jcm2

Med-MDPI/jcm_2/jcm-09-06-01731.txt

@@ -0,0 +1 @@
+To date, MRI studies focused on brain sexual dimorphism have not explored the presence of specific neural patterns in gender dysphoria (GD) using gender discrimination tasks. Considering the central role of body image in GD, the present study aims to evaluate brain activation patterns with 3T-scanner functional MRI (fMRI) during gender face discrimination task in a sample of 20 hormone-naïve transgender and 20 cisgender individuals. Additionally, participants were asked to complete psychometric measures. The between-group analysis of average blood oxygenation level dependent (BOLD) activations of female vs. male face contrast showed a significant positive cluster in the bilateral precuneus in transmen when compared to the ciswomen. In addition, the transwomen group compared to the cismen showed higher activations also in the precuneus, as well as in the posterior cingulate gyrus, the angular gyrus and the lateral occipital cortices. Moreover, the activation of precuneus, angular gyrus, lateral occipital cortices and posterior cingulate gyrus was significantly associated with higher levels of body uneasiness. These results show for the first time the existence of a possible specific GD-neural pattern. However, it remains unclear if the differences in brain phenotype of transgender people may be the result of a sex-atypical neural development or of a lifelong experience of gender non-conformity.Gender incongruence (GI) is defined by a marked and persistent discrepancy between an individual’s experienced gender and the assigned sex [1]. When this condition is associated with a clinically significant distress or impairment in social, occupational or other important areas of functioning, it is referred to as gender dysphoria (GD) [2]. Individuals whose gender identity does not completely and/or permanently match their sex characteristics may describe themselves as trans or transgender. On the other hand, individuals whose gender identity does match their sex characteristics are referred to as cisgender.The potential impact of biologic and social/cultural factors in the etiology of GI/GD remains under debate. However, biology seems to play a major role [3,4]. In the last years, the idea of the sexual dimorphic brain as the anatomic substrate of psychosexual development has been more widely pursued and research has focused on the influence and shaping role of genes and gonadal hormones on sexual differentiation of the brain [5]. In particular, gonadal steroids may influence the development of brain structures and circuits through an organizational effect during early development and an activational effect later in life [3,6].Brain structural and functional differences—resulting from the interaction of genes and sex hormones with the developing brain—are thought to be the developmental substratum of gender identity and factors interfering with this complex process may be involved in or at least favor, GI/GD [4].Morphologic and functional brain characteristics of transgender individuals have been investigated in vivo with magnetic resonance imaging (MRI), in order to substantiate this etiological hypothesis with a biologic background. Regarding brain morphology, several studies supported this hypothesis especially in transwomen, showing a pattern in line with gender identity with regards to cortical thickness, regional gray matter volumes and white matter microstructure [7,8,9,10,11]. On the other hand, findings seem to be more conflicting in transmen when evaluating morphologic brain characteristics. Particularly, some authors reported a gray matter pattern superposable to the assigned sex [7,12], while other studies found structural characteristics similar to those observed in cisgender males [13,14]. Additionally, in transmen, morphologic patterns different from both cisgender groups have been described in literature [8,15,16,17]. The discrepant results of the aforementioned studies could be partly attributed to several limitations including inhomogeneity of selected samples concerning sexual orientation, GD levels, age and previous hormonal treatments [18].Given the fact that cognitive abilities are often gender-specific, some studies evaluated differences in functional brain characteristics during task-based MRI studies. During a verbal fluency task, Soleman et al. [19] did not find significant differences in brain activation pattern between transgender adolescents and cisgender controls. Contrastingly, regarding visuospatial cognitive functioning, two studies reported atypical brain activation during a mental rotation task in transgender individuals, different from that observed in the cisgender control group [18,20]. In addition. differences in resting-state brain networks in transgender women compared to cisgender ones have been reported [21,22]. Moreover, combining both resting-state functional connectivity and behavioral data, gender identity in transgender and in cisgender persons has been recently examined, showing that machine learning algorithms could predict distinct brain connectivity patterns in cis and trans people [23].Further studies have focused on functional MRI connectivity evaluating differences between transgender and cisgender persons in cerebral networks involved in own body perception, given its importance in GD [14,15,24,25,26,27]. Several authors hypothesized that GD could be based on the disconnection of fronto-parietal networks involved in the processing of own body image. This may lead transgender individuals to be unable to incorporate typical body characteristics of their gender assigned at birth into their own body representation in the brain [28]. Particularly, a central role seems to be played by the default mode network, which represents a connectivity network involved in mind-wandering and self-referential thinking [29] and by the salience network, which could be involved in processing different stimuli from own body [30,31]. Some evidences suggest that differences in networks connectivity related to body perception and self-identification may represent a neurobiologic correlate of GD.To the best of our knowledge, no MRI study has explored the presence of possible specific neural activation pattern in transgender people by using a gender discrimination task. For this reason, considering the central role of body image in GD development, in the present study we evaluated brain activation patterns related to face gender discrimination in a sample of hormone-naïve transgender and cisgender individuals by using fMRI.Forty subjects, namely 20 cisgender controls (10 cismen, 10 ciswomen) and 20 transgender individuals (10 transmen, 10 transwomen) of similar age (mean ± SD age = 28.45 ± 5.17 and 29.55 ± 10.90 years, respectively for cisgender and for transgender individuals; t = 0.408 p = 0.68), took part in the experiment.The 20 trans individuals belonged to a consecutive series of subjects referring for the first time to the center for GI/GD at the University of Florence. They were enrolled in the present study if the following inclusion criteria were met:Age older than 18 years;Diagnosis of GD made by mental care providers experienced in GD during consultations and by assessing DSM 5 criteria [2,32].Age older than 18 years;Diagnosis of GD made by mental care providers experienced in GD during consultations and by assessing DSM 5 criteria [2,32].The exclusion criteria included:The use at any point in life of hormonal therapy.Gender affirming surgery performed.Illiteracy.Mental retardation.Disorder of sexual development (DSD).Severe and unstable psychiatric conditions (e.g., psychotic disorders, depressive disorder with suicidal ideation) assessed by mental health professionals experienced in GD during consultations and by assessing DSM 5 criteria [2]The use at any point in life of hormonal therapy.Gender affirming surgery performed.Illiteracy.Mental retardation.Disorder of sexual development (DSD).Severe and unstable psychiatric conditions (e.g., psychotic disorders, depressive disorder with suicidal ideation) assessed by mental health professionals experienced in GD during consultations and by assessing DSM 5 criteria [2]The cisgender group was enrolled by means of local advertisement at the University of Florence, when the following inclusion criteria were met: age older than 18 years, absence of GD or psychiatric disorders. The exclusion criteria included:The use in the previous six months of any hormonal treatment and of any psychiatric medication;Illiteracy;Mental retardation;DSD;Severe and unstable psychiatric conditions (e.g., psychotic disorders, depressive disorder with suicidal ideation);Pregnancy or current lactation.The use in the previous six months of any hormonal treatment and of any psychiatric medication;Illiteracy;Mental retardation;DSD;Severe and unstable psychiatric conditions (e.g., psychotic disorders, depressive disorder with suicidal ideation);Pregnancy or current lactation.All participants had normal or corrected-to-normal visual acuity and normal hearing, by self-report. Participants were screened to ensure that they satisfied MRI safety requirements and showed no structural brain abnormalities on T1 or FLAIR MRI sequences (see below) obtained before the fMRI task.All participants underwent a physical examination, with measurement of height, weight and body mass index (BMI).In addition. they were asked to complete the body uneasiness test (BUT) [33], the gender identity/gender dysphoria questionnaire for adolescents and adults [34] and the symptom checklist-90 revised (SCL-90-R) [35].The BUT is a self-rating scale exploring different areas of body-related psychopathology, including dissatisfaction with the body and its weight (weight phobia), avoidance, compulsive control behavior (compulsive self-monitoring), experience of separation and strangeness from the body (depersonalization) and specific worries about certain body parts, characteristics or functions. The subjects were asked to rate 34 different body image experiences (BUT A) and 37 body parts (BUT B) on a six-point Likert scale (from 1 = never to 6 = always), indicating how often they happen to dislike each experience or each body part. Higher scores indicate greater body uneasiness. BUT scores were analyzed by considering the total score of the test (global severity index), the number of disliked body parts (positive symptoms total, PST) and the mean intensity of dislike of all disliked body parts (positive symptom distress index, PSDI) [33].The gender identity/gender dysphoria questionnaire for adolescents and adults (GIDYQ-AA) is a 27-item questionnaire evaluating GD (17). Each item is rated on a 5-point response scale, considering the past 12 months as time frame. The response options are: always, often, sometimes, rarely or never, coded 1 to 5, respectively. Lower scores are associated with higher GD, being a score of three suggested as critical threshold for GD diagnosis. Internal coherence was satisfactory for the Italian validated version (α value of about 0.97) [34].Furthermore, levels of psychopathologic distress were specifically investigated by means of the Italian version of the symptom checklist (SCL-90-R) [35], which was answered for the week preceding the clinical assessment. The 90 items of the questionnaire are rated on a five-point Likert scale (from 0 to 4) and are grouped together into nine domains (somatization, obsessive–compulsive thoughts, interpersonal sensitivity, depression, anxiety, hostility, phobic anxiety, paranoid conceptions and psychotic behavior). In this study, we utilized the general severity index (SCL-GSI), indicating the overall psychological distress.To evaluate sexual orientation dimensionally, a visual analog scale (VAS) was used, rating 0 when sexual attraction was exclusively towards people of the same perceived gender (i.e., towards men in cis- and transmen and towards women for cis- and transwomen) to 10, when sexual attraction was exclusively towards the opposite gender from the perceived one (i.e., towards women for cis- and transmen and towards men for cis- and transwomen) [36].Written informed consent was obtained from each participant and was approved by the local ethical committee (2013/0016117) in agreement with the 1964 Helsinki declaration and its later amendments or comparable ethical standards.The intense body-related distress experienced by some transgender people [37,38,39] may result in focusing on gender-specific body characteristics. Body sexual dimorphic parts may represent indeed for some transgender people the inner source of their sufferance and frequently of their low social acceptance. Taking this in mind, we speculated that the gender discrimination task, which requires focalizing on gender dimorphic facial features, may activate a more intense emotional reaction related to the painful remind of one’s gender incongruence.Stimuli were administered by means of psycho-toolbox 3 (20) and custom Matlab code. A desktop computer was employed to display the visual stimuli. Participants viewed stimuli on a MRI-compatible display system (SensaVue fMRI, Invivo Corporation, Gainesville, FL, USA) with a mirror attached to the head coil. The experiment was composed by two sessions: a training session and a task session. The training session consisted in a block of five trials, to ensure understanding of the instructions. The task session was composed by 40 trials. The protocol was a Gender Face Detection task (hereafter GFD). Each GFD trial started with a cross of fixation positioned in the center of the screen for 3 s and was followed by a face stimulus (50% of times male; 50% of times female) for 3 s (Figure 1).Participants had to observe the face displayed in the monitor and discriminate its gender by pressing one of two possible buttons associated to the face gender perceived (male or female). At the end of each trial there was a rest phase of six seconds where a fixation cross was located in the center of the screen. To avoid any possible effect of learning or cognitive strategies, stimuli were equally in number and randomly presented. The models of faces employed in this experiment were selected from a validated database (Karolinska database) [40]. The faces were transformed in black and white, equated in luminance and the hair contour was removed with Photoshop. Two male and two female original faces with neutral expression were chosen. We created for each face the respective face of the opposite gender by means of FaceApp to obtain 4 male and 4 female faces matched for identity for a total of 8 different stimuli. FaceApp is a software developed by a Russian company, Wireless Lab, which uses GANs networks to generate highly realistic transformations of faces’ pictures. The distinctive features which make the identity unique remain unchanged, whereas the algorithm allows to transform a face to make it smile, look younger, older or change gender [41,42,43].MRI acquisitions were performed on a 3T scanner (Ingenia, Philips Healthcare, Best, The Netherlands) equipped with Omega HP gradients with maximum amplitude of 45 mT/m and slew rate of 200 T/m/s for each axis. All subjects underwent 3D T1-weighted imaging and fMRI, using a 32-phased-array-element head coil.T1-weighted MR images were acquired with a sagittal high-resolution 3D sequence (repetition time [TR] = 8 ms, echo time [TE] = 3.7 ms, inversion time [TI] = 925.6 ms, flip angle [FA] = 8°, slice thickness = 1 mm, field of view [FOV] = 240 mm × 240 mm, number of slices = 191, matrix size = 352 × 352).T2-weighted 3-dimensional fluid-attenuated inversion recovery (FLAIR)–volume isotropic turbo spin–echo acquisition images were acquired on the coronal plane (repetition time [TR] = 8000 ms, echo time [TE] = 355 ms, inversion time [TI] = 2400 ms, variable [10°–180°] flip angle, echo–train length = 110, slice thickness = 1 mm, field of view = 256 × 256 mm, matrix = 232 × 232, 155–175 sections, number of excitations = 1, sensitivity encoding factor = 3.0, fat suppression with spectral selection attenuated inversion recovery; acquisition time 5 min and 28 s).For the fMRI experiment we employed a T2*-weighted echo-planar imaging (EPI) sequence (TR/TE = 3000/35 ms, FA = 90°, slice thickness = 3.5 mm, FOV = 240 mm × 240 mm, number of slices = 42, matrix size = 240 × 240). One hundred and sixty five scans were acquired, for a total acquisition time of about 8 min, from which the first 5 scans were discarded.Volumes acquired with fMRI were analyzed using the FMRIB Software Library (www.fmrib.ox.ac.uk/fsl). Canonical preprocessing was applied (first 5 time points removed, slice-time correction with custom timings, motion correction and intensity normalization). As filtering steps, we adopted the following: temporal high-pass with cutoff at 50 s; spatial smoothing using a 4 mm full width half-maximum Gaussian kernel. Co-registration of fMRI images to the individual high-resolution T1-weighted image was performed using a 6-degree of freedom registration. The individual high resolution T1-weighted images were co-registered to the standard space Montreal Neurological Institute 152 (MNI152) brain with an affine transformation (12 degree of freedom) followed by a nonlinear transformation. fMRI images were co-registered to the MNI152 standard space using the transformation previously computed when co-registering the individual high-resolution T1-weighted images to the MNI152 standard space.Time points in the fMRI data set that were affected by large motion, namely displacement >1.5 mm of the absolute mean displacement, were identified from motion correction parameters (motion correction FMRIB’s linear image registration tool) and accounted for in a confound matrix at the subject-level analysis. Each stimulus delta functions sequence was convolved with a double gamma hemodynamic response function, whereas the temporal derivative was included in the model and temporal filtering applied.To explore activity related to the gender face perception, a general linear model contrast was set at subject-level analysis to analyze volumes correspondent to the time interval of face presentation where the two explanatory variables (EV) were set as volumes of female presented face and male presented face.Each model EV was convolved with a double gamma hemodynamic response function, whereas temporal derivatives were included, and temporal filtering applied. This contrast was assessed for within and between groups’ analyses.To establish between-group differences of the five comparisons of major interest (cismen vs. ciswoman, transmen vs. cismen, transwomen vs. ciswomen, cismen vs. transwomen and ciswomen vs. transmen), we used an unpaired t-test with a mixed effects model taking into account Bonferroni correction of p-value for the considered multiple comparisons.Because the experimental design involved randomized intervals stimuli, we reduced autocorrelation in the data applying voxel-wise pre-whitening. To establish between-group differences, we used an unpaired t-test with a mixed effects model. All group analyses were performed in the MNI152 standard space T1-weighted template. For all statistical analyses, the resulting Z (Gaussianized T/F) statistic images were thresholded using clusters determined by Z >2.3 and a (corrected) cluster significance threshold of p < 0.01. To anatomically map the significant clusters in the resulting Z statistic images with labels of maximum probability, we used the Harvard–Oxford cortical and subcortical structural probabilistic atlases [44,45]. Moreover, to localize the occipital face area and Face Fusiform Area we used coordinates defined in previous studies [46,47].Continuous variables were reported as mean ± standard deviation. The independent sample t-test was used to compare continuous variables. Univariate analysis of variance (ANOVA) was used to compare the continuous variables among groups, entering age and BMI as a covariate, when appropriate.Post hoc paired contrasts with Tukey’s B tests were performed for the pairwise comparison among the groups. Pearson’s correlation was used to evaluate the associations between different variables within each group. Bonferroni correction was applied for multiple comparisons. Differences between groups were evaluated in multivariate models (adjusting for BMI and age) by means of analysis of covariance (ANCOVA). Finally, linear and logistic regression analyses were used for multivariate analysis (adjusting for BMI and age) whenever appropriate. All analyses were performed using SPSS version 25 (SPSS, Inc., Chicago, IL, USA).Table 1 reports the socio-demographic and clinical variables of groups and their differences in an age-adjusted model. transwomen showed significantly lower BMI than other groups (p = 0.035). No differences were found among groups in terms of sexual orientation (p = 0.155). In addition. no differences between transmen and transwomen were found in terms of GD intensity and onset, according the GIDYQ-AA and the clinical interview (all p > 0.05).The within-group analysis of differential BOLD activations of female vs. male face contrast did not reveal any significant cluster of activation. However, the between-group analysis of differential BOLD activations of the female vs. male face contrast showed several statistically significant clusters. One cluster of positive statistic sign (Z > 2.3, p < 0.01) was observed in the bilateral of the precuneus in the transmen group when compared with the ciswomen group (Figure 2 and Table 2).Four clusters of positive statistic sign (Z > 2.3, p < 0.01) were observed in the transwomen group when compared with the cismen group (Figure 3 and Table 3).The first cluster was located in the right lateral occipital cortex (including the occipital face area—OFA) and the right angular gyrus (r-AG). The second cluster comprised the bilateral posterior division of the cingulate gyrus (PCG) and the precuneus. The third cluster included the left lateral occipital cortex (including the OFA) and the left angular gyrus (l-AG). The fourth cluster was located bilaterally in the precuneus and extended in the left lateral occipital cortex.All other between group analyses did not demonstrate any significant cluster of differential activation. No significant results were reported for the negative sign statistics.In the entire sample, several measures of general psychopathology and body uneasiness, as well as anthropometric measures were considered. Differences among groups were adjusted for age and BMI, which may affect results (Table 4).When BUT was analyzed, trans people showed significantly higher body uneasiness levels compared with the cisgender groups (p < 0.0001; Figure 4A). Accordingly, scores of several BUT subscales (including “avoidance”, “body image concerns”, “depersonalization” and “positive symptoms distress index”) were significantly higher in trans groups compared to the cis ones (all p < 0.02, Figure 4B–E for BUT AV, BIC, DEP and PSDI, respectively). In addition. groups showed significant differences in terms of weight phobia subscales (BUT WP), with transwomen and transmen reporting higher scores compared to cismen (p < 0.02, Figure 4F). Finally, transwomen showed significant higher compulsive self-monitoring scores (BUT CSM) compared to cismen (p < 0.02, Figure 4G).Considering body uneasiness related to different body parts (BUT B, Table 5), transwomen showed significantly higher distress towards sexual dimorphic characteristics of the face (i.e., forehead, brows, nose, mouth, chin, moustache, beard), whereas transmen towards breast, when compared to all other groups (all p < 0.005). Other correlations are showed in Table 5.Because body uneasiness is age-and BMI-correlated [33], all the following results were also adjusted for the aforementioned variables.Since a higher differential activation in the precuneus was observed in both transmen and transwomen (vs. ciswomen and cismen, respectively), the whole sample was considered for the associations between precuneus ROI Z-score and BUT. Indeed, a positive association between precuneus ROI Z-score and body avoidance (BUT AV) as well as depersonalization (BUT DEP) scores (both p < 0.02; Figure 5A,B, respectively) was observed.As a higher differential activation in the posterior cingulate gyrus, in the angular gyrus and in the lateral occipital cortices was observed only in transwomen group when compared to cismen one, only these groups were considered for the following associations. In particular, the posterior cingulate gyrus Z-score was associated with body uneasiness global score (BUT GSI), body image concerns (BIC), depersonalization (BUT DEP) and avoidance (BUT AV) subscales was observed (all p < 0.02, Figure 5C–F, respectively). Beta values are reported in Figure 5. Associations between BUT GSI and BIC were not confirmed after corrections for multiple comparisons even though they showed a high significant level (p = 0.017 and 0.009, respectively).In addition, angular gyrus ROI Z-score was associated with body uneasiness global score (BUT GSI), as well with body uneasiness towards body parts total score (BUT PSDI) (both p < 0.02, Figure 6A,B, respectively). angular gyrus ROI Z-scores was associated with BUT subscales related to body image avoidance behavior (AV) and body image concerns (BIC; both p < 0.02; Figure 6C,D, respectively). Moreover, considering individual body parts, angular gyrus ROI-Z score was associated with sexual dimorphic facial characteristics, including brows, eyes, nose, chin, moustache and beard (all p < 0.02). When left lateral occipital cortex Z-score was considered, a positive association with both BUT BIC and depersonalization scales (BUT DEP) were found (both p < 0.05, Figure 6E,F, respectively). Beta values are reported in Figure 6.Finally, right lateral occipital cortex Z-Score was associated with BUT global score (BUT GSI, 0.633 p = 0.006, Figure 7A), with body uneasiness towards body parts total score (BUT PSDI, 0.619, p = 0.012, Figure 7B) as well as with BUT BIC, CSM and AV subscales (all p < 0.02; Figure 7C–E, respectively). Beta values are reported in Figure 7.This is the first neuroimaging study evaluating the face gender discrimination and its psychopathologic correlates in a sample of transgender people, compared to a sample of cisgender ones. By using validated questionnaires and fMRI, we demonstrated for the first time specific neural activation patterns in transgender people. However, it remains unclear if differences in the brain phenotype of transgender people are the result of a sex-atypical neural development or of a lifelong experience of gender non-conformity. The strengths of the present study include (i) a comprehensive design, integrating neural correlates with psychological functioning; (ii) the homogeneity of the transgender people sample with respect to age, onset of feelings of GD, sexual orientation and hormonal treatment (all hormone-naïve) [5]. The main limitation is in the cross-sectional design of the study.The main results are the following: (i) the transmen group showed a higher differential activation in the precuneus, as compared with the ciswomen one; (ii) the transwomen group, when compared to the cismen one, showed a higher differential activation in the precuneus, in the posterior cingulate gyrus, in the angular gyrus and in the lateral occipital cortices (at variance from the transmen vs. the ciswomen groups); (iii) according to psychometric evaluation (BUT), transwomen showed significantly higher distress towards sexual dimorphic characteristics of the face than transmen; (iv) differential activation in the precuneus, angular gyrus, lateral occipital cortices and posterior cingulate gyrus areas was associated with higher levels of body uneasiness.Among the aforementioned results, the most relevant is the higher differential activation in the precuneus for transgender groups when compared to the cisgender ones of the same assigned sex at birth (transwomen vs. cismen; transmen vs. ciswomen). precuneus is a superior parietal region recruited by a wide spectrum of tasks like visuo-spatial abilities, episodic memory retrieval and self-processing operations [48]. However, more recently, it has been proposed that the precuneus may