toxicities in immunotherapy.
The introduction of immune checkpoint inhibitors has demonstrated significant improvements in survival for subsets of cancer patients. However, they carry significant and sometimes life-threatening toxicities. Prompt prediction and monitoring of immune toxicities have the potential to maximise the benefits of immune checkpoint therapy. Herein, we develop a digital nanopillar SERS platform that achieves real-time single cytokine counting and enables dynamic tracking of immune toxicities in cancer patients receiving immune checkpoint inhibitor treatment - broader applications are anticipated in other disease indications. By analysing four prospective cytokine biomarkers that initiate inflammatory responses, the digital nanopillar SERS assay achieves both highly specific and highly sensitive cytokine detection down to attomolar level. Significantly, we report the capability of the assay to longitudinally monitor 10 melanoma patients during immune inhibitor blockade treatment. Here, we show that elevated cytokine concentrations predict for higher risk of developing severe immune toxicities in our pilot cohort of patients.
Introduction
The advent of immune checkpoint therapy has revolutionised the landscape of traditional cancer treatment and is believed to constitute the backbone of managing certain malignancies1–3. By capitalising on the blockade of immune checkpoint inhibitors to take the brakes off parts of the immune system, this emerging therapy has achieved great success producing long-lasting responses (e.g., 10 years or more) in a small but significant fraction of patients3–6. Nevertheless, upon the blockade of immune checkpoint molecules, the activated and potentiated immune reaction predisposes patients to a significant risk of immune-related adverse events (irAEs), which can occur in up to 80% of patients receiving immune checkpoint therapy7–9. The high incidence of irAEs, which may manifest at any time during treatment, can offset the clinical benefits, lead to premature therapy cessation, and even be life-threatening for certain patients10–12. To assist the successful implementation of immune checkpoint therapy, the use of predictive biomarkers for early identification and vigilant monitoring of irAEs is thus critical and a pressing need in avoiding or ameliorating detrimental effects and adjusting therapeutic options.
Cytokines, small signalling proteins, are promising candidates to indicate the occurrence of irAEs due to their prominent role in modulating the anti-cancer immune responses, including enhancing antigen priming, recruiting immune cells into the tumour microenvironment, and upregulating certain immune checkpoint molecules9,13,14. Particularly, excessive cytokine secretion has been implicated in severe inflammation as a major constituent leading to irAEs. For example, the overproduction of fibroblast growth factor 2 (FGF-2)15–18, granulocyte colony-stimulating factor (G-CSF)19, granulocyte-macrophage colony-stimulating factor (GM-CSF)20, and fractalkine (CX3CL1)21 have been found to participate in immune-related inflammatory disease (e.g., rheumatoid arthritis, autoimmune gastritis, and Crohn’s disease). These inflammatory cytokines have recently been reported to indicate irAEs for melanoma patients who underwent immune checkpoint therapy9. The clinical deployment of cytokine analysis for irAE monitoring is challenging and requires a technology that can (i) determine the selected cytokines with great sensitivity9, especially at the onset of irAE development, where the cytokine concentrations are likely to be the lowest; as well as (ii) simultaneously detect a panel of cytokines to reflect the complex interplay of cytokine signalling pathways22 and the variable irAE symptoms among patients.
Conventional cytokine analyses such as immunosorbent assays have limited clinical applicability for irAE assessment due to their limited capacity to detect low cytokine concentrations in blood as well as for assessing a panel of cytokines in a single sample simultaneously. Recently, advances in micro/nanomaterial-based systems have provided a promising suite of technologies that improve the conventional assays by overcoming the above limitations23,24. Encouragingly, the unique advantages of micro/nanomaterial-based systems convey an attractive option for cytokine analysis with the desired results of high sensitivity and multiplexing. The high specific surface area of these miniaturised materials increases mass transfer subsequently enhancing the interaction with target molecules and thus improving the detection sensitivity25. The capabilities of micro/nanomaterial fabrication techniques permit individually separated compartments sufficiently discrete to hold single molecules and hence encompasses a promising strategy for counting assays that can further push the sensitivity of the traditional assays24,26,27. Moreover, the physicochemical properties of nanostructured materials can be exploited to simultaneously label multiple targets (e.g., various spectral signatures) for high-throughput parallel measurements28–30. Therefore, by combining the potential of micro/nanomaterial systems with the need for sensitive irAE monitoring, we have developed a platform for sensitive and multiplex cytokine counting analysis.
Combining the use of (a) discrete single cytokine nanopillar array chip with discretely separated compartments, (b) control of target concentrations to follow a Poisson distribution, and (c) the recognition of target by single-particle active surface-enhanced Raman scattering (SERS) nanotags with a confocal Raman microscope allows accurate and in situ counting of a multi-cytokine panel (FGF-2, G-CSF, GM-CSF, and CX3CL1). Different from the fluorescence-based digital counting strategies24,26, the strikingly narrow spectral peaks of SERS (~1–2 nm) in comparison to fluorescence (~50 nm) makes this platform intrinsically ideal for multiplexed cytokine analysis28.
In this work, we present a digital nanopillar SERS platform that enables the specific cytokine quantification down to attomolar levels and the application in melanoma patients receiving immune checkpoint blockade therapy. Beyond the capability to predict irAEs in melanoma patients receiving therapy, the digital nanopillar SERS assay could potentially be extended to other cytokine-associated immune responses such as excessive immune activation due to viral or bacterial infections (such as COVID-19).
Results
Digital nanopillar SERS platform for parallel profiling of single cytokine
Our concept of digital nanopillar SERS platform for cytokine analysis relies on Rayleigh criterion separation, probability-driven Poisson distribution, single-particle active SERS nanotags, and confocal SERS mapping (Fig. 1). To precisely fabricate the pillar array, we opted to use an electron beam lithographic approach to write the array into a photon-sensitive material followed by physical vapour deposition of gold to create the gold-topped pillars, and selectively reactive ion etching to reveal the pillar structure (Supplementary Fig. 1). The nanopillar array chip consisted of 250,000 individual pillars. As shown in the scanning electron microscope (SEM) image of Fig. 1a, the cubic nanopillars have an edge-to-edge width of 1000 nm and are evenly distributed at 1000 nm intervals to suit the lateral Raman microscope resolution (~1000 nm) that fulfils the Rayleigh criterion separation required to acquire a single SERS spectrum from each pillar without spectral overlap from adjacent pillars.Fig. 1 Digital single-molecule nanopillar surface-enhanced Raman scattering (SERS) platform for parallel counting of four types of cytokines.
SEM images of a pillar array side view, b nanoboxes, and c a single nanobox on the top of a pillar; d SERS spectra of nanoboxes conjugated with 5,5-dithiobis (2-nitrobenzoic acid) (DTNB), 4-mercaptobenzoic acid (MBA), 2,3,5,6-tetrafluoro-4-mercaptobenzoic acid (TFMBA), or 2‐mercapto‐4‐methyl‐5‐thiazoleacetic acid (MMTAA) Raman reporters; e workflow for multiplex counting of cytokines, including fibroblast growth factor 2 (FGF-2), granulocyte colony-stimulating factor (G-CSF), granulocyte-macrophage colony-stimulating factor (GM-CSF), and fractalkine (CX3CL1). Data from one independent experiment.
By using specific gold-thiol chemistry with the linker molecule dithiobis (succinimidyl propionate) (DSP), the gold-topped pillars were selectively functionalised with target recognition antibodies (anti-FGF-2, anti-G-CSF, anti-GM-CSF, and anti-CX3CL1) and acted as the small compartments to capture and confine the individual cytokine. Upon DSP binding on the gold-topped pillars through gold-thiol bond, DSP uses N-hydroxysuccimide (NHS) ester to react with the amine groups of the antibodies31,32. The successful antibody conjugation on gold-topped pillar surfaces was confirmed by matrix-assisted laser desorption ionisation-time of flight mass spectrometry (MALDI-TOF MS) (Supplementary Fig. 2), which showed high molecular weight fragments derived from antibodies. Furthermore, spectroscopic ellipsometry was utilised to estimate the antibody density on pillar surfaces. Based on the obtained film thickness of 18.5 nm, the calculated antibody surface density was 5.5 mg/m2 using the Cuypers model33, which was in agreement with the reported antibody density on substrate surfaces34. Though these characterisations indicated the presence of antibodies on the pillar array, it was not possible to assess the exact distribution of the four structurally related (same immunoglobulin G family) antibodies on a single pillar with an area of 1 µm2. As an advantage of the digital read-out with a large redundancy of pillars, it is not essential to have all four types of antibodies equally distributed on a single pillar for the assay to work. The combined surface area of all antibody-conjugated pillars provides an excess of cytokine binding sites, which maximises successful cytokine capture within the pillar array. Supplementary Fig. 3 shows the SERS mapping images of an equimolar cytokine solution (1031 aM) that provided a similar signal count for the FGF-2, GM-CSF, G-CSF, and CX3CL1 SERS nanotags, indicating a required distribution of four kinds of antibodies conjugated to the array of pillars. Instrumental to the digital counting of cytokines, we controlled the target concentration based on the principle of Poisson distribution where the ratio of cytokine molecules to pillar number was <1:10, ensuring a 99% probability that there was either one cytokine molecule or zero per pillar24. At a ratio of 1:10, 10% of all pillars were occupied or activated with the cytokine molecules.
Following the capture of cytokines on nanopillars, SERS nanotags were applied to recognise the captured cytokines. The preparation of SERS nanotags was performed by the co-conjugating of Raman reporter and target antibody onto gold–silver alloy nanoboxes. Specifically, an average size of 80 nm gold–silver alloy nanoboxes were firstly synthesised using a rapid and aqueous phase approach35 as indicated in the SEM image in Fig. 1b. Supplementary Fig. 4a, b shows the transmission electron microscope (TEM) image of the nanoboxes with the hollow inner structure and a wall thickness of around 15 nm. Nanoparticle tracking analysis (NTA), which allows the tracking and detection of single particles, shows the nanoboxes have a mode size of 77 nm (D10 = 67.6 nm and D90 = 110.6 nm) (Supplementary Fig. 4c). UV-vis extinction spectroscopy demonstrates the nanoboxes possess a surface plasmon resonance (SPR) peak at 610 nm (Supplementary Fig. 4d). The resonance frequency of the nanoboxes enables a more sensitive signal readout with 632.8 nm laser excitation29, which also has a higher Raman scattering efficiency than 785 nm laser. Thereafter, four types of Raman reporters (5,5-dithiobis (2-nitrobenzoic acid) (DTNB), 4-mercaptobenzoic acid (MBA), 2,3,5,6-tetrafluoro-4-mercaptobenzoic acid (TFMBA), and 2‐mercapto‐4‐methyl‐5‐thiazoleacetic acid (MMTAA)) that generate unique Raman signals (1330 cm−1, 1080 cm−1, 1380 cm−1, and 1288 cm−1) were coupled with their corresponding detection antibodies onto nanoboxes as specific SERS nanotags for identification of FGF-2, G-CSF, GM-CSF, and CX3CL1, respectively. As shown in Fig. 1d, the four SERS nanotags provide the strong and non-overlapping Raman signals, which facilitates the multiplexing analysis of four cytokines. The assignment of the major Raman peaks from the four Raman reporters was summarised into Supplementary Table 1. To evaluate the SERS enhancement property of the nanoboxes, we calculated the enhancement factor (EF) of the four Raman reporters on the nanoboxes. Based on the labelled characteristic peaks in Supplementary Fig. 5, the calculated EFs of DTNB, MBA, TFMBA, and MMTAA were 8.14 × 106, 1.46 × 107, 4.01 × 107, and 3.26 × 107, respectively. The obtained EFs were higher than the reported spherical gold nanoparticles and pure silver nanocubes36 and comparable to the reported hollow nanocubes37, illustrating the high SERS property of the nanoboxes. To investigate the SERS nanotag stability, we monitored the Raman signal intensity over 7 days. As shown in Supplementary Fig. 6, Raman signal intensity variations are less than 5% in the SERS spectra, suggesting the good stability of the prepared SERS nanotags.
The following SERS mapping generated false-colour images for counting single cytokine molecules. Under the Raman microscope, the pillar array was visualised as a blue and black grid by representing the specific Raman shifts corresponding to the silicon signals (520 cm−1), in which the blue colour was assigned to silicon signals showing silicon substrates and the black colour indicated the gold-topped pillars because of the lack of silicon signals. The representation of the four colours of the SERS nanotags (red, green, purple, and cyan) on the gold-topped pillars (i.e., black) reflected cytokine molecule occupation (FGF-2, G-CSF, GM-CSF, and CX3CL1). Elevating the sensing area (or gold-topped pillars) from the silicon substrate was selected as a strategy to minimise the false-positive events. By using the confocal function of the Raman microscope, the laser was selectively focused on the gold-topped pillars, thus largely removing the background signals from potentially non-specifically adsorbed SERS nanotags on the silicon substrate. Finally, the specific SERS nanotag signals present or absent on the gold-topped pillars were counted and represented as percentage of active pillars used for total cytokine quantification. For statistical calculations, SERS mapping was applied for scanning 6480 pillars. This digital counting mode, therefore, has the potential to reach the ultimate sensitivity of single molecule cytokine detection.
Demonstration of the single-particle SERS activity of gold–silver alloy nanoboxes
The successful implementation of digital nanopillar SERS assay necessitates the use of single-particle active plasmonic nanostructures that give a clearly detectable signal for each of the single cytokine binding event. The single-particle SERS detection sensitivity is essential to this assay development as the single-particle inactive plasmonic nanoparticles (e.g., spherical gold nanoparticles)38 would unavoidably result in an underestimate of cytokine concentration.
We evaluated the single-particle SERS activity of the prepared anisotropic nanoboxes by acquiring the signals from the individual nanoboxes that were labelled with DTNB reporters. The use of DTNB, a non-resonant Raman reporter, guaranteed the Raman signal enhancement was solely contributed from the nanobox-generated electromagnetic field. As seen in the SEM image (Fig. 2a), two clearly separated DTNB-labelled nanoboxes were deposited on the silicon wafer (highlighted in red circles). The corresponding Raman image (Fig. 2b) displayed several bright Raman spots originating from these individual nanoboxes. The elongated bright Raman spots in the SERS mapping image were probably caused by the slight aggregation of several nanoboxes during sample preparation processes (e.g., centrifugation)38, which was difficult to visually resolve in the SEM image (Fig. 2a). However, unlike the intensity-based assay, the aggregated nanoboxes as SERS nanotags to target cytokine will not skew the digital readout result, because each cytokine will occupy a single pillar following Poisson distribution and both aggregated and individual nanoparticles are regarded as a single binding event that truly reflects the target number39. We then acquired the SERS spectra from two individual SERS nanoboxes (Fig. 2c, (1) and (2)) and two separate spots of bare silicon (Fig. 2c, (3) and (4)). The presence of nanoboxes showed the characteristic Raman signal at 1330 cm−1 from DTNB, whereas the silicon spectra (3) and (4) lacked the specific peak. This observation demonstrated the single-particle SERS activity of nanoboxes, which was largely attributed to the enhanced electromagnetic fields of nanoboxes on specific regions (e.g., tips and corners)40,41 and thereby facilitated the sensitive and accurate counting of cytokines. Based on the acquired SERS mapping image, the median (interquartile range) of the DTNB peak intensity (1330 cm−1) in the presence and absence of nanoboxes were 183.03 a.u. (149.48–243.35 a.u.) and 18.07 a.u. (15.51–23.12 a.u.), respectively. Furthermore, the mean ± standard deviation of the DTNB peak intensity with nanoboxes (213.41 ± 85.03 a.u.) distinguished clearly from the position without nanoboxes (18.79 ± 6.01 a.u.), which demonstrated the feasibility of correctly identifying the presence of nanoboxes.Fig. 2 Demonstration of the single-particle SERS activity of DTNB-labelled nanoboxes.
a SEM image and b corresponding SERS mapping image of DTNB-labelled nanoboxes on a silicon substrate; c representative SERS spectra of numbered locations indicated in a and b. The red dotted line shows the characteristic peak at 1330 cm−1 from DTNB. Data from one independent experiment. Source data are provided in the Source Data file.
Fig. 3 Study of confocal height on Raman signal intensity.
SERS mapping of FGF-2 SERS nanotags on the silicon substrate with changing confocal height of a 0 nm, b 500 nm, c 1000 nm, and d 1500 nm; selected Raman spectra obtained from e red circles and f blue circles of SERS images. Red dotted lines in e and f indicate peak signal at 1330 cm−1 from DTNB. Data from one independent experiment. Source data are provided in the Source Data file.
Fig. 4 Specificity of digital nanopillar SERS platform for FGF-2 cytokine detection.
Representative confocal SERS images in the presence of a target FGF-2 (1031 aM), and negative controls with non-target controls b G-CSF (1031 aM), c GM-CSF (1031 aM), d CX3CL1 (1031 aM), and e PBS. The median (interquartile range) of active pillars per scanning image for FGF-2, G-CSF, GM-CSF, CX3CL1, and PBS was 72 (63.5–76.75), 1.5 (1.5–2), 2 (1–4), 0.5 (0–1.25), and 1 (1–1.75), respectively. Data from one independent experiment.
Optimisation of digital nanopillar SERS platform for cytokine detection
The reliable detection of single cytokine molecules by the digital nanopillar SERS platform depends on the geometric features of the pillar array (i.e., pillar height, cross-section area of pillar) and assay conditions (i.e., incubation time for sample and SERS nanotags).
We first sought to investigate the effect of pillar height on Raman signal intensity to differentiate signals from non-specifically bound SERS nanotags on the silicon substrate and specifically bound SERS nanotags on the gold-topped pillars. FGF-2 SERS nanotags were randomly deposited on the silicon substrate to mimic the non-specific binding scenario and the Raman mappings were perfomed by moving the objective along the z-axis direction with different heights (0 nm, 500 nm, 1000 nm, and 1500 nm) to compare the signal intensity. In Fig. 3, a–d show the false-colour SERS images and e, f the corresponding SERS spectra with characteristic DTNB reporter peak at 1330 cm−1 acquired from the circled spots in a–d. At the height of 0 nm where the SERS nanotags were in focus, we noticed bright Raman spots (Fig. 3a) and strong Raman signals (black line in Fig. 3e, f). With increasing z heights to 500 nm and 1000 nm, the Raman spots decreased (Fig. 3b, c) and the signal intensity weakened/disappeared (red and blue lines in Fig. 3e, f) as the nanoboxes became increasingly out of focus. A further increase to 1500 nm did not remarkably weaken Raman signals compared to the height of 1000 nm (Fig. 3d). Hence, for the fabrication of the pillar array chip, we selected a pillar height of 1000 nm to greatly reduce the potential interference from non-specific signals.
The cross-section area of the pillars provides the space for cytokine binding and labelling with SERS nanotags. To study the effect of pillar cross-section area, we fabricated array chips with pillars of various widths (250 nm, 500 nm, and 1000 nm) (Supplementary Fig. 7a–c) and functionalised with anti-FGF-2 antibody. Each chip consisted of 250,000 individual pillars. We then analysed a sample that contained ~25,000 molecules of FGF-2 (40 µL, 1031 aM), which should result in 10% active pillars (ratio FGF-2: pillars of 0.1). As seen in Supplementary Fig. 7d–f, an increasing pillar cross-section area results in a higher fraction of active pillars. In reference to the expected active pillar percentage (10%), the 250 nm and 500 nm pillar arrays produced lower active pillars (2% and 6%), which suggested a significant loss of target recognition by SERS nanotags. For the 1000 nm wide pillars, the active pillar percentage was 11%, close to the nominal value of 10%. We further tested a sample with 260 aM FGF-2 (i.e., 2.5% active pillars) on the pillar array chips with 250, 500, and 1000 nm pillar widths. The capture efficiency of these three chips was summarised in Supplementary Table 2. In comparison to the pillar array of 250 nm and 500 nm sizes, the 1000 nm provided an improved capture efficiency. As the accessible target recognition surface area per pillar increases, it can possibly promote the thermodynamics and kinetics for higher surface binding and capture efficiency25. Consequently, the 1000 nm pillar array was adopted in the subsequent experiments.
An optimal incubation time of cytokine and SERS nanotags on the pillar array can shorten the operation time and reduce the potential risk of nonspecific binding that could lead to false-positive counting. We thus studied the effect of incubation time of cytokine with SERS nanotags for 30 to 90 min in a solution of 1031 aM FGF-2 and FGF-2 SERS nanotags. As suggested by Supplementary Fig. 8, the increase in incubation time gives rise to a higher proportion of active pillars. In comparison with the theoretical active pillar percentage (10%), both 30 min and 60 min incubation time were able to provide a desirable active pillar percentage (11% and 13%, respectively). A longer incubation time (90 min), however, reported an active pillar percentage (20%) two times higher than the theoretical, indicating the occurrence of nonspecific absorption of SERS nanotags on the pillar array chip. Thus, we selected 30 min incubation time for further digital nanopillar SERS measurements.
Specificity of the digital nanopillar SERS platform for cytokine detection
Accurate and reliable recognition of the specific target is essential for cytokine quantification in clinical samples. To demonstrate the detection specificity of the digital nanopillar SERS assay, we prepared an anti-FGF-2 antibody functionalised pillar array and measured samples containing target FGF-2 cytokine and controls (G-CSF, GM-CSF, CX3CL1, and PBS). It was observed that only the presence of FGF-2 activated significant amounts of pillars whereas the negative controls only generated negligible active pillars (Fig. 4), indicating the high specificity for FGF-2 detection. Similarly, we studied the specific detection of G-CSF, GM-CSF, and CX3CL1, as shown in Supplementary Figs. 9–11, in which the typical Raman images displayed high proportions of active pillars in the presence of specific targets but not for the negative controls.
To further investigate the specificity of binding between SERS nanotag and antibody-functionalised pillar, we performed SEM analysis to “closely” inspect the pillar array for the presence or absence of SERS nanotags. As a representative model, we selected to image FGF-2 SERS nanotags on the anti-FGF-2-functionalised pillar array after sample incubation with FGF-2 cytokine and non-target controls (Fig. 5). As expected, we observed the cubic nanoparticles on the top of pillars in the presence of FGF-2 due to the successful recognition of SERS nanotags. On the contrary, pillar arrays did not display a significant number of FGF-2 SERS nanotags with non-target controls. Consequently, the consistent Raman and SEM data demonstrate the capability of the assay for specific target cytokine counting. The ability to selectively identify these four cytokines in the designed assay is critically important for their usage in clinical samples.Fig. 5 Specificity of the digital nanopillar SERS platform for FGF-2 cytokine detection.
Representative SEM images of pillar array incubated with FGF-2 SERS nanotags in the presence of a, b FGF-2 (1031 aM), c G-CSF (1031 aM), d GM-CSF (1031 aM), e CX3CL1 (1031 aM), and f PBS. The red circles highlight the existence of SERS nanotags. Panel b is the magnified SEM image of the red-highlighted section in a. It is noted that nanofabrication debris on the sidewall of the pillars can also be seen. Data from one independent experiment.
Sensitivity of the digital nanopillar SERS platform for cytokine detection
As there is typically low abundance of cytokines in clinical samples, the technique based on cytokine detection is expected to possess sufficient sensitivity to reliably assess irAEs9. To investigate the sensitivity and dynamic detection range of the digital nanopillar SERS assay, we firstly titrated the designated concentration of one target cytokine (FGF-2) on the pillar array chip with 250,000 pillars. To comply with the Poisson distribution, the upper number of cytokine molecules in the sample is 25,000 which should result in 10% activated pillars. Based on this upper molecule number, we were motivated to challenge the assay by serially diluting the number of cytokine molecule in the sample from 25,000 (1031 aM), 6305 (260 aM), 631 (26 aM), and 63 (2.6 aM). As suggested by the Raman images in Supplementary Fig. 12 and with the decrease in FGF-2 molecules, the percentage of active pillars decreased correspondingly from 9.39% for 1031 aM, 6.59% for 260 aM, 1.12% for 26 aM, and 0.62% for 2.6 aM, showing a strong correlation that facilitates quantitative cytokine analysis.
Subsequently, we were interested in exploring the multiplexing capability of SERS to investigate the digital nanopillar SERS assay’s dynamic range for the simultaneous quantification of all studied cytokines. As the targets independently follow Poisson distribution, each of the cytokine was separately controlled to activate less than 10% pillars. The specific SERS nanotags provided unique signals for each cytokine that was visualised in the false-colour SERS images by a different colour, thereby enabling in situ and simultaneous cytokine detection. As suggested by the confocal SERS images in Fig. 6, an increase in cytokine concentration corresponded with a higher percentage of active pillars. To facilitate quantitative measurements of the cytokines, we calculated the logarithmic transformation of the percentage of active pillars versus cytokine concentration (Supplementary Fig. 13) confirming the strong statistical and potentially clinically relevant correlation (coefficient of determination (R2) >0.97) observed in the SERS images.Fig. 6 Sensitivity for the simultaneous detection of four cytokines.
Representative confocal SERS images of fibroblast growth factor 2 (FGF-2), granulocyte colony-stimulating factor (GM-CSF), granulocyte colony-stimulating factor (G-CSF), and fractalkine (CX3CL1) with the concentration of a 2.6 aM, b 26 aM, c 260 aM, d 1031 aM. Colour scale bars indicate Raman intensities from 5,5-dithiobis (2-nitrobenzoic acid) (DTNB), 4-mercaptobenzoic acid (MBA), 2,3,5,6-tetrafluoro-4-mercaptobenzoic acid (TFMBA), or 2‐mercapto‐4‐methyl‐5‐thiazoleacetic acid (MMTAA). The median (interquartile range) of active pillars per scanning image of FGF-2, G-CSF, GM-CSF, CX3CL1 for 2.6 aM: 3 (1.5–3), 1 (1–2), 2 (1–3), 2 (1–3); 26 aM: 8 (5.5–10), 10 (9–13), 7 (6–10), 8 (6–10); 260 aM: 40 (36–48), 40 (35–52), 39 (35–50), 37 (36–49); and 1031 aM: 79 (61.5–97), 78 (72–87.5), 88 (68.5–97), 79 (64–95), respectively. Data represents one experiment from three independent tests.
To further investigate the multiplexing quantification performance of the digital nanopillar SERS assay in human serum, we spiked standard cytokines in human serum and tested the dynamic range. Supplementary Fig. 14 shows the linear relationship curves for the four targets. Because of the more complicated sample matrix composition in human samples, the lowest detectable cytokine concentration (5.2 aM) was higher than the PBS solution (2.6 aM).
At a cytokine to pillar ratio of 1:10, we studied the probability of each pillar being occupied by different molecule numbers. To experimentally investigate the number of molecules on a single pillar, we analysed a cytokine mixture that contained all four target cytokines at equal concentration (i.e., ~6250 molecules per cytokine). To visualise and count molecule binding events on a single pillar, we labelled the captured cytokines with the four SERS nanotags that provide clearly distinguishable signals. Under Poisson distribution, the likelihood of having two or more molecules on a single pillar is <0.45% (Supplementary Table 3), which underlies the digital counting principle24. Compared to the theoretical Poisson distribution, the experiment data reported a close but slightly higher value, which was probably due to minor non-specific binding of SERS nanotags on the pillars.
The high sensitivity (attomolar level) of the digital nanopillar SERS assay can be ascribed to the following factors: the digital counting strategy, the single-particle SERS activity of the nanoboxes, and the use of pillars to suit confocal Raman mapping that efficiently excludes false-positive signals. Commercially available methods with potential for trace analysis of cytokines include the single-molecule ELISA Simona by Quanterix and electrochemical luminescence assay42,43 by Meso Scale Discovery. Compared to these two methods, the developed digital nanopillar SERS platform enabled in situ multiplexed detection of four cytokines with comparable sensitivity. Unlike the issues of photo bleaching and poor multiplexing analysis often encountered in fluorescence44 and luminescence assays45, SERS provides the advantage of high multiplexing (e.g., 31-plex)46–48 with the narrow Raman linewidth and high photo stability of the Raman reporters. In addition, this digital nanopillar SERS platform can provide more accurate quantification of cytokines by reducing the false-positive signals with the confocal setting, thus eventually help clinicians to monitor irAEs during immune checkpoint therapy. The highly sensitive readout for multiple targets also indicated the capability of this assay for cytokine detection to assess irAEs in clinically relevant samples.
Evaluation of digital nanopillar SERS platform on simulated patient samples
The detection of trace concentrations of cytokines in serum samples is difficult because plasma samples contain a high abundance of non-target molecules (e.g., serum albumin and other proteins) that can potentially interfere with cytokine detection and lead to inaccurate clinical results. To evaluate the capability of the digital nanopillar SERS assay in accurately counting single cytokine molecules, we opted to perform a recovery test in simulated patient plasma samples (i.e., healthy human serum spiked with 1 fM of FGF-2, G-CSF, GM-CSF, and CX3CL1). The rapid scan rate (i.e., 0.05 s for Raman signal integration) facilitated the detection of Raman signals from FGF-2, G-CSF, GM-CSF, and CX3CL1 SERS nanotags rather than the non-target molecules present in human serum due to their low Raman cross-section. As a representative example, Supplementary Fig. 15 shows the Raman signal distribution of the FGF-2 SERS nanotags on five different spots on the pillar array obtained from the recovery test without noticeable Raman signals from other molecules. It is worth noting that unlike the solution-based DTNB labelled SERS nanotag spectra in Fig. 1d, some of the peaks at 1556 cm−1 and 1330 cm−1 in Supplementary Fig. 15 had a similar intensity, which was probably because of the different orientation of the anisotropic nanoboxes on the substrate relative to the polarisation of excitation laser49. Supplementary Tables 4 and 5 show the cytokine concentrations in healthy human serum and human serum spiked with 1 fM cytokine standards determined by digital nanopillar SERS platform, respectively. On five independent pillar arrays, the measured concentrations had the relative standard deviation (RSD) below 9.0% and the Kruskal–Wallis test showed no statistical differences among these results (p » 0.05). Overall, the observed inter-chip variation should enable accurate identification of disease progression to severe irAEs (e.g., grade 3 or 4), but may encounter some challenges in discriminating mild progressing to moderate irAEs (e.g., grade 1 or 2). The assay enabled trace determination of the four targets in simulated human serum as suggested by the target recovery rates of 80.00% to 137.00% with RSD from 16.02% to 21.80% (Supplementary Table 6). Importantly, the ability to measure reliably cytokines at attomolar levels in simulated human serum samples holds promise for detecting early changes in cytokine concentrations as predictors for the emergence of irAEs in immune checkpoint blockade treated patients.
To validate the accuracy of the digital nanopillar SERS assay, we compared the assay with commercially available ELISA kits (one kit for each cytokine tested) for cytokine quantification. To represent a potential clinical scenario of a patient developing irAEs during immune checkpoint blockade therapy, we prepared three samples with increasing concentrations of cytokines (spike in experiments into fetal bovine serum (FBS)) and subsequently analysed these samples with our digital nanopillar SERS assay and the commercial ELISA kits. FBS was used as complex sample matrix devoid of human cytokines. As the limits of detection for the ELISA kits (FGF-2 = 0.95 pM, G-CSF = 1.66 pM, GM-CSF = 1.11 pM, and CX3CL1 = 17.86 pM) were above the attomolar level, the simulated samples were prepared to suit the detection range of these kits. For the digital nanopillar SERS assay, the samples were diluted correspondingly and generated consistent results with the ELISA kits as shown in Supplementary Table 7. No statistical differences were found between ELISA and digital nanopillar SERS results based on Mann–Whitney test. Furthermore, we compared the detection of four cytokines in human serum with digital nanopillar assay and ELISA kits (Supplementary Table 8). The cytokine levels in human serum were below the limit of detection for the conventional ELISA kits, whereas their concentration was quantified by digital nanopillar SERS platform. For the human serum spiked with standard cytokines, the digital SERS platform generated similar results to ELISA without significant differences by Mann–Whitney test. Collectively, the digital nanopillar SERS platform showcased the ability to robustly and accurately quantify cytokines in complicated samples, which is significant for the prospect of dynamic correlation monitoring of irAEs in clinical samples.
Following the demonstration of the accuracy of digital nanopillar SERS platform, we tested the four cytokine levels in ten healthy people (Supplementary Table 9). These ten healthy people showed cytokine concentrations beyond the conventional ELISA capability to accurately quantify, which was consistent with previous reports9,50,51.
Dynamic correlation monitoring of irAEs in melanoma patients receiving immune checkpoint blockade treatment
Having established the feasibility of digital nanopillar SERS in simulated clinical samples, we applied the platform for longitudinally monitoring irAEs in ten melanoma patients (2–3 time points per patient, 26 samples in total) who underwent immune checkpoint blockade therapy (Supplementary Table 10). By diluting the patient samples to follow Poisson distribution, we quantified the cytokine concentration using digital nanopillar SERS platform. Based on the clinical assessments, the patients were classified into two categories: (i) developed severe irAEs (grades 3 and 4) and needed hospitalisation and dedicated treatment (Patients 1, 2, 3, 4, 5); and (ii) developed minor irAEs (grades 1 and 2) that could be managed with immunosuppressants (e.g., corticosteroids) or exhibited no symptom of irAEs (Patients 6, 7, 8, 9, 10).
As a representative case, Fig. 7 shows two cytokine profiles of a patient with severe irAEs (Patient 1) and a patient with mild irAEs (Patient 1). For Patient 1 who received ipilimumab (cytotoxic T-lymphocyte antigen-4 (CTLA-4) inhibitor) and was checked on days 7, 21, and 42, the confocal SERS images showed an increase in active pillars with the continuation of treatment (Fig. 7a–c), suggesting an elevation of the cytokine levels that could potentially trigger the severe irAEs. In agreement with the Raman images, the quantitative counting results for the four cytokines also corroborated the increase of cytokine concentrations peaking in sub-fM levels (Fig. 7d). These cytokine levels were below the limit of detection of conventional ELISA kits (pM level). Importantly, we observed significantly elevated cytokine concentrations in Patient 1 serum on day 42 compared to days 7 and 21. This patient showed the onset of grade 4 irAEs (i.e., colitis) 13 days later (day 55), consistent with the concept that higher cytokine levels correlate with increased risk of developing irAEs9. To further evaluate the utility of these four biomarkers as a signature in identifying and characterising irAEs, we analysed all the counting data from Patient 1 by applying linear discriminant analysis (LDA). As seen in Fig. 7e, the LDA successfully distinguished the data on day 42 into a separate zone from days 7 and 21, which may indicate the potential value of biomarkers in monitoring irAEs development. We further demonstrated Patient 1 LDA with the use of all combinations of two (Supplementary Fig. 16) and three cytokines (Supplementary Fig. 17). Overall, the LDA with four cytokines showed improved classification over using three or less cytokines. Interestingly, considering FGF-2/G-CSF, G-CSF/GM-CSF, and G-CSF/CX3CL1, the LDA generated similar performance to the LDA with four cytokines. To further compare the classification power of FGF-2/G-CSF, G-CSF/GM-CSF, and G-CSF/CX3CL1, and four cytokines, we performed LDA of Patient 2 (Supplementary Fig. 18), which suggested a better differentiation with the use of four cytokines. Therefore, the inclusion of all four cytokines in LDA facilitated a wider and more accurate patient sample analysis. Similarly, Patients 2, 3, 4, and 5, who manifested severe irAEs were connected with higher cytokine levels (Supplementary Fig. 19) and amelioration of irAEs symptom was witnessed with a decrease of cytokine concentrations. For these severe irAEs patients, the LDA model showed a clear discrimination in cytokine profile and this could help to identify patients at risk of irAEs (Supplementary Fig. 19).Fig. 7 Digital nanopillar SERS assay for monitoring melanoma patients during immune checkpoint therapy.
For Patient 1 who developed severe irAEs, SERS images for cytokine detection on a day 7, b day 21, c day 42, d cytokine concentration graph for fibroblast growth factor 2 (FGF-2), granulocyte colony-stimulating factor (G-CSF), granulocyte-macrophage colony-stimulating factor (GM-CSF), and fractalkine (CX3CL1). The two shorter horizontal lines denote the interquartile ranges (25th and 75th percentile) and the longer horizontal lines in between denote the median (50th percentile), and e LDA analysis, respectively. For Patient 6 who developed mild irAEs, SERS images for cytokine detection on f day 0, g day 21, h day 42, i four cytokine concentration graph, the two shorter horizontal lines denote the interquartile ranges (25th and 75th percentile) and the longer horizontal lines in between denote the median (50th percentile), and j LDA analysis, respectively. IPI ipilimumab, PEMBRO pembrolizumab; G3 grade 3, G2 grade 2; SD stable disease, PR partial response. For Patient 1, the median (interquartile range) of active pillars per scanning image of FGF-2, G-CSF, GM-CSF, CX3CL1 on day 7: 14 (11–22.5), 23 (21, 29), 12 (7.5–18), 17 (9–25.5); day 21: 30 (19–37.5), 33 (19–41), 26 (17.5–36.5), 29 (21–43); and day 42: 33 (16.5–58.5), 76 (64–128.5), 25 (14–39.5), 48 (26.5–73.5), respectively. For Patient 6, the median (interquartile range) of active pillars per scanning image of FGF-2, G-CSF, GM-CSF, CX3CL1 on day 0: 18 (16–23), 49 (31.5–56), 23 (17.5–28), 20 (14.5–27); day 21: 29 (24–33.5), 53 (46.5–70), 35 (25–46), 22 (19–29.5); and day 42: 13 (8–16.5), 44 (23.5–55.5), 10 (6.5–12.5), 30 (24–34.5), respectively. The data represented three technical replicates obtained from three chips. Nine images were acquired from each chip for cytokine counting. Statistical analysis was based on Kruskal–Wallis test followed by Dunn’s test to correct multiple comparisons (two-sided). Source data are provided in the Source Data file.
As for Patient 6 who exhibited mild grade 2 irAEs on the skin during combined ipilimumab and pembrolizumab (programmed death-1 (PD-1) inhibitor) or single ipilimumab treatments, the dynamic monitoring displayed relatively stable cytokine levels on different follow-up visits (Fig. 7). Specifically, the confocal Raman images (Fig. 7f–h) and the molecular counting (Fig. 7i) in this patient serum consistently showed no significant cytokine level alterations on the three time points (days 0, 21, 42). Under this circumstance, LDA failed to clearly classify the data into separate sections (Fig. 7j). Likewise, Patients 7, 9, and 10 possessed stable cytokine levels and were diagnosed with low grade irAEs. Meanwhile, Patient 8 who showed decreasing cytokine levels did not display signs of irAEs. LDA was not able to classify Patients 7 and 8 who had mild irAEs and did not show irAEs, but it recognised the minor difference in Patients 9 and 10 who showed grade 1 irAEs (Supplementary Fig. 20).
Overall, we found the preliminary evidence to suggest that significantly elevated cytokine levels have a strong correlation with the development and manifestation of severe irAEs, whereas stabile, low baseline, and decreasing cytokine concentration indicate mild and manageable irAEs. The relatively low concentrations of these four cytokines were below the detection sensitivities of commercially available ELISA kits (pM level), which limits their use in clinical studies. Importantly, the measurement at fM cytokine levels in clinical samples is consistent with the median concentrations of cytokine measured by using digital ELISA51. The successful demonstration of the digital nanopillar SERS platform in dynamic detection of cytokines in patient serum provides a potential approach for the future accurate early detection, characterisation, and monitoring of irAEs in clinical settings. However, it is important to note that the cytokine concentration changes are not directly correlated with the treatment response according to response evaluation criteria in solid tumours (RECIST). In our pilot study, some melanoma patients showed higher levels of cytokines compared to the healthy controls. The power of our digital nanopillar SERS platform lies in the capability to longitudinally monitor cytokines in individual patients over time.
Discussion
Despite the frequent occurrence of irAEs in immune checkpoint therapy, particularly for the combination treatment, the prediction of the emergence of irAEs remains elusive. Mounting data suggest a potential role of cytokines as predictive markers for irAE monitoring in immune checkpoint therapy9,11,13,52. Although promising, accurate quantification of these biomarkers was often not possible due to the dearth in technologies with sufficient detection sensitivity. Typically, either cytokines above the detection limit of immunosorbent assay were selected11, or relative cytokine quantification9 was performed for investigating irAEs. The former approach has the drawback of potentially excluding the low abundance cytokines of significance in irAEs. As for relative quantification9,53, the cytokine concentrations are determined by relating to a standard that had the observed median fluorescence value closest to the median of the test sample. The relative concentration, however, may fail to represent accurate cytokine levels and thus needs further exploration. Our developed digital nanopillar SERS assay offers early data suggestive of a potential approach to the above-mentioned challenges and provides the possibility to study trace amounts of a panel of cytokines in an accurate quantification manner as well as concomitantly providing attomolar level sensitivity. The current proof-of-principle approach has measured four of the potential inflammatory and/or immune toxicity-related cytokines for the prediction of emergence, characterisation, and/or quantifiable correlation with irAEs in melanoma patients.
By leveraging the narrow line width of Raman spectra, the developed digital SERS counting assay shows the ability to sensitively and simultaneously detect multiple cytokines. The adoption of the novel digital quantification mode in SERS using gold–silver alloy nanoboxes further improves the high sensitivity of SERS technology. Notably, the digital counting strategy offers an option for reproducible SERS quantification by avoiding the common Raman signal fluctuations induced by ensemble measurements. The ensemble measurement in SERS relies on the enhancement of Raman signals of molecules located in or near the “hot spots” (i.e., strong electromagnetic fields)30. Due to the random distribution and various efficiencies of “hot spots”, it can result in the discrepancies in acquired SERS intensities for inter-laboratory and even intra-laboratory tests29. To circumvent the impact of Raman intensity fluctuation on accurate quantification, we employed the digital SERS signals from the single-SERS-active nanoboxes on discrete pillar arrays to enumerate the targets and only count the “yes” or “no” signal for a robust and reproducible SERS analysis. Furthermore, the digital readout model, which regards both aggregated and single nanoparticle as a single binding event to reflect the true target number, can have a better accuracy and robustness than the intensity-based assay39.
We believe that the proposed digital nanopillar SERS assay could be used to monitor other cytokine-induced immune responses. For instance, with the outbreak of 2019 novel coronavirus (2019-nCoV), it is yet difficult to predict which infected patient will develop a strong immune response that requires hospitalisation. However, cytokines have been indicated to play a major role in the severity of immune response for critically ill patients infected with 2019-nCoV54. The specific detection of multiple cytokines at early stages of viral infection could thus potentially address this issue and help to provide the clinical care for people at the highest risk. For patients with high cytokine concentrations, the digital nanopillar SERS platform will require the sample dilution to suit Poisson distribution.
In summary, we propose a digital nanopillar SERS platform for the parallel counting of single cytokines and dynamic monitoring in the clinical context of irAE development during immune checkpoint blockade therapy. The platform achieved attomolar level sensitivity by utilising discrete pillar array compartments to hold the single cytokine and subsequently applied single-particle active nanobox-based SERS nanotags for cytokine identification and counting. The confocal Raman mapping on the pillar array offered the highest possible clinical specificity by reducing nonspecific signals and provided a “yes/no” type counting approach for reproducible Raman signal readout. The designed platform was rigorously optimised and tested in simulated clinical samples prior to the evaluation for irAE monitoring in stage IV melanoma patients receiving immune checkpoint blockade therapy. We envisaged this platform possessing the advantages of highly sensitive and multiplexing analysis capability can transit into future irAE detection methodologies after extensive validation in a large cohort of clinical samples over different time courses.
Methods
Materials
Silver nitrate (AgNO3), hydrogen tetrachloroaurate (III) trihydrate (HAuCl4·3H2O), MBA, DTNB, TFMBA, MMTAA, DSP were obtained from Sigma Aldrich. Ascorbic acid (AA) of analytical grade was purchased from MP Biomedicals. FGF-2 (223-FB), G-CSF (214-CS), GM-CSF (215-GM), CX3CL1 (365-FR) cytokines; monoclonal anti-FGF-2 (MAB233), anti-G-CSF (MAB214), anti-GM-CSF (MAB615), anti-CX3CL1 (MAB3652) antibodies; polyclonal anti-FGF-2 (AF-233), anti-G-CSF (AF-214), anti-GM-CSF (AF-215), anti-CX3CL1 (AF-365) antibodies; and FGF-2 (DY233-05), G-CSF (DY214-05), GM-CSF (DY215-05), and CX3CL1 (DY365) ELISA kits were bought from R&D Systems.
All the patient serum or plasma samples were collected at the Austin Hospital (Melbourne) under approved human ethic protocols and written informed consents were obtained from all patients before sample collection. Ethics approval was obtained from The University of Queensland Institutional Human Research Ethics Committee (approval nos. 2011001315 and 2016000876) and the following clinical assay was carried out according to the approved guidelines.
Preparation of single-particle active SERS nanotags
The preparation of SERS nanotags involved the synthesis of nanoboxes and the subsequent functionalisation with Raman reporters and antibodies. For nanobox synthesis, 45 µL of HAuCl4 (1 wt%) was added into 10 mL of ultrapure H2O (18.2 Ω cm) under magnetic stirring (800 r.p.m.) for 1 min, followed by simultaneously introducing 170 µL of AgNO3 (6 mM) and 30 µL of AA (0.1 M) into the stirring solution. Then, the formation of nanoboxes was indicated by the appearance of an apparent blue colour within 6 s and the samples were collected 1 min later by centrifuging at 600 g for 15 min.
To functionalise nanoboxes with Raman reporters and antibodies, 300 µL of nanoboxes centrifuged from 1 mL of as-prepared solution were co-incubated with one type of Raman reporters (i.e., 10 µL of DTNB, 8 µL of MBA, 10 µL of TFMBA, or 10 µL of MMTAA) and 2 µL of DSP linker for 6 h. After that, the Raman reporter and DSP functionalised nanoboxes were separated by centrifuging at 600 g for 15 min, and resuspended into 300 µL of PBS (0.1 mM). Then, 2 µg of anti-FGF-2, anti-G-CSF, anti-GM-CSF, anti-CX3CL1 antibodies were added to MBA, DTNB, TFMBA, and MMTAA labelled nanoboxes, respectively. After overnight incubation at 4 oC, the functionalised nanoboxes were purified by centrifuging at 600×g for 15 min to separate free antibodies and the final products were resuspended into 200 µL of 0.1% BSA for future use.
Fabrication of pillar arrays
The chip was made of a sensing array measuring 1 mm × 1 mm (Supplementary Fig. 1a) and consisted of 250,000 individual pillars. Each pillar was 1 µm wide, 1 µm long, and 1 µm high. The pillars were evenly spaced by 1 µm from one pillar to the next (Supplementary Fig. 1b). The pillar array was designed using Nanosuite 6.0 (Raith GmbH) and Beamer 5.9.1 (GenISys GmbH) and fabricated on a 4-inch p-type <100> silicon wafer (Bonda Technology Pte Ltd, Singapore) using electron beam lithography (EBL). The wafer accommodated 76 separate pillar arrays. Silicon wafer was first cleaned in acetone, isopropanol with sonication for 2 min each, followed by rising with deionised H2O and dehydration bake at 180 °C for 2 min. Prior to the resist coating, the wafer had undergone a further O2 plasma cleaning at 200 W for 5 min (Diener Atto, Diener Electronic GmbH). The cleaned wafer was spin-coated with two layers of polymethyl methacrylate (bottom: 495K A4 PMMA, top: 950k A4 PMMA, from MicroChemicals GmbH) using the CEE Apogee Coater (Cost Effective Equipment, LLC) at 1500 r.p.m. for 60 s each. After the coating of each layer, the wafer was baked immediately on a hot plate to prevent from intermixing of the two layers of resist. The baking time was 10 and 3 min for the bottom and top layer, respectively, at 180 °C. The thickness of the photoresist was found to be ~450 nm (top PMMA: ~250 nm, bottom PMMA: ~200 nm), characterised by white light reflectometry (FilmTek 2000M, Scientific Computing International). EBL was performed in the Raith EBPG5150 system. The patterns were exposed in EBL with an accelerated voltage of 100 kV, 150 nA of beam current (spot size ~80 nm), with step sizes of 40 nm and an electron dose of 1200 µC/cm2. The exposure time per 4-inch wafer was ~35 min, each containing 76 individual chips. After exposure, the wafer was developed in a mixture of isopropanol and methyl isobutyl ketone (3:1) for 60 s and rinsed immediately with isopropanol, followed by drying with N2. An oxygen plasma descum process, at 100 W, 60 s (Diener Atto, Diener Electronic GmbH) was carried out to remove resist residues prior to the deposition. Next, 10 nm titanium and 200 nm gold were deposited by physical vapour deposition using a Temescal FC-2000 electron beam evaporator (Ferrotec, U.S.A.). After overnight lift-off at room temperature in Remover PG (MicroChemicals GmbH, Germany), the excess material was washed off and the pillar array structure was revealed (Supplementary Fig. 1c). To create the pillar height (i.e., 1 µm), reactive ion etching (Oxford Instruments, UK) was applied for anisotropic etching of the silicon. Hereby, the deposited gold served as mask to protect the underlying silicon while the un-masked silicon was removed. Next, the wafer was coated with a protective layer of cured AZnLOF 2020 prior to wafer dicing into 76 individual sensing chips consisting of a single pillar array. Prior to use, the protective layer was washed off by consecutive washes with isopropanol and acetone and dried under a stream of nitrogen.
Pillar array functionalization
Antibody functionalisation of the gold-topped pillar array was conducted by crosslinking the antibodies to the gold surface using DSP. A solution of 5 mM DSP in dimethyl sulfoxide was pipetted onto the pillar array and incubated at room temperature for 2 h. After rinsing the pillar array with ethanol and PBS, a solution of 5 µg/mL anti-cytokine monoclonal antibody solution (100 fold dilution of antibody stock solution) in PBS was incubated overnight at 4 °C. Subsequently, the pillar array was rinsed with PBS and blocked using 1% bovine serum albumin in PBS for 1 h. Prior to use, the pillar array was rinsed with PBS. All PBS solutions were filtered through a sterile 0.22 µm syringe filter (Millex-GP, Merck, U.S.A.).
Digital nanopillar SERS profiling of cytokines
Cytokines (FGF-2, G-CSF, GM-CSF, and CX3CL1) with different concentrations in PBS (2.6 aM, 26 aM, 260 aM, and 1031 aM) were incubated with antibody functionalised pillar arrays at room temperature for 30 min, followed by washing the pillar array three times with washing buffer (0.1% BSA and 0.01% Tween 20 in PBS). SERS nanotags were then added into the pillar array for another 30 min incubation under room temperature to identify the targets. Finally, the pillar arrays were washed to remove the free SERS nanotags and were subject to confocal Raman microscope for quantification. For each sample, nine SERS images with each image has the dimension of 60 µm × 48 µm were taken on the pillar array to calculate the overall cytokine concentration.
Simulated clinical sample detection
For the recovery experiment, standard cytokines (FGF-2, G-CSF, GM-CSF, and CX3CL1) with the concentration of 1 fM were added into healthy human serum and then diluted ten times with PBS to quantify.
For the quantification of cytokines in FBS, three simulated clinical samples were prepared by titrating various concentrations of standard cytokines into 10% FBS: Sample 1 (FGF-2 = 3.64 pM, G-CSF = 3.19 pM, GM-CSF = 4.29 pM, and CX3CL1 = 28.57 fM); Sample 2 (FGF-2 = 7.28 pM, G-CSF = 6.38 pM, GM-CSF = 8.58 pM, and CX3CL1 = 571.4 fM); and Sample 3 (FGF-2 = 14.56 pM, G-CSF = 12.76 pM, GM-CSF = 17.16 pM, and CX3CL1 = 1142.8 fM). These three samples were then detected directly using the commercial ELISA kits or digital nanopillar SERS assay with a further dilution of 105, 2 × 105, and 4 × 105, respectively.
Spectroscopic ellipsometry
The antibody film thickness was measured by in-solution spectroscopic ellipsometry (M2000V JA Woollam Co., Inc. USA) using gold-coated substrates and flow cell (QSense® Ellipsometry, Biolin Scientific, Sweden). Measurements were performed at an angle of 65°. Data analysis was performed by CompleteEASE® software using a B-Spline data fit and Cauchy model to calculate the antibody film thickness.
MALDI-TOF MS
The antibody-functionalised nanopillar array chip was subjected to tryptic digest prior to analysis. Sequencing-grade trypsin was made to 50 ng/µL in 25 mM ammonium bicarbonate, and sprayed over the chip using a Bruker Imageprep instrument (Bruker, USA). After trypsin deposition, the chip was incubated in a humid environment at 40 °C for 3 h. Subsequently, the chip was sprayed with a matrix solution, 10 g/L α-cyano-4-hydroxycinnamic acid in 50% acetonitrile with 0.2% trifluoroacetic acid. Next, the chip was analysed with a Bruker Ultraflex III MALDI-TOF mass spectrometer (Bruker, USA) in positive linear mode using Flex Imaging 4.0 (Bruker, USA) with a pixel size of 60 µm. Data were collected from 2 k–30 k m/z, at a laser repetition rate of 200 Hz. Data were normalised using the root mean square approach and visualised using Flex Imaging 4.0 (Bruker, USA) and SCILS LAB 2017a software. For the SCILS LAB analysis, the data were imported using a convolution baseline subtraction, and displayed using root mean squared normalisation.
Instrumentations
SEM images of pillar arrays and nanoboxes were taken on a JEOL-7100 field emission (FE)-SEM (20 kV voltage). TEM images of nanoboxes were taken on a JEOL-2100 microscope (200 kV voltage). NTA of nanobox size distribution was performed with Malvern NanoSight NS300. UV-vis extinction spectrum of nanoboxes was performed with a Shimadzu UV-2450 spectrophotometer. Confocal Raman mapping was conducted on a WITec alpha 300 R spectrometer using 632.8 nm He–Ne laser with the power of 35 mW, a grating of 600 g/mm used with EMCCD camera, spectral resolution of 1.390 cm−1 to 2.114 cm−1, confocal pinhole size of 100 µm, 100× air objective with NA of 0.90, and 0.05 s integration time. The theoretical spot size was 857.80 nm based on the Abbe diffraction limit (i.e., d = 1.22λ/NA). The scanning area was set to have 60 µm × 48 µm with 86 points per line and 69 lines per image. For each pillar array, nine separate scanning areas were taken in total and the total active pillars were used for quantification. The SERS mapping images for counting were taken by focusing the laser on the top of the pillar surfaces. Specifically, the laser was firstly focused on the silicon substrates by obtaining the strongest silicon signals (520 cm−1) and then the 100× objective was moved up in z-axis direction of 1 µm for SERS scanning. The system was calibrated with the first-order photo peak of silicon at 520 cm−1.
Data analysis
To assign the SERS nanotag membership for DTNB, MBA, TFMBA, and MMTAA, Project Five 5.0 software from WITec was utilised to create four filters, which summed a spectral range of 40 cm−1 with the centre position at the characteristic Raman peak of each reporter and subtracted the background with a polynomial algorithm. Specifically, the filter ranges of four Raman reporters DTNB-, MBA-, TFMBA-, and MMTAA-coated SERS nanotags were (1310–1350 cm−1), (1060–1100 cm−1), (1360–1400 cm−1), and (1268–1308 cm−1), respectively. All the SERS images were analysed by using threshold intensity to determine the successful binding events. Specifically, the threshold intensity of FGF-2, G-CSF, GM-CSF, and CX3CL1 was set at 5000, 4000, 5000, and 5000, respectively. For each image, the threshold intensity was doubled-checked and adjusted based on the true Raman peaks in the spectra. Statistical analysis assuming unequal variances was conducted with Kruskal–Wallis test among three groups or Mann–Whitney test between two groups with GraphPad Prism 8.4. To control the error appropriately, we performed multiple comparisons using Dunn’s test. LDA of clinical samples was performed in R software (3.6.2) with the MASS package (7.3-52). The active pillars in SERS images were counted with Image J software.
Supplementary information
Supplementary Information
Peer Review File
Source data
Source Data
Peer review information
Nature Communications thanks Chongwen Wang and Isaac Pence for their contribution to the peer review of this work. Peer reviewer reports are available.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
The online version contains supplementary material available at 10.1038/s41467-021-21431-w.
Acknowledgements
The authors acknowledge grants received by our laboratory from the National Breast Cancer Foundation of Australia (CG-12-07) and the ARC DP (160102836 and 210103151). These grants have significantly contributed to the environment to stimulate the research described here. J.L. acknowledges support from the Australian Government Research Training Program Scholarship. A.W. and A.A.I.S. thank the National Health and Medical Research Council for funding (APP1173669 and APP1175047). A.B. is the recipient of a Fellowship from the Victorian Government Department of Health and Human Services acting through the Victorian Cancer Agency. We acknowledge the facilities, and the scientific and technical assistance, of the Australian Microscopy & Microanalysis Research Facility at the Centre for Microscopy and Microanalysis, The University of Queensland. We appreciate to receive the technical and scientific guidance from Queensland node of the Australian National Fabrication Facility (Q-ANFF) in confocal Raman mapping and spectroscopic ellipsometry measurement.
Author contributions
J.L., A.W., A.I.S., H-H.C., Y.W., A.B., P.M., and M.T. contributed to the design of the experiments and analysing the data. J.L. and A.W. performed the experiments and prepared the manuscript. All authors read, commented, and edited the manuscript, and assisted during the revision process.
Data availability
Data supporting the findings of this work are available within this paper and the supporting information files. A reporting summary of this work is available as a Supplementary file. Source data are provided with this paper.
Competing interests
The authors declare no competing interests. | IPILIMUMAB, PEMBROLIZUMAB | DrugsGivenReaction | CC BY | 33597530 | 19,690,448 | 2021-02-17 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Pancreatic toxicity'. | A digital single-molecule nanopillar SERS platform for predicting and monitoring immune toxicities in immunotherapy.
The introduction of immune checkpoint inhibitors has demonstrated significant improvements in survival for subsets of cancer patients. However, they carry significant and sometimes life-threatening toxicities. Prompt prediction and monitoring of immune toxicities have the potential to maximise the benefits of immune checkpoint therapy. Herein, we develop a digital nanopillar SERS platform that achieves real-time single cytokine counting and enables dynamic tracking of immune toxicities in cancer patients receiving immune checkpoint inhibitor treatment - broader applications are anticipated in other disease indications. By analysing four prospective cytokine biomarkers that initiate inflammatory responses, the digital nanopillar SERS assay achieves both highly specific and highly sensitive cytokine detection down to attomolar level. Significantly, we report the capability of the assay to longitudinally monitor 10 melanoma patients during immune inhibitor blockade treatment. Here, we show that elevated cytokine concentrations predict for higher risk of developing severe immune toxicities in our pilot cohort of patients.
Introduction
The advent of immune checkpoint therapy has revolutionised the landscape of traditional cancer treatment and is believed to constitute the backbone of managing certain malignancies1–3. By capitalising on the blockade of immune checkpoint inhibitors to take the brakes off parts of the immune system, this emerging therapy has achieved great success producing long-lasting responses (e.g., 10 years or more) in a small but significant fraction of patients3–6. Nevertheless, upon the blockade of immune checkpoint molecules, the activated and potentiated immune reaction predisposes patients to a significant risk of immune-related adverse events (irAEs), which can occur in up to 80% of patients receiving immune checkpoint therapy7–9. The high incidence of irAEs, which may manifest at any time during treatment, can offset the clinical benefits, lead to premature therapy cessation, and even be life-threatening for certain patients10–12. To assist the successful implementation of immune checkpoint therapy, the use of predictive biomarkers for early identification and vigilant monitoring of irAEs is thus critical and a pressing need in avoiding or ameliorating detrimental effects and adjusting therapeutic options.
Cytokines, small signalling proteins, are promising candidates to indicate the occurrence of irAEs due to their prominent role in modulating the anti-cancer immune responses, including enhancing antigen priming, recruiting immune cells into the tumour microenvironment, and upregulating certain immune checkpoint molecules9,13,14. Particularly, excessive cytokine secretion has been implicated in severe inflammation as a major constituent leading to irAEs. For example, the overproduction of fibroblast growth factor 2 (FGF-2)15–18, granulocyte colony-stimulating factor (G-CSF)19, granulocyte-macrophage colony-stimulating factor (GM-CSF)20, and fractalkine (CX3CL1)21 have been found to participate in immune-related inflammatory disease (e.g., rheumatoid arthritis, autoimmune gastritis, and Crohn’s disease). These inflammatory cytokines have recently been reported to indicate irAEs for melanoma patients who underwent immune checkpoint therapy9. The clinical deployment of cytokine analysis for irAE monitoring is challenging and requires a technology that can (i) determine the selected cytokines with great sensitivity9, especially at the onset of irAE development, where the cytokine concentrations are likely to be the lowest; as well as (ii) simultaneously detect a panel of cytokines to reflect the complex interplay of cytokine signalling pathways22 and the variable irAE symptoms among patients.
Conventional cytokine analyses such as immunosorbent assays have limited clinical applicability for irAE assessment due to their limited capacity to detect low cytokine concentrations in blood as well as for assessing a panel of cytokines in a single sample simultaneously. Recently, advances in micro/nanomaterial-based systems have provided a promising suite of technologies that improve the conventional assays by overcoming the above limitations23,24. Encouragingly, the unique advantages of micro/nanomaterial-based systems convey an attractive option for cytokine analysis with the desired results of high sensitivity and multiplexing. The high specific surface area of these miniaturised materials increases mass transfer subsequently enhancing the interaction with target molecules and thus improving the detection sensitivity25. The capabilities of micro/nanomaterial fabrication techniques permit individually separated compartments sufficiently discrete to hold single molecules and hence encompasses a promising strategy for counting assays that can further push the sensitivity of the traditional assays24,26,27. Moreover, the physicochemical properties of nanostructured materials can be exploited to simultaneously label multiple targets (e.g., various spectral signatures) for high-throughput parallel measurements28–30. Therefore, by combining the potential of micro/nanomaterial systems with the need for sensitive irAE monitoring, we have developed a platform for sensitive and multiplex cytokine counting analysis.
Combining the use of (a) discrete single cytokine nanopillar array chip with discretely separated compartments, (b) control of target concentrations to follow a Poisson distribution, and (c) the recognition of target by single-particle active surface-enhanced Raman scattering (SERS) nanotags with a confocal Raman microscope allows accurate and in situ counting of a multi-cytokine panel (FGF-2, G-CSF, GM-CSF, and CX3CL1). Different from the fluorescence-based digital counting strategies24,26, the strikingly narrow spectral peaks of SERS (~1–2 nm) in comparison to fluorescence (~50 nm) makes this platform intrinsically ideal for multiplexed cytokine analysis28.
In this work, we present a digital nanopillar SERS platform that enables the specific cytokine quantification down to attomolar levels and the application in melanoma patients receiving immune checkpoint blockade therapy. Beyond the capability to predict irAEs in melanoma patients receiving therapy, the digital nanopillar SERS assay could potentially be extended to other cytokine-associated immune responses such as excessive immune activation due to viral or bacterial infections (such as COVID-19).
Results
Digital nanopillar SERS platform for parallel profiling of single cytokine
Our concept of digital nanopillar SERS platform for cytokine analysis relies on Rayleigh criterion separation, probability-driven Poisson distribution, single-particle active SERS nanotags, and confocal SERS mapping (Fig. 1). To precisely fabricate the pillar array, we opted to use an electron beam lithographic approach to write the array into a photon-sensitive material followed by physical vapour deposition of gold to create the gold-topped pillars, and selectively reactive ion etching to reveal the pillar structure (Supplementary Fig. 1). The nanopillar array chip consisted of 250,000 individual pillars. As shown in the scanning electron microscope (SEM) image of Fig. 1a, the cubic nanopillars have an edge-to-edge width of 1000 nm and are evenly distributed at 1000 nm intervals to suit the lateral Raman microscope resolution (~1000 nm) that fulfils the Rayleigh criterion separation required to acquire a single SERS spectrum from each pillar without spectral overlap from adjacent pillars.Fig. 1 Digital single-molecule nanopillar surface-enhanced Raman scattering (SERS) platform for parallel counting of four types of cytokines.
SEM images of a pillar array side view, b nanoboxes, and c a single nanobox on the top of a pillar; d SERS spectra of nanoboxes conjugated with 5,5-dithiobis (2-nitrobenzoic acid) (DTNB), 4-mercaptobenzoic acid (MBA), 2,3,5,6-tetrafluoro-4-mercaptobenzoic acid (TFMBA), or 2‐mercapto‐4‐methyl‐5‐thiazoleacetic acid (MMTAA) Raman reporters; e workflow for multiplex counting of cytokines, including fibroblast growth factor 2 (FGF-2), granulocyte colony-stimulating factor (G-CSF), granulocyte-macrophage colony-stimulating factor (GM-CSF), and fractalkine (CX3CL1). Data from one independent experiment.
By using specific gold-thiol chemistry with the linker molecule dithiobis (succinimidyl propionate) (DSP), the gold-topped pillars were selectively functionalised with target recognition antibodies (anti-FGF-2, anti-G-CSF, anti-GM-CSF, and anti-CX3CL1) and acted as the small compartments to capture and confine the individual cytokine. Upon DSP binding on the gold-topped pillars through gold-thiol bond, DSP uses N-hydroxysuccimide (NHS) ester to react with the amine groups of the antibodies31,32. The successful antibody conjugation on gold-topped pillar surfaces was confirmed by matrix-assisted laser desorption ionisation-time of flight mass spectrometry (MALDI-TOF MS) (Supplementary Fig. 2), which showed high molecular weight fragments derived from antibodies. Furthermore, spectroscopic ellipsometry was utilised to estimate the antibody density on pillar surfaces. Based on the obtained film thickness of 18.5 nm, the calculated antibody surface density was 5.5 mg/m2 using the Cuypers model33, which was in agreement with the reported antibody density on substrate surfaces34. Though these characterisations indicated the presence of antibodies on the pillar array, it was not possible to assess the exact distribution of the four structurally related (same immunoglobulin G family) antibodies on a single pillar with an area of 1 µm2. As an advantage of the digital read-out with a large redundancy of pillars, it is not essential to have all four types of antibodies equally distributed on a single pillar for the assay to work. The combined surface area of all antibody-conjugated pillars provides an excess of cytokine binding sites, which maximises successful cytokine capture within the pillar array. Supplementary Fig. 3 shows the SERS mapping images of an equimolar cytokine solution (1031 aM) that provided a similar signal count for the FGF-2, GM-CSF, G-CSF, and CX3CL1 SERS nanotags, indicating a required distribution of four kinds of antibodies conjugated to the array of pillars. Instrumental to the digital counting of cytokines, we controlled the target concentration based on the principle of Poisson distribution where the ratio of cytokine molecules to pillar number was <1:10, ensuring a 99% probability that there was either one cytokine molecule or zero per pillar24. At a ratio of 1:10, 10% of all pillars were occupied or activated with the cytokine molecules.
Following the capture of cytokines on nanopillars, SERS nanotags were applied to recognise the captured cytokines. The preparation of SERS nanotags was performed by the co-conjugating of Raman reporter and target antibody onto gold–silver alloy nanoboxes. Specifically, an average size of 80 nm gold–silver alloy nanoboxes were firstly synthesised using a rapid and aqueous phase approach35 as indicated in the SEM image in Fig. 1b. Supplementary Fig. 4a, b shows the transmission electron microscope (TEM) image of the nanoboxes with the hollow inner structure and a wall thickness of around 15 nm. Nanoparticle tracking analysis (NTA), which allows the tracking and detection of single particles, shows the nanoboxes have a mode size of 77 nm (D10 = 67.6 nm and D90 = 110.6 nm) (Supplementary Fig. 4c). UV-vis extinction spectroscopy demonstrates the nanoboxes possess a surface plasmon resonance (SPR) peak at 610 nm (Supplementary Fig. 4d). The resonance frequency of the nanoboxes enables a more sensitive signal readout with 632.8 nm laser excitation29, which also has a higher Raman scattering efficiency than 785 nm laser. Thereafter, four types of Raman reporters (5,5-dithiobis (2-nitrobenzoic acid) (DTNB), 4-mercaptobenzoic acid (MBA), 2,3,5,6-tetrafluoro-4-mercaptobenzoic acid (TFMBA), and 2‐mercapto‐4‐methyl‐5‐thiazoleacetic acid (MMTAA)) that generate unique Raman signals (1330 cm−1, 1080 cm−1, 1380 cm−1, and 1288 cm−1) were coupled with their corresponding detection antibodies onto nanoboxes as specific SERS nanotags for identification of FGF-2, G-CSF, GM-CSF, and CX3CL1, respectively. As shown in Fig. 1d, the four SERS nanotags provide the strong and non-overlapping Raman signals, which facilitates the multiplexing analysis of four cytokines. The assignment of the major Raman peaks from the four Raman reporters was summarised into Supplementary Table 1. To evaluate the SERS enhancement property of the nanoboxes, we calculated the enhancement factor (EF) of the four Raman reporters on the nanoboxes. Based on the labelled characteristic peaks in Supplementary Fig. 5, the calculated EFs of DTNB, MBA, TFMBA, and MMTAA were 8.14 × 106, 1.46 × 107, 4.01 × 107, and 3.26 × 107, respectively. The obtained EFs were higher than the reported spherical gold nanoparticles and pure silver nanocubes36 and comparable to the reported hollow nanocubes37, illustrating the high SERS property of the nanoboxes. To investigate the SERS nanotag stability, we monitored the Raman signal intensity over 7 days. As shown in Supplementary Fig. 6, Raman signal intensity variations are less than 5% in the SERS spectra, suggesting the good stability of the prepared SERS nanotags.
The following SERS mapping generated false-colour images for counting single cytokine molecules. Under the Raman microscope, the pillar array was visualised as a blue and black grid by representing the specific Raman shifts corresponding to the silicon signals (520 cm−1), in which the blue colour was assigned to silicon signals showing silicon substrates and the black colour indicated the gold-topped pillars because of the lack of silicon signals. The representation of the four colours of the SERS nanotags (red, green, purple, and cyan) on the gold-topped pillars (i.e., black) reflected cytokine molecule occupation (FGF-2, G-CSF, GM-CSF, and CX3CL1). Elevating the sensing area (or gold-topped pillars) from the silicon substrate was selected as a strategy to minimise the false-positive events. By using the confocal function of the Raman microscope, the laser was selectively focused on the gold-topped pillars, thus largely removing the background signals from potentially non-specifically adsorbed SERS nanotags on the silicon substrate. Finally, the specific SERS nanotag signals present or absent on the gold-topped pillars were counted and represented as percentage of active pillars used for total cytokine quantification. For statistical calculations, SERS mapping was applied for scanning 6480 pillars. This digital counting mode, therefore, has the potential to reach the ultimate sensitivity of single molecule cytokine detection.
Demonstration of the single-particle SERS activity of gold–silver alloy nanoboxes
The successful implementation of digital nanopillar SERS assay necessitates the use of single-particle active plasmonic nanostructures that give a clearly detectable signal for each of the single cytokine binding event. The single-particle SERS detection sensitivity is essential to this assay development as the single-particle inactive plasmonic nanoparticles (e.g., spherical gold nanoparticles)38 would unavoidably result in an underestimate of cytokine concentration.