also be implicated in the face perception elaboration. Indeed, the precuneus seems to be activated by identity recognition of familiar and famous persons as well as during the encoding of one’s own identity [49]. In addition. the precuneus seems to be involved in empathic judgements, as reported by the Ochsner’s fMRI study [50]. Along this line, the significant higher precuneus differential activation in the transgender groups on the face discrimination task, suggests that this region may play a crucial role in the gender identification process in transgender people, but not in the cisgender ones. We can postulate that this higher differential activation may be related to a more intense emotional involvement in transgender people while discriminating faces’ gender. Indeed, body sexual dimorphic characteristics represent for some transgender people the inner source of their sufferance and a painful remind of their gender incongruence. The experienced body-related distress may also result in a selective attention, with an intense focus on gender-specific body characteristics. In line, the gender discrimination task requires focalizing on gender dimorphic facial features. This may lead to a negative emotional arousal in transgender people due to an associative mechanism between gender-dimorphic features of the displayed faces and their own.Indeed, trans people show significant higher levels of body uneasiness (BUT) compared to cisgender ones (see Figure 4). This result confirms previous studies showing transgender individuals being more dissatisfied with their body than cisgender persons [37,51], suggesting that the body is the primary source of their suffering. The possible more intense emotional involvement in transgender vs. cisgender people during a gender identification task seems here to be corroborated by the positive correlation between the precuneus ROI Z-score with body avoidance (BUT AV) and depersonalization (BUT DEP) BUT subscales (Figure 5). In line, it has been reported that the precuneus is involved in representing different body identities and in discriminating what is perceived as part of the Self from what is perceived as stranger or unfamiliar [52].An alternative explanation can be hypothesized. Transgender people sufferance may be linked not only to gender incongruence itself, but also to the fear and/or the experience of not being socially accepted. This may be particularly true regarding facial-dimorphism characteristics which are socially visible and on which transgender people may activate a selective focus. Therefore, the hyperactivation of the precuneus may reflect the anticipated and/or experienced discrimination in transgender people in regards of self-recognition and definition.Notably, the between group analysis of the transwomen group vs. the cismen group (differently from the transmen vs. the ciswomen groups) revealed additional clusters of statistically significant higher activations beyond the precuneus, including posterior cingulate gyrus, the angular gyrus and the lateral occipital cortices. The posterior cingulate gyrus activation in a face perception task is not surprising. This region resides below the precuneus and has often been reported to be co-activated with the precuneus in face perception tasks assessing discrimination between famous and less famous faces [49].The angular gyrus is located in the posterior part of the inferior parietal lobule and is activated in face-voice integration during person recognition [53,54]. In particular, Lee et al. [55] demonstrated that face components can be reconstructed from angular gyrus fMRI activity patterns, suggesting a striking role of this region in the memory and perceptual identification of faces. Moreover, fMRI studies reported that angular gyrus is involved in perception and processing of the ‘dispositions and intentions of other individuals’, i.e., social evaluation of faces [56,57] and is a core region in conceptual processing [58].The significant positive association of both posterior cingulate gyrus and angular gyrus Z-score with body uneasiness levels here observed (Figure 5C–E and Figure 6A–D) further supports the hypothesis of a more intense emotional involvement during gender recognition task in trans people compare to cis ones. In addition. the previously reported involvement of angular gyrus in social evaluation of faces [58] may acquire a more intense significance in trans individuals (especially transwomen), who often focus on low social (and/or interiorized) acceptance of their facial sexually dimorphic features [59]. Indeed, angular gyrus ROI-Z score showed an association with sexually dimorphic facial characteristics, including brows, eyes, nose, chin, moustache and beard.Considering the activations of the lateral occipital cortices, it is worth to note that the clusters identified by our analysis included the occipital face area (OFA), but not the Fusiform Face Area (FFA). OFA and FFA have crucial functions in face perception. Lesion studies have shown that both the OFA and the FFA are necessary to assign an identity to an observed face. In fact, a malfunctioning of one of the two can lead to apperceptive prosopagnosia [60,61,62]. In particular, FFA is involved in the elaboration of face parts and their spatial configurations [63], whereas OFA is selectively employed to analyze the single face parts, e.g., mouth, nose or eyes shapes [64,65]. The higher differential activation of OFA in transwomen when compared to cismen (differently from the transmen vs. the ciswomen groups) suggests that during the cognitive process of gender identification of an observed face, these subjects recruit a lateral occipital network based on the elaboration of face elements shape. In line with this possibility, body uneasiness towards facial parts (BUT B) showed significant higher scores in transwomen with respect to all other groups. In particular, transwomen reported significantly higher distress towards sexual dimorphic characteristics of the face (i.e., forehead, brows, nose, mouth, chin, moustache, beard) when compared to transmen (Table 4). Accordingly, the intensity of the activation of these brain regions (both left and right lateral occipital cortices) was significantly associated with higher levels of dissatisfaction with physical appearance (Figure 6E,F and Figure 7A–D).Notably the combination between body uneasiness (BUT) and the coupling of angular gyrus and OFA activation in transwomen, with respect to cismen, strongly support the possibility that the analysis of the face parts is at the basis of the gender judgement in this category of transgender subjects. As a possible explanation for both the association between higher emotional attitude towards face observation in transwomen and its association with specific cerebral activation can be related with more intense shaping effect of male sex steroids during puberty on facial features. Indeed, during pubertal development, testosterone accentuates typical facial dimorphic characteristics (i.e., large jaws, prominent mandible and zygoma) as well as Adam’s apple, increasing the bodily incongruence with the perceived gender in socially exposed physical areas. Pressure to conform to gender stereotypes in western world may be considered an alternative explanation. Therefore, the differences reported in this study could be explained by the effort exerted by transwomen to adhere to the Western female stereotype also by valorizing/hiding facial parts. This may include make-up practices that require a deep inspection of women facial parts. In support, transmen when compared with ciswomen did not show activation in those brain regions related to the inspection of facial parts.Considering the small sample size, the results of the present study should be considered as preliminary. Furthermore, they should be considered in the light of some limitations, such as the use of self-reported measures. The main purpose of this work was to evaluate the neural correlates of face gender discrimination in transgenders. For this reason, we measured the differential activation deriving by the contrast female vs. male faces and not the absolute activations induced by the perception of male or female faces. Since other investigations conducted with different tasks have shown that sexual attraction can affect brain activations [25,66,67], we cannot exclude the contribution of sexual orientation in our results. However, we believe that the sexual orientation may marginally if ever have influenced our results. This convincement is based not only on the contrast we used, but also on the lack of additional areas in the other between groups’ analyses. This notwithstanding, to overcome this limitation, future studies that explores the potential relation between sexual orientation and the face gender discrimination has to be conducted to clarify this issue.Additionally, in this work we exclusively employed definite male vs female faces to explore the neural correlates of gender face perception. For this reason, we believe that further experiments using more morphing stages will be necessary to elucidate potential differences in face perception of transgender people and allow the investigation of mechanisms underlying this process.Finally, we admit that based on our results we can only hypothesize that the bilateral precuneus engagement observed in both transgender groups can be crucial in the gender face discrimination. In fact, the precuneus involvement can be associated to cognitive processes that have not been controlled here, like attention, response inhibition or conscious information processing [68,69,70]. Therefore, our experimental design cannot measure how much the precuneus recruitment in the gender face discrimination was due to the higher attention triggered by observing female or male faces, to the presence of inhibitory processes and/or to the conscious information processing related to the face observed.Our results show for the first time that GD is associated with specific brain activation patterns during gender face discrimination task. In particular, the precuneus seems to play a crucial role in gender face identity perception in the transgender groups, but not in the cis-gender ones. Additionally, transwomen, with respect to cismen, showed a differential activation of the posterior parietal cortex, the angular gyrus and the lateral occipital cortices (including the OFA, but not the FFA). Notably, these regions are all engaged in the elaboration of face parts and for this reason we can speculate that discriminating faces on the basis of gender may involve a higher attentional demand in transwomen than in cismen. This may be the result of a selective focus on gender-dimorphic facial characteristics by transwomen because of the body uneasiness itself as well as the fear of not being accepted in a transphobic society.The difference in brain phenotypes of transgender people compared to cisgender controls may suggests a sex-atypical development of the brain. However, further research is needed to clarify if these differences depend on a pre-natal sex-atypical neural development or are the consequence of a lifelong inner and social gender non-conformity experience.Conceptualization, A.D.F., G.G., M.P.V. and J.R.; methodology, A.D.F., G.G., G.C., M.M. (Mario Mascalchi); software, G.G., M.M. (Mario Mascalchi), S.O. and E.C.; validation, A.D.F., G.G., and M.P.V.; formal analysis, A.D.F., G.G., S.O., M.P.V. and G.C.; investigation, A.D.F., G.G., J.R., and M.M. (Mario Maggi); resources, A.D.F., G.G., L.V., C.S., A.R., A.C., M.P.V.; data curation, A.D.F., G.G., S.O., V.R., M.M. (Mario Maggi) and M.M. (Mario Mascalchi); writing—original draft preparation, A.D.F., G.G., C.C., J.R., F.M., A.R.; writing—review and editing, A.D.F., G.G., C.C., J.R., F.M., A.R.; visualization; supervision, A.D.F., G.G., M.M. (Mario Mascalchi), M.M. (Mario Maggi). M.P.V.; project administration, A.D.F., G.G. and M.M. (Mario Mascalchi); funding acquisition, M.M. (Mario Maggi). All authors have read and agree to the published version of the manuscript.This research was partially funded by the European Society for Sexual Medicine Grant Number [RG16-05].We thank the European Society for Sexual Medicine for the financial support. The APC received no external funding.The authors declare no conflict of interest.Task. Each trial started with a cross of fixation positioned in the center of the screen and was displayed for 3 s. Then, a face stimulus (50% male; 50% female) replaced it and was presented for 3 s. Participants had to observe the face displayed in the monitor and report its gender. At the end of each trial there was a fixation cross located in the center of the screen for 6 s (rest phase).Ciswomen vs. transmen. Between-group analysis during the face presentation (female → male) shows clusters of significantly (p < 0.01) higher blood oxygen level dependent (BOLD) effect in transmen group when compared with ciswomen. The activations are included in the precuneus. Cluster formation threshold was set at 2.3 of Z statistic value. Coordinates are reported in Montreal Neurological Institute space.Cismen vs. transwomen. Between-group analysis during the face presentation (female → male) shows clusters of significantly (p < 0.01) higher blood oxygen level dependent (BOLD) effect in transwomen group when compared with cismen. They include bilaterally the precuneus, the posterior cingulate gyrus, the angular gyrus and the lateral occipital cortices (including the occipital face area). Coordinates are reported in Montreal Neurological Institute space.(A) Body uneasiness global severity index (BUT GSI), (B) body avoidance (BUT AV), (C) body image concerns (BUT BIC), (D) depersonalization (DEP), (E) positive symptom distress index (PSDI), (F) weight phobia (BUT WP) and (G) compulsive self-monitoring (BUT CSM) according to different groups as derived by ANCOVA and Post hoc Tukey’s B test after adjustment for age and body mass index. * transwomen and transmen vs. ciswomen and cismen; ∞ transwomen and transmen vs. cismen; § transwomen vs. other groups. ANCOVA= Analysis of covariance.Association between precuneus and posterior cingulate gyrus ROI Z-scores with body uneasiness test-related scales (A,B) and (C)–(F), respectively). ROI= region of interest.(A)–(D) Association between angular gyrus ROI Z-scores with body uneasiness test related scales; (E,F) association between left lateral occipital cortex Z-score with body uneasiness test scales. ROI= region of interest.(A)–(E) Association between right lateral occipital cortex Z-score and body uneasiness test related scales.Summary of means, standard deviations and statistical differences of the main sociodemographic characteristics between cismen, ciswomen, transmen and transwomen as derived by analysis of covariance (ANCOVA) and Post hoc Tukey’s B test after adjustment for age.∞ transwomen and transmen vs. ciswomen and cismen; ∞∞ transwomen vs. cismen; ¥ cismen vs. other groups; ** transmen vs. cismen vs. transwomen; ££ transmen vs. ciswomen and cismen. VAS = visual analog scale. GIDYQ-AA = gender identity gender dysphoria questionnaire for adolescents and adults; GD = Gender Dysphoria; ANCOVA = Analysis of covariance. Boldfaced numbers highlight statically significant differences between groups.Clusters of significantly (p < 0.01) higher differential activation in the transmen Group when compared with ciswomen for female→male face contrast.Clusters of significantly (p < 0.01) higher differential activation in the transwomen group when compared with cismen for female → male face contrast.Summary of means, standard deviations and statistical differences in terms of general psychopathology (SCL-90R), gender dysphoria levels (GIDYQ-AA) and body uneasiness (BUT) between Cismen, Ciswomen, Transmen and Transwomen, as derived by ANCOVA and Post-hoc Tukey B test after adjustment for Age and BMI.§ transwomen vs. other groups; ¥ cismen vs. other groups; ∞ transwomen and transmen vs. ciswomen and cismen; * transwomen and transmen vs. ciswomen and cismen; BMI = Body Mass Index; SCL-90 = symptom checklist-90; GIDYQ-AA = gender identity/gender dysphoria questionnaire for adolescents and adults; BUT = body uneasiness test; GSI = global severity index; WP = weight phobia; BIC = body image concerns; AV = avoidance; CSM = compulsive self-monitoring; DEP = depersonalization; PSDI = positive symptom distress index. Boldfaced numbers highlight statically significant differences between groups.Summary of means, standard deviations and statistical differences in dislike of body parts (BUT-B) between male cismen, ciswomen, transmen and transwomen, as derived by ANCOVA and Post hoc Tukey’s B test after adjustment for age and BMI.§ transwomen vs. other groups; ¥ cismen vs. other groups; ∞ transwomen and transmen vs. ciswomen and cismen; ∞∞ transwomen vs. cismen; £ transmen vs. other groups; ££ transmen vs. ciswomen and cismen. ANCOVA = Analysis of covariance. Boldfaced numbers highlight statically significant differences between groups.

Med-MDPI/jcm_2/jcm-09-06-01732.txt

@@ -0,0 +1 @@
+These authors contributed equally to this work.Background: Insulin may play a key role in bone metabolism, where the anabolic effect predominates. This study aims to analyze the relationship between insulin resistance and bone quality using the trabecular bone score (TBS) and three-dimensional dual-energy X-ray absorptiometry (3D-DXA) in non-diabetic postmenopausal women by determining cortical and trabecular compartments. Methods: A cross-sectional study was conducted in non-diabetic postmenopausal women with suspected or diagnosed osteoporosis. The inclusion criteria were no menstruation for more than 12 months and low bone mass or osteoporosis as defined by DXA. Glucose was calculated using a Hitachi 917 auto-analyzer. Insulin was determined using an enzyme-linked immunosorbent assay (EIA). Insulin resistance was estimated using a homeostasis model assessment of insulin resistance (HOMA-IR). DXA, 3D-DXA, and TBS were thus collected. Moreover, we examined bone parameters according to quartile of insulin, hemoglobin A1C (HbA1c), and HOMA-IR. Results: In this study, we included 381 postmenopausal women. Women located in quartile 4 (Q4) of HOMA-IR had higher values of volumetric bone mineral density (vBMD) but not TBS. The increase was higher in the trabecular compartment (16.4%) than in the cortical compartment (6.4%). Similar results were obtained for insulin. Analysis of the quartiles by HbA1c showed no differences in densitometry values, however women in Q4 had lower levels of TBS. After adjusting for BMI, statistical significance was maintained for TBS, insulin, HOMA-IR, and HbA1c. Conclusions: In non-diabetic postmenopausal women there was a direct relationship between insulin resistance and vBMD, whose effect is directly related to greater weight. TBS had an inverse relationship with HbA1c, insulin, and insulin resistance unrelated to weight. This might be explained by the formation of advanced glycosylation products (AGEs) in the bone matrix, which reduces bone deformation capacity and resistance, as well as increases fragility.Insulin binds to their receptors in pre-osteoblasts and osteoblasts by stimulating two metabolic pathways: mitogen-activated protein kinase (MAPK) and phosphatidylinositol 3-kinase pathway (PI3-K/AKT), which favor the growth, proliferation, and survival of these cells [1,2]. Insulin behaves like an anabolic hormone that increases bone mass. Osteoblast activation produces the release of receptor activator of NF-kB ligand (RANKL) that binds to osteoclast receptor activator of NF-kB (RANK) receptors located. This simulates their proliferation and maturation. Thus, insulin plays a key role in bone metabolism [3].Hyperinsulinism and insulin resistance are the key etiopathogenic elements of type 2 diabetes. Hyperinsulinism could be responsible for the increase in bone mass observed in diabetic patients [4] and non-diabetic postmenopausal women [5]. However, not all studies show uniform results. Some report a positive association between circulating insulin levels and bone mineral density (BMD), which did not change after weight adjustment [6]. Others show an association that disappears when weight is adjusted [7,8], while others show differing results, i.e., a lack of association or negative association between insulin and bone mass [9]. Despite this increase in cortical and trabecular bone mass, patients with type 2 diabetes have an increased fracture risk [10]. Factors related to bone quality are likely to be involved [10].There have been recently developed procedures that analyze bone quality, such as the trabecular bone score (TBS). TBS is derived from the lumbar dual-energy X-ray absorptiometry (DXA) images and uses software to analyze grayscale. It correlates with three-dimensional parameters of bone microarchitecture such as trabecular connectivity, the number of trabeculae, and their separation [11]. A decline in the TBS is associated with an increased risk of fracture [12]. Three-dimensional (3D) densitometry permits an assessment of the shape and intrinsic material properties that determine bone strength. Hip DXA is transformed using segmentation algorithms to provide a 3D analysis of the cortical and trabecular compartments [13].The aim of this study was to analyze the relationship between insulin resistance and bone quality measured using TBS and three-dimensional dual-energy X-ray absorptiometry (3D-DXA) in non-diabetic postmenopausal women by determining the cortical and trabecular compartments, as well as how this is linked to body mass index (BMI) and age.A cross-sectional study was conducted in non-diabetic postmenopausal women who were suspected or diagnosed with osteoporosis. The inclusion criteria were postmenopausal women (no menstruation for more than 12 months) and a low bone mass or osteoporosis, as defined by DXA. We randomly included individuals from Rio Hortega University Hospital’s Densitometry Unit (Spain). Patients had been diagnosed with osteoporosis or suspected of low bone mass by clinical criteria according to the National Osteoporosis Foundation’s Clinician’s Guide to Prevention and Treatment of Osteoporosis [14]. A DXA was appointed for patients with this diagnosis. The exclusion criteria were type 1 and 2 diabetes, and lack of informed consent. Diagnosis of diabetes was established according to the American Diabetes Association (ADA) 2019 criteria (hemoglobin A1C (HbA1C) ≥ 6.5% and/or glucose ≥ 126 mg/dl). A protocol collected demographic data, lifestyle factors, previous illnesses, and past and present medication. The BMI was calculated. A venous sample was taken by venipuncture between 8 a.m. and 11 a.m. after 8 h of fasting. The serum was immediately frozen at −20 °C prior to testing. Glucose was calculated using a Hitachi 917 auto-analyzer. Insulin was determined using an enzyme-linked immunosorbent assay (EIA). Insulin resistance was estimated using the homeostasis model assessment of insulin resistance (HOMA-IR or HOMA model: fasting glucose × fasting insulin).A DXA scan was performed using a Prodigy scanner (GE healthcare, Madison, WI, USA), according to the manufacturer’s recommendations. The software 3D-SHAPER (2.6 version, Galgo Medical S.L, Barcelona, Spain) was also used. This method used a statistical 3D model of the proximal femur’s form and density, and was built from the quantitative computed tomography (QCT) database (Caucasian men and women). The details of this method’s modeling can be found in Winzenrieth et al. and Humbert et al. [12,13]. The study was made from DXA exploration in order to obtain a 3D model specific to the patient’s proximal femur, generating measurements in 3D from the total area of interest in the femur. The volumetric BMD (vBMD, mg/cm3), bone mineral content (BMC) (g), and volume (cm3) were calculated in the trabecular, cortical, and integral (trabecular + cortical) compartments, respectively. The cortical thickness (Cth, mm) and BMD of the cortical surface (sBMD cortical, mg/cm2, obtained by the multiplication of cortical vBMD (mg/cm3) and Cth (cm)) provided additional analysis for the cortical region. The models and 3D-SHAPER measurements’ precision were evaluated against a QCT [12,13]. The average form precision—i.e., the average distance between external limits of the femur geometry—were derived from 3D-SHAPER and QCT, and the result was 0.93 mm. Regarding bone density and cortical bone thickness, the correlation coefficients between 3D-SHAPER and the measurements derived from QCT were 0.86, 0.93, and 0.91 for trabecular vBMD, cortical vBMD, and cortical thickness, respectively [12,13].TBS was evaluated at the lumbar level (L1-L4) using TBSiNsight 2.1 (Med-Imaps, Merignac, France).Continuous variables were expressed as the mean ± standard deviation (SD), while categorical variables were expressed as the absolute (n) and relative (%) frequencies. A chi-squared test was used to compare the categorical variables. The distribution of variables was analyzed using the Kolmogorov–Smirnov test. In the case of parametric variables, we applied an analysis of variance (ANOVA) t-test. In the case of non-parametric variables, groups were compared using the Mann–Whitney U test (two groups) or Kruskal–Wallis test (more than two groups). Insulin, HOMA-IR, glucose, and HbA1c were stratified in quartiles. P-values were adjusted according to past and present medication, lifestyle factors, and the presence of osteoporosis and osteoporosis risk factors. A multiple linear regression was applied to study the association between different clinical parameters (i.e., BMI, insulin, HOMA-IR, glucose, and HbA1c) and the densitometry parameters. In the multiple linear regression, we established three models that depend on the adjusted parameters. In Model 1, the linear regression analyses were adjusted by past and present medication, lifestyle factors, and the presence of osteoporosis and osteoporosis risk factors. In Model 2, they were adjusted by past and present medication, lifestyle factors, age, and the presence of osteoporosis and osteoporosis risk factors. Lastly, in Model 3, they were adjusted by past and present medication, lifestyle factors, age, BMI, and the presence of osteoporosis and osteoporosis risk factors. We applied a Bonferroni adjustment in all comparisons. A value of p < 0.05 was considered significant. All analyses were performed using the SPSS version 22.0 statistical package (SPSS, Chicago, IL, USA).The experimental protocol was approved by the Río Hortega University Hospital of Valladolid Ethics Committee and complied fully with the Declaration of Helsinki (2008) of the World Medical Association and Spanish data protection law (LO 15/1999) and specifications (RD 1720/2007). All patients who agreed to participate gave signed written consent.Our study sample included 381 non-diabetic postmenopausal women (mean age 62 ± 9 years, mean BMI 26 ± 4). The mean age of menopause was 47 ± 6 years and the menarche was 13 ± 1.5 years. There were 29% (n = 115) smokers and 1.8% (n = 7) with excessive alcohol intake. There were 6.82% (n = 26) in treatment with thiazides, 11.02% (n = 42) with serotonin receptor inhibitors, 0.26% (n = 1) with androgenic inhibitors, 3.67% (n = 14) with beta blockers, 9.7% (n = 37) with thyroid hormone, 19.42% (n = 74) with bone antiresorptives, 1.04% (n = 4) with TSH, 2.88% (n = 11) with anabolic therapy, 10.23% (n = 39) with corticosteroids, and 2.62% (n = 10) were treated with strontium ranelate. Mean glucose values were 90 ± 24 mg/dl, mean insulin 13 ± 14 UI/l, mean HOMA-IR 3.3 ± 4.6, and mean glycated hemoglobin A1c (HbA1c) 5.6 ± 0.4%.Table 1 summarizes the clinical characteristics between osteoporotic and non-osteoporotic patients. In the preliminary analysis, we separately assessed osteoporotic individuals. There were no differences between osteoporotic and non-osteoporotic patients except that osteoporotic women were older. In addition, all the osteoporotic patients were treated and we did not find significant statistical differences in the osteoporotic group in the comparison, according the quartiles of HOMA-IR, insulin, and HbA1c (Table 2). Therefore, all participants were analyzed globally. The results