We evaluated the single-particle SERS activity of the prepared anisotropic nanoboxes by acquiring the signals from the individual nanoboxes that were labelled with DTNB reporters. The use of DTNB, a non-resonant Raman reporter, guaranteed the Raman signal enhancement was solely contributed from the nanobox-generated electromagnetic field. As seen in the SEM image (Fig. 2a), two clearly separated DTNB-labelled nanoboxes were deposited on the silicon wafer (highlighted in red circles). The corresponding Raman image (Fig. 2b) displayed several bright Raman spots originating from these individual nanoboxes. The elongated bright Raman spots in the SERS mapping image were probably caused by the slight aggregation of several nanoboxes during sample preparation processes (e.g., centrifugation)38, which was difficult to visually resolve in the SEM image (Fig. 2a). However, unlike the intensity-based assay, the aggregated nanoboxes as SERS nanotags to target cytokine will not skew the digital readout result, because each cytokine will occupy a single pillar following Poisson distribution and both aggregated and individual nanoparticles are regarded as a single binding event that truly reflects the target number39. We then acquired the SERS spectra from two individual SERS nanoboxes (Fig. 2c, (1) and (2)) and two separate spots of bare silicon (Fig. 2c, (3) and (4)). The presence of nanoboxes showed the characteristic Raman signal at 1330 cm−1 from DTNB, whereas the silicon spectra (3) and (4) lacked the specific peak. This observation demonstrated the single-particle SERS activity of nanoboxes, which was largely attributed to the enhanced electromagnetic fields of nanoboxes on specific regions (e.g., tips and corners)40,41 and thereby facilitated the sensitive and accurate counting of cytokines. Based on the acquired SERS mapping image, the median (interquartile range) of the DTNB peak intensity (1330 cm−1) in the presence and absence of nanoboxes were 183.03 a.u. (149.48–243.35 a.u.) and 18.07 a.u. (15.51–23.12 a.u.), respectively. Furthermore, the mean ± standard deviation of the DTNB peak intensity with nanoboxes (213.41 ± 85.03 a.u.) distinguished clearly from the position without nanoboxes (18.79 ± 6.01 a.u.), which demonstrated the feasibility of correctly identifying the presence of nanoboxes.Fig. 2 Demonstration of the single-particle SERS activity of DTNB-labelled nanoboxes.
a SEM image and b corresponding SERS mapping image of DTNB-labelled nanoboxes on a silicon substrate; c representative SERS spectra of numbered locations indicated in a and b. The red dotted line shows the characteristic peak at 1330 cm−1 from DTNB. Data from one independent experiment. Source data are provided in the Source Data file.
Fig. 3 Study of confocal height on Raman signal intensity.
SERS mapping of FGF-2 SERS nanotags on the silicon substrate with changing confocal height of a 0 nm, b 500 nm, c 1000 nm, and d 1500 nm; selected Raman spectra obtained from e red circles and f blue circles of SERS images. Red dotted lines in e and f indicate peak signal at 1330 cm−1 from DTNB. Data from one independent experiment. Source data are provided in the Source Data file.
Fig. 4 Specificity of digital nanopillar SERS platform for FGF-2 cytokine detection.
Representative confocal SERS images in the presence of a target FGF-2 (1031 aM), and negative controls with non-target controls b G-CSF (1031 aM), c GM-CSF (1031 aM), d CX3CL1 (1031 aM), and e PBS. The median (interquartile range) of active pillars per scanning image for FGF-2, G-CSF, GM-CSF, CX3CL1, and PBS was 72 (63.5–76.75), 1.5 (1.5–2), 2 (1–4), 0.5 (0–1.25), and 1 (1–1.75), respectively. Data from one independent experiment.
Optimisation of digital nanopillar SERS platform for cytokine detection
The reliable detection of single cytokine molecules by the digital nanopillar SERS platform depends on the geometric features of the pillar array (i.e., pillar height, cross-section area of pillar) and assay conditions (i.e., incubation time for sample and SERS nanotags).
We first sought to investigate the effect of pillar height on Raman signal intensity to differentiate signals from non-specifically bound SERS nanotags on the silicon substrate and specifically bound SERS nanotags on the gold-topped pillars. FGF-2 SERS nanotags were randomly deposited on the silicon substrate to mimic the non-specific binding scenario and the Raman mappings were perfomed by moving the objective along the z-axis direction with different heights (0 nm, 500 nm, 1000 nm, and 1500 nm) to compare the signal intensity. In Fig. 3, a–d show the false-colour SERS images and e, f the corresponding SERS spectra with characteristic DTNB reporter peak at 1330 cm−1 acquired from the circled spots in a–d. At the height of 0 nm where the SERS nanotags were in focus, we noticed bright Raman spots (Fig. 3a) and strong Raman signals (black line in Fig. 3e, f). With increasing z heights to 500 nm and 1000 nm, the Raman spots decreased (Fig. 3b, c) and the signal intensity weakened/disappeared (red and blue lines in Fig. 3e, f) as the nanoboxes became increasingly out of focus. A further increase to 1500 nm did not remarkably weaken Raman signals compared to the height of 1000 nm (Fig. 3d). Hence, for the fabrication of the pillar array chip, we selected a pillar height of 1000 nm to greatly reduce the potential interference from non-specific signals.
The cross-section area of the pillars provides the space for cytokine binding and labelling with SERS nanotags. To study the effect of pillar cross-section area, we fabricated array chips with pillars of various widths (250 nm, 500 nm, and 1000 nm) (Supplementary Fig. 7a–c) and functionalised with anti-FGF-2 antibody. Each chip consisted of 250,000 individual pillars. We then analysed a sample that contained ~25,000 molecules of FGF-2 (40 µL, 1031 aM), which should result in 10% active pillars (ratio FGF-2: pillars of 0.1). As seen in Supplementary Fig. 7d–f, an increasing pillar cross-section area results in a higher fraction of active pillars. In reference to the expected active pillar percentage (10%), the 250 nm and 500 nm pillar arrays produced lower active pillars (2% and 6%), which suggested a significant loss of target recognition by SERS nanotags. For the 1000 nm wide pillars, the active pillar percentage was 11%, close to the nominal value of 10%. We further tested a sample with 260 aM FGF-2 (i.e., 2.5% active pillars) on the pillar array chips with 250, 500, and 1000 nm pillar widths. The capture efficiency of these three chips was summarised in Supplementary Table 2. In comparison to the pillar array of 250 nm and 500 nm sizes, the 1000 nm provided an improved capture efficiency. As the accessible target recognition surface area per pillar increases, it can possibly promote the thermodynamics and kinetics for higher surface binding and capture efficiency25. Consequently, the 1000 nm pillar array was adopted in the subsequent experiments.
An optimal incubation time of cytokine and SERS nanotags on the pillar array can shorten the operation time and reduce the potential risk of nonspecific binding that could lead to false-positive counting. We thus studied the effect of incubation time of cytokine with SERS nanotags for 30 to 90 min in a solution of 1031 aM FGF-2 and FGF-2 SERS nanotags. As suggested by Supplementary Fig. 8, the increase in incubation time gives rise to a higher proportion of active pillars. In comparison with the theoretical active pillar percentage (10%), both 30 min and 60 min incubation time were able to provide a desirable active pillar percentage (11% and 13%, respectively). A longer incubation time (90 min), however, reported an active pillar percentage (20%) two times higher than the theoretical, indicating the occurrence of nonspecific absorption of SERS nanotags on the pillar array chip. Thus, we selected 30 min incubation time for further digital nanopillar SERS measurements.
Specificity of the digital nanopillar SERS platform for cytokine detection
Accurate and reliable recognition of the specific target is essential for cytokine quantification in clinical samples. To demonstrate the detection specificity of the digital nanopillar SERS assay, we prepared an anti-FGF-2 antibody functionalised pillar array and measured samples containing target FGF-2 cytokine and controls (G-CSF, GM-CSF, CX3CL1, and PBS). It was observed that only the presence of FGF-2 activated significant amounts of pillars whereas the negative controls only generated negligible active pillars (Fig. 4), indicating the high specificity for FGF-2 detection. Similarly, we studied the specific detection of G-CSF, GM-CSF, and CX3CL1, as shown in Supplementary Figs. 9–11, in which the typical Raman images displayed high proportions of active pillars in the presence of specific targets but not for the negative controls.
To further investigate the specificity of binding between SERS nanotag and antibody-functionalised pillar, we performed SEM analysis to “closely” inspect the pillar array for the presence or absence of SERS nanotags. As a representative model, we selected to image FGF-2 SERS nanotags on the anti-FGF-2-functionalised pillar array after sample incubation with FGF-2 cytokine and non-target controls (Fig. 5). As expected, we observed the cubic nanoparticles on the top of pillars in the presence of FGF-2 due to the successful recognition of SERS nanotags. On the contrary, pillar arrays did not display a significant number of FGF-2 SERS nanotags with non-target controls. Consequently, the consistent Raman and SEM data demonstrate the capability of the assay for specific target cytokine counting. The ability to selectively identify these four cytokines in the designed assay is critically important for their usage in clinical samples.Fig. 5 Specificity of the digital nanopillar SERS platform for FGF-2 cytokine detection.
Representative SEM images of pillar array incubated with FGF-2 SERS nanotags in the presence of a, b FGF-2 (1031 aM), c G-CSF (1031 aM), d GM-CSF (1031 aM), e CX3CL1 (1031 aM), and f PBS. The red circles highlight the existence of SERS nanotags. Panel b is the magnified SEM image of the red-highlighted section in a. It is noted that nanofabrication debris on the sidewall of the pillars can also be seen. Data from one independent experiment.
Sensitivity of the digital nanopillar SERS platform for cytokine detection
As there is typically low abundance of cytokines in clinical samples, the technique based on cytokine detection is expected to possess sufficient sensitivity to reliably assess irAEs9. To investigate the sensitivity and dynamic detection range of the digital nanopillar SERS assay, we firstly titrated the designated concentration of one target cytokine (FGF-2) on the pillar array chip with 250,000 pillars. To comply with the Poisson distribution, the upper number of cytokine molecules in the sample is 25,000 which should result in 10% activated pillars. Based on this upper molecule number, we were motivated to challenge the assay by serially diluting the number of cytokine molecule in the sample from 25,000 (1031 aM), 6305 (260 aM), 631 (26 aM), and 63 (2.6 aM). As suggested by the Raman images in Supplementary Fig. 12 and with the decrease in FGF-2 molecules, the percentage of active pillars decreased correspondingly from 9.39% for 1031 aM, 6.59% for 260 aM, 1.12% for 26 aM, and 0.62% for 2.6 aM, showing a strong correlation that facilitates quantitative cytokine analysis.
Subsequently, we were interested in exploring the multiplexing capability of SERS to investigate the digital nanopillar SERS assay’s dynamic range for the simultaneous quantification of all studied cytokines. As the targets independently follow Poisson distribution, each of the cytokine was separately controlled to activate less than 10% pillars. The specific SERS nanotags provided unique signals for each cytokine that was visualised in the false-colour SERS images by a different colour, thereby enabling in situ and simultaneous cytokine detection. As suggested by the confocal SERS images in Fig. 6, an increase in cytokine concentration corresponded with a higher percentage of active pillars. To facilitate quantitative measurements of the cytokines, we calculated the logarithmic transformation of the percentage of active pillars versus cytokine concentration (Supplementary Fig. 13) confirming the strong statistical and potentially clinically relevant correlation (coefficient of determination (R2) >0.97) observed in the SERS images.Fig. 6 Sensitivity for the simultaneous detection of four cytokines.
Representative confocal SERS images of fibroblast growth factor 2 (FGF-2), granulocyte colony-stimulating factor (GM-CSF), granulocyte colony-stimulating factor (G-CSF), and fractalkine (CX3CL1) with the concentration of a 2.6 aM, b 26 aM, c 260 aM, d 1031 aM. Colour scale bars indicate Raman intensities from 5,5-dithiobis (2-nitrobenzoic acid) (DTNB), 4-mercaptobenzoic acid (MBA), 2,3,5,6-tetrafluoro-4-mercaptobenzoic acid (TFMBA), or 2‐mercapto‐4‐methyl‐5‐thiazoleacetic acid (MMTAA). The median (interquartile range) of active pillars per scanning image of FGF-2, G-CSF, GM-CSF, CX3CL1 for 2.6 aM: 3 (1.5–3), 1 (1–2), 2 (1–3), 2 (1–3); 26 aM: 8 (5.5–10), 10 (9–13), 7 (6–10), 8 (6–10); 260 aM: 40 (36–48), 40 (35–52), 39 (35–50), 37 (36–49); and 1031 aM: 79 (61.5–97), 78 (72–87.5), 88 (68.5–97), 79 (64–95), respectively. Data represents one experiment from three independent tests.
To further investigate the multiplexing quantification performance of the digital nanopillar SERS assay in human serum, we spiked standard cytokines in human serum and tested the dynamic range. Supplementary Fig. 14 shows the linear relationship curves for the four targets. Because of the more complicated sample matrix composition in human samples, the lowest detectable cytokine concentration (5.2 aM) was higher than the PBS solution (2.6 aM).
At a cytokine to pillar ratio of 1:10, we studied the probability of each pillar being occupied by different molecule numbers. To experimentally investigate the number of molecules on a single pillar, we analysed a cytokine mixture that contained all four target cytokines at equal concentration (i.e., ~6250 molecules per cytokine). To visualise and count molecule binding events on a single pillar, we labelled the captured cytokines with the four SERS nanotags that provide clearly distinguishable signals. Under Poisson distribution, the likelihood of having two or more molecules on a single pillar is <0.45% (Supplementary Table 3), which underlies the digital counting principle24. Compared to the theoretical Poisson distribution, the experiment data reported a close but slightly higher value, which was probably due to minor non-specific binding of SERS nanotags on the pillars.
The high sensitivity (attomolar level) of the digital nanopillar SERS assay can be ascribed to the following factors: the digital counting strategy, the single-particle SERS activity of the nanoboxes, and the use of pillars to suit confocal Raman mapping that efficiently excludes false-positive signals. Commercially available methods with potential for trace analysis of cytokines include the single-molecule ELISA Simona by Quanterix and electrochemical luminescence assay42,43 by Meso Scale Discovery. Compared to these two methods, the developed digital nanopillar SERS platform enabled in situ multiplexed detection of four cytokines with comparable sensitivity. Unlike the issues of photo bleaching and poor multiplexing analysis often encountered in fluorescence44 and luminescence assays45, SERS provides the advantage of high multiplexing (e.g., 31-plex)46–48 with the narrow Raman linewidth and high photo stability of the Raman reporters. In addition, this digital nanopillar SERS platform can provide more accurate quantification of cytokines by reducing the false-positive signals with the confocal setting, thus eventually help clinicians to monitor irAEs during immune checkpoint therapy. The highly sensitive readout for multiple targets also indicated the capability of this assay for cytokine detection to assess irAEs in clinically relevant samples.
Evaluation of digital nanopillar SERS platform on simulated patient samples
The detection of trace concentrations of cytokines in serum samples is difficult because plasma samples contain a high abundance of non-target molecules (e.g., serum albumin and other proteins) that can potentially interfere with cytokine detection and lead to inaccurate clinical results. To evaluate the capability of the digital nanopillar SERS assay in accurately counting single cytokine molecules, we opted to perform a recovery test in simulated patient plasma samples (i.e., healthy human serum spiked with 1 fM of FGF-2, G-CSF, GM-CSF, and CX3CL1). The rapid scan rate (i.e., 0.05 s for Raman signal integration) facilitated the detection of Raman signals from FGF-2, G-CSF, GM-CSF, and CX3CL1 SERS nanotags rather than the non-target molecules present in human serum due to their low Raman cross-section. As a representative example, Supplementary Fig. 15 shows the Raman signal distribution of the FGF-2 SERS nanotags on five different spots on the pillar array obtained from the recovery test without noticeable Raman signals from other molecules. It is worth noting that unlike the solution-based DTNB labelled SERS nanotag spectra in Fig. 1d, some of the peaks at 1556 cm−1 and 1330 cm−1 in Supplementary Fig. 15 had a similar intensity, which was probably because of the different orientation of the anisotropic nanoboxes on the substrate relative to the polarisation of excitation laser49. Supplementary Tables 4 and 5 show the cytokine concentrations in healthy human serum and human serum spiked with 1 fM cytokine standards determined by digital nanopillar SERS platform, respectively. On five independent pillar arrays, the measured concentrations had the relative standard deviation (RSD) below 9.0% and the Kruskal–Wallis test showed no statistical differences among these results (p » 0.05). Overall, the observed inter-chip variation should enable accurate identification of disease progression to severe irAEs (e.g., grade 3 or 4), but may encounter some challenges in discriminating mild progressing to moderate irAEs (e.g., grade 1 or 2). The assay enabled trace determination of the four targets in simulated human serum as suggested by the target recovery rates of 80.00% to 137.00% with RSD from 16.02% to 21.80% (Supplementary Table 6). Importantly, the ability to measure reliably cytokines at attomolar levels in simulated human serum samples holds promise for detecting early changes in cytokine concentrations as predictors for the emergence of irAEs in immune checkpoint blockade treated patients.
To validate the accuracy of the digital nanopillar SERS assay, we compared the assay with commercially available ELISA kits (one kit for each cytokine tested) for cytokine quantification. To represent a potential clinical scenario of a patient developing irAEs during immune checkpoint blockade therapy, we prepared three samples with increasing concentrations of cytokines (spike in experiments into fetal bovine serum (FBS)) and subsequently analysed these samples with our digital nanopillar SERS assay and the commercial ELISA kits. FBS was used as complex sample matrix devoid of human cytokines. As the limits of detection for the ELISA kits (FGF-2 = 0.95 pM, G-CSF = 1.66 pM, GM-CSF = 1.11 pM, and CX3CL1 = 17.86 pM) were above the attomolar level, the simulated samples were prepared to suit the detection range of these kits. For the digital nanopillar SERS assay, the samples were diluted correspondingly and generated consistent results with the ELISA kits as shown in Supplementary Table 7. No statistical differences were found between ELISA and digital nanopillar SERS results based on Mann–Whitney test. Furthermore, we compared the detection of four cytokines in human serum with digital nanopillar assay and ELISA kits (Supplementary Table 8). The cytokine levels in human serum were below the limit of detection for the conventional ELISA kits, whereas their concentration was quantified by digital nanopillar SERS platform. For the human serum spiked with standard cytokines, the digital SERS platform generated similar results to ELISA without significant differences by Mann–Whitney test. Collectively, the digital nanopillar SERS platform showcased the ability to robustly and accurately quantify cytokines in complicated samples, which is significant for the prospect of dynamic correlation monitoring of irAEs in clinical samples.
Following the demonstration of the accuracy of digital nanopillar SERS platform, we tested the four cytokine levels in ten healthy people (Supplementary Table 9). These ten healthy people showed cytokine concentrations beyond the conventional ELISA capability to accurately quantify, which was consistent with previous reports9,50,51.
Dynamic correlation monitoring of irAEs in melanoma patients receiving immune checkpoint blockade treatment
Having established the feasibility of digital nanopillar SERS in simulated clinical samples, we applied the platform for longitudinally monitoring irAEs in ten melanoma patients (2–3 time points per patient, 26 samples in total) who underwent immune checkpoint blockade therapy (Supplementary Table 10). By diluting the patient samples to follow Poisson distribution, we quantified the cytokine concentration using digital nanopillar SERS platform. Based on the clinical assessments, the patients were classified into two categories: (i) developed severe irAEs (grades 3 and 4) and needed hospitalisation and dedicated treatment (Patients 1, 2, 3, 4, 5); and (ii) developed minor irAEs (grades 1 and 2) that could be managed with immunosuppressants (e.g., corticosteroids) or exhibited no symptom of irAEs (Patients 6, 7, 8, 9, 10).
As a representative case, Fig. 7 shows two cytokine profiles of a patient with severe irAEs (Patient 1) and a patient with mild irAEs (Patient 1). For Patient 1 who received ipilimumab (cytotoxic T-lymphocyte antigen-4 (CTLA-4) inhibitor) and was checked on days 7, 21, and 42, the confocal SERS images showed an increase in active pillars with the continuation of treatment (Fig. 7a–c), suggesting an elevation of the cytokine levels that could potentially trigger the severe irAEs. In agreement with the Raman images, the quantitative counting results for the four cytokines also corroborated the increase of cytokine concentrations peaking in sub-fM levels (Fig. 7d). These cytokine levels were below the limit of detection of conventional ELISA kits (pM level). Importantly, we observed significantly elevated cytokine concentrations in Patient 1 serum on day 42 compared to days 7 and 21. This patient showed the onset of grade 4 irAEs (i.e., colitis) 13 days later (day 55), consistent with the concept that higher cytokine levels correlate with increased risk of developing irAEs9. To further evaluate the utility of these four biomarkers as a signature in identifying and characterising irAEs, we analysed all the counting data from Patient 1 by applying linear discriminant analysis (LDA). As seen in Fig. 7e, the LDA successfully distinguished the data on day 42 into a separate zone from days 7 and 21, which may indicate the potential value of biomarkers in monitoring irAEs development. We further demonstrated Patient 1 LDA with the use of all combinations of two (Supplementary Fig. 16) and three cytokines (Supplementary Fig. 17). Overall, the LDA with four cytokines showed improved classification over using three or less cytokines. Interestingly, considering FGF-2/G-CSF, G-CSF/GM-CSF, and G-CSF/CX3CL1, the LDA generated similar performance to the LDA with four cytokines. To further compare the classification power of FGF-2/G-CSF, G-CSF/GM-CSF, and G-CSF/CX3CL1, and four cytokines, we performed LDA of Patient 2 (Supplementary Fig. 18), which suggested a better differentiation with the use of four cytokines. Therefore, the inclusion of all four cytokines in LDA facilitated a wider and more accurate patient sample analysis. Similarly, Patients 2, 3, 4, and 5, who manifested severe irAEs were connected with higher cytokine levels (Supplementary Fig. 19) and amelioration of irAEs symptom was witnessed with a decrease of cytokine concentrations. For these severe irAEs patients, the LDA model showed a clear discrimination in cytokine profile and this could help to identify patients at risk of irAEs (Supplementary Fig. 19).Fig. 7 Digital nanopillar SERS assay for monitoring melanoma patients during immune checkpoint therapy.
For Patient 1 who developed severe irAEs, SERS images for cytokine detection on a day 7, b day 21, c day 42, d cytokine concentration graph for fibroblast growth factor 2 (FGF-2), granulocyte colony-stimulating factor (G-CSF), granulocyte-macrophage colony-stimulating factor (GM-CSF), and fractalkine (CX3CL1). The two shorter horizontal lines denote the interquartile ranges (25th and 75th percentile) and the longer horizontal lines in between denote the median (50th percentile), and e LDA analysis, respectively. For Patient 6 who developed mild irAEs, SERS images for cytokine detection on f day 0, g day 21, h day 42, i four cytokine concentration graph, the two shorter horizontal lines denote the interquartile ranges (25th and 75th percentile) and the longer horizontal lines in between denote the median (50th percentile), and j LDA analysis, respectively. IPI ipilimumab, PEMBRO pembrolizumab; G3 grade 3, G2 grade 2; SD stable disease, PR partial response. For Patient 1, the median (interquartile range) of active pillars per scanning image of FGF-2, G-CSF, GM-CSF, CX3CL1 on day 7: 14 (11–22.5), 23 (21, 29), 12 (7.5–18), 17 (9–25.5); day 21: 30 (19–37.5), 33 (19–41), 26 (17.5–36.5), 29 (21–43); and day 42: 33 (16.5–58.5), 76 (64–128.5), 25 (14–39.5), 48 (26.5–73.5), respectively. For Patient 6, the median (interquartile range) of active pillars per scanning image of FGF-2, G-CSF, GM-CSF, CX3CL1 on day 0: 18 (16–23), 49 (31.5–56), 23 (17.5–28), 20 (14.5–27); day 21: 29 (24–33.5), 53 (46.5–70), 35 (25–46), 22 (19–29.5); and day 42: 13 (8–16.5), 44 (23.5–55.5), 10 (6.5–12.5), 30 (24–34.5), respectively. The data represented three technical replicates obtained from three chips. Nine images were acquired from each chip for cytokine counting. Statistical analysis was based on Kruskal–Wallis test followed by Dunn’s test to correct multiple comparisons (two-sided). Source data are provided in the Source Data file.
As for Patient 6 who exhibited mild grade 2 irAEs on the skin during combined ipilimumab and pembrolizumab (programmed death-1 (PD-1) inhibitor) or single ipilimumab treatments, the dynamic monitoring displayed relatively stable cytokine levels on different follow-up visits (Fig. 7). Specifically, the confocal Raman images (Fig. 7f–h) and the molecular counting (Fig. 7i) in this patient serum consistently showed no significant cytokine level alterations on the three time points (days 0, 21, 42). Under this circumstance, LDA failed to clearly classify the data into separate sections (Fig. 7j). Likewise, Patients 7, 9, and 10 possessed stable cytokine levels and were diagnosed with low grade irAEs. Meanwhile, Patient 8 who showed decreasing cytokine levels did not display signs of irAEs. LDA was not able to classify Patients 7 and 8 who had mild irAEs and did not show irAEs, but it recognised the minor difference in Patients 9 and 10 who showed grade 1 irAEs (Supplementary Fig. 20).
Overall, we found the preliminary evidence to suggest that significantly elevated cytokine levels have a strong correlation with the development and manifestation of severe irAEs, whereas stabile, low baseline, and decreasing cytokine concentration indicate mild and manageable irAEs. The relatively low concentrations of these four cytokines were below the detection sensitivities of commercially available ELISA kits (pM level), which limits their use in clinical studies. Importantly, the measurement at fM cytokine levels in clinical samples is consistent with the median concentrations of cytokine measured by using digital ELISA51. The successful demonstration of the digital nanopillar SERS platform in dynamic detection of cytokines in patient serum provides a potential approach for the future accurate early detection, characterisation, and monitoring of irAEs in clinical settings. However, it is important to note that the cytokine concentration changes are not directly correlated with the treatment response according to response evaluation criteria in solid tumours (RECIST). In our pilot study, some melanoma patients showed higher levels of cytokines compared to the healthy controls. The power of our digital nanopillar SERS platform lies in the capability to longitudinally monitor cytokines in individual patients over time.
Discussion
Despite the frequent occurrence of irAEs in immune checkpoint therapy, particularly for the combination treatment, the prediction of the emergence of irAEs remains elusive. Mounting data suggest a potential role of cytokines as predictive markers for irAE monitoring in immune checkpoint therapy9,11,13,52. Although promising, accurate quantification of these biomarkers was often not possible due to the dearth in technologies with sufficient detection sensitivity. Typically, either cytokines above the detection limit of immunosorbent assay were selected11, or relative cytokine quantification9 was performed for investigating irAEs. The former approach has the drawback of potentially excluding the low abundance cytokines of significance in irAEs. As for relative quantification9,53, the cytokine concentrations are determined by relating to a standard that had the observed median fluorescence value closest to the median of the test sample. The relative concentration, however, may fail to represent accurate cytokine levels and thus needs further exploration. Our developed digital nanopillar SERS assay offers early data suggestive of a potential approach to the above-mentioned challenges and provides the possibility to study trace amounts of a panel of cytokines in an accurate quantification manner as well as concomitantly providing attomolar level sensitivity. The current proof-of-principle approach has measured four of the potential inflammatory and/or immune toxicity-related cytokines for the prediction of emergence, characterisation, and/or quantifiable correlation with irAEs in melanoma patients.
By leveraging the narrow line width of Raman spectra, the developed digital SERS counting assay shows the ability to sensitively and simultaneously detect multiple cytokines. The adoption of the novel digital quantification mode in SERS using gold–silver alloy nanoboxes further improves the high sensitivity of SERS technology. Notably, the digital counting strategy offers an option for reproducible SERS quantification by avoiding the common Raman signal fluctuations induced by ensemble measurements. The ensemble measurement in SERS relies on the enhancement of Raman signals of molecules located in or near the “hot spots” (i.e., strong electromagnetic fields)30. Due to the random distribution and various efficiencies of “hot spots”, it can result in the discrepancies in acquired SERS intensities for inter-laboratory and even intra-laboratory tests29. To circumvent the impact of Raman intensity fluctuation on accurate quantification, we employed the digital SERS signals from the single-SERS-active nanoboxes on discrete pillar arrays to enumerate the targets and only count the “yes” or “no” signal for a robust and reproducible SERS analysis. Furthermore, the digital readout model, which regards both aggregated and single nanoparticle as a single binding event to reflect the true target number, can have a better accuracy and robustness than the intensity-based assay39.
We believe that the proposed digital nanopillar SERS assay could be used to monitor other cytokine-induced immune responses. For instance, with the outbreak of 2019 novel coronavirus (2019-nCoV), it is yet difficult to predict which infected patient will develop a strong immune response that requires hospitalisation. However, cytokines have been indicated to play a major role in the severity of immune response for critically ill patients infected with 2019-nCoV54. The specific detection of multiple cytokines at early stages of viral infection could thus potentially address this issue and help to provide the clinical care for people at the highest risk. For patients with high cytokine concentrations, the digital nanopillar SERS platform will require the sample dilution to suit Poisson distribution.
In summary, we propose a digital nanopillar SERS platform for the parallel counting of single cytokines and dynamic monitoring in the clinical context of irAE development during immune checkpoint blockade therapy. The platform achieved attomolar level sensitivity by utilising discrete pillar array compartments to hold the single cytokine and subsequently applied single-particle active nanobox-based SERS nanotags for cytokine identification and counting. The confocal Raman mapping on the pillar array offered the highest possible clinical specificity by reducing nonspecific signals and provided a “yes/no” type counting approach for reproducible Raman signal readout. The designed platform was rigorously optimised and tested in simulated clinical samples prior to the evaluation for irAE monitoring in stage IV melanoma patients receiving immune checkpoint blockade therapy. We envisaged this platform possessing the advantages of highly sensitive and multiplexing analysis capability can transit into future irAE detection methodologies after extensive validation in a large cohort of clinical samples over different time courses.
Methods
Materials
Silver nitrate (AgNO3), hydrogen tetrachloroaurate (III) trihydrate (HAuCl4·3H2O), MBA, DTNB, TFMBA, MMTAA, DSP were obtained from Sigma Aldrich. Ascorbic acid (AA) of analytical grade was purchased from MP Biomedicals. FGF-2 (223-FB), G-CSF (214-CS), GM-CSF (215-GM), CX3CL1 (365-FR) cytokines; monoclonal anti-FGF-2 (MAB233), anti-G-CSF (MAB214), anti-GM-CSF (MAB615), anti-CX3CL1 (MAB3652) antibodies; polyclonal anti-FGF-2 (AF-233), anti-G-CSF (AF-214), anti-GM-CSF (AF-215), anti-CX3CL1 (AF-365) antibodies; and FGF-2 (DY233-05), G-CSF (DY214-05), GM-CSF (DY215-05), and CX3CL1 (DY365) ELISA kits were bought from R&D Systems.
All the patient serum or plasma samples were collected at the Austin Hospital (Melbourne) under approved human ethic protocols and written informed consents were obtained from all patients before sample collection. Ethics approval was obtained from The University of Queensland Institutional Human Research Ethics Committee (approval nos. 2011001315 and 2016000876) and the following clinical assay was carried out according to the approved guidelines.
Preparation of single-particle active SERS nanotags
The preparation of SERS nanotags involved the synthesis of nanoboxes and the subsequent functionalisation with Raman reporters and antibodies. For nanobox synthesis, 45 µL of HAuCl4 (1 wt%) was added into 10 mL of ultrapure H2O (18.2 Ω cm) under magnetic stirring (800 r.p.m.) for 1 min, followed by simultaneously introducing 170 µL of AgNO3 (6 mM) and 30 µL of AA (0.1 M) into the stirring solution. Then, the formation of nanoboxes was indicated by the appearance of an apparent blue colour within 6 s and the samples were collected 1 min later by centrifuging at 600 g for 15 min.
To functionalise nanoboxes with Raman reporters and antibodies, 300 µL of nanoboxes centrifuged from 1 mL of as-prepared solution were co-incubated with one type of Raman reporters (i.e., 10 µL of DTNB, 8 µL of MBA, 10 µL of TFMBA, or 10 µL of MMTAA) and 2 µL of DSP linker for 6 h. After that, the Raman reporter and DSP functionalised nanoboxes were separated by centrifuging at 600 g for 15 min, and resuspended into 300 µL of PBS (0.1 mM). Then, 2 µg of anti-FGF-2, anti-G-CSF, anti-GM-CSF, anti-CX3CL1 antibodies were added to MBA, DTNB, TFMBA, and MMTAA labelled nanoboxes, respectively. After overnight incubation at 4 oC, the functionalised nanoboxes were purified by centrifuging at 600×g for 15 min to separate free antibodies and the final products were resuspended into 200 µL of 0.1% BSA for future use.
Fabrication of pillar arrays
The chip was made of a sensing array measuring 1 mm × 1 mm (Supplementary Fig. 1a) and consisted of 250,000 individual pillars. Each pillar was 1 µm wide, 1 µm long, and 1 µm high. The pillars were evenly spaced by 1 µm from one pillar to the next (Supplementary Fig. 1b). The pillar array was designed using Nanosuite 6.0 (Raith GmbH) and Beamer 5.9.1 (GenISys GmbH) and fabricated on a 4-inch p-type <100> silicon wafer (Bonda Technology Pte Ltd, Singapore) using electron beam lithography (EBL). The wafer accommodated 76 separate pillar arrays. Silicon wafer was first cleaned in acetone, isopropanol with sonication for 2 min each, followed by rising with deionised H2O and dehydration bake at 180 °C for 2 min. Prior to the resist coating, the wafer had undergone a further O2 plasma cleaning at 200 W for 5 min (Diener Atto, Diener Electronic GmbH). The cleaned wafer was spin-coated with two layers of polymethyl methacrylate (bottom: 495K A4 PMMA, top: 950k A4 PMMA, from MicroChemicals GmbH) using the CEE Apogee Coater (Cost Effective Equipment, LLC) at 1500 r.p.m. for 60 s each. After the coating of each layer, the wafer was baked immediately on a hot plate to prevent from intermixing of the two layers of resist. The baking time was 10 and 3 min for the bottom and top layer, respectively, at 180 °C. The thickness of the photoresist was found to be ~450 nm (top PMMA: ~250 nm, bottom PMMA: ~200 nm), characterised by white light reflectometry (FilmTek 2000M, Scientific Computing International). EBL was performed in the Raith EBPG5150 system. The patterns were exposed in EBL with an accelerated voltage of 100 kV, 150 nA of beam current (spot size ~80 nm), with step sizes of 40 nm and an electron dose of 1200 µC/cm2. The exposure time per 4-inch wafer was ~35 min, each containing 76 individual chips. After exposure, the wafer was developed in a mixture of isopropanol and methyl isobutyl ketone (3:1) for 60 s and rinsed immediately with isopropanol, followed by drying with N2. An oxygen plasma descum process, at 100 W, 60 s (Diener Atto, Diener Electronic GmbH) was carried out to remove resist residues prior to the deposition. Next, 10 nm titanium and 200 nm gold were deposited by physical vapour deposition using a Temescal FC-2000 electron beam evaporator (Ferrotec, U.S.A.). After overnight lift-off at room temperature in Remover PG (MicroChemicals GmbH, Germany), the excess material was washed off and the pillar array structure was revealed (Supplementary Fig. 1c). To create the pillar height (i.e., 1 µm), reactive ion etching (Oxford Instruments, UK) was applied for anisotropic etching of the silicon. Hereby, the deposited gold served as mask to protect the underlying silicon while the un-masked silicon was removed. Next, the wafer was coated with a protective layer of cured AZnLOF 2020 prior to wafer dicing into 76 individual sensing chips consisting of a single pillar array. Prior to use, the protective layer was washed off by consecutive washes with isopropanol and acetone and dried under a stream of nitrogen.
Pillar array functionalization
Antibody functionalisation of the gold-topped pillar array was conducted by crosslinking the antibodies to the gold surface using DSP. A solution of 5 mM DSP in dimethyl sulfoxide was pipetted onto the pillar array and incubated at room temperature for 2 h. After rinsing the pillar array with ethanol and PBS, a solution of 5 µg/mL anti-cytokine monoclonal antibody solution (100 fold dilution of antibody stock solution) in PBS was incubated overnight at 4 °C. Subsequently, the pillar array was rinsed with PBS and blocked using 1% bovine serum albumin in PBS for 1 h. Prior to use, the pillar array was rinsed with PBS. All PBS solutions were filtered through a sterile 0.22 µm syringe filter (Millex-GP, Merck, U.S.A.).
Digital nanopillar SERS profiling of cytokines
Cytokines (FGF-2, G-CSF, GM-CSF, and CX3CL1) with different concentrations in PBS (2.6 aM, 26 aM, 260 aM, and 1031 aM) were incubated with antibody functionalised pillar arrays at room temperature for 30 min, followed by washing the pillar array three times with washing buffer (0.1% BSA and 0.01% Tween 20 in PBS). SERS nanotags were then added into the pillar array for another 30 min incubation under room temperature to identify the targets. Finally, the pillar arrays were washed to remove the free SERS nanotags and were subject to confocal Raman microscope for quantification. For each sample, nine SERS images with each image has the dimension of 60 µm × 48 µm were taken on the pillar array to calculate the overall cytokine concentration.
Simulated clinical sample detection
For the recovery experiment, standard cytokines (FGF-2, G-CSF, GM-CSF, and CX3CL1) with the concentration of 1 fM were added into healthy human serum and then diluted ten times with PBS to quantify.
For the quantification of cytokines in FBS, three simulated clinical samples were prepared by titrating various concentrations of standard cytokines into 10% FBS: Sample 1 (FGF-2 = 3.64 pM, G-CSF = 3.19 pM, GM-CSF = 4.29 pM, and CX3CL1 = 28.57 fM); Sample 2 (FGF-2 = 7.28 pM, G-CSF = 6.38 pM, GM-CSF = 8.58 pM, and CX3CL1 = 571.4 fM); and Sample 3 (FGF-2 = 14.56 pM, G-CSF = 12.76 pM, GM-CSF = 17.16 pM, and CX3CL1 = 1142.8 fM). These three samples were then detected directly using the commercial ELISA kits or digital nanopillar SERS assay with a further dilution of 105, 2 × 105, and 4 × 105, respectively.
Spectroscopic ellipsometry
The antibody film thickness was measured by in-solution spectroscopic ellipsometry (M2000V JA Woollam Co., Inc. USA) using gold-coated substrates and flow cell (QSense® Ellipsometry, Biolin Scientific, Sweden). Measurements were performed at an angle of 65°. Data analysis was performed by CompleteEASE® software using a B-Spline data fit and Cauchy model to calculate the antibody film thickness.
MALDI-TOF MS
The antibody-functionalised nanopillar array chip was subjected to tryptic digest prior to analysis. Sequencing-grade trypsin was made to 50 ng/µL in 25 mM ammonium bicarbonate, and sprayed over the chip using a Bruker Imageprep instrument (Bruker, USA). After trypsin deposition, the chip was incubated in a humid environment at 40 °C for 3 h. Subsequently, the chip was sprayed with a matrix solution, 10 g/L α-cyano-4-hydroxycinnamic acid in 50% acetonitrile with 0.2% trifluoroacetic acid. Next, the chip was analysed with a Bruker Ultraflex III MALDI-TOF mass spectrometer (Bruker, USA) in positive linear mode using Flex Imaging 4.0 (Bruker, USA) with a pixel size of 60 µm. Data were collected from 2 k–30 k m/z, at a laser repetition rate of 200 Hz. Data were normalised using the root mean square approach and visualised using Flex Imaging 4.0 (Bruker, USA) and SCILS LAB 2017a software. For the SCILS LAB analysis, the data were imported using a convolution baseline subtraction, and displayed using root mean squared normalisation.
Instrumentations
SEM images of pillar arrays and nanoboxes were taken on a JEOL-7100 field emission (FE)-SEM (20 kV voltage). TEM images of nanoboxes were taken on a JEOL-2100 microscope (200 kV voltage). NTA of nanobox size distribution was performed with Malvern NanoSight NS300. UV-vis extinction spectrum of nanoboxes was performed with a Shimadzu UV-2450 spectrophotometer. Confocal Raman mapping was conducted on a WITec alpha 300 R spectrometer using 632.8 nm He–Ne laser with the power of 35 mW, a grating of 600 g/mm used with EMCCD camera, spectral resolution of 1.390 cm−1 to 2.114 cm−1, confocal pinhole size of 100 µm, 100× air objective with NA of 0.90, and 0.05 s integration time. The theoretical spot size was 857.80 nm based on the Abbe diffraction limit (i.e., d = 1.22λ/NA). The scanning area was set to have 60 µm × 48 µm with 86 points per line and 69 lines per image. For each pillar array, nine separate scanning areas were taken in total and the total active pillars were used for quantification. The SERS mapping images for counting were taken by focusing the laser on the top of the pillar surfaces. Specifically, the laser was firstly focused on the silicon substrates by obtaining the strongest silicon signals (520 cm−1) and then the 100× objective was moved up in z-axis direction of 1 µm for SERS scanning. The system was calibrated with the first-order photo peak of silicon at 520 cm−1.
Data analysis
To assign the SERS nanotag membership for DTNB, MBA, TFMBA, and MMTAA, Project Five 5.0 software from WITec was utilised to create four filters, which summed a spectral range of 40 cm−1 with the centre position at the characteristic Raman peak of each reporter and subtracted the background with a polynomial algorithm. Specifically, the filter ranges of four Raman reporters DTNB-, MBA-, TFMBA-, and MMTAA-coated SERS nanotags were (1310–1350 cm−1), (1060–1100 cm−1), (1360–1400 cm−1), and (1268–1308 cm−1), respectively. All the SERS images were analysed by using threshold intensity to determine the successful binding events. Specifically, the threshold intensity of FGF-2, G-CSF, GM-CSF, and CX3CL1 was set at 5000, 4000, 5000, and 5000, respectively. For each image, the threshold intensity was doubled-checked and adjusted based on the true Raman peaks in the spectra. Statistical analysis assuming unequal variances was conducted with Kruskal–Wallis test among three groups or Mann–Whitney test between two groups with GraphPad Prism 8.4. To control the error appropriately, we performed multiple comparisons using Dunn’s test. LDA of clinical samples was performed in R software (3.6.2) with the MASS package (7.3-52). The active pillars in SERS images were counted with Image J software.
Supplementary information
Supplementary Information
Peer Review File
Source data
Source Data
Peer review information
Nature Communications thanks Chongwen Wang and Isaac Pence for their contribution to the peer review of this work. Peer reviewer reports are available.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
The online version contains supplementary material available at 10.1038/s41467-021-21431-w.
Acknowledgements
The authors acknowledge grants received by our laboratory from the National Breast Cancer Foundation of Australia (CG-12-07) and the ARC DP (160102836 and 210103151). These grants have significantly contributed to the environment to stimulate the research described here. J.L. acknowledges support from the Australian Government Research Training Program Scholarship. A.W. and A.A.I.S. thank the National Health and Medical Research Council for funding (APP1173669 and APP1175047). A.B. is the recipient of a Fellowship from the Victorian Government Department of Health and Human Services acting through the Victorian Cancer Agency. We acknowledge the facilities, and the scientific and technical assistance, of the Australian Microscopy & Microanalysis Research Facility at the Centre for Microscopy and Microanalysis, The University of Queensland. We appreciate to receive the technical and scientific guidance from Queensland node of the Australian National Fabrication Facility (Q-ANFF) in confocal Raman mapping and spectroscopic ellipsometry measurement.
Author contributions
J.L., A.W., A.I.S., H-H.C., Y.W., A.B., P.M., and M.T. contributed to the design of the experiments and analysing the data. J.L. and A.W. performed the experiments and prepared the manuscript. All authors read, commented, and edited the manuscript, and assisted during the revision process.
Data availability
Data supporting the findings of this work are available within this paper and the supporting information files. A reporting summary of this work is available as a Supplementary file. Source data are provided with this paper.
Competing interests
The authors declare no competing interests. | IPILIMUMAB, PEMBROLIZUMAB | DrugsGivenReaction | CC BY | 33597530 | 19,690,448 | 2021-02-17 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Pulmonary toxicity'. | A digital single-molecule nanopillar SERS platform for predicting and monitoring immune toxicities in immunotherapy.
The introduction of immune checkpoint inhibitors has demonstrated significant improvements in survival for subsets of cancer patients. However, they carry significant and sometimes life-threatening toxicities. Prompt prediction and monitoring of immune toxicities have the potential to maximise the benefits of immune checkpoint therapy. Herein, we develop a digital nanopillar SERS platform that achieves real-time single cytokine counting and enables dynamic tracking of immune toxicities in cancer patients receiving immune checkpoint inhibitor treatment - broader applications are anticipated in other disease indications. By analysing four prospective cytokine biomarkers that initiate inflammatory responses, the digital nanopillar SERS assay achieves both highly specific and highly sensitive cytokine detection down to attomolar level. Significantly, we report the capability of the assay to longitudinally monitor 10 melanoma patients during immune inhibitor blockade treatment. Here, we show that elevated cytokine concentrations predict for higher risk of developing severe immune toxicities in our pilot cohort of patients.
Introduction
The advent of immune checkpoint therapy has revolutionised the landscape of traditional cancer treatment and is believed to constitute the backbone of managing certain malignancies1–3. By capitalising on the blockade of immune checkpoint inhibitors to take the brakes off parts of the immune system, this emerging therapy has achieved great success producing long-lasting responses (e.g., 10 years or more) in a small but significant fraction of patients3–6. Nevertheless, upon the blockade of immune checkpoint molecules, the activated and potentiated immune reaction predisposes patients to a significant risk of immune-related adverse events (irAEs), which can occur in up to 80% of patients receiving immune checkpoint therapy7–9. The high incidence of irAEs, which may manifest at any time during treatment, can offset the clinical benefits, lead to premature therapy cessation, and even be life-threatening for certain patients10–12. To assist the successful implementation of immune checkpoint therapy, the use of predictive biomarkers for early identification and vigilant monitoring of irAEs is thus critical and a pressing need in avoiding or ameliorating detrimental effects and adjusting therapeutic options.
Cytokines, small signalling proteins, are promising candidates to indicate the occurrence of irAEs due to their prominent role in modulating the anti-cancer immune responses, including enhancing antigen priming, recruiting immune cells into the tumour microenvironment, and upregulating certain immune checkpoint molecules9,13,14. Particularly, excessive cytokine secretion has been implicated in severe inflammation as a major constituent leading to irAEs. For example, the overproduction of fibroblast growth factor 2 (FGF-2)15–18, granulocyte colony-stimulating factor (G-CSF)19, granulocyte-macrophage colony-stimulating factor (GM-CSF)20, and fractalkine (CX3CL1)21 have been found to participate in immune-related inflammatory disease (e.g., rheumatoid arthritis, autoimmune gastritis, and Crohn’s disease). These inflammatory cytokines have recently been reported to indicate irAEs for melanoma patients who underwent immune checkpoint therapy9. The clinical deployment of cytokine analysis for irAE monitoring is challenging and requires a technology that can (i) determine the selected cytokines with great sensitivity9, especially at the onset of irAE development, where the cytokine concentrations are likely to be the lowest; as well as (ii) simultaneously detect a panel of cytokines to reflect the complex interplay of cytokine signalling pathways22 and the variable irAE symptoms among patients.
Conventional cytokine analyses such as immunosorbent assays have limited clinical applicability for irAE assessment due to their limited capacity to detect low cytokine concentrations in blood as well as for assessing a panel of cytokines in a single sample simultaneously. Recently, advances in micro/nanomaterial-based systems have provided a promising suite of technologies that improve the conventional assays by overcoming the above limitations23,24. Encouragingly, the unique advantages of micro/nanomaterial-based systems convey an attractive option for cytokine analysis with the desired results of high sensitivity and multiplexing. The high specific surface area of these miniaturised materials increases mass transfer subsequently enhancing the interaction with target molecules and thus improving the detection sensitivity25. The capabilities of micro/nanomaterial fabrication techniques permit individually separated compartments sufficiently discrete to hold single molecules and hence encompasses a promising strategy for counting assays that can further push the sensitivity of the traditional assays24,26,27. Moreover, the physicochemical properties of nanostructured materials can be exploited to simultaneously label multiple targets (e.g., various spectral signatures) for high-throughput parallel measurements28–30. Therefore, by combining the potential of micro/nanomaterial systems with the need for sensitive irAE monitoring, we have developed a platform for sensitive and multiplex cytokine counting analysis.
Combining the use of (a) discrete single cytokine nanopillar array chip with discretely separated compartments, (b) control of target concentrations to follow a Poisson distribution, and (c) the recognition of target by single-particle active surface-enhanced Raman scattering (SERS) nanotags with a confocal Raman microscope allows accurate and in situ counting of a multi-cytokine panel (FGF-2, G-CSF, GM-CSF, and CX3CL1). Different from the fluorescence-based digital counting strategies24,26, the strikingly narrow spectral peaks of SERS (~1–2 nm) in comparison to fluorescence (~50 nm) makes this platform intrinsically ideal for multiplexed cytokine analysis28.
In this work, we present a digital nanopillar SERS platform that enables the specific cytokine quantification down to attomolar levels and the application in melanoma patients receiving immune checkpoint blockade therapy. Beyond the capability to predict irAEs in melanoma patients receiving therapy, the digital nanopillar SERS assay could potentially be extended to other cytokine-associated immune responses such as excessive immune activation due to viral or bacterial infections (such as COVID-19).
Results
Digital nanopillar SERS platform for parallel profiling of single cytokine
Our concept of digital nanopillar SERS platform for cytokine analysis relies on Rayleigh criterion separation, probability-driven Poisson distribution, single-particle active SERS nanotags, and confocal SERS mapping (Fig. 1). To precisely fabricate the pillar array, we opted to use an electron beam lithographic approach to write the array into a photon-sensitive material followed by physical vapour deposition of gold to create the gold-topped pillars, and selectively reactive ion etching to reveal the pillar structure (Supplementary Fig. 1). The nanopillar array chip consisted of 250,000 individual pillars. As shown in the scanning electron microscope (SEM) image of Fig. 1a, the cubic nanopillars have an edge-to-edge width of 1000 nm and are evenly distributed at 1000 nm intervals to suit the lateral Raman microscope resolution (~1000 nm) that fulfils the Rayleigh criterion separation required to acquire a single SERS spectrum from each pillar without spectral overlap from adjacent pillars.Fig. 1 Digital single-molecule nanopillar surface-enhanced Raman scattering (SERS) platform for parallel counting of four types of cytokines.
SEM images of a pillar array side view, b nanoboxes, and c a single nanobox on the top of a pillar; d SERS spectra of nanoboxes conjugated with 5,5-dithiobis (2-nitrobenzoic acid) (DTNB), 4-mercaptobenzoic acid (MBA), 2,3,5,6-tetrafluoro-4-mercaptobenzoic acid (TFMBA), or 2‐mercapto‐4‐methyl‐5‐thiazoleacetic acid (MMTAA) Raman reporters; e workflow for multiplex counting of cytokines, including fibroblast growth factor 2 (FGF-2), granulocyte colony-stimulating factor (G-CSF), granulocyte-macrophage colony-stimulating factor (GM-CSF), and fractalkine (CX3CL1). Data from one independent experiment.
By using specific gold-thiol chemistry with the linker molecule dithiobis (succinimidyl propionate) (DSP), the gold-topped pillars were selectively functionalised with target recognition antibodies (anti-FGF-2, anti-G-CSF, anti-GM-CSF, and anti-CX3CL1) and acted as the small compartments to capture and confine the individual cytokine. Upon DSP binding on the gold-topped pillars through gold-thiol bond, DSP uses N-hydroxysuccimide (NHS) ester to react with the amine groups of the antibodies31,32. The successful antibody conjugation on gold-topped pillar surfaces was confirmed by matrix-assisted laser desorption ionisation-time of flight mass spectrometry (MALDI-TOF MS) (Supplementary Fig. 2), which showed high molecular weight fragments derived from antibodies. Furthermore, spectroscopic ellipsometry was utilised to estimate the antibody density on pillar surfaces. Based on the obtained film thickness of 18.5 nm, the calculated antibody surface density was 5.5 mg/m2 using the Cuypers model33, which was in agreement with the reported antibody density on substrate surfaces34. Though these characterisations indicated the presence of antibodies on the pillar array, it was not possible to assess the exact distribution of the four structurally related (same immunoglobulin G family) antibodies on a single pillar with an area of 1 µm2. As an advantage of the digital read-out with a large redundancy of pillars, it is not essential to have all four types of antibodies equally distributed on a single pillar for the assay to work. The combined surface area of all antibody-conjugated pillars provides an excess of cytokine binding sites, which maximises successful cytokine capture within the pillar array. Supplementary Fig. 3 shows the SERS mapping images of an equimolar cytokine solution (1031 aM) that provided a similar signal count for the FGF-2, GM-CSF, G-CSF, and CX3CL1 SERS nanotags, indicating a required distribution of four kinds of antibodies conjugated to the array of pillars. Instrumental to the digital counting of cytokines, we controlled the target concentration based on the principle of Poisson distribution where the ratio of cytokine molecules to pillar number was <1:10, ensuring a 99% probability that there was either one cytokine molecule or zero per pillar24. At a ratio of 1:10, 10% of all pillars were occupied or activated with the cytokine molecules.
Following the capture of cytokines on nanopillars, SERS nanotags were applied to recognise the captured cytokines. The preparation of SERS nanotags was performed by the co-conjugating of Raman reporter and target antibody onto gold–silver alloy nanoboxes. Specifically, an average size of 80 nm gold–silver alloy nanoboxes were firstly synthesised using a rapid and aqueous phase approach35 as indicated in the SEM image in Fig. 1b. Supplementary Fig. 4a, b shows the transmission electron microscope (TEM) image of the nanoboxes with the hollow inner structure and a wall thickness of around 15 nm. Nanoparticle tracking analysis (NTA), which allows the tracking and detection of single particles, shows the nanoboxes have a mode size of 77 nm (D10 = 67.6 nm and D90 = 110.6 nm) (Supplementary Fig. 4c). UV-vis extinction spectroscopy demonstrates the nanoboxes possess a surface plasmon resonance (SPR) peak at 610 nm (Supplementary Fig. 4d). The resonance frequency of the nanoboxes enables a more sensitive signal readout with 632.8 nm laser excitation29, which also has a higher Raman scattering efficiency than 785 nm laser. Thereafter, four types of Raman reporters (5,5-dithiobis (2-nitrobenzoic acid) (DTNB), 4-mercaptobenzoic acid (MBA), 2,3,5,6-tetrafluoro-4-mercaptobenzoic acid (TFMBA), and 2‐mercapto‐4‐methyl‐5‐thiazoleacetic acid (MMTAA)) that generate unique Raman signals (1330 cm−1, 1080 cm−1, 1380 cm−1, and 1288 cm−1) were coupled with their corresponding detection antibodies onto nanoboxes as specific SERS nanotags for identification of FGF-2, G-CSF, GM-CSF, and CX3CL1, respectively. As shown in Fig. 1d, the four SERS nanotags provide the strong and non-overlapping Raman signals, which facilitates the multiplexing analysis of four cytokines. The assignment of the major Raman peaks from the four Raman reporters was summarised into Supplementary Table 1. To evaluate the SERS enhancement property of the nanoboxes, we calculated the enhancement factor (EF) of the four Raman reporters on the nanoboxes. Based on the labelled characteristic peaks in Supplementary Fig. 5, the calculated EFs of DTNB, MBA, TFMBA, and MMTAA were 8.14 × 106, 1.46 × 107, 4.01 × 107, and 3.26 × 107, respectively. The obtained EFs were higher than the reported spherical gold nanoparticles and pure silver nanocubes36 and comparable to the reported hollow nanocubes37, illustrating the high SERS property of the nanoboxes. To investigate the SERS nanotag stability, we monitored the Raman signal intensity over 7 days. As shown in Supplementary Fig. 6, Raman signal intensity variations are less than 5% in the SERS spectra, suggesting the good stability of the prepared SERS nanotags.
The following SERS mapping generated false-colour images for counting single cytokine molecules. Under the Raman microscope, the pillar array was visualised as a blue and black grid by representing the specific Raman shifts corresponding to the silicon signals (520 cm−1), in which the blue colour was assigned to silicon signals showing silicon substrates and the black colour indicated the gold-topped pillars because of the lack of silicon signals. The representation of the four colours of the SERS nanotags (red, green, purple, and cyan) on the gold-topped pillars (i.e., black) reflected cytokine molecule occupation (FGF-2, G-CSF, GM-CSF, and CX3CL1). Elevating the sensing area (or gold-topped pillars) from the silicon substrate was selected as a strategy to minimise the false-positive events. By using the confocal function of the Raman microscope, the laser was selectively focused on the gold-topped pillars, thus largely removing the background signals from potentially non-specifically adsorbed SERS nanotags on the silicon substrate. Finally, the specific SERS nanotag signals present or absent on the gold-topped pillars were counted and represented as percentage of active pillars used for total cytokine quantification. For statistical calculations, SERS mapping was applied for scanning 6480 pillars. This digital counting mode, therefore, has the potential to reach the ultimate sensitivity of single molecule cytokine detection.
Demonstration of the single-particle SERS activity of gold–silver alloy nanoboxes
The successful implementation of digital nanopillar SERS assay necessitates the use of single-particle active plasmonic nanostructures that give a clearly detectable signal for each of the single cytokine binding event. The single-particle SERS detection sensitivity is essential to this assay development as the single-particle inactive plasmonic nanoparticles (e.g., spherical gold nanoparticles)38 would unavoidably result in an underestimate of cytokine concentration.
We evaluated the single-particle SERS activity of the prepared anisotropic nanoboxes by acquiring the signals from the individual nanoboxes that were labelled with DTNB reporters. The use of DTNB, a non-resonant Raman reporter, guaranteed the Raman signal enhancement was solely contributed from the nanobox-generated electromagnetic field. As seen in the SEM image (Fig. 2a), two clearly separated DTNB-labelled nanoboxes were deposited on the silicon wafer (highlighted in red circles). The corresponding Raman image (Fig. 2b) displayed several bright Raman spots originating from these individual nanoboxes. The elongated bright Raman spots in the SERS mapping image were probably caused by the slight aggregation of several nanoboxes during sample preparation processes (e.g., centrifugation)38, which was difficult to visually resolve in the SEM image (Fig. 2a). However, unlike the intensity-based assay, the aggregated nanoboxes as SERS nanotags to target cytokine will not skew the digital readout result, because each cytokine will occupy a single pillar following Poisson distribution and both aggregated and individual nanoparticles are regarded as a single binding event that truly reflects the target number39. We then acquired the SERS spectra from two individual SERS nanoboxes (Fig. 2c, (1) and (2)) and two separate spots of bare silicon (Fig. 2c, (3) and (4)). The presence of nanoboxes showed the characteristic Raman signal at 1330 cm−1 from DTNB, whereas the silicon spectra (3) and (4) lacked the specific peak. This observation demonstrated the single-particle SERS activity of nanoboxes, which was largely attributed to the enhanced electromagnetic fields of nanoboxes on specific regions (e.g., tips and corners)40,41 and thereby facilitated the sensitive and accurate counting of cytokines. Based on the acquired SERS mapping image, the median (interquartile range) of the DTNB peak intensity (1330 cm−1) in the presence and absence of nanoboxes were 183.03 a.u. (149.48–243.35 a.u.) and 18.07 a.u. (15.51–23.12 a.u.), respectively. Furthermore, the mean ± standard deviation of the DTNB peak intensity with nanoboxes (213.41 ± 85.03 a.u.) distinguished clearly from the position without nanoboxes (18.79 ± 6.01 a.u.), which demonstrated the feasibility of correctly identifying the presence of nanoboxes.Fig. 2 Demonstration of the single-particle SERS activity of DTNB-labelled nanoboxes.
a SEM image and b corresponding SERS mapping image of DTNB-labelled nanoboxes on a silicon substrate; c representative SERS spectra of numbered locations indicated in a and b. The red dotted line shows the characteristic peak at 1330 cm−1 from DTNB. Data from one independent experiment. Source data are provided in the Source Data file.
Fig. 3 Study of confocal height on Raman signal intensity.
SERS mapping of FGF-2 SERS nanotags on the silicon substrate with changing confocal height of a 0 nm, b 500 nm, c 1000 nm, and d 1500 nm; selected Raman spectra obtained from e red circles and f blue circles of SERS images. Red dotted lines in e and f indicate peak signal at 1330 cm−1 from DTNB. Data from one independent experiment. Source data are provided in the Source Data file.
Fig. 4 Specificity of digital nanopillar SERS platform for FGF-2 cytokine detection.
Representative confocal SERS images in the presence of a target FGF-2 (1031 aM), and negative controls with non-target controls b G-CSF (1031 aM), c GM-CSF (1031 aM), d CX3CL1 (1031 aM), and e PBS. The median (interquartile range) of active pillars per scanning image for FGF-2, G-CSF, GM-CSF, CX3CL1, and PBS was 72 (63.5–76.75), 1.5 (1.5–2), 2 (1–4), 0.5 (0–1.25), and 1 (1–1.75), respectively. Data from one independent experiment.
Optimisation of digital nanopillar SERS platform for cytokine detection
The reliable detection of single cytokine molecules by the digital nanopillar SERS platform depends on the geometric features of the pillar array (i.e., pillar height, cross-section area of pillar) and assay conditions (i.e., incubation time for sample and SERS nanotags).
We first sought to investigate the effect of pillar height on Raman signal intensity to differentiate signals from non-specifically bound SERS nanotags on the silicon substrate and specifically bound SERS nanotags on the gold-topped pillars. FGF-2 SERS nanotags were randomly deposited on the silicon substrate to mimic the non-specific binding scenario and the Raman mappings were perfomed by moving the objective along the z-axis direction with different heights (0 nm, 500 nm, 1000 nm, and 1500 nm) to compare the signal intensity. In Fig. 3, a–d show the false-colour SERS images and e, f the corresponding SERS spectra with characteristic DTNB reporter peak at 1330 cm−1 acquired from the circled spots in a–d. At the height of 0 nm where the SERS nanotags were in focus, we noticed bright Raman spots (Fig. 3a) and strong Raman signals (black line in Fig. 3e, f). With increasing z heights to 500 nm and 1000 nm, the Raman spots decreased (Fig. 3b, c) and the signal intensity weakened/disappeared (red and blue lines in Fig. 3e, f) as the nanoboxes became increasingly out of focus. A further increase to 1500 nm did not remarkably weaken Raman signals compared to the height of 1000 nm (Fig. 3d). Hence, for the fabrication of the pillar array chip, we selected a pillar height of 1000 nm to greatly reduce the potential interference from non-specific signals.
The cross-section area of the pillars provides the space for cytokine binding and labelling with SERS nanotags. To study the effect of pillar cross-section area, we fabricated array chips with pillars of various widths (250 nm, 500 nm, and 1000 nm) (Supplementary Fig. 7a–c) and functionalised with anti-FGF-2 antibody. Each chip consisted of 250,000 individual pillars. We then analysed a sample that contained ~25,000 molecules of FGF-2 (40 µL, 1031 aM), which should result in 10% active pillars (ratio FGF-2: pillars of 0.1). As seen in Supplementary Fig. 7d–f, an increasing pillar cross-section area results in a higher fraction of active pillars. In reference to the expected active pillar percentage (10%), the 250 nm and 500 nm pillar arrays produced lower active pillars (2% and 6%), which suggested a significant loss of target recognition by SERS nanotags. For the 1000 nm wide pillars, the active pillar percentage was 11%, close to the nominal value of 10%. We further tested a sample with 260 aM FGF-2 (i.e., 2.5% active pillars) on the pillar array chips with 250, 500, and 1000 nm pillar widths. The capture efficiency of these three chips was summarised in Supplementary Table 2. In comparison to the pillar array of 250 nm and 500 nm sizes, the 1000 nm provided an improved capture efficiency. As the accessible target recognition surface area per pillar increases, it can possibly promote the thermodynamics and kinetics for higher surface binding and capture efficiency25. Consequently, the 1000 nm pillar array was adopted in the subsequent experiments.
An optimal incubation time of cytokine and SERS nanotags on the pillar array can shorten the operation time and reduce the potential risk of nonspecific binding that could lead to false-positive counting. We thus studied the effect of incubation time of cytokine with SERS nanotags for 30 to 90 min in a solution of 1031 aM FGF-2 and FGF-2 SERS nanotags. As suggested by Supplementary Fig. 8, the increase in incubation time gives rise to a higher proportion of active pillars. In comparison with the theoretical active pillar percentage (10%), both 30 min and 60 min incubation time were able to provide a desirable active pillar percentage (11% and 13%, respectively). A longer incubation time (90 min), however, reported an active pillar percentage (20%) two times higher than the theoretical, indicating the occurrence of nonspecific absorption of SERS nanotags on the pillar array chip. Thus, we selected 30 min incubation time for further digital nanopillar SERS measurements.
Specificity of the digital nanopillar SERS platform for cytokine detection
Accurate and reliable recognition of the specific target is essential for cytokine quantification in clinical samples. To demonstrate the detection specificity of the digital nanopillar SERS assay, we prepared an anti-FGF-2 antibody functionalised pillar array and measured samples containing target FGF-2 cytokine and controls (G-CSF, GM-CSF, CX3CL1, and PBS). It was observed that only the presence of FGF-2 activated significant amounts of pillars whereas the negative controls only generated negligible active pillars (Fig. 4), indicating the high specificity for FGF-2 detection. Similarly, we studied the specific detection of G-CSF, GM-CSF, and CX3CL1, as shown in Supplementary Figs. 9–11, in which the typical Raman images displayed high proportions of active pillars in the presence of specific targets but not for the negative controls.
To further investigate the specificity of binding between SERS nanotag and antibody-functionalised pillar, we performed SEM analysis to “closely” inspect the pillar array for the presence or absence of SERS nanotags. As a representative model, we selected to image FGF-2 SERS nanotags on the anti-FGF-2-functionalised pillar array after sample incubation with FGF-2 cytokine and non-target controls (Fig. 5). As expected, we observed the cubic nanoparticles on the top of pillars in the presence of FGF-2 due to the successful recognition of SERS nanotags. On the contrary, pillar arrays did not display a significant number of FGF-2 SERS nanotags with non-target controls. Consequently, the consistent Raman and SEM data demonstrate the capability of the assay for specific target cytokine counting. The ability to selectively identify these four cytokines in the designed assay is critically important for their usage in clinical samples.Fig. 5 Specificity of the digital nanopillar SERS platform for FGF-2 cytokine detection.
Representative SEM images of pillar array incubated with FGF-2 SERS nanotags in the presence of a, b FGF-2 (1031 aM), c G-CSF (1031 aM), d GM-CSF (1031 aM), e CX3CL1 (1031 aM), and f PBS. The red circles highlight the existence of SERS nanotags. Panel b is the magnified SEM image of the red-highlighted section in a. It is noted that nanofabrication debris on the sidewall of the pillars can also be seen. Data from one independent experiment.
Sensitivity of the digital nanopillar SERS platform for cytokine detection
As there is typically low abundance of cytokines in clinical samples, the technique based on cytokine detection is expected to possess sufficient sensitivity to reliably assess irAEs9. To investigate the sensitivity and dynamic detection range of the digital nanopillar SERS assay, we firstly titrated the designated concentration of one target cytokine (FGF-2) on the pillar array chip with 250,000 pillars. To comply with the Poisson distribution, the upper number of cytokine molecules in the sample is 25,000 which should result in 10% activated pillars. Based on this upper molecule number, we were motivated to challenge the assay by serially diluting the number of cytokine molecule in the sample from 25,000 (1031 aM), 6305 (260 aM), 631 (26 aM), and 63 (2.6 aM). As suggested by the Raman images in Supplementary Fig. 12 and with the decrease in FGF-2 molecules, the percentage of active pillars decreased correspondingly from 9.39% for 1031 aM, 6.59% for 260 aM, 1.12% for 26 aM, and 0.62% for 2.6 aM, showing a strong correlation that facilitates quantitative cytokine analysis.
Subsequently, we were interested in exploring the multiplexing capability of SERS to investigate the digital nanopillar SERS assay’s dynamic range for the simultaneous quantification of all studied cytokines. As the targets independently follow Poisson distribution, each of the cytokine was separately controlled to activate less than 10% pillars. The specific SERS nanotags provided unique signals for each cytokine that was visualised in the false-colour SERS images by a different colour, thereby enabling in situ and simultaneous cytokine detection. As suggested by the confocal SERS images in Fig. 6, an increase in cytokine concentration corresponded with a higher percentage of active pillars. To facilitate quantitative measurements of the cytokines, we calculated the logarithmic transformation of the percentage of active pillars versus cytokine concentration (Supplementary Fig. 13) confirming the strong statistical and potentially clinically relevant correlation (coefficient of determination (R2) >0.97) observed in the SERS images.Fig. 6 Sensitivity for the simultaneous detection of four cytokines.
Representative confocal SERS images of fibroblast growth factor 2 (FGF-2), granulocyte colony-stimulating factor (GM-CSF), granulocyte colony-stimulating factor (G-CSF), and fractalkine (CX3CL1) with the concentration of a 2.6 aM, b 26 aM, c 260 aM, d 1031 aM. Colour scale bars indicate Raman intensities from 5,5-dithiobis (2-nitrobenzoic acid) (DTNB), 4-mercaptobenzoic acid (MBA), 2,3,5,6-tetrafluoro-4-mercaptobenzoic acid (TFMBA), or 2‐mercapto‐4‐methyl‐5‐thiazoleacetic acid (MMTAA). The median (interquartile range) of active pillars per scanning image of FGF-2, G-CSF, GM-CSF, CX3CL1 for 2.6 aM: 3 (1.5–3), 1 (1–2), 2 (1–3), 2 (1–3); 26 aM: 8 (5.5–10), 10 (9–13), 7 (6–10), 8 (6–10); 260 aM: 40 (36–48), 40 (35–52), 39 (35–50), 37 (36–49); and 1031 aM: 79 (61.5–97), 78 (72–87.5), 88 (68.5–97), 79 (64–95), respectively. Data represents one experiment from three independent tests.
To further investigate the multiplexing quantification performance of the digital nanopillar SERS assay in human serum, we spiked standard cytokines in human serum and tested the dynamic range. Supplementary Fig. 14 shows the linear relationship curves for the four targets. Because of the more complicated sample matrix composition in human samples, the lowest detectable cytokine concentration (5.2 aM) was higher than the PBS solution (2.6 aM).
At a cytokine to pillar ratio of 1:10, we studied the probability of each pillar being occupied by different molecule numbers. To experimentally investigate the number of molecules on a single pillar, we analysed a cytokine mixture that contained all four target cytokines at equal concentration (i.e., ~6250 molecules per cytokine). To visualise and count molecule binding events on a single pillar, we labelled the captured cytokines with the four SERS nanotags that provide clearly distinguishable signals. Under Poisson distribution, the likelihood of having two or more molecules on a single pillar is <0.45% (Supplementary Table 3), which underlies the digital counting principle24. Compared to the theoretical Poisson distribution, the experiment data reported a close but slightly higher value, which was probably due to minor non-specific binding of SERS nanotags on the pillars.
The high sensitivity (attomolar level) of the digital nanopillar SERS assay can be ascribed to the following factors: the digital counting strategy, the single-particle SERS activity of the nanoboxes, and the use of pillars to suit confocal Raman mapping that efficiently excludes false-positive signals. Commercially available methods with potential for trace analysis of cytokines include the single-molecule ELISA Simona by Quanterix and electrochemical luminescence assay42,43 by Meso Scale Discovery. Compared to these two methods, the developed digital nanopillar SERS platform enabled in situ multiplexed detection of four cytokines with comparable sensitivity. Unlike the issues of photo bleaching and poor multiplexing analysis often encountered in fluorescence44 and luminescence assays45, SERS provides the advantage of high multiplexing (e.g., 31-plex)46–48 with the narrow Raman linewidth and high photo stability of the Raman reporters. In addition, this digital nanopillar SERS platform can provide more accurate quantification of cytokines by reducing the false-positive signals with the confocal setting, thus eventually help clinicians to monitor irAEs during immune checkpoint therapy. The highly sensitive readout for multiple targets also indicated the capability of this assay for cytokine detection to assess irAEs in clinically relevant samples.
Evaluation of digital nanopillar SERS platform on simulated patient samples
The detection of trace concentrations of cytokines in serum samples is difficult because plasma samples contain a high abundance of non-target molecules (e.g., serum albumin and other proteins) that can potentially interfere with cytokine detection and lead to inaccurate clinical results. To evaluate the capability of the digital nanopillar SERS assay in accurately counting single cytokine molecules, we opted to perform a recovery test in simulated patient plasma samples (i.e., healthy human serum spiked with 1 fM of FGF-2, G-CSF, GM-CSF, and CX3CL1). The rapid scan rate (i.e., 0.05 s for Raman signal integration) facilitated the detection of Raman signals from FGF-2, G-CSF, GM-CSF, and CX3CL1 SERS nanotags rather than the non-target molecules present in human serum due to their low Raman cross-section. As a representative example, Supplementary Fig. 15 shows the Raman signal distribution of the FGF-2 SERS nanotags on five different spots on the pillar array obtained from the recovery test without noticeable Raman signals from other molecules. It is worth noting that unlike the solution-based DTNB labelled SERS nanotag spectra in Fig. 1d, some of the peaks at 1556 cm−1 and 1330 cm−1 in Supplementary Fig. 15 had a similar intensity, which was probably because of the different orientation of the anisotropic nanoboxes on the substrate relative to the polarisation of excitation laser49. Supplementary Tables 4 and 5 show the cytokine concentrations in healthy human serum and human serum spiked with 1 fM cytokine standards determined by digital nanopillar SERS platform, respectively. On five independent pillar arrays, the measured concentrations had the relative standard deviation (RSD) below 9.0% and the Kruskal–Wallis test showed no statistical differences among these results (p » 0.05). Overall, the observed inter-chip variation should enable accurate identification of disease progression to severe irAEs (e.g., grade 3 or 4), but may encounter some challenges in discriminating mild progressing to moderate irAEs (e.g., grade 1 or 2). The assay enabled trace determination of the four targets in simulated human serum as suggested by the target recovery rates of 80.00% to 137.00% with RSD from 16.02% to 21.80% (Supplementary Table 6). Importantly, the ability to measure reliably cytokines at attomolar levels in simulated human serum samples holds promise for detecting early changes in cytokine concentrations as predictors for the emergence of irAEs in immune checkpoint blockade treated patients.
To validate the accuracy of the digital nanopillar SERS assay, we compared the assay with commercially available ELISA kits (one kit for each cytokine tested) for cytokine quantification. To represent a potential clinical scenario of a patient developing irAEs during immune checkpoint blockade therapy, we prepared three samples with increasing concentrations of cytokines (spike in experiments into fetal bovine serum (FBS)) and subsequently analysed these samples with our digital nanopillar SERS assay and the commercial ELISA kits. FBS was used as complex sample matrix devoid of human cytokines. As the limits of detection for the ELISA kits (FGF-2 = 0.95 pM, G-CSF = 1.66 pM, GM-CSF = 1.11 pM, and CX3CL1 = 17.86 pM) were above the attomolar level, the simulated samples were prepared to suit the detection range of these kits. For the digital nanopillar SERS assay, the samples were diluted correspondingly and generated consistent results with the ELISA kits as shown in Supplementary Table 7. No statistical differences were found between ELISA and digital nanopillar SERS results based on Mann–Whitney test. Furthermore, we compared the detection of four cytokines in human serum with digital nanopillar assay and ELISA kits (Supplementary Table 8). The cytokine levels in human serum were below the limit of detection for the conventional ELISA kits, whereas their concentration was quantified by digital nanopillar SERS platform. For the human serum spiked with standard cytokines, the digital SERS platform generated similar results to ELISA without significant differences by Mann–Whitney test. Collectively, the digital nanopillar SERS platform showcased the ability to robustly and accurately quantify cytokines in complicated samples, which is significant for the prospect of dynamic correlation monitoring of irAEs in clinical samples.
Following the demonstration of the accuracy of digital nanopillar SERS platform, we tested the four cytokine levels in ten healthy people (Supplementary Table 9). These ten healthy people showed cytokine concentrations beyond the conventional ELISA capability to accurately quantify, which was consistent with previous reports9,50,51.
Dynamic correlation monitoring of irAEs in melanoma patients receiving immune checkpoint blockade treatment
Having established the feasibility of digital nanopillar SERS in simulated clinical samples, we applied the platform for longitudinally monitoring irAEs in ten melanoma patients (2–3 time points per patient, 26 samples in total) who underwent immune checkpoint blockade therapy (Supplementary Table 10). By diluting the patient samples to follow Poisson distribution, we quantified the cytokine concentration using digital nanopillar SERS platform. Based on the clinical assessments, the patients were classified into two categories: (i) developed severe irAEs (grades 3 and 4) and needed hospitalisation and dedicated treatment (Patients 1, 2, 3, 4, 5); and (ii) developed minor irAEs (grades 1 and 2) that could be managed with immunosuppressants (e.g., corticosteroids) or exhibited no symptom of irAEs (Patients 6, 7, 8, 9, 10).
As a representative case, Fig. 7 shows two cytokine profiles of a patient with severe irAEs (Patient 1) and a patient with mild irAEs (Patient 1). For Patient 1 who received ipilimumab (cytotoxic T-lymphocyte antigen-4 (CTLA-4) inhibitor) and was checked on days 7, 21, and 42, the confocal SERS images showed an increase in active pillars with the continuation of treatment (Fig. 7a–c), suggesting an elevation of the cytokine levels that could potentially trigger the severe irAEs. In agreement with the Raman images, the quantitative counting results for the four cytokines also corroborated the increase of cytokine concentrations peaking in sub-fM levels (Fig. 7d). These cytokine levels were below the limit of detection of conventional ELISA kits (pM level). Importantly, we observed significantly elevated cytokine concentrations in Patient 1 serum on day 42 compared to days 7 and 21. This patient showed the onset of grade 4 irAEs (i.e., colitis) 13 days later (day 55), consistent with the concept that higher cytokine levels correlate with increased risk of developing irAEs9. To further evaluate the utility of these four biomarkers as a signature in identifying and characterising irAEs, we analysed all the counting data from Patient 1 by applying linear discriminant analysis (LDA). As seen in Fig. 7e, the LDA successfully distinguished the data on day 42 into a separate zone from days 7 and 21, which may indicate the potential value of biomarkers in monitoring irAEs development. We further demonstrated Patient 1 LDA with the use of all combinations of two (Supplementary Fig. 16) and three cytokines (Supplementary Fig. 17). Overall, the LDA with four cytokines showed improved classification over using three or less cytokines. Interestingly, considering FGF-2/G-CSF, G-CSF/GM-CSF, and G-CSF/CX3CL1, the LDA generated similar performance to the LDA with four cytokines. To further compare the classification power of FGF-2/G-CSF, G-CSF/GM-CSF, and G-CSF/CX3CL1, and four cytokines, we performed LDA of Patient 2 (Supplementary Fig. 18), which suggested a better differentiation with the use of four cytokines. Therefore, the inclusion of all four cytokines in LDA facilitated a wider and more accurate patient sample analysis. Similarly, Patients 2, 3, 4, and 5, who manifested severe irAEs were connected with higher cytokine levels (Supplementary Fig. 19) and amelioration of irAEs symptom was witnessed with a decrease of cytokine concentrations. For these severe irAEs patients, the LDA model showed a clear discrimination in cytokine profile and this could help to identify patients at risk of irAEs (Supplementary Fig. 19).Fig. 7 Digital nanopillar SERS assay for monitoring melanoma patients during immune checkpoint therapy.
For Patient 1 who developed severe irAEs, SERS images for cytokine detection on a day 7, b day 21, c day 42, d cytokine concentration graph for fibroblast growth factor 2 (FGF-2), granulocyte colony-stimulating factor (G-CSF), granulocyte-macrophage colony-stimulating factor (GM-CSF), and fractalkine (CX3CL1). The two shorter horizontal lines denote the interquartile ranges (25th and 75th percentile) and the longer horizontal lines in between denote the median (50th percentile), and e LDA analysis, respectively. For Patient 6 who developed mild irAEs, SERS images for cytokine detection on f day 0, g day 21, h day 42, i four cytokine concentration graph, the two shorter horizontal lines denote the interquartile ranges (25th and 75th percentile) and the longer horizontal lines in between denote the median (50th percentile), and j LDA analysis, respectively. IPI ipilimumab, PEMBRO pembrolizumab; G3 grade 3, G2 grade 2; SD stable disease, PR partial response. For Patient 1, the median (interquartile range) of active pillars per scanning image of FGF-2, G-CSF, GM-CSF, CX3CL1 on day 7: 14 (11–22.5), 23 (21, 29), 12 (7.5–18), 17 (9–25.5); day 21: 30 (19–37.5), 33 (19–41), 26 (17.5–36.5), 29 (21–43); and day 42: 33 (16.5–58.5), 76 (64–128.5), 25 (14–39.5), 48 (26.5–73.5), respectively. For Patient 6, the median (interquartile range) of active pillars per scanning image of FGF-2, G-CSF, GM-CSF, CX3CL1 on day 0: 18 (16–23), 49 (31.5–56), 23 (17.5–28), 20 (14.5–27); day 21: 29 (24–33.5), 53 (46.5–70), 35 (25–46), 22 (19–29.5); and day 42: 13 (8–16.5), 44 (23.5–55.5), 10 (6.5–12.5), 30 (24–34.5), respectively. The data represented three technical replicates obtained from three chips. Nine images were acquired from each chip for cytokine counting. Statistical analysis was based on Kruskal–Wallis test followed by Dunn’s test to correct multiple comparisons (two-sided). Source data are provided in the Source Data file.
As for Patient 6 who exhibited mild grade 2 irAEs on the skin during combined ipilimumab and pembrolizumab (programmed death-1 (PD-1) inhibitor) or single ipilimumab treatments, the dynamic monitoring displayed relatively stable cytokine levels on different follow-up visits (Fig. 7). Specifically, the confocal Raman images (Fig. 7f–h) and the molecular counting (Fig. 7i) in this patient serum consistently showed no significant cytokine level alterations on the three time points (days 0, 21, 42). Under this circumstance, LDA failed to clearly classify the data into separate sections (Fig. 7j). Likewise, Patients 7, 9, and 10 possessed stable cytokine levels and were diagnosed with low grade irAEs. Meanwhile, Patient 8 who showed decreasing cytokine levels did not display signs of irAEs. LDA was not able to classify Patients 7 and 8 who had mild irAEs and did not show irAEs, but it recognised the minor difference in Patients 9 and 10 who showed grade 1 irAEs (Supplementary Fig. 20).
Overall, we found the preliminary evidence to suggest that significantly elevated cytokine levels have a strong correlation with the development and manifestation of severe irAEs, whereas stabile, low baseline, and decreasing cytokine concentration indicate mild and manageable irAEs. The relatively low concentrations of these four cytokines were below the detection sensitivities of commercially available ELISA kits (pM level), which limits their use in clinical studies. Importantly, the measurement at fM cytokine levels in clinical samples is consistent with the median concentrations of cytokine measured by using digital ELISA51. The successful demonstration of the digital nanopillar SERS platform in dynamic detection of cytokines in patient serum provides a potential approach for the future accurate early detection, characterisation, and monitoring of irAEs in clinical settings. However, it is important to note that the cytokine concentration changes are not directly correlated with the treatment response according to response evaluation criteria in solid tumours (RECIST). In our pilot study, some melanoma patients showed higher levels of cytokines compared to the healthy controls. The power of our digital nanopillar SERS platform lies in the capability to longitudinally monitor cytokines in individual patients over time.
Discussion
Despite the frequent occurrence of irAEs in immune checkpoint therapy, particularly for the combination treatment, the prediction of the emergence of irAEs remains elusive. Mounting data suggest a potential role of cytokines as predictive markers for irAE monitoring in immune checkpoint therapy9,11,13,52. Although promising, accurate quantification of these biomarkers was often not possible due to the dearth in technologies with sufficient detection sensitivity. Typically, either cytokines above the detection limit of immunosorbent assay were selected11, or relative cytokine quantification9 was performed for investigating irAEs. The former approach has the drawback of potentially excluding the low abundance cytokines of significance in irAEs. As for relative quantification9,53, the cytokine concentrations are determined by relating to a standard that had the observed median fluorescence value closest to the median of the test sample. The relative concentration, however, may fail to represent accurate cytokine levels and thus needs further exploration. Our developed digital nanopillar SERS assay offers early data suggestive of a potential approach to the above-mentioned challenges and provides the possibility to study trace amounts of a panel of cytokines in an accurate quantification manner as well as concomitantly providing attomolar level sensitivity. The current proof-of-principle approach has measured four of the potential inflammatory and/or immune toxicity-related cytokines for the prediction of emergence, characterisation, and/or quantifiable correlation with irAEs in melanoma patients.
By leveraging the narrow line width of Raman spectra, the developed digital SERS counting assay shows the ability to sensitively and simultaneously detect multiple cytokines. The adoption of the novel digital quantification mode in SERS using gold–silver alloy nanoboxes further improves the high sensitivity of SERS technology. Notably, the digital counting strategy offers an option for reproducible SERS quantification by avoiding the common Raman signal fluctuations induced by ensemble measurements. The ensemble measurement in SERS relies on the enhancement of Raman signals of molecules located in or near the “hot spots” (i.e., strong electromagnetic fields)30. Due to the random distribution and various efficiencies of “hot spots”, it can result in the discrepancies in acquired SERS intensities for inter-laboratory and even intra-laboratory tests29. To circumvent the impact of Raman intensity fluctuation on accurate quantification, we employed the digital SERS signals from the single-SERS-active nanoboxes on discrete pillar arrays to enumerate the targets and only count the “yes” or “no” signal for a robust and reproducible SERS analysis. Furthermore, the digital readout model, which regards both aggregated and single nanoparticle as a single binding event to reflect the true target number, can have a better accuracy and robustness than the intensity-based assay39.
We believe that the proposed digital nanopillar SERS assay could be used to monitor other cytokine-induced immune responses. For instance, with the outbreak of 2019 novel coronavirus (2019-nCoV), it is yet difficult to predict which infected patient will develop a strong immune response that requires hospitalisation. However, cytokines have been indicated to play a major role in the severity of immune response for critically ill patients infected with 2019-nCoV54. The specific detection of multiple cytokines at early stages of viral infection could thus potentially address this issue and help to provide the clinical care for people at the highest risk. For patients with high cytokine concentrations, the digital nanopillar SERS platform will require the sample dilution to suit Poisson distribution.
In summary, we propose a digital nanopillar SERS platform for the parallel counting of single cytokines and dynamic monitoring in the clinical context of irAE development during immune checkpoint blockade therapy. The platform achieved attomolar level sensitivity by utilising discrete pillar array compartments to hold the single cytokine and subsequently applied single-particle active nanobox-based SERS nanotags for cytokine identification and counting. The confocal Raman mapping on the pillar array offered the highest possible clinical specificity by reducing nonspecific signals and provided a “yes/no” type counting approach for reproducible Raman signal readout. The designed platform was rigorously optimised and tested in simulated clinical samples prior to the evaluation for irAE monitoring in stage IV melanoma patients receiving immune checkpoint blockade therapy. We envisaged this platform possessing the advantages of highly sensitive and multiplexing analysis capability can transit into future irAE detection methodologies after extensive validation in a large cohort of clinical samples over different time courses.
Methods
Materials
Silver nitrate (AgNO3), hydrogen tetrachloroaurate (III) trihydrate (HAuCl4·3H2O), MBA, DTNB, TFMBA, MMTAA, DSP were obtained from Sigma Aldrich. Ascorbic acid (AA) of analytical grade was purchased from MP Biomedicals. FGF-2 (223-FB), G-CSF (214-CS), GM-CSF (215-GM), CX3CL1 (365-FR) cytokines; monoclonal anti-FGF-2 (MAB233), anti-G-CSF (MAB214), anti-GM-CSF (MAB615), anti-CX3CL1 (MAB3652) antibodies; polyclonal anti-FGF-2 (AF-233), anti-G-CSF (AF-214), anti-GM-CSF (AF-215), anti-CX3CL1 (AF-365) antibodies; and FGF-2 (DY233-05), G-CSF (DY214-05), GM-CSF (DY215-05), and CX3CL1 (DY365) ELISA kits were bought from R&D Systems.
All the patient serum or plasma samples were collected at the Austin Hospital (Melbourne) under approved human ethic protocols and written informed consents were obtained from all patients before sample collection. Ethics approval was obtained from The University of Queensland Institutional Human Research Ethics Committee (approval nos. 2011001315 and 2016000876) and the following clinical assay was carried out according to the approved guidelines.
Preparation of single-particle active SERS nanotags
The preparation of SERS nanotags involved the synthesis of nanoboxes and the subsequent functionalisation with Raman reporters and antibodies. For nanobox synthesis, 45 µL of HAuCl4 (1 wt%) was added into 10 mL of ultrapure H2O (18.2 Ω cm) under magnetic stirring (800 r.p.m.) for 1 min, followed by simultaneously introducing 170 µL of AgNO3 (6 mM) and 30 µL of AA (0.1 M) into the stirring solution. Then, the formation of nanoboxes was indicated by the appearance of an apparent blue colour within 6 s and the samples were collected 1 min later by centrifuging at 600 g for 15 min.
To functionalise nanoboxes with Raman reporters and antibodies, 300 µL of nanoboxes centrifuged from 1 mL of as-prepared solution were co-incubated with one type of Raman reporters (i.e., 10 µL of DTNB, 8 µL of MBA, 10 µL of TFMBA, or 10 µL of MMTAA) and 2 µL of DSP linker for 6 h. After that, the Raman reporter and DSP functionalised nanoboxes were separated by centrifuging at 600 g for 15 min, and resuspended into 300 µL of PBS (0.1 mM). Then, 2 µg of anti-FGF-2, anti-G-CSF, anti-GM-CSF, anti-CX3CL1 antibodies were added to MBA, DTNB, TFMBA, and MMTAA labelled nanoboxes, respectively. After overnight incubation at 4 oC, the functionalised nanoboxes were purified by centrifuging at 600×g for 15 min to separate free antibodies and the final products were resuspended into 200 µL of 0.1% BSA for future use.
Fabrication of pillar arrays
The chip was made of a sensing array measuring 1 mm × 1 mm (Supplementary Fig. 1a) and consisted of 250,000 individual pillars. Each pillar was 1 µm wide, 1 µm long, and 1 µm high. The pillars were evenly spaced by 1 µm from one pillar to the next (Supplementary Fig. 1b). The pillar array was designed using Nanosuite 6.0 (Raith GmbH) and Beamer 5.9.1 (GenISys GmbH) and fabricated on a 4-inch p-type <100> silicon wafer (Bonda Technology Pte Ltd, Singapore) using electron beam lithography (EBL). The wafer accommodated 76 separate pillar arrays. Silicon wafer was first cleaned in acetone, isopropanol with sonication for 2 min each, followed by rising with deionised H2O and dehydration bake at 180 °C for 2 min. Prior to the resist coating, the wafer had undergone a further O2 plasma cleaning at 200 W for 5 min (Diener Atto, Diener Electronic GmbH). The cleaned wafer was spin-coated with two layers of polymethyl methacrylate (bottom: 495K A4 PMMA, top: 950k A4 PMMA, from MicroChemicals GmbH) using the CEE Apogee Coater (Cost Effective Equipment, LLC) at 1500 r.p.m. for 60 s each. After the coating of each layer, the wafer was baked immediately on a hot plate to prevent from intermixing of the two layers of resist. The baking time was 10 and 3 min for the bottom and top layer, respectively, at 180 °C. The thickness of the photoresist was found to be ~450 nm (top PMMA: ~250 nm, bottom PMMA: ~200 nm), characterised by white light reflectometry (FilmTek 2000M, Scientific Computing International). EBL was performed in the Raith EBPG5150 system. The patterns were exposed in EBL with an accelerated voltage of 100 kV, 150 nA of beam current (spot size ~80 nm), with step sizes of 40 nm and an electron dose of 1200 µC/cm2. The exposure time per 4-inch wafer was ~35 min, each containing 76 individual chips. After exposure, the wafer was developed in a mixture of isopropanol and methyl isobutyl ketone (3:1) for 60 s and rinsed immediately with isopropanol, followed by drying with N2. An oxygen plasma descum process, at 100 W, 60 s (Diener Atto, Diener Electronic GmbH) was carried out to remove resist residues prior to the deposition. Next, 10 nm titanium and 200 nm gold were deposited by physical vapour deposition using a Temescal FC-2000 electron beam evaporator (Ferrotec, U.S.A.). After overnight lift-off at room temperature in Remover PG (MicroChemicals GmbH, Germany), the excess material was washed off and the pillar array structure was revealed (Supplementary Fig. 1c). To create the pillar height (i.e., 1 µm), reactive ion etching (Oxford Instruments, UK) was applied for anisotropic etching of the silicon. Hereby, the deposited gold served as mask to protect the underlying silicon while the un-masked silicon was removed. Next, the wafer was coated with a protective layer of cured AZnLOF 2020 prior to wafer dicing into 76 individual sensing chips consisting of a single pillar array. Prior to use, the protective layer was washed off by consecutive washes with isopropanol and acetone and dried under a stream of nitrogen.
Pillar array functionalization
Antibody functionalisation of the gold-topped pillar array was conducted by crosslinking the antibodies to the gold surface using DSP. A solution of 5 mM DSP in dimethyl sulfoxide was pipetted onto the pillar array and incubated at room temperature for 2 h. After rinsing the pillar array with ethanol and PBS, a solution of 5 µg/mL anti-cytokine monoclonal antibody solution (100 fold dilution of antibody stock solution) in PBS was incubated overnight at 4 °C. Subsequently, the pillar array was rinsed with PBS and blocked using 1% bovine serum albumin in PBS for 1 h. Prior to use, the pillar array was rinsed with PBS. All PBS solutions were filtered through a sterile 0.22 µm syringe filter (Millex-GP, Merck, U.S.A.).
Digital nanopillar SERS profiling of cytokines
Cytokines (FGF-2, G-CSF, GM-CSF, and CX3CL1) with different concentrations in PBS (2.6 aM, 26 aM, 260 aM, and 1031 aM) were incubated with antibody functionalised pillar arrays at room temperature for 30 min, followed by washing the pillar array three times with washing buffer (0.1% BSA and 0.01% Tween 20 in PBS). SERS nanotags were then added into the pillar array for another 30 min incubation under room temperature to identify the targets. Finally, the pillar arrays were washed to remove the free SERS nanotags and were subject to confocal Raman microscope for quantification. For each sample, nine SERS images with each image has the dimension of 60 µm × 48 µm were taken on the pillar array to calculate the overall cytokine concentration.
Simulated clinical sample detection
For the recovery experiment, standard cytokines (FGF-2, G-CSF, GM-CSF, and CX3CL1) with the concentration of 1 fM were added into healthy human serum and then diluted ten times with PBS to quantify.
For the quantification of cytokines in FBS, three simulated clinical samples were prepared by titrating various concentrations of standard cytokines into 10% FBS: Sample 1 (FGF-2 = 3.64 pM, G-CSF = 3.19 pM, GM-CSF = 4.29 pM, and CX3CL1 = 28.57 fM); Sample 2 (FGF-2 = 7.28 pM, G-CSF = 6.38 pM, GM-CSF = 8.58 pM, and CX3CL1 = 571.4 fM); and Sample 3 (FGF-2 = 14.56 pM, G-CSF = 12.76 pM, GM-CSF = 17.16 pM, and CX3CL1 = 1142.8 fM). These three samples were then detected directly using the commercial ELISA kits or digital nanopillar SERS assay with a further dilution of 105, 2 × 105, and 4 × 105, respectively.
Spectroscopic ellipsometry
The antibody film thickness was measured by in-solution spectroscopic ellipsometry (M2000V JA Woollam Co., Inc. USA) using gold-coated substrates and flow cell (QSense® Ellipsometry, Biolin Scientific, Sweden). Measurements were performed at an angle of 65°. Data analysis was performed by CompleteEASE® software using a B-Spline data fit and Cauchy model to calculate the antibody film thickness.
MALDI-TOF MS
The antibody-functionalised nanopillar array chip was subjected to tryptic digest prior to analysis. Sequencing-grade trypsin was made to 50 ng/µL in 25 mM ammonium bicarbonate, and sprayed over the chip using a Bruker Imageprep instrument (Bruker, USA). After trypsin deposition, the chip was incubated in a humid environment at 40 °C for 3 h. Subsequently, the chip was sprayed with a matrix solution, 10 g/L α-cyano-4-hydroxycinnamic acid in 50% acetonitrile with 0.2% trifluoroacetic acid. Next, the chip was analysed with a Bruker Ultraflex III MALDI-TOF mass spectrometer (Bruker, USA) in positive linear mode using Flex Imaging 4.0 (Bruker, USA) with a pixel size of 60 µm. Data were collected from 2 k–30 k m/z, at a laser repetition rate of 200 Hz. Data were normalised using the root mean square approach and visualised using Flex Imaging 4.0 (Bruker, USA) and SCILS LAB 2017a software. For the SCILS LAB analysis, the data were imported using a convolution baseline subtraction, and displayed using root mean squared normalisation.
Instrumentations
SEM images of pillar arrays and nanoboxes were taken on a JEOL-7100 field emission (FE)-SEM (20 kV voltage). TEM images of nanoboxes were taken on a JEOL-2100 microscope (200 kV voltage). NTA of nanobox size distribution was performed with Malvern NanoSight NS300. UV-vis extinction spectrum of nanoboxes was performed with a Shimadzu UV-2450 spectrophotometer. Confocal Raman mapping was conducted on a WITec alpha 300 R spectrometer using 632.8 nm He–Ne laser with the power of 35 mW, a grating of 600 g/mm used with EMCCD camera, spectral resolution of 1.390 cm−1 to 2.114 cm−1, confocal pinhole size of 100 µm, 100× air objective with NA of 0.90, and 0.05 s integration time. The theoretical spot size was 857.80 nm based on the Abbe diffraction limit (i.e., d = 1.22λ/NA). The scanning area was set to have 60 µm × 48 µm with 86 points per line and 69 lines per image. For each pillar array, nine separate scanning areas were taken in total and the total active pillars were used for quantification. The SERS mapping images for counting were taken by focusing the laser on the top of the pillar surfaces. Specifically, the laser was firstly focused on the silicon substrates by obtaining the strongest silicon signals (520 cm−1) and then the 100× objective was moved up in z-axis direction of 1 µm for SERS scanning. The system was calibrated with the first-order photo peak of silicon at 520 cm−1.
Data analysis
To assign the SERS nanotag membership for DTNB, MBA, TFMBA, and MMTAA, Project Five 5.0 software from WITec was utilised to create four filters, which summed a spectral range of 40 cm−1 with the centre position at the characteristic Raman peak of each reporter and subtracted the background with a polynomial algorithm. Specifically, the filter ranges of four Raman reporters DTNB-, MBA-, TFMBA-, and MMTAA-coated SERS nanotags were (1310–1350 cm−1), (1060–1100 cm−1), (1360–1400 cm−1), and (1268–1308 cm−1), respectively. All the SERS images were analysed by using threshold intensity to determine the successful binding events. Specifically, the threshold intensity of FGF-2, G-CSF, GM-CSF, and CX3CL1 was set at 5000, 4000, 5000, and 5000, respectively. For each image, the threshold intensity was doubled-checked and adjusted based on the true Raman peaks in the spectra. Statistical analysis assuming unequal variances was conducted with Kruskal–Wallis test among three groups or Mann–Whitney test between two groups with GraphPad Prism 8.4. To control the error appropriately, we performed multiple comparisons using Dunn’s test. LDA of clinical samples was performed in R software (3.6.2) with the MASS package (7.3-52). The active pillars in SERS images were counted with Image J software.
Supplementary information
Supplementary Information
Peer Review File
Source data
Source Data
Peer review information
Nature Communications thanks Chongwen Wang and Isaac Pence for their contribution to the peer review of this work. Peer reviewer reports are available.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
The online version contains supplementary material available at 10.1038/s41467-021-21431-w.
Acknowledgements
The authors acknowledge grants received by our laboratory from the National Breast Cancer Foundation of Australia (CG-12-07) and the ARC DP (160102836 and 210103151). These grants have significantly contributed to the environment to stimulate the research described here. J.L. acknowledges support from the Australian Government Research Training Program Scholarship. A.W. and A.A.I.S. thank the National Health and Medical Research Council for funding (APP1173669 and APP1175047). A.B. is the recipient of a Fellowship from the Victorian Government Department of Health and Human Services acting through the Victorian Cancer Agency. We acknowledge the facilities, and the scientific and technical assistance, of the Australian Microscopy & Microanalysis Research Facility at the Centre for Microscopy and Microanalysis, The University of Queensland. We appreciate to receive the technical and scientific guidance from Queensland node of the Australian National Fabrication Facility (Q-ANFF) in confocal Raman mapping and spectroscopic ellipsometry measurement.
Author contributions
J.L., A.W., A.I.S., H-H.C., Y.W., A.B., P.M., and M.T. contributed to the design of the experiments and analysing the data. J.L. and A.W. performed the experiments and prepared the manuscript. All authors read, commented, and edited the manuscript, and assisted during the revision process.
Data availability
Data supporting the findings of this work are available within this paper and the supporting information files. A reporting summary of this work is available as a Supplementary file. Source data are provided with this paper.
Competing interests
The authors declare no competing interests. | IPILIMUMAB | DrugsGivenReaction | CC BY | 33597530 | 19,690,420 | 2021-02-17 |
What was the dosage of drug 'PEMBROLIZUMAB'? | A digital single-molecule nanopillar SERS platform for predicting and monitoring immune toxicities in immunotherapy.
The introduction of immune checkpoint inhibitors has demonstrated significant improvements in survival for subsets of cancer patients. However, they carry significant and sometimes life-threatening toxicities. Prompt prediction and monitoring of immune toxicities have the potential to maximise the benefits of immune checkpoint therapy. Herein, we develop a digital nanopillar SERS platform that achieves real-time single cytokine counting and enables dynamic tracking of immune toxicities in cancer patients receiving immune checkpoint inhibitor treatment - broader applications are anticipated in other disease indications. By analysing four prospective cytokine biomarkers that initiate inflammatory responses, the digital nanopillar SERS assay achieves both highly specific and highly sensitive cytokine detection down to attomolar level. Significantly, we report the capability of the assay to longitudinally monitor 10 melanoma patients during immune inhibitor blockade treatment. Here, we show that elevated cytokine concentrations predict for higher risk of developing severe immune toxicities in our pilot cohort of patients.
Introduction
The advent of immune checkpoint therapy has revolutionised the landscape of traditional cancer treatment and is believed to constitute the backbone of managing certain malignancies1–3. By capitalising on the blockade of immune checkpoint inhibitors to take the brakes off parts of the immune system, this emerging therapy has achieved great success producing long-lasting responses (e.g., 10 years or more) in a small but significant fraction of patients3–6. Nevertheless, upon the blockade of immune checkpoint molecules, the activated and potentiated immune reaction predisposes patients to a significant risk of immune-related adverse events (irAEs), which can occur in up to 80% of patients receiving immune checkpoint therapy7–9. The high incidence of irAEs, which may manifest at any time during treatment, can offset the clinical benefits, lead to premature therapy cessation, and even be life-threatening for certain patients10–12. To assist the successful implementation of immune checkpoint therapy, the use of predictive biomarkers for early identification and vigilant monitoring of irAEs is thus critical and a pressing need in avoiding or ameliorating detrimental effects and adjusting therapeutic options.
Cytokines, small signalling proteins, are promising candidates to indicate the occurrence of irAEs due to their prominent role in modulating the anti-cancer immune responses, including enhancing antigen priming, recruiting immune cells into the tumour microenvironment, and upregulating certain immune checkpoint molecules9,13,14. Particularly, excessive cytokine secretion has been implicated in severe inflammation as a major constituent leading to irAEs. For example, the overproduction of fibroblast growth factor 2 (FGF-2)15–18, granulocyte colony-stimulating factor (G-CSF)19, granulocyte-macrophage colony-stimulating factor (GM-CSF)20, and fractalkine (CX3CL1)21 have been found to participate in immune-related inflammatory disease (e.g., rheumatoid arthritis, autoimmune gastritis, and Crohn’s disease). These inflammatory cytokines have recently been reported to indicate irAEs for melanoma patients who underwent immune checkpoint therapy9. The clinical deployment of cytokine analysis for irAE monitoring is challenging and requires a technology that can (i) determine the selected cytokines with great sensitivity9, especially at the onset of irAE development, where the cytokine concentrations are likely to be the lowest; as well as (ii) simultaneously detect a panel of cytokines to reflect the complex interplay of cytokine signalling pathways22 and the variable irAE symptoms among patients.
Conventional cytokine analyses such as immunosorbent assays have limited clinical applicability for irAE assessment due to their limited capacity to detect low cytokine concentrations in blood as well as for assessing a panel of cytokines in a single sample simultaneously. Recently, advances in micro/nanomaterial-based systems have provided a promising suite of technologies that improve the conventional assays by overcoming the above limitations23,24. Encouragingly, the unique advantages of micro/nanomaterial-based systems convey an attractive option for cytokine analysis with the desired results of high sensitivity and multiplexing. The high specific surface area of these miniaturised materials increases mass transfer subsequently enhancing the interaction with target molecules and thus improving the detection sensitivity25. The capabilities of micro/nanomaterial fabrication techniques permit individually separated compartments sufficiently discrete to hold single molecules and hence encompasses a promising strategy for counting assays that can further push the sensitivity of the traditional assays24,26,27. Moreover, the physicochemical properties of nanostructured materials can be exploited to simultaneously label multiple targets (e.g., various spectral signatures) for high-throughput parallel measurements28–30. Therefore, by combining the potential of micro/nanomaterial systems with the need for sensitive irAE monitoring, we have developed a platform for sensitive and multiplex cytokine counting analysis.
Combining the use of (a) discrete single cytokine nanopillar array chip with discretely separated compartments, (b) control of target concentrations to follow a Poisson distribution, and (c) the recognition of target by single-particle active surface-enhanced Raman scattering (SERS) nanotags with a confocal Raman microscope allows accurate and in situ counting of a multi-cytokine panel (FGF-2, G-CSF, GM-CSF, and CX3CL1). Different from the fluorescence-based digital counting strategies24,26, the strikingly narrow spectral peaks of SERS (~1–2 nm) in comparison to fluorescence (~50 nm) makes this platform intrinsically ideal for multiplexed cytokine analysis28.
In this work, we present a digital nanopillar SERS platform that enables the specific cytokine quantification down to attomolar levels and the application in melanoma patients receiving immune checkpoint blockade therapy. Beyond the capability to predict irAEs in melanoma patients receiving therapy, the digital nanopillar SERS assay could potentially be extended to other cytokine-associated immune responses such as excessive immune activation due to viral or bacterial infections (such as COVID-19).
Results
Digital nanopillar SERS platform for parallel profiling of single cytokine
Our concept of digital nanopillar SERS platform for cytokine analysis relies on Rayleigh criterion separation, probability-driven Poisson distribution, single-particle active SERS nanotags, and confocal SERS mapping (Fig. 1). To precisely fabricate the pillar array, we opted to use an electron beam lithographic approach to write the array into a photon-sensitive material followed by physical vapour deposition of gold to create the gold-topped pillars, and selectively reactive ion etching to reveal the pillar structure (Supplementary Fig. 1). The nanopillar array chip consisted of 250,000 individual pillars. As shown in the scanning electron microscope (SEM) image of Fig. 1a, the cubic nanopillars have an edge-to-edge width of 1000 nm and are evenly distributed at 1000 nm intervals to suit the lateral Raman microscope resolution (~1000 nm) that fulfils the Rayleigh criterion separation required to acquire a single SERS spectrum from each pillar without spectral overlap from adjacent pillars.Fig. 1 Digital single-molecule nanopillar surface-enhanced Raman scattering (SERS) platform for parallel counting of four types of cytokines.
SEM images of a pillar array side view, b nanoboxes, and c a single nanobox on the top of a pillar; d SERS spectra of nanoboxes conjugated with 5,5-dithiobis (2-nitrobenzoic acid) (DTNB), 4-mercaptobenzoic acid (MBA), 2,3,5,6-tetrafluoro-4-mercaptobenzoic acid (TFMBA), or 2‐mercapto‐4‐methyl‐5‐thiazoleacetic acid (MMTAA) Raman reporters; e workflow for multiplex counting of cytokines, including fibroblast growth factor 2 (FGF-2), granulocyte colony-stimulating factor (G-CSF), granulocyte-macrophage colony-stimulating factor (GM-CSF), and fractalkine (CX3CL1). Data from one independent experiment.
By using specific gold-thiol chemistry with the linker molecule dithiobis (succinimidyl propionate) (DSP), the gold-topped pillars were selectively functionalised with target recognition antibodies (anti-FGF-2, anti-G-CSF, anti-GM-CSF, and anti-CX3CL1) and acted as the small compartments to capture and confine the individual cytokine. Upon DSP binding on the gold-topped pillars through gold-thiol bond, DSP uses N-hydroxysuccimide (NHS) ester to react with the amine groups of the antibodies31,32. The successful antibody conjugation on gold-topped pillar surfaces was confirmed by matrix-assisted laser desorption ionisation-time of flight mass spectrometry (MALDI-TOF MS) (Supplementary Fig. 2), which showed high molecular weight fragments derived from antibodies. Furthermore, spectroscopic ellipsometry was utilised to estimate the antibody density on pillar surfaces. Based on the obtained film thickness of 18.5 nm, the calculated antibody surface density was 5.5 mg/m2 using the Cuypers model33, which was in agreement with the reported antibody density on substrate surfaces34. Though these characterisations indicated the presence of antibodies on the pillar array, it was not possible to assess the exact distribution of the four structurally related (same immunoglobulin G family) antibodies on a single pillar with an area of 1 µm2. As an advantage of the digital read-out with a large redundancy of pillars, it is not essential to have all four types of antibodies equally distributed on a single pillar for the assay to work. The combined surface area of all antibody-conjugated pillars provides an excess of cytokine binding sites, which maximises successful cytokine capture within the pillar array. Supplementary Fig. 3 shows the SERS mapping images of an equimolar cytokine solution (1031 aM) that provided a similar signal count for the FGF-2, GM-CSF, G-CSF, and CX3CL1 SERS nanotags, indicating a required distribution of four kinds of antibodies conjugated to the array of pillars. Instrumental to the digital counting of cytokines, we controlled the target concentration based on the principle of Poisson distribution where the ratio of cytokine molecules to pillar number was <1:10, ensuring a 99% probability that there was either one cytokine molecule or zero per pillar24. At a ratio of 1:10, 10% of all pillars were occupied or activated with the cytokine molecules.
Following the capture of cytokines on nanopillars, SERS nanotags were applied to recognise the captured cytokines. The preparation of SERS nanotags was performed by the co-conjugating of Raman reporter and target antibody onto gold–silver alloy nanoboxes. Specifically, an average size of 80 nm gold–silver alloy nanoboxes were firstly synthesised using a rapid and aqueous phase approach35 as indicated in the SEM image in Fig. 1b. Supplementary Fig. 4a, b shows the transmission electron microscope (TEM) image of the nanoboxes with the hollow inner structure and a wall thickness of around 15 nm. Nanoparticle tracking analysis (NTA), which allows the tracking and detection of single particles, shows the nanoboxes have a mode size of 77 nm (D10 = 67.6 nm and D90 = 110.6 nm) (Supplementary Fig. 4c). UV-vis extinction spectroscopy demonstrates the nanoboxes possess a surface plasmon resonance (SPR) peak at 610 nm (Supplementary Fig. 4d). The resonance frequency of the nanoboxes enables a more sensitive signal readout with 632.8 nm laser excitation29, which also has a higher Raman scattering efficiency than 785 nm laser. Thereafter, four types of Raman reporters (5,5-dithiobis (2-nitrobenzoic acid) (DTNB), 4-mercaptobenzoic acid (MBA), 2,3,5,6-tetrafluoro-4-mercaptobenzoic acid (TFMBA), and 2‐mercapto‐4‐methyl‐5‐thiazoleacetic acid (MMTAA)) that generate unique Raman signals (1330 cm−1, 1080 cm−1, 1380 cm−1, and 1288 cm−1) were coupled with their corresponding detection antibodies onto nanoboxes as specific SERS nanotags for identification of FGF-2, G-CSF, GM-CSF, and CX3CL1, respectively. As shown in Fig. 1d, the four SERS nanotags provide the strong and non-overlapping Raman signals, which facilitates the multiplexing analysis of four cytokines. The assignment of the major Raman peaks from the four Raman reporters was summarised into Supplementary Table 1. To evaluate the SERS enhancement property of the nanoboxes, we calculated the enhancement factor (EF) of the four Raman reporters on the nanoboxes. Based on the labelled characteristic peaks in Supplementary Fig. 5, the calculated EFs of DTNB, MBA, TFMBA, and MMTAA were 8.14 × 106, 1.46 × 107, 4.01 × 107, and 3.26 × 107, respectively. The obtained EFs were higher than the reported spherical gold nanoparticles and pure silver nanocubes36 and comparable to the reported hollow nanocubes37, illustrating the high SERS property of the nanoboxes. To investigate the SERS nanotag stability, we monitored the Raman signal intensity over 7 days. As shown in Supplementary Fig. 6, Raman signal intensity variations are less than 5% in the SERS spectra, suggesting the good stability of the prepared SERS nanotags.
The following SERS mapping generated false-colour images for counting single cytokine molecules. Under the Raman microscope, the pillar array was visualised as a blue and black grid by representing the specific Raman shifts corresponding to the silicon signals (520 cm−1), in which the blue colour was assigned to silicon signals showing silicon substrates and the black colour indicated the gold-topped pillars because of the lack of silicon signals. The representation of the four colours of the SERS nanotags (red, green, purple, and cyan) on the gold-topped pillars (i.e., black) reflected cytokine molecule occupation (FGF-2, G-CSF, GM-CSF, and CX3CL1). Elevating the sensing area (or gold-topped pillars) from the silicon substrate was selected as a strategy to minimise the false-positive events. By using the confocal function of the Raman microscope, the laser was selectively focused on the gold-topped pillars, thus largely removing the background signals from potentially non-specifically adsorbed SERS nanotags on the silicon substrate. Finally, the specific SERS nanotag signals present or absent on the gold-topped pillars were counted and represented as percentage of active pillars used for total cytokine quantification. For statistical calculations, SERS mapping was applied for scanning 6480 pillars. This digital counting mode, therefore, has the potential to reach the ultimate sensitivity of single molecule cytokine detection.
Demonstration of the single-particle SERS activity of gold–silver alloy nanoboxes
The successful implementation of digital nanopillar SERS assay necessitates the use of single-particle active plasmonic nanostructures that give a clearly detectable signal for each of the single cytokine binding event. The single-particle SERS detection sensitivity is essential to this assay development as the single-particle inactive plasmonic nanoparticles (e.g., spherical gold nanoparticles)38 would unavoidably result in an underestimate of cytokine concentration.
We evaluated the single-particle SERS activity of the prepared anisotropic nanoboxes by acquiring the signals from the individual nanoboxes that were labelled with DTNB reporters. The use of DTNB, a non-resonant Raman reporter, guaranteed the Raman signal enhancement was solely contributed from the nanobox-generated electromagnetic field. As seen in the SEM image (Fig. 2a), two clearly separated DTNB-labelled nanoboxes were deposited on the silicon wafer (highlighted in red circles). The corresponding Raman image (Fig. 2b) displayed several bright Raman spots originating from these individual nanoboxes. The elongated bright Raman spots in the SERS mapping image were probably caused by the slight aggregation of several nanoboxes during sample preparation processes (e.g., centrifugation)38, which was difficult to visually resolve in the SEM image (Fig. 2a). However, unlike the intensity-based assay, the aggregated nanoboxes as SERS nanotags to target cytokine will not skew the digital readout result, because each cytokine will occupy a single pillar following Poisson distribution and both aggregated and individual nanoparticles are regarded as a single binding event that truly reflects the target number39. We then acquired the SERS spectra from two individual SERS nanoboxes (Fig. 2c, (1) and (2)) and two separate spots of bare silicon (Fig. 2c, (3) and (4)). The presence of nanoboxes showed the characteristic Raman signal at 1330 cm−1 from DTNB, whereas the silicon spectra (3) and (4) lacked the specific peak. This observation demonstrated the single-particle SERS activity of nanoboxes, which was largely attributed to the enhanced electromagnetic fields of nanoboxes on specific regions (e.g., tips and corners)40,41 and thereby facilitated the sensitive and accurate counting of cytokines. Based on the acquired SERS mapping image, the median (interquartile range) of the DTNB peak intensity (1330 cm−1) in the presence and absence of nanoboxes were 183.03 a.u. (149.48–243.35 a.u.) and 18.07 a.u. (15.51–23.12 a.u.), respectively. Furthermore, the mean ± standard deviation of the DTNB peak intensity with nanoboxes (213.41 ± 85.03 a.u.) distinguished clearly from the position without nanoboxes (18.79 ± 6.01 a.u.), which demonstrated the feasibility of correctly identifying the presence of nanoboxes.Fig. 2 Demonstration of the single-particle SERS activity of DTNB-labelled nanoboxes.
a SEM image and b corresponding SERS mapping image of DTNB-labelled nanoboxes on a silicon substrate; c representative SERS spectra of numbered locations indicated in a and b. The red dotted line shows the characteristic peak at 1330 cm−1 from DTNB. Data from one independent experiment. Source data are provided in the Source Data file.
Fig. 3 Study of confocal height on Raman signal intensity.
SERS mapping of FGF-2 SERS nanotags on the silicon substrate with changing confocal height of a 0 nm, b 500 nm, c 1000 nm, and d 1500 nm; selected Raman spectra obtained from e red circles and f blue circles of SERS images. Red dotted lines in e and f indicate peak signal at 1330 cm−1 from DTNB. Data from one independent experiment. Source data are provided in the Source Data file.
Fig. 4 Specificity of digital nanopillar SERS platform for FGF-2 cytokine detection.
Representative confocal SERS images in the presence of a target FGF-2 (1031 aM), and negative controls with non-target controls b G-CSF (1031 aM), c GM-CSF (1031 aM), d CX3CL1 (1031 aM), and e PBS. The median (interquartile range) of active pillars per scanning image for FGF-2, G-CSF, GM-CSF, CX3CL1, and PBS was 72 (63.5–76.75), 1.5 (1.5–2), 2 (1–4), 0.5 (0–1.25), and 1 (1–1.75), respectively. Data from one independent experiment.
Optimisation of digital nanopillar SERS platform for cytokine detection
The reliable detection of single cytokine molecules by the digital nanopillar SERS platform depends on the geometric features of the pillar array (i.e., pillar height, cross-section area of pillar) and assay conditions (i.e., incubation time for sample and SERS nanotags).
We first sought to investigate the effect of pillar height on Raman signal intensity to differentiate signals from non-specifically bound SERS nanotags on the silicon substrate and specifically bound SERS nanotags on the gold-topped pillars. FGF-2 SERS nanotags were randomly deposited on the silicon substrate to mimic the non-specific binding scenario and the Raman mappings were perfomed by moving the objective along the z-axis direction with different heights (0 nm, 500 nm, 1000 nm, and 1500 nm) to compare the signal intensity. In Fig. 3, a–d show the false-colour SERS images and e, f the corresponding SERS spectra with characteristic DTNB reporter peak at 1330 cm−1 acquired from the circled spots in a–d. At the height of 0 nm where the SERS nanotags were in focus, we noticed bright Raman spots (Fig. 3a) and strong Raman signals (black line in Fig. 3e, f). With increasing z heights to 500 nm and 1000 nm, the Raman spots decreased (Fig. 3b, c) and the signal intensity weakened/disappeared (red and blue lines in Fig. 3e, f) as the nanoboxes became increasingly out of focus. A further increase to 1500 nm did not remarkably weaken Raman signals compared to the height of 1000 nm (Fig. 3d). Hence, for the fabrication of the pillar array chip, we selected a pillar height of 1000 nm to greatly reduce the potential interference from non-specific signals.
The cross-section area of the pillars provides the space for cytokine binding and labelling with SERS nanotags. To study the effect of pillar cross-section area, we fabricated array chips with pillars of various widths (250 nm, 500 nm, and 1000 nm) (Supplementary Fig. 7a–c) and functionalised with anti-FGF-2 antibody. Each chip consisted of 250,000 individual pillars. We then analysed a sample that contained ~25,000 molecules of FGF-2 (40 µL, 1031 aM), which should result in 10% active pillars (ratio FGF-2: pillars of 0.1). As seen in Supplementary Fig. 7d–f, an increasing pillar cross-section area results in a higher fraction of active pillars. In reference to the expected active pillar percentage (10%), the 250 nm and 500 nm pillar arrays produced lower active pillars (2% and 6%), which suggested a significant loss of target recognition by SERS nanotags. For the 1000 nm wide pillars, the active pillar percentage was 11%, close to the nominal value of 10%. We further tested a sample with 260 aM FGF-2 (i.e., 2.5% active pillars) on the pillar array chips with 250, 500, and 1000 nm pillar widths. The capture efficiency of these three chips was summarised in Supplementary Table 2. In comparison to the pillar array of 250 nm and 500 nm sizes, the 1000 nm provided an improved capture efficiency. As the accessible target recognition surface area per pillar increases, it can possibly promote the thermodynamics and kinetics for higher surface binding and capture efficiency25. Consequently, the 1000 nm pillar array was adopted in the subsequent experiments.
An optimal incubation time of cytokine and SERS nanotags on the pillar array can shorten the operation time and reduce the potential risk of nonspecific binding that could lead to false-positive counting. We thus studied the effect of incubation time of cytokine with SERS nanotags for 30 to 90 min in a solution of 1031 aM FGF-2 and FGF-2 SERS nanotags. As suggested by Supplementary Fig. 8, the increase in incubation time gives rise to a higher proportion of active pillars. In comparison with the theoretical active pillar percentage (10%), both 30 min and 60 min incubation time were able to provide a desirable active pillar percentage (11% and 13%, respectively). A longer incubation time (90 min), however, reported an active pillar percentage (20%) two times higher than the theoretical, indicating the occurrence of nonspecific absorption of SERS nanotags on the pillar array chip. Thus, we selected 30 min incubation time for further digital nanopillar SERS measurements.
Specificity of the digital nanopillar SERS platform for cytokine detection
Accurate and reliable recognition of the specific target is essential for cytokine quantification in clinical samples. To demonstrate the detection specificity of the digital nanopillar SERS assay, we prepared an anti-FGF-2 antibody functionalised pillar array and measured samples containing target FGF-2 cytokine and controls (G-CSF, GM-CSF, CX3CL1, and PBS). It was observed that only the presence of FGF-2 activated significant amounts of pillars whereas the negative controls only generated negligible active pillars (Fig. 4), indicating the high specificity for FGF-2 detection. Similarly, we studied the specific detection of G-CSF, GM-CSF, and CX3CL1, as shown in Supplementary Figs. 9–11, in which the typical Raman images displayed high proportions of active pillars in the presence of specific targets but not for the negative controls.
To further investigate the specificity of binding between SERS nanotag and antibody-functionalised pillar, we performed SEM analysis to “closely” inspect the pillar array for the presence or absence of SERS nanotags. As a representative model, we selected to image FGF-2 SERS nanotags on the anti-FGF-2-functionalised pillar array after sample incubation with FGF-2 cytokine and non-target controls (Fig. 5). As expected, we observed the cubic nanoparticles on the top of pillars in the presence of FGF-2 due to the successful recognition of SERS nanotags. On the contrary, pillar arrays did not display a significant number of FGF-2 SERS nanotags with non-target controls. Consequently, the consistent Raman and SEM data demonstrate the capability of the assay for specific target cytokine counting. The ability to selectively identify these four cytokines in the designed assay is critically important for their usage in clinical samples.Fig. 5 Specificity of the digital nanopillar SERS platform for FGF-2 cytokine detection.
Representative SEM images of pillar array incubated with FGF-2 SERS nanotags in the presence of a, b FGF-2 (1031 aM), c G-CSF (1031 aM), d GM-CSF (1031 aM), e CX3CL1 (1031 aM), and f PBS. The red circles highlight the existence of SERS nanotags. Panel b is the magnified SEM image of the red-highlighted section in a. It is noted that nanofabrication debris on the sidewall of the pillars can also be seen. Data from one independent experiment.
Sensitivity of the digital nanopillar SERS platform for cytokine detection
As there is typically low abundance of cytokines in clinical samples, the technique based on cytokine detection is expected to possess sufficient sensitivity to reliably assess irAEs9. To investigate the sensitivity and dynamic detection range of the digital nanopillar SERS assay, we firstly titrated the designated concentration of one target cytokine (FGF-2) on the pillar array chip with 250,000 pillars. To comply with the Poisson distribution, the upper number of cytokine molecules in the sample is 25,000 which should result in 10% activated pillars. Based on this upper molecule number, we were motivated to challenge the assay by serially diluting the number of cytokine molecule in the sample from 25,000 (1031 aM), 6305 (260 aM), 631 (26 aM), and 63 (2.6 aM). As suggested by the Raman images in Supplementary Fig. 12 and with the decrease in FGF-2 molecules, the percentage of active pillars decreased correspondingly from 9.39% for 1031 aM, 6.59% for 260 aM, 1.12% for 26 aM, and 0.62% for 2.6 aM, showing a strong correlation that facilitates quantitative cytokine analysis.
Subsequently, we were interested in exploring the multiplexing capability of SERS to investigate the digital nanopillar SERS assay’s dynamic range for the simultaneous quantification of all studied cytokines. As the targets independently follow Poisson distribution, each of the cytokine was separately controlled to activate less than 10% pillars. The specific SERS nanotags provided unique signals for each cytokine that was visualised in the false-colour SERS images by a different colour, thereby enabling in situ and simultaneous cytokine detection. As suggested by the confocal SERS images in Fig. 6, an increase in cytokine concentration corresponded with a higher percentage of active pillars. To facilitate quantitative measurements of the cytokines, we calculated the logarithmic transformation of the percentage of active pillars versus cytokine concentration (Supplementary Fig. 13) confirming the strong statistical and potentially clinically relevant correlation (coefficient of determination (R2) >0.97) observed in the SERS images.Fig. 6 Sensitivity for the simultaneous detection of four cytokines.
Representative confocal SERS images of fibroblast growth factor 2 (FGF-2), granulocyte colony-stimulating factor (GM-CSF), granulocyte colony-stimulating factor (G-CSF), and fractalkine (CX3CL1) with the concentration of a 2.6 aM, b 26 aM, c 260 aM, d 1031 aM. Colour scale bars indicate Raman intensities from 5,5-dithiobis (2-nitrobenzoic acid) (DTNB), 4-mercaptobenzoic acid (MBA), 2,3,5,6-tetrafluoro-4-mercaptobenzoic acid (TFMBA), or 2‐mercapto‐4‐methyl‐5‐thiazoleacetic acid (MMTAA). The median (interquartile range) of active pillars per scanning image of FGF-2, G-CSF, GM-CSF, CX3CL1 for 2.6 aM: 3 (1.5–3), 1 (1–2), 2 (1–3), 2 (1–3); 26 aM: 8 (5.5–10), 10 (9–13), 7 (6–10), 8 (6–10); 260 aM: 40 (36–48), 40 (35–52), 39 (35–50), 37 (36–49); and 1031 aM: 79 (61.5–97), 78 (72–87.5), 88 (68.5–97), 79 (64–95), respectively. Data represents one experiment from three independent tests.
To further investigate the multiplexing quantification performance of the digital nanopillar SERS assay in human serum, we spiked standard cytokines in human serum and tested the dynamic range. Supplementary Fig. 14 shows the linear relationship curves for the four targets. Because of the more complicated sample matrix composition in human samples, the lowest detectable cytokine concentration (5.2 aM) was higher than the PBS solution (2.6 aM).
At a cytokine to pillar ratio of 1:10, we studied the probability of each pillar being occupied by different molecule numbers. To experimentally investigate the number of molecules on a single pillar, we analysed a cytokine mixture that contained all four target cytokines at equal concentration (i.e., ~6250 molecules per cytokine). To visualise and count molecule binding events on a single pillar, we labelled the captured cytokines with the four SERS nanotags that provide clearly distinguishable signals. Under Poisson distribution, the likelihood of having two or more molecules on a single pillar is <0.45% (Supplementary Table 3), which underlies the digital counting principle24. Compared to the theoretical Poisson distribution, the experiment data reported a close but slightly higher value, which was probably due to minor non-specific binding of SERS nanotags on the pillars.
The high sensitivity (attomolar level) of the digital nanopillar SERS assay can be ascribed to the following factors: the digital counting strategy, the single-particle SERS activity of the nanoboxes, and the use of pillars to suit confocal Raman mapping that efficiently excludes false-positive signals. Commercially available methods with potential for trace analysis of cytokines include the single-molecule ELISA Simona by Quanterix and electrochemical luminescence assay42,43 by Meso Scale Discovery. Compared to these two methods, the developed digital nanopillar SERS platform enabled in situ multiplexed detection of four cytokines with comparable sensitivity. Unlike the issues of photo bleaching and poor multiplexing analysis often encountered in fluorescence44 and luminescence assays45, SERS provides the advantage of high multiplexing (e.g., 31-plex)46–48 with the narrow Raman linewidth and high photo stability of the Raman reporters. In addition, this digital nanopillar SERS platform can provide more accurate quantification of cytokines by reducing the false-positive signals with the confocal setting, thus eventually help clinicians to monitor irAEs during immune checkpoint therapy. The highly sensitive readout for multiple targets also indicated the capability of this assay for cytokine detection to assess irAEs in clinically relevant samples.
Evaluation of digital nanopillar SERS platform on simulated patient samples
The detection of trace concentrations of cytokines in serum samples is difficult because plasma samples contain a high abundance of non-target molecules (e.g., serum albumin and other proteins) that can potentially interfere with cytokine detection and lead to inaccurate clinical results. To evaluate the capability of the digital nanopillar SERS assay in accurately counting single cytokine molecules, we opted to perform a recovery test in simulated patient plasma samples (i.e., healthy human serum spiked with 1 fM of FGF-2, G-CSF, GM-CSF, and CX3CL1). The rapid scan rate (i.e., 0.05 s for Raman signal integration) facilitated the detection of Raman signals from FGF-2, G-CSF, GM-CSF, and CX3CL1 SERS nanotags rather than the non-target molecules present in human serum due to their low Raman cross-section. As a representative example, Supplementary Fig. 15 shows the Raman signal distribution of the FGF-2 SERS nanotags on five different spots on the pillar array obtained from the recovery test without noticeable Raman signals from other molecules. It is worth noting that unlike the solution-based DTNB labelled SERS nanotag spectra in Fig. 1d, some of the peaks at 1556 cm−1 and 1330 cm−1 in Supplementary Fig. 15 had a similar intensity, which was probably because of the different orientation of the anisotropic nanoboxes on the substrate relative to the polarisation of excitation laser49. Supplementary Tables 4 and 5 show the cytokine concentrations in healthy human serum and human serum spiked with 1 fM cytokine standards determined by digital nanopillar SERS platform, respectively. On five independent pillar arrays, the measured concentrations had the relative standard deviation (RSD) below 9.0% and the Kruskal–Wallis test showed no statistical differences among these results (p » 0.05). Overall, the observed inter-chip variation should enable accurate identification of disease progression to severe irAEs (e.g., grade 3 or 4), but may encounter some challenges in discriminating mild progressing to moderate irAEs (e.g., grade 1 or 2). The assay enabled trace determination of the four targets in simulated human serum as suggested by the target recovery rates of 80.00% to 137.00% with RSD from 16.02% to 21.80% (Supplementary Table 6). Importantly, the ability to measure reliably cytokines at attomolar levels in simulated human serum samples holds promise for detecting early changes in cytokine concentrations as predictors for the emergence of irAEs in immune checkpoint blockade treated patients.
To validate the accuracy of the digital nanopillar SERS assay, we compared the assay with commercially available ELISA kits (one kit for each cytokine tested) for cytokine quantification. To represent a potential clinical scenario of a patient developing irAEs during immune checkpoint blockade therapy, we prepared three samples with increasing concentrations of cytokines (spike in experiments into fetal bovine serum (FBS)) and subsequently analysed these samples with our digital nanopillar SERS assay and the commercial ELISA kits. FBS was used as complex sample matrix devoid of human cytokines. As the limits of detection for the ELISA kits (FGF-2 = 0.95 pM, G-CSF = 1.66 pM, GM-CSF = 1.11 pM, and CX3CL1 = 17.86 pM) were above the attomolar level, the simulated samples were prepared to suit the detection range of these kits. For the digital nanopillar SERS assay, the samples were diluted correspondingly and generated consistent results with the ELISA kits as shown in Supplementary Table 7. No statistical differences were found between ELISA and digital nanopillar SERS results based on Mann–Whitney test. Furthermore, we compared the detection of four cytokines in human serum with digital nanopillar assay and ELISA kits (Supplementary Table 8). The cytokine levels in human serum were below the limit of detection for the conventional ELISA kits, whereas their concentration was quantified by digital nanopillar SERS platform. For the human serum spiked with standard cytokines, the digital SERS platform generated similar results to ELISA without significant differences by Mann–Whitney test. Collectively, the digital nanopillar SERS platform showcased the ability to robustly and accurately quantify cytokines in complicated samples, which is significant for the prospect of dynamic correlation monitoring of irAEs in clinical samples.
Following the demonstration of the accuracy of digital nanopillar SERS platform, we tested the four cytokine levels in ten healthy people (Supplementary Table 9). These ten healthy people showed cytokine concentrations beyond the conventional ELISA capability to accurately quantify, which was consistent with previous reports9,50,51.
Dynamic correlation monitoring of irAEs in melanoma patients receiving immune checkpoint blockade treatment
Having established the feasibility of digital nanopillar SERS in simulated clinical samples, we applied the platform for longitudinally monitoring irAEs in ten melanoma patients (2–3 time points per patient, 26 samples in total) who underwent immune checkpoint blockade therapy (Supplementary Table 10). By diluting the patient samples to follow Poisson distribution, we quantified the cytokine concentration using digital nanopillar SERS platform. Based on the clinical assessments, the patients were classified into two categories: (i) developed severe irAEs (grades 3 and 4) and needed hospitalisation and dedicated treatment (Patients 1, 2, 3, 4, 5); and (ii) developed minor irAEs (grades 1 and 2) that could be managed with immunosuppressants (e.g., corticosteroids) or exhibited no symptom of irAEs (Patients 6, 7, 8, 9, 10).
As a representative case, Fig. 7 shows two cytokine profiles of a patient with severe irAEs (Patient 1) and a patient with mild irAEs (Patient 1). For Patient 1 who received ipilimumab (cytotoxic T-lymphocyte antigen-4 (CTLA-4) inhibitor) and was checked on days 7, 21, and 42, the confocal SERS images showed an increase in active pillars with the continuation of treatment (Fig. 7a–c), suggesting an elevation of the cytokine levels that could potentially trigger the severe irAEs. In agreement with the Raman images, the quantitative counting results for the four cytokines also corroborated the increase of cytokine concentrations peaking in sub-fM levels (Fig. 7d). These cytokine levels were below the limit of detection of conventional ELISA kits (pM level). Importantly, we observed significantly elevated cytokine concentrations in Patient 1 serum on day 42 compared to days 7 and 21. This patient showed the onset of grade 4 irAEs (i.e., colitis) 13 days later (day 55), consistent with the concept that higher cytokine levels correlate with increased risk of developing irAEs9. To further evaluate the utility of these four biomarkers as a signature in identifying and characterising irAEs, we analysed all the counting data from Patient 1 by applying linear discriminant analysis (LDA). As seen in Fig. 7e, the LDA successfully distinguished the data on day 42 into a separate zone from days 7 and 21, which may indicate the potential value of biomarkers in monitoring irAEs development. We further demonstrated Patient 1 LDA with the use of all combinations of two (Supplementary Fig. 16) and three cytokines (Supplementary Fig. 17). Overall, the LDA with four cytokines showed improved classification over using three or less cytokines. Interestingly, considering FGF-2/G-CSF, G-CSF/GM-CSF, and G-CSF/CX3CL1, the LDA generated similar performance to the LDA with four cytokines. To further compare the classification power of FGF-2/G-CSF, G-CSF/GM-CSF, and G-CSF/CX3CL1, and four cytokines, we performed LDA of Patient 2 (Supplementary Fig. 18), which suggested a better differentiation with the use of four cytokines. Therefore, the inclusion of all four cytokines in LDA facilitated a wider and more accurate patient sample analysis. Similarly, Patients 2, 3, 4, and 5, who manifested severe irAEs were connected with higher cytokine levels (Supplementary Fig. 19) and amelioration of irAEs symptom was witnessed with a decrease of cytokine concentrations. For these severe irAEs patients, the LDA model showed a clear discrimination in cytokine profile and this could help to identify patients at risk of irAEs (Supplementary Fig. 19).Fig. 7 Digital nanopillar SERS assay for monitoring melanoma patients during immune checkpoint therapy.
For Patient 1 who developed severe irAEs, SERS images for cytokine detection on a day 7, b day 21, c day 42, d cytokine concentration graph for fibroblast growth factor 2 (FGF-2), granulocyte colony-stimulating factor (G-CSF), granulocyte-macrophage colony-stimulating factor (GM-CSF), and fractalkine (CX3CL1). The two shorter horizontal lines denote the interquartile ranges (25th and 75th percentile) and the longer horizontal lines in between denote the median (50th percentile), and e LDA analysis, respectively. For Patient 6 who developed mild irAEs, SERS images for cytokine detection on f day 0, g day 21, h day 42, i four cytokine concentration graph, the two shorter horizontal lines denote the interquartile ranges (25th and 75th percentile) and the longer horizontal lines in between denote the median (50th percentile), and j LDA analysis, respectively. IPI ipilimumab, PEMBRO pembrolizumab; G3 grade 3, G2 grade 2; SD stable disease, PR partial response. For Patient 1, the median (interquartile range) of active pillars per scanning image of FGF-2, G-CSF, GM-CSF, CX3CL1 on day 7: 14 (11–22.5), 23 (21, 29), 12 (7.5–18), 17 (9–25.5); day 21: 30 (19–37.5), 33 (19–41), 26 (17.5–36.5), 29 (21–43); and day 42: 33 (16.5–58.5), 76 (64–128.5), 25 (14–39.5), 48 (26.5–73.5), respectively. For Patient 6, the median (interquartile range) of active pillars per scanning image of FGF-2, G-CSF, GM-CSF, CX3CL1 on day 0: 18 (16–23), 49 (31.5–56), 23 (17.5–28), 20 (14.5–27); day 21: 29 (24–33.5), 53 (46.5–70), 35 (25–46), 22 (19–29.5); and day 42: 13 (8–16.5), 44 (23.5–55.5), 10 (6.5–12.5), 30 (24–34.5), respectively. The data represented three technical replicates obtained from three chips. Nine images were acquired from each chip for cytokine counting. Statistical analysis was based on Kruskal–Wallis test followed by Dunn’s test to correct multiple comparisons (two-sided). Source data are provided in the Source Data file.
As for Patient 6 who exhibited mild grade 2 irAEs on the skin during combined ipilimumab and pembrolizumab (programmed death-1 (PD-1) inhibitor) or single ipilimumab treatments, the dynamic monitoring displayed relatively stable cytokine levels on different follow-up visits (Fig. 7). Specifically, the confocal Raman images (Fig. 7f–h) and the molecular counting (Fig. 7i) in this patient serum consistently showed no significant cytokine level alterations on the three time points (days 0, 21, 42). Under this circumstance, LDA failed to clearly classify the data into separate sections (Fig. 7j). Likewise, Patients 7, 9, and 10 possessed stable cytokine levels and were diagnosed with low grade irAEs. Meanwhile, Patient 8 who showed decreasing cytokine levels did not display signs of irAEs. LDA was not able to classify Patients 7 and 8 who had mild irAEs and did not show irAEs, but it recognised the minor difference in Patients 9 and 10 who showed grade 1 irAEs (Supplementary Fig. 20).
Overall, we found the preliminary evidence to suggest that significantly elevated cytokine levels have a strong correlation with the development and manifestation of severe irAEs, whereas stabile, low baseline, and decreasing cytokine concentration indicate mild and manageable irAEs. The relatively low concentrations of these four cytokines were below the detection sensitivities of commercially available ELISA kits (pM level), which limits their use in clinical studies. Importantly, the measurement at fM cytokine levels in clinical samples is consistent with the median concentrations of cytokine measured by using digital ELISA51. The successful demonstration of the digital nanopillar SERS platform in dynamic detection of cytokines in patient serum provides a potential approach for the future accurate early detection, characterisation, and monitoring of irAEs in clinical settings. However, it is important to note that the cytokine concentration changes are not directly correlated with the treatment response according to response evaluation criteria in solid tumours (RECIST). In our pilot study, some melanoma patients showed higher levels of cytokines compared to the healthy controls. The power of our digital nanopillar SERS platform lies in the capability to longitudinally monitor cytokines in individual patients over time.
Discussion
Despite the frequent occurrence of irAEs in immune checkpoint therapy, particularly for the combination treatment, the prediction of the emergence of irAEs remains elusive. Mounting data suggest a potential role of cytokines as predictive markers for irAE monitoring in immune checkpoint therapy9,11,13,52. Although promising, accurate quantification of these biomarkers was often not possible due to the dearth in technologies with sufficient detection sensitivity. Typically, either cytokines above the detection limit of immunosorbent assay were selected11, or relative cytokine quantification9 was performed for investigating irAEs. The former approach has the drawback of potentially excluding the low abundance cytokines of significance in irAEs. As for relative quantification9,53, the cytokine concentrations are determined by relating to a standard that had the observed median fluorescence value closest to the median of the test sample. The relative concentration, however, may fail to represent accurate cytokine levels and thus needs further exploration. Our developed digital nanopillar SERS assay offers early data suggestive of a potential approach to the above-mentioned challenges and provides the possibility to study trace amounts of a panel of cytokines in an accurate quantification manner as well as concomitantly providing attomolar level sensitivity. The current proof-of-principle approach has measured four of the potential inflammatory and/or immune toxicity-related cytokines for the prediction of emergence, characterisation, and/or quantifiable correlation with irAEs in melanoma patients.
By leveraging the narrow line width of Raman spectra, the developed digital SERS counting assay shows the ability to sensitively and simultaneously detect multiple cytokines. The adoption of the novel digital quantification mode in SERS using gold–silver alloy nanoboxes further improves the high sensitivity of SERS technology. Notably, the digital counting strategy offers an option for reproducible SERS quantification by avoiding the common Raman signal fluctuations induced by ensemble measurements. The ensemble measurement in SERS relies on the enhancement of Raman signals of molecules located in or near the “hot spots” (i.e., strong electromagnetic fields)30. Due to the random distribution and various efficiencies of “hot spots”, it can result in the discrepancies in acquired SERS intensities for inter-laboratory and even intra-laboratory tests29. To circumvent the impact of Raman intensity fluctuation on accurate quantification, we employed the digital SERS signals from the single-SERS-active nanoboxes on discrete pillar arrays to enumerate the targets and only count the “yes” or “no” signal for a robust and reproducible SERS analysis. Furthermore, the digital readout model, which regards both aggregated and single nanoparticle as a single binding event to reflect the true target number, can have a better accuracy and robustness than the intensity-based assay39.