were adjusted by the presence of osteoporosis.Table 2 shows the general patient characteristics, dividing the patient’s HOMA-IR, insulin, and HbA1C by quartiles. The women in the fourth quartile (Q4) of HOMA-IR and insulin had a higher BMI, and there were differences in smoking in the two groups. When we considered HbA1C, in addition to BMI, they had an older age. The results of the analysis of densitometry parameters divided patients according to quartiles of HOMA-IR, insulin, and HbA1c (Table 3). Women in Q4 HOMA-IR had higher values of vBMD but not TBS. The increase was higher in the trabecular compartment (16.4%) than in the cortical compartment (6.4%). The data for insulin were similar (Table 3). When the quartiles were assessed by HbA1c, no significant differences were observed in densitometry values. Yet in the TBS, women in Q4 had lower levels (Table 3). Table 4 summarizes the results of the linear regression analysis of the relationship between BMI, insulin, HOMA-IR, glucose, HBA1c, and the densitometry parameters. We established three models that depend on the adjusted parameters. In Model 1, the linear regression analyses were adjusted by past and present medication, lifestyle factors, and the presence of osteoporosis and osteoporosis risk factors. In Model 2, they were adjusted by past and present medication, lifestyle factors, age, and the presence of osteoporosis and osteoporosis risk factors. Lastly, in Model 3, they were adjusted by past and present medication, lifestyle factors, age, BMI, and the presence of osteoporosis and osteoporosis risk factors. While assessing the general characteristics of our population, we obtained a significant difference in BMI in our samples (Table 2). This fact, in combination with the literature on the importance of weight in bone quality and structure, lead us to include it and adjust it accordingly.Comparing the results in our models, there were barely any differences between Model 1 and 2 in statistical significance. However, in Model 3, statistical significance disappeared in the densitometry parameters (BMD total, Sdensitometry, vBMD trabecular, vBMD cortical, vBMD integral, and mCT). Our results showed that when adjusting for BMI, the statistical significance was maintained for insulin, HOMA-IR, TBS, and HbA1c. BMI had a direct and significant relationship with glucose, insulin, HOMA-IR, and HbA1c (Table 4).We found that higher values of HOMA-IR and insulin were associated with increases in vBMD, at both the cortical and trabecular levels. However, after adjusting for BMI, this relationship disappeared. The vBMD can be used to determine bone material properties like stiffness and strength, standing for a great asset of bone fracture prediction [15]. QCT is the gold standard for assessing volumetric density in the cortical and trabecular compartments [16,17]. However, QCT, compared to DXA, exposes the patient to higher radiation. Device availability is limited and the cost is relatively high, which limits the use of QCT for routine patient explorations and monitoring [18]. Recently, 3D models were generated from DXA images [19,20]. These new techniques—i.e., 3D modeling—have been proposed to solve these limitations [12]. These techniques use a statistical 3D shape and a density model of the proximal femur constructed from a database of QCT scans. The model is subsequently recorded from a standard hip DXA. Humbert et al. [19] used DXA projections to extrapolate 3D density distributions for trabecular and cortical regions, as well as the femoral shape of the femur and cortical thickness. For the extrapolated vBMD in the trabecular and cortical regions and for the mean cortical thickness, they found correlation coefficients of 0.86, 0.93, and 0.91 between 3D-DXA and QCT measurements, respectively. Biomechanical descriptors obtained through FE simulations integrate femur shape, cortical thickness, and volumetric distribution of BMD using a 3D-Shaper. These improve the discrimination of facture occurrence, especially when compared to the classical use of areal BMD or vBMD. Indeed, one compartment of the two (trabecular or cortical) can be more impacted than the other [21] and the two compartments can react differently [22,23]. The potential of this new approach can be a good surrogate of QCT, in the context of osteoporosis diagnosis and drug treatment monitoring. The fact that DXA is less invasive than QTC makes a DXA-based 3D finite element (FE) model possible, especially in clinical practice as a routine patient screening.Other studies have analyzed this relationship, but none has used the technique employed in the present study. To our knowledge, we are the first to report the relationship between 3D-DXA parameters according to insulin, HbA1c, and HOMA-IR in non-diabetic postmenopausal women. Shanbhogue et al. found an association between insulin resistance and total, trabecular, and cortical vBMD in White American women using HR-QCT [6]. Their effect was not modified after adjusting for weight. Haffner et al. found a direct relationship between insulin levels and vBMD in the femoral neck, yet the relationship disappeared after weight adjustment [24]. Similar results were obtained by Srikanthem et al. [7] and Napoli et al. [8], yet these study included 19% of diabetic patients. Other authors observed an inverse relationship with lower femoral neck strength relative load, while Asian studies found different results [9,25]. Our results showed that after adjusting for BMI, the negative association between vBMD and insulin resistance disappeared, which indicates that obesity is a key element. A small increase was observed at the cortical level and it is important to adjust the results by taking BMI into consideration, as we do in Model 3. None of our patients were diabetic although there was a high percentage of osteoporotic patients. We analyzed osteoporotic and non-osteoporotic patients together and our results showed no differences between them.It has been suggested that being overweight, expressed through BMI, has protective effects on the skeleton [26], which might be explained by various mechanisms [27]. Leptin, a cytokine produced by adipocytes, is higher in individuals with major fat content. In vitro studies have shown that leptin stimulates osteoblastogenesis without affecting mature osteoblasts [28]. However, excess weight could be harmful due to the release of inflammatory cytokines by the visceral adipose tissue [29]. Interleucin-6 (IL-6) and tumor necrosis factor-alpha (TNF-∝) increase the expression of c-fmc gene, RANK, and RANKL, all of which stimulate osteoclastogenesis. The osteocyte behaves like a mechanostat in response to the mechanical overload of being overweight. It releases IGF-I, which acts on receptors located in the osteoblasts, thus increasing bone formation [30]. In turn, this molecule blocks the action of sclerostin, an inhibitor of the metabolic Wnt pathway. A direct relationship between sclerostin and insulin resistance has been found in obese patients and patients with type 2 diabetes [31,32]. Sclerostin inhibits bone formation but the effect on bones may be limited by the release of IGF-I. Our results showed a relationship between TBS, HOMA-IR, and HbA1c that was maintained after being adjusted by other parameters, including BMI. Women in the upper quartile of HbA1c had the lowest TBS values. As TBS measures bone quality and predicts the risk of fracture [33,34], numerous studies in diabetics have shown a decline in TBS that is inversely related to insulin resistance. In our study of a non-diabetic population, we found that women in Q4 of HbA1c had the lowest TBS levels, which was maintained after adjusting for age and BMI. Recently, there have been two studies published that analyzed the relationship between HOMA-IR- and TBS-measured insulin resistance in non-diabetic populations [35,36]; their results were similar to ours. Although their sample populations were both heterogeneous (including both men and women) and pre- and postmenopausal, they also had a small sample.HbA1c assesses the metabolic situation in relation to blood glucose. Elevated glycaemia facilitates the non-enzymatic glycation that forms intramolecular bonds at the level of collagen, which forms the bone matrix, leading to the formation of so-called advanced glycosylation products (AGEs) [37]. AGEs are located in the middle region of collagen fibers and reduce bone deformation capacity as well as decrease resistance, which thus increases fragility. AGEs act on osteoblasts and osteoclasts. Moreover, they act on specific osteoblast receptors and decrease their proliferation and differentiation. They also activate the NF-kb pathway in osteoblasts, increasing the release of inflammatory cytokines that act on osteoclasts, thus stimulating their proliferation and activation, and increasing bone resorption [38]. These findings have some congruence with studies of TBS in type 1 diabetes where a relationship between fracture and TBS and fracture and HbA1c have been described [39].Although our study provides a comprehensive evaluation of this topic, there were some limitations, which were determined by the absence of osteoporotic women without treatment. Moreover, we did not have data regarding the analysis of phosphocalcic metabolism or bone remodeling markers. The strengths were determined by its sample size, the homogeneity of the studied population, and to the best of our knowledge we are the first group to report on the relationship between 3D-DXA parameters according to insulin, HbA1c, and HOMA-IR in non-diabetic postmenopausal women.In conclusion, we found a direct relationship between insulin resistance and vBMD in non-diabetic postmenopausal patients, whose effect is directly related to higher weight and had the greatest effect at the trabecular level. Thus, bodyweight is fundamental when evalutating a patient’s overall fracture risk. TBS was inversely related to HbA1c and unrelated to weight, indicating a deterioration in bone quality that could justify an increase in fracture rusk due to glycation affecting the bone architecture and impairing its correct healing.Study design: J.L.P.-C., F.C.-S., and R.U.-M.; patient recruitment: F.C.-S., R.U.-M., Á.R.-d.T., J.G., M.R.-M. and J.L.P.-C.; experimental work: F.C.-S., R.U.-M., Á.R.-d.T., J.G., M.R.-M. and J.L.P.-C.; data analysis: R.U.-M. and J.L.P.-C.; data interpretation: F.C.-S., R.U.-M., Á.R.-d.T., J.G., A.D.-L., J.M.F.-G., and J.L.P.-C.; drafting manuscript: F.C.-S., R.U.-M., A.D.-L., J.M.F.-G., and J.L.P.-C.; revising manuscript: F.C.-S., R.U.-M., Á.R.-d.T., J.G., A.D.-L., J.M.F.-G., M.R.-M. and J.L.P.-C.; approving final version of the manuscript: J.L.P.-C.; coordination of the study: J.L.P.-C. All authors have read and agreed to the published version of the manuscript.This research received no external funding.The authors declare no conflict of interest.General characteristics of the non-osteoporotic and osteoporotic subjects.BMI: Body mass index.General characteristics of the study population stratified by quartiles of HOMA-IR, insulin and HbA1c.BMI: Body mass index. HbA1c: Glycated hemoglobin A1c. HOMA-IR: homeostasis model assessment insulin resistance.Densitometry parameters according quartiles of HOMA-IR, insulin, and HbA1c.HOMA-IR: homeostasis model assessment-insulin resistance. HbA1c: glycated hemoglobin A1c. BMD: bone mineral density. vBMD: volumetric bone mineral density. mCT: micro computed tomography. TBS: trabecular bone score.Lineal regression analysis between the different parameters (BMI, insulin, HOMA-IR, glucose, and HBA1c) and the densitometry parameters. In Model 1, the linear regression analyses were adjusted by past and present medication, lifestyle factors, and the presence of osteoporosis and osteoporosis risk factors. In Model 2, they were adjusted by past and present medication, lifestyle factors, age, and the presence of osteoporosis and osteoporosis risk factors. Lastly, in Model 3, they were adjusted by past and present medication, lifestyle factors, age, BMI, and the presence of osteoporosis and osteoporosis risk factors.ρ: Correlation coefficient.

Med-MDPI/jcm_2/jcm-09-06-01733.txt

@@ -0,0 +1 @@
+A complete list of the members of the COVID@HULP Working Group is provided in the Supplementary Materials.Background: Since the confirmation of the first patient infected with SARS-CoV-2 in Spain in January 2020, the epidemic has grown rapidly, with the greatest impact on the region of Madrid. This article describes the first 2226 adult patients with COVID-19, consecutively admitted to La Paz University Hospital in Madrid. Methods: Our cohort included all patients consecutively hospitalized who had a final outcome (death or discharge) in a 1286-bed hospital of Madrid (Spain) from 25 February  (first case admitted) to 19 April 2020. The data were manually entered into an electronic case report form, which was monitored prior to the analysis. Results: We consecutively included 2226 adult patients admitted to the hospital who either died (460) or were discharged (1766). The patients’ median age was 61 years, and 51.8% were women. The most common comorbidity was arterial hypertension (41.3%), and the most common symptom on admission was fever (71.2%). The median time from disease onset to hospital admission was 6 days. The overall mortality was 20.7% and was higher in men (26.6% vs. 15.1%). Seventy-five patients with a final outcome were transferred to the intensive care unit (ICU) (3.4%). Most patients admitted to the ICU were men, and the median age was 64 years. Baseline laboratory values on admission were consistent with an impaired immune-inflammatory profile. Conclusions: We provide a description of the first large cohort of hospitalized patients with COVID-19 in Europe. Advanced age, male sex, the presence of comorbidities and abnormal laboratory values were more common among the patients with fatal outcomes.As of this writing, Spain has the second highest number of confirmed severe acute respiratory coronavirus 2 (SARS-CoV-2) infections worldwide after the United States. The first infection in Spain was confirmed on 31 January 2020, in La Gomera in the Canary Islands [1]. In the Madrid region, the first infection was registered on 25 February 2020. Madrid is a densely populated area with 6.7 million inhabitants [2] and has felt the greatest impact from the pandemic in Spain. The number of confirmed cases in Madrid was 58,819 as of 25 April (26.3% of cases in Spain) [3], reaching a peak of 3419 new cases on 30 March.The progression of the outbreak in Madrid is similar to that observed in the most affected areas in Western countries, such as the Lombardy region in Italy and New York City in the US. The healthcare systems of these regions are under massive stress, and the cumulative COVID-19 mortality per 100,000 inhabitants since the start of the pandemic has been high: 132 deaths in Lombardy, 140 in New York City and 190 in Madrid (as of 25 April) [4,5,6].The La Paz University Hospital is a large teaching hospital with a catchment area of 527,366 inhabitants in the north of Madrid. Shortly after the outbreak, the hospital’s procedures were adapted to cope with the rise in COVID-19 cases. By 25 April, the hospital had admitted over 2500 patients with COVID-19, one of the largest single-site cohorts in Europe. These patients’ clinical information was collected using a standardized protocol. In this article, we describe the first 2226 adult patients consecutively admitted with a confirmed diagnosis of SARS-CoV-2 infection to La Paz University Hospital and who had died or been discharged by 19 April.Our study included all individuals, 18 years or older, who were hospitalized in the wards (or emergency department, due to the lack of available beds in the wards) of La Paz University Hospital with a diagnosis of COVID-19 and who either died or were discharged by 19 April. Patients discharged from the emergency department after a stay of less than 24 h were not considered hospitalized and were not included in this analysis.We employed a modified version of the electronic case record form (eCRF) for severe acute respiratory infections, developed by the World Health Organization/International Severe Acute Respiratory and Emerging Infection Consortium [7]. Our eCRF includes 372 variables, grouped into demographics, medical history, infection-exposure history, symptoms, complications, treatments (excluding clinical trials) and disease progression during hospitalization (see Supplementary Material).We collected the clinical data directly extracting the information from the hospital’s database, when possible, or by a manual and individual review of the patients’ electronic clinical records, including the clinical notes (DXC-HCIS- Healthcare Information System). The clinical data collected at hospital admission included age, sex, smoking status, transmission, comorbidities, symptoms on admission, respiratory status and time from disease onset. Complications during hospitalization and intensive care unit (ICU) admission were also recorded. The data collection effort was conducted by a volunteer team of resident doctors and senior medical students. Data monitoring was conducted by our hospital’s Central Clinical Research Unit.Laboratory results (hematology, biochemistry, microbiology) were extracted from various hospital data management systems, and information regarding the drugs used during hospitalization was extracted from the electronic prescription system. Continuous variables are presented as means and standard deviations (SD) or medians and interquartile ranges (IQR), and categorical variables are listed as numbers and percentages (%). To analyze predictors of in-hospital death, we employed a multivariate logistic regression model. We selected the variables for inclusion in the model on the basis of previous findings and considering the total number of deaths in our study to avoid overfitting the model. In a sensitivity analysis to explore potential collider bias, we restricted the fit of the model to patients admitted before 20 March(results did not materially change). We performed the statistical calculations using R (version 3.4.0) [8].The study was approved by the Research Ethics Committee of La Paz University Hospital (PI-4072) and by the Spanish Agency of Medicines and Medical Devices (HUL-AIN-2020-01) and was registered in the European Union Electronic Register of Post-Authorization Studies (EUPAS34331).A total of 3127 patients were consecutively treated in the emergency department of La Paz University Hospital between 25 February  and 19 April 2020. Of these, 2226 adult patients were hospitalized and either died (460, 20.7%) or were discharged (1766, 79.3%) and were therefore included in our analysis. Figure 1 shows the bed occupancy by patients with COVID-19 over time, with a peak of 1033 beds, 106 of which were in the ICU (compared with 30 ICU beds before the COVID-19 pandemic).The median time from clinical onset to hospital admission was 6 days (IQR, 3–9). At admission, the patients had a median age of 61 (IQR, 46–78) years, 52% were women, 41% had arterial hypertension, 19% had chronic heart disease, and 17% had diabetes mellitus. The most common symptoms at admission were fever, cough and dyspnea, and the median oxygen saturation at admission was 95% (IQR, 92–97). The most common complications during hospitalization were acute confusional syndrome, acute kidney failure and acute respiratory distress syndrome (Table 1).Most of the patients were treated with drugs presumed to have antiviral activity against SARS-CoV-2. The most frequent combination was hydroxychloroquine plus azithromycin followed by hydroxychloroquine in isolation.At the time of the analysis, 237 patients had been admitted to the ICU, 116 remained in the ICU, 55 had died, 20 were discharged from the hospital, and 46 remained confined to a standard hospital bed. Table 2 shows the demographic characteristics, comorbidities and respiratory status on the day of emergency department admission of the 75 patients (3.4%) transferred to the ICU who had died or been discharged by 19 April. Compared with the entire cohort, the patients admitted to the ICU were older (median age, 64 vs. 61 years), had a higher male/female ratio (3.2 vs. 0.93) and had a higher prevalence of arterial hypertension (52 vs. 41.3%), obesity (30.7% vs. 10.9%), diabetes mellitus (28.0% vs. 17.1%) and chronic obstructive pulmonary disease (17.3% vs. 6.9%).Table 3 shows the mortality by age group and sex for the 2226 patients. The overall mortality was 26.6% for the men and 15.1% for the women. Mortality increased with age, reaching over 60% for patients over 80 years of age.Table 4 shows the laboratory findings at admission for the entire cohort and for the ICU subgroup. In the cohort, the baseline creatine kinase, creatinine, D-dimer, ferritin, lactate dehydrogenase, procalcitonin, C-reactive protein and high-sensitivity cardiac troponin I levels and prothrombin times were higher among the non-survivors than the survivors. In the cohort admitted to the ICU, the most notable differences with the whole cohort were higher levels of D-dimer, ferritin, C reactive protein and troponin. Within the ICU cohort, the lymphocyte counts and procalcitonin and C-reactive protein levels at hospital admission were also clearly higher in the patients who died compared with those who survived.In the multivariable logistic regression model, we found that male sex, older age, an oxygen saturation <90% on admission, lower lymphocyte count and high C-reactive protein levels were associated with a high probability of death (Table 5).To the best of our knowledge, this is the first report of a large cohort of patients hospitalized with COVID-19 in Europe. Our cohort includes all patients with a final outcome (discharge or death) consecutively admitted to our hospital during the worst phase of COVID-19 in Madrid’s hospital system.Similar to other cohorts [9,10,11,12,13,14,15], the hospitalized patients in Madrid were elderly and had numerous comorbidities, the most common of which were arterial hypertension and diabetes. Our male/female ratio was 0.9, which is lower than the 1.5 reported in the series from Wuhan (China) [12] and New York City [10]. In Madrid, the male/female ratio for individuals older than 60 and 75 years was 0.74 and 0.61, respectively [16]. Differences compared with other cohorts could be partially explained by the different male/female ratio in the Madrid population pyramid. Despite our lower male/female ratio compared with other reports, the mortality for each age group was notably higher for the male patients than for the female patients, as reported in other cohorts. It is relevant that one third of the patients included in our cohort were nursing homes residents.The overall mortality (20.7%) in our series by age group was similar to that of the New York cohort (21%) [10] and lower than that of a Wuhan cohort (28.3%) [12]. In our cohort, older age and the presence of comorbidities were more common among the patients with fatal outcomes, both for the entire cohort and for those admitted to the ICU.The most frequent symptoms at admission were fever, cough and dyspnea; myalgia and diarrhea were also common. Notably, 12.8% of the patients had anosmia as a presenting symptom, as described in other cohorts [17]. The time from disease onset was short (6 days), and the patients’ respiratory status on admission (as reflected by oxygen saturation) was generally poor, with half of the patients presenting an oxygen saturation of < 95%. General symptoms, such as diarrhea, myalgia, headache and anosmia, were more common among survivors, while respiratory symptoms such as dyspnea and sputum production were more prevalent among the non-survivors, who also had a lower median oxygen saturation on admission (90% vs. 96%). In any case, these findings need to be confirmed in other large cohorts and meta-analyses.The laboratory values on admission were consistent with an impaired immune-inflammatory profile, characterized by lymphopenia and elevated D-dimer, procalcitonin, ferritin and C-reactive protein levels. Most of these abnormal laboratory readings were more common in the patients with fatal outcomes. Creatine kinase and troponin levels on admission were also higher in the patients with fatal outcomes, a finding also reported in other series [18].Most of our patients underwent treatments presumed to have activity against SARS-CoV-2. The indication for treatment in Spain has changed during the course of the pandemic but has always been based on the indications by the Spanish Ministry of Health and the availability of treatments. Our hospital has also participated in several clinical trials, the results of which we cannot provide in this manuscript (remdesivir, tocilizumab, sarilumab). We have included in this article the use of treatments with potentially antiviral effects available in our hospital but that have not been included in a clinical trial (hydroxychloroquine, azithromycin and lopinavir/ritonavir). It is also noteworthy that, at the start of the pandemic in Spain, only symptomatic treatment was indicated. At present, there are no clinical trial data that support the use of any of these treatments for improving the outcomes of patients with COVID-19.This study has a number of limitations. First, the data were collected from various databases, both manually and automatically. The data manually entered into the eCRF were monitored and curated. Second, we did not conduct a follow-up of the patients after discharge. Third, our reported mortality rates might change once the entire cohort of hospitalized patients has been analyzed.In summary, this study provides initial data on the clinical and laboratory features and outcomes of patients hospitalized with COVID-19 infection in a large teaching hospital in Madrid during the peak of the pandemic in Spain.The following are available online at https://www.mdpi.com/2077-0383/9/6/1733/s1, 1. COVID@HULP Working Group list of members and Acknowledgments, 2. Electronic case record form (eCRF).Conceptualization, A.M.B., A.J.C., E.R. and J.R.A.; Data curation, J.M.-V., M.Q., A.M.-V., G.M.-A. and D.P.A.; Formal analysis, A.M.B., A.J.C., J.M.-V., E.R. and J.F.; Investigation, A.M.B., F.A., R.Á.-S., J.M.-V., M.Q., J.C.F., R.M.T.S.-O., J.G.-R., A.M.-V., A.B., E.R., G.M.-A., N.G.-A., M.C.N., M.M.-d.-G., F.M.R., F.R.-B., A.M.-Q., A.R.N., J.M., C.J.C.S. and J.F.; Methodology, A.M.B., A.J.C., E.R., G.M.-A., J.F. and J.R.A.; Supervision, A.M.B., M.Q., E.R.C., C.P.S., J.J.R. and J.R.A.; Validation, J.F.; Visualization, J.R.A.; Writing – original draft, A.M.B., A.J.C., E.R. and J.F.; Writing – review & editing, A.M.B., A.J.C., E.R., M.A.H., J.F. and J.R.A. All authors have read and agreed to the published version of the manuscript.This research received no external funding.We thank all La Paz Hospital staff, whose selfless dedication is allowing us to face the worst healthcare challenge of our history. We also want to acknowledge Hospital Management for its support to the Covid@HULP project.The authors declare no conflict of interest.Patients treated by the emergency department per day between 25 February 2020, and 19 April 2020. The blue lines indicate patients admitted and hospitalized, and the red lines indicate patients discharged from the emergency department. The green line represents the percentage of hospital beds (including ICU) occupied by patients with COVID-19.Characteristics at admission and complications during hospitalization.Abbreviations: (f), Fisher’s Exact Test; HIV, human immunodeficiency virus; IQR, interquartile range; SatO2, oxygen saturation; (w), Wilcoxon Rank Sum Test.Demographics, Comorbidities and Respiratory Status on Admission of the Intensive Care Patients.Abbreviations: (f), Fisher’s Exact Test; HIV, human immunodeficiency virus; IQR, interquartile range; SatO2, oxygen saturation; (w), Wilcoxon Rank Sum Test.Mortality distribution by age group and sex.Laboratory findings on admission.Abbreviations: ALT, alanine aminotransferase; ICU, intensive care unit; IL, interleukin; IQR, interquartile range; NR, normal range.Risk factors associated with in-hospital death.Abbreviations: SatO2, oxygen saturation.