We believe that the proposed digital nanopillar SERS assay could be used to monitor other cytokine-induced immune responses. For instance, with the outbreak of 2019 novel coronavirus (2019-nCoV), it is yet difficult to predict which infected patient will develop a strong immune response that requires hospitalisation. However, cytokines have been indicated to play a major role in the severity of immune response for critically ill patients infected with 2019-nCoV54. The specific detection of multiple cytokines at early stages of viral infection could thus potentially address this issue and help to provide the clinical care for people at the highest risk. For patients with high cytokine concentrations, the digital nanopillar SERS platform will require the sample dilution to suit Poisson distribution.
In summary, we propose a digital nanopillar SERS platform for the parallel counting of single cytokines and dynamic monitoring in the clinical context of irAE development during immune checkpoint blockade therapy. The platform achieved attomolar level sensitivity by utilising discrete pillar array compartments to hold the single cytokine and subsequently applied single-particle active nanobox-based SERS nanotags for cytokine identification and counting. The confocal Raman mapping on the pillar array offered the highest possible clinical specificity by reducing nonspecific signals and provided a “yes/no” type counting approach for reproducible Raman signal readout. The designed platform was rigorously optimised and tested in simulated clinical samples prior to the evaluation for irAE monitoring in stage IV melanoma patients receiving immune checkpoint blockade therapy. We envisaged this platform possessing the advantages of highly sensitive and multiplexing analysis capability can transit into future irAE detection methodologies after extensive validation in a large cohort of clinical samples over different time courses.
Methods
Materials
Silver nitrate (AgNO3), hydrogen tetrachloroaurate (III) trihydrate (HAuCl4·3H2O), MBA, DTNB, TFMBA, MMTAA, DSP were obtained from Sigma Aldrich. Ascorbic acid (AA) of analytical grade was purchased from MP Biomedicals. FGF-2 (223-FB), G-CSF (214-CS), GM-CSF (215-GM), CX3CL1 (365-FR) cytokines; monoclonal anti-FGF-2 (MAB233), anti-G-CSF (MAB214), anti-GM-CSF (MAB615), anti-CX3CL1 (MAB3652) antibodies; polyclonal anti-FGF-2 (AF-233), anti-G-CSF (AF-214), anti-GM-CSF (AF-215), anti-CX3CL1 (AF-365) antibodies; and FGF-2 (DY233-05), G-CSF (DY214-05), GM-CSF (DY215-05), and CX3CL1 (DY365) ELISA kits were bought from R&D Systems.
All the patient serum or plasma samples were collected at the Austin Hospital (Melbourne) under approved human ethic protocols and written informed consents were obtained from all patients before sample collection. Ethics approval was obtained from The University of Queensland Institutional Human Research Ethics Committee (approval nos. 2011001315 and 2016000876) and the following clinical assay was carried out according to the approved guidelines.
Preparation of single-particle active SERS nanotags
The preparation of SERS nanotags involved the synthesis of nanoboxes and the subsequent functionalisation with Raman reporters and antibodies. For nanobox synthesis, 45 µL of HAuCl4 (1 wt%) was added into 10 mL of ultrapure H2O (18.2 Ω cm) under magnetic stirring (800 r.p.m.) for 1 min, followed by simultaneously introducing 170 µL of AgNO3 (6 mM) and 30 µL of AA (0.1 M) into the stirring solution. Then, the formation of nanoboxes was indicated by the appearance of an apparent blue colour within 6 s and the samples were collected 1 min later by centrifuging at 600 g for 15 min.
To functionalise nanoboxes with Raman reporters and antibodies, 300 µL of nanoboxes centrifuged from 1 mL of as-prepared solution were co-incubated with one type of Raman reporters (i.e., 10 µL of DTNB, 8 µL of MBA, 10 µL of TFMBA, or 10 µL of MMTAA) and 2 µL of DSP linker for 6 h. After that, the Raman reporter and DSP functionalised nanoboxes were separated by centrifuging at 600 g for 15 min, and resuspended into 300 µL of PBS (0.1 mM). Then, 2 µg of anti-FGF-2, anti-G-CSF, anti-GM-CSF, anti-CX3CL1 antibodies were added to MBA, DTNB, TFMBA, and MMTAA labelled nanoboxes, respectively. After overnight incubation at 4 oC, the functionalised nanoboxes were purified by centrifuging at 600×g for 15 min to separate free antibodies and the final products were resuspended into 200 µL of 0.1% BSA for future use.
Fabrication of pillar arrays
The chip was made of a sensing array measuring 1 mm × 1 mm (Supplementary Fig. 1a) and consisted of 250,000 individual pillars. Each pillar was 1 µm wide, 1 µm long, and 1 µm high. The pillars were evenly spaced by 1 µm from one pillar to the next (Supplementary Fig. 1b). The pillar array was designed using Nanosuite 6.0 (Raith GmbH) and Beamer 5.9.1 (GenISys GmbH) and fabricated on a 4-inch p-type <100> silicon wafer (Bonda Technology Pte Ltd, Singapore) using electron beam lithography (EBL). The wafer accommodated 76 separate pillar arrays. Silicon wafer was first cleaned in acetone, isopropanol with sonication for 2 min each, followed by rising with deionised H2O and dehydration bake at 180 °C for 2 min. Prior to the resist coating, the wafer had undergone a further O2 plasma cleaning at 200 W for 5 min (Diener Atto, Diener Electronic GmbH). The cleaned wafer was spin-coated with two layers of polymethyl methacrylate (bottom: 495K A4 PMMA, top: 950k A4 PMMA, from MicroChemicals GmbH) using the CEE Apogee Coater (Cost Effective Equipment, LLC) at 1500 r.p.m. for 60 s each. After the coating of each layer, the wafer was baked immediately on a hot plate to prevent from intermixing of the two layers of resist. The baking time was 10 and 3 min for the bottom and top layer, respectively, at 180 °C. The thickness of the photoresist was found to be ~450 nm (top PMMA: ~250 nm, bottom PMMA: ~200 nm), characterised by white light reflectometry (FilmTek 2000M, Scientific Computing International). EBL was performed in the Raith EBPG5150 system. The patterns were exposed in EBL with an accelerated voltage of 100 kV, 150 nA of beam current (spot size ~80 nm), with step sizes of 40 nm and an electron dose of 1200 µC/cm2. The exposure time per 4-inch wafer was ~35 min, each containing 76 individual chips. After exposure, the wafer was developed in a mixture of isopropanol and methyl isobutyl ketone (3:1) for 60 s and rinsed immediately with isopropanol, followed by drying with N2. An oxygen plasma descum process, at 100 W, 60 s (Diener Atto, Diener Electronic GmbH) was carried out to remove resist residues prior to the deposition. Next, 10 nm titanium and 200 nm gold were deposited by physical vapour deposition using a Temescal FC-2000 electron beam evaporator (Ferrotec, U.S.A.). After overnight lift-off at room temperature in Remover PG (MicroChemicals GmbH, Germany), the excess material was washed off and the pillar array structure was revealed (Supplementary Fig. 1c). To create the pillar height (i.e., 1 µm), reactive ion etching (Oxford Instruments, UK) was applied for anisotropic etching of the silicon. Hereby, the deposited gold served as mask to protect the underlying silicon while the un-masked silicon was removed. Next, the wafer was coated with a protective layer of cured AZnLOF 2020 prior to wafer dicing into 76 individual sensing chips consisting of a single pillar array. Prior to use, the protective layer was washed off by consecutive washes with isopropanol and acetone and dried under a stream of nitrogen.
Pillar array functionalization
Antibody functionalisation of the gold-topped pillar array was conducted by crosslinking the antibodies to the gold surface using DSP. A solution of 5 mM DSP in dimethyl sulfoxide was pipetted onto the pillar array and incubated at room temperature for 2 h. After rinsing the pillar array with ethanol and PBS, a solution of 5 µg/mL anti-cytokine monoclonal antibody solution (100 fold dilution of antibody stock solution) in PBS was incubated overnight at 4 °C. Subsequently, the pillar array was rinsed with PBS and blocked using 1% bovine serum albumin in PBS for 1 h. Prior to use, the pillar array was rinsed with PBS. All PBS solutions were filtered through a sterile 0.22 µm syringe filter (Millex-GP, Merck, U.S.A.).
Digital nanopillar SERS profiling of cytokines
Cytokines (FGF-2, G-CSF, GM-CSF, and CX3CL1) with different concentrations in PBS (2.6 aM, 26 aM, 260 aM, and 1031 aM) were incubated with antibody functionalised pillar arrays at room temperature for 30 min, followed by washing the pillar array three times with washing buffer (0.1% BSA and 0.01% Tween 20 in PBS). SERS nanotags were then added into the pillar array for another 30 min incubation under room temperature to identify the targets. Finally, the pillar arrays were washed to remove the free SERS nanotags and were subject to confocal Raman microscope for quantification. For each sample, nine SERS images with each image has the dimension of 60 µm × 48 µm were taken on the pillar array to calculate the overall cytokine concentration.
Simulated clinical sample detection
For the recovery experiment, standard cytokines (FGF-2, G-CSF, GM-CSF, and CX3CL1) with the concentration of 1 fM were added into healthy human serum and then diluted ten times with PBS to quantify.
For the quantification of cytokines in FBS, three simulated clinical samples were prepared by titrating various concentrations of standard cytokines into 10% FBS: Sample 1 (FGF-2 = 3.64 pM, G-CSF = 3.19 pM, GM-CSF = 4.29 pM, and CX3CL1 = 28.57 fM); Sample 2 (FGF-2 = 7.28 pM, G-CSF = 6.38 pM, GM-CSF = 8.58 pM, and CX3CL1 = 571.4 fM); and Sample 3 (FGF-2 = 14.56 pM, G-CSF = 12.76 pM, GM-CSF = 17.16 pM, and CX3CL1 = 1142.8 fM). These three samples were then detected directly using the commercial ELISA kits or digital nanopillar SERS assay with a further dilution of 105, 2 × 105, and 4 × 105, respectively.
Spectroscopic ellipsometry
The antibody film thickness was measured by in-solution spectroscopic ellipsometry (M2000V JA Woollam Co., Inc. USA) using gold-coated substrates and flow cell (QSense® Ellipsometry, Biolin Scientific, Sweden). Measurements were performed at an angle of 65°. Data analysis was performed by CompleteEASE® software using a B-Spline data fit and Cauchy model to calculate the antibody film thickness.
MALDI-TOF MS
The antibody-functionalised nanopillar array chip was subjected to tryptic digest prior to analysis. Sequencing-grade trypsin was made to 50 ng/µL in 25 mM ammonium bicarbonate, and sprayed over the chip using a Bruker Imageprep instrument (Bruker, USA). After trypsin deposition, the chip was incubated in a humid environment at 40 °C for 3 h. Subsequently, the chip was sprayed with a matrix solution, 10 g/L α-cyano-4-hydroxycinnamic acid in 50% acetonitrile with 0.2% trifluoroacetic acid. Next, the chip was analysed with a Bruker Ultraflex III MALDI-TOF mass spectrometer (Bruker, USA) in positive linear mode using Flex Imaging 4.0 (Bruker, USA) with a pixel size of 60 µm. Data were collected from 2 k–30 k m/z, at a laser repetition rate of 200 Hz. Data were normalised using the root mean square approach and visualised using Flex Imaging 4.0 (Bruker, USA) and SCILS LAB 2017a software. For the SCILS LAB analysis, the data were imported using a convolution baseline subtraction, and displayed using root mean squared normalisation.
Instrumentations
SEM images of pillar arrays and nanoboxes were taken on a JEOL-7100 field emission (FE)-SEM (20 kV voltage). TEM images of nanoboxes were taken on a JEOL-2100 microscope (200 kV voltage). NTA of nanobox size distribution was performed with Malvern NanoSight NS300. UV-vis extinction spectrum of nanoboxes was performed with a Shimadzu UV-2450 spectrophotometer. Confocal Raman mapping was conducted on a WITec alpha 300 R spectrometer using 632.8 nm He–Ne laser with the power of 35 mW, a grating of 600 g/mm used with EMCCD camera, spectral resolution of 1.390 cm−1 to 2.114 cm−1, confocal pinhole size of 100 µm, 100× air objective with NA of 0.90, and 0.05 s integration time. The theoretical spot size was 857.80 nm based on the Abbe diffraction limit (i.e., d = 1.22λ/NA). The scanning area was set to have 60 µm × 48 µm with 86 points per line and 69 lines per image. For each pillar array, nine separate scanning areas were taken in total and the total active pillars were used for quantification. The SERS mapping images for counting were taken by focusing the laser on the top of the pillar surfaces. Specifically, the laser was firstly focused on the silicon substrates by obtaining the strongest silicon signals (520 cm−1) and then the 100× objective was moved up in z-axis direction of 1 µm for SERS scanning. The system was calibrated with the first-order photo peak of silicon at 520 cm−1.
Data analysis
To assign the SERS nanotag membership for DTNB, MBA, TFMBA, and MMTAA, Project Five 5.0 software from WITec was utilised to create four filters, which summed a spectral range of 40 cm−1 with the centre position at the characteristic Raman peak of each reporter and subtracted the background with a polynomial algorithm. Specifically, the filter ranges of four Raman reporters DTNB-, MBA-, TFMBA-, and MMTAA-coated SERS nanotags were (1310–1350 cm−1), (1060–1100 cm−1), (1360–1400 cm−1), and (1268–1308 cm−1), respectively. All the SERS images were analysed by using threshold intensity to determine the successful binding events. Specifically, the threshold intensity of FGF-2, G-CSF, GM-CSF, and CX3CL1 was set at 5000, 4000, 5000, and 5000, respectively. For each image, the threshold intensity was doubled-checked and adjusted based on the true Raman peaks in the spectra. Statistical analysis assuming unequal variances was conducted with Kruskal–Wallis test among three groups or Mann–Whitney test between two groups with GraphPad Prism 8.4. To control the error appropriately, we performed multiple comparisons using Dunn’s test. LDA of clinical samples was performed in R software (3.6.2) with the MASS package (7.3-52). The active pillars in SERS images were counted with Image J software.
Supplementary information
Supplementary Information
Peer Review File
Source data
Source Data
Peer review information
Nature Communications thanks Chongwen Wang and Isaac Pence for their contribution to the peer review of this work. Peer reviewer reports are available.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
The online version contains supplementary material available at 10.1038/s41467-021-21431-w.
Acknowledgements
The authors acknowledge grants received by our laboratory from the National Breast Cancer Foundation of Australia (CG-12-07) and the ARC DP (160102836 and 210103151). These grants have significantly contributed to the environment to stimulate the research described here. J.L. acknowledges support from the Australian Government Research Training Program Scholarship. A.W. and A.A.I.S. thank the National Health and Medical Research Council for funding (APP1173669 and APP1175047). A.B. is the recipient of a Fellowship from the Victorian Government Department of Health and Human Services acting through the Victorian Cancer Agency. We acknowledge the facilities, and the scientific and technical assistance, of the Australian Microscopy & Microanalysis Research Facility at the Centre for Microscopy and Microanalysis, The University of Queensland. We appreciate to receive the technical and scientific guidance from Queensland node of the Australian National Fabrication Facility (Q-ANFF) in confocal Raman mapping and spectroscopic ellipsometry measurement.
Author contributions
J.L., A.W., A.I.S., H-H.C., Y.W., A.B., P.M., and M.T. contributed to the design of the experiments and analysing the data. J.L. and A.W. performed the experiments and prepared the manuscript. All authors read, commented, and edited the manuscript, and assisted during the revision process.
Data availability
Data supporting the findings of this work are available within this paper and the supporting information files. A reporting summary of this work is available as a Supplementary file. Source data are provided with this paper.
Competing interests
The authors declare no competing interests. | UNK, Q3WK | DrugDosageText | CC BY | 33597530 | 19,690,448 | 2021-02-17 |
What was the outcome of reaction 'Pancreatic toxicity'? | A digital single-molecule nanopillar SERS platform for predicting and monitoring immune toxicities in immunotherapy.
The introduction of immune checkpoint inhibitors has demonstrated significant improvements in survival for subsets of cancer patients. However, they carry significant and sometimes life-threatening toxicities. Prompt prediction and monitoring of immune toxicities have the potential to maximise the benefits of immune checkpoint therapy. Herein, we develop a digital nanopillar SERS platform that achieves real-time single cytokine counting and enables dynamic tracking of immune toxicities in cancer patients receiving immune checkpoint inhibitor treatment - broader applications are anticipated in other disease indications. By analysing four prospective cytokine biomarkers that initiate inflammatory responses, the digital nanopillar SERS assay achieves both highly specific and highly sensitive cytokine detection down to attomolar level. Significantly, we report the capability of the assay to longitudinally monitor 10 melanoma patients during immune inhibitor blockade treatment. Here, we show that elevated cytokine concentrations predict for higher risk of developing severe immune toxicities in our pilot cohort of patients.
Introduction
The advent of immune checkpoint therapy has revolutionised the landscape of traditional cancer treatment and is believed to constitute the backbone of managing certain malignancies1–3. By capitalising on the blockade of immune checkpoint inhibitors to take the brakes off parts of the immune system, this emerging therapy has achieved great success producing long-lasting responses (e.g., 10 years or more) in a small but significant fraction of patients3–6. Nevertheless, upon the blockade of immune checkpoint molecules, the activated and potentiated immune reaction predisposes patients to a significant risk of immune-related adverse events (irAEs), which can occur in up to 80% of patients receiving immune checkpoint therapy7–9. The high incidence of irAEs, which may manifest at any time during treatment, can offset the clinical benefits, lead to premature therapy cessation, and even be life-threatening for certain patients10–12. To assist the successful implementation of immune checkpoint therapy, the use of predictive biomarkers for early identification and vigilant monitoring of irAEs is thus critical and a pressing need in avoiding or ameliorating detrimental effects and adjusting therapeutic options.
Cytokines, small signalling proteins, are promising candidates to indicate the occurrence of irAEs due to their prominent role in modulating the anti-cancer immune responses, including enhancing antigen priming, recruiting immune cells into the tumour microenvironment, and upregulating certain immune checkpoint molecules9,13,14. Particularly, excessive cytokine secretion has been implicated in severe inflammation as a major constituent leading to irAEs. For example, the overproduction of fibroblast growth factor 2 (FGF-2)15–18, granulocyte colony-stimulating factor (G-CSF)19, granulocyte-macrophage colony-stimulating factor (GM-CSF)20, and fractalkine (CX3CL1)21 have been found to participate in immune-related inflammatory disease (e.g., rheumatoid arthritis, autoimmune gastritis, and Crohn’s disease). These inflammatory cytokines have recently been reported to indicate irAEs for melanoma patients who underwent immune checkpoint therapy9. The clinical deployment of cytokine analysis for irAE monitoring is challenging and requires a technology that can (i) determine the selected cytokines with great sensitivity9, especially at the onset of irAE development, where the cytokine concentrations are likely to be the lowest; as well as (ii) simultaneously detect a panel of cytokines to reflect the complex interplay of cytokine signalling pathways22 and the variable irAE symptoms among patients.
Conventional cytokine analyses such as immunosorbent assays have limited clinical applicability for irAE assessment due to their limited capacity to detect low cytokine concentrations in blood as well as for assessing a panel of cytokines in a single sample simultaneously. Recently, advances in micro/nanomaterial-based systems have provided a promising suite of technologies that improve the conventional assays by overcoming the above limitations23,24. Encouragingly, the unique advantages of micro/nanomaterial-based systems convey an attractive option for cytokine analysis with the desired results of high sensitivity and multiplexing. The high specific surface area of these miniaturised materials increases mass transfer subsequently enhancing the interaction with target molecules and thus improving the detection sensitivity25. The capabilities of micro/nanomaterial fabrication techniques permit individually separated compartments sufficiently discrete to hold single molecules and hence encompasses a promising strategy for counting assays that can further push the sensitivity of the traditional assays24,26,27. Moreover, the physicochemical properties of nanostructured materials can be exploited to simultaneously label multiple targets (e.g., various spectral signatures) for high-throughput parallel measurements28–30. Therefore, by combining the potential of micro/nanomaterial systems with the need for sensitive irAE monitoring, we have developed a platform for sensitive and multiplex cytokine counting analysis.
Combining the use of (a) discrete single cytokine nanopillar array chip with discretely separated compartments, (b) control of target concentrations to follow a Poisson distribution, and (c) the recognition of target by single-particle active surface-enhanced Raman scattering (SERS) nanotags with a confocal Raman microscope allows accurate and in situ counting of a multi-cytokine panel (FGF-2, G-CSF, GM-CSF, and CX3CL1). Different from the fluorescence-based digital counting strategies24,26, the strikingly narrow spectral peaks of SERS (~1–2 nm) in comparison to fluorescence (~50 nm) makes this platform intrinsically ideal for multiplexed cytokine analysis28.
In this work, we present a digital nanopillar SERS platform that enables the specific cytokine quantification down to attomolar levels and the application in melanoma patients receiving immune checkpoint blockade therapy. Beyond the capability to predict irAEs in melanoma patients receiving therapy, the digital nanopillar SERS assay could potentially be extended to other cytokine-associated immune responses such as excessive immune activation due to viral or bacterial infections (such as COVID-19).
Results
Digital nanopillar SERS platform for parallel profiling of single cytokine
Our concept of digital nanopillar SERS platform for cytokine analysis relies on Rayleigh criterion separation, probability-driven Poisson distribution, single-particle active SERS nanotags, and confocal SERS mapping (Fig. 1). To precisely fabricate the pillar array, we opted to use an electron beam lithographic approach to write the array into a photon-sensitive material followed by physical vapour deposition of gold to create the gold-topped pillars, and selectively reactive ion etching to reveal the pillar structure (Supplementary Fig. 1). The nanopillar array chip consisted of 250,000 individual pillars. As shown in the scanning electron microscope (SEM) image of Fig. 1a, the cubic nanopillars have an edge-to-edge width of 1000 nm and are evenly distributed at 1000 nm intervals to suit the lateral Raman microscope resolution (~1000 nm) that fulfils the Rayleigh criterion separation required to acquire a single SERS spectrum from each pillar without spectral overlap from adjacent pillars.Fig. 1 Digital single-molecule nanopillar surface-enhanced Raman scattering (SERS) platform for parallel counting of four types of cytokines.
SEM images of a pillar array side view, b nanoboxes, and c a single nanobox on the top of a pillar; d SERS spectra of nanoboxes conjugated with 5,5-dithiobis (2-nitrobenzoic acid) (DTNB), 4-mercaptobenzoic acid (MBA), 2,3,5,6-tetrafluoro-4-mercaptobenzoic acid (TFMBA), or 2‐mercapto‐4‐methyl‐5‐thiazoleacetic acid (MMTAA) Raman reporters; e workflow for multiplex counting of cytokines, including fibroblast growth factor 2 (FGF-2), granulocyte colony-stimulating factor (G-CSF), granulocyte-macrophage colony-stimulating factor (GM-CSF), and fractalkine (CX3CL1). Data from one independent experiment.
By using specific gold-thiol chemistry with the linker molecule dithiobis (succinimidyl propionate) (DSP), the gold-topped pillars were selectively functionalised with target recognition antibodies (anti-FGF-2, anti-G-CSF, anti-GM-CSF, and anti-CX3CL1) and acted as the small compartments to capture and confine the individual cytokine. Upon DSP binding on the gold-topped pillars through gold-thiol bond, DSP uses N-hydroxysuccimide (NHS) ester to react with the amine groups of the antibodies31,32. The successful antibody conjugation on gold-topped pillar surfaces was confirmed by matrix-assisted laser desorption ionisation-time of flight mass spectrometry (MALDI-TOF MS) (Supplementary Fig. 2), which showed high molecular weight fragments derived from antibodies. Furthermore, spectroscopic ellipsometry was utilised to estimate the antibody density on pillar surfaces. Based on the obtained film thickness of 18.5 nm, the calculated antibody surface density was 5.5 mg/m2 using the Cuypers model33, which was in agreement with the reported antibody density on substrate surfaces34. Though these characterisations indicated the presence of antibodies on the pillar array, it was not possible to assess the exact distribution of the four structurally related (same immunoglobulin G family) antibodies on a single pillar with an area of 1 µm2. As an advantage of the digital read-out with a large redundancy of pillars, it is not essential to have all four types of antibodies equally distributed on a single pillar for the assay to work. The combined surface area of all antibody-conjugated pillars provides an excess of cytokine binding sites, which maximises successful cytokine capture within the pillar array. Supplementary Fig. 3 shows the SERS mapping images of an equimolar cytokine solution (1031 aM) that provided a similar signal count for the FGF-2, GM-CSF, G-CSF, and CX3CL1 SERS nanotags, indicating a required distribution of four kinds of antibodies conjugated to the array of pillars. Instrumental to the digital counting of cytokines, we controlled the target concentration based on the principle of Poisson distribution where the ratio of cytokine molecules to pillar number was <1:10, ensuring a 99% probability that there was either one cytokine molecule or zero per pillar24. At a ratio of 1:10, 10% of all pillars were occupied or activated with the cytokine molecules.
Following the capture of cytokines on nanopillars, SERS nanotags were applied to recognise the captured cytokines. The preparation of SERS nanotags was performed by the co-conjugating of Raman reporter and target antibody onto gold–silver alloy nanoboxes. Specifically, an average size of 80 nm gold–silver alloy nanoboxes were firstly synthesised using a rapid and aqueous phase approach35 as indicated in the SEM image in Fig. 1b. Supplementary Fig. 4a, b shows the transmission electron microscope (TEM) image of the nanoboxes with the hollow inner structure and a wall thickness of around 15 nm. Nanoparticle tracking analysis (NTA), which allows the tracking and detection of single particles, shows the nanoboxes have a mode size of 77 nm (D10 = 67.6 nm and D90 = 110.6 nm) (Supplementary Fig. 4c). UV-vis extinction spectroscopy demonstrates the nanoboxes possess a surface plasmon resonance (SPR) peak at 610 nm (Supplementary Fig. 4d). The resonance frequency of the nanoboxes enables a more sensitive signal readout with 632.8 nm laser excitation29, which also has a higher Raman scattering efficiency than 785 nm laser. Thereafter, four types of Raman reporters (5,5-dithiobis (2-nitrobenzoic acid) (DTNB), 4-mercaptobenzoic acid (MBA), 2,3,5,6-tetrafluoro-4-mercaptobenzoic acid (TFMBA), and 2‐mercapto‐4‐methyl‐5‐thiazoleacetic acid (MMTAA)) that generate unique Raman signals (1330 cm−1, 1080 cm−1, 1380 cm−1, and 1288 cm−1) were coupled with their corresponding detection antibodies onto nanoboxes as specific SERS nanotags for identification of FGF-2, G-CSF, GM-CSF, and CX3CL1, respectively. As shown in Fig. 1d, the four SERS nanotags provide the strong and non-overlapping Raman signals, which facilitates the multiplexing analysis of four cytokines. The assignment of the major Raman peaks from the four Raman reporters was summarised into Supplementary Table 1. To evaluate the SERS enhancement property of the nanoboxes, we calculated the enhancement factor (EF) of the four Raman reporters on the nanoboxes. Based on the labelled characteristic peaks in Supplementary Fig. 5, the calculated EFs of DTNB, MBA, TFMBA, and MMTAA were 8.14 × 106, 1.46 × 107, 4.01 × 107, and 3.26 × 107, respectively. The obtained EFs were higher than the reported spherical gold nanoparticles and pure silver nanocubes36 and comparable to the reported hollow nanocubes37, illustrating the high SERS property of the nanoboxes. To investigate the SERS nanotag stability, we monitored the Raman signal intensity over 7 days. As shown in Supplementary Fig. 6, Raman signal intensity variations are less than 5% in the SERS spectra, suggesting the good stability of the prepared SERS nanotags.
The following SERS mapping generated false-colour images for counting single cytokine molecules. Under the Raman microscope, the pillar array was visualised as a blue and black grid by representing the specific Raman shifts corresponding to the silicon signals (520 cm−1), in which the blue colour was assigned to silicon signals showing silicon substrates and the black colour indicated the gold-topped pillars because of the lack of silicon signals. The representation of the four colours of the SERS nanotags (red, green, purple, and cyan) on the gold-topped pillars (i.e., black) reflected cytokine molecule occupation (FGF-2, G-CSF, GM-CSF, and CX3CL1). Elevating the sensing area (or gold-topped pillars) from the silicon substrate was selected as a strategy to minimise the false-positive events. By using the confocal function of the Raman microscope, the laser was selectively focused on the gold-topped pillars, thus largely removing the background signals from potentially non-specifically adsorbed SERS nanotags on the silicon substrate. Finally, the specific SERS nanotag signals present or absent on the gold-topped pillars were counted and represented as percentage of active pillars used for total cytokine quantification. For statistical calculations, SERS mapping was applied for scanning 6480 pillars. This digital counting mode, therefore, has the potential to reach the ultimate sensitivity of single molecule cytokine detection.
Demonstration of the single-particle SERS activity of gold–silver alloy nanoboxes
The successful implementation of digital nanopillar SERS assay necessitates the use of single-particle active plasmonic nanostructures that give a clearly detectable signal for each of the single cytokine binding event. The single-particle SERS detection sensitivity is essential to this assay development as the single-particle inactive plasmonic nanoparticles (e.g., spherical gold nanoparticles)38 would unavoidably result in an underestimate of cytokine concentration.
We evaluated the single-particle SERS activity of the prepared anisotropic nanoboxes by acquiring the signals from the individual nanoboxes that were labelled with DTNB reporters. The use of DTNB, a non-resonant Raman reporter, guaranteed the Raman signal enhancement was solely contributed from the nanobox-generated electromagnetic field. As seen in the SEM image (Fig. 2a), two clearly separated DTNB-labelled nanoboxes were deposited on the silicon wafer (highlighted in red circles). The corresponding Raman image (Fig. 2b) displayed several bright Raman spots originating from these individual nanoboxes. The elongated bright Raman spots in the SERS mapping image were probably caused by the slight aggregation of several nanoboxes during sample preparation processes (e.g., centrifugation)38, which was difficult to visually resolve in the SEM image (Fig. 2a). However, unlike the intensity-based assay, the aggregated nanoboxes as SERS nanotags to target cytokine will not skew the digital readout result, because each cytokine will occupy a single pillar following Poisson distribution and both aggregated and individual nanoparticles are regarded as a single binding event that truly reflects the target number39. We then acquired the SERS spectra from two individual SERS nanoboxes (Fig. 2c, (1) and (2)) and two separate spots of bare silicon (Fig. 2c, (3) and (4)). The presence of nanoboxes showed the characteristic Raman signal at 1330 cm−1 from DTNB, whereas the silicon spectra (3) and (4) lacked the specific peak. This observation demonstrated the single-particle SERS activity of nanoboxes, which was largely attributed to the enhanced electromagnetic fields of nanoboxes on specific regions (e.g., tips and corners)40,41 and thereby facilitated the sensitive and accurate counting of cytokines. Based on the acquired SERS mapping image, the median (interquartile range) of the DTNB peak intensity (1330 cm−1) in the presence and absence of nanoboxes were 183.03 a.u. (149.48–243.35 a.u.) and 18.07 a.u. (15.51–23.12 a.u.), respectively. Furthermore, the mean ± standard deviation of the DTNB peak intensity with nanoboxes (213.41 ± 85.03 a.u.) distinguished clearly from the position without nanoboxes (18.79 ± 6.01 a.u.), which demonstrated the feasibility of correctly identifying the presence of nanoboxes.Fig. 2 Demonstration of the single-particle SERS activity of DTNB-labelled nanoboxes.
a SEM image and b corresponding SERS mapping image of DTNB-labelled nanoboxes on a silicon substrate; c representative SERS spectra of numbered locations indicated in a and b. The red dotted line shows the characteristic peak at 1330 cm−1 from DTNB. Data from one independent experiment. Source data are provided in the Source Data file.
Fig. 3 Study of confocal height on Raman signal intensity.
SERS mapping of FGF-2 SERS nanotags on the silicon substrate with changing confocal height of a 0 nm, b 500 nm, c 1000 nm, and d 1500 nm; selected Raman spectra obtained from e red circles and f blue circles of SERS images. Red dotted lines in e and f indicate peak signal at 1330 cm−1 from DTNB. Data from one independent experiment. Source data are provided in the Source Data file.
Fig. 4 Specificity of digital nanopillar SERS platform for FGF-2 cytokine detection.
Representative confocal SERS images in the presence of a target FGF-2 (1031 aM), and negative controls with non-target controls b G-CSF (1031 aM), c GM-CSF (1031 aM), d CX3CL1 (1031 aM), and e PBS. The median (interquartile range) of active pillars per scanning image for FGF-2, G-CSF, GM-CSF, CX3CL1, and PBS was 72 (63.5–76.75), 1.5 (1.5–2), 2 (1–4), 0.5 (0–1.25), and 1 (1–1.75), respectively. Data from one independent experiment.
Optimisation of digital nanopillar SERS platform for cytokine detection
The reliable detection of single cytokine molecules by the digital nanopillar SERS platform depends on the geometric features of the pillar array (i.e., pillar height, cross-section area of pillar) and assay conditions (i.e., incubation time for sample and SERS nanotags).
We first sought to investigate the effect of pillar height on Raman signal intensity to differentiate signals from non-specifically bound SERS nanotags on the silicon substrate and specifically bound SERS nanotags on the gold-topped pillars. FGF-2 SERS nanotags were randomly deposited on the silicon substrate to mimic the non-specific binding scenario and the Raman mappings were perfomed by moving the objective along the z-axis direction with different heights (0 nm, 500 nm, 1000 nm, and 1500 nm) to compare the signal intensity. In Fig. 3, a–d show the false-colour SERS images and e, f the corresponding SERS spectra with characteristic DTNB reporter peak at 1330 cm−1 acquired from the circled spots in a–d. At the height of 0 nm where the SERS nanotags were in focus, we noticed bright Raman spots (Fig. 3a) and strong Raman signals (black line in Fig. 3e, f). With increasing z heights to 500 nm and 1000 nm, the Raman spots decreased (Fig. 3b, c) and the signal intensity weakened/disappeared (red and blue lines in Fig. 3e, f) as the nanoboxes became increasingly out of focus. A further increase to 1500 nm did not remarkably weaken Raman signals compared to the height of 1000 nm (Fig. 3d). Hence, for the fabrication of the pillar array chip, we selected a pillar height of 1000 nm to greatly reduce the potential interference from non-specific signals.
The cross-section area of the pillars provides the space for cytokine binding and labelling with SERS nanotags. To study the effect of pillar cross-section area, we fabricated array chips with pillars of various widths (250 nm, 500 nm, and 1000 nm) (Supplementary Fig. 7a–c) and functionalised with anti-FGF-2 antibody. Each chip consisted of 250,000 individual pillars. We then analysed a sample that contained ~25,000 molecules of FGF-2 (40 µL, 1031 aM), which should result in 10% active pillars (ratio FGF-2: pillars of 0.1). As seen in Supplementary Fig. 7d–f, an increasing pillar cross-section area results in a higher fraction of active pillars. In reference to the expected active pillar percentage (10%), the 250 nm and 500 nm pillar arrays produced lower active pillars (2% and 6%), which suggested a significant loss of target recognition by SERS nanotags. For the 1000 nm wide pillars, the active pillar percentage was 11%, close to the nominal value of 10%. We further tested a sample with 260 aM FGF-2 (i.e., 2.5% active pillars) on the pillar array chips with 250, 500, and 1000 nm pillar widths. The capture efficiency of these three chips was summarised in Supplementary Table 2. In comparison to the pillar array of 250 nm and 500 nm sizes, the 1000 nm provided an improved capture efficiency. As the accessible target recognition surface area per pillar increases, it can possibly promote the thermodynamics and kinetics for higher surface binding and capture efficiency25. Consequently, the 1000 nm pillar array was adopted in the subsequent experiments.
An optimal incubation time of cytokine and SERS nanotags on the pillar array can shorten the operation time and reduce the potential risk of nonspecific binding that could lead to false-positive counting. We thus studied the effect of incubation time of cytokine with SERS nanotags for 30 to 90 min in a solution of 1031 aM FGF-2 and FGF-2 SERS nanotags. As suggested by Supplementary Fig. 8, the increase in incubation time gives rise to a higher proportion of active pillars. In comparison with the theoretical active pillar percentage (10%), both 30 min and 60 min incubation time were able to provide a desirable active pillar percentage (11% and 13%, respectively). A longer incubation time (90 min), however, reported an active pillar percentage (20%) two times higher than the theoretical, indicating the occurrence of nonspecific absorption of SERS nanotags on the pillar array chip. Thus, we selected 30 min incubation time for further digital nanopillar SERS measurements.
Specificity of the digital nanopillar SERS platform for cytokine detection
Accurate and reliable recognition of the specific target is essential for cytokine quantification in clinical samples. To demonstrate the detection specificity of the digital nanopillar SERS assay, we prepared an anti-FGF-2 antibody functionalised pillar array and measured samples containing target FGF-2 cytokine and controls (G-CSF, GM-CSF, CX3CL1, and PBS). It was observed that only the presence of FGF-2 activated significant amounts of pillars whereas the negative controls only generated negligible active pillars (Fig. 4), indicating the high specificity for FGF-2 detection. Similarly, we studied the specific detection of G-CSF, GM-CSF, and CX3CL1, as shown in Supplementary Figs. 9–11, in which the typical Raman images displayed high proportions of active pillars in the presence of specific targets but not for the negative controls.
To further investigate the specificity of binding between SERS nanotag and antibody-functionalised pillar, we performed SEM analysis to “closely” inspect the pillar array for the presence or absence of SERS nanotags. As a representative model, we selected to image FGF-2 SERS nanotags on the anti-FGF-2-functionalised pillar array after sample incubation with FGF-2 cytokine and non-target controls (Fig. 5). As expected, we observed the cubic nanoparticles on the top of pillars in the presence of FGF-2 due to the successful recognition of SERS nanotags. On the contrary, pillar arrays did not display a significant number of FGF-2 SERS nanotags with non-target controls. Consequently, the consistent Raman and SEM data demonstrate the capability of the assay for specific target cytokine counting. The ability to selectively identify these four cytokines in the designed assay is critically important for their usage in clinical samples.Fig. 5 Specificity of the digital nanopillar SERS platform for FGF-2 cytokine detection.
Representative SEM images of pillar array incubated with FGF-2 SERS nanotags in the presence of a, b FGF-2 (1031 aM), c G-CSF (1031 aM), d GM-CSF (1031 aM), e CX3CL1 (1031 aM), and f PBS. The red circles highlight the existence of SERS nanotags. Panel b is the magnified SEM image of the red-highlighted section in a. It is noted that nanofabrication debris on the sidewall of the pillars can also be seen. Data from one independent experiment.
Sensitivity of the digital nanopillar SERS platform for cytokine detection
As there is typically low abundance of cytokines in clinical samples, the technique based on cytokine detection is expected to possess sufficient sensitivity to reliably assess irAEs9. To investigate the sensitivity and dynamic detection range of the digital nanopillar SERS assay, we firstly titrated the designated concentration of one target cytokine (FGF-2) on the pillar array chip with 250,000 pillars. To comply with the Poisson distribution, the upper number of cytokine molecules in the sample is 25,000 which should result in 10% activated pillars. Based on this upper molecule number, we were motivated to challenge the assay by serially diluting the number of cytokine molecule in the sample from 25,000 (1031 aM), 6305 (260 aM), 631 (26 aM), and 63 (2.6 aM). As suggested by the Raman images in Supplementary Fig. 12 and with the decrease in FGF-2 molecules, the percentage of active pillars decreased correspondingly from 9.39% for 1031 aM, 6.59% for 260 aM, 1.12% for 26 aM, and 0.62% for 2.6 aM, showing a strong correlation that facilitates quantitative cytokine analysis.
Subsequently, we were interested in exploring the multiplexing capability of SERS to investigate the digital nanopillar SERS assay’s dynamic range for the simultaneous quantification of all studied cytokines. As the targets independently follow Poisson distribution, each of the cytokine was separately controlled to activate less than 10% pillars. The specific SERS nanotags provided unique signals for each cytokine that was visualised in the false-colour SERS images by a different colour, thereby enabling in situ and simultaneous cytokine detection. As suggested by the confocal SERS images in Fig. 6, an increase in cytokine concentration corresponded with a higher percentage of active pillars. To facilitate quantitative measurements of the cytokines, we calculated the logarithmic transformation of the percentage of active pillars versus cytokine concentration (Supplementary Fig. 13) confirming the strong statistical and potentially clinically relevant correlation (coefficient of determination (R2) >0.97) observed in the SERS images.Fig. 6 Sensitivity for the simultaneous detection of four cytokines.
Representative confocal SERS images of fibroblast growth factor 2 (FGF-2), granulocyte colony-stimulating factor (GM-CSF), granulocyte colony-stimulating factor (G-CSF), and fractalkine (CX3CL1) with the concentration of a 2.6 aM, b 26 aM, c 260 aM, d 1031 aM. Colour scale bars indicate Raman intensities from 5,5-dithiobis (2-nitrobenzoic acid) (DTNB), 4-mercaptobenzoic acid (MBA), 2,3,5,6-tetrafluoro-4-mercaptobenzoic acid (TFMBA), or 2‐mercapto‐4‐methyl‐5‐thiazoleacetic acid (MMTAA). The median (interquartile range) of active pillars per scanning image of FGF-2, G-CSF, GM-CSF, CX3CL1 for 2.6 aM: 3 (1.5–3), 1 (1–2), 2 (1–3), 2 (1–3); 26 aM: 8 (5.5–10), 10 (9–13), 7 (6–10), 8 (6–10); 260 aM: 40 (36–48), 40 (35–52), 39 (35–50), 37 (36–49); and 1031 aM: 79 (61.5–97), 78 (72–87.5), 88 (68.5–97), 79 (64–95), respectively. Data represents one experiment from three independent tests.
To further investigate the multiplexing quantification performance of the digital nanopillar SERS assay in human serum, we spiked standard cytokines in human serum and tested the dynamic range. Supplementary Fig. 14 shows the linear relationship curves for the four targets. Because of the more complicated sample matrix composition in human samples, the lowest detectable cytokine concentration (5.2 aM) was higher than the PBS solution (2.6 aM).
At a cytokine to pillar ratio of 1:10, we studied the probability of each pillar being occupied by different molecule numbers. To experimentally investigate the number of molecules on a single pillar, we analysed a cytokine mixture that contained all four target cytokines at equal concentration (i.e., ~6250 molecules per cytokine). To visualise and count molecule binding events on a single pillar, we labelled the captured cytokines with the four SERS nanotags that provide clearly distinguishable signals. Under Poisson distribution, the likelihood of having two or more molecules on a single pillar is <0.45% (Supplementary Table 3), which underlies the digital counting principle24. Compared to the theoretical Poisson distribution, the experiment data reported a close but slightly higher value, which was probably due to minor non-specific binding of SERS nanotags on the pillars.
The high sensitivity (attomolar level) of the digital nanopillar SERS assay can be ascribed to the following factors: the digital counting strategy, the single-particle SERS activity of the nanoboxes, and the use of pillars to suit confocal Raman mapping that efficiently excludes false-positive signals. Commercially available methods with potential for trace analysis of cytokines include the single-molecule ELISA Simona by Quanterix and electrochemical luminescence assay42,43 by Meso Scale Discovery. Compared to these two methods, the developed digital nanopillar SERS platform enabled in situ multiplexed detection of four cytokines with comparable sensitivity. Unlike the issues of photo bleaching and poor multiplexing analysis often encountered in fluorescence44 and luminescence assays45, SERS provides the advantage of high multiplexing (e.g., 31-plex)46–48 with the narrow Raman linewidth and high photo stability of the Raman reporters. In addition, this digital nanopillar SERS platform can provide more accurate quantification of cytokines by reducing the false-positive signals with the confocal setting, thus eventually help clinicians to monitor irAEs during immune checkpoint therapy. The highly sensitive readout for multiple targets also indicated the capability of this assay for cytokine detection to assess irAEs in clinically relevant samples.
Evaluation of digital nanopillar SERS platform on simulated patient samples
The detection of trace concentrations of cytokines in serum samples is difficult because plasma samples contain a high abundance of non-target molecules (e.g., serum albumin and other proteins) that can potentially interfere with cytokine detection and lead to inaccurate clinical results. To evaluate the capability of the digital nanopillar SERS assay in accurately counting single cytokine molecules, we opted to perform a recovery test in simulated patient plasma samples (i.e., healthy human serum spiked with 1 fM of FGF-2, G-CSF, GM-CSF, and CX3CL1). The rapid scan rate (i.e., 0.05 s for Raman signal integration) facilitated the detection of Raman signals from FGF-2, G-CSF, GM-CSF, and CX3CL1 SERS nanotags rather than the non-target molecules present in human serum due to their low Raman cross-section. As a representative example, Supplementary Fig. 15 shows the Raman signal distribution of the FGF-2 SERS nanotags on five different spots on the pillar array obtained from the recovery test without noticeable Raman signals from other molecules. It is worth noting that unlike the solution-based DTNB labelled SERS nanotag spectra in Fig. 1d, some of the peaks at 1556 cm−1 and 1330 cm−1 in Supplementary Fig. 15 had a similar intensity, which was probably because of the different orientation of the anisotropic nanoboxes on the substrate relative to the polarisation of excitation laser49. Supplementary Tables 4 and 5 show the cytokine concentrations in healthy human serum and human serum spiked with 1 fM cytokine standards determined by digital nanopillar SERS platform, respectively. On five independent pillar arrays, the measured concentrations had the relative standard deviation (RSD) below 9.0% and the Kruskal–Wallis test showed no statistical differences among these results (p » 0.05). Overall, the observed inter-chip variation should enable accurate identification of disease progression to severe irAEs (e.g., grade 3 or 4), but may encounter some challenges in discriminating mild progressing to moderate irAEs (e.g., grade 1 or 2). The assay enabled trace determination of the four targets in simulated human serum as suggested by the target recovery rates of 80.00% to 137.00% with RSD from 16.02% to 21.80% (Supplementary Table 6). Importantly, the ability to measure reliably cytokines at attomolar levels in simulated human serum samples holds promise for detecting early changes in cytokine concentrations as predictors for the emergence of irAEs in immune checkpoint blockade treated patients.
To validate the accuracy of the digital nanopillar SERS assay, we compared the assay with commercially available ELISA kits (one kit for each cytokine tested) for cytokine quantification. To represent a potential clinical scenario of a patient developing irAEs during immune checkpoint blockade therapy, we prepared three samples with increasing concentrations of cytokines (spike in experiments into fetal bovine serum (FBS)) and subsequently analysed these samples with our digital nanopillar SERS assay and the commercial ELISA kits. FBS was used as complex sample matrix devoid of human cytokines. As the limits of detection for the ELISA kits (FGF-2 = 0.95 pM, G-CSF = 1.66 pM, GM-CSF = 1.11 pM, and CX3CL1 = 17.86 pM) were above the attomolar level, the simulated samples were prepared to suit the detection range of these kits. For the digital nanopillar SERS assay, the samples were diluted correspondingly and generated consistent results with the ELISA kits as shown in Supplementary Table 7. No statistical differences were found between ELISA and digital nanopillar SERS results based on Mann–Whitney test. Furthermore, we compared the detection of four cytokines in human serum with digital nanopillar assay and ELISA kits (Supplementary Table 8). The cytokine levels in human serum were below the limit of detection for the conventional ELISA kits, whereas their concentration was quantified by digital nanopillar SERS platform. For the human serum spiked with standard cytokines, the digital SERS platform generated similar results to ELISA without significant differences by Mann–Whitney test. Collectively, the digital nanopillar SERS platform showcased the ability to robustly and accurately quantify cytokines in complicated samples, which is significant for the prospect of dynamic correlation monitoring of irAEs in clinical samples.
Following the demonstration of the accuracy of digital nanopillar SERS platform, we tested the four cytokine levels in ten healthy people (Supplementary Table 9). These ten healthy people showed cytokine concentrations beyond the conventional ELISA capability to accurately quantify, which was consistent with previous reports9,50,51.
Dynamic correlation monitoring of irAEs in melanoma patients receiving immune checkpoint blockade treatment
Having established the feasibility of digital nanopillar SERS in simulated clinical samples, we applied the platform for longitudinally monitoring irAEs in ten melanoma patients (2–3 time points per patient, 26 samples in total) who underwent immune checkpoint blockade therapy (Supplementary Table 10). By diluting the patient samples to follow Poisson distribution, we quantified the cytokine concentration using digital nanopillar SERS platform. Based on the clinical assessments, the patients were classified into two categories: (i) developed severe irAEs (grades 3 and 4) and needed hospitalisation and dedicated treatment (Patients 1, 2, 3, 4, 5); and (ii) developed minor irAEs (grades 1 and 2) that could be managed with immunosuppressants (e.g., corticosteroids) or exhibited no symptom of irAEs (Patients 6, 7, 8, 9, 10).
As a representative case, Fig. 7 shows two cytokine profiles of a patient with severe irAEs (Patient 1) and a patient with mild irAEs (Patient 1). For Patient 1 who received ipilimumab (cytotoxic T-lymphocyte antigen-4 (CTLA-4) inhibitor) and was checked on days 7, 21, and 42, the confocal SERS images showed an increase in active pillars with the continuation of treatment (Fig. 7a–c), suggesting an elevation of the cytokine levels that could potentially trigger the severe irAEs. In agreement with the Raman images, the quantitative counting results for the four cytokines also corroborated the increase of cytokine concentrations peaking in sub-fM levels (Fig. 7d). These cytokine levels were below the limit of detection of conventional ELISA kits (pM level). Importantly, we observed significantly elevated cytokine concentrations in Patient 1 serum on day 42 compared to days 7 and 21. This patient showed the onset of grade 4 irAEs (i.e., colitis) 13 days later (day 55), consistent with the concept that higher cytokine levels correlate with increased risk of developing irAEs9. To further evaluate the utility of these four biomarkers as a signature in identifying and characterising irAEs, we analysed all the counting data from Patient 1 by applying linear discriminant analysis (LDA). As seen in Fig. 7e, the LDA successfully distinguished the data on day 42 into a separate zone from days 7 and 21, which may indicate the potential value of biomarkers in monitoring irAEs development. We further demonstrated Patient 1 LDA with the use of all combinations of two (Supplementary Fig. 16) and three cytokines (Supplementary Fig. 17). Overall, the LDA with four cytokines showed improved classification over using three or less cytokines. Interestingly, considering FGF-2/G-CSF, G-CSF/GM-CSF, and G-CSF/CX3CL1, the LDA generated similar performance to the LDA with four cytokines. To further compare the classification power of FGF-2/G-CSF, G-CSF/GM-CSF, and G-CSF/CX3CL1, and four cytokines, we performed LDA of Patient 2 (Supplementary Fig. 18), which suggested a better differentiation with the use of four cytokines. Therefore, the inclusion of all four cytokines in LDA facilitated a wider and more accurate patient sample analysis. Similarly, Patients 2, 3, 4, and 5, who manifested severe irAEs were connected with higher cytokine levels (Supplementary Fig. 19) and amelioration of irAEs symptom was witnessed with a decrease of cytokine concentrations. For these severe irAEs patients, the LDA model showed a clear discrimination in cytokine profile and this could help to identify patients at risk of irAEs (Supplementary Fig. 19).Fig. 7 Digital nanopillar SERS assay for monitoring melanoma patients during immune checkpoint therapy.
For Patient 1 who developed severe irAEs, SERS images for cytokine detection on a day 7, b day 21, c day 42, d cytokine concentration graph for fibroblast growth factor 2 (FGF-2), granulocyte colony-stimulating factor (G-CSF), granulocyte-macrophage colony-stimulating factor (GM-CSF), and fractalkine (CX3CL1). The two shorter horizontal lines denote the interquartile ranges (25th and 75th percentile) and the longer horizontal lines in between denote the median (50th percentile), and e LDA analysis, respectively. For Patient 6 who developed mild irAEs, SERS images for cytokine detection on f day 0, g day 21, h day 42, i four cytokine concentration graph, the two shorter horizontal lines denote the interquartile ranges (25th and 75th percentile) and the longer horizontal lines in between denote the median (50th percentile), and j LDA analysis, respectively. IPI ipilimumab, PEMBRO pembrolizumab; G3 grade 3, G2 grade 2; SD stable disease, PR partial response. For Patient 1, the median (interquartile range) of active pillars per scanning image of FGF-2, G-CSF, GM-CSF, CX3CL1 on day 7: 14 (11–22.5), 23 (21, 29), 12 (7.5–18), 17 (9–25.5); day 21: 30 (19–37.5), 33 (19–41), 26 (17.5–36.5), 29 (21–43); and day 42: 33 (16.5–58.5), 76 (64–128.5), 25 (14–39.5), 48 (26.5–73.5), respectively. For Patient 6, the median (interquartile range) of active pillars per scanning image of FGF-2, G-CSF, GM-CSF, CX3CL1 on day 0: 18 (16–23), 49 (31.5–56), 23 (17.5–28), 20 (14.5–27); day 21: 29 (24–33.5), 53 (46.5–70), 35 (25–46), 22 (19–29.5); and day 42: 13 (8–16.5), 44 (23.5–55.5), 10 (6.5–12.5), 30 (24–34.5), respectively. The data represented three technical replicates obtained from three chips. Nine images were acquired from each chip for cytokine counting. Statistical analysis was based on Kruskal–Wallis test followed by Dunn’s test to correct multiple comparisons (two-sided). Source data are provided in the Source Data file.
As for Patient 6 who exhibited mild grade 2 irAEs on the skin during combined ipilimumab and pembrolizumab (programmed death-1 (PD-1) inhibitor) or single ipilimumab treatments, the dynamic monitoring displayed relatively stable cytokine levels on different follow-up visits (Fig. 7). Specifically, the confocal Raman images (Fig. 7f–h) and the molecular counting (Fig. 7i) in this patient serum consistently showed no significant cytokine level alterations on the three time points (days 0, 21, 42). Under this circumstance, LDA failed to clearly classify the data into separate sections (Fig. 7j). Likewise, Patients 7, 9, and 10 possessed stable cytokine levels and were diagnosed with low grade irAEs. Meanwhile, Patient 8 who showed decreasing cytokine levels did not display signs of irAEs. LDA was not able to classify Patients 7 and 8 who had mild irAEs and did not show irAEs, but it recognised the minor difference in Patients 9 and 10 who showed grade 1 irAEs (Supplementary Fig. 20).
Overall, we found the preliminary evidence to suggest that significantly elevated cytokine levels have a strong correlation with the development and manifestation of severe irAEs, whereas stabile, low baseline, and decreasing cytokine concentration indicate mild and manageable irAEs. The relatively low concentrations of these four cytokines were below the detection sensitivities of commercially available ELISA kits (pM level), which limits their use in clinical studies. Importantly, the measurement at fM cytokine levels in clinical samples is consistent with the median concentrations of cytokine measured by using digital ELISA51. The successful demonstration of the digital nanopillar SERS platform in dynamic detection of cytokines in patient serum provides a potential approach for the future accurate early detection, characterisation, and monitoring of irAEs in clinical settings. However, it is important to note that the cytokine concentration changes are not directly correlated with the treatment response according to response evaluation criteria in solid tumours (RECIST). In our pilot study, some melanoma patients showed higher levels of cytokines compared to the healthy controls. The power of our digital nanopillar SERS platform lies in the capability to longitudinally monitor cytokines in individual patients over time.
Discussion
Despite the frequent occurrence of irAEs in immune checkpoint therapy, particularly for the combination treatment, the prediction of the emergence of irAEs remains elusive. Mounting data suggest a potential role of cytokines as predictive markers for irAE monitoring in immune checkpoint therapy9,11,13,52. Although promising, accurate quantification of these biomarkers was often not possible due to the dearth in technologies with sufficient detection sensitivity. Typically, either cytokines above the detection limit of immunosorbent assay were selected11, or relative cytokine quantification9 was performed for investigating irAEs. The former approach has the drawback of potentially excluding the low abundance cytokines of significance in irAEs. As for relative quantification9,53, the cytokine concentrations are determined by relating to a standard that had the observed median fluorescence value closest to the median of the test sample. The relative concentration, however, may fail to represent accurate cytokine levels and thus needs further exploration. Our developed digital nanopillar SERS assay offers early data suggestive of a potential approach to the above-mentioned challenges and provides the possibility to study trace amounts of a panel of cytokines in an accurate quantification manner as well as concomitantly providing attomolar level sensitivity. The current proof-of-principle approach has measured four of the potential inflammatory and/or immune toxicity-related cytokines for the prediction of emergence, characterisation, and/or quantifiable correlation with irAEs in melanoma patients.
By leveraging the narrow line width of Raman spectra, the developed digital SERS counting assay shows the ability to sensitively and simultaneously detect multiple cytokines. The adoption of the novel digital quantification mode in SERS using gold–silver alloy nanoboxes further improves the high sensitivity of SERS technology. Notably, the digital counting strategy offers an option for reproducible SERS quantification by avoiding the common Raman signal fluctuations induced by ensemble measurements. The ensemble measurement in SERS relies on the enhancement of Raman signals of molecules located in or near the “hot spots” (i.e., strong electromagnetic fields)30. Due to the random distribution and various efficiencies of “hot spots”, it can result in the discrepancies in acquired SERS intensities for inter-laboratory and even intra-laboratory tests29. To circumvent the impact of Raman intensity fluctuation on accurate quantification, we employed the digital SERS signals from the single-SERS-active nanoboxes on discrete pillar arrays to enumerate the targets and only count the “yes” or “no” signal for a robust and reproducible SERS analysis. Furthermore, the digital readout model, which regards both aggregated and single nanoparticle as a single binding event to reflect the true target number, can have a better accuracy and robustness than the intensity-based assay39.
We believe that the proposed digital nanopillar SERS assay could be used to monitor other cytokine-induced immune responses. For instance, with the outbreak of 2019 novel coronavirus (2019-nCoV), it is yet difficult to predict which infected patient will develop a strong immune response that requires hospitalisation. However, cytokines have been indicated to play a major role in the severity of immune response for critically ill patients infected with 2019-nCoV54. The specific detection of multiple cytokines at early stages of viral infection could thus potentially address this issue and help to provide the clinical care for people at the highest risk. For patients with high cytokine concentrations, the digital nanopillar SERS platform will require the sample dilution to suit Poisson distribution.
In summary, we propose a digital nanopillar SERS platform for the parallel counting of single cytokines and dynamic monitoring in the clinical context of irAE development during immune checkpoint blockade therapy. The platform achieved attomolar level sensitivity by utilising discrete pillar array compartments to hold the single cytokine and subsequently applied single-particle active nanobox-based SERS nanotags for cytokine identification and counting. The confocal Raman mapping on the pillar array offered the highest possible clinical specificity by reducing nonspecific signals and provided a “yes/no” type counting approach for reproducible Raman signal readout. The designed platform was rigorously optimised and tested in simulated clinical samples prior to the evaluation for irAE monitoring in stage IV melanoma patients receiving immune checkpoint blockade therapy. We envisaged this platform possessing the advantages of highly sensitive and multiplexing analysis capability can transit into future irAE detection methodologies after extensive validation in a large cohort of clinical samples over different time courses.
Methods
Materials
Silver nitrate (AgNO3), hydrogen tetrachloroaurate (III) trihydrate (HAuCl4·3H2O), MBA, DTNB, TFMBA, MMTAA, DSP were obtained from Sigma Aldrich. Ascorbic acid (AA) of analytical grade was purchased from MP Biomedicals. FGF-2 (223-FB), G-CSF (214-CS), GM-CSF (215-GM), CX3CL1 (365-FR) cytokines; monoclonal anti-FGF-2 (MAB233), anti-G-CSF (MAB214), anti-GM-CSF (MAB615), anti-CX3CL1 (MAB3652) antibodies; polyclonal anti-FGF-2 (AF-233), anti-G-CSF (AF-214), anti-GM-CSF (AF-215), anti-CX3CL1 (AF-365) antibodies; and FGF-2 (DY233-05), G-CSF (DY214-05), GM-CSF (DY215-05), and CX3CL1 (DY365) ELISA kits were bought from R&D Systems.
All the patient serum or plasma samples were collected at the Austin Hospital (Melbourne) under approved human ethic protocols and written informed consents were obtained from all patients before sample collection. Ethics approval was obtained from The University of Queensland Institutional Human Research Ethics Committee (approval nos. 2011001315 and 2016000876) and the following clinical assay was carried out according to the approved guidelines.
Preparation of single-particle active SERS nanotags
The preparation of SERS nanotags involved the synthesis of nanoboxes and the subsequent functionalisation with Raman reporters and antibodies. For nanobox synthesis, 45 µL of HAuCl4 (1 wt%) was added into 10 mL of ultrapure H2O (18.2 Ω cm) under magnetic stirring (800 r.p.m.) for 1 min, followed by simultaneously introducing 170 µL of AgNO3 (6 mM) and 30 µL of AA (0.1 M) into the stirring solution. Then, the formation of nanoboxes was indicated by the appearance of an apparent blue colour within 6 s and the samples were collected 1 min later by centrifuging at 600 g for 15 min.
To functionalise nanoboxes with Raman reporters and antibodies, 300 µL of nanoboxes centrifuged from 1 mL of as-prepared solution were co-incubated with one type of Raman reporters (i.e., 10 µL of DTNB, 8 µL of MBA, 10 µL of TFMBA, or 10 µL of MMTAA) and 2 µL of DSP linker for 6 h. After that, the Raman reporter and DSP functionalised nanoboxes were separated by centrifuging at 600 g for 15 min, and resuspended into 300 µL of PBS (0.1 mM). Then, 2 µg of anti-FGF-2, anti-G-CSF, anti-GM-CSF, anti-CX3CL1 antibodies were added to MBA, DTNB, TFMBA, and MMTAA labelled nanoboxes, respectively. After overnight incubation at 4 oC, the functionalised nanoboxes were purified by centrifuging at 600×g for 15 min to separate free antibodies and the final products were resuspended into 200 µL of 0.1% BSA for future use.
Fabrication of pillar arrays
The chip was made of a sensing array measuring 1 mm × 1 mm (Supplementary Fig. 1a) and consisted of 250,000 individual pillars. Each pillar was 1 µm wide, 1 µm long, and 1 µm high. The pillars were evenly spaced by 1 µm from one pillar to the next (Supplementary Fig. 1b). The pillar array was designed using Nanosuite 6.0 (Raith GmbH) and Beamer 5.9.1 (GenISys GmbH) and fabricated on a 4-inch p-type <100> silicon wafer (Bonda Technology Pte Ltd, Singapore) using electron beam lithography (EBL). The wafer accommodated 76 separate pillar arrays. Silicon wafer was first cleaned in acetone, isopropanol with sonication for 2 min each, followed by rising with deionised H2O and dehydration bake at 180 °C for 2 min. Prior to the resist coating, the wafer had undergone a further O2 plasma cleaning at 200 W for 5 min (Diener Atto, Diener Electronic GmbH). The cleaned wafer was spin-coated with two layers of polymethyl methacrylate (bottom: 495K A4 PMMA, top: 950k A4 PMMA, from MicroChemicals GmbH) using the CEE Apogee Coater (Cost Effective Equipment, LLC) at 1500 r.p.m. for 60 s each. After the coating of each layer, the wafer was baked immediately on a hot plate to prevent from intermixing of the two layers of resist. The baking time was 10 and 3 min for the bottom and top layer, respectively, at 180 °C. The thickness of the photoresist was found to be ~450 nm (top PMMA: ~250 nm, bottom PMMA: ~200 nm), characterised by white light reflectometry (FilmTek 2000M, Scientific Computing International). EBL was performed in the Raith EBPG5150 system. The patterns were exposed in EBL with an accelerated voltage of 100 kV, 150 nA of beam current (spot size ~80 nm), with step sizes of 40 nm and an electron dose of 1200 µC/cm2. The exposure time per 4-inch wafer was ~35 min, each containing 76 individual chips. After exposure, the wafer was developed in a mixture of isopropanol and methyl isobutyl ketone (3:1) for 60 s and rinsed immediately with isopropanol, followed by drying with N2. An oxygen plasma descum process, at 100 W, 60 s (Diener Atto, Diener Electronic GmbH) was carried out to remove resist residues prior to the deposition. Next, 10 nm titanium and 200 nm gold were deposited by physical vapour deposition using a Temescal FC-2000 electron beam evaporator (Ferrotec, U.S.A.). After overnight lift-off at room temperature in Remover PG (MicroChemicals GmbH, Germany), the excess material was washed off and the pillar array structure was revealed (Supplementary Fig. 1c). To create the pillar height (i.e., 1 µm), reactive ion etching (Oxford Instruments, UK) was applied for anisotropic etching of the silicon. Hereby, the deposited gold served as mask to protect the underlying silicon while the un-masked silicon was removed. Next, the wafer was coated with a protective layer of cured AZnLOF 2020 prior to wafer dicing into 76 individual sensing chips consisting of a single pillar array. Prior to use, the protective layer was washed off by consecutive washes with isopropanol and acetone and dried under a stream of nitrogen.
Pillar array functionalization
Antibody functionalisation of the gold-topped pillar array was conducted by crosslinking the antibodies to the gold surface using DSP. A solution of 5 mM DSP in dimethyl sulfoxide was pipetted onto the pillar array and incubated at room temperature for 2 h. After rinsing the pillar array with ethanol and PBS, a solution of 5 µg/mL anti-cytokine monoclonal antibody solution (100 fold dilution of antibody stock solution) in PBS was incubated overnight at 4 °C. Subsequently, the pillar array was rinsed with PBS and blocked using 1% bovine serum albumin in PBS for 1 h. Prior to use, the pillar array was rinsed with PBS. All PBS solutions were filtered through a sterile 0.22 µm syringe filter (Millex-GP, Merck, U.S.A.).
Digital nanopillar SERS profiling of cytokines
Cytokines (FGF-2, G-CSF, GM-CSF, and CX3CL1) with different concentrations in PBS (2.6 aM, 26 aM, 260 aM, and 1031 aM) were incubated with antibody functionalised pillar arrays at room temperature for 30 min, followed by washing the pillar array three times with washing buffer (0.1% BSA and 0.01% Tween 20 in PBS). SERS nanotags were then added into the pillar array for another 30 min incubation under room temperature to identify the targets. Finally, the pillar arrays were washed to remove the free SERS nanotags and were subject to confocal Raman microscope for quantification. For each sample, nine SERS images with each image has the dimension of 60 µm × 48 µm were taken on the pillar array to calculate the overall cytokine concentration.
Simulated clinical sample detection
For the recovery experiment, standard cytokines (FGF-2, G-CSF, GM-CSF, and CX3CL1) with the concentration of 1 fM were added into healthy human serum and then diluted ten times with PBS to quantify.
For the quantification of cytokines in FBS, three simulated clinical samples were prepared by titrating various concentrations of standard cytokines into 10% FBS: Sample 1 (FGF-2 = 3.64 pM, G-CSF = 3.19 pM, GM-CSF = 4.29 pM, and CX3CL1 = 28.57 fM); Sample 2 (FGF-2 = 7.28 pM, G-CSF = 6.38 pM, GM-CSF = 8.58 pM, and CX3CL1 = 571.4 fM); and Sample 3 (FGF-2 = 14.56 pM, G-CSF = 12.76 pM, GM-CSF = 17.16 pM, and CX3CL1 = 1142.8 fM). These three samples were then detected directly using the commercial ELISA kits or digital nanopillar SERS assay with a further dilution of 105, 2 × 105, and 4 × 105, respectively.
Spectroscopic ellipsometry
The antibody film thickness was measured by in-solution spectroscopic ellipsometry (M2000V JA Woollam Co., Inc. USA) using gold-coated substrates and flow cell (QSense® Ellipsometry, Biolin Scientific, Sweden). Measurements were performed at an angle of 65°. Data analysis was performed by CompleteEASE® software using a B-Spline data fit and Cauchy model to calculate the antibody film thickness.
MALDI-TOF MS
The antibody-functionalised nanopillar array chip was subjected to tryptic digest prior to analysis. Sequencing-grade trypsin was made to 50 ng/µL in 25 mM ammonium bicarbonate, and sprayed over the chip using a Bruker Imageprep instrument (Bruker, USA). After trypsin deposition, the chip was incubated in a humid environment at 40 °C for 3 h. Subsequently, the chip was sprayed with a matrix solution, 10 g/L α-cyano-4-hydroxycinnamic acid in 50% acetonitrile with 0.2% trifluoroacetic acid. Next, the chip was analysed with a Bruker Ultraflex III MALDI-TOF mass spectrometer (Bruker, USA) in positive linear mode using Flex Imaging 4.0 (Bruker, USA) with a pixel size of 60 µm. Data were collected from 2 k–30 k m/z, at a laser repetition rate of 200 Hz. Data were normalised using the root mean square approach and visualised using Flex Imaging 4.0 (Bruker, USA) and SCILS LAB 2017a software. For the SCILS LAB analysis, the data were imported using a convolution baseline subtraction, and displayed using root mean squared normalisation.
Instrumentations
SEM images of pillar arrays and nanoboxes were taken on a JEOL-7100 field emission (FE)-SEM (20 kV voltage). TEM images of nanoboxes were taken on a JEOL-2100 microscope (200 kV voltage). NTA of nanobox size distribution was performed with Malvern NanoSight NS300. UV-vis extinction spectrum of nanoboxes was performed with a Shimadzu UV-2450 spectrophotometer. Confocal Raman mapping was conducted on a WITec alpha 300 R spectrometer using 632.8 nm He–Ne laser with the power of 35 mW, a grating of 600 g/mm used with EMCCD camera, spectral resolution of 1.390 cm−1 to 2.114 cm−1, confocal pinhole size of 100 µm, 100× air objective with NA of 0.90, and 0.05 s integration time. The theoretical spot size was 857.80 nm based on the Abbe diffraction limit (i.e., d = 1.22λ/NA). The scanning area was set to have 60 µm × 48 µm with 86 points per line and 69 lines per image. For each pillar array, nine separate scanning areas were taken in total and the total active pillars were used for quantification. The SERS mapping images for counting were taken by focusing the laser on the top of the pillar surfaces. Specifically, the laser was firstly focused on the silicon substrates by obtaining the strongest silicon signals (520 cm−1) and then the 100× objective was moved up in z-axis direction of 1 µm for SERS scanning. The system was calibrated with the first-order photo peak of silicon at 520 cm−1.
Data analysis
To assign the SERS nanotag membership for DTNB, MBA, TFMBA, and MMTAA, Project Five 5.0 software from WITec was utilised to create four filters, which summed a spectral range of 40 cm−1 with the centre position at the characteristic Raman peak of each reporter and subtracted the background with a polynomial algorithm. Specifically, the filter ranges of four Raman reporters DTNB-, MBA-, TFMBA-, and MMTAA-coated SERS nanotags were (1310–1350 cm−1), (1060–1100 cm−1), (1360–1400 cm−1), and (1268–1308 cm−1), respectively. All the SERS images were analysed by using threshold intensity to determine the successful binding events. Specifically, the threshold intensity of FGF-2, G-CSF, GM-CSF, and CX3CL1 was set at 5000, 4000, 5000, and 5000, respectively. For each image, the threshold intensity was doubled-checked and adjusted based on the true Raman peaks in the spectra. Statistical analysis assuming unequal variances was conducted with Kruskal–Wallis test among three groups or Mann–Whitney test between two groups with GraphPad Prism 8.4. To control the error appropriately, we performed multiple comparisons using Dunn’s test. LDA of clinical samples was performed in R software (3.6.2) with the MASS package (7.3-52). The active pillars in SERS images were counted with Image J software.
Supplementary information
Supplementary Information
Peer Review File
Source data
Source Data
Peer review information
Nature Communications thanks Chongwen Wang and Isaac Pence for their contribution to the peer review of this work. Peer reviewer reports are available.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
The online version contains supplementary material available at 10.1038/s41467-021-21431-w.
Acknowledgements
The authors acknowledge grants received by our laboratory from the National Breast Cancer Foundation of Australia (CG-12-07) and the ARC DP (160102836 and 210103151). These grants have significantly contributed to the environment to stimulate the research described here. J.L. acknowledges support from the Australian Government Research Training Program Scholarship. A.W. and A.A.I.S. thank the National Health and Medical Research Council for funding (APP1173669 and APP1175047). A.B. is the recipient of a Fellowship from the Victorian Government Department of Health and Human Services acting through the Victorian Cancer Agency. We acknowledge the facilities, and the scientific and technical assistance, of the Australian Microscopy & Microanalysis Research Facility at the Centre for Microscopy and Microanalysis, The University of Queensland. We appreciate to receive the technical and scientific guidance from Queensland node of the Australian National Fabrication Facility (Q-ANFF) in confocal Raman mapping and spectroscopic ellipsometry measurement.
Author contributions
J.L., A.W., A.I.S., H-H.C., Y.W., A.B., P.M., and M.T. contributed to the design of the experiments and analysing the data. J.L. and A.W. performed the experiments and prepared the manuscript. All authors read, commented, and edited the manuscript, and assisted during the revision process.
Data availability
Data supporting the findings of this work are available within this paper and the supporting information files. A reporting summary of this work is available as a Supplementary file. Source data are provided with this paper.
Competing interests
The authors declare no competing interests. | Recovering | ReactionOutcome | CC BY | 33597530 | 19,690,448 | 2021-02-17 |
What was the outcome of reaction 'Pulmonary toxicity'? | A digital single-molecule nanopillar SERS platform for predicting and monitoring immune toxicities in immunotherapy.
The introduction of immune checkpoint inhibitors has demonstrated significant improvements in survival for subsets of cancer patients. However, they carry significant and sometimes life-threatening toxicities. Prompt prediction and monitoring of immune toxicities have the potential to maximise the benefits of immune checkpoint therapy. Herein, we develop a digital nanopillar SERS platform that achieves real-time single cytokine counting and enables dynamic tracking of immune toxicities in cancer patients receiving immune checkpoint inhibitor treatment - broader applications are anticipated in other disease indications. By analysing four prospective cytokine biomarkers that initiate inflammatory responses, the digital nanopillar SERS assay achieves both highly specific and highly sensitive cytokine detection down to attomolar level. Significantly, we report the capability of the assay to longitudinally monitor 10 melanoma patients during immune inhibitor blockade treatment. Here, we show that elevated cytokine concentrations predict for higher risk of developing severe immune toxicities in our pilot cohort of patients.
Introduction
The advent of immune checkpoint therapy has revolutionised the landscape of traditional cancer treatment and is believed to constitute the backbone of managing certain malignancies1–3. By capitalising on the blockade of immune checkpoint inhibitors to take the brakes off parts of the immune system, this emerging therapy has achieved great success producing long-lasting responses (e.g., 10 years or more) in a small but significant fraction of patients3–6. Nevertheless, upon the blockade of immune checkpoint molecules, the activated and potentiated immune reaction predisposes patients to a significant risk of immune-related adverse events (irAEs), which can occur in up to 80% of patients receiving immune checkpoint therapy7–9. The high incidence of irAEs, which may manifest at any time during treatment, can offset the clinical benefits, lead to premature therapy cessation, and even be life-threatening for certain patients10–12. To assist the successful implementation of immune checkpoint therapy, the use of predictive biomarkers for early identification and vigilant monitoring of irAEs is thus critical and a pressing need in avoiding or ameliorating detrimental effects and adjusting therapeutic options.
Cytokines, small signalling proteins, are promising candidates to indicate the occurrence of irAEs due to their prominent role in modulating the anti-cancer immune responses, including enhancing antigen priming, recruiting immune cells into the tumour microenvironment, and upregulating certain immune checkpoint molecules9,13,14. Particularly, excessive cytokine secretion has been implicated in severe inflammation as a major constituent leading to irAEs. For example, the overproduction of fibroblast growth factor 2 (FGF-2)15–18, granulocyte colony-stimulating factor (G-CSF)19, granulocyte-macrophage colony-stimulating factor (GM-CSF)20, and fractalkine (CX3CL1)21 have been found to participate in immune-related inflammatory disease (e.g., rheumatoid arthritis, autoimmune gastritis, and Crohn’s disease). These inflammatory cytokines have recently been reported to indicate irAEs for melanoma patients who underwent immune checkpoint therapy9. The clinical deployment of cytokine analysis for irAE monitoring is challenging and requires a technology that can (i) determine the selected cytokines with great sensitivity9, especially at the onset of irAE development, where the cytokine concentrations are likely to be the lowest; as well as (ii) simultaneously detect a panel of cytokines to reflect the complex interplay of cytokine signalling pathways22 and the variable irAE symptoms among patients.
Conventional cytokine analyses such as immunosorbent assays have limited clinical applicability for irAE assessment due to their limited capacity to detect low cytokine concentrations in blood as well as for assessing a panel of cytokines in a single sample simultaneously. Recently, advances in micro/nanomaterial-based systems have provided a promising suite of technologies that improve the conventional assays by overcoming the above limitations23,24. Encouragingly, the unique advantages of micro/nanomaterial-based systems convey an attractive option for cytokine analysis with the desired results of high sensitivity and multiplexing. The high specific surface area of these miniaturised materials increases mass transfer subsequently enhancing the interaction with target molecules and thus improving the detection sensitivity25. The capabilities of micro/nanomaterial fabrication techniques permit individually separated compartments sufficiently discrete to hold single molecules and hence encompasses a promising strategy for counting assays that can further push the sensitivity of the traditional assays24,26,27. Moreover, the physicochemical properties of nanostructured materials can be exploited to simultaneously label multiple targets (e.g., various spectral signatures) for high-throughput parallel measurements28–30. Therefore, by combining the potential of micro/nanomaterial systems with the need for sensitive irAE monitoring, we have developed a platform for sensitive and multiplex cytokine counting analysis.
Combining the use of (a) discrete single cytokine nanopillar array chip with discretely separated compartments, (b) control of target concentrations to follow a Poisson distribution, and (c) the recognition of target by single-particle active surface-enhanced Raman scattering (SERS) nanotags with a confocal Raman microscope allows accurate and in situ counting of a multi-cytokine panel (FGF-2, G-CSF, GM-CSF, and CX3CL1). Different from the fluorescence-based digital counting strategies24,26, the strikingly narrow spectral peaks of SERS (~1–2 nm) in comparison to fluorescence (~50 nm) makes this platform intrinsically ideal for multiplexed cytokine analysis28.
In this work, we present a digital nanopillar SERS platform that enables the specific cytokine quantification down to attomolar levels and the application in melanoma patients receiving immune checkpoint blockade therapy. Beyond the capability to predict irAEs in melanoma patients receiving therapy, the digital nanopillar SERS assay could potentially be extended to other cytokine-associated immune responses such as excessive immune activation due to viral or bacterial infections (such as COVID-19).
Results
Digital nanopillar SERS platform for parallel profiling of single cytokine
Our concept of digital nanopillar SERS platform for cytokine analysis relies on Rayleigh criterion separation, probability-driven Poisson distribution, single-particle active SERS nanotags, and confocal SERS mapping (Fig. 1). To precisely fabricate the pillar array, we opted to use an electron beam lithographic approach to write the array into a photon-sensitive material followed by physical vapour deposition of gold to create the gold-topped pillars, and selectively reactive ion etching to reveal the pillar structure (Supplementary Fig. 1). The nanopillar array chip consisted of 250,000 individual pillars. As shown in the scanning electron microscope (SEM) image of Fig. 1a, the cubic nanopillars have an edge-to-edge width of 1000 nm and are evenly distributed at 1000 nm intervals to suit the lateral Raman microscope resolution (~1000 nm) that fulfils the Rayleigh criterion separation required to acquire a single SERS spectrum from each pillar without spectral overlap from adjacent pillars.Fig. 1 Digital single-molecule nanopillar surface-enhanced Raman scattering (SERS) platform for parallel counting of four types of cytokines.
SEM images of a pillar array side view, b nanoboxes, and c a single nanobox on the top of a pillar; d SERS spectra of nanoboxes conjugated with 5,5-dithiobis (2-nitrobenzoic acid) (DTNB), 4-mercaptobenzoic acid (MBA), 2,3,5,6-tetrafluoro-4-mercaptobenzoic acid (TFMBA), or 2‐mercapto‐4‐methyl‐5‐thiazoleacetic acid (MMTAA) Raman reporters; e workflow for multiplex counting of cytokines, including fibroblast growth factor 2 (FGF-2), granulocyte colony-stimulating factor (G-CSF), granulocyte-macrophage colony-stimulating factor (GM-CSF), and fractalkine (CX3CL1). Data from one independent experiment.
By using specific gold-thiol chemistry with the linker molecule dithiobis (succinimidyl propionate) (DSP), the gold-topped pillars were selectively functionalised with target recognition antibodies (anti-FGF-2, anti-G-CSF, anti-GM-CSF, and anti-CX3CL1) and acted as the small compartments to capture and confine the individual cytokine. Upon DSP binding on the gold-topped pillars through gold-thiol bond, DSP uses N-hydroxysuccimide (NHS) ester to react with the amine groups of the antibodies31,32. The successful antibody conjugation on gold-topped pillar surfaces was confirmed by matrix-assisted laser desorption ionisation-time of flight mass spectrometry (MALDI-TOF MS) (Supplementary Fig. 2), which showed high molecular weight fragments derived from antibodies. Furthermore, spectroscopic ellipsometry was utilised to estimate the antibody density on pillar surfaces. Based on the obtained film thickness of 18.5 nm, the calculated antibody surface density was 5.5 mg/m2 using the Cuypers model33, which was in agreement with the reported antibody density on substrate surfaces34. Though these characterisations indicated the presence of antibodies on the pillar array, it was not possible to assess the exact distribution of the four structurally related (same immunoglobulin G family) antibodies on a single pillar with an area of 1 µm2. As an advantage of the digital read-out with a large redundancy of pillars, it is not essential to have all four types of antibodies equally distributed on a single pillar for the assay to work. The combined surface area of all antibody-conjugated pillars provides an excess of cytokine binding sites, which maximises successful cytokine capture within the pillar array. Supplementary Fig. 3 shows the SERS mapping images of an equimolar cytokine solution (1031 aM) that provided a similar signal count for the FGF-2, GM-CSF, G-CSF, and CX3CL1 SERS nanotags, indicating a required distribution of four kinds of antibodies conjugated to the array of pillars. Instrumental to the digital counting of cytokines, we controlled the target concentration based on the principle of Poisson distribution where the ratio of cytokine molecules to pillar number was <1:10, ensuring a 99% probability that there was either one cytokine molecule or zero per pillar24. At a ratio of 1:10, 10% of all pillars were occupied or activated with the cytokine molecules.
Following the capture of cytokines on nanopillars, SERS nanotags were applied to recognise the captured cytokines. The preparation of SERS nanotags was performed by the co-conjugating of Raman reporter and target antibody onto gold–silver alloy nanoboxes. Specifically, an average size of 80 nm gold–silver alloy nanoboxes were firstly synthesised using a rapid and aqueous phase approach35 as indicated in the SEM image in Fig. 1b. Supplementary Fig. 4a, b shows the transmission electron microscope (TEM) image of the nanoboxes with the hollow inner structure and a wall thickness of around 15 nm. Nanoparticle tracking analysis (NTA), which allows the tracking and detection of single particles, shows the nanoboxes have a mode size of 77 nm (D10 = 67.6 nm and D90 = 110.6 nm) (Supplementary Fig. 4c). UV-vis extinction spectroscopy demonstrates the nanoboxes possess a surface plasmon resonance (SPR) peak at 610 nm (Supplementary Fig. 4d). The resonance frequency of the nanoboxes enables a more sensitive signal readout with 632.8 nm laser excitation29, which also has a higher Raman scattering efficiency than 785 nm laser. Thereafter, four types of Raman reporters (5,5-dithiobis (2-nitrobenzoic acid) (DTNB), 4-mercaptobenzoic acid (MBA), 2,3,5,6-tetrafluoro-4-mercaptobenzoic acid (TFMBA), and 2‐mercapto‐4‐methyl‐5‐thiazoleacetic acid (MMTAA)) that generate unique Raman signals (1330 cm−1, 1080 cm−1, 1380 cm−1, and 1288 cm−1) were coupled with their corresponding detection antibodies onto nanoboxes as specific SERS nanotags for identification of FGF-2, G-CSF, GM-CSF, and CX3CL1, respectively. As shown in Fig. 1d, the four SERS nanotags provide the strong and non-overlapping Raman signals, which facilitates the multiplexing analysis of four cytokines. The assignment of the major Raman peaks from the four Raman reporters was summarised into Supplementary Table 1. To evaluate the SERS enhancement property of the nanoboxes, we calculated the enhancement factor (EF) of the four Raman reporters on the nanoboxes. Based on the labelled characteristic peaks in Supplementary Fig. 5, the calculated EFs of DTNB, MBA, TFMBA, and MMTAA were 8.14 × 106, 1.46 × 107, 4.01 × 107, and 3.26 × 107, respectively. The obtained EFs were higher than the reported spherical gold nanoparticles and pure silver nanocubes36 and comparable to the reported hollow nanocubes37, illustrating the high SERS property of the nanoboxes. To investigate the SERS nanotag stability, we monitored the Raman signal intensity over 7 days. As shown in Supplementary Fig. 6, Raman signal intensity variations are less than 5% in the SERS spectra, suggesting the good stability of the prepared SERS nanotags.
The following SERS mapping generated false-colour images for counting single cytokine molecules. Under the Raman microscope, the pillar array was visualised as a blue and black grid by representing the specific Raman shifts corresponding to the silicon signals (520 cm−1), in which the blue colour was assigned to silicon signals showing silicon substrates and the black colour indicated the gold-topped pillars because of the lack of silicon signals. The representation of the four colours of the SERS nanotags (red, green, purple, and cyan) on the gold-topped pillars (i.e., black) reflected cytokine molecule occupation (FGF-2, G-CSF, GM-CSF, and CX3CL1). Elevating the sensing area (or gold-topped pillars) from the silicon substrate was selected as a strategy to minimise the false-positive events. By using the confocal function of the Raman microscope, the laser was selectively focused on the gold-topped pillars, thus largely removing the background signals from potentially non-specifically adsorbed SERS nanotags on the silicon substrate. Finally, the specific SERS nanotag signals present or absent on the gold-topped pillars were counted and represented as percentage of active pillars used for total cytokine quantification. For statistical calculations, SERS mapping was applied for scanning 6480 pillars. This digital counting mode, therefore, has the potential to reach the ultimate sensitivity of single molecule cytokine detection.
Demonstration of the single-particle SERS activity of gold–silver alloy nanoboxes
The successful implementation of digital nanopillar SERS assay necessitates the use of single-particle active plasmonic nanostructures that give a clearly detectable signal for each of the single cytokine binding event. The single-particle SERS detection sensitivity is essential to this assay development as the single-particle inactive plasmonic nanoparticles (e.g., spherical gold nanoparticles)38 would unavoidably result in an underestimate of cytokine concentration.
We evaluated the single-particle SERS activity of the prepared anisotropic nanoboxes by acquiring the signals from the individual nanoboxes that were labelled with DTNB reporters. The use of DTNB, a non-resonant Raman reporter, guaranteed the Raman signal enhancement was solely contributed from the nanobox-generated electromagnetic field. As seen in the SEM image (Fig. 2a), two clearly separated DTNB-labelled nanoboxes were deposited on the silicon wafer (highlighted in red circles). The corresponding Raman image (Fig. 2b) displayed several bright Raman spots originating from these individual nanoboxes. The elongated bright Raman spots in the SERS mapping image were probably caused by the slight aggregation of several nanoboxes during sample preparation processes (e.g., centrifugation)38, which was difficult to visually resolve in the SEM image (Fig. 2a). However, unlike the intensity-based assay, the aggregated nanoboxes as SERS nanotags to target cytokine will not skew the digital readout result, because each cytokine will occupy a single pillar following Poisson distribution and both aggregated and individual nanoparticles are regarded as a single binding event that truly reflects the target number39. We then acquired the SERS spectra from two individual SERS nanoboxes (Fig. 2c, (1) and (2)) and two separate spots of bare silicon (Fig. 2c, (3) and (4)). The presence of nanoboxes showed the characteristic Raman signal at 1330 cm−1 from DTNB, whereas the silicon spectra (3) and (4) lacked the specific peak. This observation demonstrated the single-particle SERS activity of nanoboxes, which was largely attributed to the enhanced electromagnetic fields of nanoboxes on specific regions (e.g., tips and corners)40,41 and thereby facilitated the sensitive and accurate counting of cytokines. Based on the acquired SERS mapping image, the median (interquartile range) of the DTNB peak intensity (1330 cm−1) in the presence and absence of nanoboxes were 183.03 a.u. (149.48–243.35 a.u.) and 18.07 a.u. (15.51–23.12 a.u.), respectively. Furthermore, the mean ± standard deviation of the DTNB peak intensity with nanoboxes (213.41 ± 85.03 a.u.) distinguished clearly from the position without nanoboxes (18.79 ± 6.01 a.u.), which demonstrated the feasibility of correctly identifying the presence of nanoboxes.Fig. 2 Demonstration of the single-particle SERS activity of DTNB-labelled nanoboxes.
a SEM image and b corresponding SERS mapping image of DTNB-labelled nanoboxes on a silicon substrate; c representative SERS spectra of numbered locations indicated in a and b. The red dotted line shows the characteristic peak at 1330 cm−1 from DTNB. Data from one independent experiment. Source data are provided in the Source Data file.
Fig. 3 Study of confocal height on Raman signal intensity.
SERS mapping of FGF-2 SERS nanotags on the silicon substrate with changing confocal height of a 0 nm, b 500 nm, c 1000 nm, and d 1500 nm; selected Raman spectra obtained from e red circles and f blue circles of SERS images. Red dotted lines in e and f indicate peak signal at 1330 cm−1 from DTNB. Data from one independent experiment. Source data are provided in the Source Data file.
Fig. 4 Specificity of digital nanopillar SERS platform for FGF-2 cytokine detection.
Representative confocal SERS images in the presence of a target FGF-2 (1031 aM), and negative controls with non-target controls b G-CSF (1031 aM), c GM-CSF (1031 aM), d CX3CL1 (1031 aM), and e PBS. The median (interquartile range) of active pillars per scanning image for FGF-2, G-CSF, GM-CSF, CX3CL1, and PBS was 72 (63.5–76.75), 1.5 (1.5–2), 2 (1–4), 0.5 (0–1.25), and 1 (1–1.75), respectively. Data from one independent experiment.
Optimisation of digital nanopillar SERS platform for cytokine detection
The reliable detection of single cytokine molecules by the digital nanopillar SERS platform depends on the geometric features of the pillar array (i.e., pillar height, cross-section area of pillar) and assay conditions (i.e., incubation time for sample and SERS nanotags).
We first sought to investigate the effect of pillar height on Raman signal intensity to differentiate signals from non-specifically bound SERS nanotags on the silicon substrate and specifically bound SERS nanotags on the gold-topped pillars. FGF-2 SERS nanotags were randomly deposited on the silicon substrate to mimic the non-specific binding scenario and the Raman mappings were perfomed by moving the objective along the z-axis direction with different heights (0 nm, 500 nm, 1000 nm, and 1500 nm) to compare the signal intensity. In Fig. 3, a–d show the false-colour SERS images and e, f the corresponding SERS spectra with characteristic DTNB reporter peak at 1330 cm−1 acquired from the circled spots in a–d. At the height of 0 nm where the SERS nanotags were in focus, we noticed bright Raman spots (Fig. 3a) and strong Raman signals (black line in Fig. 3e, f). With increasing z heights to 500 nm and 1000 nm, the Raman spots decreased (Fig. 3b, c) and the signal intensity weakened/disappeared (red and blue lines in Fig. 3e, f) as the nanoboxes became increasingly out of focus. A further increase to 1500 nm did not remarkably weaken Raman signals compared to the height of 1000 nm (Fig. 3d). Hence, for the fabrication of the pillar array chip, we selected a pillar height of 1000 nm to greatly reduce the potential interference from non-specific signals.
The cross-section area of the pillars provides the space for cytokine binding and labelling with SERS nanotags. To study the effect of pillar cross-section area, we fabricated array chips with pillars of various widths (250 nm, 500 nm, and 1000 nm) (Supplementary Fig. 7a–c) and functionalised with anti-FGF-2 antibody. Each chip consisted of 250,000 individual pillars. We then analysed a sample that contained ~25,000 molecules of FGF-2 (40 µL, 1031 aM), which should result in 10% active pillars (ratio FGF-2: pillars of 0.1). As seen in Supplementary Fig. 7d–f, an increasing pillar cross-section area results in a higher fraction of active pillars. In reference to the expected active pillar percentage (10%), the 250 nm and 500 nm pillar arrays produced lower active pillars (2% and 6%), which suggested a significant loss of target recognition by SERS nanotags. For the 1000 nm wide pillars, the active pillar percentage was 11%, close to the nominal value of 10%. We further tested a sample with 260 aM FGF-2 (i.e., 2.5% active pillars) on the pillar array chips with 250, 500, and 1000 nm pillar widths. The capture efficiency of these three chips was summarised in Supplementary Table 2. In comparison to the pillar array of 250 nm and 500 nm sizes, the 1000 nm provided an improved capture efficiency. As the accessible target recognition surface area per pillar increases, it can possibly promote the thermodynamics and kinetics for higher surface binding and capture efficiency25. Consequently, the 1000 nm pillar array was adopted in the subsequent experiments.
An optimal incubation time of cytokine and SERS nanotags on the pillar array can shorten the operation time and reduce the potential risk of nonspecific binding that could lead to false-positive counting. We thus studied the effect of incubation time of cytokine with SERS nanotags for 30 to 90 min in a solution of 1031 aM FGF-2 and FGF-2 SERS nanotags. As suggested by Supplementary Fig. 8, the increase in incubation time gives rise to a higher proportion of active pillars. In comparison with the theoretical active pillar percentage (10%), both 30 min and 60 min incubation time were able to provide a desirable active pillar percentage (11% and 13%, respectively). A longer incubation time (90 min), however, reported an active pillar percentage (20%) two times higher than the theoretical, indicating the occurrence of nonspecific absorption of SERS nanotags on the pillar array chip. Thus, we selected 30 min incubation time for further digital nanopillar SERS measurements.
Specificity of the digital nanopillar SERS platform for cytokine detection
Accurate and reliable recognition of the specific target is essential for cytokine quantification in clinical samples. To demonstrate the detection specificity of the digital nanopillar SERS assay, we prepared an anti-FGF-2 antibody functionalised pillar array and measured samples containing target FGF-2 cytokine and controls (G-CSF, GM-CSF, CX3CL1, and PBS). It was observed that only the presence of FGF-2 activated significant amounts of pillars whereas the negative controls only generated negligible active pillars (Fig. 4), indicating the high specificity for FGF-2 detection. Similarly, we studied the specific detection of G-CSF, GM-CSF, and CX3CL1, as shown in Supplementary Figs. 9–11, in which the typical Raman images displayed high proportions of active pillars in the presence of specific targets but not for the negative controls.
To further investigate the specificity of binding between SERS nanotag and antibody-functionalised pillar, we performed SEM analysis to “closely” inspect the pillar array for the presence or absence of SERS nanotags. As a representative model, we selected to image FGF-2 SERS nanotags on the anti-FGF-2-functionalised pillar array after sample incubation with FGF-2 cytokine and non-target controls (Fig. 5). As expected, we observed the cubic nanoparticles on the top of pillars in the presence of FGF-2 due to the successful recognition of SERS nanotags. On the contrary, pillar arrays did not display a significant number of FGF-2 SERS nanotags with non-target controls. Consequently, the consistent Raman and SEM data demonstrate the capability of the assay for specific target cytokine counting. The ability to selectively identify these four cytokines in the designed assay is critically important for their usage in clinical samples.Fig. 5 Specificity of the digital nanopillar SERS platform for FGF-2 cytokine detection.
Representative SEM images of pillar array incubated with FGF-2 SERS nanotags in the presence of a, b FGF-2 (1031 aM), c G-CSF (1031 aM), d GM-CSF (1031 aM), e CX3CL1 (1031 aM), and f PBS. The red circles highlight the existence of SERS nanotags. Panel b is the magnified SEM image of the red-highlighted section in a. It is noted that nanofabrication debris on the sidewall of the pillars can also be seen. Data from one independent experiment.
Sensitivity of the digital nanopillar SERS platform for cytokine detection
As there is typically low abundance of cytokines in clinical samples, the technique based on cytokine detection is expected to possess sufficient sensitivity to reliably assess irAEs9. To investigate the sensitivity and dynamic detection range of the digital nanopillar SERS assay, we firstly titrated the designated concentration of one target cytokine (FGF-2) on the pillar array chip with 250,000 pillars. To comply with the Poisson distribution, the upper number of cytokine molecules in the sample is 25,000 which should result in 10% activated pillars. Based on this upper molecule number, we were motivated to challenge the assay by serially diluting the number of cytokine molecule in the sample from 25,000 (1031 aM), 6305 (260 aM), 631 (26 aM), and 63 (2.6 aM). As suggested by the Raman images in Supplementary Fig. 12 and with the decrease in FGF-2 molecules, the percentage of active pillars decreased correspondingly from 9.39% for 1031 aM, 6.59% for 260 aM, 1.12% for 26 aM, and 0.62% for 2.6 aM, showing a strong correlation that facilitates quantitative cytokine analysis.
Subsequently, we were interested in exploring the multiplexing capability of SERS to investigate the digital nanopillar SERS assay’s dynamic range for the simultaneous quantification of all studied cytokines. As the targets independently follow Poisson distribution, each of the cytokine was separately controlled to activate less than 10% pillars. The specific SERS nanotags provided unique signals for each cytokine that was visualised in the false-colour SERS images by a different colour, thereby enabling in situ and simultaneous cytokine detection. As suggested by the confocal SERS images in Fig. 6, an increase in cytokine concentration corresponded with a higher percentage of active pillars. To facilitate quantitative measurements of the cytokines, we calculated the logarithmic transformation of the percentage of active pillars versus cytokine concentration (Supplementary Fig. 13) confirming the strong statistical and potentially clinically relevant correlation (coefficient of determination (R2) >0.97) observed in the SERS images.Fig. 6 Sensitivity for the simultaneous detection of four cytokines.
Representative confocal SERS images of fibroblast growth factor 2 (FGF-2), granulocyte colony-stimulating factor (GM-CSF), granulocyte colony-stimulating factor (G-CSF), and fractalkine (CX3CL1) with the concentration of a 2.6 aM, b 26 aM, c 260 aM, d 1031 aM. Colour scale bars indicate Raman intensities from 5,5-dithiobis (2-nitrobenzoic acid) (DTNB), 4-mercaptobenzoic acid (MBA), 2,3,5,6-tetrafluoro-4-mercaptobenzoic acid (TFMBA), or 2‐mercapto‐4‐methyl‐5‐thiazoleacetic acid (MMTAA). The median (interquartile range) of active pillars per scanning image of FGF-2, G-CSF, GM-CSF, CX3CL1 for 2.6 aM: 3 (1.5–3), 1 (1–2), 2 (1–3), 2 (1–3); 26 aM: 8 (5.5–10), 10 (9–13), 7 (6–10), 8 (6–10); 260 aM: 40 (36–48), 40 (35–52), 39 (35–50), 37 (36–49); and 1031 aM: 79 (61.5–97), 78 (72–87.5), 88 (68.5–97), 79 (64–95), respectively. Data represents one experiment from three independent tests.
To further investigate the multiplexing quantification performance of the digital nanopillar SERS assay in human serum, we spiked standard cytokines in human serum and tested the dynamic range. Supplementary Fig. 14 shows the linear relationship curves for the four targets. Because of the more complicated sample matrix composition in human samples, the lowest detectable cytokine concentration (5.2 aM) was higher than the PBS solution (2.6 aM).
At a cytokine to pillar ratio of 1:10, we studied the probability of each pillar being occupied by different molecule numbers. To experimentally investigate the number of molecules on a single pillar, we analysed a cytokine mixture that contained all four target cytokines at equal concentration (i.e., ~6250 molecules per cytokine). To visualise and count molecule binding events on a single pillar, we labelled the captured cytokines with the four SERS nanotags that provide clearly distinguishable signals. Under Poisson distribution, the likelihood of having two or more molecules on a single pillar is <0.45% (Supplementary Table 3), which underlies the digital counting principle24. Compared to the theoretical Poisson distribution, the experiment data reported a close but slightly higher value, which was probably due to minor non-specific binding of SERS nanotags on the pillars.
The high sensitivity (attomolar level) of the digital nanopillar SERS assay can be ascribed to the following factors: the digital counting strategy, the single-particle SERS activity of the nanoboxes, and the use of pillars to suit confocal Raman mapping that efficiently excludes false-positive signals. Commercially available methods with potential for trace analysis of cytokines include the single-molecule ELISA Simona by Quanterix and electrochemical luminescence assay42,43 by Meso Scale Discovery. Compared to these two methods, the developed digital nanopillar SERS platform enabled in situ multiplexed detection of four cytokines with comparable sensitivity. Unlike the issues of photo bleaching and poor multiplexing analysis often encountered in fluorescence44 and luminescence assays45, SERS provides the advantage of high multiplexing (e.g., 31-plex)46–48 with the narrow Raman linewidth and high photo stability of the Raman reporters. In addition, this digital nanopillar SERS platform can provide more accurate quantification of cytokines by reducing the false-positive signals with the confocal setting, thus eventually help clinicians to monitor irAEs during immune checkpoint therapy. The highly sensitive readout for multiple targets also indicated the capability of this assay for cytokine detection to assess irAEs in clinically relevant samples.
Evaluation of digital nanopillar SERS platform on simulated patient samples
The detection of trace concentrations of cytokines in serum samples is difficult because plasma samples contain a high abundance of non-target molecules (e.g., serum albumin and other proteins) that can potentially interfere with cytokine detection and lead to inaccurate clinical results. To evaluate the capability of the digital nanopillar SERS assay in accurately counting single cytokine molecules, we opted to perform a recovery test in simulated patient plasma samples (i.e., healthy human serum spiked with 1 fM of FGF-2, G-CSF, GM-CSF, and CX3CL1). The rapid scan rate (i.e., 0.05 s for Raman signal integration) facilitated the detection of Raman signals from FGF-2, G-CSF, GM-CSF, and CX3CL1 SERS nanotags rather than the non-target molecules present in human serum due to their low Raman cross-section. As a representative example, Supplementary Fig. 15 shows the Raman signal distribution of the FGF-2 SERS nanotags on five different spots on the pillar array obtained from the recovery test without noticeable Raman signals from other molecules. It is worth noting that unlike the solution-based DTNB labelled SERS nanotag spectra in Fig. 1d, some of the peaks at 1556 cm−1 and 1330 cm−1 in Supplementary Fig. 15 had a similar intensity, which was probably because of the different orientation of the anisotropic nanoboxes on the substrate relative to the polarisation of excitation laser49. Supplementary Tables 4 and 5 show the cytokine concentrations in healthy human serum and human serum spiked with 1 fM cytokine standards determined by digital nanopillar SERS platform, respectively. On five independent pillar arrays, the measured concentrations had the relative standard deviation (RSD) below 9.0% and the Kruskal–Wallis test showed no statistical differences among these results (p » 0.05). Overall, the observed inter-chip variation should enable accurate identification of disease progression to severe irAEs (e.g., grade 3 or 4), but may encounter some challenges in discriminating mild progressing to moderate irAEs (e.g., grade 1 or 2). The assay enabled trace determination of the four targets in simulated human serum as suggested by the target recovery rates of 80.00% to 137.00% with RSD from 16.02% to 21.80% (Supplementary Table 6). Importantly, the ability to measure reliably cytokines at attomolar levels in simulated human serum samples holds promise for detecting early changes in cytokine concentrations as predictors for the emergence of irAEs in immune checkpoint blockade treated patients.
To validate the accuracy of the digital nanopillar SERS assay, we compared the assay with commercially available ELISA kits (one kit for each cytokine tested) for cytokine quantification. To represent a potential clinical scenario of a patient developing irAEs during immune checkpoint blockade therapy, we prepared three samples with increasing concentrations of cytokines (spike in experiments into fetal bovine serum (FBS)) and subsequently analysed these samples with our digital nanopillar SERS assay and the commercial ELISA kits. FBS was used as complex sample matrix devoid of human cytokines. As the limits of detection for the ELISA kits (FGF-2 = 0.95 pM, G-CSF = 1.66 pM, GM-CSF = 1.11 pM, and CX3CL1 = 17.86 pM) were above the attomolar level, the simulated samples were prepared to suit the detection range of these kits. For the digital nanopillar SERS assay, the samples were diluted correspondingly and generated consistent results with the ELISA kits as shown in Supplementary Table 7. No statistical differences were found between ELISA and digital nanopillar SERS results based on Mann–Whitney test. Furthermore, we compared the detection of four cytokines in human serum with digital nanopillar assay and ELISA kits (Supplementary Table 8). The cytokine levels in human serum were below the limit of detection for the conventional ELISA kits, whereas their concentration was quantified by digital nanopillar SERS platform. For the human serum spiked with standard cytokines, the digital SERS platform generated similar results to ELISA without significant differences by Mann–Whitney test. Collectively, the digital nanopillar SERS platform showcased the ability to robustly and accurately quantify cytokines in complicated samples, which is significant for the prospect of dynamic correlation monitoring of irAEs in clinical samples.
Following the demonstration of the accuracy of digital nanopillar SERS platform, we tested the four cytokine levels in ten healthy people (Supplementary Table 9). These ten healthy people showed cytokine concentrations beyond the conventional ELISA capability to accurately quantify, which was consistent with previous reports9,50,51.
Dynamic correlation monitoring of irAEs in melanoma patients receiving immune checkpoint blockade treatment
Having established the feasibility of digital nanopillar SERS in simulated clinical samples, we applied the platform for longitudinally monitoring irAEs in ten melanoma patients (2–3 time points per patient, 26 samples in total) who underwent immune checkpoint blockade therapy (Supplementary Table 10). By diluting the patient samples to follow Poisson distribution, we quantified the cytokine concentration using digital nanopillar SERS platform. Based on the clinical assessments, the patients were classified into two categories: (i) developed severe irAEs (grades 3 and 4) and needed hospitalisation and dedicated treatment (Patients 1, 2, 3, 4, 5); and (ii) developed minor irAEs (grades 1 and 2) that could be managed with immunosuppressants (e.g., corticosteroids) or exhibited no symptom of irAEs (Patients 6, 7, 8, 9, 10).
As a representative case, Fig. 7 shows two cytokine profiles of a patient with severe irAEs (Patient 1) and a patient with mild irAEs (Patient 1). For Patient 1 who received ipilimumab (cytotoxic T-lymphocyte antigen-4 (CTLA-4) inhibitor) and was checked on days 7, 21, and 42, the confocal SERS images showed an increase in active pillars with the continuation of treatment (Fig. 7a–c), suggesting an elevation of the cytokine levels that could potentially trigger the severe irAEs. In agreement with the Raman images, the quantitative counting results for the four cytokines also corroborated the increase of cytokine concentrations peaking in sub-fM levels (Fig. 7d). These cytokine levels were below the limit of detection of conventional ELISA kits (pM level). Importantly, we observed significantly elevated cytokine concentrations in Patient 1 serum on day 42 compared to days 7 and 21. This patient showed the onset of grade 4 irAEs (i.e., colitis) 13 days later (day 55), consistent with the concept that higher cytokine levels correlate with increased risk of developing irAEs9. To further evaluate the utility of these four biomarkers as a signature in identifying and characterising irAEs, we analysed all the counting data from Patient 1 by applying linear discriminant analysis (LDA). As seen in Fig. 7e, the LDA successfully distinguished the data on day 42 into a separate zone from days 7 and 21, which may indicate the potential value of biomarkers in monitoring irAEs development. We further demonstrated Patient 1 LDA with the use of all combinations of two (Supplementary Fig. 16) and three cytokines (Supplementary Fig. 17). Overall, the LDA with four cytokines showed improved classification over using three or less cytokines. Interestingly, considering FGF-2/G-CSF, G-CSF/GM-CSF, and G-CSF/CX3CL1, the LDA generated similar performance to the LDA with four cytokines. To further compare the classification power of FGF-2/G-CSF, G-CSF/GM-CSF, and G-CSF/CX3CL1, and four cytokines, we performed LDA of Patient 2 (Supplementary Fig. 18), which suggested a better differentiation with the use of four cytokines. Therefore, the inclusion of all four cytokines in LDA facilitated a wider and more accurate patient sample analysis. Similarly, Patients 2, 3, 4, and 5, who manifested severe irAEs were connected with higher cytokine levels (Supplementary Fig. 19) and amelioration of irAEs symptom was witnessed with a decrease of cytokine concentrations. For these severe irAEs patients, the LDA model showed a clear discrimination in cytokine profile and this could help to identify patients at risk of irAEs (Supplementary Fig. 19).Fig. 7 Digital nanopillar SERS assay for monitoring melanoma patients during immune checkpoint therapy.
For Patient 1 who developed severe irAEs, SERS images for cytokine detection on a day 7, b day 21, c day 42, d cytokine concentration graph for fibroblast growth factor 2 (FGF-2), granulocyte colony-stimulating factor (G-CSF), granulocyte-macrophage colony-stimulating factor (GM-CSF), and fractalkine (CX3CL1). The two shorter horizontal lines denote the interquartile ranges (25th and 75th percentile) and the longer horizontal lines in between denote the median (50th percentile), and e LDA analysis, respectively. For Patient 6 who developed mild irAEs, SERS images for cytokine detection on f day 0, g day 21, h day 42, i four cytokine concentration graph, the two shorter horizontal lines denote the interquartile ranges (25th and 75th percentile) and the longer horizontal lines in between denote the median (50th percentile), and j LDA analysis, respectively. IPI ipilimumab, PEMBRO pembrolizumab; G3 grade 3, G2 grade 2; SD stable disease, PR partial response. For Patient 1, the median (interquartile range) of active pillars per scanning image of FGF-2, G-CSF, GM-CSF, CX3CL1 on day 7: 14 (11–22.5), 23 (21, 29), 12 (7.5–18), 17 (9–25.5); day 21: 30 (19–37.5), 33 (19–41), 26 (17.5–36.5), 29 (21–43); and day 42: 33 (16.5–58.5), 76 (64–128.5), 25 (14–39.5), 48 (26.5–73.5), respectively. For Patient 6, the median (interquartile range) of active pillars per scanning image of FGF-2, G-CSF, GM-CSF, CX3CL1 on day 0: 18 (16–23), 49 (31.5–56), 23 (17.5–28), 20 (14.5–27); day 21: 29 (24–33.5), 53 (46.5–70), 35 (25–46), 22 (19–29.5); and day 42: 13 (8–16.5), 44 (23.5–55.5), 10 (6.5–12.5), 30 (24–34.5), respectively. The data represented three technical replicates obtained from three chips. Nine images were acquired from each chip for cytokine counting. Statistical analysis was based on Kruskal–Wallis test followed by Dunn’s test to correct multiple comparisons (two-sided). Source data are provided in the Source Data file.
As for Patient 6 who exhibited mild grade 2 irAEs on the skin during combined ipilimumab and pembrolizumab (programmed death-1 (PD-1) inhibitor) or single ipilimumab treatments, the dynamic monitoring displayed relatively stable cytokine levels on different follow-up visits (Fig. 7). Specifically, the confocal Raman images (Fig. 7f–h) and the molecular counting (Fig. 7i) in this patient serum consistently showed no significant cytokine level alterations on the three time points (days 0, 21, 42). Under this circumstance, LDA failed to clearly classify the data into separate sections (Fig. 7j). Likewise, Patients 7, 9, and 10 possessed stable cytokine levels and were diagnosed with low grade irAEs. Meanwhile, Patient 8 who showed decreasing cytokine levels did not display signs of irAEs. LDA was not able to classify Patients 7 and 8 who had mild irAEs and did not show irAEs, but it recognised the minor difference in Patients 9 and 10 who showed grade 1 irAEs (Supplementary Fig. 20).
Overall, we found the preliminary evidence to suggest that significantly elevated cytokine levels have a strong correlation with the development and manifestation of severe irAEs, whereas stabile, low baseline, and decreasing cytokine concentration indicate mild and manageable irAEs. The relatively low concentrations of these four cytokines were below the detection sensitivities of commercially available ELISA kits (pM level), which limits their use in clinical studies. Importantly, the measurement at fM cytokine levels in clinical samples is consistent with the median concentrations of cytokine measured by using digital ELISA51. The successful demonstration of the digital nanopillar SERS platform in dynamic detection of cytokines in patient serum provides a potential approach for the future accurate early detection, characterisation, and monitoring of irAEs in clinical settings. However, it is important to note that the cytokine concentration changes are not directly correlated with the treatment response according to response evaluation criteria in solid tumours (RECIST). In our pilot study, some melanoma patients showed higher levels of cytokines compared to the healthy controls. The power of our digital nanopillar SERS platform lies in the capability to longitudinally monitor cytokines in individual patients over time.
Discussion
Despite the frequent occurrence of irAEs in immune checkpoint therapy, particularly for the combination treatment, the prediction of the emergence of irAEs remains elusive. Mounting data suggest a potential role of cytokines as predictive markers for irAE monitoring in immune checkpoint therapy9,11,13,52. Although promising, accurate quantification of these biomarkers was often not possible due to the dearth in technologies with sufficient detection sensitivity. Typically, either cytokines above the detection limit of immunosorbent assay were selected11, or relative cytokine quantification9 was performed for investigating irAEs. The former approach has the drawback of potentially excluding the low abundance cytokines of significance in irAEs. As for relative quantification9,53, the cytokine concentrations are determined by relating to a standard that had the observed median fluorescence value closest to the median of the test sample. The relative concentration, however, may fail to represent accurate cytokine levels and thus needs further exploration. Our developed digital nanopillar SERS assay offers early data suggestive of a potential approach to the above-mentioned challenges and provides the possibility to study trace amounts of a panel of cytokines in an accurate quantification manner as well as concomitantly providing attomolar level sensitivity. The current proof-of-principle approach has measured four of the potential inflammatory and/or immune toxicity-related cytokines for the prediction of emergence, characterisation, and/or quantifiable correlation with irAEs in melanoma patients.
By leveraging the narrow line width of Raman spectra, the developed digital SERS counting assay shows the ability to sensitively and simultaneously detect multiple cytokines. The adoption of the novel digital quantification mode in SERS using gold–silver alloy nanoboxes further improves the high sensitivity of SERS technology. Notably, the digital counting strategy offers an option for reproducible SERS quantification by avoiding the common Raman signal fluctuations induced by ensemble measurements. The ensemble measurement in SERS relies on the enhancement of Raman signals of molecules located in or near the “hot spots” (i.e., strong electromagnetic fields)30. Due to the random distribution and various efficiencies of “hot spots”, it can result in the discrepancies in acquired SERS intensities for inter-laboratory and even intra-laboratory tests29. To circumvent the impact of Raman intensity fluctuation on accurate quantification, we employed the digital SERS signals from the single-SERS-active nanoboxes on discrete pillar arrays to enumerate the targets and only count the “yes” or “no” signal for a robust and reproducible SERS analysis. Furthermore, the digital readout model, which regards both aggregated and single nanoparticle as a single binding event to reflect the true target number, can have a better accuracy and robustness than the intensity-based assay39.
We believe that the proposed digital nanopillar SERS assay could be used to monitor other cytokine-induced immune responses. For instance, with the outbreak of 2019 novel coronavirus (2019-nCoV), it is yet difficult to predict which infected patient will develop a strong immune response that requires hospitalisation. However, cytokines have been indicated to play a major role in the severity of immune response for critically ill patients infected with 2019-nCoV54. The specific detection of multiple cytokines at early stages of viral infection could thus potentially address this issue and help to provide the clinical care for people at the highest risk. For patients with high cytokine concentrations, the digital nanopillar SERS platform will require the sample dilution to suit Poisson distribution.
In summary, we propose a digital nanopillar SERS platform for the parallel counting of single cytokines and dynamic monitoring in the clinical context of irAE development during immune checkpoint blockade therapy. The platform achieved attomolar level sensitivity by utilising discrete pillar array compartments to hold the single cytokine and subsequently applied single-particle active nanobox-based SERS nanotags for cytokine identification and counting. The confocal Raman mapping on the pillar array offered the highest possible clinical specificity by reducing nonspecific signals and provided a “yes/no” type counting approach for reproducible Raman signal readout. The designed platform was rigorously optimised and tested in simulated clinical samples prior to the evaluation for irAE monitoring in stage IV melanoma patients receiving immune checkpoint blockade therapy. We envisaged this platform possessing the advantages of highly sensitive and multiplexing analysis capability can transit into future irAE detection methodologies after extensive validation in a large cohort of clinical samples over different time courses.
Methods
Materials
Silver nitrate (AgNO3), hydrogen tetrachloroaurate (III) trihydrate (HAuCl4·3H2O), MBA, DTNB, TFMBA, MMTAA, DSP were obtained from Sigma Aldrich. Ascorbic acid (AA) of analytical grade was purchased from MP Biomedicals. FGF-2 (223-FB), G-CSF (214-CS), GM-CSF (215-GM), CX3CL1 (365-FR) cytokines; monoclonal anti-FGF-2 (MAB233), anti-G-CSF (MAB214), anti-GM-CSF (MAB615), anti-CX3CL1 (MAB3652) antibodies; polyclonal anti-FGF-2 (AF-233), anti-G-CSF (AF-214), anti-GM-CSF (AF-215), anti-CX3CL1 (AF-365) antibodies; and FGF-2 (DY233-05), G-CSF (DY214-05), GM-CSF (DY215-05), and CX3CL1 (DY365) ELISA kits were bought from R&D Systems.
All the patient serum or plasma samples were collected at the Austin Hospital (Melbourne) under approved human ethic protocols and written informed consents were obtained from all patients before sample collection. Ethics approval was obtained from The University of Queensland Institutional Human Research Ethics Committee (approval nos. 2011001315 and 2016000876) and the following clinical assay was carried out according to the approved guidelines.
Preparation of single-particle active SERS nanotags
The preparation of SERS nanotags involved the synthesis of nanoboxes and the subsequent functionalisation with Raman reporters and antibodies. For nanobox synthesis, 45 µL of HAuCl4 (1 wt%) was added into 10 mL of ultrapure H2O (18.2 Ω cm) under magnetic stirring (800 r.p.m.) for 1 min, followed by simultaneously introducing 170 µL of AgNO3 (6 mM) and 30 µL of AA (0.1 M) into the stirring solution. Then, the formation of nanoboxes was indicated by the appearance of an apparent blue colour within 6 s and the samples were collected 1 min later by centrifuging at 600 g for 15 min.
To functionalise nanoboxes with Raman reporters and antibodies, 300 µL of nanoboxes centrifuged from 1 mL of as-prepared solution were co-incubated with one type of Raman reporters (i.e., 10 µL of DTNB, 8 µL of MBA, 10 µL of TFMBA, or 10 µL of MMTAA) and 2 µL of DSP linker for 6 h. After that, the Raman reporter and DSP functionalised nanoboxes were separated by centrifuging at 600 g for 15 min, and resuspended into 300 µL of PBS (0.1 mM). Then, 2 µg of anti-FGF-2, anti-G-CSF, anti-GM-CSF, anti-CX3CL1 antibodies were added to MBA, DTNB, TFMBA, and MMTAA labelled nanoboxes, respectively. After overnight incubation at 4 oC, the functionalised nanoboxes were purified by centrifuging at 600×g for 15 min to separate free antibodies and the final products were resuspended into 200 µL of 0.1% BSA for future use.
Fabrication of pillar arrays
The chip was made of a sensing array measuring 1 mm × 1 mm (Supplementary Fig. 1a) and consisted of 250,000 individual pillars. Each pillar was 1 µm wide, 1 µm long, and 1 µm high. The pillars were evenly spaced by 1 µm from one pillar to the next (Supplementary Fig. 1b). The pillar array was designed using Nanosuite 6.0 (Raith GmbH) and Beamer 5.9.1 (GenISys GmbH) and fabricated on a 4-inch p-type <100> silicon wafer (Bonda Technology Pte Ltd, Singapore) using electron beam lithography (EBL). The wafer accommodated 76 separate pillar arrays. Silicon wafer was first cleaned in acetone, isopropanol with sonication for 2 min each, followed by rising with deionised H2O and dehydration bake at 180 °C for 2 min. Prior to the resist coating, the wafer had undergone a further O2 plasma cleaning at 200 W for 5 min (Diener Atto, Diener Electronic GmbH). The cleaned wafer was spin-coated with two layers of polymethyl methacrylate (bottom: 495K A4 PMMA, top: 950k A4 PMMA, from MicroChemicals GmbH) using the CEE Apogee Coater (Cost Effective Equipment, LLC) at 1500 r.p.m. for 60 s each. After the coating of each layer, the wafer was baked immediately on a hot plate to prevent from intermixing of the two layers of resist. The baking time was 10 and 3 min for the bottom and top layer, respectively, at 180 °C. The thickness of the photoresist was found to be ~450 nm (top PMMA: ~250 nm, bottom PMMA: ~200 nm), characterised by white light reflectometry (FilmTek 2000M, Scientific Computing International). EBL was performed in the Raith EBPG5150 system. The patterns were exposed in EBL with an accelerated voltage of 100 kV, 150 nA of beam current (spot size ~80 nm), with step sizes of 40 nm and an electron dose of 1200 µC/cm2. The exposure time per 4-inch wafer was ~35 min, each containing 76 individual chips. After exposure, the wafer was developed in a mixture of isopropanol and methyl isobutyl ketone (3:1) for 60 s and rinsed immediately with isopropanol, followed by drying with N2. An oxygen plasma descum process, at 100 W, 60 s (Diener Atto, Diener Electronic GmbH) was carried out to remove resist residues prior to the deposition. Next, 10 nm titanium and 200 nm gold were deposited by physical vapour deposition using a Temescal FC-2000 electron beam evaporator (Ferrotec, U.S.A.). After overnight lift-off at room temperature in Remover PG (MicroChemicals GmbH, Germany), the excess material was washed off and the pillar array structure was revealed (Supplementary Fig. 1c). To create the pillar height (i.e., 1 µm), reactive ion etching (Oxford Instruments, UK) was applied for anisotropic etching of the silicon. Hereby, the deposited gold served as mask to protect the underlying silicon while the un-masked silicon was removed. Next, the wafer was coated with a protective layer of cured AZnLOF 2020 prior to wafer dicing into 76 individual sensing chips consisting of a single pillar array. Prior to use, the protective layer was washed off by consecutive washes with isopropanol and acetone and dried under a stream of nitrogen.
Pillar array functionalization
Antibody functionalisation of the gold-topped pillar array was conducted by crosslinking the antibodies to the gold surface using DSP. A solution of 5 mM DSP in dimethyl sulfoxide was pipetted onto the pillar array and incubated at room temperature for 2 h. After rinsing the pillar array with ethanol and PBS, a solution of 5 µg/mL anti-cytokine monoclonal antibody solution (100 fold dilution of antibody stock solution) in PBS was incubated overnight at 4 °C. Subsequently, the pillar array was rinsed with PBS and blocked using 1% bovine serum albumin in PBS for 1 h. Prior to use, the pillar array was rinsed with PBS. All PBS solutions were filtered through a sterile 0.22 µm syringe filter (Millex-GP, Merck, U.S.A.).
Digital nanopillar SERS profiling of cytokines
Cytokines (FGF-2, G-CSF, GM-CSF, and CX3CL1) with different concentrations in PBS (2.6 aM, 26 aM, 260 aM, and 1031 aM) were incubated with antibody functionalised pillar arrays at room temperature for 30 min, followed by washing the pillar array three times with washing buffer (0.1% BSA and 0.01% Tween 20 in PBS). SERS nanotags were then added into the pillar array for another 30 min incubation under room temperature to identify the targets. Finally, the pillar arrays were washed to remove the free SERS nanotags and were subject to confocal Raman microscope for quantification. For each sample, nine SERS images with each image has the dimension of 60 µm × 48 µm were taken on the pillar array to calculate the overall cytokine concentration.
Simulated clinical sample detection
For the recovery experiment, standard cytokines (FGF-2, G-CSF, GM-CSF, and CX3CL1) with the concentration of 1 fM were added into healthy human serum and then diluted ten times with PBS to quantify.
For the quantification of cytokines in FBS, three simulated clinical samples were prepared by titrating various concentrations of standard cytokines into 10% FBS: Sample 1 (FGF-2 = 3.64 pM, G-CSF = 3.19 pM, GM-CSF = 4.29 pM, and CX3CL1 = 28.57 fM); Sample 2 (FGF-2 = 7.28 pM, G-CSF = 6.38 pM, GM-CSF = 8.58 pM, and CX3CL1 = 571.4 fM); and Sample 3 (FGF-2 = 14.56 pM, G-CSF = 12.76 pM, GM-CSF = 17.16 pM, and CX3CL1 = 1142.8 fM). These three samples were then detected directly using the commercial ELISA kits or digital nanopillar SERS assay with a further dilution of 105, 2 × 105, and 4 × 105, respectively.
Spectroscopic ellipsometry
The antibody film thickness was measured by in-solution spectroscopic ellipsometry (M2000V JA Woollam Co., Inc. USA) using gold-coated substrates and flow cell (QSense® Ellipsometry, Biolin Scientific, Sweden). Measurements were performed at an angle of 65°. Data analysis was performed by CompleteEASE® software using a B-Spline data fit and Cauchy model to calculate the antibody film thickness.
MALDI-TOF MS
The antibody-functionalised nanopillar array chip was subjected to tryptic digest prior to analysis. Sequencing-grade trypsin was made to 50 ng/µL in 25 mM ammonium bicarbonate, and sprayed over the chip using a Bruker Imageprep instrument (Bruker, USA). After trypsin deposition, the chip was incubated in a humid environment at 40 °C for 3 h. Subsequently, the chip was sprayed with a matrix solution, 10 g/L α-cyano-4-hydroxycinnamic acid in 50% acetonitrile with 0.2% trifluoroacetic acid. Next, the chip was analysed with a Bruker Ultraflex III MALDI-TOF mass spectrometer (Bruker, USA) in positive linear mode using Flex Imaging 4.0 (Bruker, USA) with a pixel size of 60 µm. Data were collected from 2 k–30 k m/z, at a laser repetition rate of 200 Hz. Data were normalised using the root mean square approach and visualised using Flex Imaging 4.0 (Bruker, USA) and SCILS LAB 2017a software. For the SCILS LAB analysis, the data were imported using a convolution baseline subtraction, and displayed using root mean squared normalisation.
Instrumentations
SEM images of pillar arrays and nanoboxes were taken on a JEOL-7100 field emission (FE)-SEM (20 kV voltage). TEM images of nanoboxes were taken on a JEOL-2100 microscope (200 kV voltage). NTA of nanobox size distribution was performed with Malvern NanoSight NS300. UV-vis extinction spectrum of nanoboxes was performed with a Shimadzu UV-2450 spectrophotometer. Confocal Raman mapping was conducted on a WITec alpha 300 R spectrometer using 632.8 nm He–Ne laser with the power of 35 mW, a grating of 600 g/mm used with EMCCD camera, spectral resolution of 1.390 cm−1 to 2.114 cm−1, confocal pinhole size of 100 µm, 100× air objective with NA of 0.90, and 0.05 s integration time. The theoretical spot size was 857.80 nm based on the Abbe diffraction limit (i.e., d = 1.22λ/NA). The scanning area was set to have 60 µm × 48 µm with 86 points per line and 69 lines per image. For each pillar array, nine separate scanning areas were taken in total and the total active pillars were used for quantification. The SERS mapping images for counting were taken by focusing the laser on the top of the pillar surfaces. Specifically, the laser was firstly focused on the silicon substrates by obtaining the strongest silicon signals (520 cm−1) and then the 100× objective was moved up in z-axis direction of 1 µm for SERS scanning. The system was calibrated with the first-order photo peak of silicon at 520 cm−1.
Data analysis
To assign the SERS nanotag membership for DTNB, MBA, TFMBA, and MMTAA, Project Five 5.0 software from WITec was utilised to create four filters, which summed a spectral range of 40 cm−1 with the centre position at the characteristic Raman peak of each reporter and subtracted the background with a polynomial algorithm. Specifically, the filter ranges of four Raman reporters DTNB-, MBA-, TFMBA-, and MMTAA-coated SERS nanotags were (1310–1350 cm−1), (1060–1100 cm−1), (1360–1400 cm−1), and (1268–1308 cm−1), respectively. All the SERS images were analysed by using threshold intensity to determine the successful binding events. Specifically, the threshold intensity of FGF-2, G-CSF, GM-CSF, and CX3CL1 was set at 5000, 4000, 5000, and 5000, respectively. For each image, the threshold intensity was doubled-checked and adjusted based on the true Raman peaks in the spectra. Statistical analysis assuming unequal variances was conducted with Kruskal–Wallis test among three groups or Mann–Whitney test between two groups with GraphPad Prism 8.4. To control the error appropriately, we performed multiple comparisons using Dunn’s test. LDA of clinical samples was performed in R software (3.6.2) with the MASS package (7.3-52). The active pillars in SERS images were counted with Image J software.
Supplementary information
Supplementary Information
Peer Review File
Source data
Source Data
Peer review information
Nature Communications thanks Chongwen Wang and Isaac Pence for their contribution to the peer review of this work. Peer reviewer reports are available.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
The online version contains supplementary material available at 10.1038/s41467-021-21431-w.
Acknowledgements
The authors acknowledge grants received by our laboratory from the National Breast Cancer Foundation of Australia (CG-12-07) and the ARC DP (160102836 and 210103151). These grants have significantly contributed to the environment to stimulate the research described here. J.L. acknowledges support from the Australian Government Research Training Program Scholarship. A.W. and A.A.I.S. thank the National Health and Medical Research Council for funding (APP1173669 and APP1175047). A.B. is the recipient of a Fellowship from the Victorian Government Department of Health and Human Services acting through the Victorian Cancer Agency. We acknowledge the facilities, and the scientific and technical assistance, of the Australian Microscopy & Microanalysis Research Facility at the Centre for Microscopy and Microanalysis, The University of Queensland. We appreciate to receive the technical and scientific guidance from Queensland node of the Australian National Fabrication Facility (Q-ANFF) in confocal Raman mapping and spectroscopic ellipsometry measurement.
Author contributions
J.L., A.W., A.I.S., H-H.C., Y.W., A.B., P.M., and M.T. contributed to the design of the experiments and analysing the data. J.L. and A.W. performed the experiments and prepared the manuscript. All authors read, commented, and edited the manuscript, and assisted during the revision process.
Data availability
Data supporting the findings of this work are available within this paper and the supporting information files. A reporting summary of this work is available as a Supplementary file. Source data are provided with this paper.
Competing interests
The authors declare no competing interests. | Recovering | ReactionOutcome | CC BY | 33597530 | 19,690,420 | 2021-02-17 |
What was the outcome of reaction 'Skin toxicity'? | A digital single-molecule nanopillar SERS platform for predicting and monitoring immune toxicities in immunotherapy.
The introduction of immune checkpoint inhibitors has demonstrated significant improvements in survival for subsets of cancer patients. However, they carry significant and sometimes life-threatening toxicities. Prompt prediction and monitoring of immune toxicities have the potential to maximise the benefits of immune checkpoint therapy. Herein, we develop a digital nanopillar SERS platform that achieves real-time single cytokine counting and enables dynamic tracking of immune toxicities in cancer patients receiving immune checkpoint inhibitor treatment - broader applications are anticipated in other disease indications. By analysing four prospective cytokine biomarkers that initiate inflammatory responses, the digital nanopillar SERS assay achieves both highly specific and highly sensitive cytokine detection down to attomolar level. Significantly, we report the capability of the assay to longitudinally monitor 10 melanoma patients during immune inhibitor blockade treatment. Here, we show that elevated cytokine concentrations predict for higher risk of developing severe immune toxicities in our pilot cohort of patients.
Introduction
The advent of immune checkpoint therapy has revolutionised the landscape of traditional cancer treatment and is believed to constitute the backbone of managing certain malignancies1–3. By capitalising on the blockade of immune checkpoint inhibitors to take the brakes off parts of the immune system, this emerging therapy has achieved great success producing long-lasting responses (e.g., 10 years or more) in a small but significant fraction of patients3–6. Nevertheless, upon the blockade of immune checkpoint molecules, the activated and potentiated immune reaction predisposes patients to a significant risk of immune-related adverse events (irAEs), which can occur in up to 80% of patients receiving immune checkpoint therapy7–9. The high incidence of irAEs, which may manifest at any time during treatment, can offset the clinical benefits, lead to premature therapy cessation, and even be life-threatening for certain patients10–12. To assist the successful implementation of immune checkpoint therapy, the use of predictive biomarkers for early identification and vigilant monitoring of irAEs is thus critical and a pressing need in avoiding or ameliorating detrimental effects and adjusting therapeutic options.
Cytokines, small signalling proteins, are promising candidates to indicate the occurrence of irAEs due to their prominent role in modulating the anti-cancer immune responses, including enhancing antigen priming, recruiting immune cells into the tumour microenvironment, and upregulating certain immune checkpoint molecules9,13,14. Particularly, excessive cytokine secretion has been implicated in severe inflammation as a major constituent leading to irAEs. For example, the overproduction of fibroblast growth factor 2 (FGF-2)15–18, granulocyte colony-stimulating factor (G-CSF)19, granulocyte-macrophage colony-stimulating factor (GM-CSF)20, and fractalkine (CX3CL1)21 have been found to participate in immune-related inflammatory disease (e.g., rheumatoid arthritis, autoimmune gastritis, and Crohn’s disease). These inflammatory cytokines have recently been reported to indicate irAEs for melanoma patients who underwent immune checkpoint therapy9. The clinical deployment of cytokine analysis for irAE monitoring is challenging and requires a technology that can (i) determine the selected cytokines with great sensitivity9, especially at the onset of irAE development, where the cytokine concentrations are likely to be the lowest; as well as (ii) simultaneously detect a panel of cytokines to reflect the complex interplay of cytokine signalling pathways22 and the variable irAE symptoms among patients.
Conventional cytokine analyses such as immunosorbent assays have limited clinical applicability for irAE assessment due to their limited capacity to detect low cytokine concentrations in blood as well as for assessing a panel of cytokines in a single sample simultaneously. Recently, advances in micro/nanomaterial-based systems have provided a promising suite of technologies that improve the conventional assays by overcoming the above limitations23,24. Encouragingly, the unique advantages of micro/nanomaterial-based systems convey an attractive option for cytokine analysis with the desired results of high sensitivity and multiplexing. The high specific surface area of these miniaturised materials increases mass transfer subsequently enhancing the interaction with target molecules and thus improving the detection sensitivity25. The capabilities of micro/nanomaterial fabrication techniques permit individually separated compartments sufficiently discrete to hold single molecules and hence encompasses a promising strategy for counting assays that can further push the sensitivity of the traditional assays24,26,27. Moreover, the physicochemical properties of nanostructured materials can be exploited to simultaneously label multiple targets (e.g., various spectral signatures) for high-throughput parallel measurements28–30. Therefore, by combining the potential of micro/nanomaterial systems with the need for sensitive irAE monitoring, we have developed a platform for sensitive and multiplex cytokine counting analysis.
Combining the use of (a) discrete single cytokine nanopillar array chip with discretely separated compartments, (b) control of target concentrations to follow a Poisson distribution, and (c) the recognition of target by single-particle active surface-enhanced Raman scattering (SERS) nanotags with a confocal Raman microscope allows accurate and in situ counting of a multi-cytokine panel (FGF-2, G-CSF, GM-CSF, and CX3CL1). Different from the fluorescence-based digital counting strategies24,26, the strikingly narrow spectral peaks of SERS (~1–2 nm) in comparison to fluorescence (~50 nm) makes this platform intrinsically ideal for multiplexed cytokine analysis28.
In this work, we present a digital nanopillar SERS platform that enables the specific cytokine quantification down to attomolar levels and the application in melanoma patients receiving immune checkpoint blockade therapy. Beyond the capability to predict irAEs in melanoma patients receiving therapy, the digital nanopillar SERS assay could potentially be extended to other cytokine-associated immune responses such as excessive immune activation due to viral or bacterial infections (such as COVID-19).
Results
Digital nanopillar SERS platform for parallel profiling of single cytokine
Our concept of digital nanopillar SERS platform for cytokine analysis relies on Rayleigh criterion separation, probability-driven Poisson distribution, single-particle active SERS nanotags, and confocal SERS mapping (Fig. 1). To precisely fabricate the pillar array, we opted to use an electron beam lithographic approach to write the array into a photon-sensitive material followed by physical vapour deposition of gold to create the gold-topped pillars, and selectively reactive ion etching to reveal the pillar structure (Supplementary Fig. 1). The nanopillar array chip consisted of 250,000 individual pillars. As shown in the scanning electron microscope (SEM) image of Fig. 1a, the cubic nanopillars have an edge-to-edge width of 1000 nm and are evenly distributed at 1000 nm intervals to suit the lateral Raman microscope resolution (~1000 nm) that fulfils the Rayleigh criterion separation required to acquire a single SERS spectrum from each pillar without spectral overlap from adjacent pillars.Fig. 1 Digital single-molecule nanopillar surface-enhanced Raman scattering (SERS) platform for parallel counting of four types of cytokines.
SEM images of a pillar array side view, b nanoboxes, and c a single nanobox on the top of a pillar; d SERS spectra of nanoboxes conjugated with 5,5-dithiobis (2-nitrobenzoic acid) (DTNB), 4-mercaptobenzoic acid (MBA), 2,3,5,6-tetrafluoro-4-mercaptobenzoic acid (TFMBA), or 2‐mercapto‐4‐methyl‐5‐thiazoleacetic acid (MMTAA) Raman reporters; e workflow for multiplex counting of cytokines, including fibroblast growth factor 2 (FGF-2), granulocyte colony-stimulating factor (G-CSF), granulocyte-macrophage colony-stimulating factor (GM-CSF), and fractalkine (CX3CL1). Data from one independent experiment.
By using specific gold-thiol chemistry with the linker molecule dithiobis (succinimidyl propionate) (DSP), the gold-topped pillars were selectively functionalised with target recognition antibodies (anti-FGF-2, anti-G-CSF, anti-GM-CSF, and anti-CX3CL1) and acted as the small compartments to capture and confine the individual cytokine. Upon DSP binding on the gold-topped pillars through gold-thiol bond, DSP uses N-hydroxysuccimide (NHS) ester to react with the amine groups of the antibodies31,32. The successful antibody conjugation on gold-topped pillar surfaces was confirmed by matrix-assisted laser desorption ionisation-time of flight mass spectrometry (MALDI-TOF MS) (Supplementary Fig. 2), which showed high molecular weight fragments derived from antibodies. Furthermore, spectroscopic ellipsometry was utilised to estimate the antibody density on pillar surfaces. Based on the obtained film thickness of 18.5 nm, the calculated antibody surface density was 5.5 mg/m2 using the Cuypers model33, which was in agreement with the reported antibody density on substrate surfaces34. Though these characterisations indicated the presence of antibodies on the pillar array, it was not possible to assess the exact distribution of the four structurally related (same immunoglobulin G family) antibodies on a single pillar with an area of 1 µm2. As an advantage of the digital read-out with a large redundancy of pillars, it is not essential to have all four types of antibodies equally distributed on a single pillar for the assay to work. The combined surface area of all antibody-conjugated pillars provides an excess of cytokine binding sites, which maximises successful cytokine capture within the pillar array. Supplementary Fig. 3 shows the SERS mapping images of an equimolar cytokine solution (1031 aM) that provided a similar signal count for the FGF-2, GM-CSF, G-CSF, and CX3CL1 SERS nanotags, indicating a required distribution of four kinds of antibodies conjugated to the array of pillars. Instrumental to the digital counting of cytokines, we controlled the target concentration based on the principle of Poisson distribution where the ratio of cytokine molecules to pillar number was <1:10, ensuring a 99% probability that there was either one cytokine molecule or zero per pillar24. At a ratio of 1:10, 10% of all pillars were occupied or activated with the cytokine molecules.
Following the capture of cytokines on nanopillars, SERS nanotags were applied to recognise the captured cytokines. The preparation of SERS nanotags was performed by the co-conjugating of Raman reporter and target antibody onto gold–silver alloy nanoboxes. Specifically, an average size of 80 nm gold–silver alloy nanoboxes were firstly synthesised using a rapid and aqueous phase approach35 as indicated in the SEM image in Fig. 1b. Supplementary Fig. 4a, b shows the transmission electron microscope (TEM) image of the nanoboxes with the hollow inner structure and a wall thickness of around 15 nm. Nanoparticle tracking analysis (NTA), which allows the tracking and detection of single particles, shows the nanoboxes have a mode size of 77 nm (D10 = 67.6 nm and D90 = 110.6 nm) (Supplementary Fig. 4c). UV-vis extinction spectroscopy demonstrates the nanoboxes possess a surface plasmon resonance (SPR) peak at 610 nm (Supplementary Fig. 4d). The resonance frequency of the nanoboxes enables a more sensitive signal readout with 632.8 nm laser excitation29, which also has a higher Raman scattering efficiency than 785 nm laser. Thereafter, four types of Raman reporters (5,5-dithiobis (2-nitrobenzoic acid) (DTNB), 4-mercaptobenzoic acid (MBA), 2,3,5,6-tetrafluoro-4-mercaptobenzoic acid (TFMBA), and 2‐mercapto‐4‐methyl‐5‐thiazoleacetic acid (MMTAA)) that generate unique Raman signals (1330 cm−1, 1080 cm−1, 1380 cm−1, and 1288 cm−1) were coupled with their corresponding detection antibodies onto nanoboxes as specific SERS nanotags for identification of FGF-2, G-CSF, GM-CSF, and CX3CL1, respectively. As shown in Fig. 1d, the four SERS nanotags provide the strong and non-overlapping Raman signals, which facilitates the multiplexing analysis of four cytokines. The assignment of the major Raman peaks from the four Raman reporters was summarised into Supplementary Table 1. To evaluate the SERS enhancement property of the nanoboxes, we calculated the enhancement factor (EF) of the four Raman reporters on the nanoboxes. Based on the labelled characteristic peaks in Supplementary Fig. 5, the calculated EFs of DTNB, MBA, TFMBA, and MMTAA were 8.14 × 106, 1.46 × 107, 4.01 × 107, and 3.26 × 107, respectively. The obtained EFs were higher than the reported spherical gold nanoparticles and pure silver nanocubes36 and comparable to the reported hollow nanocubes37, illustrating the high SERS property of the nanoboxes. To investigate the SERS nanotag stability, we monitored the Raman signal intensity over 7 days. As shown in Supplementary Fig. 6, Raman signal intensity variations are less than 5% in the SERS spectra, suggesting the good stability of the prepared SERS nanotags.