Med-MDPI/jcm_2/jcm-09-06-01734.txt

@@ -0,0 +1 @@
+One thousand and eighty patients, having prolonged bleeding times, frequent epistaxis, menorrhagia or easy bruising or other bleeding manifestations, and excluding those with von Willebrand’s disease, were evaluated for platelet dense granule deficiency. The mean diameter of platelet dense granules was determined for all patients using image analysis. Four hundred and ninety-nine had “classic” dense (delta) granule storage pool deficiency (δ-SPD). Five hundred and eighty-one individuals (53.8%) were found to have a normal mean number of dense granules, but for some of these patients, the dense granules were smaller than for the controls. Of the patients having a normal number of dense granules, 165 (28.4%) were found to have significantly smaller granules than the platelets obtained from the control subjects. Their average granule diameter was 123.35 ± 0.86 nm, that is more than three standard deviations below the mean of the control data. Total δ-granule storage pool volumes (TDGV)/platelet were calculated using these measurements. Individuals with δ-SPD had half the number of granules (2.25 ± 0.04 DG/PL) and storage pool volume (3.88 ± 1.06 × 106 nm3) when compared to our control data (4.64 ± 0.11 DG/PL; 10.79 × 106 nm3 ± 0.42). Individuals having a bleeding history but a normal average of small dense granules had a calculated storage pool volume statistically different than controls and essentially the same storage pool volume as patients with δ-SPD. We have identified a sub-classification of δ-SPD that we have defined as micro-granular storage pool deficiency (δ-MGSPD).Storage pool deficiency (SPD) is a heterogeneous bleeding disorder related to a decreased number of specific platelet organelles resulting in platelet dysfunction. Deficiencies of platelet alpha granules (α-SPD), dense (delta) granules (δ-SPD) and combined deficiencies of both alpha and delta granules, including their respective substance contents, will all manifest with common bleeding symptoms. The clinical presentation of a patient with SPD often includes a history of easy bruising, frequent epistaxis, surgical bleeding, heavy menses and/or post-partum hemorrhage with a normal or prolonged screening test of platelet function in terms of the bleeding time or closure time with or without abnormal aggregation tests. Both hereditary [1,2,3,4,5,6,7,8] and acquired [9,10] etiologies have been reported. Chédiak–Higashi [1,11], Hermansky–Pudlak [12,13,14,15], and Wiskott–Aldrich [16] syndromes are examples of hereditary δ-SPDs. Acquired δ-SPDs have been reported in subjects from myeloproliferative disorders [17] cirrhosis [18,19], and uremia [20,21,22].Patients presenting with a history of bleeding predominantly affecting the skin, mucous membranes, gastrointestinal, gynecological, and urinary tracts, or who have had a significant post-surgical bleeding complication, should be evaluated for a platelet quantitative or qualitative disorder. It is not uncommon for patients to remain without a diagnosis following platelet aggregation studies if von Willebrand antigen and activity levels and factor VIII clotting activity are all normal. In such instances, δ-SPD should be considered a potential etiology. Nieuwenhuis et al. reported that of 106 patients diagnosed with δ-SPD, 23 individuals (13 congenital and 10 acquired δ-SPD) had normal aggregation responses to ADP, epinephrine, and collagen [23]. A similar study by Israels et al. reported that of 153 new patients referred for a work-up of abnormal bleeding had a prolonged bleeding time and of these, 46 were evaluated for platelet dense granule deficiency and decreased ATP release [24]. Ten percent (5/46) were found to have a dense granule deficiency, yet had no aggregation defect [24]. Both groups of investigators, using a normal reference range of 3–8 dense granules per platelet (DG/PL), concluded that δ-SPD is common, heterogeneous and cannot be excluded using aggregation assays alone. We have similar unpublished data that was presented at an International Society on Thrombosis and Haemostasis meeting in Toronto, Canada [25] and has been confirmed in a report by Brunet et al. [26]. More recently, others have reported that dense granule reference ranges may vary by age group [27].We include the electron microscope (EM) as an essential tool to evaluate patients with unexplained bleeding for potential dense granule deficiency (δ-SPD), using the platelet whole mount technique originally described by Bull [28] and refined by White [12,13,24,29]. Our methodology follows classic protocols that have been established in the literature over the past five decades; we do not have energy dispersive X-ray (EDX) available to use when evaluating platelets. Initially, we identified a cohort of patients (n = 13) with bleeding and elevated numbers of dense granules (δ-SPD); nine of whom had an abnormal aggregation response to either ADP or epinephrine. Many of these patients had significantly smaller DGs than those of normal subjects. Subsequently, we tested the hypothesis that the determination of the DG size and volume may be useful in the evaluation of patients with significant bleeding symptoms with normal to elevated numbers of DGs and who were not found to have von Willebrand disease (VWD). We report a variant of δ-SPD that we have termed microgranular (delta) storage pool deficiency (δ-MGSPD).This study was based upon the clinical and laboratory observations of 13 individuals seen at our hospital who presented with abnormal bleeding and unusual laboratory test results, including elevated numbers of platelet dense granules enumerated by the utilization of platelet whole mount electron microscopy. The patient demographics, including 9 females and 4 males, whose ages ranged from 13 to 80 years, are shown in Table 1. Nine out of 13 had a family history of bleeding. The ethnic make-up included 10 Caucasians (2 Eastern European Jewish), 1 African American, 1 Hispanic and 1 Laotian. The severity of the bleeding ranged from easy bruising and menorrhagia to life threatening bleeding after invasive procedures. While many had been treated with fresh frozen plasma, cryoprecipitate or vitamin K for their bleeding episodes, none had been empirically treated with platelet transfusion or with desmopressin (DDAVP). Six of the 13 individuals had a bleeding time (BT) that was prolonged at >8.5 min (BTs were done with a Simplate device, with the cuts parallel to the antecubital crease). There was no correlation between the bleeding time and the extent of clinical bleeding. Twelve individuals had platelet aggregation studies performed (the 13th specimen clotted and was not reordered). As has been reported for dense granule deficiency [12,23], the aggregation response was variable. All had a normal aggregation response to 1 and 4 µg/mL of equine tendon collagen. Four individuals had a normal response to both ADP and epinephrine, while 4 had a suboptimal response to either ADP or epinephrine, and 5 had a suboptimal response to both ADP and epinephrine. During the assessment by transmission electron microscopy (TEM), their platelet dense granules appeared to be smaller than usual. There was no correlation between the aggregation response to ADP and epinephrine and the dense granules size. We were unable to assess the subsequent cases included in this study for correlations as the majority of cases were submitted from a number of different institutions without accompanying clinical and laboratory test data.This is a descriptive case series that utilized the blood collected from 49 healthy adults, used as quality control donors for aggregation assays, to establish control values of the platelet dense granule number, dense granule diameter and the concentration of adenine nucleotides extracted from platelets to determine the ATP:ADP ratios. We presented the data collected over a twenty-year period from 1080 individuals with unexplained bleeding, suggestive of a platelet function disorder, who had been assessed for diagnostic purposes using whole mount TEM (wmTEM), dense granule diameters, and the infrequent determination of adenine nucleotide ratio. Many subject blood samples originated from institutions other than the authors’ institutions and for most samples, clinical symptoms and associated laboratory assessment results were not provided. All the samples included in this study were submitted to our laboratory for the evaluation of the unexplained bleeding. When the clinical history was provided, all the subjects had reported symptoms suggestive of a platelet function disorder or other hemostasis defect. Platelet dense granule stability in whole blood acid dextrose citrate vacutainers and shipped via air freight to our laboratory was unaltered if received within 48 h of venipuncture [30]. To validate the dense granule stability in the blood samples that were shipped to our laboratory, we compared the mean number of DGs/PL in whole blood samples obtained from 35 healthy student volunteers, all in their 20s with a balance of both sexes. Two vials were drawn, with one sample analyzed immediately after venipuncture and another one shipped airfreight to Kansas City, Missouri. Each shipment (each contained 5–7 samples with 1 shipment per week) was immediately reshipped to our laboratory and processed 48 h post-venipuncture for comparison. There was no significant difference between the dense granule number in the samples assessed at the day of venipuncture (4.4 ± 0.2 DG/PL) compared with blood that was shipped and analyzed 48 h post-venipuncture (4.3 ± 0.2 DG/PL) [30]. The results were also within the normal range of DG/PL as established in the literature [13,25]. All the procedures were performed using Internal Review Board (IRB)-approved protocols as specified by Federal guidelines and standards and conforming to the rules of the Declaration of Helsinki of 1975.Peripheral blood was acquired from subjects by routine venipuncture and placed into 7 mL or 10 mL ACD (acid citrate dextrose) Vacutainer® brand evacuated blood collection tubes containing 22.0 mg/mL trisodium citrate, 8.0 mg/mL citric acid and 24.5 mg/mL dextrose. Platelet rich plasma (PRP) was obtained by centrifugation at 100 g for 15 min. Each vial of centrifuged blood was incubated at 37° in a sand bath for 30 min, allowing a “rest” period for the platelets prior to the subsequent processes. The determination of the PRP platelet count was made using a Coulter Hematology Analyzer (STK-R or S, Beckman Coulter, Inc., Fullerton, CA). Aliquots of each sample were processed for whole mount transmission electron microscopy (wmTEM) and the extraction of adenosine nucleotides. These samples were not utilized for light transmission aggregometry (LTA).The whole mount technique for the enumeration of platelet delta granules has been previously described [13,28,31,32]. In summary, drops of PRP were placed upon multiple electron microscope (EM) grids coated with parlodion and stabilized with evaporated carbon, for 5–10 min at room temperature (approximately 20 µL for each grid). The platelets adhered to the surface of the carbon-coated plastic support film without becoming activated. Subsequently, the grids were blotted with bibulous paper to remove the excess PRP and were “washed” by placing the grids upon drops of distilled water briefly and blotted again; the wash step was repeated three times. Platelet whole mount preparations were allowed to air dry before proceeding to the TEM. Dense granules were readily identified in air-dried whole cell preparations due to their calcium content in contrast to the epoxy-embedded and sectioned platelets that are representative of only a thin sliver of a platelet’s cytosol with other organelles, including alpha granules and lysosomes, also having electron density imparted via heavy metal fixation techniques and post-section staining (Figure 1).Once the aliquots of PRP were acquired for the preparation of whole mounts and extractions, an additional centrifugation at 500× g for 15 min and the removal of the plasma layer were conducted, the overlaying buffy coat was fixed with 3% glutaraldehyde fixative and buffered with 0.2 M sodium cacodylate for 2 h. Subsequently, the fixed buffy coat was divided and processed for the routine embedment for transmission electron microscopy (TEM), high-voltage transmission electron microscopy (HVEM) and an uranaffin reaction [12].The uranaffin reaction used for this study was a modified technique (slightly lower pH and decreased reaction time). The portions of glutaraldehyde-fixed buffy coat were immersed in saturated aqueous uranyl acetate (pH 3.3) for two hours, followed by a graded series of ethanolic dehydration steps, a transitional dehydration step with absolute acetone and the subsequent infiltration and embedment in Spurr’s epoxy resin [33,34]. The tissues prepared in this manner were sectioned for both routine TEM and for HVEM. The platelets subjected to the uranaffin reaction were evaluated by routine TEM and classified based upon their morphology as described by Weiss et al. [12]. The granules with electron-dense material occupying more than 50% of their interiors were classified as Type 1 (Figure 2A); solid granules occupying less than 50% of the membrane-limited vacuoles were classified as Type 2 (Figure 2A); granules with fragmented cores were classified as Type 3 (Figure 2A); and “empty sacs” vacuoles with uranaffin-positive membranes without an electron-dense interior were classified as Type 4 (Figure 2A) [12].Buffy coat samples processed for the uranaffin reaction from 10 normal subjects and 10 δ-SPD cases and embedded in epoxy were sectioned 0.75–1.0 micron thick for HVEM. The “thick” sections were placed on copper mesh electron microscopy grids and air dried without any post-section staining procedures. These 20 samples were evaluated by HVEM (1.0MV) at the NIH Biotechnology Resource at the Wadsworth Center in Albany, New York, to determine platelet dense granule shape and size from a series of electron photomicrographs recorded at different specimen tilt angles (Figure 3). The investigation included the image analysis of corresponding wmTEM photographs for comparison with the HVEM images performed as a blind study.The extraction and determination of platelet adenosine nucleotide content were performed by standard methods [3,35,36,37,38]. A 500 µL sample of PRP was mixed with 450 µL of 96% ethanol (EtOH) and 50 µL of 0.1M EDTA, pH 7.4, for 10 s, and centrifuged for 15 min at 14,000× g. The supernatant was stored at −25 °C until assayed.The ATP and ADP contents of the platelets were determined by the bioluminescence of firefly luciferase method. Using a Lumat LB9501-2 Luminometer (Wallac Inc., Gaithersburg, MD, USA) and known quantities of ATP (Sigma ATP Bioluminescent Assay Kit (Cat. FL-AA), Sigma Chemical Co., St. Louis, MO, USA), a standard curve was determined prior to the analysis of adenine nucleotides extracted from PRP preparations. The procedure determined the total ATP and ADP concentrations extracted from the lysed platelets and was not used to determine the adenine nucleotides contained solely within the storage pool dense granules (DGs). The ATP analysis was quantified first. To determine the ADP content of the platelets, the extracted ADP was converted to ATP using creatinine phosphokinase and phosphocreatine, and quantified using the bioluminescence assay [3]. The amount of ADP in the lysate could then be determined by the subtraction of the original ATP concentration in the lysates from the results of the second assay that included ADP converted to ATP [3].For this study, a total of 100 contiguous air-dried platelets were assessed for DG content; platelets (PL) that were partially obscured by a grid bar or that exhibited preparation artifacts were excluded from the evaluation. The total number of DGs enumerated within the whole mounted platelets included in the assessment were divided by 100 (total PL observed) and the average number of DG/PL was determined. During the enumeration of DGs, 12 random platelet fields were photographed at 10,200X using a Philips 410LS TEM. The platelets were representative of the preparation and included 12–20 platelets for subsequent image analysis to accurately measure the DG diameter and to calculate the DG volume. The platelets that were photographed had at least 4 DG/PL for purposes of generating a reasonable N for statistical validity. No statistical difference was found for DG size between DGs measured in platelets with 1–4 DG/PL versus platelets having more than 10 DG/PL. The image analysis employed NIH Image 1.57®, Image Pro 5.1© (MediaCybernetics, Silver Spring, MD, USA) and proprietary software written by two of the co-authors (BR and SR). The diameter of each digitized DG was used to calculate the volume of each DG (4/3 πr3). The mean DG volume (MDGV) for each platelet was calculated from the sum of DG values divided by the number of DGs evaluated. The total dense granule volume (TDGV) was calculated by multiplying the MDGV by the mean number of DG per PL. The confirmation of DG spherical morphology used 1 µm thick plastic sections of uranaffin-reacted PLs photographed at different specimen tilt angles with the HVEM (Figure 3). Using Image Pro 10©, another co-author (MR) utilized artificial intelligent macros to validate all the measurements obtained initially for the 13 index cases (Supplementary Figure S1: Comparing AI vs. Manual Measurements of Index Cases, Supplementary Table S1: AI, Supplementary Table S2: Total Group DG Numbers and Volumes).The values for the average DG/PL and TDGV/PL were calculated. The values for the “normal” (control) subjects were obtained and averaged for comparison with the values calculated for individuals who had a bleeding history. For each calculation, the normal values were based upon a range including 3 standard errors of the mean (SE) from the average of all control samples. Student t and chi square tests were performed using SigmaPlot© 12 (SSPS, Inc., Chicago, IL, USA).The platelets of 49 normal (control) subjects contained 4.64 ± 0.11 (SE) DG/PL, similar to the values previously reported [4,12,13,38,39,40]. The mean diameter of the control dense granules was 143.14 ± 1.56 nm (Table 2) which was also similar to that previously reported [32,41].This study included 1080 individuals, none of whom were thrombocytopenic and had platelet electron microscopy (EM) and image analysis performed for the evaluation of unexplained bleeding. Four hundred and ninety-nine (499) patients had a classic dense granule deficiency with an average number of DG/PL of 2.25 ± 0.04 (Table 2). The average diameter of their dense granules (151.91 ± 23.74 nm) was similar to the average diameter of normal subjects’ DGs (143.0 ± 1.56 nm (SE)). Three hundred and twenty-eight (328) patients having bleeding symptoms were all found to be within normal limits using the analyses employed in this study. A total of 88 individuals were found to have an elevated number of DGs (6.64 ± 0.11 DG/PL) with a normal diameter (141.76 ± 0.17 nm) but significantly elevated TDGV (14.23 ± 0.54 × 106 nm3) compared to all the other groups which is of unknown significance; three of these individuals were known to be hyper-aggregable, but most presented with bleeding symptoms. One hundred and sixty-five (165) patients were found to have an average dense granule diameter more than three standard errors of the mean below the mean of normal subjects. Their platelets contained 4.31 ± 0.36 DG/PL (no statistical difference) with a mean dense granule diameter of 123.36 ± 0.86 nm (p < 0.001) compared to the controls. The calculated mean volume of their DGs was significantly decreased compared to normal (p < 0.001). The total DG storage pool volume per PL was 5.99 ± 1.13 x 106 nm3, which was significantly decreased compared to the value obtained for the normal subjects (10.79 ± 0.39 × 106 nm3, p < 0.001). The TDGV per PL for these individuals was found to be similar in magnitude to that seen in classic platelet dense granule deficiency (3.88 ± 1.06 × 106 nm3). We have termed this defect δ-MGSPD, a dense granule storage pool volume deficiency (SPVD). There was no correlation between the platelet count, the DG number, nor the total volume of DG/PL for individuals with SPVD. For both normal subjects and patients with classic dense granule deficiency (DGD), there was a very strong relationship (r = 0.87) between the dense granule number and the total volume of dense granules (Table 2). Using the dense granule number and the total volume of dense granules, individuals with δ-MGSPD were easily separable from the continuum of normal subjects and patients with DGD.The distribution of the dense granule size was also evaluated to determine if the decrease in the dense granule volume in δ-MGSPD was due to a loss of the larger sized granules or whether the volume decrease was due to all the granules being smaller than normal. Patients with δ-MGSPD were found to have a population of platelet dense granules that were smaller than those of the control group. Although the volume of the largest dense granules in individuals with δ-MGSPD was similar to normal subjects, the percentage of larger granules was decreased. A scatter box plot of patients and control subjects is shown in Figure 4 to exemplify these differences; individuals with δ-MGSPD had a dense granule distribution that shifted to smaller sizes compared to normal subjects. Individuals with δ-MGSPD had 59.4% of their granules <1.5 × 106 nm3 in volume, whereas none of the granules of normal subjects were as small. Forty-three percent (43%) of the “classic” SPD patients had a dense granule volume (DGV) <1.5 × 106 nm3.In addition to our morphometric calculation of a total storage pool volume, we classified DGs using the uranaffin reaction described by Richards, et al. [33] and later employed by Weiss et al. (1993) to determine the distribution of the DG type [12]. Figure 2 shows that individuals classified as δ-MGSPD have a granule “type” distribution similar to that of normal individuals and this classification is not similar to that of empty granule/empty sac syndrome [12,40].Since the calculation of platelet dense granule volume was based on the presumption that platelet dense granules are spherical, we calculated their volume by standard formulas of solid geometry. To confirm the assumption, we examined the three-dimensional shape of DGs using HVEM tilt series of uranaffin-eacted epoxy sections (Figure 4). All the uranaffin-enhanced DGs were found to be spherical (excluding an occasional granule undergoing degranulation). Although some granules initially appeared to be rod-shaped or elliptical in morphology, the subsequent tilting of the sample section relative to the beam demonstrated that these granules were actually two or more granules that were superimposed with similar positions in the specimen plane, but were at different locations within the thickness of the sample. When the specimen was tilted to different orientations relative to the beam, two distinct particles with circular cross-sections were revealed. These data confirmed our hypothesis that platelet dense granules are indeed spherical and validated our volume calculations.Given the clinical similarity between the individuals with DGD and δ-MGSPD, and given the similarity in total DGV between these two disorders, we next determined the ADP and ATP concentrations in the platelets of individuals with δ-MGSPD. If δ-MGSPD is a bleeding disorder analogous to DGD, then the ADP and ATP contents of these platelets would be expected to be decreased. In the normal subjects, there was a very strong correlation between the ADP content and the TDGV (r = 0.91), suggesting that the ADP assay was primarily reflecting the ADP content of the dense granules (Table 2). The ADP and ATP contents of platelets from individuals with δ-MGSPD were decreased compared to the normal platelets. The ADP and ATP concentration in δ-MGSPD were similar to those seen in classic DGD (Table 2). For the normal subjects and individuals with δ-MGSPD, there was a strong correlation between the TDGV and the concentrations of ADP (r = 0.56), ATP and the ADP + ATP content (r = 0.70). Individuals with δ-MGSPD had ADP and ADP + ATP concentrations that were indistinguishable from those of individuals with dense granule deficiency and the two groups had similar TDGVs (Table 2).At the onset of our studies of patients having unexplained bleeding, we used the platelet whole mount technique for electron microscopy to determine whether these patients had a normal number of DG. We sought to establish that the normal subjects assayed in our laboratory had similar values to control subjects cited in the literature [13,23,24,25,29]. There are a number of assays that have been used to evaluate unexplained bleeding including lumiaggregometry, the measurement of 5-HT content, and mepacrine assays as other means for the diagnosis of δ-SPD [41,42,43,44,45,46,47]; however, the platelet whole mount TEM techniques remains the gold standard [41,48]. The term “gold standard” is used in other contexts as well, such as for the diagnosis of inherited platelet secretion disorders, whereas radioactive serotonin incorporation and release is cited as the gold standard for δ-granule release [49]. Since the technique was first described by Bull and refined by White, the technique has been reliably utilized for 50 years for objective morphologic, not functional, assessments [13,23,24,26]. The number of DG/PL that we have established as normal in our laboratory is 4.54 ± 0.29 and that the lower limit for a normal DG number, at three standard errors from the mean, is 3.68 DG/PL; these values are similar to those published in other studies of δ-SPD [13,23,24,26]. These findings are at odds with a single report suggesting that the normal range of DGs is significantly lower at 1.2-4.0 DG/PL, which is radically different than what the rest of the literature suggests. Those investigators utilized energy dispersive X-ray (EDX) analysis to categorize DGs based upon the identification of calcium and phosphorus, both elements that are known to concentrate in DGs [50]. The use of EDX for the determination of SPD is novel and an infrequently available technique in clinical electron microscopy (EM) laboratories.For those individuals found to have a normal to elevated number of DGs, we were able to determine that 165/581 (28.4%) had a mean DG diameter statistically smaller than that of our normal subject group. The adenine nucleotide pool was found to be very similar to the means of δ-SPD patients and significantly less than the control values (Table 2).Individuals having a bleeding tendency in association with a normal platelet count should be evaluated for von Willebrand disease. If normal, a qualitative platelet defect or δ-SPD should be considered. δ-SPD has historically been diagnosed using platelet aggregation assays, by the firefly luciferase adenine nucleotide assay, whole mount electron microscopy and more recently by flow cytometry using quinacrine (mepacrine)-labeled platelets [10,51,52]. We report the use of image analysis to assess the dense granule volume in patients with a significant bleeding history who appear to have a normal number of dense granules.Platelet dense granule or delta granule (DG) deficiency (δ-SPD) is a heterogeneous bleeding diathesis, often characterized by prolonged bleeding time (BT) or closure time, variable abnormalities in platelet aggregation, decreased adenine nucleotides and decreased DGs by EM [31,32,50,51,52,53,54,55,56,57]. In our experience, it is not infrequent to find that patients with a significant bleeding history may not be diagnosed with any particular disorder following platelet aggregation and release studies. Such patients are often labeled as being “VWD-like” and remain a diagnostic enigma presently labeled as “Bleeding of Unknown Cause” [58,59,60,61]. Further research is needed to determine whether the prevalence of δ-SPD and its microgranular variant (δ-MGSPD) is an even more common entity than VWD as a cause of unexplained mucocutaneous bleeding. The advent of whole exome sequencing should also better define such patients [62].The limitations of our study include a number of assessments that were not made such as age and sex correlation, ethnic background and family studies, the use of a bleeding assessment tool for the objective scoring of bleeding symptom severity, and the review of each patient’s complete blood cell count and prior hemostasis testing. Many of the patients included in this study had samples that had been referred from a number of other hospitals and health centers without accompanying medical history. Blood was also submitted in acid citrate dextrose (ACD) tubes, diluting some of the values obtained by CBC. As previously stated, we have reported that blood samples in ACD tubes are stable for at least 48 h post-venipuncture for the assessment of DGs and/or VWD profile testing [30]. The adenine nucleotide results included in this report, while limited, were critically scrutinized for accuracy with adequate quality control.The essential point to be concluded from this study is that δ-SPD may be a significant etiology for patients presenting with bleeding histories that are difficult to diagnose by traditional means. In our experience, it is apparent that δ-SPD can clinically mimic von Willebrand disease, as both disorders present with the same symptoms. Many of the patients that we have evaluated have previously been “labeled” as having a “von Willebrand-like syndrome” as they have a normal aggregation assays, normal von Willebrand antigens and activity, collagen-binding assays and factor VIII clotting factor activity. Previous studies have reported the disorder that cannot be excluded with simple aggregation tests [23,24,25]. There have been no population-based studies to determine the incidence of δ-SPD and the prevalence of inherited platelet disorders is unknown [63]. Unfortunately, a specific etiology of the disorder has yet to be established. The inheritance pattern is often described as autosomal dominant, but both autosomal recessive and acquired SPDs have been reported. Delta granule storage pool deficiency is well documented in Hermansky–Pudlak, Chediak–Higashi, and Wiskott–Aldrich syndromes, all of which have known mutations and none of which are common to one another. The etiology of δ-SPD appears to be multifactorial and it is often unrecognized, especially when extensive laboratory workup, including whole mount EM, fails to identify the cause of unexplained bleeding. The utilization of TEM cannot determine functional secretions defects and our group of patients with normal numbers of DGs and TDGV may in fact have a functional defect. The results of the present study demonstrate that variations of δ-SPD do exist. Weiss et al. [12] described a variant δ-SPD in which DGs were conspicuously absent but had numerous “empty” uranaffin reactive membrane limited vacuoles which was subsequently termed the Empty Sac Syndrome by McNicol et al. in 1994 [40]. We now report a new variant δ-SPD which we have termed “microgranular delta storage pool disease” (δ-MGSPD) in which prominent bleeding history is evident, aggregation studies may/may not be abnormal, and the wmTEM of the dense granule number is within normal limits, but the mean diameter of DGs is significantly less than that of normal subjects. Further scrutiny of the platelet DGs including image analysis for DG measurement and the determination of the adenine nucleotide content may be required to establish an etiology for some cases of undiagnosed bleeding disorders. It may even go undiagnosed following routine electron microscopic evaluation. These findings may also be important as a basis to investigate possible altered mechanism(s) of platelet formation/maturation and/or function that lead to the qualitative platelet dysfunction related to diminished dense granule constituents.The following are available online at https://www.mdpi.com/2077-0383/9/6/1734/s1, Supplementary Figure S1: Comparing AI vs. Manual Measurements of Index Cases, Supplementary Table S1: AI, Supplementary Table S2: Total Group DG Numbers and Volumes.All authors had significant contributions that are summarized as follows: The index cases were all patients of N.A.L. The case series was conceptualized by W.T.G.III, M.R.S. and N.A.L. Referral of patients by P.A.K. and N.A.L. Laboratory methodology and assessments were performed by E.P.C. and S.R. High-voltage electron microscopy was provided by J.N.T. Software development was provided by S.R. B.R. and commercial software utilization performed by E.P.C. and W.T.G.III Validation of software was made by S.R. and B.R. and utilization of artificial intelligence to validate software was performed by M.R. Formal analysis was performed by all authors. The original manuscript writing was by W.T.G.III with revisions, editing, and reviews provided by M.R.S., J.N.T., B.R., P.A.K. and N.A.L. All authors have read and agreed to the published version of the manuscript.This research received no external funding.The high-voltage electron microscope (HVEM) was partially supported by “The Biological Microscopy and Image Reconstruction Resource at the Wadsworth Center under a PHS Grant RR 01219 from NIH’s NCRR.The authors declare no conflict of interest.Normal platelets obtained from the peripheral blood. (A) Thin section of the epoxy-embedded platelets demonstrates a number of various organelles. Dense granules are not readily seen in the sections of platelets as not all will be within the plane of the section; the majority of the “dark” bodies are alpha granules. (B) A whole mounted and air-dried platelet includes all the granules within the cell. Dense granules appear opaque black due to the calcium content of these organelles that vary in both size and density. Many obscure gray shadows may be seen; these represent the alpha granules that are not electron dense and are in the number range of 50–80 per platelet.The uranaffin reaction demonstrates the four stages of dense granule development. (A) Epoxy-embedded platelets that reacted with uranyl acetate were categorized into 4 stages of development with Type 1 granules being fully developed; an empty vacuole (Type 4 granules) represents the beginning of the earliest phase of a dense granule. (B) This graph demonstrates the distribution of dense granule types for the control (normal subjects) and microgranular storage pool deficiency presented in this report. The data for our control and microgranular (delta) granule storage pool deficiency (δ-MGSPD) subjects are identical to the 4 stages of dense granule development in normal patients as described by Weiss and co-investigators [12]; the microgranular storage pool deficiency has a normal distribution of dense granule development in contrast to the empty sac/granule syndrome.Thick sections of a platelet observed in a tilt series using a high voltage TEM. This demonstrates the spherical nature of dense granules; overlapping dense granules (DGs) may appear oblong.The graph illustrates that the patients categorized as microgranular storage pool deficient (δ-MGSPD) are similar in total dense granule volume (TDGV) per platelet to classic dense granule deficiency (δ-SPD) when compared to the TDGV of control subjects, unexplained bleeding patients and those who were found to have elevated numbers of granules. This graph includes all the subjects included in the study.Demographics of the index patient cohort.Characterization of the studied subjects.Mann–Whitney Rank Sum: * p < 0.001, value is significantly different than the control value ** p < 0.05, value is increased compared to the control value.

Med-MDPI/jcm_2/jcm-09-06-01735.txt

@@ -0,0 +1 @@
+Twin anemia polycythemia sequence (TAPS) is a rare complication of monochorionic diamniotic (MCDA) twins. Middle cerebral artery peak systolic velocity (MCA-PSV) measurements are used to screen for TAPS while fetal or neonatal hemoglobin levels are required for definitive diagnosis. We sought to perform a systematic review of the efficacy of MCA-PSV in diagnosing TAPS. Search criteria were developed using relevant terms to query the Pubmed, Embase, and SCOPUS electronic databases. Publications reporting diagnostic characteristics of MCA-PSV measurements (i.e., sensitivity, specificity or receiver operator curves) were included. Each article was assessed for bias using the Quality Assessment of Diagnostic Accuracy Studies II (QUADAS II) tool. Results were assessed for uniformity to determine whether meta-analysis was feasible. Data were presented in tabular form. Among publications, five met the inclusion criteria. QUADAS II analysis revealed that four of the publications were highly likely to have bias in multiple areas. Meta-analysis was precluded by non-uniformity between definitions of TAPS by MCA-PSV and neonatal or fetal hemoglobin levels. High-quality prospective studies with consistent definitions and ultrasound surveillance protocols are still required to determine the efficacy of MCA-PSV in diagnosing TAPS. Other ultrasound findings (e.g., placenta echogenicity discordance) may augment Doppler studies.Twin Anemia-Polycythemia Sequence (TAPS) is a rare complication of monochorionic diamniotic (MCDA) twin pregnancies first reported following fetoscopic laser surgery (FLS) for twin to twin transfusion syndrome (TTTS) in 2006 [1]. TAPS is caused by chronic blood transfusion through shared placental vessels leading to substantial morbidity as the twins become increasingly anemic and plethoric. While the disease may follow FLS (referred to here as “post-laser TAPS” (plTAPS)) in 3–10% of MCDA pregnancies, spontaneous TAPS (sTAPS), first described in 2007 by Lopriore et al., may develop 2–5% of the time [2,3,4,5,6,7,8]. Middle cerebral artery peak systolic velocity (MCA-PSV), as measured during Doppler velocimetry studies, is used to screen for TAPS. Absolute criteria (i.e., MCA-PSV < 1.0 MoMs in the plethoric twin plus MCA-PSV > 1.5 MoMs in the anemic twin, MoM = multiples of the median) and relative criteria (e.g., ΔMCA-PSV > 0.373 or > 0.5 MoMs) have been proposed for diagnosis [5,9,10]. Definitive diagnosis is obtained by comparing twin serum hemoglobin (Hb) levels with or without reticulocyte count ratios either postnatally or by intrauterine umbilical cord blood sampling. Treatment modalities vary by gestational age (GA) and include expectant management, fetoscopic laser surgery, intrauterine transfusion (IUT) with or without partial exchange transfusion (PET), selective reduction and preterm delivery [11,12].Routine use of MCA Doppler velocimetry to detect TAPS is currently recommended by the International Society for Ultrasound in Obstetrics and Gynecology (ISUOG) [13]. However, in 2013 the Society for Maternal Fetal Medicine (SMFM) recommended against this practice stating that there is “no evidence that monitoring for TAPS with MCA PSV Doppler at any time, including >26 weeks, improves outcomes [14]”. Since 2013, TAPS has received considerable attention in the literature and the first randomized clinical trial for treatment of TAPS is currently underway [15]. Following these developments, there has been increased enthusiasm for screening protocols utilizing serial MCA Doppler velocimetry [16].MCA Doppler velocimetry is also the principle screening test for fetal anemia in the setting of maternal alloimmunization to fetal red blood cell antigens [17]. The use of MCA-PSV values to screen for fetal anemia from alloimmunization has been established following extensive research, but with a false positive rate of 10% still prevailing [18]. The method, which may be affected by a range of physiologic parameters, is less well studied in TAPS, which is a unique physiologic entity. Furthermore, MCA Doppler velocimetry may be less efficacious for detecting polycythemia than for anemia [19]. Finally, the definitive diagnosis of TAPS and verification of MCA Doppler velocimetry as a screening tool is confounded at early GAs by the inability to sample fetal Hb due to technical limitations and concern regarding perinatal risks. In this study, we sought to exhaustively review the literature to determine the diagnostic efficacy of MCA Doppler velocimetry in screening for TAPS.We performed a systematic review following PRISMA guidelines [20]. The research question of interest was to determine the sensitivity, specificity, positive predictive value, and negative predictive value of MCA Doppler velocimetry for antenatal diagnosis of TAPS. Eligible studies included those where MCA-PSV values were collected to screen for TAPS and fetal or neonatal serum Hb concentrations were also collected as the “gold standard” diagnostic test. Studies written in English during or after 2006 (i.e., when TAPS was first described in the literature) were considered [1]. Case reports, case series and literature that did not undergo peer review (i.e., abstracts, clinical commentary) were excluded. To locate eligible studies, a search strategy was developed and applied to the Pubmed, Embase and SCOPUS electronic databases. The following terms were queried in April of 2020: “twin anemia polycythemia sequence,” “TAPS” and “feto-fetal transfusion.” In the SCOPUS search, only publications in the “Medicine” and “Health Professions” subject areas were included. Abstracts for all publications retrieved by this search were reviewed by two of the authors (C.B. and E.B.). Publications that appeared as though they may include antenatal screening for TAPS based on abstract review were obtained for complete review. A secondary search (i.e., “by hand”) was performed by reviewing the bibliographies of these publications. The aim of this process was to generate a comprehensive list of primary studies suitable to answer the research question.Documents retrieved by the primary and secondary searches were reviewed in their entirety. Two independent reviewers (C.B. and E.B.) carefully examined each document and extracted relevant data including the author, PMID number, year of publication, journal and study type (i.e., prospective vs. retrospective). Studies that compared antenatal screening for TAPS by MCA-PSV values to fetal or neonatal serum Hb levels, including either sensitivity/specificity calculations or derivation of a receiver-operative curve (ROC) were considered for analysis. For these studies, we extracted study size, number of TAPS cases, number of plTAPS and sTAPS, stage and GA at TAPS diagnosis and the date range over which women were clinically evaluated. Outcome measures including the sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and area under the ROC were also collected. Uniformity of the studies was assessed for the possibility of performing meta-analysis. Because determination of fetal Hb levels may not be possible for pregnancies at earlier GAs, it was also noted whether calculations were stratified by GA. Any discrepancies in extracted data or study inclusion were arbitrated and resolved by a third author (R.P.).The Quality Assessment of Diagnostic Accuracy Studies II (QUADAS II) methodology was applied to the selected studies to determine their likelihood of bias and applicability in answering the research question [21]. QUADAS II includes a structured list of 11 “signaling questions” evaluating for bias in the following four domains: (1) patient selection, (2) collection and interpretation of an index test (i.e., MCA-PSV to screen for TAPS), (3) collection and interpretation of a reference standard (i.e., Hb levels to definitively diagnose TAPS), (4) flow/timing of the study. Questions may be answered “yes,” “no,” or “unclear.” Additional questions may be added at the users’ discretion to address possible biases unique to the research question. We added two questions: (1) “Was a clearly defined schedule described for timing of initial and follow-up application of the index test?” (2) “Were reference test values from repeat intrauterine transfusions or reference test values following birth in neonates who had a previous intrauterine transfusion used to calculate sensitivity/specificity?” MCA-PSV is usually measured serially as recommended by ISUOG; the first question was added because a missed diagnosis following inadequate antenatal screening may lead to underestimation of the test’s sensitivity. The second question was added for two reasons: (1) Hb values at the time of intervention, not postnatal Hb values, should be compared with MCA-PSV values at the time of TAPS diagnosis for fetuses that underwent transfusion. (2) The efficacy of MCA-PSV values in predicting fetal anemia decreases following intrauterine transfusion [22]. Based on answers to these 13 questions (see Table 1), the likelihood of bias in the four domains was assessed as “HIGH,” “LOW,” or “UNCLEAR.” Similarly, but without use of signaling questions, concerns about applicability of the studies to the four domains are assessed as “HIGH,” “LOW,” or “UNCLEAR.”The aim of this study was to collate and summarize the results of primary studies and, if possible, perform meta-analysis to estimate the pooled sensitivity and specificity of MCA-PSV values for diagnosing TAPS. Data were summarized in tabular form. Microsoft Excel (Redmond, WA) was used to tabulate search results, extracted data, and QUADAS II results.The primary search of the Pubmed, Embase, and SCOPUS databases produced 432 publications. Based on abstract review, 54 of these publications were selected to be reviewed in their entirety [1,2,4,5,7,9,10,11,12,16,19,23,24,25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51,52,53,54,55,56,57,58,59,60,61,62,63,64,65]. The secondary search of bibliographies produced three additional publications which were also reviewed in their entirety for a total 57 publications [6,66,67]. The publications are detailed in Table 1 and reasons for exclusion are summarized in Figure 1. Five of the 57 publications, all from the primary search, described a study of MCDA twins where serial ultrasounds with MCA-PSV values were collected to detect TAPS with direct comparison to Hb values [5,9,10,19,47]. After applying QUADAS II methodology to these five studies, four were determined highly likely to have bias in patient selection (Figure 2) [5,9,10,47]. Three were found highly likely to have bias in conduction or interpretation of the index test (i.e., MCA-PSV Doppler values) as well as conduction or interpretation of the standard reference (i.e., Hb levels) [5,10,47]. All of the studies were determined to have low likelihood of bias in flow and timing. The answers to the QUADAS II signaling questions provide rationale for these assessments of bias and are detailed in Table 2. Regarding patient selection, studies by Slaghekke et al. and Veujoz et al. only included women with a diagnosis of TAPS without any uncomplicated MCDA twins [9,47]. Hence, the number of true negatives and false positives, or women without TAPS, appear to have been based on women who were ultimately diagnosed with TAPS with MCA-PSV values and Hb levels collected prior to diagnosis or after treatment and resolution of the disease. Studies by De Sousa et al. and Tollenaar et al. excluded twins that were missing either MCA-PCV values or Hb levels [5,10]. As protocols for serial MCA-PSV analysis (or compliance with such a protocol) over the study periods were not available, the reasons these values were missing, and women excluded, could not be determined. One study appeared to include postnatal Hb levels to calculate sensitivity, specificity, PPV, and NPV in twins that had a prior IUT and another used MCA-PSV/Hb pairs from the time of repeat IUT for the same purpose [10,47]. None of the studies were found to have concerns regarding the applicability of patient selection, the index test, or the standard reference. Figure 2 Legend: Results of QUADAS II application. QUADAS II consists of four assessments of bias and three assessments of applicability to the study question. A judgement of “HIGH,” “LOW,” or “UNCLEAR” is made for each assessment of each study considered. A “HIGH” judgement means that the study methodology is likely to introduce bias (for bias assessments) or likely lacks applicability to the study question (for applicability assessments). A “LOW” means there is less likelihood of bias or lack of applicability. Answers to the “signaling questions” which support the bias judgements which are detailed in Table 3.Demographic characteristics of women from each of the selected studies are detailed in Table 3. The studies included 779 women and 122 cases of TAPS. There may be overlap in TAPS cases between studies by Tollenaar et al. and Slaghekke et al. as they are from the same institution (Leiden University, Netherlands) with overlapping study periods [10,47]. All but one study (Slaghekke et al.) reported that the incidence of TAPS ranged from 4.6% to 35.3% (sTAPS and plTAPS together). Tollenaar et al. did not report the incidence of plTAPS and sTAPS separately for all cases (see fotenote, Table 3) [10]. Among the remaining studies, the incidences of plTAPS (total n = 44) and sTAPS (total n = 43) were reported separately, but each type was included together in calculations of sensitivity, specificity, PPV and NPV. Among studies where plTAPS and sTAPS were delineated and incidence was reported, the combined incidence of sTAPS was 4.8% (31/643). The majority of TAPS was diagnosed after 24 weeks with delivery (mostly by Cesarean section) in the late 3rd trimester (i.e., >32 weeks). Only Veujoz et al. described the disease severity by stages [9]. Table 4 shows sensitivity, specificity, PPV, and NPV of MCA-PSV values in predicting the diagnosis of TAPS by differences in intertwin Hb levels. All studies limited analysis to patients where an MCA-PSV value was collected within one week of determining Hb levels by cordocentesis (antenatally) or venipuncture (postnatally). Studies by Tollenaar et al. and Fishel-Bartal et al. defined prenatal TAPS with absolute values (i.e., MCA-PSV > 1.5 MoMs and MCA-PSV < 1.0 MoMs) as well as the relative difference between values [10,19]. De Sousa et al. considered only relative values, and the remaining two studies used only absolute values [5,9,47]. Collectively, four different criteria for a positive MCA-PSV screening test were used between the five studies. None of the studies stratified calculations of the diagnostic characteristics of MCA-PSV values by GA or treatment received. The criteria for definitive diagnosis of TAPS was similarly heterogeneous. Each study included numerical cutoffs for diagnosis; however, three evaluated the difference in Hb levels between twins for diagnosis of TAPS, while the remaining two used individual Hb levels to diagnose anemia and polycythemia separately. Two of the studies also