The following SERS mapping generated false-colour images for counting single cytokine molecules. Under the Raman microscope, the pillar array was visualised as a blue and black grid by representing the specific Raman shifts corresponding to the silicon signals (520 cm−1), in which the blue colour was assigned to silicon signals showing silicon substrates and the black colour indicated the gold-topped pillars because of the lack of silicon signals. The representation of the four colours of the SERS nanotags (red, green, purple, and cyan) on the gold-topped pillars (i.e., black) reflected cytokine molecule occupation (FGF-2, G-CSF, GM-CSF, and CX3CL1). Elevating the sensing area (or gold-topped pillars) from the silicon substrate was selected as a strategy to minimise the false-positive events. By using the confocal function of the Raman microscope, the laser was selectively focused on the gold-topped pillars, thus largely removing the background signals from potentially non-specifically adsorbed SERS nanotags on the silicon substrate. Finally, the specific SERS nanotag signals present or absent on the gold-topped pillars were counted and represented as percentage of active pillars used for total cytokine quantification. For statistical calculations, SERS mapping was applied for scanning 6480 pillars. This digital counting mode, therefore, has the potential to reach the ultimate sensitivity of single molecule cytokine detection.
Demonstration of the single-particle SERS activity of gold–silver alloy nanoboxes
The successful implementation of digital nanopillar SERS assay necessitates the use of single-particle active plasmonic nanostructures that give a clearly detectable signal for each of the single cytokine binding event. The single-particle SERS detection sensitivity is essential to this assay development as the single-particle inactive plasmonic nanoparticles (e.g., spherical gold nanoparticles)38 would unavoidably result in an underestimate of cytokine concentration.
We evaluated the single-particle SERS activity of the prepared anisotropic nanoboxes by acquiring the signals from the individual nanoboxes that were labelled with DTNB reporters. The use of DTNB, a non-resonant Raman reporter, guaranteed the Raman signal enhancement was solely contributed from the nanobox-generated electromagnetic field. As seen in the SEM image (Fig. 2a), two clearly separated DTNB-labelled nanoboxes were deposited on the silicon wafer (highlighted in red circles). The corresponding Raman image (Fig. 2b) displayed several bright Raman spots originating from these individual nanoboxes. The elongated bright Raman spots in the SERS mapping image were probably caused by the slight aggregation of several nanoboxes during sample preparation processes (e.g., centrifugation)38, which was difficult to visually resolve in the SEM image (Fig. 2a). However, unlike the intensity-based assay, the aggregated nanoboxes as SERS nanotags to target cytokine will not skew the digital readout result, because each cytokine will occupy a single pillar following Poisson distribution and both aggregated and individual nanoparticles are regarded as a single binding event that truly reflects the target number39. We then acquired the SERS spectra from two individual SERS nanoboxes (Fig. 2c, (1) and (2)) and two separate spots of bare silicon (Fig. 2c, (3) and (4)). The presence of nanoboxes showed the characteristic Raman signal at 1330 cm−1 from DTNB, whereas the silicon spectra (3) and (4) lacked the specific peak. This observation demonstrated the single-particle SERS activity of nanoboxes, which was largely attributed to the enhanced electromagnetic fields of nanoboxes on specific regions (e.g., tips and corners)40,41 and thereby facilitated the sensitive and accurate counting of cytokines. Based on the acquired SERS mapping image, the median (interquartile range) of the DTNB peak intensity (1330 cm−1) in the presence and absence of nanoboxes were 183.03 a.u. (149.48–243.35 a.u.) and 18.07 a.u. (15.51–23.12 a.u.), respectively. Furthermore, the mean ± standard deviation of the DTNB peak intensity with nanoboxes (213.41 ± 85.03 a.u.) distinguished clearly from the position without nanoboxes (18.79 ± 6.01 a.u.), which demonstrated the feasibility of correctly identifying the presence of nanoboxes.Fig. 2 Demonstration of the single-particle SERS activity of DTNB-labelled nanoboxes.
a SEM image and b corresponding SERS mapping image of DTNB-labelled nanoboxes on a silicon substrate; c representative SERS spectra of numbered locations indicated in a and b. The red dotted line shows the characteristic peak at 1330 cm−1 from DTNB. Data from one independent experiment. Source data are provided in the Source Data file.
Fig. 3 Study of confocal height on Raman signal intensity.
SERS mapping of FGF-2 SERS nanotags on the silicon substrate with changing confocal height of a 0 nm, b 500 nm, c 1000 nm, and d 1500 nm; selected Raman spectra obtained from e red circles and f blue circles of SERS images. Red dotted lines in e and f indicate peak signal at 1330 cm−1 from DTNB. Data from one independent experiment. Source data are provided in the Source Data file.
Fig. 4 Specificity of digital nanopillar SERS platform for FGF-2 cytokine detection.
Representative confocal SERS images in the presence of a target FGF-2 (1031 aM), and negative controls with non-target controls b G-CSF (1031 aM), c GM-CSF (1031 aM), d CX3CL1 (1031 aM), and e PBS. The median (interquartile range) of active pillars per scanning image for FGF-2, G-CSF, GM-CSF, CX3CL1, and PBS was 72 (63.5–76.75), 1.5 (1.5–2), 2 (1–4), 0.5 (0–1.25), and 1 (1–1.75), respectively. Data from one independent experiment.
Optimisation of digital nanopillar SERS platform for cytokine detection
The reliable detection of single cytokine molecules by the digital nanopillar SERS platform depends on the geometric features of the pillar array (i.e., pillar height, cross-section area of pillar) and assay conditions (i.e., incubation time for sample and SERS nanotags).
We first sought to investigate the effect of pillar height on Raman signal intensity to differentiate signals from non-specifically bound SERS nanotags on the silicon substrate and specifically bound SERS nanotags on the gold-topped pillars. FGF-2 SERS nanotags were randomly deposited on the silicon substrate to mimic the non-specific binding scenario and the Raman mappings were perfomed by moving the objective along the z-axis direction with different heights (0 nm, 500 nm, 1000 nm, and 1500 nm) to compare the signal intensity. In Fig. 3, a–d show the false-colour SERS images and e, f the corresponding SERS spectra with characteristic DTNB reporter peak at 1330 cm−1 acquired from the circled spots in a–d. At the height of 0 nm where the SERS nanotags were in focus, we noticed bright Raman spots (Fig. 3a) and strong Raman signals (black line in Fig. 3e, f). With increasing z heights to 500 nm and 1000 nm, the Raman spots decreased (Fig. 3b, c) and the signal intensity weakened/disappeared (red and blue lines in Fig. 3e, f) as the nanoboxes became increasingly out of focus. A further increase to 1500 nm did not remarkably weaken Raman signals compared to the height of 1000 nm (Fig. 3d). Hence, for the fabrication of the pillar array chip, we selected a pillar height of 1000 nm to greatly reduce the potential interference from non-specific signals.
The cross-section area of the pillars provides the space for cytokine binding and labelling with SERS nanotags. To study the effect of pillar cross-section area, we fabricated array chips with pillars of various widths (250 nm, 500 nm, and 1000 nm) (Supplementary Fig. 7a–c) and functionalised with anti-FGF-2 antibody. Each chip consisted of 250,000 individual pillars. We then analysed a sample that contained ~25,000 molecules of FGF-2 (40 µL, 1031 aM), which should result in 10% active pillars (ratio FGF-2: pillars of 0.1). As seen in Supplementary Fig. 7d–f, an increasing pillar cross-section area results in a higher fraction of active pillars. In reference to the expected active pillar percentage (10%), the 250 nm and 500 nm pillar arrays produced lower active pillars (2% and 6%), which suggested a significant loss of target recognition by SERS nanotags. For the 1000 nm wide pillars, the active pillar percentage was 11%, close to the nominal value of 10%. We further tested a sample with 260 aM FGF-2 (i.e., 2.5% active pillars) on the pillar array chips with 250, 500, and 1000 nm pillar widths. The capture efficiency of these three chips was summarised in Supplementary Table 2. In comparison to the pillar array of 250 nm and 500 nm sizes, the 1000 nm provided an improved capture efficiency. As the accessible target recognition surface area per pillar increases, it can possibly promote the thermodynamics and kinetics for higher surface binding and capture efficiency25. Consequently, the 1000 nm pillar array was adopted in the subsequent experiments.
An optimal incubation time of cytokine and SERS nanotags on the pillar array can shorten the operation time and reduce the potential risk of nonspecific binding that could lead to false-positive counting. We thus studied the effect of incubation time of cytokine with SERS nanotags for 30 to 90 min in a solution of 1031 aM FGF-2 and FGF-2 SERS nanotags. As suggested by Supplementary Fig. 8, the increase in incubation time gives rise to a higher proportion of active pillars. In comparison with the theoretical active pillar percentage (10%), both 30 min and 60 min incubation time were able to provide a desirable active pillar percentage (11% and 13%, respectively). A longer incubation time (90 min), however, reported an active pillar percentage (20%) two times higher than the theoretical, indicating the occurrence of nonspecific absorption of SERS nanotags on the pillar array chip. Thus, we selected 30 min incubation time for further digital nanopillar SERS measurements.
Specificity of the digital nanopillar SERS platform for cytokine detection
Accurate and reliable recognition of the specific target is essential for cytokine quantification in clinical samples. To demonstrate the detection specificity of the digital nanopillar SERS assay, we prepared an anti-FGF-2 antibody functionalised pillar array and measured samples containing target FGF-2 cytokine and controls (G-CSF, GM-CSF, CX3CL1, and PBS). It was observed that only the presence of FGF-2 activated significant amounts of pillars whereas the negative controls only generated negligible active pillars (Fig. 4), indicating the high specificity for FGF-2 detection. Similarly, we studied the specific detection of G-CSF, GM-CSF, and CX3CL1, as shown in Supplementary Figs. 9–11, in which the typical Raman images displayed high proportions of active pillars in the presence of specific targets but not for the negative controls.
To further investigate the specificity of binding between SERS nanotag and antibody-functionalised pillar, we performed SEM analysis to “closely” inspect the pillar array for the presence or absence of SERS nanotags. As a representative model, we selected to image FGF-2 SERS nanotags on the anti-FGF-2-functionalised pillar array after sample incubation with FGF-2 cytokine and non-target controls (Fig. 5). As expected, we observed the cubic nanoparticles on the top of pillars in the presence of FGF-2 due to the successful recognition of SERS nanotags. On the contrary, pillar arrays did not display a significant number of FGF-2 SERS nanotags with non-target controls. Consequently, the consistent Raman and SEM data demonstrate the capability of the assay for specific target cytokine counting. The ability to selectively identify these four cytokines in the designed assay is critically important for their usage in clinical samples.Fig. 5 Specificity of the digital nanopillar SERS platform for FGF-2 cytokine detection.
Representative SEM images of pillar array incubated with FGF-2 SERS nanotags in the presence of a, b FGF-2 (1031 aM), c G-CSF (1031 aM), d GM-CSF (1031 aM), e CX3CL1 (1031 aM), and f PBS. The red circles highlight the existence of SERS nanotags. Panel b is the magnified SEM image of the red-highlighted section in a. It is noted that nanofabrication debris on the sidewall of the pillars can also be seen. Data from one independent experiment.
Sensitivity of the digital nanopillar SERS platform for cytokine detection
As there is typically low abundance of cytokines in clinical samples, the technique based on cytokine detection is expected to possess sufficient sensitivity to reliably assess irAEs9. To investigate the sensitivity and dynamic detection range of the digital nanopillar SERS assay, we firstly titrated the designated concentration of one target cytokine (FGF-2) on the pillar array chip with 250,000 pillars. To comply with the Poisson distribution, the upper number of cytokine molecules in the sample is 25,000 which should result in 10% activated pillars. Based on this upper molecule number, we were motivated to challenge the assay by serially diluting the number of cytokine molecule in the sample from 25,000 (1031 aM), 6305 (260 aM), 631 (26 aM), and 63 (2.6 aM). As suggested by the Raman images in Supplementary Fig. 12 and with the decrease in FGF-2 molecules, the percentage of active pillars decreased correspondingly from 9.39% for 1031 aM, 6.59% for 260 aM, 1.12% for 26 aM, and 0.62% for 2.6 aM, showing a strong correlation that facilitates quantitative cytokine analysis.
Subsequently, we were interested in exploring the multiplexing capability of SERS to investigate the digital nanopillar SERS assay’s dynamic range for the simultaneous quantification of all studied cytokines. As the targets independently follow Poisson distribution, each of the cytokine was separately controlled to activate less than 10% pillars. The specific SERS nanotags provided unique signals for each cytokine that was visualised in the false-colour SERS images by a different colour, thereby enabling in situ and simultaneous cytokine detection. As suggested by the confocal SERS images in Fig. 6, an increase in cytokine concentration corresponded with a higher percentage of active pillars. To facilitate quantitative measurements of the cytokines, we calculated the logarithmic transformation of the percentage of active pillars versus cytokine concentration (Supplementary Fig. 13) confirming the strong statistical and potentially clinically relevant correlation (coefficient of determination (R2) >0.97) observed in the SERS images.Fig. 6 Sensitivity for the simultaneous detection of four cytokines.
Representative confocal SERS images of fibroblast growth factor 2 (FGF-2), granulocyte colony-stimulating factor (GM-CSF), granulocyte colony-stimulating factor (G-CSF), and fractalkine (CX3CL1) with the concentration of a 2.6 aM, b 26 aM, c 260 aM, d 1031 aM. Colour scale bars indicate Raman intensities from 5,5-dithiobis (2-nitrobenzoic acid) (DTNB), 4-mercaptobenzoic acid (MBA), 2,3,5,6-tetrafluoro-4-mercaptobenzoic acid (TFMBA), or 2‐mercapto‐4‐methyl‐5‐thiazoleacetic acid (MMTAA). The median (interquartile range) of active pillars per scanning image of FGF-2, G-CSF, GM-CSF, CX3CL1 for 2.6 aM: 3 (1.5–3), 1 (1–2), 2 (1–3), 2 (1–3); 26 aM: 8 (5.5–10), 10 (9–13), 7 (6–10), 8 (6–10); 260 aM: 40 (36–48), 40 (35–52), 39 (35–50), 37 (36–49); and 1031 aM: 79 (61.5–97), 78 (72–87.5), 88 (68.5–97), 79 (64–95), respectively. Data represents one experiment from three independent tests.
To further investigate the multiplexing quantification performance of the digital nanopillar SERS assay in human serum, we spiked standard cytokines in human serum and tested the dynamic range. Supplementary Fig. 14 shows the linear relationship curves for the four targets. Because of the more complicated sample matrix composition in human samples, the lowest detectable cytokine concentration (5.2 aM) was higher than the PBS solution (2.6 aM).
At a cytokine to pillar ratio of 1:10, we studied the probability of each pillar being occupied by different molecule numbers. To experimentally investigate the number of molecules on a single pillar, we analysed a cytokine mixture that contained all four target cytokines at equal concentration (i.e., ~6250 molecules per cytokine). To visualise and count molecule binding events on a single pillar, we labelled the captured cytokines with the four SERS nanotags that provide clearly distinguishable signals. Under Poisson distribution, the likelihood of having two or more molecules on a single pillar is <0.45% (Supplementary Table 3), which underlies the digital counting principle24. Compared to the theoretical Poisson distribution, the experiment data reported a close but slightly higher value, which was probably due to minor non-specific binding of SERS nanotags on the pillars.
The high sensitivity (attomolar level) of the digital nanopillar SERS assay can be ascribed to the following factors: the digital counting strategy, the single-particle SERS activity of the nanoboxes, and the use of pillars to suit confocal Raman mapping that efficiently excludes false-positive signals. Commercially available methods with potential for trace analysis of cytokines include the single-molecule ELISA Simona by Quanterix and electrochemical luminescence assay42,43 by Meso Scale Discovery. Compared to these two methods, the developed digital nanopillar SERS platform enabled in situ multiplexed detection of four cytokines with comparable sensitivity. Unlike the issues of photo bleaching and poor multiplexing analysis often encountered in fluorescence44 and luminescence assays45, SERS provides the advantage of high multiplexing (e.g., 31-plex)46–48 with the narrow Raman linewidth and high photo stability of the Raman reporters. In addition, this digital nanopillar SERS platform can provide more accurate quantification of cytokines by reducing the false-positive signals with the confocal setting, thus eventually help clinicians to monitor irAEs during immune checkpoint therapy. The highly sensitive readout for multiple targets also indicated the capability of this assay for cytokine detection to assess irAEs in clinically relevant samples.
Evaluation of digital nanopillar SERS platform on simulated patient samples
The detection of trace concentrations of cytokines in serum samples is difficult because plasma samples contain a high abundance of non-target molecules (e.g., serum albumin and other proteins) that can potentially interfere with cytokine detection and lead to inaccurate clinical results. To evaluate the capability of the digital nanopillar SERS assay in accurately counting single cytokine molecules, we opted to perform a recovery test in simulated patient plasma samples (i.e., healthy human serum spiked with 1 fM of FGF-2, G-CSF, GM-CSF, and CX3CL1). The rapid scan rate (i.e., 0.05 s for Raman signal integration) facilitated the detection of Raman signals from FGF-2, G-CSF, GM-CSF, and CX3CL1 SERS nanotags rather than the non-target molecules present in human serum due to their low Raman cross-section. As a representative example, Supplementary Fig. 15 shows the Raman signal distribution of the FGF-2 SERS nanotags on five different spots on the pillar array obtained from the recovery test without noticeable Raman signals from other molecules. It is worth noting that unlike the solution-based DTNB labelled SERS nanotag spectra in Fig. 1d, some of the peaks at 1556 cm−1 and 1330 cm−1 in Supplementary Fig. 15 had a similar intensity, which was probably because of the different orientation of the anisotropic nanoboxes on the substrate relative to the polarisation of excitation laser49. Supplementary Tables 4 and 5 show the cytokine concentrations in healthy human serum and human serum spiked with 1 fM cytokine standards determined by digital nanopillar SERS platform, respectively. On five independent pillar arrays, the measured concentrations had the relative standard deviation (RSD) below 9.0% and the Kruskal–Wallis test showed no statistical differences among these results (p » 0.05). Overall, the observed inter-chip variation should enable accurate identification of disease progression to severe irAEs (e.g., grade 3 or 4), but may encounter some challenges in discriminating mild progressing to moderate irAEs (e.g., grade 1 or 2). The assay enabled trace determination of the four targets in simulated human serum as suggested by the target recovery rates of 80.00% to 137.00% with RSD from 16.02% to 21.80% (Supplementary Table 6). Importantly, the ability to measure reliably cytokines at attomolar levels in simulated human serum samples holds promise for detecting early changes in cytokine concentrations as predictors for the emergence of irAEs in immune checkpoint blockade treated patients.
To validate the accuracy of the digital nanopillar SERS assay, we compared the assay with commercially available ELISA kits (one kit for each cytokine tested) for cytokine quantification. To represent a potential clinical scenario of a patient developing irAEs during immune checkpoint blockade therapy, we prepared three samples with increasing concentrations of cytokines (spike in experiments into fetal bovine serum (FBS)) and subsequently analysed these samples with our digital nanopillar SERS assay and the commercial ELISA kits. FBS was used as complex sample matrix devoid of human cytokines. As the limits of detection for the ELISA kits (FGF-2 = 0.95 pM, G-CSF = 1.66 pM, GM-CSF = 1.11 pM, and CX3CL1 = 17.86 pM) were above the attomolar level, the simulated samples were prepared to suit the detection range of these kits. For the digital nanopillar SERS assay, the samples were diluted correspondingly and generated consistent results with the ELISA kits as shown in Supplementary Table 7. No statistical differences were found between ELISA and digital nanopillar SERS results based on Mann–Whitney test. Furthermore, we compared the detection of four cytokines in human serum with digital nanopillar assay and ELISA kits (Supplementary Table 8). The cytokine levels in human serum were below the limit of detection for the conventional ELISA kits, whereas their concentration was quantified by digital nanopillar SERS platform. For the human serum spiked with standard cytokines, the digital SERS platform generated similar results to ELISA without significant differences by Mann–Whitney test. Collectively, the digital nanopillar SERS platform showcased the ability to robustly and accurately quantify cytokines in complicated samples, which is significant for the prospect of dynamic correlation monitoring of irAEs in clinical samples.
Following the demonstration of the accuracy of digital nanopillar SERS platform, we tested the four cytokine levels in ten healthy people (Supplementary Table 9). These ten healthy people showed cytokine concentrations beyond the conventional ELISA capability to accurately quantify, which was consistent with previous reports9,50,51.
Dynamic correlation monitoring of irAEs in melanoma patients receiving immune checkpoint blockade treatment
Having established the feasibility of digital nanopillar SERS in simulated clinical samples, we applied the platform for longitudinally monitoring irAEs in ten melanoma patients (2–3 time points per patient, 26 samples in total) who underwent immune checkpoint blockade therapy (Supplementary Table 10). By diluting the patient samples to follow Poisson distribution, we quantified the cytokine concentration using digital nanopillar SERS platform. Based on the clinical assessments, the patients were classified into two categories: (i) developed severe irAEs (grades 3 and 4) and needed hospitalisation and dedicated treatment (Patients 1, 2, 3, 4, 5); and (ii) developed minor irAEs (grades 1 and 2) that could be managed with immunosuppressants (e.g., corticosteroids) or exhibited no symptom of irAEs (Patients 6, 7, 8, 9, 10).
As a representative case, Fig. 7 shows two cytokine profiles of a patient with severe irAEs (Patient 1) and a patient with mild irAEs (Patient 1). For Patient 1 who received ipilimumab (cytotoxic T-lymphocyte antigen-4 (CTLA-4) inhibitor) and was checked on days 7, 21, and 42, the confocal SERS images showed an increase in active pillars with the continuation of treatment (Fig. 7a–c), suggesting an elevation of the cytokine levels that could potentially trigger the severe irAEs. In agreement with the Raman images, the quantitative counting results for the four cytokines also corroborated the increase of cytokine concentrations peaking in sub-fM levels (Fig. 7d). These cytokine levels were below the limit of detection of conventional ELISA kits (pM level). Importantly, we observed significantly elevated cytokine concentrations in Patient 1 serum on day 42 compared to days 7 and 21. This patient showed the onset of grade 4 irAEs (i.e., colitis) 13 days later (day 55), consistent with the concept that higher cytokine levels correlate with increased risk of developing irAEs9. To further evaluate the utility of these four biomarkers as a signature in identifying and characterising irAEs, we analysed all the counting data from Patient 1 by applying linear discriminant analysis (LDA). As seen in Fig. 7e, the LDA successfully distinguished the data on day 42 into a separate zone from days 7 and 21, which may indicate the potential value of biomarkers in monitoring irAEs development. We further demonstrated Patient 1 LDA with the use of all combinations of two (Supplementary Fig. 16) and three cytokines (Supplementary Fig. 17). Overall, the LDA with four cytokines showed improved classification over using three or less cytokines. Interestingly, considering FGF-2/G-CSF, G-CSF/GM-CSF, and G-CSF/CX3CL1, the LDA generated similar performance to the LDA with four cytokines. To further compare the classification power of FGF-2/G-CSF, G-CSF/GM-CSF, and G-CSF/CX3CL1, and four cytokines, we performed LDA of Patient 2 (Supplementary Fig. 18), which suggested a better differentiation with the use of four cytokines. Therefore, the inclusion of all four cytokines in LDA facilitated a wider and more accurate patient sample analysis. Similarly, Patients 2, 3, 4, and 5, who manifested severe irAEs were connected with higher cytokine levels (Supplementary Fig. 19) and amelioration of irAEs symptom was witnessed with a decrease of cytokine concentrations. For these severe irAEs patients, the LDA model showed a clear discrimination in cytokine profile and this could help to identify patients at risk of irAEs (Supplementary Fig. 19).Fig. 7 Digital nanopillar SERS assay for monitoring melanoma patients during immune checkpoint therapy.
For Patient 1 who developed severe irAEs, SERS images for cytokine detection on a day 7, b day 21, c day 42, d cytokine concentration graph for fibroblast growth factor 2 (FGF-2), granulocyte colony-stimulating factor (G-CSF), granulocyte-macrophage colony-stimulating factor (GM-CSF), and fractalkine (CX3CL1). The two shorter horizontal lines denote the interquartile ranges (25th and 75th percentile) and the longer horizontal lines in between denote the median (50th percentile), and e LDA analysis, respectively. For Patient 6 who developed mild irAEs, SERS images for cytokine detection on f day 0, g day 21, h day 42, i four cytokine concentration graph, the two shorter horizontal lines denote the interquartile ranges (25th and 75th percentile) and the longer horizontal lines in between denote the median (50th percentile), and j LDA analysis, respectively. IPI ipilimumab, PEMBRO pembrolizumab; G3 grade 3, G2 grade 2; SD stable disease, PR partial response. For Patient 1, the median (interquartile range) of active pillars per scanning image of FGF-2, G-CSF, GM-CSF, CX3CL1 on day 7: 14 (11–22.5), 23 (21, 29), 12 (7.5–18), 17 (9–25.5); day 21: 30 (19–37.5), 33 (19–41), 26 (17.5–36.5), 29 (21–43); and day 42: 33 (16.5–58.5), 76 (64–128.5), 25 (14–39.5), 48 (26.5–73.5), respectively. For Patient 6, the median (interquartile range) of active pillars per scanning image of FGF-2, G-CSF, GM-CSF, CX3CL1 on day 0: 18 (16–23), 49 (31.5–56), 23 (17.5–28), 20 (14.5–27); day 21: 29 (24–33.5), 53 (46.5–70), 35 (25–46), 22 (19–29.5); and day 42: 13 (8–16.5), 44 (23.5–55.5), 10 (6.5–12.5), 30 (24–34.5), respectively. The data represented three technical replicates obtained from three chips. Nine images were acquired from each chip for cytokine counting. Statistical analysis was based on Kruskal–Wallis test followed by Dunn’s test to correct multiple comparisons (two-sided). Source data are provided in the Source Data file.
As for Patient 6 who exhibited mild grade 2 irAEs on the skin during combined ipilimumab and pembrolizumab (programmed death-1 (PD-1) inhibitor) or single ipilimumab treatments, the dynamic monitoring displayed relatively stable cytokine levels on different follow-up visits (Fig. 7). Specifically, the confocal Raman images (Fig. 7f–h) and the molecular counting (Fig. 7i) in this patient serum consistently showed no significant cytokine level alterations on the three time points (days 0, 21, 42). Under this circumstance, LDA failed to clearly classify the data into separate sections (Fig. 7j). Likewise, Patients 7, 9, and 10 possessed stable cytokine levels and were diagnosed with low grade irAEs. Meanwhile, Patient 8 who showed decreasing cytokine levels did not display signs of irAEs. LDA was not able to classify Patients 7 and 8 who had mild irAEs and did not show irAEs, but it recognised the minor difference in Patients 9 and 10 who showed grade 1 irAEs (Supplementary Fig. 20).
Overall, we found the preliminary evidence to suggest that significantly elevated cytokine levels have a strong correlation with the development and manifestation of severe irAEs, whereas stabile, low baseline, and decreasing cytokine concentration indicate mild and manageable irAEs. The relatively low concentrations of these four cytokines were below the detection sensitivities of commercially available ELISA kits (pM level), which limits their use in clinical studies. Importantly, the measurement at fM cytokine levels in clinical samples is consistent with the median concentrations of cytokine measured by using digital ELISA51. The successful demonstration of the digital nanopillar SERS platform in dynamic detection of cytokines in patient serum provides a potential approach for the future accurate early detection, characterisation, and monitoring of irAEs in clinical settings. However, it is important to note that the cytokine concentration changes are not directly correlated with the treatment response according to response evaluation criteria in solid tumours (RECIST). In our pilot study, some melanoma patients showed higher levels of cytokines compared to the healthy controls. The power of our digital nanopillar SERS platform lies in the capability to longitudinally monitor cytokines in individual patients over time.
Discussion
Despite the frequent occurrence of irAEs in immune checkpoint therapy, particularly for the combination treatment, the prediction of the emergence of irAEs remains elusive. Mounting data suggest a potential role of cytokines as predictive markers for irAE monitoring in immune checkpoint therapy9,11,13,52. Although promising, accurate quantification of these biomarkers was often not possible due to the dearth in technologies with sufficient detection sensitivity. Typically, either cytokines above the detection limit of immunosorbent assay were selected11, or relative cytokine quantification9 was performed for investigating irAEs. The former approach has the drawback of potentially excluding the low abundance cytokines of significance in irAEs. As for relative quantification9,53, the cytokine concentrations are determined by relating to a standard that had the observed median fluorescence value closest to the median of the test sample. The relative concentration, however, may fail to represent accurate cytokine levels and thus needs further exploration. Our developed digital nanopillar SERS assay offers early data suggestive of a potential approach to the above-mentioned challenges and provides the possibility to study trace amounts of a panel of cytokines in an accurate quantification manner as well as concomitantly providing attomolar level sensitivity. The current proof-of-principle approach has measured four of the potential inflammatory and/or immune toxicity-related cytokines for the prediction of emergence, characterisation, and/or quantifiable correlation with irAEs in melanoma patients.
By leveraging the narrow line width of Raman spectra, the developed digital SERS counting assay shows the ability to sensitively and simultaneously detect multiple cytokines. The adoption of the novel digital quantification mode in SERS using gold–silver alloy nanoboxes further improves the high sensitivity of SERS technology. Notably, the digital counting strategy offers an option for reproducible SERS quantification by avoiding the common Raman signal fluctuations induced by ensemble measurements. The ensemble measurement in SERS relies on the enhancement of Raman signals of molecules located in or near the “hot spots” (i.e., strong electromagnetic fields)30. Due to the random distribution and various efficiencies of “hot spots”, it can result in the discrepancies in acquired SERS intensities for inter-laboratory and even intra-laboratory tests29. To circumvent the impact of Raman intensity fluctuation on accurate quantification, we employed the digital SERS signals from the single-SERS-active nanoboxes on discrete pillar arrays to enumerate the targets and only count the “yes” or “no” signal for a robust and reproducible SERS analysis. Furthermore, the digital readout model, which regards both aggregated and single nanoparticle as a single binding event to reflect the true target number, can have a better accuracy and robustness than the intensity-based assay39.
We believe that the proposed digital nanopillar SERS assay could be used to monitor other cytokine-induced immune responses. For instance, with the outbreak of 2019 novel coronavirus (2019-nCoV), it is yet difficult to predict which infected patient will develop a strong immune response that requires hospitalisation. However, cytokines have been indicated to play a major role in the severity of immune response for critically ill patients infected with 2019-nCoV54. The specific detection of multiple cytokines at early stages of viral infection could thus potentially address this issue and help to provide the clinical care for people at the highest risk. For patients with high cytokine concentrations, the digital nanopillar SERS platform will require the sample dilution to suit Poisson distribution.
In summary, we propose a digital nanopillar SERS platform for the parallel counting of single cytokines and dynamic monitoring in the clinical context of irAE development during immune checkpoint blockade therapy. The platform achieved attomolar level sensitivity by utilising discrete pillar array compartments to hold the single cytokine and subsequently applied single-particle active nanobox-based SERS nanotags for cytokine identification and counting. The confocal Raman mapping on the pillar array offered the highest possible clinical specificity by reducing nonspecific signals and provided a “yes/no” type counting approach for reproducible Raman signal readout. The designed platform was rigorously optimised and tested in simulated clinical samples prior to the evaluation for irAE monitoring in stage IV melanoma patients receiving immune checkpoint blockade therapy. We envisaged this platform possessing the advantages of highly sensitive and multiplexing analysis capability can transit into future irAE detection methodologies after extensive validation in a large cohort of clinical samples over different time courses.
Methods
Materials
Silver nitrate (AgNO3), hydrogen tetrachloroaurate (III) trihydrate (HAuCl4·3H2O), MBA, DTNB, TFMBA, MMTAA, DSP were obtained from Sigma Aldrich. Ascorbic acid (AA) of analytical grade was purchased from MP Biomedicals. FGF-2 (223-FB), G-CSF (214-CS), GM-CSF (215-GM), CX3CL1 (365-FR) cytokines; monoclonal anti-FGF-2 (MAB233), anti-G-CSF (MAB214), anti-GM-CSF (MAB615), anti-CX3CL1 (MAB3652) antibodies; polyclonal anti-FGF-2 (AF-233), anti-G-CSF (AF-214), anti-GM-CSF (AF-215), anti-CX3CL1 (AF-365) antibodies; and FGF-2 (DY233-05), G-CSF (DY214-05), GM-CSF (DY215-05), and CX3CL1 (DY365) ELISA kits were bought from R&D Systems.
All the patient serum or plasma samples were collected at the Austin Hospital (Melbourne) under approved human ethic protocols and written informed consents were obtained from all patients before sample collection. Ethics approval was obtained from The University of Queensland Institutional Human Research Ethics Committee (approval nos. 2011001315 and 2016000876) and the following clinical assay was carried out according to the approved guidelines.
Preparation of single-particle active SERS nanotags
The preparation of SERS nanotags involved the synthesis of nanoboxes and the subsequent functionalisation with Raman reporters and antibodies. For nanobox synthesis, 45 µL of HAuCl4 (1 wt%) was added into 10 mL of ultrapure H2O (18.2 Ω cm) under magnetic stirring (800 r.p.m.) for 1 min, followed by simultaneously introducing 170 µL of AgNO3 (6 mM) and 30 µL of AA (0.1 M) into the stirring solution. Then, the formation of nanoboxes was indicated by the appearance of an apparent blue colour within 6 s and the samples were collected 1 min later by centrifuging at 600 g for 15 min.
To functionalise nanoboxes with Raman reporters and antibodies, 300 µL of nanoboxes centrifuged from 1 mL of as-prepared solution were co-incubated with one type of Raman reporters (i.e., 10 µL of DTNB, 8 µL of MBA, 10 µL of TFMBA, or 10 µL of MMTAA) and 2 µL of DSP linker for 6 h. After that, the Raman reporter and DSP functionalised nanoboxes were separated by centrifuging at 600 g for 15 min, and resuspended into 300 µL of PBS (0.1 mM). Then, 2 µg of anti-FGF-2, anti-G-CSF, anti-GM-CSF, anti-CX3CL1 antibodies were added to MBA, DTNB, TFMBA, and MMTAA labelled nanoboxes, respectively. After overnight incubation at 4 oC, the functionalised nanoboxes were purified by centrifuging at 600×g for 15 min to separate free antibodies and the final products were resuspended into 200 µL of 0.1% BSA for future use.
Fabrication of pillar arrays
The chip was made of a sensing array measuring 1 mm × 1 mm (Supplementary Fig. 1a) and consisted of 250,000 individual pillars. Each pillar was 1 µm wide, 1 µm long, and 1 µm high. The pillars were evenly spaced by 1 µm from one pillar to the next (Supplementary Fig. 1b). The pillar array was designed using Nanosuite 6.0 (Raith GmbH) and Beamer 5.9.1 (GenISys GmbH) and fabricated on a 4-inch p-type <100> silicon wafer (Bonda Technology Pte Ltd, Singapore) using electron beam lithography (EBL). The wafer accommodated 76 separate pillar arrays. Silicon wafer was first cleaned in acetone, isopropanol with sonication for 2 min each, followed by rising with deionised H2O and dehydration bake at 180 °C for 2 min. Prior to the resist coating, the wafer had undergone a further O2 plasma cleaning at 200 W for 5 min (Diener Atto, Diener Electronic GmbH). The cleaned wafer was spin-coated with two layers of polymethyl methacrylate (bottom: 495K A4 PMMA, top: 950k A4 PMMA, from MicroChemicals GmbH) using the CEE Apogee Coater (Cost Effective Equipment, LLC) at 1500 r.p.m. for 60 s each. After the coating of each layer, the wafer was baked immediately on a hot plate to prevent from intermixing of the two layers of resist. The baking time was 10 and 3 min for the bottom and top layer, respectively, at 180 °C. The thickness of the photoresist was found to be ~450 nm (top PMMA: ~250 nm, bottom PMMA: ~200 nm), characterised by white light reflectometry (FilmTek 2000M, Scientific Computing International). EBL was performed in the Raith EBPG5150 system. The patterns were exposed in EBL with an accelerated voltage of 100 kV, 150 nA of beam current (spot size ~80 nm), with step sizes of 40 nm and an electron dose of 1200 µC/cm2. The exposure time per 4-inch wafer was ~35 min, each containing 76 individual chips. After exposure, the wafer was developed in a mixture of isopropanol and methyl isobutyl ketone (3:1) for 60 s and rinsed immediately with isopropanol, followed by drying with N2. An oxygen plasma descum process, at 100 W, 60 s (Diener Atto, Diener Electronic GmbH) was carried out to remove resist residues prior to the deposition. Next, 10 nm titanium and 200 nm gold were deposited by physical vapour deposition using a Temescal FC-2000 electron beam evaporator (Ferrotec, U.S.A.). After overnight lift-off at room temperature in Remover PG (MicroChemicals GmbH, Germany), the excess material was washed off and the pillar array structure was revealed (Supplementary Fig. 1c). To create the pillar height (i.e., 1 µm), reactive ion etching (Oxford Instruments, UK) was applied for anisotropic etching of the silicon. Hereby, the deposited gold served as mask to protect the underlying silicon while the un-masked silicon was removed. Next, the wafer was coated with a protective layer of cured AZnLOF 2020 prior to wafer dicing into 76 individual sensing chips consisting of a single pillar array. Prior to use, the protective layer was washed off by consecutive washes with isopropanol and acetone and dried under a stream of nitrogen.
Pillar array functionalization
Antibody functionalisation of the gold-topped pillar array was conducted by crosslinking the antibodies to the gold surface using DSP. A solution of 5 mM DSP in dimethyl sulfoxide was pipetted onto the pillar array and incubated at room temperature for 2 h. After rinsing the pillar array with ethanol and PBS, a solution of 5 µg/mL anti-cytokine monoclonal antibody solution (100 fold dilution of antibody stock solution) in PBS was incubated overnight at 4 °C. Subsequently, the pillar array was rinsed with PBS and blocked using 1% bovine serum albumin in PBS for 1 h. Prior to use, the pillar array was rinsed with PBS. All PBS solutions were filtered through a sterile 0.22 µm syringe filter (Millex-GP, Merck, U.S.A.).
Digital nanopillar SERS profiling of cytokines
Cytokines (FGF-2, G-CSF, GM-CSF, and CX3CL1) with different concentrations in PBS (2.6 aM, 26 aM, 260 aM, and 1031 aM) were incubated with antibody functionalised pillar arrays at room temperature for 30 min, followed by washing the pillar array three times with washing buffer (0.1% BSA and 0.01% Tween 20 in PBS). SERS nanotags were then added into the pillar array for another 30 min incubation under room temperature to identify the targets. Finally, the pillar arrays were washed to remove the free SERS nanotags and were subject to confocal Raman microscope for quantification. For each sample, nine SERS images with each image has the dimension of 60 µm × 48 µm were taken on the pillar array to calculate the overall cytokine concentration.
Simulated clinical sample detection
For the recovery experiment, standard cytokines (FGF-2, G-CSF, GM-CSF, and CX3CL1) with the concentration of 1 fM were added into healthy human serum and then diluted ten times with PBS to quantify.
For the quantification of cytokines in FBS, three simulated clinical samples were prepared by titrating various concentrations of standard cytokines into 10% FBS: Sample 1 (FGF-2 = 3.64 pM, G-CSF = 3.19 pM, GM-CSF = 4.29 pM, and CX3CL1 = 28.57 fM); Sample 2 (FGF-2 = 7.28 pM, G-CSF = 6.38 pM, GM-CSF = 8.58 pM, and CX3CL1 = 571.4 fM); and Sample 3 (FGF-2 = 14.56 pM, G-CSF = 12.76 pM, GM-CSF = 17.16 pM, and CX3CL1 = 1142.8 fM). These three samples were then detected directly using the commercial ELISA kits or digital nanopillar SERS assay with a further dilution of 105, 2 × 105, and 4 × 105, respectively.
Spectroscopic ellipsometry
The antibody film thickness was measured by in-solution spectroscopic ellipsometry (M2000V JA Woollam Co., Inc. USA) using gold-coated substrates and flow cell (QSense® Ellipsometry, Biolin Scientific, Sweden). Measurements were performed at an angle of 65°. Data analysis was performed by CompleteEASE® software using a B-Spline data fit and Cauchy model to calculate the antibody film thickness.
MALDI-TOF MS
The antibody-functionalised nanopillar array chip was subjected to tryptic digest prior to analysis. Sequencing-grade trypsin was made to 50 ng/µL in 25 mM ammonium bicarbonate, and sprayed over the chip using a Bruker Imageprep instrument (Bruker, USA). After trypsin deposition, the chip was incubated in a humid environment at 40 °C for 3 h. Subsequently, the chip was sprayed with a matrix solution, 10 g/L α-cyano-4-hydroxycinnamic acid in 50% acetonitrile with 0.2% trifluoroacetic acid. Next, the chip was analysed with a Bruker Ultraflex III MALDI-TOF mass spectrometer (Bruker, USA) in positive linear mode using Flex Imaging 4.0 (Bruker, USA) with a pixel size of 60 µm. Data were collected from 2 k–30 k m/z, at a laser repetition rate of 200 Hz. Data were normalised using the root mean square approach and visualised using Flex Imaging 4.0 (Bruker, USA) and SCILS LAB 2017a software. For the SCILS LAB analysis, the data were imported using a convolution baseline subtraction, and displayed using root mean squared normalisation.
Instrumentations
SEM images of pillar arrays and nanoboxes were taken on a JEOL-7100 field emission (FE)-SEM (20 kV voltage). TEM images of nanoboxes were taken on a JEOL-2100 microscope (200 kV voltage). NTA of nanobox size distribution was performed with Malvern NanoSight NS300. UV-vis extinction spectrum of nanoboxes was performed with a Shimadzu UV-2450 spectrophotometer. Confocal Raman mapping was conducted on a WITec alpha 300 R spectrometer using 632.8 nm He–Ne laser with the power of 35 mW, a grating of 600 g/mm used with EMCCD camera, spectral resolution of 1.390 cm−1 to 2.114 cm−1, confocal pinhole size of 100 µm, 100× air objective with NA of 0.90, and 0.05 s integration time. The theoretical spot size was 857.80 nm based on the Abbe diffraction limit (i.e., d = 1.22λ/NA). The scanning area was set to have 60 µm × 48 µm with 86 points per line and 69 lines per image. For each pillar array, nine separate scanning areas were taken in total and the total active pillars were used for quantification. The SERS mapping images for counting were taken by focusing the laser on the top of the pillar surfaces. Specifically, the laser was firstly focused on the silicon substrates by obtaining the strongest silicon signals (520 cm−1) and then the 100× objective was moved up in z-axis direction of 1 µm for SERS scanning. The system was calibrated with the first-order photo peak of silicon at 520 cm−1.
Data analysis
To assign the SERS nanotag membership for DTNB, MBA, TFMBA, and MMTAA, Project Five 5.0 software from WITec was utilised to create four filters, which summed a spectral range of 40 cm−1 with the centre position at the characteristic Raman peak of each reporter and subtracted the background with a polynomial algorithm. Specifically, the filter ranges of four Raman reporters DTNB-, MBA-, TFMBA-, and MMTAA-coated SERS nanotags were (1310–1350 cm−1), (1060–1100 cm−1), (1360–1400 cm−1), and (1268–1308 cm−1), respectively. All the SERS images were analysed by using threshold intensity to determine the successful binding events. Specifically, the threshold intensity of FGF-2, G-CSF, GM-CSF, and CX3CL1 was set at 5000, 4000, 5000, and 5000, respectively. For each image, the threshold intensity was doubled-checked and adjusted based on the true Raman peaks in the spectra. Statistical analysis assuming unequal variances was conducted with Kruskal–Wallis test among three groups or Mann–Whitney test between two groups with GraphPad Prism 8.4. To control the error appropriately, we performed multiple comparisons using Dunn’s test. LDA of clinical samples was performed in R software (3.6.2) with the MASS package (7.3-52). The active pillars in SERS images were counted with Image J software.
Supplementary information
Supplementary Information
Peer Review File
Source data
Source Data
Peer review information
Nature Communications thanks Chongwen Wang and Isaac Pence for their contribution to the peer review of this work. Peer reviewer reports are available.
Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
Supplementary information
The online version contains supplementary material available at 10.1038/s41467-021-21431-w.
Acknowledgements
The authors acknowledge grants received by our laboratory from the National Breast Cancer Foundation of Australia (CG-12-07) and the ARC DP (160102836 and 210103151). These grants have significantly contributed to the environment to stimulate the research described here. J.L. acknowledges support from the Australian Government Research Training Program Scholarship. A.W. and A.A.I.S. thank the National Health and Medical Research Council for funding (APP1173669 and APP1175047). A.B. is the recipient of a Fellowship from the Victorian Government Department of Health and Human Services acting through the Victorian Cancer Agency. We acknowledge the facilities, and the scientific and technical assistance, of the Australian Microscopy & Microanalysis Research Facility at the Centre for Microscopy and Microanalysis, The University of Queensland. We appreciate to receive the technical and scientific guidance from Queensland node of the Australian National Fabrication Facility (Q-ANFF) in confocal Raman mapping and spectroscopic ellipsometry measurement.
Author contributions
J.L., A.W., A.I.S., H-H.C., Y.W., A.B., P.M., and M.T. contributed to the design of the experiments and analysing the data. J.L. and A.W. performed the experiments and prepared the manuscript. All authors read, commented, and edited the manuscript, and assisted during the revision process.
Data availability
Data supporting the findings of this work are available within this paper and the supporting information files. A reporting summary of this work is available as a Supplementary file. Source data are provided with this paper.
Competing interests
The authors declare no competing interests. | Recovering | ReactionOutcome | CC BY | 33597530 | 19,690,449 | 2021-02-17 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Transient aphasia'. | Transient aphasia following general anesthesia in patient undergoing laparoscopic gynecologic surgery: A case report and literature review.
Aphasia could be a manifestation of postoperative psychoemotional disorder. Early differential diagnosis is important.
1 INTRODUCTION
This case report describes a healthy 45‐year‐old woman who received total intravenous anesthesia for laparoscopic hysterectomy, and then exhibited postoperative transient aphasia. Droperidol was given and the patient resolved. The author hypothesized that aphasia is a manifestation of psychoemotional disorder. Early differential diagnosis is important.
Laparoscopic gynecological surgery has been widely employed under general anesthesia. Although the safety and effectiveness of these techniques have greatly improved over the past decades, some problematic effects may still occur. Lithotomy position, Trendelenburg position, and CO2 pneumoperitoneum may increase the risk of neurologic complications.
1
,
2
,
3
,
4
There have been isolated case reports of psychogenic language disorder during surgery.
5
However, cognitive and neurologic complications after prolonged Trendelenburg position are extremely rare and more often reported in elderly patients, or those with underlying venous disease.
6
Furthermore, aphasia is rarely reported in perioperative settings. In this report, we present a patient who developed transient aphasia after extubation after laparoscopic hysterectomy. The patient provided written consent for this case report.
2 CASE PRESENTATION
A 45‐year‐old woman was admitted for laparoscopic hysterectomy under general anesthesia. She weighed 60 kg and was 162 cm tall. She had no history of psychiatric illness, medical disease, or drug abuse. Her physical examination did not reveal any significant findings. The preoperative electrocardiography and laboratory data, which included complete blood count, liver function, kidney function, and coagulation profile, were unremarkable.
Upon arrival in the operating room, routine monitoring, including electrocardiograph (ECG) monitoring, noninvasive blood pressure, and pulse oximetry, was applied. Lactated Ringer's solution (500 mL) was rapidly infused intravenously during the induction of anesthesia. Scopolamine (0.3 mg) and dexamethasone (5 mg) were administered before induction. Propofol (2 mg/kg) and fentanyl (4 µg/kg) were used for induction. When the patient lost consciousness, 0.6 mg/kg of rocuronium was given to facilitate tracheal intubation. Anesthesia was maintained by the continuous infusion of propofol (6‐7 mg/kg/h) and remifentanil (0.15‐0.2 µg/kg/min). The patient was placed in the Trendelenburg position, and intra‐abdominal pressure was maintained at 11‐14 cmH2O during surgery. After surgery, the patient was extubated when spontaneous breathing and consciousness recovered. The surgery lasted 80 minutes. Neuromuscular blockade was monitored by TOF‐Watch, and no additional rocuronium was given during surgery.
After extubation, the patient manifested symptoms of dyspnea, showing repeated effort to breathe. She was a little anxious but could obey loud verbal commands, such as opening the mouth and holding the anesthetist's hand. The TOF value was above 90% though our first concern was a possible weakness of muscle tone. We gave the patient neostigmine (0.04 mg/kg) and atropine (0.02 mg/kg). However, the patient became more anxious and irritable. We tried to pacify her with reassuring remarks that the surgery was successful and that she would be sent back to her husband soon. However, the patient frequently pointed to her throat, and we began to notice that she had difficulty speaking. She remained conscious and seemed to understand what we were saying to her. She tried to answer but could not utter a word. She could move her mouth, lips, tongue, head, and extremities upon command. Her pupils were equal in size and reactive to light. Her lungs were clear, and vital signs were stable. Arterial blood gas analysis showed no abnormal results. The patient asked for paper by gesture and wrote down the following words, “I want to speak, but I can't say a word” and “help me.” The patient was getting increasingly anxious, became uncooperative, and was in tears.
3 DIAGNOSIS AND TREATMENT
We administered 2.5 mg of droperidol, and she fell asleep. When she woke up 20 minutes later, she could speak in a hoarse voice and was transferred to the ward. She became more peaceful when she saw her husband. After 4 hours, the patient was able to talk normally. The aphasia lasted for 1 hour, and the whole episode lasted for 5.5 hours. She told the anesthetist that she could recall the events during recovery and felt happy to be able to talk again. She recalled that approximately 4 years ago, she experienced an episode of transient aphasia after a bitter quarrel with her husband.
4 OUTCOME AND FOLLOW‐UP
A neurology consultation suggested that the patient should undergo more extensive studies before any definite diagnosis could be made. Therefore, imaging (MRI) was arranged immediately. No definite organic lesions were found in an MRI of the brain. She was discharged without sequelae on postsurgery day 3. She was followed up for 3 months, and no complications were reported.
5 DISCUSSION
The differential diagnosis of such a clinical presentation of sudden and transient aphasia could be related to a residual effect of the anesthesia, transient ischemic attack (TIA), structural injury of vocal function,
7
or psychoemotional change.
8
,
9
The available literature on acute aphasia after anesthesia is limited.
Transient ischemic attack is a transient neurological event that usually persists for minutes and can occur due to hypotension, hypoxia, or embolism. However, the patient did not have hypoxia or hypotension. Thromboembolism was considered because the patient was placed in the lithotomy position. Lithotomy position and CO2 pneumoperitoneum may result in increased lower extremity venous pressure during hysterectomy, representing a risk factor for deep venous thrombosis.
10
And it was reported that steep Trendelenburg, along with several other risk factors, increased the odds of venous thromboembolism formation by 2.4 [95% confidence interval (95% CI) 1.9‐5.0].
11
When TIA is suspected, the “ABCD2” score could be used for the assessment of early stroke risk (“ABCD2” score ≥ 4).
12
The “ABCD2” score in this patient was 3 (speech disturbance without weakness [1 point] and duration of symptoms > 60 minutes [2 points]). And the patient was with a Glasgow score of 15 and a normal MRI results after surgery. The patient was followed up for 3 months because there is a 10%‐20% risk of major stroke after TIA in the subsequent 3 months.
13
All the evidences supported that the patient did not suffer from TIA.
There are several reports of acute neurological events, including cerebral edema, in patients undergoing laparoscopic gynecological surgery.
2
,
3
Any rise in arterial carbon dioxide can possibly result in increased cerebral blood flow. A steep Trendelenburg position can alter the cerebral blood volume and, thus, increase intracranial pressure. Moreover, head‐down tilt causes cephalad fluid shift and increases capillary pressure in the head, leading to an increase in extracranial vascular pressure that may cause facial, laryngeal, pharyngeal edema, and nasal congestion.
1
,
4
However, in our case, the carbon dioxide levels were measured by capnography and were kept at the normal limit. The surgery lasted for 80 minutes, and no significant facial edema existed. Because MRI was normal, brain edema was not considered.
Residual neuromuscular blockade could affect the speaking function. The patient could write down what she wanted to say, appropriately response to verbal command, indicating well‐reversed neuromuscular blockade. Structural injury, including the luxation of the arytenoid cartilage and laryngeal nerve injury, may occur after tracheal intubation and affect pronunciation.
7
,
14
,
15
The aphonia or aphasia in these cases is usually consistent and requires treatment. However, for our case the aphasia was transient, so structural damage was excluded. Pre‐existing neurologic disease should be screened for postoperative speaking dysfunction. Willson et al
16
reported a patient with hemiplegic migraine who developed atypical migraine with apneic spells, aphasia, and hemiparesis following general anesthesia. By history and MRI assessment, these possibilities were excluded in our case.
Postoperative delirium may present as speaking disorder. Medications including midazolam and scopolamine have been proved to increase delirium.
17
,
18
And in a case of a healthy 12‐year‐old girl who received preoperative midazolam, Drobish et al
19
reported postoperative transient associative agnosia and expressive aphasia. The administration of flumazenil led to the immediate and lasting resolution of her symptoms. Confusion assessment method (CAM) is the most often used tool for delirium evaluation. The diagnosis of delirium by CAM requires the presence of features 1 (acute onset and fluctuation course) and 2 (inattention) and either 3 (disorganized thinking) or 4 (altered level of consciousness). Given the presentation as it is, the patient does not probably meet these criteria. Though the EEG was not recorded during the episode, the Brief Psychiatric Rating Scale of the patient was 10, which did not support the diagnosis of psychiatric disorder. Moreover, Broca's aphasia patient's cannot speak or write and with Wernicke's they can speak but cannot convey a clear message. The presentation of the patient may support that the aphasia is manifestation of psychological in nature. Droperidol was effective for the sedation of the patient; however, ruling out dangerous episodes by CT scan or MRI maybe important before droperidol administration.
Psychogenic cognitive disturbance varies in presentation, including aphasia, aphagia, and quadriplegia.
8
,
20
,
21
In this case, the patient stayed awake and remembered the entire recovery process. The patient has a history of transient aphasia after a bitter quarrel with her husband 4 years ago. Stress and anxiety could have triggered her aphasia. She could be soothed by seeing her family, which highlights the importance of emotional support in treating perioperative psychogenic disorders. Enhanced pre‐operative education and early company of the family may improve the outcome.
6 CONCLUSION
In summary, transient aphasia can occur after general anesthesia. The possibility of cerebral complications and the injury of the laryngeal‐pharyngeal structures should be ruled out in appropriate clinical settings. A previous history of psychoemotional disorder should not be dismissed when reviewing a patient preoperatively.
CONFLICT OF INTEREST
None declared.
AUTHOR CONTRIBUTION
ZL and HD: were the patient's anesthetists, reviewed the literature, and contributed to manuscript revision; JZ and MY: reviewed the literature and contributed to manuscript drafting; all authors issued final approval for the version to be submitted.
ETHICAL APPROVAL
The patient provided signed informed consent to the publishment of the case. All the procedures performed in this study were in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.
ACKNOWLEDGMENTS
We thoroughly appreciated the patient and her family for collaboration with us. Consent statement: Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
Data were not available for share. | BUTYLSCOPOLAMINE BROMIDE, CALCIUM CHLORIDE\POTASSIUM CHLORIDE\SODIUM LACTATE, DEXAMETHASONE, FENTANYL, PROPOFOL, REMIFENTANIL, ROCURONIUM BROMIDE | DrugsGivenReaction | CC BY-NC-ND | 33598216 | 19,156,275 | 2021-02 |
What is the weight of the patient? | Transient aphasia following general anesthesia in patient undergoing laparoscopic gynecologic surgery: A case report and literature review.
Aphasia could be a manifestation of postoperative psychoemotional disorder. Early differential diagnosis is important.
1 INTRODUCTION
This case report describes a healthy 45‐year‐old woman who received total intravenous anesthesia for laparoscopic hysterectomy, and then exhibited postoperative transient aphasia. Droperidol was given and the patient resolved. The author hypothesized that aphasia is a manifestation of psychoemotional disorder. Early differential diagnosis is important.
Laparoscopic gynecological surgery has been widely employed under general anesthesia. Although the safety and effectiveness of these techniques have greatly improved over the past decades, some problematic effects may still occur. Lithotomy position, Trendelenburg position, and CO2 pneumoperitoneum may increase the risk of neurologic complications.
1
,
2
,
3
,
4
There have been isolated case reports of psychogenic language disorder during surgery.
5
However, cognitive and neurologic complications after prolonged Trendelenburg position are extremely rare and more often reported in elderly patients, or those with underlying venous disease.
6
Furthermore, aphasia is rarely reported in perioperative settings. In this report, we present a patient who developed transient aphasia after extubation after laparoscopic hysterectomy. The patient provided written consent for this case report.
2 CASE PRESENTATION
A 45‐year‐old woman was admitted for laparoscopic hysterectomy under general anesthesia. She weighed 60 kg and was 162 cm tall. She had no history of psychiatric illness, medical disease, or drug abuse. Her physical examination did not reveal any significant findings. The preoperative electrocardiography and laboratory data, which included complete blood count, liver function, kidney function, and coagulation profile, were unremarkable.
Upon arrival in the operating room, routine monitoring, including electrocardiograph (ECG) monitoring, noninvasive blood pressure, and pulse oximetry, was applied. Lactated Ringer's solution (500 mL) was rapidly infused intravenously during the induction of anesthesia. Scopolamine (0.3 mg) and dexamethasone (5 mg) were administered before induction. Propofol (2 mg/kg) and fentanyl (4 µg/kg) were used for induction. When the patient lost consciousness, 0.6 mg/kg of rocuronium was given to facilitate tracheal intubation. Anesthesia was maintained by the continuous infusion of propofol (6‐7 mg/kg/h) and remifentanil (0.15‐0.2 µg/kg/min). The patient was placed in the Trendelenburg position, and intra‐abdominal pressure was maintained at 11‐14 cmH2O during surgery. After surgery, the patient was extubated when spontaneous breathing and consciousness recovered. The surgery lasted 80 minutes. Neuromuscular blockade was monitored by TOF‐Watch, and no additional rocuronium was given during surgery.
After extubation, the patient manifested symptoms of dyspnea, showing repeated effort to breathe. She was a little anxious but could obey loud verbal commands, such as opening the mouth and holding the anesthetist's hand. The TOF value was above 90% though our first concern was a possible weakness of muscle tone. We gave the patient neostigmine (0.04 mg/kg) and atropine (0.02 mg/kg). However, the patient became more anxious and irritable. We tried to pacify her with reassuring remarks that the surgery was successful and that she would be sent back to her husband soon. However, the patient frequently pointed to her throat, and we began to notice that she had difficulty speaking. She remained conscious and seemed to understand what we were saying to her. She tried to answer but could not utter a word. She could move her mouth, lips, tongue, head, and extremities upon command. Her pupils were equal in size and reactive to light. Her lungs were clear, and vital signs were stable. Arterial blood gas analysis showed no abnormal results. The patient asked for paper by gesture and wrote down the following words, “I want to speak, but I can't say a word” and “help me.” The patient was getting increasingly anxious, became uncooperative, and was in tears.
3 DIAGNOSIS AND TREATMENT
We administered 2.5 mg of droperidol, and she fell asleep. When she woke up 20 minutes later, she could speak in a hoarse voice and was transferred to the ward. She became more peaceful when she saw her husband. After 4 hours, the patient was able to talk normally. The aphasia lasted for 1 hour, and the whole episode lasted for 5.5 hours. She told the anesthetist that she could recall the events during recovery and felt happy to be able to talk again. She recalled that approximately 4 years ago, she experienced an episode of transient aphasia after a bitter quarrel with her husband.
4 OUTCOME AND FOLLOW‐UP
A neurology consultation suggested that the patient should undergo more extensive studies before any definite diagnosis could be made. Therefore, imaging (MRI) was arranged immediately. No definite organic lesions were found in an MRI of the brain. She was discharged without sequelae on postsurgery day 3. She was followed up for 3 months, and no complications were reported.
5 DISCUSSION
The differential diagnosis of such a clinical presentation of sudden and transient aphasia could be related to a residual effect of the anesthesia, transient ischemic attack (TIA), structural injury of vocal function,
7
or psychoemotional change.
8
,
9
The available literature on acute aphasia after anesthesia is limited.
Transient ischemic attack is a transient neurological event that usually persists for minutes and can occur due to hypotension, hypoxia, or embolism. However, the patient did not have hypoxia or hypotension. Thromboembolism was considered because the patient was placed in the lithotomy position. Lithotomy position and CO2 pneumoperitoneum may result in increased lower extremity venous pressure during hysterectomy, representing a risk factor for deep venous thrombosis.
10
And it was reported that steep Trendelenburg, along with several other risk factors, increased the odds of venous thromboembolism formation by 2.4 [95% confidence interval (95% CI) 1.9‐5.0].
11
When TIA is suspected, the “ABCD2” score could be used for the assessment of early stroke risk (“ABCD2” score ≥ 4).
12
The “ABCD2” score in this patient was 3 (speech disturbance without weakness [1 point] and duration of symptoms > 60 minutes [2 points]). And the patient was with a Glasgow score of 15 and a normal MRI results after surgery. The patient was followed up for 3 months because there is a 10%‐20% risk of major stroke after TIA in the subsequent 3 months.
13
All the evidences supported that the patient did not suffer from TIA.
There are several reports of acute neurological events, including cerebral edema, in patients undergoing laparoscopic gynecological surgery.
2
,
3
Any rise in arterial carbon dioxide can possibly result in increased cerebral blood flow. A steep Trendelenburg position can alter the cerebral blood volume and, thus, increase intracranial pressure. Moreover, head‐down tilt causes cephalad fluid shift and increases capillary pressure in the head, leading to an increase in extracranial vascular pressure that may cause facial, laryngeal, pharyngeal edema, and nasal congestion.
1
,
4
However, in our case, the carbon dioxide levels were measured by capnography and were kept at the normal limit. The surgery lasted for 80 minutes, and no significant facial edema existed. Because MRI was normal, brain edema was not considered.
Residual neuromuscular blockade could affect the speaking function. The patient could write down what she wanted to say, appropriately response to verbal command, indicating well‐reversed neuromuscular blockade. Structural injury, including the luxation of the arytenoid cartilage and laryngeal nerve injury, may occur after tracheal intubation and affect pronunciation.
7
,
14
,
15
The aphonia or aphasia in these cases is usually consistent and requires treatment. However, for our case the aphasia was transient, so structural damage was excluded. Pre‐existing neurologic disease should be screened for postoperative speaking dysfunction. Willson et al
16
reported a patient with hemiplegic migraine who developed atypical migraine with apneic spells, aphasia, and hemiparesis following general anesthesia. By history and MRI assessment, these possibilities were excluded in our case.
Postoperative delirium may present as speaking disorder. Medications including midazolam and scopolamine have been proved to increase delirium.
17
,
18
And in a case of a healthy 12‐year‐old girl who received preoperative midazolam, Drobish et al
19
reported postoperative transient associative agnosia and expressive aphasia. The administration of flumazenil led to the immediate and lasting resolution of her symptoms. Confusion assessment method (CAM) is the most often used tool for delirium evaluation. The diagnosis of delirium by CAM requires the presence of features 1 (acute onset and fluctuation course) and 2 (inattention) and either 3 (disorganized thinking) or 4 (altered level of consciousness). Given the presentation as it is, the patient does not probably meet these criteria. Though the EEG was not recorded during the episode, the Brief Psychiatric Rating Scale of the patient was 10, which did not support the diagnosis of psychiatric disorder. Moreover, Broca's aphasia patient's cannot speak or write and with Wernicke's they can speak but cannot convey a clear message. The presentation of the patient may support that the aphasia is manifestation of psychological in nature. Droperidol was effective for the sedation of the patient; however, ruling out dangerous episodes by CT scan or MRI maybe important before droperidol administration.
Psychogenic cognitive disturbance varies in presentation, including aphasia, aphagia, and quadriplegia.
8
,
20
,
21
In this case, the patient stayed awake and remembered the entire recovery process. The patient has a history of transient aphasia after a bitter quarrel with her husband 4 years ago. Stress and anxiety could have triggered her aphasia. She could be soothed by seeing her family, which highlights the importance of emotional support in treating perioperative psychogenic disorders. Enhanced pre‐operative education and early company of the family may improve the outcome.
6 CONCLUSION
In summary, transient aphasia can occur after general anesthesia. The possibility of cerebral complications and the injury of the laryngeal‐pharyngeal structures should be ruled out in appropriate clinical settings. A previous history of psychoemotional disorder should not be dismissed when reviewing a patient preoperatively.
CONFLICT OF INTEREST
None declared.
AUTHOR CONTRIBUTION
ZL and HD: were the patient's anesthetists, reviewed the literature, and contributed to manuscript revision; JZ and MY: reviewed the literature and contributed to manuscript drafting; all authors issued final approval for the version to be submitted.
ETHICAL APPROVAL
The patient provided signed informed consent to the publishment of the case. All the procedures performed in this study were in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.
ACKNOWLEDGMENTS
We thoroughly appreciated the patient and her family for collaboration with us. Consent statement: Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
Data were not available for share. | 60 kg. | Weight | CC BY-NC-ND | 33598216 | 19,156,275 | 2021-02 |
What was the administration route of drug 'CALCIUM CHLORIDE\POTASSIUM CHLORIDE\SODIUM LACTATE'? | Transient aphasia following general anesthesia in patient undergoing laparoscopic gynecologic surgery: A case report and literature review.
Aphasia could be a manifestation of postoperative psychoemotional disorder. Early differential diagnosis is important.
1 INTRODUCTION
This case report describes a healthy 45‐year‐old woman who received total intravenous anesthesia for laparoscopic hysterectomy, and then exhibited postoperative transient aphasia. Droperidol was given and the patient resolved. The author hypothesized that aphasia is a manifestation of psychoemotional disorder. Early differential diagnosis is important.
Laparoscopic gynecological surgery has been widely employed under general anesthesia. Although the safety and effectiveness of these techniques have greatly improved over the past decades, some problematic effects may still occur. Lithotomy position, Trendelenburg position, and CO2 pneumoperitoneum may increase the risk of neurologic complications.
1
,
2
,
3
,
4
There have been isolated case reports of psychogenic language disorder during surgery.
5
However, cognitive and neurologic complications after prolonged Trendelenburg position are extremely rare and more often reported in elderly patients, or those with underlying venous disease.
6
Furthermore, aphasia is rarely reported in perioperative settings. In this report, we present a patient who developed transient aphasia after extubation after laparoscopic hysterectomy. The patient provided written consent for this case report.
2 CASE PRESENTATION
A 45‐year‐old woman was admitted for laparoscopic hysterectomy under general anesthesia. She weighed 60 kg and was 162 cm tall. She had no history of psychiatric illness, medical disease, or drug abuse. Her physical examination did not reveal any significant findings. The preoperative electrocardiography and laboratory data, which included complete blood count, liver function, kidney function, and coagulation profile, were unremarkable.
Upon arrival in the operating room, routine monitoring, including electrocardiograph (ECG) monitoring, noninvasive blood pressure, and pulse oximetry, was applied. Lactated Ringer's solution (500 mL) was rapidly infused intravenously during the induction of anesthesia. Scopolamine (0.3 mg) and dexamethasone (5 mg) were administered before induction. Propofol (2 mg/kg) and fentanyl (4 µg/kg) were used for induction. When the patient lost consciousness, 0.6 mg/kg of rocuronium was given to facilitate tracheal intubation. Anesthesia was maintained by the continuous infusion of propofol (6‐7 mg/kg/h) and remifentanil (0.15‐0.2 µg/kg/min). The patient was placed in the Trendelenburg position, and intra‐abdominal pressure was maintained at 11‐14 cmH2O during surgery. After surgery, the patient was extubated when spontaneous breathing and consciousness recovered. The surgery lasted 80 minutes. Neuromuscular blockade was monitored by TOF‐Watch, and no additional rocuronium was given during surgery.
After extubation, the patient manifested symptoms of dyspnea, showing repeated effort to breathe. She was a little anxious but could obey loud verbal commands, such as opening the mouth and holding the anesthetist's hand. The TOF value was above 90% though our first concern was a possible weakness of muscle tone. We gave the patient neostigmine (0.04 mg/kg) and atropine (0.02 mg/kg). However, the patient became more anxious and irritable. We tried to pacify her with reassuring remarks that the surgery was successful and that she would be sent back to her husband soon. However, the patient frequently pointed to her throat, and we began to notice that she had difficulty speaking. She remained conscious and seemed to understand what we were saying to her. She tried to answer but could not utter a word. She could move her mouth, lips, tongue, head, and extremities upon command. Her pupils were equal in size and reactive to light. Her lungs were clear, and vital signs were stable. Arterial blood gas analysis showed no abnormal results. The patient asked for paper by gesture and wrote down the following words, “I want to speak, but I can't say a word” and “help me.” The patient was getting increasingly anxious, became uncooperative, and was in tears.
3 DIAGNOSIS AND TREATMENT
We administered 2.5 mg of droperidol, and she fell asleep. When she woke up 20 minutes later, she could speak in a hoarse voice and was transferred to the ward. She became more peaceful when she saw her husband. After 4 hours, the patient was able to talk normally. The aphasia lasted for 1 hour, and the whole episode lasted for 5.5 hours. She told the anesthetist that she could recall the events during recovery and felt happy to be able to talk again. She recalled that approximately 4 years ago, she experienced an episode of transient aphasia after a bitter quarrel with her husband.
4 OUTCOME AND FOLLOW‐UP
A neurology consultation suggested that the patient should undergo more extensive studies before any definite diagnosis could be made. Therefore, imaging (MRI) was arranged immediately. No definite organic lesions were found in an MRI of the brain. She was discharged without sequelae on postsurgery day 3. She was followed up for 3 months, and no complications were reported.
5 DISCUSSION
The differential diagnosis of such a clinical presentation of sudden and transient aphasia could be related to a residual effect of the anesthesia, transient ischemic attack (TIA), structural injury of vocal function,
7
or psychoemotional change.
8
,
9
The available literature on acute aphasia after anesthesia is limited.
Transient ischemic attack is a transient neurological event that usually persists for minutes and can occur due to hypotension, hypoxia, or embolism. However, the patient did not have hypoxia or hypotension. Thromboembolism was considered because the patient was placed in the lithotomy position. Lithotomy position and CO2 pneumoperitoneum may result in increased lower extremity venous pressure during hysterectomy, representing a risk factor for deep venous thrombosis.
10
And it was reported that steep Trendelenburg, along with several other risk factors, increased the odds of venous thromboembolism formation by 2.4 [95% confidence interval (95% CI) 1.9‐5.0].
11
When TIA is suspected, the “ABCD2” score could be used for the assessment of early stroke risk (“ABCD2” score ≥ 4).