required the reticulocyte count ratio of the polycythemic twin to the anemic twin to be greater than 1.7 [9,10]. Collectively, the five studies had four different definitions of definitive TAPS. The sensitivities, specificities, PPVs, and NPVs were reported for four of the five studies, while Bartal et al. reported ROC characteristics [19]. Finally, none of the studies stratified calculations by GA or intrauterine treatment for any of the calculations.Heterogeneity between the studies precluded meta-analysis for pooled estimates of PPV, NPV, sensitivity, or specificity.This systematic review arrived at several important findings on the efficacy of MCA-PSV values in diagnosing TAPS: (1) Five studies (two prospective, three retrospective) were found that describe the diagnostic characteristics of MCA-PSV values. (2) plTAPS and sTAPS were considered together in calculations of diagnostic characteristics for all studies. (3) No study stratified calculations of the diagnostic characteristics by GA or treatment. (4) Diagnostic criteria (i.e., “cutoff values”) for the screening test and the diagnostic test were not consistently defined between studies. (5) Because of their retrospective nature, most of the studies likely exhibit bias in patient selection and conduction/interpretation of the index test (MCA-PSV values) and standard reference (Hb levels).The above findings suggest significant limitations in the diagnosis and management of TAPS. First, false positive and false negative diagnoses are of major concern. Currently, fetal intervention may be undertaken based on MCA Doppler velocimetry which appears to have PPVs ranging from 70% to 100%; however, these PPVs are estimated in retrospective studies with a high likelihood of bias, particularly in patient selection [5,10,47]. This is not surprising as each study includes data from before 2016, when ISUOG first recommended universal serial MCA Doppler velocimetry in MCDA twins [13]. Prospective data by Fishel-Bartal et al. suggest that absolute MCA-PSV values perform modestly in diagnosing TAPS (AUC = 0.687, 95% CI (0.547–0.827) for anemia and AUC = 0.617, 95% CI (0.505–0.728) for polycythemia) [19]. While the same study shows better performance for relative MCA-PSV values (i.e., ΔMCA-PSV, AUC = 0.871, 95% CI (0.757–0.985)), these data are yet to be corroborated in another high-quality prospective study. Furthermore, the incidence of sTAPS in this study (10.1%) is more than twice as high as reported elsewhere (1.2–4.9%, except for De Sousa et al. with an incidence of 9.1%) increasing the likelihood that PPV is overestimated [2,3,4,5,6,7]. Hence, the risk of intervention for TAPS when disease is not truly present may be substantial. As the stage of TAPS is only reported in one of the studies, the degree to which MCA-PSV values may over- or underestimate severe outcomes, which provide greater impetus for intervention, is also unclear. Of note, the work by Fishel-Bartal et al. highlights an emerging trend of favoring ΔMCA-PSV over absolute MCA-PSV values because of the former’s reported higher sensitivity [19]. Tollenaar et al. retrospectively observed that twins meeting ΔMCA-PSV criteria, but not absolute criteria, have similar postnatal outcomes to twins meeting absolute criteria. This led that group to propose a new staging system for TAPS based on ΔMCA-PSV [10]. Fishel-Bartal et al. suggest that the superiority of ΔMCA-PSV values may be related to the poor predictive ability of MCA-PSV < 1.0 MoM in diagnosing polycythemia; however, Slaghekke et al. report high sensitivity and specificity using this method (Table 4) [19,47]. Further prospective data are required to validate (or invalidate) MCA-PSV < 1.0 MoM and ΔMCA-PSV in diagnosing polycythemia and TAPS, respectively.The evaluation of diagnostic performance is further complicated by the inability to measure Hb levels when TAPS is diagnosed at an early GA or following FLS. The treatment algorithm published in Leiden recommends FLS prior to 28 weeks, preterm delivery after 32 weeks, and IUT with PET at intermediate GAs for Stage II or greater TAPS [42]. At GAs greater than 32 weeks, MCV-PSV values and Hb levels may be measured in close temporal proximity because delivery is likely to occur soon regardless of whether TAPS is diagnosed (Figure 3A). Thus, the incidence of all four possible outcomes relating MCA-PSV to Hb (true and false positives, true and false negatives) may be determined. These values are required to determine sensitivity, specificity, PPV, and NPV. For GAs of 28 to 32 weeks, Hb would only be measured when IUT/PET is performed in the setting of Stage II or greater TAPS. As such, true and false negatives cannot be determined (Figure 3B). It is possible that undiagnosed TAPS (i.e., false negatives) account for some portion of the fetal demise observed in otherwise normal appearing MCDA twins [68,69,70]. For GA less than 28 weeks, none of the four possible outcomes may be determined because Hb levels are not typically measured in the setting of FLS (Figure 3C). The diagnostic efficacy of MCA-PSV values cannot be determined if the gold standard diagnostic test (i.e., Hb levels) is unavailable. Using one or more surrogate markers (a so called “gold alloy”) for diagnosis, such as the “starry liver sign”, discordant placental echogenicity, high-output cardiac failure or polycythemic thrombosis (i.e., fetal limb necrosis, bowel infarction with perforation or intracranial hemorrhage) is one method to address such diagnostic quandaries; however, use of this approach was not encountered in our search of the literature [29,71,72].Following diagnosis, our understanding of the role for various intrauterine treatments is evolving. A recent multicenter study (n = 370) reports a high level of variance in the treatments used (i.e., FLS, IUT/PET, expectant management, early delivery and selective feticide) and the GAs at which they are employed. An advantage in pregnancy prolongation is reported for FLS, but comparative treatments were performed at later GAs [73]. Several smaller studies also address treatment (total n = 185); however, each combined plTAPS with sTAPS, had unsystematic methods of patient inclusion, and directly compared treatments performed across the full range of gestational ages [11,26,52]. Further work is required to provide a valid basis for the diagnosis and treatment of TAPS. sTAPS and plTAPS should be considered separately as data is sparse to suggest whether MCA-PSV values behave similarly in these two physiologically distinct scenarios. We are unaware of any data that separate the two entities as part of an evaluation of MCA-PSV as a screening test. Reliable false positive rates for MCA-PSV may help determine whether the risk of intervention is appropriate. At early GAs, surrogate markers for the disease will be required to determine estimates for the incidence of false positive diagnosis. Further work is also required to elucidate the benefits of intrauterine treatment. While grave outcomes such as neonatal death, ischemic loss of fetal limbs, and cerebellar disruption have been described, it is unclear whether a treatment protocol based on MCA-PSV values can prevent these outcomes [1,74,75,76].In summary, this systematic review of MCA-PSV values for diagnosis of TAPS reveals that studies supporting this methodology are dissimilar and at risk for bias. The resulting estimates of the PPV of MCA-PSV measurements may be unreliable, and thus lead to attempted treatment of normal twins, particularly in the setting of Stage II disease. Further work is required to reliably estimate the diagnostic performance of MCA-PSV values and determine the risks and benefits of intrauterine treatment of TAPS. Therefore, in the current practice with limited strong supporting data, it is pragmatic to continue surveillance with antenatal monitoring of MCA-PSV for MCDA twins, but to interpret the results with caution. Intervention should be more strongly considered when additional abnormalities are noted such as cardiac failure, abnormal Doppler flow in umbilical artery or ductus venosus, discordant placental echogenicity or hydrops. Even in the presence of normal MCA-PSV in MCDA twins, it is prudent to consider evaluating for other signs of TAPS during ultrasound surveillance. Until more rigorous, large, prospective and well-controlled studies are conducted, many questions related to screening for TAPS will remain unanswered.Conceptualization: C.O.B., E.P.B., R.P., E.H.-A., A.J., K.J.M.J.; Methodology: C.O.B., E.P.B., R.P.; Data Curation: C.O.B., E.P.B.; Formal Analysis: C.O.B., E.P.B., R.P., D.L.; Writing, Review, and Editing: C.O.B., E.P.B., R.P., E.H.-A., A.J., K.J.M.J., D.L.; Supervision: R.P., A.J., K.J.M.J. All authors have read and agreed to the published version of the manuscript.This research received no external funding.No funding was required for the completion of this manuscript.The authors report no conflict of interest.Flowchart of Study Inclusion.QUADAS II Systematic Assessment of Bias for Included Studies.TAPS Screening by Gestational Age (or Treatment Modality). Ability to determine true or false positives and negatives according to the availability of Hb levels (i.e., the gold standard) at (A) greater than 32 weeks, (B) 28–32 weeks, and (C) less than 28 weeks gestational age. Values shaded in red or green can be determined whereas values shaded in grey cannot be determined.Studies Select for Full Review for MCA-PSV diagnosis of TAPS.Abbreviations: Hb = hemoglobin concentration, MCA-PSV = middle cerebral artery peak systolic velocity.QUADAS II Signaling Questions for Included Studies.* A retrospective study design was applied without explanation of a standardized protocol for the timing of either the index test (MCA-PSV) or reference standard (Hb) during the retrospective period. Patients were not included when MCA-PSV (73/256, 29%) or Hb data (3/256, 1%) were not available. Patients were also not included if the interval between MCA-PSV and Hb data was greater than one week. ƚ These studies were limited to patients with TAPS (i.e., not consecutive or random MCDA twins). Sensitivity and specificity were calculated using Hb and MCA-PSV values at times when patients met, versus did not meet, the criteria for diagnosis (i.e., multiple MCA-PSV, Hb pairs per patient). Slaghekke et al. did not include patients with missing data for Hb or MCA-PSV. ǂ A retrospective study design was applied without explanation of a standardized protocol for the timing of either the index test (MCA-PSV) or reference standard (Hb) during the retrospective period. Patients were excluded when MCA-PSV (221/351, 63%) or Hb (50/351, 14%) data were not available. Patients were also excluded if the interval between MCA-PSV and Hb data was greater than one week. § These studies did not describe a schedule for application of MCA-PSV measurement (i.e., patients considered, starting gestational age and frequency of follow up) or report ultrasound findings that might have prompted more frequent Doppler studies. ¶ Sensitivity and specificity calculations were performed using “postnatal intertwin Hb difference,” however 10 cases of intrauterine transfusion were reported. # Sensitivity and specificity calculations were performed using MCA-PSV/Hb pairs at the time of repeat IUT.Demographic characteristics of women from the selected studies.Abbreviations: FLS = fetoscopic laser surgery, pl = post-laser, Hb = hemoglobin concentration, IUT = intrauterine transfusion MCA-PSV = middle cerebral artery peak systolic velocity, ΔMCA-PSV = difference in MCA-PSV between twins, MoM = multiples of the median, PET = partial exchange transfusion, s = spontaneous, TAPS = twin anemia polycythemia sequence, wk = week. * Of 35 TAPS cases, type was specified for 29: There were 7 sTAPS and 22 plTAPS.Diagnostic characteristics of TAPS by MCA-PSV compared with Hb difference.Abbreviations: AUC = area under the curve, Hb = hemoglobin concentration, MCA-PSV = middle cerebral artery peak systolic velocity, ΔMCA-PSV = difference in MCA-PSV between twins, MoM = multiples of the median, TAPS = twin anemia polycythemia sequence, wk = week.

Med-MDPI/jcm_2/jcm-09-06-01736.txt

@@ -0,0 +1 @@
+Previous studies have found gender differences in body-related attentional bias (AB), with women showing AB towards weight-related body parts. However, few studies have assessed the relationship between body-related AB and muscularity dissatisfaction (MD) in men. This study aimed to assess the presence of muscle-related AB in men, using a combination of a virtual reality (VR) embodiment-based technique and eye-tracking (ET) technology. Twenty men with high MD and 20 with low MD, owned a virtual avatar that had the same silhouette and body mass index as the participant. To analyze the gaze data, muscle-related areas of interest (M-AOIs) and nonmuscle-related areas of interest (NM-AOIs) were defined. The complete fixation time and the number of fixations on each AOI were recorded. Mixed between (group)-within (AOI_condition) analyses of variance showed a statistically significant interaction between group and time (p < 0.05) in both AB measures. Follow-up analyses revealed an AB towards M-AOIs only in men with high MD. Overall, men with high MD spent more time looking and displayed a higher number of fixations on M-AOIs, specifically the chest and shoulders, compared to men with low MD. This study provides new information about the relationship between MD and body-related AB in men. Combining VR with ET technologies presents interesting opportunities in the study of body image in men.Body image studies have been typically conducted in women, with fewer studies focusing on men. In fact, research on male body image is considered a relatively recent field of study [1], with important gaps, such as those in research on male body image and eating disorders [2]. Body image has been described as a multidimensional construct reflecting the mental representation a person has of their physical appearance [3], which includes perceptual, cognitive, attitudinal, and affective components [4]. Individuals with body image dissatisfaction usually experience negative and dysfunctional cognitions and emotions (sadness, anger, or disgust) related to the way they perceive, evaluate, and feel about their own body [1,5]. Among adult men, body image dissatisfaction is common phenomenon, with a prevalence ranging from 9% to 28% [6] and 8 to 64% [7] in the US population.There is consensus that body dissatisfaction in men is mainly associated with two central factors, muscularity and body fat [8,9]. For instance, the ideal male body usually presented in Western media is that of a lean muscular body with well-developed chest, arms, shoulders, and legs [10,11], with the internalization of the ideal muscular body as a mediating factor for concerns associated with both muscularity and body fat [12]. Previous studies have found that the pathological drive for muscularity and body weight concerns among men are related to greater depressive symptomatology, low self-esteem, more physical exercise, and abuse of anabolic-androgenic steroids [10,13,14]. Likewise, persistent and intense concerns about the lack of muscularity and/or excessive body fat might lead to body image disorders, such as muscle dysmorphia [15].According to the DSM-5 [16], individuals with muscle dysmorphia usually experience a pathological preoccupation with the idea that his or her body build is too small or insufficiently muscular. Other studies have suggested that individuals with a high risk of developing muscle dysmorphia might share some of the psychopathological symptoms of individuals with anorexia nervosa (AN), such as a negative affective state, body image disorders, more physical exercise, and significant avoidance or negative body checking strategies [17,18]. For instance, previous studies suggest that young men with symptoms of muscle dysmorphia reported similar body-checking behaviors than individuals with eating disorders [19]. This suggests an implicit tendency to pay more attention to certain types of information (e.g., disorder-relevant information) over others, which is also referred to as attentional bias (AB) [20]. Dysfunctional body-related AB presumably maintains body dissatisfaction by processing negative information that is consistent with dysfunctional cognitive schemas (such as I am getting a fatter belly or my arms are not muscular enough), while schema-inconsistent information is not noticed or processed equally, usually being neglected visually [21]. Accordingly, these dysfunctional cognitive schemas and body-related AB elicit a negative affective state and compensatory or avoidance behaviors [20].Previous studies have found that women with eating disorders (ED) and individuals with high levels of body dissatisfaction show AB towards “disorder-salient stimuli,” including thin bodies, weight-related body areas, or self-reported unattractive body areas [22,23,24,25].According to those studies that assess body-related AB among men, two main paradigms have been proposed. In the first paradigm, different competing stimuli (usually bodies with different shapes) are presented at the same time. Previous studies using well-established AB methods, such as the dot probe task and eye tracking (ET) technology, have observed specific body-related AB towards idealized thin and muscular bodies over other types of body sizes or neutral images in men with high body dissatisfaction [26,27]. Similar results were found among men at high risk of developing muscle dysmorphia after conducting a visual probe task [28].In the second paradigm, participants are presented single images of their own or other types of bodies, during which attentional patterns towards specific body areas are measured. Two previous studies conducted with ET technology found preliminary evidence of AB towards weight-related body areas among men with a high drive for thinness [29] and AB towards upper body areas (e.g., chest, shoulders, and abdomen) in male college students when exposed to photographs of attractive males [30]. Comparing self-reported attractive vs. unattractive body parts, Cordes, Vocks, Düsing, Bauer, and Waldorf [31] reported that male weight trainers showed greater AB, as assessed by ET technology, towards their self-reported attractive body areas when exposed to pictures of a muscular athletic body. Additionally, a recent ET study found that men with muscle dysmorphia showed AB towards their self-reported unattractive body parts when looking at pictures of their own bodies [32]. In summary, results on body-related AB in men indicate that there was AB towards muscular bodies when bodies with different shapes were presented. There is also evidence suggesting that men with muscle dysmorphia show AB towards self-reported unattractive body parts [32].In the field of body image and body-related AB, several methodologies have been used to measure different attentional processes. Some of these methodologies are based on reaction-time measures, such as the Stroop tasks, dot-probe tasks, and visual search tasks. The Stroop tasks are usually employed to measure attentional maintenance and avoidance, while the dot-probe or visual search tasks are used to distinguish between attentional mechanisms (e.g., attentional maintenance and avoidance mechanisms) [33]. However, some of these reaction time measures might not be equally useful in assessing change or dynamic attentional patterns during a certain amount of time [33,34]. To do so, ET technology has been recommended, which can provide a continuous and dynamic measure of AB. In fact, all the studies mentioned above (e.g., [29,31,32]) used fixed ET devices.Although ET technology is a useful tool for assessing body-related AB in healthy and clinical samples with eating disorders [35], it also presents some important limitations, such as the lack of external validity [36]. The use of virtual reality (VR) technology may provide interesting new insights into the assessment of body-related AB, as well as help to overcome this drawback, by adding ET devices into a head-mounted display (HMD). For instance, it provides researchers with an accurate, objective and real-time measurement of an individual’s gaze patterns while participants are fully immersed in a VR scenario that represents a real-life situation [37]. In addition, based on the rubber hand illusion paradigm [38], the use of an embodiment-based technique induces the illusion of ownership of a virtual body (i.e., participants regard the virtual body as their own). This avatar may even simulate a real-size 3D simulation of the body of an individual with their specific physical features [39]. Finally, the use of VR embodiment-based techniques has already shown promising results in changing dysfunctional body representations in individuals with AN [40,41] and in healthy participants [42,43,44], in a paradigm recently referred to as embodied medicine [45].A recent study assessed gender differences in body-related AB using an ET and VR embodiment-based technique. Participants were first exposed to an avatar with the same measurements as their own bodies, then to a larger-sized virtual body before being exposed to the first virtual body [46]. The authors found that women showed AB towards weight-related body parts, while men showed AB towards muscle-related body parts. Body dissatisfaction levels among men and women did not significantly affect the results [46].The current study aimed to provide further information about the relationship between body-related attention and muscularity dissatisfaction (MD) among men. Specifically, body-related attention towards particular muscle- or nonmuscle-related body parts were assessed in men with high vs. low levels of MD who were exposed to a virtual avatar that had their real silhouette and body mass index (BMI).To the best of our knowledge, this is the first study assessing body-related AB and comparing muscle- to nonmuscle-related areas of interest (AOIs) in men, in contrast to other studies using different methodologies to compare groups of AOIs (e.g., self-reported unattractive vs. attractive AOIs). Additionally, this study presents an innovative assessment procedure in which VR and ET technologies were combined to assess body-related AB in men. Based on the previous research [32,46], it was expected that all men, regardless of their MD levels, would show AB towards muscle-related body areas. Furthermore, men with high MD were expected to show greater AB than those with low MD.Forty male college students from the University of Barcelona, comprising 20 men with low MD (Mage = 24.65, SD = 3.94, MBMI = 24.84, SD = 1.98) and 20 men with high MD (Mage = 24.05, SD = 3.31, MBMI = 23.71, SD = 2.30), participated in the study. All participants were voluntarily recruited through campus flyers or by direct contact. The exclusion criteria were: a BMI of less than 17 (moderate thinness) or more than 30 (obesity; according to the World Health Organization [47]), a self-reported diagnosis of current muscle dysmorphia or an eating disorder, and other self-reported current severe mental disorders (e.g., schizophrenia or bipolar disorder). Visual deficits that impeded VR exposure, epilepsy, and clinical cardiac arrhythmia were also criteria for exclusion.In accordance with the body items loading for muscularity of the Male Body Attitudes Scale (MBAS; 8) and the Male Body Checking Questionnaire (MBCQ) [48], the same AOIs were individually drawn onto a 2D frontal view image of the male avatar and were labeled as muscle-related body parts (M-AOIs), i.e., the chest, arms, shoulders, abdomen, and lower legs. The remaining body areas were labeled as nonmuscle-related body parts (NM-AOIs), i.e., head, neck, hands, stomach, hips, waist, thighs, and feet (Figure 1). The same M-AOI and NM-AOI labels were used in a previous study conducted by our group [46].The term fixation is defined as the act of sustaining one’s gaze on a single location for a minimum amount of time, typically 100–200 ms [49]. The following measures were used to assess AB: Number of fixations on AOIs: Number of fixations on a specified group of AOIs (M-AOIs or NM-AOIs).Complete fixation time on AOIs: Sum of the durations of the fixations on a specified group of AOIs (M-AOIs or NM-AOIs) in milliseconds.Number of fixations on AOIs: Number of fixations on a specified group of AOIs (M-AOIs or NM-AOIs).Complete fixation time on AOIs: Sum of the durations of the fixations on a specified group of AOIs (M-AOIs or NM-AOIs) in milliseconds.Male Body Attitudes Scale [8] is a self-report questionnaire that assesses male body image. It includes 3 scales (muscularity dissatisfaction, pursuit of low body fat and height) and 24 items that are scored on a 6-point Likert scale, ranging from 1 (never) to 6 (always). After a confirmatory factor analysis, the Spanish version of this questionnaire was developed, which includes 22 items after excluding the 2 items of the height scale [50]. For the purposes of this study, only the muscularity subscale (MBAS-M) was administered, which contains 10 items. Higher average scores indicate greater MD, with scores ranging from 1 to 6. Both the original and Spanish version of this questionnaire presents robust validity indices and good reliability, with Cronbach’s alpha ranging from 0.85 to 0.90 [50]. Cronbach’s alpha for the MBAS-M scale in the current study was 0.87.All participants were exposed to an immersive virtual scenario through a VR head-mounted display (HMD HTC VIVE-Pro, HTC Corporation, New Taipei City, Taiwan). In addition, two HTC VIVE-Pro controllers and three additional body trackers were used to achieve full-body motion tracking. VR body trackers allow the movement of objects or people within the virtual environment with high fidelity. HMD HTC VIVE-Pro has dual OLED panels at a resolution of 1440 × 1600 pixels per eye (2880 × 1600 combined), with a refresh rate of 90 Hz. It also has 110° vision and is easily adaptable to facial features. The entire VR equipment was connected to a VR-ready computer with a powerful processor (Intel Core i7-8700k with 3.70 GHz, Intel Corporation, Santa Clara, CA, USA) and a graphic card (Nvidia RTX 2080, Nvidia Corporation, Santa Clara, CA, USA) to run the VR environments.The VR HMD FOVE Eye Tracking (FOVE, Inc., Torrance, CA, USA) was used to detect and register eye movements. The headset uses incorporated position and orientation eye-tracking systems. The FOVE display has a resolution of 2560 × 1440 pixels and creates 70 frames per second. Infrared eye tracking sensors create 120 frames per second, with an accuracy level of less than 1°.FOVE Setup v0.16.0 (FOVE, Inc., Torrance, CA, USA) and Unity 3D 2018 v2.10 (Unity Technologies, San Francisco, CA, USA) were used to create virtual simulations. Unity 3D was used to develop the object-oriented programming code and integrate the elements within a virtual environment. Virtual avatars were created using the software Blender 2.78 and Gimp for improved texturing. The virtual environment was a simple room without any furnishings, but with a large mirror on the front wall that was placed 1.5 m in front of the avatar. The room also had a slightly open door to avoid possible feelings of being locked in. The avatar represented a young male, wearing a tight white t-shirt, blue jeans, and a swimming cap to avoid any influence of hairstyle.This study was approved by the ethics committee of the University of Barcelona. Participants had to provide informed consent before entering the study. Each participant was informed of the procedure and about data confidentiality and was told that they could withdraw from the study at any point without consequences. Confidentiality was ensured by assigning a different identification code.Each participant was weighed and measured to calculate their BMI. Additionally, the researchers interviewed the participants to ensure that they did not meet the exclusion criteria.To create the avatar, a frontal and lateral photo of the whole body of the participant was taken using a camera. Each participant was placed at a fixed position 2 m from the camera and was told to stand still with their arms slightly raised and legs slightly apart. The BMI of the participant was manually inputted into the software, while the body size of the avatar was automatically adjusted based on the BMI value. The participant’s photo and virtual silhouette were then matched by manually adjusting (e.g., increasing or decreasing) the different parts of the virtual silhouette (shoulders, arms, chest, waist, stomach, hip, thighs, and legs) to fit the silhouette of the participant (for an illustrative example, see Figure 2). In the meantime, the Muscularity subscale (MBAS-M) of the MBAS was administered.Participants were exposed to the VR environment through the HMD HTC VIVE-Pro. Once inside the virtual environment, each participant was able to observe himself in the first-person perspective and to look at himself in the mirror (in the third-person perspective). Full-body illusion (FBI) with the virtual body was induced using visual-motor and visual–tactile stimulation (for an illustration, see Figure 2). The visual-motor stimulation procedure was adapted from study of Waltemate, Gall, Roth, Botsch, and Latoschik [51] and consisted of synchronizing the movement of the participant with that of the avatar using motion capture sensors placed on the hands, feet, and waist. The visual–tactile stimulation procedure consisted of synchronizing visual and tactile stimulations. While the different areas of the body (upper and lower limbs and stomach) were touched on the participant, each participant observed the same areas being touched on the avatar at the same time, which was performed by a virtual controller. The visual–tactile procedure was adapted from previous studies on embodiment involving the abdomen [40] or the whole body [46]. Both procedures lasted for 3 min.Finally, to assess body-related AB, the HMD was replaced with the VR HMD FOVE Eye Tracking headset. The virtual room displayed on the HMD FOVE was the same as that in the previous VR environment, with the participant’s real-size avatar reflected in the mirror.First, the accuracy of the ET recordings was measured using a 9-point calibration procedure. The individual’s gaze was then tracked while they were asked to observe their real-size virtual body in the mirror for 30 s, a similar recording time to that used in previous studies [23,24]. Throughout this process, the participant was instructed to avoid abrupt head movements. During the process and as a cover story, the participant was told to remain still while the position of the avatar was being recalibrated.The OGAMA (Open Gaze and Mouse Analyzer) software was used to transform the ET raw data into suitable quantitative data. Complete fixation time and the number of fixations on the AOIs were calculated for each participant. This was achieved by summing up the separate complete fixation times and the number of fixations on the M-AOIs vs. on the NM-AOIs. An additional data transformation was conducted by calculating the difference between M-AOIs and NM-AOIs (e.g., for the complete fixation time, 2510 M-AOIs − 2110 NM-AOIs = 400). Therefore, a positive outcome indicated that the participant had been looking more at the muscle-related body parts than at the nonmuscle-related body parts, while a negative outcome indicated the opposite.Statistical analyses were conducted with the statistical software IBM SPSS Statistics v.24. An a priori alpha level of 0.05 was considered for the study.All the participants were divided into two groups: one with high MD and the other with low MD. Since the original test did not provide a standardized cut-off score for dividing the groups [8], and since other methods for splitting the sample (e.g., recruiting the top and bottom thirds) would have notably reduced the sample size, it was decided to use the median score of the MBAS-M (Me−MBAS-M = 3) as a cut-off score for this study.Mixed between (group)-within (M-vs-NM AOIs condition) analyses of variance (ANOVA) were conducted with both body-related AB measures (complete fixation time and number of fixations).Furthermore, independent-sample t tests were run to determine if there were significant differences in body-related AB between men with high and low levels of MD. These analyses were conducted for the groups of AOIs (e.g., difference between M-AOIs and NM-AOIs) and for single M-AOIs (e.g., arms, shoulders, chest, etc.).Finally, Pearson’s correlations analyses were conducted to assess the relationship between BMI and body-related AB measures separately for men with high and low MD.All assumptions required for two-way mixed ANOVAs were met. There were no outliers in the data, as determined by the inspection of a boxplot. Scores for both ET measures for each group were normally distributed as determined by the Shapiro–Wilk test (p > 0.05), and there was homogeneity of variances as assessed by Levene’s test for equality of variances (p > 0.05). Regarding the assumptions required for independent-sample t tests, nonparametric Mann–Whitney U tests were used in the analyses conducted with the single M-AOIs, since data for some of the variables were not normally distributed.The mean MBAS-M score was 2.23 (SD = 0.39) among men with low MD and 3.66 (SD = 0.60) among those with high MD. Group differences in age, BMI, and muscularity dissatisfaction levels were assessed using independent t test analyses. Results showed that the groups did not differ significantly on measures such as age and BMI (p > 0.05), but did differ significantly in muscularity dissatisfaction levels (p < 0.001). The mean and standard deviations of the two ET measures are shown in Table 1.Two-way mixed ANOVA showed statistically significant interactions between the group and AOI_condition on complete fixation time (F (1, 38) = 8953, p = 0.005, partial η2 = 0.191), and number of fixations (F (1, 38) = 6275, p = 0.017, partial η2 = 0.142). Follow-up analyses were conducted, and mean differences (MD) ± standard error (SE) are specified. These analyses showed that men with high MD spent significantly more time looking at M-AOIs and presented a significantly higher number of fixations on M-AOIs than on NM-AOIs (complete fixation time, MD = 5.750 ± SE = 1.923, p = 0.005, d = 0.65 and number of fixations, MD = 9.50 ± SE = 2.40, p < 0.001, d = 0.82). On the other hand, men with low MD did not show significant differences in AB between M-AOIs and NM-AOIs (complete fixation time, MD = 2.387 ± SE = 1.923, p = 0.222, d = 0.28 and number of fixations, MD = 1.00 ± SE = 2.40, p = 0.679, d = 0.10). Overall, they spent a similar amount of time on M-AOIs and NM-AOIs and presented an equal number of fixations on the two types of areas.Furthermore, considering differences between M-AOIs and NM-AOIs, men with high MD spent more time looking at M-AOIs and displayed a higher number of fixations on M-AOIs compared to individuals with low MD (see Figure 3). Independent-sample t tests revealed that these group differences were significant (complete fixation time, t (38) = 2.992, p < 0.01, d = 0.95 and number of fixations, t (38) = 2.005, p = 0.01, d = 0.79).Mann–Whitney U tests were run to assess group differences in AB measures in each single M-AOIs. Distributions of the complete fixation time and the number of fixations were similar for men with high and low MD, as determined by visual inspection. The median for complete fixation time and number of fixations was significantly higher in men with high MD than in those with low MD for the chest (complete fixation time, U = 302, z = 2.769, p = 0.005 and number of fixations, U = 301, z = 2.741, p = 0.006) and shoulders (complete fixation time, U = 272, z = −2.004, p = 0.045), using an exact sampling distribution for U. Overall, men with high MD spent more time looking at the upper body areas (arms, shoulders, and chest), while both groups spent an equal amount of time looking at the abdomen and the lower legs (see Figure 4).Finally, Pearson’s correlations analyses were conducted to assess the relationship between BMI and body-related AB measures separately for each group. There were no significant relationships between BMI and complete fixation time (p > 0.05) either in men with high MD or in men with low MD. In addition, there was no significant relationship (p > 0.05) between BMI and number of fixations in men with low MD. However, there was a statistically significant, moderately negative correlation between BMI and number of fixations (r (18) = −0.50, p = 0.024) in men with high MD, with BMI explaining 25% of the variation in number of fixations. This means that, among men with high MD, those with higher BMI values looked at M-AOIs less frequently.This study aimed to assess body-related AB towards specific muscle-related or nonmuscle-related body parts in men with high vs. low MD. We also present a new body-related AB assessment methodology that involves VR and ET technologies. In contrast to previous studies measuring gaze behavioral patterns in individuals looking at a picture or a photograph of their own body, we exposed participants to a virtual avatar with their real silhouette and BMI in a virtual room with a large mirror in front of them. As expected, men with high MD spent more time and looked more frequently at muscle-related body areas, especially the shoulders and chest, than those with low MD. Additionally, AB towards specific muscle-related body areas was found only among individuals with high MD.Consistent with the findings of Porras-Garcia, Ferrer-Garcia, Ghita et al. [46] that there is AB towards muscle-related body parts in men, our results provide further evidence about the relationship between body-related AB and MD. Although both studies used a similar assessment procedure with VR embodiment-based techniques and ET devices, there are some key differences between the studies. The current study found significant differences in body-related AB between men with high and low MD, which is in contrast to the previous study that did not find such AB differences between men with high and low body dissatisfaction. Although body dissatisfaction is certainly associated with muscularity in men [9], there are other body factors (e.g., concerns with body fat or slimness [9,10,11]) that might also influence male body image. Consequently, differences in body-related AB towards muscle-related body areas might be easier to detect between men with high vs. low MD than between men with high vs. low body dissatisfaction. Another important difference between the two studies was the technical improvements made to the software used in this study, such as developing a more precise method to create the avatar (by taking a frontal and lateral photo and incorporating the information of the BMI of each participant) and improving the full-body tracking system to conduct visual-motor stimulation.Our results also support those of previous studies by providing further evidence of AB towards muscle-related body parts in men with a high drive for muscularity [28,32]. However, there are several methodological differences between the current study and others in this field. For instance, as described before, some of the studies on men with high body dissatisfaction [26] or at high risk of developing muscle dysmorphia [28] assessed AB for competing stimuli, such as different body shapes (e.g., muscular bodies vs. normal or overweight bodies), while others (e.g., [31,32]) used a similar free-viewing single-body paradigm but measured AB towards self-reported attractive vs. unattractive body areas. Although these previous studies used a successful and well-established methodology to define areas of interest [21,35], evidence for AB towards specific muscle-related body areas could not be clearly established, since there were muscle-related body areas in both the self-reported attractive and unattractive body parts [31,32]. In this study and in two previous studies, a different methodology was used to define areas of interest, in which an individual’s gaze behavior was analyzed using the same definition of areas of interest for all participants (e.g., weight- vs. nonweight-related body areas [46,50]). These areas of interest were defined based on well-established questionnaires that assess the same body image construct, such as the Muscularity subscale of the MBAS [8] and the MBCQ [48].In men with low MD, differences in AB measures towards muscle-related and nonmuscle-related body parts did not reach significance, suggesting that they generally scanned over the whole body compared to men with high MD. This is in line with previous research reporting a similar behavior among healthy women [22,23,25] and healthy men [28,46].Regarding specific muscle-related body areas, we observed that, regardless of the MD group, all men paid more attention to the chest and abdomen. These results support previous findings reported by Cordes et al. [31] in which a similar attentional behavior was reported among men towards the same two body areas. Moreover, there were significant differences in AB towards the chest between men with high and low MD for both AB measures, while the same was observed for AB towards the shoulders, but only for the complete fixation time. This suggests that these two body areas may be particularly salient for men with high MD. Surprisingly, there were no significant group differences for other muscle-related body areas such as the abdomen, arms and lower legs, indicating that the two groups spent similar amounts of time looking at these specific body parts. This could be because these three body areas might be related not only to muscularity but also to body fat. As mentioned before, male body dissatisfaction is associated with two central factors, muscularity and body fat concerns [8,9]. Thus, paying attention to these body areas might not necessarily imply AB towards muscularity, but might be influenced by other body image concerns as well. The high prevalence of body dissatisfaction among healthy men [6,7] might also explain why men with high and low MD displayed similar gaze patterns towards these specific body areas. Finally, preliminary evidence was found suggesting that, among men with high MD, those with higher BMI values looked at M-AOIs less frequently. Even though these results were found in only one of the two AB measures (number of fixations), they may suggest a distinct visual attentional pattern toward the body between men with high MD and high BMI compared with men with high MD but within a normal BMI range. However, the results of the current study do not allow us to state conclusively that men with high MD and BMI pay more attention to weight-related body areas, as has been found among women with high BMI values [52]. Future studies are needed to confirm and extend these preliminary findings.This study had some limitations. Several key variables in male body image and muscularity, such as body fat concerns [8], regular physical activity, self-esteem, and the use of anabolic-androgenic steroids and related substances [10], were not properly controlled in this study. These variables can influence concerns on muscularity and, thus, might also affect body-related AB. Furthermore, no screening questionnaires or structured clinical interviews were used to properly assess the presence of muscle dysmorphia, eating disorders, or other mental disorders. Another limitation is related to the BMI formula, since its ability to account for body mass composition (i.e., lean body mass or muscle mass) has been regarded as limited: using the fat free mass index (FFMI) instead of BMI would have provided a better measure for muscle mass [53]. However, certain reliable and accurate fat free mass index measures (e.g., dual-energy X-ray absorptiometry [54]) were impractical for use in the current study. The use of other anthropometric measures such as triceps skinfold or abdominal circumference would also have been more invasive for our participants, since we would have had to literally touch certain body parts to perform the measurements; this might have increased distress levels among our participants and might have affected the results obtained on the VR-ET task. Finally, even though significant group differences were found between men with high and low MD, the use of the median as an “artificial categorization” of a continuous variable might imply some limitations that should be considered: for instance, in terms of the statistical power and accuracy of the relations estimated and in terms of the reduction of the relations observed between variables [55]. On the other hand, other researchers have suggested that the costs of artificial categorization using the median in terms of power are relatively small and are more than offset by the gains that it provides in terms of interpretability [56].Regarding the VR software, some limitations should also be considered. Despite the avatar sharing the same silhouette and BMI as the participant, the general appearance of the virtual body (e.g., clothes, skin color, etc.) was clearly different. Therefore, it is debatable whether participants truly showed an AB toward their own body, as was the intention, or whether they considered the virtual body as different from their own. However, this limitation is shared by all studies that use 3D computer-generated bodies to assess body image disturbances or body-related AB. In fact, a recent study found that computer-generated bodies yield poorer discrimination and increase errors in body perception research [57]. A possible way to overcome this limitation has been presented in recent VR studies, which allow the reproduction of an exact 3D biometric avatar with all the individual’s features [58]. Accordingly, the use of 3D biometric avatars might notably improve studies on body image and body-related AB that use VR. Finally, the fact that two different VR HMDs were used during the whole assessment task (HTC-VIVE-PRO for inducing FBI and HMD FOVE-ET for measuring gaze) might have reduced FBI levels when participants had to change from one VR HMD to the other. This limitation could have been overcome by using the new generation of VR HMD with advanced ET devices (e.g., HTC VIVE-Pro Eye), which has a complete full-body tracking system and can be used to elicit FBI over the virtual body and measure gaze patterns.This study might have noteworthy implications for future research and clinical practice involving body image in men. Combining VR and ET technologies to assess AB (e.g., towards body, food, or other stimuli) might lead to new possibilities in the upcoming years. For instance, the use of immersive VR environments might enable the reproduction of real-life environments, such as gyms and dressing rooms, to assess individuals’ gaze patterns towards their own bodies and/or towards other avatars with different body sizes. Additionally, both technologies could be used to implement interventions based on retraining dysfunctional body-related AB. Some studies have already attempted to retrain AB in women with high body dissatisfaction [59,60,61] or in individuals with eating disorders [62], in an approach referred as attentional bias modification training [62].Considering the similarities between muscle dysmorphia and some eating disorders, not only in their etiology [17,18] but also in their treatment approaches [63], a similar AB modification approach could be also used in men with high MD and/or with muscle dysmorphia diagnosis and as a complement of cognitive behavioral body exposure therapies, for instance, by systematically retraining individuals with MD and/or with muscle dysmorphia to attend equally at their muscular and nonmuscular related body areas. This sort of intervention may be suitable to reduce not only dysfunctional attentional patterns toward the body but also body dissatisfaction, as has been previously found in women with high body dissatisfaction [59,60,61].In summary, the current study provides further evidence about the relationship between body-related attention and MD in men. Our results suggest that men with high MD spend more time and look more frequently at muscle-related body areas compared to men with low MD levels. Furthermore, AB towards specific muscle-related body areas was only found among individuals with high MD. Based on the increasing prevalence of body dissatisfaction and body image disorders in men [6,7], future research should focus not only on assessing but also on retraining dysfunctional attentional patterns. The current study used an innovative VR embodiment-based procedure with an ET assessment task that might be especially suitable for future studies aiming to retrain body-related AB in men with high MD and/or with muscle dysmorphia.Conceptualization, B.P.-G., E.E.-S., M.F.-G., and J.G.-M.; methodology, B.P.-G. and E.E.-S.; validation, B.P.-G., E.E.-S., and O.C.-M.; formal analysis, B.P.-G.; investigation, E.E.-S. and O.C.-M.; resources, M.F.-G. and J.G.-M.; data curation, B.P.-G.; writing—original draft preparation, B.P.-G., E.E.-S., and O.C.-M.; writing—review and editing, B.P.-G., M.F.-G., and J.G.-M.; visualization, B.P.-G., E.E.-S., M.F.-G., and O.C.-M.; supervision, M.F.-G. and J.G.-M.; project coordinator, J.G.-M.; and funding acquisition, J.G.-M. All authors have read and agreed to the published version of the manuscript.This study was funded by the Spanish Ministry of Economy and Competitiveness (MINECO/FEDER/UE/Project PSI2015-70389-R: Development of reality-based exposure techniques for improving anorexia nervosa treatment) and by the AGAUR, Generalitat de Catalunya, 2017SGR1693.The authors declare no conflicts of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.Muscle-related and nonmuscle-related areas of interest on the male virtual avatar.Illustration of the procedures used to induce full-body illusion in the study.Differences between men with high and low muscularity dissatisfaction (MD) in mean complete fixation time (in milliseconds; ms) and mean number of fixations on muscle (M)-related vs. nonmuscle (NM)-related areas of interest (AOIs). Error bars represent standard error of the mean (SEM). (a) Complete fixation time; (b) Number of fixations.Differences between men with high and low muscularity dissatisfaction (MD) (in mean, standard deviation, and median) in complete fixation time (in milliseconds; ms) and number of fixations on individual muscle-related areas of interest (AOIs). Mann–Whitney U test analyses: * p = 0.04, ** p < 0.01. (a) Complete fixation time; (b) Number of fixations.Mean and standard deviation (SD) of the eye-tracking measures.Note: Muscle (M)-related and nonmuscle (NM)-related areas of interest (AOIs); MD, muscularity dissatisfaction. ** Follow-up analyses, p < 0.01.