12
The “ABCD2” score in this patient was 3 (speech disturbance without weakness [1 point] and duration of symptoms > 60 minutes [2 points]). And the patient was with a Glasgow score of 15 and a normal MRI results after surgery. The patient was followed up for 3 months because there is a 10%‐20% risk of major stroke after TIA in the subsequent 3 months.
13
All the evidences supported that the patient did not suffer from TIA.
There are several reports of acute neurological events, including cerebral edema, in patients undergoing laparoscopic gynecological surgery.
2
,
3
Any rise in arterial carbon dioxide can possibly result in increased cerebral blood flow. A steep Trendelenburg position can alter the cerebral blood volume and, thus, increase intracranial pressure. Moreover, head‐down tilt causes cephalad fluid shift and increases capillary pressure in the head, leading to an increase in extracranial vascular pressure that may cause facial, laryngeal, pharyngeal edema, and nasal congestion.
1
,
4
However, in our case, the carbon dioxide levels were measured by capnography and were kept at the normal limit. The surgery lasted for 80 minutes, and no significant facial edema existed. Because MRI was normal, brain edema was not considered.
Residual neuromuscular blockade could affect the speaking function. The patient could write down what she wanted to say, appropriately response to verbal command, indicating well‐reversed neuromuscular blockade. Structural injury, including the luxation of the arytenoid cartilage and laryngeal nerve injury, may occur after tracheal intubation and affect pronunciation.
7
,
14
,
15
The aphonia or aphasia in these cases is usually consistent and requires treatment. However, for our case the aphasia was transient, so structural damage was excluded. Pre‐existing neurologic disease should be screened for postoperative speaking dysfunction. Willson et al
16
reported a patient with hemiplegic migraine who developed atypical migraine with apneic spells, aphasia, and hemiparesis following general anesthesia. By history and MRI assessment, these possibilities were excluded in our case.
Postoperative delirium may present as speaking disorder. Medications including midazolam and scopolamine have been proved to increase delirium.
17
,
18
And in a case of a healthy 12‐year‐old girl who received preoperative midazolam, Drobish et al
19
reported postoperative transient associative agnosia and expressive aphasia. The administration of flumazenil led to the immediate and lasting resolution of her symptoms. Confusion assessment method (CAM) is the most often used tool for delirium evaluation. The diagnosis of delirium by CAM requires the presence of features 1 (acute onset and fluctuation course) and 2 (inattention) and either 3 (disorganized thinking) or 4 (altered level of consciousness). Given the presentation as it is, the patient does not probably meet these criteria. Though the EEG was not recorded during the episode, the Brief Psychiatric Rating Scale of the patient was 10, which did not support the diagnosis of psychiatric disorder. Moreover, Broca's aphasia patient's cannot speak or write and with Wernicke's they can speak but cannot convey a clear message. The presentation of the patient may support that the aphasia is manifestation of psychological in nature. Droperidol was effective for the sedation of the patient; however, ruling out dangerous episodes by CT scan or MRI maybe important before droperidol administration.
Psychogenic cognitive disturbance varies in presentation, including aphasia, aphagia, and quadriplegia.
8
,
20
,
21
In this case, the patient stayed awake and remembered the entire recovery process. The patient has a history of transient aphasia after a bitter quarrel with her husband 4 years ago. Stress and anxiety could have triggered her aphasia. She could be soothed by seeing her family, which highlights the importance of emotional support in treating perioperative psychogenic disorders. Enhanced pre‐operative education and early company of the family may improve the outcome.
6 CONCLUSION
In summary, transient aphasia can occur after general anesthesia. The possibility of cerebral complications and the injury of the laryngeal‐pharyngeal structures should be ruled out in appropriate clinical settings. A previous history of psychoemotional disorder should not be dismissed when reviewing a patient preoperatively.
CONFLICT OF INTEREST
None declared.
AUTHOR CONTRIBUTION
ZL and HD: were the patient's anesthetists, reviewed the literature, and contributed to manuscript revision; JZ and MY: reviewed the literature and contributed to manuscript drafting; all authors issued final approval for the version to be submitted.
ETHICAL APPROVAL
The patient provided signed informed consent to the publishment of the case. All the procedures performed in this study were in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.
ACKNOWLEDGMENTS
We thoroughly appreciated the patient and her family for collaboration with us. Consent statement: Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
Data were not available for share. | Intravenous (not otherwise specified) | DrugAdministrationRoute | CC BY-NC-ND | 33598216 | 19,156,275 | 2021-02 |
What was the administration route of drug 'FENTANYL'? | Transient aphasia following general anesthesia in patient undergoing laparoscopic gynecologic surgery: A case report and literature review.
Aphasia could be a manifestation of postoperative psychoemotional disorder. Early differential diagnosis is important.
1 INTRODUCTION
This case report describes a healthy 45‐year‐old woman who received total intravenous anesthesia for laparoscopic hysterectomy, and then exhibited postoperative transient aphasia. Droperidol was given and the patient resolved. The author hypothesized that aphasia is a manifestation of psychoemotional disorder. Early differential diagnosis is important.
Laparoscopic gynecological surgery has been widely employed under general anesthesia. Although the safety and effectiveness of these techniques have greatly improved over the past decades, some problematic effects may still occur. Lithotomy position, Trendelenburg position, and CO2 pneumoperitoneum may increase the risk of neurologic complications.
1
,
2
,
3
,
4
There have been isolated case reports of psychogenic language disorder during surgery.
5
However, cognitive and neurologic complications after prolonged Trendelenburg position are extremely rare and more often reported in elderly patients, or those with underlying venous disease.
6
Furthermore, aphasia is rarely reported in perioperative settings. In this report, we present a patient who developed transient aphasia after extubation after laparoscopic hysterectomy. The patient provided written consent for this case report.
2 CASE PRESENTATION
A 45‐year‐old woman was admitted for laparoscopic hysterectomy under general anesthesia. She weighed 60 kg and was 162 cm tall. She had no history of psychiatric illness, medical disease, or drug abuse. Her physical examination did not reveal any significant findings. The preoperative electrocardiography and laboratory data, which included complete blood count, liver function, kidney function, and coagulation profile, were unremarkable.
Upon arrival in the operating room, routine monitoring, including electrocardiograph (ECG) monitoring, noninvasive blood pressure, and pulse oximetry, was applied. Lactated Ringer's solution (500 mL) was rapidly infused intravenously during the induction of anesthesia. Scopolamine (0.3 mg) and dexamethasone (5 mg) were administered before induction. Propofol (2 mg/kg) and fentanyl (4 µg/kg) were used for induction. When the patient lost consciousness, 0.6 mg/kg of rocuronium was given to facilitate tracheal intubation. Anesthesia was maintained by the continuous infusion of propofol (6‐7 mg/kg/h) and remifentanil (0.15‐0.2 µg/kg/min). The patient was placed in the Trendelenburg position, and intra‐abdominal pressure was maintained at 11‐14 cmH2O during surgery. After surgery, the patient was extubated when spontaneous breathing and consciousness recovered. The surgery lasted 80 minutes. Neuromuscular blockade was monitored by TOF‐Watch, and no additional rocuronium was given during surgery.
After extubation, the patient manifested symptoms of dyspnea, showing repeated effort to breathe. She was a little anxious but could obey loud verbal commands, such as opening the mouth and holding the anesthetist's hand. The TOF value was above 90% though our first concern was a possible weakness of muscle tone. We gave the patient neostigmine (0.04 mg/kg) and atropine (0.02 mg/kg). However, the patient became more anxious and irritable. We tried to pacify her with reassuring remarks that the surgery was successful and that she would be sent back to her husband soon. However, the patient frequently pointed to her throat, and we began to notice that she had difficulty speaking. She remained conscious and seemed to understand what we were saying to her. She tried to answer but could not utter a word. She could move her mouth, lips, tongue, head, and extremities upon command. Her pupils were equal in size and reactive to light. Her lungs were clear, and vital signs were stable. Arterial blood gas analysis showed no abnormal results. The patient asked for paper by gesture and wrote down the following words, “I want to speak, but I can't say a word” and “help me.” The patient was getting increasingly anxious, became uncooperative, and was in tears.
3 DIAGNOSIS AND TREATMENT
We administered 2.5 mg of droperidol, and she fell asleep. When she woke up 20 minutes later, she could speak in a hoarse voice and was transferred to the ward. She became more peaceful when she saw her husband. After 4 hours, the patient was able to talk normally. The aphasia lasted for 1 hour, and the whole episode lasted for 5.5 hours. She told the anesthetist that she could recall the events during recovery and felt happy to be able to talk again. She recalled that approximately 4 years ago, she experienced an episode of transient aphasia after a bitter quarrel with her husband.
4 OUTCOME AND FOLLOW‐UP
A neurology consultation suggested that the patient should undergo more extensive studies before any definite diagnosis could be made. Therefore, imaging (MRI) was arranged immediately. No definite organic lesions were found in an MRI of the brain. She was discharged without sequelae on postsurgery day 3. She was followed up for 3 months, and no complications were reported.
5 DISCUSSION
The differential diagnosis of such a clinical presentation of sudden and transient aphasia could be related to a residual effect of the anesthesia, transient ischemic attack (TIA), structural injury of vocal function,
7
or psychoemotional change.
8
,
9
The available literature on acute aphasia after anesthesia is limited.
Transient ischemic attack is a transient neurological event that usually persists for minutes and can occur due to hypotension, hypoxia, or embolism. However, the patient did not have hypoxia or hypotension. Thromboembolism was considered because the patient was placed in the lithotomy position. Lithotomy position and CO2 pneumoperitoneum may result in increased lower extremity venous pressure during hysterectomy, representing a risk factor for deep venous thrombosis.
10
And it was reported that steep Trendelenburg, along with several other risk factors, increased the odds of venous thromboembolism formation by 2.4 [95% confidence interval (95% CI) 1.9‐5.0].
11
When TIA is suspected, the “ABCD2” score could be used for the assessment of early stroke risk (“ABCD2” score ≥ 4).
12
The “ABCD2” score in this patient was 3 (speech disturbance without weakness [1 point] and duration of symptoms > 60 minutes [2 points]). And the patient was with a Glasgow score of 15 and a normal MRI results after surgery. The patient was followed up for 3 months because there is a 10%‐20% risk of major stroke after TIA in the subsequent 3 months.
13
All the evidences supported that the patient did not suffer from TIA.
There are several reports of acute neurological events, including cerebral edema, in patients undergoing laparoscopic gynecological surgery.
2
,
3
Any rise in arterial carbon dioxide can possibly result in increased cerebral blood flow. A steep Trendelenburg position can alter the cerebral blood volume and, thus, increase intracranial pressure. Moreover, head‐down tilt causes cephalad fluid shift and increases capillary pressure in the head, leading to an increase in extracranial vascular pressure that may cause facial, laryngeal, pharyngeal edema, and nasal congestion.
1
,
4
However, in our case, the carbon dioxide levels were measured by capnography and were kept at the normal limit. The surgery lasted for 80 minutes, and no significant facial edema existed. Because MRI was normal, brain edema was not considered.
Residual neuromuscular blockade could affect the speaking function. The patient could write down what she wanted to say, appropriately response to verbal command, indicating well‐reversed neuromuscular blockade. Structural injury, including the luxation of the arytenoid cartilage and laryngeal nerve injury, may occur after tracheal intubation and affect pronunciation.
7
,
14
,
15
The aphonia or aphasia in these cases is usually consistent and requires treatment. However, for our case the aphasia was transient, so structural damage was excluded. Pre‐existing neurologic disease should be screened for postoperative speaking dysfunction. Willson et al
16
reported a patient with hemiplegic migraine who developed atypical migraine with apneic spells, aphasia, and hemiparesis following general anesthesia. By history and MRI assessment, these possibilities were excluded in our case.
Postoperative delirium may present as speaking disorder. Medications including midazolam and scopolamine have been proved to increase delirium.
17
,
18
And in a case of a healthy 12‐year‐old girl who received preoperative midazolam, Drobish et al
19
reported postoperative transient associative agnosia and expressive aphasia. The administration of flumazenil led to the immediate and lasting resolution of her symptoms. Confusion assessment method (CAM) is the most often used tool for delirium evaluation. The diagnosis of delirium by CAM requires the presence of features 1 (acute onset and fluctuation course) and 2 (inattention) and either 3 (disorganized thinking) or 4 (altered level of consciousness). Given the presentation as it is, the patient does not probably meet these criteria. Though the EEG was not recorded during the episode, the Brief Psychiatric Rating Scale of the patient was 10, which did not support the diagnosis of psychiatric disorder. Moreover, Broca's aphasia patient's cannot speak or write and with Wernicke's they can speak but cannot convey a clear message. The presentation of the patient may support that the aphasia is manifestation of psychological in nature. Droperidol was effective for the sedation of the patient; however, ruling out dangerous episodes by CT scan or MRI maybe important before droperidol administration.
Psychogenic cognitive disturbance varies in presentation, including aphasia, aphagia, and quadriplegia.
8
,
20
,
21
In this case, the patient stayed awake and remembered the entire recovery process. The patient has a history of transient aphasia after a bitter quarrel with her husband 4 years ago. Stress and anxiety could have triggered her aphasia. She could be soothed by seeing her family, which highlights the importance of emotional support in treating perioperative psychogenic disorders. Enhanced pre‐operative education and early company of the family may improve the outcome.
6 CONCLUSION
In summary, transient aphasia can occur after general anesthesia. The possibility of cerebral complications and the injury of the laryngeal‐pharyngeal structures should be ruled out in appropriate clinical settings. A previous history of psychoemotional disorder should not be dismissed when reviewing a patient preoperatively.
CONFLICT OF INTEREST
None declared.
AUTHOR CONTRIBUTION
ZL and HD: were the patient's anesthetists, reviewed the literature, and contributed to manuscript revision; JZ and MY: reviewed the literature and contributed to manuscript drafting; all authors issued final approval for the version to be submitted.
ETHICAL APPROVAL
The patient provided signed informed consent to the publishment of the case. All the procedures performed in this study were in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.
ACKNOWLEDGMENTS
We thoroughly appreciated the patient and her family for collaboration with us. Consent statement: Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
Data were not available for share. | Intravenous (not otherwise specified) | DrugAdministrationRoute | CC BY-NC-ND | 33598216 | 19,156,275 | 2021-02 |
What was the administration route of drug 'PROPOFOL'? | Transient aphasia following general anesthesia in patient undergoing laparoscopic gynecologic surgery: A case report and literature review.
Aphasia could be a manifestation of postoperative psychoemotional disorder. Early differential diagnosis is important.
1 INTRODUCTION
This case report describes a healthy 45‐year‐old woman who received total intravenous anesthesia for laparoscopic hysterectomy, and then exhibited postoperative transient aphasia. Droperidol was given and the patient resolved. The author hypothesized that aphasia is a manifestation of psychoemotional disorder. Early differential diagnosis is important.
Laparoscopic gynecological surgery has been widely employed under general anesthesia. Although the safety and effectiveness of these techniques have greatly improved over the past decades, some problematic effects may still occur. Lithotomy position, Trendelenburg position, and CO2 pneumoperitoneum may increase the risk of neurologic complications.
1
,
2
,
3
,
4
There have been isolated case reports of psychogenic language disorder during surgery.
5
However, cognitive and neurologic complications after prolonged Trendelenburg position are extremely rare and more often reported in elderly patients, or those with underlying venous disease.
6
Furthermore, aphasia is rarely reported in perioperative settings. In this report, we present a patient who developed transient aphasia after extubation after laparoscopic hysterectomy. The patient provided written consent for this case report.
2 CASE PRESENTATION
A 45‐year‐old woman was admitted for laparoscopic hysterectomy under general anesthesia. She weighed 60 kg and was 162 cm tall. She had no history of psychiatric illness, medical disease, or drug abuse. Her physical examination did not reveal any significant findings. The preoperative electrocardiography and laboratory data, which included complete blood count, liver function, kidney function, and coagulation profile, were unremarkable.
Upon arrival in the operating room, routine monitoring, including electrocardiograph (ECG) monitoring, noninvasive blood pressure, and pulse oximetry, was applied. Lactated Ringer's solution (500 mL) was rapidly infused intravenously during the induction of anesthesia. Scopolamine (0.3 mg) and dexamethasone (5 mg) were administered before induction. Propofol (2 mg/kg) and fentanyl (4 µg/kg) were used for induction. When the patient lost consciousness, 0.6 mg/kg of rocuronium was given to facilitate tracheal intubation. Anesthesia was maintained by the continuous infusion of propofol (6‐7 mg/kg/h) and remifentanil (0.15‐0.2 µg/kg/min). The patient was placed in the Trendelenburg position, and intra‐abdominal pressure was maintained at 11‐14 cmH2O during surgery. After surgery, the patient was extubated when spontaneous breathing and consciousness recovered. The surgery lasted 80 minutes. Neuromuscular blockade was monitored by TOF‐Watch, and no additional rocuronium was given during surgery.
After extubation, the patient manifested symptoms of dyspnea, showing repeated effort to breathe. She was a little anxious but could obey loud verbal commands, such as opening the mouth and holding the anesthetist's hand. The TOF value was above 90% though our first concern was a possible weakness of muscle tone. We gave the patient neostigmine (0.04 mg/kg) and atropine (0.02 mg/kg). However, the patient became more anxious and irritable. We tried to pacify her with reassuring remarks that the surgery was successful and that she would be sent back to her husband soon. However, the patient frequently pointed to her throat, and we began to notice that she had difficulty speaking. She remained conscious and seemed to understand what we were saying to her. She tried to answer but could not utter a word. She could move her mouth, lips, tongue, head, and extremities upon command. Her pupils were equal in size and reactive to light. Her lungs were clear, and vital signs were stable. Arterial blood gas analysis showed no abnormal results. The patient asked for paper by gesture and wrote down the following words, “I want to speak, but I can't say a word” and “help me.” The patient was getting increasingly anxious, became uncooperative, and was in tears.
3 DIAGNOSIS AND TREATMENT
We administered 2.5 mg of droperidol, and she fell asleep. When she woke up 20 minutes later, she could speak in a hoarse voice and was transferred to the ward. She became more peaceful when she saw her husband. After 4 hours, the patient was able to talk normally. The aphasia lasted for 1 hour, and the whole episode lasted for 5.5 hours. She told the anesthetist that she could recall the events during recovery and felt happy to be able to talk again. She recalled that approximately 4 years ago, she experienced an episode of transient aphasia after a bitter quarrel with her husband.
4 OUTCOME AND FOLLOW‐UP
A neurology consultation suggested that the patient should undergo more extensive studies before any definite diagnosis could be made. Therefore, imaging (MRI) was arranged immediately. No definite organic lesions were found in an MRI of the brain. She was discharged without sequelae on postsurgery day 3. She was followed up for 3 months, and no complications were reported.
5 DISCUSSION
The differential diagnosis of such a clinical presentation of sudden and transient aphasia could be related to a residual effect of the anesthesia, transient ischemic attack (TIA), structural injury of vocal function,
7
or psychoemotional change.
8
,
9
The available literature on acute aphasia after anesthesia is limited.
Transient ischemic attack is a transient neurological event that usually persists for minutes and can occur due to hypotension, hypoxia, or embolism. However, the patient did not have hypoxia or hypotension. Thromboembolism was considered because the patient was placed in the lithotomy position. Lithotomy position and CO2 pneumoperitoneum may result in increased lower extremity venous pressure during hysterectomy, representing a risk factor for deep venous thrombosis.
10
And it was reported that steep Trendelenburg, along with several other risk factors, increased the odds of venous thromboembolism formation by 2.4 [95% confidence interval (95% CI) 1.9‐5.0].
11
When TIA is suspected, the “ABCD2” score could be used for the assessment of early stroke risk (“ABCD2” score ≥ 4).
12
The “ABCD2” score in this patient was 3 (speech disturbance without weakness [1 point] and duration of symptoms > 60 minutes [2 points]). And the patient was with a Glasgow score of 15 and a normal MRI results after surgery. The patient was followed up for 3 months because there is a 10%‐20% risk of major stroke after TIA in the subsequent 3 months.
13
All the evidences supported that the patient did not suffer from TIA.
There are several reports of acute neurological events, including cerebral edema, in patients undergoing laparoscopic gynecological surgery.
2
,
3
Any rise in arterial carbon dioxide can possibly result in increased cerebral blood flow. A steep Trendelenburg position can alter the cerebral blood volume and, thus, increase intracranial pressure. Moreover, head‐down tilt causes cephalad fluid shift and increases capillary pressure in the head, leading to an increase in extracranial vascular pressure that may cause facial, laryngeal, pharyngeal edema, and nasal congestion.
1
,
4
However, in our case, the carbon dioxide levels were measured by capnography and were kept at the normal limit. The surgery lasted for 80 minutes, and no significant facial edema existed. Because MRI was normal, brain edema was not considered.
Residual neuromuscular blockade could affect the speaking function. The patient could write down what she wanted to say, appropriately response to verbal command, indicating well‐reversed neuromuscular blockade. Structural injury, including the luxation of the arytenoid cartilage and laryngeal nerve injury, may occur after tracheal intubation and affect pronunciation.
7
,
14
,
15
The aphonia or aphasia in these cases is usually consistent and requires treatment. However, for our case the aphasia was transient, so structural damage was excluded. Pre‐existing neurologic disease should be screened for postoperative speaking dysfunction. Willson et al
16
reported a patient with hemiplegic migraine who developed atypical migraine with apneic spells, aphasia, and hemiparesis following general anesthesia. By history and MRI assessment, these possibilities were excluded in our case.
Postoperative delirium may present as speaking disorder. Medications including midazolam and scopolamine have been proved to increase delirium.
17
,
18
And in a case of a healthy 12‐year‐old girl who received preoperative midazolam, Drobish et al
19
reported postoperative transient associative agnosia and expressive aphasia. The administration of flumazenil led to the immediate and lasting resolution of her symptoms. Confusion assessment method (CAM) is the most often used tool for delirium evaluation. The diagnosis of delirium by CAM requires the presence of features 1 (acute onset and fluctuation course) and 2 (inattention) and either 3 (disorganized thinking) or 4 (altered level of consciousness). Given the presentation as it is, the patient does not probably meet these criteria. Though the EEG was not recorded during the episode, the Brief Psychiatric Rating Scale of the patient was 10, which did not support the diagnosis of psychiatric disorder. Moreover, Broca's aphasia patient's cannot speak or write and with Wernicke's they can speak but cannot convey a clear message. The presentation of the patient may support that the aphasia is manifestation of psychological in nature. Droperidol was effective for the sedation of the patient; however, ruling out dangerous episodes by CT scan or MRI maybe important before droperidol administration.
Psychogenic cognitive disturbance varies in presentation, including aphasia, aphagia, and quadriplegia.
8
,
20
,
21
In this case, the patient stayed awake and remembered the entire recovery process. The patient has a history of transient aphasia after a bitter quarrel with her husband 4 years ago. Stress and anxiety could have triggered her aphasia. She could be soothed by seeing her family, which highlights the importance of emotional support in treating perioperative psychogenic disorders. Enhanced pre‐operative education and early company of the family may improve the outcome.
6 CONCLUSION
In summary, transient aphasia can occur after general anesthesia. The possibility of cerebral complications and the injury of the laryngeal‐pharyngeal structures should be ruled out in appropriate clinical settings. A previous history of psychoemotional disorder should not be dismissed when reviewing a patient preoperatively.
CONFLICT OF INTEREST
None declared.
AUTHOR CONTRIBUTION
ZL and HD: were the patient's anesthetists, reviewed the literature, and contributed to manuscript revision; JZ and MY: reviewed the literature and contributed to manuscript drafting; all authors issued final approval for the version to be submitted.
ETHICAL APPROVAL
The patient provided signed informed consent to the publishment of the case. All the procedures performed in this study were in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.
ACKNOWLEDGMENTS
We thoroughly appreciated the patient and her family for collaboration with us. Consent statement: Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
Data were not available for share. | Intravenous (not otherwise specified) | DrugAdministrationRoute | CC BY-NC-ND | 33598216 | 19,156,275 | 2021-02 |
What was the dosage of drug 'PROPOFOL'? | Transient aphasia following general anesthesia in patient undergoing laparoscopic gynecologic surgery: A case report and literature review.
Aphasia could be a manifestation of postoperative psychoemotional disorder. Early differential diagnosis is important.
1 INTRODUCTION
This case report describes a healthy 45‐year‐old woman who received total intravenous anesthesia for laparoscopic hysterectomy, and then exhibited postoperative transient aphasia. Droperidol was given and the patient resolved. The author hypothesized that aphasia is a manifestation of psychoemotional disorder. Early differential diagnosis is important.
Laparoscopic gynecological surgery has been widely employed under general anesthesia. Although the safety and effectiveness of these techniques have greatly improved over the past decades, some problematic effects may still occur. Lithotomy position, Trendelenburg position, and CO2 pneumoperitoneum may increase the risk of neurologic complications.
1
,
2
,
3
,
4
There have been isolated case reports of psychogenic language disorder during surgery.
5
However, cognitive and neurologic complications after prolonged Trendelenburg position are extremely rare and more often reported in elderly patients, or those with underlying venous disease.
6
Furthermore, aphasia is rarely reported in perioperative settings. In this report, we present a patient who developed transient aphasia after extubation after laparoscopic hysterectomy. The patient provided written consent for this case report.
2 CASE PRESENTATION
A 45‐year‐old woman was admitted for laparoscopic hysterectomy under general anesthesia. She weighed 60 kg and was 162 cm tall. She had no history of psychiatric illness, medical disease, or drug abuse. Her physical examination did not reveal any significant findings. The preoperative electrocardiography and laboratory data, which included complete blood count, liver function, kidney function, and coagulation profile, were unremarkable.
Upon arrival in the operating room, routine monitoring, including electrocardiograph (ECG) monitoring, noninvasive blood pressure, and pulse oximetry, was applied. Lactated Ringer's solution (500 mL) was rapidly infused intravenously during the induction of anesthesia. Scopolamine (0.3 mg) and dexamethasone (5 mg) were administered before induction. Propofol (2 mg/kg) and fentanyl (4 µg/kg) were used for induction. When the patient lost consciousness, 0.6 mg/kg of rocuronium was given to facilitate tracheal intubation. Anesthesia was maintained by the continuous infusion of propofol (6‐7 mg/kg/h) and remifentanil (0.15‐0.2 µg/kg/min). The patient was placed in the Trendelenburg position, and intra‐abdominal pressure was maintained at 11‐14 cmH2O during surgery. After surgery, the patient was extubated when spontaneous breathing and consciousness recovered. The surgery lasted 80 minutes. Neuromuscular blockade was monitored by TOF‐Watch, and no additional rocuronium was given during surgery.
After extubation, the patient manifested symptoms of dyspnea, showing repeated effort to breathe. She was a little anxious but could obey loud verbal commands, such as opening the mouth and holding the anesthetist's hand. The TOF value was above 90% though our first concern was a possible weakness of muscle tone. We gave the patient neostigmine (0.04 mg/kg) and atropine (0.02 mg/kg). However, the patient became more anxious and irritable. We tried to pacify her with reassuring remarks that the surgery was successful and that she would be sent back to her husband soon. However, the patient frequently pointed to her throat, and we began to notice that she had difficulty speaking. She remained conscious and seemed to understand what we were saying to her. She tried to answer but could not utter a word. She could move her mouth, lips, tongue, head, and extremities upon command. Her pupils were equal in size and reactive to light. Her lungs were clear, and vital signs were stable. Arterial blood gas analysis showed no abnormal results. The patient asked for paper by gesture and wrote down the following words, “I want to speak, but I can't say a word” and “help me.” The patient was getting increasingly anxious, became uncooperative, and was in tears.
3 DIAGNOSIS AND TREATMENT
We administered 2.5 mg of droperidol, and she fell asleep. When she woke up 20 minutes later, she could speak in a hoarse voice and was transferred to the ward. She became more peaceful when she saw her husband. After 4 hours, the patient was able to talk normally. The aphasia lasted for 1 hour, and the whole episode lasted for 5.5 hours. She told the anesthetist that she could recall the events during recovery and felt happy to be able to talk again. She recalled that approximately 4 years ago, she experienced an episode of transient aphasia after a bitter quarrel with her husband.
4 OUTCOME AND FOLLOW‐UP
A neurology consultation suggested that the patient should undergo more extensive studies before any definite diagnosis could be made. Therefore, imaging (MRI) was arranged immediately. No definite organic lesions were found in an MRI of the brain. She was discharged without sequelae on postsurgery day 3. She was followed up for 3 months, and no complications were reported.
5 DISCUSSION
The differential diagnosis of such a clinical presentation of sudden and transient aphasia could be related to a residual effect of the anesthesia, transient ischemic attack (TIA), structural injury of vocal function,
7
or psychoemotional change.
8
,
9
The available literature on acute aphasia after anesthesia is limited.
Transient ischemic attack is a transient neurological event that usually persists for minutes and can occur due to hypotension, hypoxia, or embolism. However, the patient did not have hypoxia or hypotension. Thromboembolism was considered because the patient was placed in the lithotomy position. Lithotomy position and CO2 pneumoperitoneum may result in increased lower extremity venous pressure during hysterectomy, representing a risk factor for deep venous thrombosis.
10
And it was reported that steep Trendelenburg, along with several other risk factors, increased the odds of venous thromboembolism formation by 2.4 [95% confidence interval (95% CI) 1.9‐5.0].
11
When TIA is suspected, the “ABCD2” score could be used for the assessment of early stroke risk (“ABCD2” score ≥ 4).
12
The “ABCD2” score in this patient was 3 (speech disturbance without weakness [1 point] and duration of symptoms > 60 minutes [2 points]). And the patient was with a Glasgow score of 15 and a normal MRI results after surgery. The patient was followed up for 3 months because there is a 10%‐20% risk of major stroke after TIA in the subsequent 3 months.
13
All the evidences supported that the patient did not suffer from TIA.
There are several reports of acute neurological events, including cerebral edema, in patients undergoing laparoscopic gynecological surgery.
2
,
3
Any rise in arterial carbon dioxide can possibly result in increased cerebral blood flow. A steep Trendelenburg position can alter the cerebral blood volume and, thus, increase intracranial pressure. Moreover, head‐down tilt causes cephalad fluid shift and increases capillary pressure in the head, leading to an increase in extracranial vascular pressure that may cause facial, laryngeal, pharyngeal edema, and nasal congestion.
1
,
4
However, in our case, the carbon dioxide levels were measured by capnography and were kept at the normal limit. The surgery lasted for 80 minutes, and no significant facial edema existed. Because MRI was normal, brain edema was not considered.
Residual neuromuscular blockade could affect the speaking function. The patient could write down what she wanted to say, appropriately response to verbal command, indicating well‐reversed neuromuscular blockade. Structural injury, including the luxation of the arytenoid cartilage and laryngeal nerve injury, may occur after tracheal intubation and affect pronunciation.
7
,
14
,
15
The aphonia or aphasia in these cases is usually consistent and requires treatment. However, for our case the aphasia was transient, so structural damage was excluded. Pre‐existing neurologic disease should be screened for postoperative speaking dysfunction. Willson et al
16
reported a patient with hemiplegic migraine who developed atypical migraine with apneic spells, aphasia, and hemiparesis following general anesthesia. By history and MRI assessment, these possibilities were excluded in our case.
Postoperative delirium may present as speaking disorder. Medications including midazolam and scopolamine have been proved to increase delirium.
17
,
18
And in a case of a healthy 12‐year‐old girl who received preoperative midazolam, Drobish et al
19
reported postoperative transient associative agnosia and expressive aphasia. The administration of flumazenil led to the immediate and lasting resolution of her symptoms. Confusion assessment method (CAM) is the most often used tool for delirium evaluation. The diagnosis of delirium by CAM requires the presence of features 1 (acute onset and fluctuation course) and 2 (inattention) and either 3 (disorganized thinking) or 4 (altered level of consciousness). Given the presentation as it is, the patient does not probably meet these criteria. Though the EEG was not recorded during the episode, the Brief Psychiatric Rating Scale of the patient was 10, which did not support the diagnosis of psychiatric disorder. Moreover, Broca's aphasia patient's cannot speak or write and with Wernicke's they can speak but cannot convey a clear message. The presentation of the patient may support that the aphasia is manifestation of psychological in nature. Droperidol was effective for the sedation of the patient; however, ruling out dangerous episodes by CT scan or MRI maybe important before droperidol administration.
Psychogenic cognitive disturbance varies in presentation, including aphasia, aphagia, and quadriplegia.
8
,
20
,
21
In this case, the patient stayed awake and remembered the entire recovery process. The patient has a history of transient aphasia after a bitter quarrel with her husband 4 years ago. Stress and anxiety could have triggered her aphasia. She could be soothed by seeing her family, which highlights the importance of emotional support in treating perioperative psychogenic disorders. Enhanced pre‐operative education and early company of the family may improve the outcome.
6 CONCLUSION
In summary, transient aphasia can occur after general anesthesia. The possibility of cerebral complications and the injury of the laryngeal‐pharyngeal structures should be ruled out in appropriate clinical settings. A previous history of psychoemotional disorder should not be dismissed when reviewing a patient preoperatively.
CONFLICT OF INTEREST
None declared.
AUTHOR CONTRIBUTION
ZL and HD: were the patient's anesthetists, reviewed the literature, and contributed to manuscript revision; JZ and MY: reviewed the literature and contributed to manuscript drafting; all authors issued final approval for the version to be submitted.
ETHICAL APPROVAL
The patient provided signed informed consent to the publishment of the case. All the procedures performed in this study were in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.
ACKNOWLEDGMENTS
We thoroughly appreciated the patient and her family for collaboration with us. Consent statement: Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
Data were not available for share. | 6?7 MG/KG/H | DrugDosageText | CC BY-NC-ND | 33598216 | 19,156,275 | 2021-02 |
What was the dosage of drug 'REMIFENTANIL'? | Transient aphasia following general anesthesia in patient undergoing laparoscopic gynecologic surgery: A case report and literature review.
Aphasia could be a manifestation of postoperative psychoemotional disorder. Early differential diagnosis is important.
1 INTRODUCTION
This case report describes a healthy 45‐year‐old woman who received total intravenous anesthesia for laparoscopic hysterectomy, and then exhibited postoperative transient aphasia. Droperidol was given and the patient resolved. The author hypothesized that aphasia is a manifestation of psychoemotional disorder. Early differential diagnosis is important.
Laparoscopic gynecological surgery has been widely employed under general anesthesia. Although the safety and effectiveness of these techniques have greatly improved over the past decades, some problematic effects may still occur. Lithotomy position, Trendelenburg position, and CO2 pneumoperitoneum may increase the risk of neurologic complications.
1
,
2
,
3
,
4
There have been isolated case reports of psychogenic language disorder during surgery.
5
However, cognitive and neurologic complications after prolonged Trendelenburg position are extremely rare and more often reported in elderly patients, or those with underlying venous disease.
6
Furthermore, aphasia is rarely reported in perioperative settings. In this report, we present a patient who developed transient aphasia after extubation after laparoscopic hysterectomy. The patient provided written consent for this case report.
2 CASE PRESENTATION
A 45‐year‐old woman was admitted for laparoscopic hysterectomy under general anesthesia. She weighed 60 kg and was 162 cm tall. She had no history of psychiatric illness, medical disease, or drug abuse. Her physical examination did not reveal any significant findings. The preoperative electrocardiography and laboratory data, which included complete blood count, liver function, kidney function, and coagulation profile, were unremarkable.
Upon arrival in the operating room, routine monitoring, including electrocardiograph (ECG) monitoring, noninvasive blood pressure, and pulse oximetry, was applied. Lactated Ringer's solution (500 mL) was rapidly infused intravenously during the induction of anesthesia. Scopolamine (0.3 mg) and dexamethasone (5 mg) were administered before induction. Propofol (2 mg/kg) and fentanyl (4 µg/kg) were used for induction. When the patient lost consciousness, 0.6 mg/kg of rocuronium was given to facilitate tracheal intubation. Anesthesia was maintained by the continuous infusion of propofol (6‐7 mg/kg/h) and remifentanil (0.15‐0.2 µg/kg/min). The patient was placed in the Trendelenburg position, and intra‐abdominal pressure was maintained at 11‐14 cmH2O during surgery. After surgery, the patient was extubated when spontaneous breathing and consciousness recovered. The surgery lasted 80 minutes. Neuromuscular blockade was monitored by TOF‐Watch, and no additional rocuronium was given during surgery.
After extubation, the patient manifested symptoms of dyspnea, showing repeated effort to breathe. She was a little anxious but could obey loud verbal commands, such as opening the mouth and holding the anesthetist's hand. The TOF value was above 90% though our first concern was a possible weakness of muscle tone. We gave the patient neostigmine (0.04 mg/kg) and atropine (0.02 mg/kg). However, the patient became more anxious and irritable. We tried to pacify her with reassuring remarks that the surgery was successful and that she would be sent back to her husband soon. However, the patient frequently pointed to her throat, and we began to notice that she had difficulty speaking. She remained conscious and seemed to understand what we were saying to her. She tried to answer but could not utter a word. She could move her mouth, lips, tongue, head, and extremities upon command. Her pupils were equal in size and reactive to light. Her lungs were clear, and vital signs were stable. Arterial blood gas analysis showed no abnormal results. The patient asked for paper by gesture and wrote down the following words, “I want to speak, but I can't say a word” and “help me.” The patient was getting increasingly anxious, became uncooperative, and was in tears.
3 DIAGNOSIS AND TREATMENT
We administered 2.5 mg of droperidol, and she fell asleep. When she woke up 20 minutes later, she could speak in a hoarse voice and was transferred to the ward. She became more peaceful when she saw her husband. After 4 hours, the patient was able to talk normally. The aphasia lasted for 1 hour, and the whole episode lasted for 5.5 hours. She told the anesthetist that she could recall the events during recovery and felt happy to be able to talk again. She recalled that approximately 4 years ago, she experienced an episode of transient aphasia after a bitter quarrel with her husband.
4 OUTCOME AND FOLLOW‐UP
A neurology consultation suggested that the patient should undergo more extensive studies before any definite diagnosis could be made. Therefore, imaging (MRI) was arranged immediately. No definite organic lesions were found in an MRI of the brain. She was discharged without sequelae on postsurgery day 3. She was followed up for 3 months, and no complications were reported.
5 DISCUSSION
The differential diagnosis of such a clinical presentation of sudden and transient aphasia could be related to a residual effect of the anesthesia, transient ischemic attack (TIA), structural injury of vocal function,
7
or psychoemotional change.
8
,
9
The available literature on acute aphasia after anesthesia is limited.
Transient ischemic attack is a transient neurological event that usually persists for minutes and can occur due to hypotension, hypoxia, or embolism. However, the patient did not have hypoxia or hypotension. Thromboembolism was considered because the patient was placed in the lithotomy position. Lithotomy position and CO2 pneumoperitoneum may result in increased lower extremity venous pressure during hysterectomy, representing a risk factor for deep venous thrombosis.
10
And it was reported that steep Trendelenburg, along with several other risk factors, increased the odds of venous thromboembolism formation by 2.4 [95% confidence interval (95% CI) 1.9‐5.0].
11
When TIA is suspected, the “ABCD2” score could be used for the assessment of early stroke risk (“ABCD2” score ≥ 4).
12
The “ABCD2” score in this patient was 3 (speech disturbance without weakness [1 point] and duration of symptoms > 60 minutes [2 points]). And the patient was with a Glasgow score of 15 and a normal MRI results after surgery. The patient was followed up for 3 months because there is a 10%‐20% risk of major stroke after TIA in the subsequent 3 months.
13
All the evidences supported that the patient did not suffer from TIA.
There are several reports of acute neurological events, including cerebral edema, in patients undergoing laparoscopic gynecological surgery.
2
,
3
Any rise in arterial carbon dioxide can possibly result in increased cerebral blood flow. A steep Trendelenburg position can alter the cerebral blood volume and, thus, increase intracranial pressure. Moreover, head‐down tilt causes cephalad fluid shift and increases capillary pressure in the head, leading to an increase in extracranial vascular pressure that may cause facial, laryngeal, pharyngeal edema, and nasal congestion.
1
,
4
However, in our case, the carbon dioxide levels were measured by capnography and were kept at the normal limit. The surgery lasted for 80 minutes, and no significant facial edema existed. Because MRI was normal, brain edema was not considered.
Residual neuromuscular blockade could affect the speaking function. The patient could write down what she wanted to say, appropriately response to verbal command, indicating well‐reversed neuromuscular blockade. Structural injury, including the luxation of the arytenoid cartilage and laryngeal nerve injury, may occur after tracheal intubation and affect pronunciation.
7
,
14
,
15
The aphonia or aphasia in these cases is usually consistent and requires treatment. However, for our case the aphasia was transient, so structural damage was excluded. Pre‐existing neurologic disease should be screened for postoperative speaking dysfunction. Willson et al
16
reported a patient with hemiplegic migraine who developed atypical migraine with apneic spells, aphasia, and hemiparesis following general anesthesia. By history and MRI assessment, these possibilities were excluded in our case.
Postoperative delirium may present as speaking disorder. Medications including midazolam and scopolamine have been proved to increase delirium.
17
,
18
And in a case of a healthy 12‐year‐old girl who received preoperative midazolam, Drobish et al
19
reported postoperative transient associative agnosia and expressive aphasia. The administration of flumazenil led to the immediate and lasting resolution of her symptoms. Confusion assessment method (CAM) is the most often used tool for delirium evaluation. The diagnosis of delirium by CAM requires the presence of features 1 (acute onset and fluctuation course) and 2 (inattention) and either 3 (disorganized thinking) or 4 (altered level of consciousness). Given the presentation as it is, the patient does not probably meet these criteria. Though the EEG was not recorded during the episode, the Brief Psychiatric Rating Scale of the patient was 10, which did not support the diagnosis of psychiatric disorder. Moreover, Broca's aphasia patient's cannot speak or write and with Wernicke's they can speak but cannot convey a clear message. The presentation of the patient may support that the aphasia is manifestation of psychological in nature. Droperidol was effective for the sedation of the patient; however, ruling out dangerous episodes by CT scan or MRI maybe important before droperidol administration.
Psychogenic cognitive disturbance varies in presentation, including aphasia, aphagia, and quadriplegia.
8
,
20
,
21
In this case, the patient stayed awake and remembered the entire recovery process. The patient has a history of transient aphasia after a bitter quarrel with her husband 4 years ago. Stress and anxiety could have triggered her aphasia. She could be soothed by seeing her family, which highlights the importance of emotional support in treating perioperative psychogenic disorders. Enhanced pre‐operative education and early company of the family may improve the outcome.
6 CONCLUSION
In summary, transient aphasia can occur after general anesthesia. The possibility of cerebral complications and the injury of the laryngeal‐pharyngeal structures should be ruled out in appropriate clinical settings. A previous history of psychoemotional disorder should not be dismissed when reviewing a patient preoperatively.
CONFLICT OF INTEREST
None declared.
AUTHOR CONTRIBUTION
ZL and HD: were the patient's anesthetists, reviewed the literature, and contributed to manuscript revision; JZ and MY: reviewed the literature and contributed to manuscript drafting; all authors issued final approval for the version to be submitted.
ETHICAL APPROVAL
The patient provided signed informed consent to the publishment of the case. All the procedures performed in this study were in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.
ACKNOWLEDGMENTS
We thoroughly appreciated the patient and her family for collaboration with us. Consent statement: Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
Data were not available for share. | 0.15?0.2 UG/KG/MIN | DrugDosageText | CC BY-NC-ND | 33598216 | 19,156,275 | 2021-02 |
What was the outcome of reaction 'Transient aphasia'? | Transient aphasia following general anesthesia in patient undergoing laparoscopic gynecologic surgery: A case report and literature review.
Aphasia could be a manifestation of postoperative psychoemotional disorder. Early differential diagnosis is important.
1 INTRODUCTION
This case report describes a healthy 45‐year‐old woman who received total intravenous anesthesia for laparoscopic hysterectomy, and then exhibited postoperative transient aphasia. Droperidol was given and the patient resolved. The author hypothesized that aphasia is a manifestation of psychoemotional disorder. Early differential diagnosis is important.
Laparoscopic gynecological surgery has been widely employed under general anesthesia. Although the safety and effectiveness of these techniques have greatly improved over the past decades, some problematic effects may still occur. Lithotomy position, Trendelenburg position, and CO2 pneumoperitoneum may increase the risk of neurologic complications.
1
,
2
,
3
,
4
There have been isolated case reports of psychogenic language disorder during surgery.
5
However, cognitive and neurologic complications after prolonged Trendelenburg position are extremely rare and more often reported in elderly patients, or those with underlying venous disease.
6
Furthermore, aphasia is rarely reported in perioperative settings. In this report, we present a patient who developed transient aphasia after extubation after laparoscopic hysterectomy. The patient provided written consent for this case report.
2 CASE PRESENTATION
A 45‐year‐old woman was admitted for laparoscopic hysterectomy under general anesthesia. She weighed 60 kg and was 162 cm tall. She had no history of psychiatric illness, medical disease, or drug abuse. Her physical examination did not reveal any significant findings. The preoperative electrocardiography and laboratory data, which included complete blood count, liver function, kidney function, and coagulation profile, were unremarkable.
Upon arrival in the operating room, routine monitoring, including electrocardiograph (ECG) monitoring, noninvasive blood pressure, and pulse oximetry, was applied. Lactated Ringer's solution (500 mL) was rapidly infused intravenously during the induction of anesthesia. Scopolamine (0.3 mg) and dexamethasone (5 mg) were administered before induction. Propofol (2 mg/kg) and fentanyl (4 µg/kg) were used for induction. When the patient lost consciousness, 0.6 mg/kg of rocuronium was given to facilitate tracheal intubation. Anesthesia was maintained by the continuous infusion of propofol (6‐7 mg/kg/h) and remifentanil (0.15‐0.2 µg/kg/min). The patient was placed in the Trendelenburg position, and intra‐abdominal pressure was maintained at 11‐14 cmH2O during surgery. After surgery, the patient was extubated when spontaneous breathing and consciousness recovered. The surgery lasted 80 minutes. Neuromuscular blockade was monitored by TOF‐Watch, and no additional rocuronium was given during surgery.
After extubation, the patient manifested symptoms of dyspnea, showing repeated effort to breathe. She was a little anxious but could obey loud verbal commands, such as opening the mouth and holding the anesthetist's hand. The TOF value was above 90% though our first concern was a possible weakness of muscle tone. We gave the patient neostigmine (0.04 mg/kg) and atropine (0.02 mg/kg). However, the patient became more anxious and irritable. We tried to pacify her with reassuring remarks that the surgery was successful and that she would be sent back to her husband soon. However, the patient frequently pointed to her throat, and we began to notice that she had difficulty speaking. She remained conscious and seemed to understand what we were saying to her. She tried to answer but could not utter a word. She could move her mouth, lips, tongue, head, and extremities upon command. Her pupils were equal in size and reactive to light. Her lungs were clear, and vital signs were stable. Arterial blood gas analysis showed no abnormal results. The patient asked for paper by gesture and wrote down the following words, “I want to speak, but I can't say a word” and “help me.” The patient was getting increasingly anxious, became uncooperative, and was in tears.
3 DIAGNOSIS AND TREATMENT
We administered 2.5 mg of droperidol, and she fell asleep. When she woke up 20 minutes later, she could speak in a hoarse voice and was transferred to the ward. She became more peaceful when she saw her husband. After 4 hours, the patient was able to talk normally. The aphasia lasted for 1 hour, and the whole episode lasted for 5.5 hours. She told the anesthetist that she could recall the events during recovery and felt happy to be able to talk again. She recalled that approximately 4 years ago, she experienced an episode of transient aphasia after a bitter quarrel with her husband.
4 OUTCOME AND FOLLOW‐UP
A neurology consultation suggested that the patient should undergo more extensive studies before any definite diagnosis could be made. Therefore, imaging (MRI) was arranged immediately. No definite organic lesions were found in an MRI of the brain. She was discharged without sequelae on postsurgery day 3. She was followed up for 3 months, and no complications were reported.
5 DISCUSSION
The differential diagnosis of such a clinical presentation of sudden and transient aphasia could be related to a residual effect of the anesthesia, transient ischemic attack (TIA), structural injury of vocal function,
7
or psychoemotional change.
8
,
9
The available literature on acute aphasia after anesthesia is limited.
Transient ischemic attack is a transient neurological event that usually persists for minutes and can occur due to hypotension, hypoxia, or embolism. However, the patient did not have hypoxia or hypotension. Thromboembolism was considered because the patient was placed in the lithotomy position. Lithotomy position and CO2 pneumoperitoneum may result in increased lower extremity venous pressure during hysterectomy, representing a risk factor for deep venous thrombosis.
10
And it was reported that steep Trendelenburg, along with several other risk factors, increased the odds of venous thromboembolism formation by 2.4 [95% confidence interval (95% CI) 1.9‐5.0].
11
When TIA is suspected, the “ABCD2” score could be used for the assessment of early stroke risk (“ABCD2” score ≥ 4).
12
The “ABCD2” score in this patient was 3 (speech disturbance without weakness [1 point] and duration of symptoms > 60 minutes [2 points]). And the patient was with a Glasgow score of 15 and a normal MRI results after surgery. The patient was followed up for 3 months because there is a 10%‐20% risk of major stroke after TIA in the subsequent 3 months.
13
All the evidences supported that the patient did not suffer from TIA.
There are several reports of acute neurological events, including cerebral edema, in patients undergoing laparoscopic gynecological surgery.
2
,
3
Any rise in arterial carbon dioxide can possibly result in increased cerebral blood flow. A steep Trendelenburg position can alter the cerebral blood volume and, thus, increase intracranial pressure. Moreover, head‐down tilt causes cephalad fluid shift and increases capillary pressure in the head, leading to an increase in extracranial vascular pressure that may cause facial, laryngeal, pharyngeal edema, and nasal congestion.
1
,
4
However, in our case, the carbon dioxide levels were measured by capnography and were kept at the normal limit. The surgery lasted for 80 minutes, and no significant facial edema existed. Because MRI was normal, brain edema was not considered.
Residual neuromuscular blockade could affect the speaking function. The patient could write down what she wanted to say, appropriately response to verbal command, indicating well‐reversed neuromuscular blockade. Structural injury, including the luxation of the arytenoid cartilage and laryngeal nerve injury, may occur after tracheal intubation and affect pronunciation.
7
,
14
,
15
The aphonia or aphasia in these cases is usually consistent and requires treatment. However, for our case the aphasia was transient, so structural damage was excluded. Pre‐existing neurologic disease should be screened for postoperative speaking dysfunction. Willson et al
16
reported a patient with hemiplegic migraine who developed atypical migraine with apneic spells, aphasia, and hemiparesis following general anesthesia. By history and MRI assessment, these possibilities were excluded in our case.
Postoperative delirium may present as speaking disorder. Medications including midazolam and scopolamine have been proved to increase delirium.
17
,
18
And in a case of a healthy 12‐year‐old girl who received preoperative midazolam, Drobish et al
19
reported postoperative transient associative agnosia and expressive aphasia. The administration of flumazenil led to the immediate and lasting resolution of her symptoms. Confusion assessment method (CAM) is the most often used tool for delirium evaluation. The diagnosis of delirium by CAM requires the presence of features 1 (acute onset and fluctuation course) and 2 (inattention) and either 3 (disorganized thinking) or 4 (altered level of consciousness). Given the presentation as it is, the patient does not probably meet these criteria. Though the EEG was not recorded during the episode, the Brief Psychiatric Rating Scale of the patient was 10, which did not support the diagnosis of psychiatric disorder. Moreover, Broca's aphasia patient's cannot speak or write and with Wernicke's they can speak but cannot convey a clear message. The presentation of the patient may support that the aphasia is manifestation of psychological in nature. Droperidol was effective for the sedation of the patient; however, ruling out dangerous episodes by CT scan or MRI maybe important before droperidol administration.
Psychogenic cognitive disturbance varies in presentation, including aphasia, aphagia, and quadriplegia.
8
,
20
,
21
In this case, the patient stayed awake and remembered the entire recovery process. The patient has a history of transient aphasia after a bitter quarrel with her husband 4 years ago. Stress and anxiety could have triggered her aphasia. She could be soothed by seeing her family, which highlights the importance of emotional support in treating perioperative psychogenic disorders. Enhanced pre‐operative education and early company of the family may improve the outcome.
6 CONCLUSION
In summary, transient aphasia can occur after general anesthesia. The possibility of cerebral complications and the injury of the laryngeal‐pharyngeal structures should be ruled out in appropriate clinical settings. A previous history of psychoemotional disorder should not be dismissed when reviewing a patient preoperatively.
CONFLICT OF INTEREST
None declared.
AUTHOR CONTRIBUTION
ZL and HD: were the patient's anesthetists, reviewed the literature, and contributed to manuscript revision; JZ and MY: reviewed the literature and contributed to manuscript drafting; all authors issued final approval for the version to be submitted.
ETHICAL APPROVAL
The patient provided signed informed consent to the publishment of the case. All the procedures performed in this study were in accordance with the ethical standards of the institutional and national research committee and with the 1964 Helsinki Declaration and its later amendments or comparable ethical standards.
ACKNOWLEDGMENTS
We thoroughly appreciated the patient and her family for collaboration with us. Consent statement: Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
Data were not available for share. | Recovered | ReactionOutcome | CC BY-NC-ND | 33598216 | 19,156,275 | 2021-02 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Pyrexia'. | Management of generalized severe periodontitis using full-mouth disinfection and systemic antibiotics in a leukemic patient before stem cell transplantation: A case report.
The full-mouth disinfection protocol implemented in this case can be integrated into established protocols for treating severe periodontitis in the context of a hematological malignancy, without any interference with the cancer treatment.
1 INTRODUCTION
Oral infection is a contributor to morbidity and mortality in leukemic patients. Full‐mouth disinfection, which terminates subgingival debridement within a short period, is applicable and effective in combination with antimicrobial agents in treating leukemic patients with severe periodontitis as an oral infection removal procedure prior to stem cell transplantation.
Hematopoietic stem cell transplantation (SCT) is a curative therapy for hematologic malignancies. Standard pretransplantation therapeutic approaches, including high‐dose chemotherapy, total body irradiation, and SCT, greatly improve the prognosis of leukemic patients. At multiple phases during SCT, patients are at risk of contracting infectious diseases due to profound and prolonged neutropenia.
1
The US National Cancer Institute indicated that odontogenic infection is a potential source of systemic infections that should be eliminated by dental treatment.
2
,
3
However, no definitive criteria for extraction or preservation of infected teeth that do not affect hematological treatment have been established for recipients with severe infections, including patients with hematological malignancies.
Here, we present a case of acute lymphoblastic leukemia with severe periodontitis prior to SCT that was treated with a regimen involving full‐mouth disinfection (FMD).
2 CASE REPORT
A 38‐year‐old man diagnosed with acute lymphoblastic leukemia and scheduled to undergo chemotherapy, total body irradiation, and SCT was referred to our department for odontogenic infection screening. After evaluation of his clinical oral condition, dental X‐ray photographs, and the presence of periodontitis‐related bacteria by quantitative polymerase chain reaction (qPCR), a diagnosis of generalized severe periodontitis (generalized periodontitis stage IV, grade C) was made. The patient had already received intravenous antimicrobial infusion for febrile neutropenia (FN) on admission to the hospital; however, periodontitis‐related bacteria were detected (Table 1). When the leukocyte count was low, we only followed the oral hygiene instructions. To remove dental plaque and supragingival calculus, scaling procedures on the gingival margin were performed at the time of recovery from FN after induction of chemotherapy. To prevent bacteremia and septicemia after SCT, FMD was considered to remove the infectious source via completion of scaling and root planing (SRP) within the limited dental treatment window of only 1 week, covering all periodontally infected teeth (Figure 1). Before the initiation of chemotherapy, oral hygiene instructions were provided to the patient to avoid risk of infection from dental foci. In the neutropenic phase, during chemotherapy and immunosuppression, oral care was provided to reduce oral complications including oral mucositis and acute periodontitis. Since his first visit to our department, the patient was expected to develop chemotherapy‐induced FN and oral mucosal disorders. We thus provided the patient with thorough oral hygiene instructions aiming to quantitatively reduce the source of infection in the mouth. Specifically, the patient was instructed on the toothbrush Bass technique regarding the toothbrush and interdental brush cleaning methods. Oral assessments were performed based on an oral assessment guide,
4
and the information was shared with the physician and nurses. A nonirritating moisturizer was used for xerostomia.
TABLE 1 Quantitative evaluation of periodontitis‐related bacteria in the deepest periodontal pocket of the patient
Baseline (count) Seven months after FMD (count)
Total bacteria 30 000 <1000
Porphyromonas gingivalis
570 <10
Tannerella forsythia
1300 <10
Treponema denticola
1100 <10
Fusobacterium nucleatum
2000 <10
John Wiley & Sons, LtdFIGURE 1 Treatment timeline. The patient was a 38‐year‐old male hematopoietic stem cell transplant (SCT) recipient with generalized severe periodontitis
The patient achieved complete remission following induction of chemotherapy consisting of cyclophosphamide vincristine, prednisolone, daunorubicin, and L‐asparaginase and consolidation chemotherapy consisting of doxorubicin, vincristine, and prednisolone. The myelosuppression grades of these chemotherapies were categorized as severe and moderate, respectively.
5
Procalcitonin (PCT) levels were measured when fever and febrile neutropenia (FN) were present. PCT was predominantly higher (7.4 ng/mL) in the presence of FN and septic shock during the first round of chemotherapy on day −116. Otherwise, the PCT was not predominant during the pretransplant oral care and at 1 month after the transplant, which is the focus of this study (Figure 1). Blood cultures were performed at the time of fever, and the only positive results between the time of admission and 1 month after transplantation were obtained from day −116 to day −111. At that time, FN and septic shock were both present, and the patient was positive for Corynebacterium striatum and Staphylococcus haemolyticus. No oral bacteria were found. Mechanical debridement for severe periodontal disease was then scheduled during the non‐neutropenic period after chemotherapy induction and consolidation. The day on which SCT was performed was designated day 0. To gain sufficient healing time of SRP for debridement of the periodontal deep pocket before SCT and to avoid reinfection from an untreated site to the treatment site in this case, we planned for FMD combined with systemic antimicrobial therapy, which could effectively be used to treat patients who had been diagnosed with generalized severe periodontitis in a short period of time.
6
After the first round of chemotherapy, we considered performing the procedure if the white blood cell count was normal. However, after consulting with the hematologist, we decided to schedule the consolidation chemotherapy; thus, the medical team considered the time when the consolidation chemotherapy was finished, the white blood cell and absolute neutrophil counts had recovered, there was no febrile neutropenia, and there was enough time for the wound to heal after FMD. As a result of this discussion, we treated for severe periodontitis in a short period from day −30 to day −22. FMD was performed from days −30 to −22 for debridement of all periodontal pockets by SRP with Gracey curettes. We performed SRP for tooth numbers 36, 37, 38, 46, 47, and 48 on day −30 (Federation Dentaire Internationale System); tooth numbers 14‐18 and 24‐28 on day −27; tooth numbers 31, 32, 33, 35 and from 41‐44 on day −24; and finally, tooth numbers 13, 21, 22, and 23, on day −22. During FMD, the patient was treated with sitafloxacin (100 mg/day for 14 days) to prevent bacteremia/septicemia, periodontopathic bacterial infection, and fever, which are possible adverse complications induced by FMD.
7
Both body temperature and C‐reactive protein (CRP) level were stable during FMD period, with a maximum body temperature of 37.0°C and maximum CRP concentration of 1.13 mg/dL (Figure 1). Allogeneic peripheral blood SCT was performed with myeloablative conditioning on day 0. The graft contained CD34+ cells at 2.66 × 106/kg. Myeloablative conditioning was based on total body irradiation at 12 Gy in six fractions on day −8 to −6, administered in combination with cytarabine 2000 mg/m2/day on days −5 to −4 and cyclophosphamide 60 mg/kg/day on days −3 to −2. In the nadir phase, the physician and nurses checked the patient's condition including oral conditions daily using the oral assessment guide to score the patient's voice, swallowing, lips, tongue, saliva, mucous membranes, gums, teeth, and dentures.
4
The records confirmed that there was no odontogenic infection. From day + 6, we also performed the same assessment after SCT. No acute exacerbation of periodontitis was detected. Oral mucositis was present from days + 2 to + 10. The attended hematologists and nurses shared this patient's treatment plan, the patient's condition, and oral assessment through conferences and we discussed the treatment plan with them. Seven and 13 months after FMD (6 and 12 months after SCT), the microbiological and clinical parameters of periodontitis, as demonstrated by the O’Leary plaque control record, decreased from 95% to 19%. Additionally, bleeding on probing (BOP), an indicator of periodontal inflammation (including gingivitis), decreased from 81.6% to 7.5% (Table 2). Probing depth (PD), defined as the distance from the bottom of the periodontal pocket to the gingival cuff, also improved. Deeper periodontal pockets indicate greater alveolar bone resorption. PD ≥4 mm and PD ≥7 mm significantly reduced after FMD from 98.3% to 4.6% and from 31.6% to 0%, respectively; the mean PD reduced from 6.1 to 2.5 mm (Table 2). At 13 months after FMD, gingival inflammation was effectively reduced (Figure 2 A,B), the teeth affected by severe periodontitis were preserved, alveolar bone was regenerated (Figure 2C,D), and functional occlusion was achieved without tooth extraction. Furthermore, qPCR data revealed no periodontitis‐related bacteria in the deepest periodontal pocket during the initial examination (Table 1).
TABLE 2 Clinical characteristics of the patient at baseline and 13 mo after FMD
Characteristics Baseline 13 mo after FMD
Plaque control record (%) 95 19
BOP (mean %) 81.6 7.5
PD ≥4 mm (mean %) 98.3 4.6
PD ≥7 mm (mean %) 31.6 0
PD (mm; mean) 6.1 2.5
Abbreviations: BOP, bleeding on probing; FMD, full‐mouth disinfection; PD, probing depth.
John Wiley & Sons, LtdFIGURE 2 Oral evaluation and dental radiographic images. The patient was a 38‐year‐old male hematopoietic stem cell transplant (SCT) recipient with generalized severe periodontitis. A, Intraoral image at the initial visit. B, Intraoral image after SCT and full‐mouth disinfection (FMD). C, Radiographic images at the initial visit. D, Radiographic images after SCT and FMD
3 DISCUSSION
We have demonstrated a case in which FMD was effective in treating generalized severe periodontitis without interfering with chemotherapy or SCT. Moreover, FMD effectively removed infectious sources and SRP was completed within only a few days throughout the entire jaw using antimicrobial agents. Upon re‐evaluation after SCT, we confirmed improvement of the periodontal status, bone regeneration at the resorption site, and observed no odontogenic infectious complications.
Full‐mouth disinfection with systemic antimicrobial therapy was conducted on this patient for three reasons: (a) The patient was diagnosed with acute lymphoblastic leukemia, indicating immunodeficiency; (b) intensive chemotherapy, total body irradiation, and SCT were scheduled at the first visit; and (c) intensive periodontal treatment reportedly reduces the occurrence of febrile neutropenia, significantly reducing CRP levels.
8
Before administering the conditioning regimen preceding hematopoietic SCT, the period available for dental treatment, including tooth extraction and periodontal basic therapy, is limited due to the development of neutropenia and thrombocytopenia. Moreover, the conventional method of nonsurgical periodontal therapy consists of SRP, which involves the debridement of periodontal pockets in the jaw quadrants (Q‐SRP).
9
,
10
However, as most periodontopathic bacteria exist in periodontal pockets as well as in several other oral mucosal sites, including the saliva,
11
the treated periodontal pockets could be reinfected from the untreated regions
12
during Q‐SRP. Quirynen et al
7
,
13
suggested a one‐stage FMD protocol instead of the conventional strategy with consecutive debridement per quadrant over a 1‐2‐week interval. On the contrary, Herrera and colleagues have suggested that antimicrobials should be administered immediately after the end of debridement and that debridement should be terminated in as short a time as possible, that is, within 1 week.
14
Yashima et al
15
found that if the SRP was completed within 1 week of maintaining the effective concentration of the antimicrobial agent (azithromycin), the clinical fungicidal efficacy of SRP was comparable to that of one‐stage full‐mouth SRP. Thus, termination of SRP within 1 week under antimicrobial administration would be equivalent in clinical efficacy to SRP within 24 hours. Furthermore, in practice, SRP below the gingival margin of the entire jaw in 1 day is known to be burdensome for patients and to cause a high frequency of fever.
7
Therefore, we planned to complete the SRP within approximately 1 week in this case while monitoring the patient's physical condition and response on the dental chair. Systematic review and meta‐analysis
16
have revealed that FMD combined with systemic administration of amoxicillin and metronidazole improved the clinical and microbiological outcomes significantly and effectively. These combined periodontitis treatments have not been approved in Japan. However, in our study, we systemically administrated sitafloxacin, which has been reported as significantly effective against periodontopathic bacteria.
17
,
18
Leukemic patients with pancytopenia and hematopoietic dysfunction are often immunodeficient, with difficulty in maintaining hemostasis, which increases the risks of postoperative infection and hemorrhage. Hence, it is preferable to conduct these treatments when patients attain normal hematopoiesis after the nadir phase, in which remission induction and consolidation chemotherapy suppress tumor cells. Thus, a desirable disinfection protocol includes (a) postinduction and consolidation chemotherapy; (b) following exit from the nadir phase, administration of antibiotics in combination with oral care; and (c) completion (with epithelialization) of the indicated dental surgical treatments before pretreatment (chemotherapy and total body irradiation) for SCT. Another study showed that reduced frequency of FN coincided with periodontal treatment.
19
These studies clearly demonstrate the importance of periodontal management in patients with leukemia and periodontal disease.
In our case, adverse events induced by dental treatment did not occur before or after SCT. Leukemic patients with severe odontogenic infection can experience infections within the periodontal deep pockets. For instance, a prospective observational study indicated that patients with periodontal inflammation had bacteremia more often than those with healthy periodontal tissue and that bleeding on probing was related to bacteremia.
20
These conditions may lead to bacteremia and septicemia before and after the SCT immunodeficient phase. Moreover, previous studies have shown that periodontal disease is associated with markedly elevated hospital charges, longer hospital stays, and higher rates of infectious complications in SCT recipients.
21
Furthermore, significantly lower frequencies of both oral and systemic infections were observed in patients undergoing chemotherapy that completed all dental treatments compared with those who did not. However, this difference was dependent on the chemotherapy grade.
5
,
22
Hence, dental treatment should be implemented prior to, during, and following SCT and chemotherapy. To this end, FMD can be integrated into existing protocols and regimens for hematology treatment without interfering with SCT. Furthermore, the lack of periodontitis‐related bacteria as detected by qPCR after FMD combined with systemic antimicrobial therapy suggests that FMD with systemic antimicrobial therapy may alter the oral microbiome of patients with severe periodontitis to a more desirable state.
6
,
16
However, the establishment of pre‐ and post‐SCT protocols based on individual patients’ oral condition is required to prevent oral adverse events. Overall, FMD can be performed rapidly and before SCT, providing effective oral care and improving the condition of leukemic patients with severe periodontitis.
CONFLICTS OF INTEREST
The authors declare that they have no conflicts of interest.
AUTHOR CONTRIBUTIONS
SM: provided clinical oral care, conducted full‐mouth disinfection, collected and analyzed the data, and primarily wrote the manuscript. KT: analyzed data and revised the manuscript. YM and MN: provided clinical oral care and analyzed the data. NH and SU: provided clinical oral care and commented on the manuscript. JK: performed the SCT, clinical care, and commented on the manuscript. TM and TN: contributed to writing, reviewing, and editing the manuscript.
ACKNOWLEDGMENTS
We thank the physicians and hospital staff of the Division of Haematology, Department of Medicine, Keio University School of Medicine, Tokyo, Japan, for providing thoughtful patient care. We would also like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The datasets used and analyzed during this study are available from the corresponding author on reasonable request. | ASPARAGINASE, CYCLOPHOSPHAMIDE, DAUNORUBICIN, DOXORUBICIN HYDROCHLORIDE, PREDNISOLONE, VINCRISTINE | DrugsGivenReaction | CC BY-NC-ND | 33598218 | 19,034,703 | 2021-02 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Septic shock'. | Management of generalized severe periodontitis using full-mouth disinfection and systemic antibiotics in a leukemic patient before stem cell transplantation: A case report.
The full-mouth disinfection protocol implemented in this case can be integrated into established protocols for treating severe periodontitis in the context of a hematological malignancy, without any interference with the cancer treatment.
1 INTRODUCTION
Oral infection is a contributor to morbidity and mortality in leukemic patients. Full‐mouth disinfection, which terminates subgingival debridement within a short period, is applicable and effective in combination with antimicrobial agents in treating leukemic patients with severe periodontitis as an oral infection removal procedure prior to stem cell transplantation.
Hematopoietic stem cell transplantation (SCT) is a curative therapy for hematologic malignancies. Standard pretransplantation therapeutic approaches, including high‐dose chemotherapy, total body irradiation, and SCT, greatly improve the prognosis of leukemic patients. At multiple phases during SCT, patients are at risk of contracting infectious diseases due to profound and prolonged neutropenia.
1
The US National Cancer Institute indicated that odontogenic infection is a potential source of systemic infections that should be eliminated by dental treatment.
2
,
3
However, no definitive criteria for extraction or preservation of infected teeth that do not affect hematological treatment have been established for recipients with severe infections, including patients with hematological malignancies.
Here, we present a case of acute lymphoblastic leukemia with severe periodontitis prior to SCT that was treated with a regimen involving full‐mouth disinfection (FMD).
2 CASE REPORT
A 38‐year‐old man diagnosed with acute lymphoblastic leukemia and scheduled to undergo chemotherapy, total body irradiation, and SCT was referred to our department for odontogenic infection screening. After evaluation of his clinical oral condition, dental X‐ray photographs, and the presence of periodontitis‐related bacteria by quantitative polymerase chain reaction (qPCR), a diagnosis of generalized severe periodontitis (generalized periodontitis stage IV, grade C) was made. The patient had already received intravenous antimicrobial infusion for febrile neutropenia (FN) on admission to the hospital; however, periodontitis‐related bacteria were detected (Table 1). When the leukocyte count was low, we only followed the oral hygiene instructions. To remove dental plaque and supragingival calculus, scaling procedures on the gingival margin were performed at the time of recovery from FN after induction of chemotherapy. To prevent bacteremia and septicemia after SCT, FMD was considered to remove the infectious source via completion of scaling and root planing (SRP) within the limited dental treatment window of only 1 week, covering all periodontally infected teeth (Figure 1). Before the initiation of chemotherapy, oral hygiene instructions were provided to the patient to avoid risk of infection from dental foci. In the neutropenic phase, during chemotherapy and immunosuppression, oral care was provided to reduce oral complications including oral mucositis and acute periodontitis. Since his first visit to our department, the patient was expected to develop chemotherapy‐induced FN and oral mucosal disorders. We thus provided the patient with thorough oral hygiene instructions aiming to quantitatively reduce the source of infection in the mouth. Specifically, the patient was instructed on the toothbrush Bass technique regarding the toothbrush and interdental brush cleaning methods. Oral assessments were performed based on an oral assessment guide,
4
and the information was shared with the physician and nurses. A nonirritating moisturizer was used for xerostomia.
TABLE 1 Quantitative evaluation of periodontitis‐related bacteria in the deepest periodontal pocket of the patient
Baseline (count) Seven months after FMD (count)
Total bacteria 30 000 <1000
Porphyromonas gingivalis
570 <10
Tannerella forsythia
1300 <10
Treponema denticola
1100 <10
Fusobacterium nucleatum
2000 <10
John Wiley & Sons, LtdFIGURE 1 Treatment timeline. The patient was a 38‐year‐old male hematopoietic stem cell transplant (SCT) recipient with generalized severe periodontitis
The patient achieved complete remission following induction of chemotherapy consisting of cyclophosphamide vincristine, prednisolone, daunorubicin, and L‐asparaginase and consolidation chemotherapy consisting of doxorubicin, vincristine, and prednisolone. The myelosuppression grades of these chemotherapies were categorized as severe and moderate, respectively.
5
Procalcitonin (PCT) levels were measured when fever and febrile neutropenia (FN) were present. PCT was predominantly higher (7.4 ng/mL) in the presence of FN and septic shock during the first round of chemotherapy on day −116. Otherwise, the PCT was not predominant during the pretransplant oral care and at 1 month after the transplant, which is the focus of this study (Figure 1). Blood cultures were performed at the time of fever, and the only positive results between the time of admission and 1 month after transplantation were obtained from day −116 to day −111. At that time, FN and septic shock were both present, and the patient was positive for Corynebacterium striatum and Staphylococcus haemolyticus. No oral bacteria were found. Mechanical debridement for severe periodontal disease was then scheduled during the non‐neutropenic period after chemotherapy induction and consolidation. The day on which SCT was performed was designated day 0. To gain sufficient healing time of SRP for debridement of the periodontal deep pocket before SCT and to avoid reinfection from an untreated site to the treatment site in this case, we planned for FMD combined with systemic antimicrobial therapy, which could effectively be used to treat patients who had been diagnosed with generalized severe periodontitis in a short period of time.
6
After the first round of chemotherapy, we considered performing the procedure if the white blood cell count was normal. However, after consulting with the hematologist, we decided to schedule the consolidation chemotherapy; thus, the medical team considered the time when the consolidation chemotherapy was finished, the white blood cell and absolute neutrophil counts had recovered, there was no febrile neutropenia, and there was enough time for the wound to heal after FMD. As a result of this discussion, we treated for severe periodontitis in a short period from day −30 to day −22. FMD was performed from days −30 to −22 for debridement of all periodontal pockets by SRP with Gracey curettes. We performed SRP for tooth numbers 36, 37, 38, 46, 47, and 48 on day −30 (Federation Dentaire Internationale System); tooth numbers 14‐18 and 24‐28 on day −27; tooth numbers 31, 32, 33, 35 and from 41‐44 on day −24; and finally, tooth numbers 13, 21, 22, and 23, on day −22. During FMD, the patient was treated with sitafloxacin (100 mg/day for 14 days) to prevent bacteremia/septicemia, periodontopathic bacterial infection, and fever, which are possible adverse complications induced by FMD.
7
Both body temperature and C‐reactive protein (CRP) level were stable during FMD period, with a maximum body temperature of 37.0°C and maximum CRP concentration of 1.13 mg/dL (Figure 1). Allogeneic peripheral blood SCT was performed with myeloablative conditioning on day 0. The graft contained CD34+ cells at 2.66 × 106/kg. Myeloablative conditioning was based on total body irradiation at 12 Gy in six fractions on day −8 to −6, administered in combination with cytarabine 2000 mg/m2/day on days −5 to −4 and cyclophosphamide 60 mg/kg/day on days −3 to −2. In the nadir phase, the physician and nurses checked the patient's condition including oral conditions daily using the oral assessment guide to score the patient's voice, swallowing, lips, tongue, saliva, mucous membranes, gums, teeth, and dentures.
4
The records confirmed that there was no odontogenic infection. From day + 6, we also performed the same assessment after SCT. No acute exacerbation of periodontitis was detected. Oral mucositis was present from days + 2 to + 10. The attended hematologists and nurses shared this patient's treatment plan, the patient's condition, and oral assessment through conferences and we discussed the treatment plan with them. Seven and 13 months after FMD (6 and 12 months after SCT), the microbiological and clinical parameters of periodontitis, as demonstrated by the O’Leary plaque control record, decreased from 95% to 19%. Additionally, bleeding on probing (BOP), an indicator of periodontal inflammation (including gingivitis), decreased from 81.6% to 7.5% (Table 2). Probing depth (PD), defined as the distance from the bottom of the periodontal pocket to the gingival cuff, also improved. Deeper periodontal pockets indicate greater alveolar bone resorption. PD ≥4 mm and PD ≥7 mm significantly reduced after FMD from 98.3% to 4.6% and from 31.6% to 0%, respectively; the mean PD reduced from 6.1 to 2.5 mm (Table 2). At 13 months after FMD, gingival inflammation was effectively reduced (Figure 2 A,B), the teeth affected by severe periodontitis were preserved, alveolar bone was regenerated (Figure 2C,D), and functional occlusion was achieved without tooth extraction. Furthermore, qPCR data revealed no periodontitis‐related bacteria in the deepest periodontal pocket during the initial examination (Table 1).
TABLE 2 Clinical characteristics of the patient at baseline and 13 mo after FMD
Characteristics Baseline 13 mo after FMD
Plaque control record (%) 95 19
BOP (mean %) 81.6 7.5
PD ≥4 mm (mean %) 98.3 4.6
PD ≥7 mm (mean %) 31.6 0
PD (mm; mean) 6.1 2.5
Abbreviations: BOP, bleeding on probing; FMD, full‐mouth disinfection; PD, probing depth.
John Wiley & Sons, LtdFIGURE 2 Oral evaluation and dental radiographic images. The patient was a 38‐year‐old male hematopoietic stem cell transplant (SCT) recipient with generalized severe periodontitis. A, Intraoral image at the initial visit. B, Intraoral image after SCT and full‐mouth disinfection (FMD). C, Radiographic images at the initial visit. D, Radiographic images after SCT and FMD
3 DISCUSSION
We have demonstrated a case in which FMD was effective in treating generalized severe periodontitis without interfering with chemotherapy or SCT. Moreover, FMD effectively removed infectious sources and SRP was completed within only a few days throughout the entire jaw using antimicrobial agents. Upon re‐evaluation after SCT, we confirmed improvement of the periodontal status, bone regeneration at the resorption site, and observed no odontogenic infectious complications.
Full‐mouth disinfection with systemic antimicrobial therapy was conducted on this patient for three reasons: (a) The patient was diagnosed with acute lymphoblastic leukemia, indicating immunodeficiency; (b) intensive chemotherapy, total body irradiation, and SCT were scheduled at the first visit; and (c) intensive periodontal treatment reportedly reduces the occurrence of febrile neutropenia, significantly reducing CRP levels.
8
Before administering the conditioning regimen preceding hematopoietic SCT, the period available for dental treatment, including tooth extraction and periodontal basic therapy, is limited due to the development of neutropenia and thrombocytopenia. Moreover, the conventional method of nonsurgical periodontal therapy consists of SRP, which involves the debridement of periodontal pockets in the jaw quadrants (Q‐SRP).
9
,
10
However, as most periodontopathic bacteria exist in periodontal pockets as well as in several other oral mucosal sites, including the saliva,
11
the treated periodontal pockets could be reinfected from the untreated regions
12
during Q‐SRP. Quirynen et al
7
,
13
suggested a one‐stage FMD protocol instead of the conventional strategy with consecutive debridement per quadrant over a 1‐2‐week interval. On the contrary, Herrera and colleagues have suggested that antimicrobials should be administered immediately after the end of debridement and that debridement should be terminated in as short a time as possible, that is, within 1 week.
14
Yashima et al
15
found that if the SRP was completed within 1 week of maintaining the effective concentration of the antimicrobial agent (azithromycin), the clinical fungicidal efficacy of SRP was comparable to that of one‐stage full‐mouth SRP. Thus, termination of SRP within 1 week under antimicrobial administration would be equivalent in clinical efficacy to SRP within 24 hours. Furthermore, in practice, SRP below the gingival margin of the entire jaw in 1 day is known to be burdensome for patients and to cause a high frequency of fever.
7
Therefore, we planned to complete the SRP within approximately 1 week in this case while monitoring the patient's physical condition and response on the dental chair. Systematic review and meta‐analysis
16
have revealed that FMD combined with systemic administration of amoxicillin and metronidazole improved the clinical and microbiological outcomes significantly and effectively. These combined periodontitis treatments have not been approved in Japan. However, in our study, we systemically administrated sitafloxacin, which has been reported as significantly effective against periodontopathic bacteria.
17
,
18
Leukemic patients with pancytopenia and hematopoietic dysfunction are often immunodeficient, with difficulty in maintaining hemostasis, which increases the risks of postoperative infection and hemorrhage. Hence, it is preferable to conduct these treatments when patients attain normal hematopoiesis after the nadir phase, in which remission induction and consolidation chemotherapy suppress tumor cells. Thus, a desirable disinfection protocol includes (a) postinduction and consolidation chemotherapy; (b) following exit from the nadir phase, administration of antibiotics in combination with oral care; and (c) completion (with epithelialization) of the indicated dental surgical treatments before pretreatment (chemotherapy and total body irradiation) for SCT. Another study showed that reduced frequency of FN coincided with periodontal treatment.
19
These studies clearly demonstrate the importance of periodontal management in patients with leukemia and periodontal disease.
In our case, adverse events induced by dental treatment did not occur before or after SCT. Leukemic patients with severe odontogenic infection can experience infections within the periodontal deep pockets. For instance, a prospective observational study indicated that patients with periodontal inflammation had bacteremia more often than those with healthy periodontal tissue and that bleeding on probing was related to bacteremia.
20
These conditions may lead to bacteremia and septicemia before and after the SCT immunodeficient phase. Moreover, previous studies have shown that periodontal disease is associated with markedly elevated hospital charges, longer hospital stays, and higher rates of infectious complications in SCT recipients.
21
Furthermore, significantly lower frequencies of both oral and systemic infections were observed in patients undergoing chemotherapy that completed all dental treatments compared with those who did not. However, this difference was dependent on the chemotherapy grade.
5
,
22
Hence, dental treatment should be implemented prior to, during, and following SCT and chemotherapy. To this end, FMD can be integrated into existing protocols and regimens for hematology treatment without interfering with SCT. Furthermore, the lack of periodontitis‐related bacteria as detected by qPCR after FMD combined with systemic antimicrobial therapy suggests that FMD with systemic antimicrobial therapy may alter the oral microbiome of patients with severe periodontitis to a more desirable state.
6
,
16
However, the establishment of pre‐ and post‐SCT protocols based on individual patients’ oral condition is required to prevent oral adverse events. Overall, FMD can be performed rapidly and before SCT, providing effective oral care and improving the condition of leukemic patients with severe periodontitis.
CONFLICTS OF INTEREST
The authors declare that they have no conflicts of interest.
AUTHOR CONTRIBUTIONS
SM: provided clinical oral care, conducted full‐mouth disinfection, collected and analyzed the data, and primarily wrote the manuscript. KT: analyzed data and revised the manuscript. YM and MN: provided clinical oral care and analyzed the data. NH and SU: provided clinical oral care and commented on the manuscript. JK: performed the SCT, clinical care, and commented on the manuscript. TM and TN: contributed to writing, reviewing, and editing the manuscript.
ACKNOWLEDGMENTS
We thank the physicians and hospital staff of the Division of Haematology, Department of Medicine, Keio University School of Medicine, Tokyo, Japan, for providing thoughtful patient care. We would also like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The datasets used and analyzed during this study are available from the corresponding author on reasonable request. | ASPARAGINASE, CYCLOPHOSPHAMIDE, DAUNORUBICIN, DOXORUBICIN HYDROCHLORIDE, PREDNISOLONE, VINCRISTINE | DrugsGivenReaction | CC BY-NC-ND | 33598218 | 19,034,703 | 2021-02 |
What was the dosage of drug 'CYCLOPHOSPHAMIDE'? | Management of generalized severe periodontitis using full-mouth disinfection and systemic antibiotics in a leukemic patient before stem cell transplantation: A case report.
The full-mouth disinfection protocol implemented in this case can be integrated into established protocols for treating severe periodontitis in the context of a hematological malignancy, without any interference with the cancer treatment.
1 INTRODUCTION
Oral infection is a contributor to morbidity and mortality in leukemic patients. Full‐mouth disinfection, which terminates subgingival debridement within a short period, is applicable and effective in combination with antimicrobial agents in treating leukemic patients with severe periodontitis as an oral infection removal procedure prior to stem cell transplantation.
Hematopoietic stem cell transplantation (SCT) is a curative therapy for hematologic malignancies. Standard pretransplantation therapeutic approaches, including high‐dose chemotherapy, total body irradiation, and SCT, greatly improve the prognosis of leukemic patients. At multiple phases during SCT, patients are at risk of contracting infectious diseases due to profound and prolonged neutropenia.
1
The US National Cancer Institute indicated that odontogenic infection is a potential source of systemic infections that should be eliminated by dental treatment.
2
,
3
However, no definitive criteria for extraction or preservation of infected teeth that do not affect hematological treatment have been established for recipients with severe infections, including patients with hematological malignancies.
Here, we present a case of acute lymphoblastic leukemia with severe periodontitis prior to SCT that was treated with a regimen involving full‐mouth disinfection (FMD).
2 CASE REPORT
A 38‐year‐old man diagnosed with acute lymphoblastic leukemia and scheduled to undergo chemotherapy, total body irradiation, and SCT was referred to our department for odontogenic infection screening. After evaluation of his clinical oral condition, dental X‐ray photographs, and the presence of periodontitis‐related bacteria by quantitative polymerase chain reaction (qPCR), a diagnosis of generalized severe periodontitis (generalized periodontitis stage IV, grade C) was made. The patient had already received intravenous antimicrobial infusion for febrile neutropenia (FN) on admission to the hospital; however, periodontitis‐related bacteria were detected (Table 1). When the leukocyte count was low, we only followed the oral hygiene instructions. To remove dental plaque and supragingival calculus, scaling procedures on the gingival margin were performed at the time of recovery from FN after induction of chemotherapy. To prevent bacteremia and septicemia after SCT, FMD was considered to remove the infectious source via completion of scaling and root planing (SRP) within the limited dental treatment window of only 1 week, covering all periodontally infected teeth (Figure 1). Before the initiation of chemotherapy, oral hygiene instructions were provided to the patient to avoid risk of infection from dental foci. In the neutropenic phase, during chemotherapy and immunosuppression, oral care was provided to reduce oral complications including oral mucositis and acute periodontitis. Since his first visit to our department, the patient was expected to develop chemotherapy‐induced FN and oral mucosal disorders. We thus provided the patient with thorough oral hygiene instructions aiming to quantitatively reduce the source of infection in the mouth. Specifically, the patient was instructed on the toothbrush Bass technique regarding the toothbrush and interdental brush cleaning methods. Oral assessments were performed based on an oral assessment guide,
4
and the information was shared with the physician and nurses. A nonirritating moisturizer was used for xerostomia.
TABLE 1 Quantitative evaluation of periodontitis‐related bacteria in the deepest periodontal pocket of the patient
Baseline (count) Seven months after FMD (count)
Total bacteria 30 000 <1000
Porphyromonas gingivalis
570 <10
Tannerella forsythia
1300 <10
Treponema denticola
1100 <10
Fusobacterium nucleatum
2000 <10
John Wiley & Sons, LtdFIGURE 1 Treatment timeline. The patient was a 38‐year‐old male hematopoietic stem cell transplant (SCT) recipient with generalized severe periodontitis
The patient achieved complete remission following induction of chemotherapy consisting of cyclophosphamide vincristine, prednisolone, daunorubicin, and L‐asparaginase and consolidation chemotherapy consisting of doxorubicin, vincristine, and prednisolone. The myelosuppression grades of these chemotherapies were categorized as severe and moderate, respectively.
5
Procalcitonin (PCT) levels were measured when fever and febrile neutropenia (FN) were present. PCT was predominantly higher (7.4 ng/mL) in the presence of FN and septic shock during the first round of chemotherapy on day −116. Otherwise, the PCT was not predominant during the pretransplant oral care and at 1 month after the transplant, which is the focus of this study (Figure 1). Blood cultures were performed at the time of fever, and the only positive results between the time of admission and 1 month after transplantation were obtained from day −116 to day −111. At that time, FN and septic shock were both present, and the patient was positive for Corynebacterium striatum and Staphylococcus haemolyticus. No oral bacteria were found. Mechanical debridement for severe periodontal disease was then scheduled during the non‐neutropenic period after chemotherapy induction and consolidation. The day on which SCT was performed was designated day 0. To gain sufficient healing time of SRP for debridement of the periodontal deep pocket before SCT and to avoid reinfection from an untreated site to the treatment site in this case, we planned for FMD combined with systemic antimicrobial therapy, which could effectively be used to treat patients who had been diagnosed with generalized severe periodontitis in a short period of time.
6
After the first round of chemotherapy, we considered performing the procedure if the white blood cell count was normal. However, after consulting with the hematologist, we decided to schedule the consolidation chemotherapy; thus, the medical team considered the time when the consolidation chemotherapy was finished, the white blood cell and absolute neutrophil counts had recovered, there was no febrile neutropenia, and there was enough time for the wound to heal after FMD. As a result of this discussion, we treated for severe periodontitis in a short period from day −30 to day −22. FMD was performed from days −30 to −22 for debridement of all periodontal pockets by SRP with Gracey curettes. We performed SRP for tooth numbers 36, 37, 38, 46, 47, and 48 on day −30 (Federation Dentaire Internationale System); tooth numbers 14‐18 and 24‐28 on day −27; tooth numbers 31, 32, 33, 35 and from 41‐44 on day −24; and finally, tooth numbers 13, 21, 22, and 23, on day −22. During FMD, the patient was treated with sitafloxacin (100 mg/day for 14 days) to prevent bacteremia/septicemia, periodontopathic bacterial infection, and fever, which are possible adverse complications induced by FMD.
7
Both body temperature and C‐reactive protein (CRP) level were stable during FMD period, with a maximum body temperature of 37.0°C and maximum CRP concentration of 1.13 mg/dL (Figure 1). Allogeneic peripheral blood SCT was performed with myeloablative conditioning on day 0. The graft contained CD34+ cells at 2.66 × 106/kg. Myeloablative conditioning was based on total body irradiation at 12 Gy in six fractions on day −8 to −6, administered in combination with cytarabine 2000 mg/m2/day on days −5 to −4 and cyclophosphamide 60 mg/kg/day on days −3 to −2. In the nadir phase, the physician and nurses checked the patient's condition including oral conditions daily using the oral assessment guide to score the patient's voice, swallowing, lips, tongue, saliva, mucous membranes, gums, teeth, and dentures.
4
The records confirmed that there was no odontogenic infection. From day + 6, we also performed the same assessment after SCT. No acute exacerbation of periodontitis was detected. Oral mucositis was present from days + 2 to + 10. The attended hematologists and nurses shared this patient's treatment plan, the patient's condition, and oral assessment through conferences and we discussed the treatment plan with them. Seven and 13 months after FMD (6 and 12 months after SCT), the microbiological and clinical parameters of periodontitis, as demonstrated by the O’Leary plaque control record, decreased from 95% to 19%. Additionally, bleeding on probing (BOP), an indicator of periodontal inflammation (including gingivitis), decreased from 81.6% to 7.5% (Table 2). Probing depth (PD), defined as the distance from the bottom of the periodontal pocket to the gingival cuff, also improved. Deeper periodontal pockets indicate greater alveolar bone resorption. PD ≥4 mm and PD ≥7 mm significantly reduced after FMD from 98.3% to 4.6% and from 31.6% to 0%, respectively; the mean PD reduced from 6.1 to 2.5 mm (Table 2). At 13 months after FMD, gingival inflammation was effectively reduced (Figure 2 A,B), the teeth affected by severe periodontitis were preserved, alveolar bone was regenerated (Figure 2C,D), and functional occlusion was achieved without tooth extraction. Furthermore, qPCR data revealed no periodontitis‐related bacteria in the deepest periodontal pocket during the initial examination (Table 1).
TABLE 2 Clinical characteristics of the patient at baseline and 13 mo after FMD
Characteristics Baseline 13 mo after FMD
Plaque control record (%) 95 19
BOP (mean %) 81.6 7.5
PD ≥4 mm (mean %) 98.3 4.6
PD ≥7 mm (mean %) 31.6 0
PD (mm; mean) 6.1 2.5
Abbreviations: BOP, bleeding on probing; FMD, full‐mouth disinfection; PD, probing depth.
John Wiley & Sons, LtdFIGURE 2 Oral evaluation and dental radiographic images. The patient was a 38‐year‐old male hematopoietic stem cell transplant (SCT) recipient with generalized severe periodontitis. A, Intraoral image at the initial visit. B, Intraoral image after SCT and full‐mouth disinfection (FMD). C, Radiographic images at the initial visit. D, Radiographic images after SCT and FMD
3 DISCUSSION
We have demonstrated a case in which FMD was effective in treating generalized severe periodontitis without interfering with chemotherapy or SCT. Moreover, FMD effectively removed infectious sources and SRP was completed within only a few days throughout the entire jaw using antimicrobial agents. Upon re‐evaluation after SCT, we confirmed improvement of the periodontal status, bone regeneration at the resorption site, and observed no odontogenic infectious complications.
Full‐mouth disinfection with systemic antimicrobial therapy was conducted on this patient for three reasons: (a) The patient was diagnosed with acute lymphoblastic leukemia, indicating immunodeficiency; (b) intensive chemotherapy, total body irradiation, and SCT were scheduled at the first visit; and (c) intensive periodontal treatment reportedly reduces the occurrence of febrile neutropenia, significantly reducing CRP levels.
8
Before administering the conditioning regimen preceding hematopoietic SCT, the period available for dental treatment, including tooth extraction and periodontal basic therapy, is limited due to the development of neutropenia and thrombocytopenia. Moreover, the conventional method of nonsurgical periodontal therapy consists of SRP, which involves the debridement of periodontal pockets in the jaw quadrants (Q‐SRP).
9
,
10
However, as most periodontopathic bacteria exist in periodontal pockets as well as in several other oral mucosal sites, including the saliva,
11
the treated periodontal pockets could be reinfected from the untreated regions
12
during Q‐SRP. Quirynen et al
7
,
13
suggested a one‐stage FMD protocol instead of the conventional strategy with consecutive debridement per quadrant over a 1‐2‐week interval. On the contrary, Herrera and colleagues have suggested that antimicrobials should be administered immediately after the end of debridement and that debridement should be terminated in as short a time as possible, that is, within 1 week.
14
Yashima et al
15
found that if the SRP was completed within 1 week of maintaining the effective concentration of the antimicrobial agent (azithromycin), the clinical fungicidal efficacy of SRP was comparable to that of one‐stage full‐mouth SRP. Thus, termination of SRP within 1 week under antimicrobial administration would be equivalent in clinical efficacy to SRP within 24 hours. Furthermore, in practice, SRP below the gingival margin of the entire jaw in 1 day is known to be burdensome for patients and to cause a high frequency of fever.
7
Therefore, we planned to complete the SRP within approximately 1 week in this case while monitoring the patient's physical condition and response on the dental chair. Systematic review and meta‐analysis
16
have revealed that FMD combined with systemic administration of amoxicillin and metronidazole improved the clinical and microbiological outcomes significantly and effectively. These combined periodontitis treatments have not been approved in Japan. However, in our study, we systemically administrated sitafloxacin, which has been reported as significantly effective against periodontopathic bacteria.
17
,
18
Leukemic patients with pancytopenia and hematopoietic dysfunction are often immunodeficient, with difficulty in maintaining hemostasis, which increases the risks of postoperative infection and hemorrhage. Hence, it is preferable to conduct these treatments when patients attain normal hematopoiesis after the nadir phase, in which remission induction and consolidation chemotherapy suppress tumor cells. Thus, a desirable disinfection protocol includes (a) postinduction and consolidation chemotherapy; (b) following exit from the nadir phase, administration of antibiotics in combination with oral care; and (c) completion (with epithelialization) of the indicated dental surgical treatments before pretreatment (chemotherapy and total body irradiation) for SCT. Another study showed that reduced frequency of FN coincided with periodontal treatment.
19
These studies clearly demonstrate the importance of periodontal management in patients with leukemia and periodontal disease.
In our case, adverse events induced by dental treatment did not occur before or after SCT. Leukemic patients with severe odontogenic infection can experience infections within the periodontal deep pockets. For instance, a prospective observational study indicated that patients with periodontal inflammation had bacteremia more often than those with healthy periodontal tissue and that bleeding on probing was related to bacteremia.
20
These conditions may lead to bacteremia and septicemia before and after the SCT immunodeficient phase. Moreover, previous studies have shown that periodontal disease is associated with markedly elevated hospital charges, longer hospital stays, and higher rates of infectious complications in SCT recipients.
21
Furthermore, significantly lower frequencies of both oral and systemic infections were observed in patients undergoing chemotherapy that completed all dental treatments compared with those who did not. However, this difference was dependent on the chemotherapy grade.
5
,
22
Hence, dental treatment should be implemented prior to, during, and following SCT and chemotherapy. To this end, FMD can be integrated into existing protocols and regimens for hematology treatment without interfering with SCT. Furthermore, the lack of periodontitis‐related bacteria as detected by qPCR after FMD combined with systemic antimicrobial therapy suggests that FMD with systemic antimicrobial therapy may alter the oral microbiome of patients with severe periodontitis to a more desirable state.
6
,
16
However, the establishment of pre‐ and post‐SCT protocols based on individual patients’ oral condition is required to prevent oral adverse events. Overall, FMD can be performed rapidly and before SCT, providing effective oral care and improving the condition of leukemic patients with severe periodontitis.
CONFLICTS OF INTEREST
The authors declare that they have no conflicts of interest.
AUTHOR CONTRIBUTIONS
SM: provided clinical oral care, conducted full‐mouth disinfection, collected and analyzed the data, and primarily wrote the manuscript. KT: analyzed data and revised the manuscript. YM and MN: provided clinical oral care and analyzed the data. NH and SU: provided clinical oral care and commented on the manuscript. JK: performed the SCT, clinical care, and commented on the manuscript. TM and TN: contributed to writing, reviewing, and editing the manuscript.
ACKNOWLEDGMENTS
We thank the physicians and hospital staff of the Division of Haematology, Department of Medicine, Keio University School of Medicine, Tokyo, Japan, for providing thoughtful patient care. We would also like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The datasets used and analyzed during this study are available from the corresponding author on reasonable request. | ON DAYS?3 TO 2. | DrugDosageText | CC BY-NC-ND | 33598218 | 19,034,703 | 2021-02 |
What was the outcome of reaction 'Febrile neutropenia'? | Management of generalized severe periodontitis using full-mouth disinfection and systemic antibiotics in a leukemic patient before stem cell transplantation: A case report.
The full-mouth disinfection protocol implemented in this case can be integrated into established protocols for treating severe periodontitis in the context of a hematological malignancy, without any interference with the cancer treatment.
1 INTRODUCTION
Oral infection is a contributor to morbidity and mortality in leukemic patients. Full‐mouth disinfection, which terminates subgingival debridement within a short period, is applicable and effective in combination with antimicrobial agents in treating leukemic patients with severe periodontitis as an oral infection removal procedure prior to stem cell transplantation.
Hematopoietic stem cell transplantation (SCT) is a curative therapy for hematologic malignancies. Standard pretransplantation therapeutic approaches, including high‐dose chemotherapy, total body irradiation, and SCT, greatly improve the prognosis of leukemic patients. At multiple phases during SCT, patients are at risk of contracting infectious diseases due to profound and prolonged neutropenia.
1
The US National Cancer Institute indicated that odontogenic infection is a potential source of systemic infections that should be eliminated by dental treatment.
2
,
3
However, no definitive criteria for extraction or preservation of infected teeth that do not affect hematological treatment have been established for recipients with severe infections, including patients with hematological malignancies.
Here, we present a case of acute lymphoblastic leukemia with severe periodontitis prior to SCT that was treated with a regimen involving full‐mouth disinfection (FMD).
2 CASE REPORT
A 38‐year‐old man diagnosed with acute lymphoblastic leukemia and scheduled to undergo chemotherapy, total body irradiation, and SCT was referred to our department for odontogenic infection screening. After evaluation of his clinical oral condition, dental X‐ray photographs, and the presence of periodontitis‐related bacteria by quantitative polymerase chain reaction (qPCR), a diagnosis of generalized severe periodontitis (generalized periodontitis stage IV, grade C) was made. The patient had already received intravenous antimicrobial infusion for febrile neutropenia (FN) on admission to the hospital; however, periodontitis‐related bacteria were detected (Table 1). When the leukocyte count was low, we only followed the oral hygiene instructions. To remove dental plaque and supragingival calculus, scaling procedures on the gingival margin were performed at the time of recovery from FN after induction of chemotherapy. To prevent bacteremia and septicemia after SCT, FMD was considered to remove the infectious source via completion of scaling and root planing (SRP) within the limited dental treatment window of only 1 week, covering all periodontally infected teeth (Figure 1). Before the initiation of chemotherapy, oral hygiene instructions were provided to the patient to avoid risk of infection from dental foci. In the neutropenic phase, during chemotherapy and immunosuppression, oral care was provided to reduce oral complications including oral mucositis and acute periodontitis. Since his first visit to our department, the patient was expected to develop chemotherapy‐induced FN and oral mucosal disorders. We thus provided the patient with thorough oral hygiene instructions aiming to quantitatively reduce the source of infection in the mouth. Specifically, the patient was instructed on the toothbrush Bass technique regarding the toothbrush and interdental brush cleaning methods. Oral assessments were performed based on an oral assessment guide,
4
and the information was shared with the physician and nurses. A nonirritating moisturizer was used for xerostomia.
TABLE 1 Quantitative evaluation of periodontitis‐related bacteria in the deepest periodontal pocket of the patient
Baseline (count) Seven months after FMD (count)
Total bacteria 30 000 <1000
Porphyromonas gingivalis
570 <10
Tannerella forsythia
1300 <10
Treponema denticola
1100 <10
Fusobacterium nucleatum
2000 <10
John Wiley & Sons, LtdFIGURE 1 Treatment timeline. The patient was a 38‐year‐old male hematopoietic stem cell transplant (SCT) recipient with generalized severe periodontitis
The patient achieved complete remission following induction of chemotherapy consisting of cyclophosphamide vincristine, prednisolone, daunorubicin, and L‐asparaginase and consolidation chemotherapy consisting of doxorubicin, vincristine, and prednisolone. The myelosuppression grades of these chemotherapies were categorized as severe and moderate, respectively.
5
Procalcitonin (PCT) levels were measured when fever and febrile neutropenia (FN) were present. PCT was predominantly higher (7.4 ng/mL) in the presence of FN and septic shock during the first round of chemotherapy on day −116. Otherwise, the PCT was not predominant during the pretransplant oral care and at 1 month after the transplant, which is the focus of this study (Figure 1). Blood cultures were performed at the time of fever, and the only positive results between the time of admission and 1 month after transplantation were obtained from day −116 to day −111. At that time, FN and septic shock were both present, and the patient was positive for Corynebacterium striatum and Staphylococcus haemolyticus. No oral bacteria were found. Mechanical debridement for severe periodontal disease was then scheduled during the non‐neutropenic period after chemotherapy induction and consolidation. The day on which SCT was performed was designated day 0. To gain sufficient healing time of SRP for debridement of the periodontal deep pocket before SCT and to avoid reinfection from an untreated site to the treatment site in this case, we planned for FMD combined with systemic antimicrobial therapy, which could effectively be used to treat patients who had been diagnosed with generalized severe periodontitis in a short period of time.
6
After the first round of chemotherapy, we considered performing the procedure if the white blood cell count was normal. However, after consulting with the hematologist, we decided to schedule the consolidation chemotherapy; thus, the medical team considered the time when the consolidation chemotherapy was finished, the white blood cell and absolute neutrophil counts had recovered, there was no febrile neutropenia, and there was enough time for the wound to heal after FMD. As a result of this discussion, we treated for severe periodontitis in a short period from day −30 to day −22. FMD was performed from days −30 to −22 for debridement of all periodontal pockets by SRP with Gracey curettes. We performed SRP for tooth numbers 36, 37, 38, 46, 47, and 48 on day −30 (Federation Dentaire Internationale System); tooth numbers 14‐18 and 24‐28 on day −27; tooth numbers 31, 32, 33, 35 and from 41‐44 on day −24; and finally, tooth numbers 13, 21, 22, and 23, on day −22. During FMD, the patient was treated with sitafloxacin (100 mg/day for 14 days) to prevent bacteremia/septicemia, periodontopathic bacterial infection, and fever, which are possible adverse complications induced by FMD.
7
Both body temperature and C‐reactive protein (CRP) level were stable during FMD period, with a maximum body temperature of 37.0°C and maximum CRP concentration of 1.13 mg/dL (Figure 1). Allogeneic peripheral blood SCT was performed with myeloablative conditioning on day 0. The graft contained CD34+ cells at 2.66 × 106/kg. Myeloablative conditioning was based on total body irradiation at 12 Gy in six fractions on day −8 to −6, administered in combination with cytarabine 2000 mg/m2/day on days −5 to −4 and cyclophosphamide 60 mg/kg/day on days −3 to −2. In the nadir phase, the physician and nurses checked the patient's condition including oral conditions daily using the oral assessment guide to score the patient's voice, swallowing, lips, tongue, saliva, mucous membranes, gums, teeth, and dentures.
4
The records confirmed that there was no odontogenic infection. From day + 6, we also performed the same assessment after SCT. No acute exacerbation of periodontitis was detected. Oral mucositis was present from days + 2 to + 10. The attended hematologists and nurses shared this patient's treatment plan, the patient's condition, and oral assessment through conferences and we discussed the treatment plan with them. Seven and 13 months after FMD (6 and 12 months after SCT), the microbiological and clinical parameters of periodontitis, as demonstrated by the O’Leary plaque control record, decreased from 95% to 19%. Additionally, bleeding on probing (BOP), an indicator of periodontal inflammation (including gingivitis), decreased from 81.6% to 7.5% (Table 2). Probing depth (PD), defined as the distance from the bottom of the periodontal pocket to the gingival cuff, also improved. Deeper periodontal pockets indicate greater alveolar bone resorption. PD ≥4 mm and PD ≥7 mm significantly reduced after FMD from 98.3% to 4.6% and from 31.6% to 0%, respectively; the mean PD reduced from 6.1 to 2.5 mm (Table 2). At 13 months after FMD, gingival inflammation was effectively reduced (Figure 2 A,B), the teeth affected by severe periodontitis were preserved, alveolar bone was regenerated (Figure 2C,D), and functional occlusion was achieved without tooth extraction. Furthermore, qPCR data revealed no periodontitis‐related bacteria in the deepest periodontal pocket during the initial examination (Table 1).
TABLE 2 Clinical characteristics of the patient at baseline and 13 mo after FMD
Characteristics Baseline 13 mo after FMD
Plaque control record (%) 95 19
BOP (mean %) 81.6 7.5
PD ≥4 mm (mean %) 98.3 4.6
PD ≥7 mm (mean %) 31.6 0
PD (mm; mean) 6.1 2.5
Abbreviations: BOP, bleeding on probing; FMD, full‐mouth disinfection; PD, probing depth.
John Wiley & Sons, LtdFIGURE 2 Oral evaluation and dental radiographic images. The patient was a 38‐year‐old male hematopoietic stem cell transplant (SCT) recipient with generalized severe periodontitis. A, Intraoral image at the initial visit. B, Intraoral image after SCT and full‐mouth disinfection (FMD). C, Radiographic images at the initial visit. D, Radiographic images after SCT and FMD
3 DISCUSSION
We have demonstrated a case in which FMD was effective in treating generalized severe periodontitis without interfering with chemotherapy or SCT. Moreover, FMD effectively removed infectious sources and SRP was completed within only a few days throughout the entire jaw using antimicrobial agents. Upon re‐evaluation after SCT, we confirmed improvement of the periodontal status, bone regeneration at the resorption site, and observed no odontogenic infectious complications.
Full‐mouth disinfection with systemic antimicrobial therapy was conducted on this patient for three reasons: (a) The patient was diagnosed with acute lymphoblastic leukemia, indicating immunodeficiency; (b) intensive chemotherapy, total body irradiation, and SCT were scheduled at the first visit; and (c) intensive periodontal treatment reportedly reduces the occurrence of febrile neutropenia, significantly reducing CRP levels.
8
Before administering the conditioning regimen preceding hematopoietic SCT, the period available for dental treatment, including tooth extraction and periodontal basic therapy, is limited due to the development of neutropenia and thrombocytopenia. Moreover, the conventional method of nonsurgical periodontal therapy consists of SRP, which involves the debridement of periodontal pockets in the jaw quadrants (Q‐SRP).
9
,
10
However, as most periodontopathic bacteria exist in periodontal pockets as well as in several other oral mucosal sites, including the saliva,
11
the treated periodontal pockets could be reinfected from the untreated regions
12
during Q‐SRP. Quirynen et al
7
,
13
suggested a one‐stage FMD protocol instead of the conventional strategy with consecutive debridement per quadrant over a 1‐2‐week interval. On the contrary, Herrera and colleagues have suggested that antimicrobials should be administered immediately after the end of debridement and that debridement should be terminated in as short a time as possible, that is, within 1 week.
14
Yashima et al
15
found that if the SRP was completed within 1 week of maintaining the effective concentration of the antimicrobial agent (azithromycin), the clinical fungicidal efficacy of SRP was comparable to that of one‐stage full‐mouth SRP. Thus, termination of SRP within 1 week under antimicrobial administration would be equivalent in clinical efficacy to SRP within 24 hours. Furthermore, in practice, SRP below the gingival margin of the entire jaw in 1 day is known to be burdensome for patients and to cause a high frequency of fever.
7
Therefore, we planned to complete the SRP within approximately 1 week in this case while monitoring the patient's physical condition and response on the dental chair. Systematic review and meta‐analysis
16
have revealed that FMD combined with systemic administration of amoxicillin and metronidazole improved the clinical and microbiological outcomes significantly and effectively. These combined periodontitis treatments have not been approved in Japan. However, in our study, we systemically administrated sitafloxacin, which has been reported as significantly effective against periodontopathic bacteria.
17
,
18
Leukemic patients with pancytopenia and hematopoietic dysfunction are often immunodeficient, with difficulty in maintaining hemostasis, which increases the risks of postoperative infection and hemorrhage. Hence, it is preferable to conduct these treatments when patients attain normal hematopoiesis after the nadir phase, in which remission induction and consolidation chemotherapy suppress tumor cells. Thus, a desirable disinfection protocol includes (a) postinduction and consolidation chemotherapy; (b) following exit from the nadir phase, administration of antibiotics in combination with oral care; and (c) completion (with epithelialization) of the indicated dental surgical treatments before pretreatment (chemotherapy and total body irradiation) for SCT. Another study showed that reduced frequency of FN coincided with periodontal treatment.
19
These studies clearly demonstrate the importance of periodontal management in patients with leukemia and periodontal disease.
In our case, adverse events induced by dental treatment did not occur before or after SCT. Leukemic patients with severe odontogenic infection can experience infections within the periodontal deep pockets. For instance, a prospective observational study indicated that patients with periodontal inflammation had bacteremia more often than those with healthy periodontal tissue and that bleeding on probing was related to bacteremia.
20
These conditions may lead to bacteremia and septicemia before and after the SCT immunodeficient phase. Moreover, previous studies have shown that periodontal disease is associated with markedly elevated hospital charges, longer hospital stays, and higher rates of infectious complications in SCT recipients.
21
Furthermore, significantly lower frequencies of both oral and systemic infections were observed in patients undergoing chemotherapy that completed all dental treatments compared with those who did not. However, this difference was dependent on the chemotherapy grade.
5
,
22
Hence, dental treatment should be implemented prior to, during, and following SCT and chemotherapy. To this end, FMD can be integrated into existing protocols and regimens for hematology treatment without interfering with SCT. Furthermore, the lack of periodontitis‐related bacteria as detected by qPCR after FMD combined with systemic antimicrobial therapy suggests that FMD with systemic antimicrobial therapy may alter the oral microbiome of patients with severe periodontitis to a more desirable state.
6
,
16
However, the establishment of pre‐ and post‐SCT protocols based on individual patients’ oral condition is required to prevent oral adverse events. Overall, FMD can be performed rapidly and before SCT, providing effective oral care and improving the condition of leukemic patients with severe periodontitis.
CONFLICTS OF INTEREST
The authors declare that they have no conflicts of interest.
AUTHOR CONTRIBUTIONS
SM: provided clinical oral care, conducted full‐mouth disinfection, collected and analyzed the data, and primarily wrote the manuscript. KT: analyzed data and revised the manuscript. YM and MN: provided clinical oral care and analyzed the data. NH and SU: provided clinical oral care and commented on the manuscript. JK: performed the SCT, clinical care, and commented on the manuscript. TM and TN: contributed to writing, reviewing, and editing the manuscript.
ACKNOWLEDGMENTS
We thank the physicians and hospital staff of the Division of Haematology, Department of Medicine, Keio University School of Medicine, Tokyo, Japan, for providing thoughtful patient care. We would also like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The datasets used and analyzed during this study are available from the corresponding author on reasonable request. | Recovered | ReactionOutcome | CC BY-NC-ND | 33598218 | 19,052,640 | 2021-02 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Bundle branch block right'. | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | AMITRIPTYLINE, CHLORPROMAZINE, DULOXETINE, MAGNESIUM OXIDE, NITRAZEPAM, RABEPRAZOLE SODIUM, TRAZODONE HYDROCHLORIDE, ZOLPIDEM TARTRATE | DrugsGivenReaction | CC BY-NC-ND | 33598249 | 18,711,570 | 2021-02 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Cardiac arrest'. | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | AMITRIPTYLINE, CHLORPROMAZINE, DULOXETINE, MAGNESIUM OXIDE, NITRAZEPAM, RABEPRAZOLE, TRAZODONE HYDROCHLORIDE, ZOLPIDEM TARTRATE | DrugsGivenReaction | CC BY-NC-ND | 33598249 | 18,738,472 | 2021-02 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Defect conduction intraventricular'. | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | AMITRIPTYLINE, CHLORPROMAZINE, DULOXETINE, MAGNESIUM OXIDE, NITRAZEPAM, RABEPRAZOLE SODIUM, TRAZODONE HYDROCHLORIDE, ZOLPIDEM TARTRATE | DrugsGivenReaction | CC BY-NC-ND | 33598249 | 18,711,570 | 2021-02 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Drug level increased'. | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | AMITRIPTYLINE, CHLORPROMAZINE, DULOXETINE, MAGNESIUM OXIDE, NITRAZEPAM, RABEPRAZOLE SODIUM, TRAZODONE HYDROCHLORIDE, ZOLPIDEM TARTRATE | DrugsGivenReaction | CC BY-NC-ND | 33598249 | 18,711,570 | 2021-02 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Electrocardiogram T wave abnormal'. | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | AMITRIPTYLINE, CHLORPROMAZINE, DULOXETINE, MAGNESIUM OXIDE, NITRAZEPAM, RABEPRAZOLE SODIUM, TRAZODONE HYDROCHLORIDE, ZOLPIDEM TARTRATE | DrugsGivenReaction | CC BY-NC-ND | 33598249 | 18,711,570 | 2021-02 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Haemodynamic instability'. | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | AMITRIPTYLINE, CHLORPROMAZINE, DULOXETINE, MAGNESIUM OXIDE, NITRAZEPAM, RABEPRAZOLE SODIUM, TRAZODONE HYDROCHLORIDE, ZOLPIDEM TARTRATE | DrugsGivenReaction | CC BY-NC-ND | 33598249 | 18,711,570 | 2021-02 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Intentional overdose'. | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | AMITRIPTYLINE, CHLORPROMAZINE, DULOXETINE, MAGNESIUM OXIDE, NITRAZEPAM, RABEPRAZOLE SODIUM, TRAZODONE HYDROCHLORIDE, ZOLPIDEM TARTRATE | DrugsGivenReaction | CC BY-NC-ND | 33598249 | 18,711,570 | 2021-02 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Liver function test increased'. | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | AMITRIPTYLINE, CHLORPROMAZINE, DULOXETINE, MAGNESIUM OXIDE, NITRAZEPAM, RABEPRAZOLE SODIUM, TRAZODONE HYDROCHLORIDE, ZOLPIDEM TARTRATE | DrugsGivenReaction | CC BY-NC-ND | 33598249 | 18,711,570 | 2021-02 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Renal function test abnormal'. | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | AMITRIPTYLINE, CHLORPROMAZINE, DULOXETINE, MAGNESIUM OXIDE, NITRAZEPAM, RABEPRAZOLE SODIUM, TRAZODONE HYDROCHLORIDE, ZOLPIDEM TARTRATE | DrugsGivenReaction | CC BY-NC-ND | 33598249 | 18,711,570 | 2021-02 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Rhythm idioventricular'. | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | AMITRIPTYLINE, CHLORPROMAZINE, DULOXETINE, MAGNESIUM OXIDE, NITRAZEPAM, RABEPRAZOLE SODIUM, TRAZODONE HYDROCHLORIDE, ZOLPIDEM TARTRATE | DrugsGivenReaction | CC BY-NC-ND | 33598249 | 18,711,570 | 2021-02 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Shock symptom'. | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | AMITRIPTYLINE, CHLORPROMAZINE, DULOXETINE, MAGNESIUM OXIDE, NITRAZEPAM, RABEPRAZOLE, TRAZODONE HYDROCHLORIDE, ZOLPIDEM TARTRATE | DrugsGivenReaction | CC BY-NC-ND | 33598249 | 18,738,472 | 2021-02 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Suicidal ideation'. | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | AMITRIPTYLINE HYDROCHLORIDE, CHLORPROMAZINE, DULOXETINE, MAGNESIUM OXIDE, NITRAZEPAM, NORTRIPTYLINE, RABEPRAZOLE, TRAZODONE HYDROCHLORIDE, ZOLPIDEM TARTRATE | DrugsGivenReaction | CC BY-NC-ND | 33598249 | 18,745,418 | 2021-02 |
Give an alphabetized list of all active substances of drugs taken by the patient who experienced 'Ventricular extrasystoles'. | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | AMITRIPTYLINE, CHLORPROMAZINE, DULOXETINE, MAGNESIUM OXIDE, NITRAZEPAM, RABEPRAZOLE SODIUM, TRAZODONE HYDROCHLORIDE, ZOLPIDEM TARTRATE | DrugsGivenReaction | CC BY-NC-ND | 33598249 | 18,711,570 | 2021-02 |
What is the weight of the patient? | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | 70.1 kg. | Weight | CC BY-NC-ND | 33598249 | 18,745,418 | 2021-02 |
What was the administration route of drug 'AMITRIPTYLINE'? | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | Oral | DrugAdministrationRoute | CC BY-NC-ND | 33598249 | 18,711,570 | 2021-02 |
What was the administration route of drug 'CHLORPROMAZINE'? | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | Oral | DrugAdministrationRoute | CC BY-NC-ND | 33598249 | 18,711,570 | 2021-02 |
What was the administration route of drug 'DULOXETINE'? | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | Oral | DrugAdministrationRoute | CC BY-NC-ND | 33598249 | 18,711,570 | 2021-02 |
What was the administration route of drug 'MAGNESIUM OXIDE'? | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | Oral | DrugAdministrationRoute | CC BY-NC-ND | 33598249 | 18,711,570 | 2021-02 |
What was the administration route of drug 'NITRAZEPAM'? | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | Oral | DrugAdministrationRoute | CC BY-NC-ND | 33598249 | 18,711,570 | 2021-02 |
What was the administration route of drug 'RABEPRAZOLE SODIUM'? | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | Oral | DrugAdministrationRoute | CC BY-NC-ND | 33598249 | 18,711,570 | 2021-02 |
What was the administration route of drug 'TRAZODONE HYDROCHLORIDE'? | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | Oral | DrugAdministrationRoute | CC BY-NC-ND | 33598249 | 18,711,570 | 2021-02 |
What was the administration route of drug 'ZOLPIDEM TARTRATE'? | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | Oral | DrugAdministrationRoute | CC BY-NC-ND | 33598249 | 18,711,570 | 2021-02 |
What was the dosage of drug 'AMITRIPTYLINE HYDROCHLORIDE'? | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | 280 DF, UNKNOWN | DrugDosageText | CC BY-NC-ND | 33598249 | 18,745,418 | 2021-02 |
What was the dosage of drug 'AMITRIPTYLINE'? | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | STRENGTH 25 MG, RESCUE CREWS FOUND 280 EMPTY PILLS OF 25 MG AMITRIPTYLINE. | DrugDosageText | CC BY-NC-ND | 33598249 | 18,711,570 | 2021-02 |
What was the dosage of drug 'CHLORPROMAZINE'? | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | UNK UNKNOWN, UNKNOWN | DrugDosageText | CC BY-NC-ND | 33598249 | 18,745,418 | 2021-02 |
What was the dosage of drug 'DULOXETINE'? | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | UNK UNKNOWN, UNKNOWN | DrugDosageText | CC BY-NC-ND | 33598249 | 18,745,418 | 2021-02 |
What was the dosage of drug 'MAGNESIUM OXIDE'? | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | UNK UNKNOWN, UNKNOWN | DrugDosageText | CC BY-NC-ND | 33598249 | 18,745,418 | 2021-02 |
What was the dosage of drug 'NITRAZEPAM'? | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | UNK UNKNOWN, UNKNOWN | DrugDosageText | CC BY-NC-ND | 33598249 | 18,745,418 | 2021-02 |
What was the dosage of drug 'NORTRIPTYLINE'? | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | UNK UNKNOWN, UNKNOWN | DrugDosageText | CC BY-NC-ND | 33598249 | 18,745,418 | 2021-02 |
What was the dosage of drug 'RABEPRAZOLE'? | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | UNK UNKNOWN, UNKNOWN | DrugDosageText | CC BY-NC-ND | 33598249 | 18,745,418 | 2021-02 |
What was the dosage of drug 'TRAZODONE HYDROCHLORIDE'? | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | UNK UNKNOWN, UNKNOWN | DrugDosageText | CC BY-NC-ND | 33598249 | 18,745,418 | 2021-02 |
What was the dosage of drug 'ZOLPIDEM TARTRATE'? | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | UNK UNKNOWN, UNKNOWN | DrugDosageText | CC BY-NC-ND | 33598249 | 18,745,418 | 2021-02 |
What was the outcome of reaction 'Bundle branch block right'? | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | Recovering | ReactionOutcome | CC BY-NC-ND | 33598249 | 18,711,570 | 2021-02 |
What was the outcome of reaction 'Cardio-respiratory arrest'? | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | Recovered | ReactionOutcome | CC BY-NC-ND | 33598249 | 18,745,418 | 2021-02 |
What was the outcome of reaction 'Defect conduction intraventricular'? | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | Recovering | ReactionOutcome | CC BY-NC-ND | 33598249 | 18,711,570 | 2021-02 |
What was the outcome of reaction 'Drug level increased'? | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | Recovering | ReactionOutcome | CC BY-NC-ND | 33598249 | 18,711,570 | 2021-02 |
What was the outcome of reaction 'Electrocardiogram T wave abnormal'? | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | Recovering | ReactionOutcome | CC BY-NC-ND | 33598249 | 18,711,570 | 2021-02 |
What was the outcome of reaction 'Haemodynamic instability'? | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | Recovering | ReactionOutcome | CC BY-NC-ND | 33598249 | 18,711,570 | 2021-02 |
What was the outcome of reaction 'Liver function test increased'? | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | Recovering | ReactionOutcome | CC BY-NC-ND | 33598249 | 18,711,570 | 2021-02 |
What was the outcome of reaction 'Renal function test abnormal'? | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | Recovering | ReactionOutcome | CC BY-NC-ND | 33598249 | 18,711,570 | 2021-02 |
What was the outcome of reaction 'Rhythm idioventricular'? | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | Recovering | ReactionOutcome | CC BY-NC-ND | 33598249 | 18,711,570 | 2021-02 |
What was the outcome of reaction 'Suicidal ideation'? | Plasma concentration of amitriptyline and metabolites after resuscitation from cardiopulmonary arrest following an overdose: A case report.
It may need to pay attention to the sustention of moderate cardiotoxicity and delayed elevation of plasma 10-hydroxynortriptyline level in severe amitriptyline overdose case.
1 INTRODUCTION
In an amitriptyline (AT) overdose case with brief cardiopulmonary arrest, the concentrations of AT and metabolites were measured. Consequently, physiological damage by ischemic‐reperfusion was not suggested to affect disposition of AT and nortriptyline (NT). However, the level of 10‐hydroxynortriptyline (10‐OH‐NT) was belatedly elevated, suggested as a cause of sustention of moderate cardiotoxicity.
AT overdose is frequent among tricyclic antidepressants (TCA)‐related toxicities and more fatal than overdose of other antidepressants, including selective serotonin reuptake inhibitors and noradrenergic and specific serotonin antidepressants.
1
Followings are the pharmacological properties that are reported as toxic effects of TCA: (a) inhibition of norepinephrine reuptake at nerve terminals; (b) direct blockade of α‐adrenergic receptors; (c) membrane stabilizing or quinine‐like effect on the myocardium; and (d) anticholinergic action.
2
Based on these effects, a patient with AT toxicity manifests several clinical complications including sinus tachycardia, prolonged QRS/QTc duration, vasodilation, hypotension, cardiogenic shock, ventricular fibrillation/tachycardia, coma, drowsiness, delirium, respiratory depression, and others.
2
These symptoms are more likely to occur at total blood concentrations of AT and NT >1000 ng/mL or total concentrations of NT and 10‐OH‐NT >300 ng/mL.
3
,
4
,
5
Regarding TCA, plasma concentrations >450 ng/mL tend to show cognitive or behavioral toxicities
6
and >2000 or 3000 ng/mL are fatal,
7
,
8
though therapeutic range is from 50 to 300 ng/mL.
6
Therefore, blood concentrations of AT, NT, and 10‐OH‐NT may provide useful information in assessing the level of toxicity and in predicting subsequent clinical outcomes in patients following an AT overdose. In drug overdose cases, information regarding ingested dose is often sparse and thus a poor predictor of clinical outcome.
2
Therefore, blood concentration data for the drug in question is extremely valuable.
Cardiopulmonary arrest (CPA) sometimes occurs in AT overdose cases. The resuscitation following CPA is often accompanied with ischemic‐reperfusion, during which reactive oxygen species are derived, inflicting injury to living grafts or cells.
9
Injury to hepatocytes or intestinal tissues can result in dysfunctional or unusual protein expression of cytochrome P450 (CYP) 2C19 or CYP2D6, alteration of enterohepatic recirculation, or a dysfunctional elimination process, which can in turn affect the metabolism, distribution, and elimination of AT, NT, and 10‐OH‐NT, resulting in alterations in the overall disposition and pharmacokinetic profile.
10
These alterations in drug disposition may also change the occurrence of both, the onset and duration of clinical features of toxicity in patients. Nonetheless, the effect of CPA on subsequent drug disposition and its clinical effects are rarely investigated and reported, despite numerous reports of AT overdose.
3
,
11
,
12
In this report, we measured the blood levels of AT, NT, and 10‐OH‐NT and evaluated the influence of CPA on the pharmacokinetics of these compounds in a patient who had taken a severe overdose of AT. We also evaluated the decontamination effect of activated charcoal (AC) to evaluate its effects in cases such as this with CPA.
2 CASE DESCRIPTION
A 55‐year‐old man (weight, 70.1 kg) with impaired awareness was found by his roommate in a warm room in his own house and transferred to our hospital. Rescue crews found 280 empty pills of 25 mg AT [corresponding to a total amount of 7 g (99.8 mg/kg)] beside him. A urine‐screening test using a commercial test kit (Triage® DOA, Sysmex Corp., Kobe, Japan) indicated TCA presence. Additionally, concurrent use of chlorpromazine, duloxetine, rabeprazole, trazodone, zolpidem tartrate, nitrazepam, and magnesium oxide was suspected. However, there was no evidence of excessive ingestion of these drugs, and the patient did not present with toxicity relevant to these drugs. Considering these observations and following comprehensive clinical assessments, we suspected that he likely ingested a massive dose of AT. The elapsed time of ingestion of the overdose could not be exactly estimated because the patient's awareness was impaired and there was no information available about anyone who had contacted him during the 12‐hours period before he was found. The roommate told us that she found him in a coma 12 hours after her last contact with him. Unfortunately, we did not have any other means of knowing if he had any pre‐existing disorders, other than depression. His comorbidity and past medical history were not available. On admission, the Glasgow Coma Scale (GCS) score was 3 (E1; V1; M1); systolic/diastolic blood pressure, 101/62 mm Hg; oxygen saturation, 94%; body temperature, 38.7°C; heart rate, 120/min; respiration rate, 20/min; blood pH, 7.022; QTc interval, 610 ms; and QRS interval, 270 ms (Figure 1A). Other relevant clinical data are included in the supporting information (Table S1). Atrial fibrillation (AF) with ventricular aberration or premature ventricular contraction, complete right bundle branch block, escape rhythm, and abnormal T wave were observed on the electrocardiogram (ECG; Figure 1A).
Figure 1 Representative electrocardiograms obtained during hospitalization. (A) On admission; (B) 10 h later; (C) 12 h later; (D) 15 h later; (E) 60 h later; (F) 112 h later
Immediately after admission, a CPA following 10 seconds of clonus occurred, and 20 minutes later, he was resuscitated; however, his awareness remained impaired. The patient was thereafter intubated and treated for shock‐like hemodynamic status with the initiation of an infusion of 0.30 mg/h (approximately 0.07 μg/kg/min) of adrenaline, 0.60 mg/h of (approximately 0.14 μg/kg/min) of noradrenaline, and 15 mg/h (approximately 3.60 μg/kg/min) of dopamine. Totally, 1.1 mg/h (approximately 0.26 μg/kg/min) of adrenaline, 0.70 mg/h (approximately 0.17 μg/kg/min) of noradrenaline, and 30 mg/h (approximately 7.13 μg/kg/min) of dopamine were administered during the subsequent 11 hours period. The blood was alkalized by administering sodium bicarbonate (500 and 120 mL/d on the first and second hospital day, respectively). Additionally, 50 g of AC was repeatedly administered (five times in total) at 3, 14, 19, 27, and 35 hours after admission (Figure 1). Gastrointestinal decontamination other than AC administration was not conducted.
Approximately 15 hours after resuscitation from CPA, the drastically elongated QTc and QRS intervals were shortened to their near‐normal values (441 and 128 ms, respectively, Figure 1D), although their values remained near the upper limits of normal for several more hours. AF with ventricular aberration or premature ventricular contraction, nonspecific intraventricular conduction delay and temporal escape rhythm were also observed at some time point (Figure 1C,D). Clinical laboratory parameters of liver and renal function were elevated during the initial 10 hours; subsequently, the parameter values reduced (Figure S1). However, these values persistently remained at higher than normal levels, although there were no significant toxic effects observed. Approximately 60 hours after admission, the patient's awareness improved [GCS score of 14 (E4; V4; M6)]. The ECG showed the indications of sinus rhythm, although nonspecific intraventricular conduction delay was sometimes observed (Figure 1E). The creatinine clearance was almost normalized (approximately 80 mL/min) at 72 hours after admission. Nutritional provision was started with a small amount of enteral nutrition sometime after the end of catecholamine administration. During the 7th and 9th days of hospitalization, delirium was observed; however, the patient did not show any signed of orientation. ECG at 112 hours after admission indicated that sinus rhythm was maintained (Figure 1F), although premature ventricular contraction was infrequently observed. On the 10th day, he was transferred to the medical psychiatry unit. In the unit, the patient reported transient suicidal feelings that gradually dissipated. On the 23rd day, he was discharged.
The plasma concentrations of AT at 14, 45, and 131 hours after admission and those of NT and its hydroxyl metabolites at 45 and 131 hours after admission were measured by liquid chromatography‐tandem mass spectrometry (Figure 2). Both AT and NT concentrations decreased over time after resuscitation following CPA. In contrast, the concentration of 10‐OH‐NT increased during this period. The elimination half‐lives of AT and NT estimated after 45 hours of admission were approximately 35 and 89 hours, respectively.
Figure 2 Time profiles of plasma levels of AT and its metabolites, QTc, QRS, heart rate, creatinine clearance, and blood pressure during the monitoring interval. AT, amitriptyline; NT, nortriptyline; 10‐OH‐NT, 10‐hydroxynortriptyline; HR, heart rate; Ccr, creatinine clearance; NIBP, noninvasive blood pressure. The black‐filled arrows under the x‐axis in top panel indicate AC administration at 3, 14, 19, 27, and 35 h after admission. The shaded area means the one above toxic concentration
3 DISCUSSION
In this overdose case with AT, AC was repeatedly administered (five times in total), following resuscitation 20 minutes after CPA. The subsequent concentration levels of AT were, then, reduced compared to the initial concentration. The repeated AC administration can complex with the TCA (doxepin) released following gradual disaggregation of the complex of doxepin with initially administered AC, thus suppressing the subsequent absorption of parent drug and metabolites secreted into the gastrointestinal tract.
13
This study also reported the repeated dosing effect of AC was observed despite the delay of three or more hours in AC administration following doxepin dosing. Although the delay between AT ingestion and AC initiation could not be confirmed in our case report, the initial measured concentration of AT (ie, 1031 ng/mL) was reduced after the repetitive AC administration. Thus, repeated AC administration and/or the clearance of the drug from the patient's system may have contributed to the reduction of AT concentrations. However, the reduction in AT concentrations between 14 and 45 hours after admission was extremely low (only 55 ng/mL). Additionally, the elimination half‐life estimated from this reduction was approximately 390 hours, while another study on AT overdose patients reported a half‐life of <10 hours when patients were treated with repeated AC administration.
14
There are several possible reasons for the discrepancy and slight reduction of AT concentration, including interruption of patient's individual drug clearance because of cellular damage by ischemic‐reperfusion; saturation of metabolic enzymes by the excessive amount of AT ingested; poor metabolism due to genetic polymorphism in CYP2C19 and CYP2D6, or anticholinergic suppression of elimination by AT and NT. Although it is unclear how the repeated AC administration and patient's clearance system effectively functioned, they may have at least partly contributed to the reduction of AT concentration. Thus, the further increase of the subsequent AT concentration would have not been shown. The repeated AC administration may have functioned as suppressing the elevation of peak concentration of AT, rather than accelerating the elimination process. Repetitive AC may have suppressed the potential increase in AT levels that could occur if the complex formed with the initial dose of AC is disaggregated. Additionally, AC may have also prevented the subsequent absorption/reabsorption of AT and its metabolites following their secretions into the gastrointestinal tract, as AT has been known to undergo enterohepatic circulation.
14
While further studies are needed, we suggest that repeated AC administration is a possible optional treatment in severe cases of AT overdose, even if CPA is presented and several hours have elapsed from the time of overdose.
A significant shortening of QRS interval from 124 to 110 ms was observed after approximately 72 hours from admission. Simultaneously, the total concentrations of AT and NT were estimated to be below the toxic range (total of 1000 ng/mL), suggesting the importance of monitoring the total concentrations of AT and NT in AT overdose cases. However, despite the reduction in total concentrations of AT and NT to less than toxic levels at 110 hours after admission, QTc and QRS intervals remained relatively high values. The QRS interval, particularly, did not fall below the upper limit of normal (100 ms). In parallel with these cardiac effects, elevation of the concentration of 10‐OH‐NT was observed, although the creatinine clearance recovered to a normal level, suggesting that 10‐OH‐NT elimination was not impaired.
15
A previous study demonstrates the cardiotoxic potential of 10‐OH‐NT, although it was less toxic than NT.
16
Additionally, the total concentrations of 10‐E‐hydroxynortriptyline and NT of > 300 ng/mL were associated with QRS prolongation.
4
Therefore, we speculated that the high levels of 10‐OH‐NT may have contributed to this sustained prolonged QTc and QRS intervals. Several factors may have contributed to the delayed elevation of 10‐OH‐NT concentration. One of these may be the genetic polymorphism in CYP2C19 and CYP2D6. Then, we estimated the phenotype of our patient, based on previous reports by Mifsud et al and Franssen et al, although the cases reported by Mifsud et al were treated with a normal dose (10 mg daily) of AT.
12
,
17
In a case with a “normal phenotype” of both CYP2C19 and CYP2D6 (ie, case example III in the report by Mifsud et al), the metabolite‐to‐parent rate (MPR; ie, NT/AT) was reported as 0.4. On the other hand, the case with concomitant use of only CYP2C19 inhibitor (ie, omeprazole) but with normal phenotype in both CYP2C19 and CYP2D6 showed the MPR as 0.2. However, another case with “normal phenotype'” in CYP2C19 and CYP2D6 (ie, case example V in the report of Mifsud et al) and concomitant use of inhibitors for CYP2C19 and CYP2D6 (omeprazole and paroxetine, respectively) indicated an MPR as 0.7, which was close to the MPR of our case (calculated as approximately 0.6). AT itself has an inhibitory effect on both CYP2C19 and CYP2D6, although CYP2D6 inhibition is reported to be clinically insignificant at a normal dose.
6
However, in the current case, a massive dose of AT was ingested, suggesting that the inhibitory effect on CYP2D6 may have somewhat functioned. In addition, considering the similarity of MPR between ours (approximately 0.6) and case's mentioned above (0.7), in which the patient had normal phenotype but was concomitantly administered with inhibitors for CYP2C19 and CYP2D6, our patient may also have possessed a normal phenotype and hers CYP2C19 and CYP2D6 may have somewhat inhibited due to the extremely high dose of AT. We also considered the possibility that the patient was a rapid metabolizer of CYP2C19, and this rapid activity was suppressed to normal levels due to the inhibitory effect of the large amount of AT ingested. However, as shown in case example I reported by Mifsud et al, the rapid metabolizer of CYP2C19 indicated a high MPR of 2.0, despite receiving a concurrent CYP2C19 inhibitor (ie, omeprazole). Therefore, with regard to CYP2C19, our patient (with MPR of 0.6) was not considered to correspond to a rapid metabolizer. Regarding CYP2D6, we focused on the ratio of 10‐OH‐NT/NT. In our case study, the ratio was calculated to be approximately 1.4 at 45 hours after admission, which was relatively close to the estimated one in case example III (approximately 3.5) and example IV (approximately 2.0) in the report by Mifsud et al
17
In these two cases, the CYP2D6 and CYP2C19 phenotypes both identified as normal, although the case example IV concomitantly received a CYP2C19 inhibitor (omeprazole). Another case (case example V in the report by Mifsud et al) with concurrent use of CYP2D6 inhibitor (paroxetine) indicated a lower 10‐OH‐NT/NT ratio (approximately 0.3), while a different case with ultra‐rapid metabolism in CYP2D6 indicated a higher 10‐OH‐NT/NT ratio (approximately 8.0). These estimates were far from the 10‐OH‐NT/NT ratio in our patient. Taken together, our analysis indicates that the patient may have possessed a “normal phenotype” in CYP2D6 as well. Levels of AT and its metabolite (ie, NT) following AT overdose may inhibit both CYP2C19 and CYP2D6 in the initial phase. Then, as concentrations of AT and NT decreased with time, the inhibitory effects on CYP may have gradually weakened, while concentrations of 10‐OH‐NT may have been elevated in a delayed manner. Besides, other factors including saturation of metabolic enzymes, dysfunction of metabolic enzymes or elimination systems by ischemic‐reperfusion damage, insufficient alkalization of urine or anticholinergic effect of AT, might have contributed in a complex manner to the delayed elevation of 10‐OH‐NT levels. Although some potential drugs that inhibit CYP2C19 (ie, duloxetine and chlorpromazine) seemed to have been prescribed, their inhibitory effects can be ignored because there was no evidence of their massive ingestion, and the patient did not present relevant toxic effects. The delayed elevation of 10‐OH‐NT levels over 100 hours after admission was also indicated in a fatal case with AT overdose by Franssen et al,
12
supporting our finding. Monitoring the level of 10‐OH‐NT may also be favorable, if possible, in severe AT overdose cases because it has toxic effects on cardiac function and its concentration can be elevated in later.
In this case report, the elimination half‐life of AT and NT was calculated to be approximately 34 and 88 hours, respectively, from the levels measured at 45 and 131 hours. A previous study reported that the elimination half‐lives of nine cases with AT overdose ranged from 15 to 43 hours in patients treated with gastric lavage and 50 g of AC.
18
Even at normal doses, the elimination half‐life of AT is reported to range from 10 to 46 hours.
6
The elimination half‐life of AT (ie, 45 hours) in our patient was approximately same with the ones reported. For NT in the case with normal dose, elimination half‐life has been reported to range from 13 to 90 hours, although shorter in ultra‐metabolizers (13‐35 hours).
6
A case of NT overdose reported an elimination half‐life of 50 hours based on levels measured from 50 to 110 hours after admission,
12
whereas another case showed a half‐life of 184 hours and sustained elevated levels of NT (468 ng/mL) until 6 days after admission.
19
Although both previous overdose cases were not treated with AC, the elimination half‐life of NT (ie, 88 hours) in our patient treated with AC was compatible with the reported ones. The elimination half‐life of 10‐OH‐NT could not be calculated in this study because its elimination phase was not observed during our evaluation period, although its half‐life has been reported to be approximately 8‐10 hours.
20
Thus, taken these, the damage, possibly induced by ischemic‐reperfusion, would have not induced any meaningful anomalous effect on the disposition of AT or NT, despite physiological damage was suspected by the sustained elevation of lactate dehydrogenase (Figure S1).
21
,
22
As long as CPA is recovered within approximately 20 minutes, as seen in our study, and appropriate treatment is provided, a patient with AT/NT overdose can survive and be managed clinically.
4 CONCLUSION
It would be desired to monitor not only the concentration of AT and NT but also those of its hydroxyl metabolites, especially in severe AT/NT overdose because 10‐OH‐NT has toxic effects on cardiac function and its blood concentration can increase behind the reduction of AT and NT concentrations. Although further studies are needed, repeated AC administration may be a possible optional treatment in severe cases of AT overdose, even after several hours have elapsed from overdose. If the patient experiences CPA, as long as appropriate treatment is provided, AT/NT‐poisoned patient can survivable and may be clinically managed like a typical AT/NT overdose case. These experiences would be beneficial for other workers if they encounter a similar AT overdose case with CPA.
CONFLICT OF INTEREST
No conflicts of interest have been declared.
AUTHORS’ CONTRIBUTION
MA, RT, and ST measured the concentrations of amitriptyline, nortriptyline, and 10‐hydroxynortriptyline and evaluated their relationship with the patient's clinical outcome. MY, KS, and MS provided technical support in quantification by a validated method. KA provided several useful medical considerations as an emergency physician. RT and SN provided several useful pharmaceutical considerations as pharmacists engaged in the treatment of this patient. SF, NS, and TH provided helpful guidance and pharmacokinetic expertise in the area of pharmaceutical sciences. All authors have read and approved the final manuscript.
ETHICAL APPROVAL
Through the provision of a chance to opt‐out, the study and reporting were approved by the ethics review board of our hospital (Approval #: k181033).
Supporting information
Supinfo
Click here for additional data file.
ACKNOWLEDGMENTS
We would like to thank Editage (www.editage.com) for English language editing. Published with written consent of the patient.
DATA AVAILABILITY STATEMENT
The data that support the findings of this study are available from the corresponding author upon reasonable request. | Recovered | ReactionOutcome | CC BY-NC-ND | 33598249 | 18,745,418 | 2021-02 |
Subsets and Splits
No community queries yet
The top public SQL queries from the community will appear here once available.