Med-MDPI/jcm_2/jcm-09-06-01737.txt

@@ -0,0 +1 @@
+Elevated levels of erythropoietin (EPO) are associated with an increased risk of death in renal transplant recipients (RTRs), but the underlying mechanisms remain unclear. Emerging data suggest that EPO stimulates production of the phosphaturic hormone fibroblast growth factor 23 (FGF23), another strong risk factor for death in RTRs. We hypothesized that the hitherto unexplained association between EPO levels and adverse outcomes may be attributable to increased levels of FGF23. We included 579 RTRs (age 51 ± 12 years, 55% males) from the TransplantLines Insulin Resistance and Inflammation Cohort study (NCT03272854). During a follow-up of 7.0 years, 121 RTRs died, of which 62 were due to cardiovascular cause. In multivariable Cox regression analysis, EPO was independently associated with all-cause (HR, 1.66; 95% CI 1.16–2.36; P = 0.005) and cardiovascular death (HR, 1.87; 95% CI 1.14–3.06; P = 0.01). However, the associations were abrogated following adjustment for FGF23 (HR, 1.28; 95% CI 0.87–1.88; P = 0.20, and HR, 1.45; 95% CI 0.84–2.48; P = 0.18, respectively). In subsequent mediation analysis, FGF23 mediated 72% and 50% of the association between EPO and all-cause and cardiovascular death, respectively. Our results underline the strong relationship between EPO and FGF23 physiology, and provide a potential mechanism underlying the relationship between increased EPO levels and adverse outcomes in RTRs.Renal transplant recipients (RTRs) have a high residual risk of all-cause and cardiovascular death, compared to the general population [1]. Previous studies demonstrated an independent association between higher circulating endogenous erythropoietin (EPO) levels and risk of all-cause and cardiovascular death among RTRs, similar to other patient populations such as chronic heart failure patients and the elderly [2,3,4]. In addition, administration of exogenous EPO may increase the risk of cardiovascular events in patients with chronic kidney disease (CKD) and end stage renal disease (ESRD) [5,6]. However, the underlying mechanisms responsible for the link between endogenous and exogenous EPO and adverse outcomes are unknown.Studies from our group and others suggest that EPO is prominently involved in fibroblast growth factor-23 (FGF23) physiology [7,8,9,10]. FGF23 is an osteocyte-derived hormone that plays an essential role in regulating phosphate and vitamin D metabolism. In RTRs, increased FGF23 levels post-transplant are independently associated with an increased risk of graft failure and death [11,12]. Hypoxia, the main stimulus for EPO synthesis, stabilizes hypoxia-inducible factor (HIF)-1α, which is a heterodimeric transcription factor that regulates oxygen homeostasis [13,14]. Subsequently, stabilized HIF1-α upregulates FGF23 production while concomitantly increasing FGF23 cleavage into inactive fragments, resulting in elevated total FGF23 levels but normal levels of intact, bioactive FGF23 [15,16,17,18]. In the current study, we hypothesized that the previously established, but hitherto unexplained association between EPO levels and adverse outcomes may be attributable to increased levels of total FGF23. Therefore, we investigated the associations between EPO and total FGF23 levels and prospective outcomes in our RTRs cohort. All RTRs (aged ≥ 18 years) who were at least 1-year post-transplantation were approached for participation in the current study during outpatient clinic visits between 2001 and 2003. All RTRs were transplanted in the University Medical Center Groningen (Groningen, the Netherlands). The study has been described in detail previously [19]. Among 847 RTRs approached for participation, 606 RTRs agreed to participate and were included. All patients provided written informed consent and the study protocol was approved by the local medical ethical committee (METc 2001/039). The study protocol adhered to principles of the Declaration of Helsinki and the Declaration of Istanbul. The co-primary endpoints of the study were all-cause and cardiovascular death. Cause of death was obtained by linking the number of the death certificate to the primary cause of death as coded by a physician from the Central Bureau of Statistics according to the International Classification of Diseases, 9th revision (ICD-9; https://icd.codes/icd9cm). CV death was defined as deaths in which the principal cause of death was cardiovascular in nature, using ICD-9 codes 410 to 447. Secondary endpoint constituted death-censored graft failure (DCGF). DCGF was defined as return to dialysis or re-transplantation. For the current analyses, we excluded RTRs who did not have plasma samples available for measuring EPO levels (n = 14) and RTRs who used exogenous EPO (n = 13) due to positive interference in EPO measurement, resulting in 579 RTRs eligible for analyses. Median follow-up time from inclusion to endpoint was 7.0 (interquartile range (IQR), 6.2 to 7.4) years. Data on the co-primary and secondary endpoints were available in all 579 participants. There was no loss-to-follow-up in the current study. Relevant donor, recipient, and transplant characteristics at baseline were extracted from the Groningen Renal Transplant Database, as described in detail previously [19]. Information on medical history and medication use was obtained from patient records. Participants’ height and weight were measured with participants wearing light indoor clothing without shoes. Blood pressure was measured according to a strict protocol as previously described [19]. Alcohol consumption and smoking behavior were recorded using a self-reported questionnaire. Smoking behavior was classified as never, former, or current smoker. Blood samples were drawn during the next outpatient clinic visit after agreeing to participate. Blood was drawn in the morning after an 8–12 h overnight fast, and all measurements were performed in samples of the same timepoint. In plasma EDTA samples frozen at −80°C, we measured plasma EPO levels using an immunoassay based on chemiluminescence (Immulite, Los Angeles, CA) [20]. We measured plasma total FGF23 levels with a human FGF23 (C-terminal) enzyme-linked immunosorbent assay (ELISA; Quidel Corp., San Diego, CA, USA) with intra-assay and interassay coefficients (CVs) of variation of <5% and <16% in blinded replicated samples, respectively [21]. The total FGF23 immunometric assay uses two antibodies directed against different epitopes within the C-terminal part of FGF23, and as such the assay detects both the intact hormone as well as C-terminal cleavage products, and therefore measures total FGF23 levels. We measured plasma ferritin levels using an electrochemiluminescence immunoassay (Modular analytics E170, Roche diagnostics, Mannheim, Germany). Renal function was determined by estimating GFR by applying the Chronic Kidney Disease Epidemiology Collaboration equation [22]. Proteinuria was defined as urinary protein excretion ≥ 0.5 g/24 h in 24-h urine collection. Serum cholesterol was measured using standard laboratory procedures. Serum creatinine was assessed using a modified version of the Jaffé method (MEGA AU 510; Merck Diagnostica, Darmstadt, Germany). Erythrocytosis was defined as hemoglobin level higher than 16.0 g/dL for women, and higher than 16.5 g/dL for men [23].Data were analyzed using IBM SPSS software, version 23.0 (SPSS Inc., Chicago, IL), R version 3.2.3 (Vienna, Austria) and STATA 14.1 (STATA Corp., College Station, TX). Data are expressed as mean ± standard deviation [SD] for normally distributed variables and as median (25th–75th interquartile range (IQR)) for variables with a skewed distribution. Categorical data are expressed as numbers (percentages). Co-linearity was tested by means of variance inflation factor (VIF) calculation, with a VIF score of lower than 5 indicating no evidence for co-linearity. We used Cox proportional hazards regression analysis to investigate the association between EPO levels and prospective outcomes. Assumptions of proportionality in Cox regression analyses were checked using Schoenfeld residuals plots and checking nonsignificance of covariates and with the global test (Model 1; EPO with death and CV death; P > 0.30 for global test). In these Cox regression analyses, we adjusted for potential confounders based on univariable associations or for factors of known biologic importance. We adjusted for age, sex, body surface area (BSA), eGFR, proteinuria, time since transplantation, presence of diabetes, systolic blood pressure (SBP), total cholesterol, and use of calcineurin inhibitors, proliferation inhibitors, and angiotensin-converting enzyme (ACE)-inhibitors and angiotensin II-receptor blockers (ARBs) (Model 1). We subsequently adjusted for potential mediators in the pathway between EPO and death, i.e., hemoglobin levels (Model 2); for ferritin (Model 3), high-sensitive C-reactive protein (hs-CRP) (Model 4), and finally for total FGF23 (Model 5). Due to skewed distribution, EPO, ferritin, hs-CRP, and total FGF23 were natural log-transformed. We repeated the Cox regression analyses between EPO and outcomes with EPO levels being divided in quartiles. Furthermore, we generated Kaplan–Meier curves to visually show the effect of increased risk of death and cardiovascular death while being in the highest EPO quartile. A log-rank test for trend was used to compare rates of death across quartiles. We also assessed the association between FGF23 levels and prospective outcomes adjusting for all potential confounders according to Model 1 and including EPO. To reflect the contribution of covariates in the different Cox regression models, we generated Supplementary Tables S1–S4 showing the strength of covariates in univariable and multivariable models. To allow comparability between the hazard ratios (HR) of covariates, HR of continuous variables in Supplementary Tables S1–S4 are shown as expressed per SD. As sensitivity analysis, we assessed the prevalence of different etiologies of CKD in total cohort and across EPO quartiles, and we adjusted the association between EPO and all-cause and cardiovascular death for etiology of CKD. Subsequently, we calculated the percentage of change in HR before and after adjustment for FGF23. Percentage change in HR was calculated as—(HR without adjustment – HR with adjustment)/(HR without adjustment – 1) × 100% [24]. Hereafter, we performed mediation analyses with the methods as previously described by Preacher and Hayes, which are based on logistic regression [25,26]. These analyses allow for testing significance and magnitude of mediation on the association between EPO and outcomes [25,26]. Overall, 0.4% of demographic data were missing and these data were imputed using regressive switching [27]. Five datasets were multiply-imputed, and results were pooled and analyzed according to Rubin’s rules [28]. In all analyses, a two-sided p-value < 0.05 was considered significant.We included 579 RTRs (mean age of 51 ± 12 years; 55% male) at a median of 6.0 (2.6–11.6) years after transplantation. Erythrocytosis was present in 27 (5%) of the included RTRs. Further demographics and clinical baseline characteristics across quartiles of EPO are shown in Table 1.Median plasma EPO levels were 17.4 (11.9–24.2) IU/L and median FGF23 levels were 137 (94–212) RU/mL. Increased FGF23 levels were noted across EPO quartiles (115 (81–168) RU/mL; 125 (88–184) RU/mL; 138 (95–212) RU/mL; and 195 (115–363) RU/mL respectively, P < 0.001). FGF23 levels were positively correlated with EPO levels (r = 0.28, P < 0.001), with a VIF of 1.15, indicating very minimal co-linearity. During a median follow-up of 7.0 (6.2–7.4) years, 121 RTRs died. Of the 121 deceased RTRs, 62 RTRs (51%) died from cardiovascular causes. Other causes of death were infection (18%), malignancy (24%), and miscellaneous causes (8%). In univariable Cox regression analyses, higher EPO levels were associated with an increased risk of all-cause death (HR per 1 ln IU/L increase, 1.74; 95% confidence interval (CI), 1.29–2.34; P < 0.001). A full list of HRs for covariates univariably with death are described in Supplementary Table S1. In multivariable Cox regression analyses, the association between EPO and all-cause death remained significant (HR, 1.66; 95% CI, 1.16-2.36; P=0.005) independent of adjustment for age, sex, BSA, eGFR, proteinuria, time since transplantation, presence of diabetes, SBP, total cholesterol, use of calcineurin inhibitors, proliferation inhibitors, ACE-inhibitors or ARB (Model 1). Further adjustment for hemoglobin, ferritin, or hs-CRP levels did not materially alter the results. However, further adjustment for FGF23 levels attenuated the association between EPO and all-cause death such that the association no longer remained significant (HR, 1.28; 95% CI, 0.87–1.88; P = 0.20) (Table 2). A full list of HRs for covariates in the multivariable model can be found in Supplementary Table S2.We identified similar results when subdividing EPO levels into quartiles (Table 3; Figure 1 with Kaplan–Meier curves showing the univariably increased risk of death across EPO quartiles). In multivariable Cox regression analysis, RTRs in the upper quartile of EPO had a more than two times higher risk of death (HR, 2.11; 95% CI, 1.15–3.86), when compared to RTRs in the lowest quartile, independent of potential confounders. In line with the association between EPO as continuous variable and death, further adjustment for FGF23 levels attenuated the association between the upper EPO quartile and risk of death (HR, 1.55; 95% CI, 0.82–2.91; Table 3; Figure 2A). A full list of HRs for covariates in the multivariable model for EPO divided in quartiles can be found in Supplementary Table S3. When we assessed the associations between EPO and cardiovascular death, we found similar findings. Higher EPO levels were associated with an increased risk of cardiovascular death in univariable analyses (Figure 1) and in all subsequent models (Table 2). However, the association no longer remained significant, both as a continuous variable and as divided in quartiles, after further adjustment for FGF23 (Table 2 and Table 3; Figure 2B).In contrast, FGF23 levels per se were strongly associated with all-cause death independent of adjustment for potential confounders including EPO (HR per RU/mL, 1.76; 95% CI, 1.33–2.34; P < 0.001). Likewise, FGF23 levels per se were also strongly associated with cardiovascular death independent of adjustment for potential confounders including EPO (HR, 1.84; 95% CI, 1.23–2.76; P = 0.003). A full list of HRs for covariates can be found in Supplementary Table S4. As sensitivity analysis, we assessed the prevalence of different etiologies of CKD in the total cohort and across quartiles of EPO (Supplementary Table S5). The most prevalent etiologies of CKD were primary glomerular disease (28%), polycystic disease (18%), and tubulo-interstitial disease (16%). Following adjustment for etiology of CKD additive to model 1, the association between EPO and all-cause death (HR, 1.57; 95% CI, 1.09–2.26; P = 0.02) and between EPO and cardiovascular death (HR, 1.75; 95% CI, 1.05–2.90; P = 0.03) remained materially unchanged. During a median follow-up of 6.9 (6.1–7.4) years, 46 RTRs developed DCGF. When we assessed the associations between EPO and DCGF, we did not find an association (HR, 0.82; 95% CI, 0.48–1.41; P = 0.48). Further adjustment for potential confounders did not ameliorate the association between EPO and DCGF. In contrast, FGF23 levels were univariately associated with DCGF (HR, 3.07; 95% CI, 2.22–4.24; P < 0.001). However, after adjustment for potential confounders including EPO, FGF23 was no longer associated with DCGF (HR, 1.57; 95% CI, 0.94–2.64; P = 0.09).Adjustment for FGF23 caused a large reduction in HR in the Cox Regression analysis in the association between EPO and all-cause and cardiovascular death (58% reduction in HR between EPO and all-cause death; and 48% reduction in HR between EPO and cardiovascular death). In subsequent mediation analyses, we identified that FGF23 was a significant mediator of the association between EPO and all-cause death (P value for indirect effect <0.05; 72% of the association was explained by FGF23; Table 4). Similarly, FGF23 explained 50% of the association between EPO and cardiovascular death (P value for indirect effect <0.05; Table 4).In this study, we show that higher endogenous EPO levels are associated with an increased risk of all-cause and cardiovascular death in RTRs, and that these associations are largely explained by variation in FGF23 levels. This study confirms recent studies about the essential role of EPO in FGF23 physiology in experimental and human models [7,8,9,10], extends these findings to RTRs, and support the notion that FGF23 is an important mediator in the association between EPO and risk of death. EPO, a hormone mainly produced in the kidney in response to hypoxia, is essential for erythropoiesis [29]. EPO controls proliferation, maturation, and also survival of erythroid progenitor cells [30]. Previously, it has been shown that high endogenous EPO levels were associated with an increased risk of all-cause and cardiovascular death in RTRs [2,3]. Similarly, in the setting of CKD and ESRD, correction of anemia with recombinant EPO led an increased risk of cardiovascular morbidity and death [5,6]. The mechanisms responsible for these adverse effects of both endogenous as exogenous EPO are unknown. In the Correction of Hemoglobin and Outcomes in Renal Insufficiency (CHOIR) trial, the highest risk of cardiovascular death was seen in patients with the highest EPO dose, suggesting that EPO resistance through inflammation and/or functional iron deficiency might be a possible link [6]. However, in the current study, the association between endogenous EPO levels and death was independent of adjustment for inflammation as well as independent of iron parameters, renal function, and standard classical cardiovascular risk factors including systolic blood pressure and cholesterol levels. Although there was a difference in prevalence of use of calcineurin and proliferation inhibitors across EPO quartiles, the association between EPO and death remained independent of adjustment for calcineurin and proliferation inhibitors. In contrast, adjustment for FGF23 markedly attenuated the association between EPO and death.Elevated total FGF23 levels have previously been shown to be associated with increased risk of death in RTRs, as well as in various other patient groups including postoperative acute kidney injury, nondialysis CKD, and ESRD [31,32,33,34,35]. FGF23 regulation is determined by a complex interplay between parathyroid hormone, 1,25-dihydroxyvitamin D, klotho, glucocorticoids, calcium, and phosphate [36,37]. In recent years, iron deficiency has been identified as an important regulator of FGF23 [38,39,40]. In addition, recent studies demonstrated that EPO stimulates murine and human FGF23 [7,8]. Clinkenbeard and colleagues reported increased FGF23 mRNA expression in vitro, ex vivo, and in vivo due to EPO treatment in UMR-106 cells, in isolated bone marrow cells, and in marrow from mice, respectively [7]. In addition, Rabadi et al. showed in experimental animal models that an acute loss of 10% blood volume led to an increase in total FGF23 and EPO levels within six hours. Furthermore, exogenous administration of EPO resulted in an acute increase in plasma total FGF23 levels similar to those seen in acute blood loss [8]. Similarly, Flamme et al. described in animal models an increase in plasma total FGF23 both after injection of recombinant human EPO and after HIF-proline hydroxylase inhibitor [41]. The present findings in our study underscore these observations and emphasize the important role of EPO in FGF23 physiology. Importantly, the current study is the first to show that prospective associations between EPO and adverse outcomes in RTRs seem to be, at least to large extent, related to increased levels of total FGF23. The mechanisms through which EPO, as reflection of tissue hypoxia, lead to increased bone marrow FGF23 transcription are currently unknown and require additional investigation. The previously performed studies showed that EPO acutely increases total FGF23 levels out of proportion to intact FGF23 (iFGF23), suggesting an upregulated FGF23 production with concomitantly increased cleavage, with as a result an increase in C-terminal FGF23 fragments. To date, it remains incompletely understood how EPO increases post-translational cleavage. Results from our group and collaborators found previously in EPO-overexpressing mice a decreased GalNT3 bone marrow mRNA expression, without differences in Fam20C or furin expression, implying that a decreased GalNT3 might play a possible role [9]. However, more investigation is imperative to unravel this mechanistic link. The downstream consequences of elevated levels of FGF23 and the subsequent excess risk of death have not been fully elucidated yet. Several previous reports have shown that iFGF23 has biologic activity through binding to several FGF23 receptors. Besides the well-known functions of iFGF23 in the regulation of renal phosphate handling and vitamin D metabolism, recent studies have shown a myriad “off-target” effects of iFGF23 on the heart and other organs. Preclinical studies demonstrated that FGF23 can lead to left ventricular hypertrophy in cardiac myocytes, and promote endothelial dysfunction [42,43]. In addition, FGF23 stimulates renal fibrosis [44], exerts pro-inflammatory effects [45], and disrupts normal immune function [46]. Most likely, the increased death risk due to elevated levels of FGF23 is attributable to a combination of these effects. Although the biologic activity of iFGF23 has unequivocally been demonstrated, the biologic activity of C-terminal FGF23 fragments remains uncertain. Previously, it has been shown that C-terminal FGF23 may function as an iFGF23 antagonist, by competing with iFGF23 for binding to its receptor, which may reduce phosphaturia and aggravate soft tissue calcification [47]. In addition, Courbabaisse et al. has shown, at least in vitro, that C-terminal FGF23 increases adult rat ventricular cardiomyocyte size by stimulation of FGF receptor 4 in the absence of co-stimulatory factor alpha-klotho, and in sickle cell disease patients that elevated cleaved FGF23 levels were associated with heart hypertrophy [48].Our study has multiple strengths as well as limitations. The major strength of the current study is the large prospective cohort of stable RTRs with detailed clinical and laboratory data available, including EPO, FGF23, hs-CRP, and ferritin levels. Additionally, no participants were lost to follow-up with respect to the endpoints, despite a considerable follow-up period. Limitations of the current study include that, due to the observational status of our single center study, we cannot exclude the possibility of residual confounding, and conclusions about causality cannot be drawn. Furthermore, we were unable to measure iFGF23 levels, since samples were not stored with protease inhibitors, and iFGF23 has been shown to be susceptible to degradation with long-term storage [49]. This precludes us to discern whether the elevated total FGF23 levels are the result of increased iFGF23 levels or due to an increase in allegedly assumed inactive C-terminal fragments. Another limitation of the study is that we only used CRP levels as inflammatory parameter, other markers of inflammation (e.g., cytokines, cell subtypes) were not available, but could possibly have contributed to the results. In addition, another limitation of current study is the use of single-time measurements hampering the possibility to track the levels of EPO and FGF23 over time with respect to each other and with respect to risk of death. However, it should be realized that most epidemiological studies use a single baseline measurement to investigate associations of variables with outcomes, which adversely affects the strength and significance of the association of these variables with outcomes. If intraindividual variability of variables is taken into account, this results in strengthening of associations that also existed for single measurements of these variables [50,51]. Finally, we want to emphasize that the mediation analyses that we performed are plain straightforward mediation analyses. Although based on literature, we have strong evidence that FGF23 is a mediator in the association of EPO with risk of death, we cannot exclude that an unmeasured cause of mortality or alternative potential mediators has influenced the currently identified results.In conclusion, we identified that elevated levels of EPO were independently associated with an increased risk of death in RTRs, and that this association was to a large extent explained by variation in FGF23 levels. Further research is needed to fully elucidate the mechanism through which this ensues and to unravel whether the currently identified results can be extrapolated to exogenous EPO in RTRs.The following are available online at https://www.mdpi.com/2077-0383/9/6/1737/s1, Supplementary Table S1: Univariable associations of variables with prospective outcomes (i.e., all-cause death and cardiovascular death), Supplementary Table S2: Reporting of all hazard ratios of all covariates included in the Cox Regression Analyses for the association between erythropoietin as continuous variable and risk of all-cause and cardiovascular death (according to model 5 [including FGF23]), Supplementary Table S3: Reporting of all hazard ratios of all covariates included in the Cox Regression Analyses for the association between quartiles of erythropoietin and risk of all-cause and cardiovascular death (according to model 5 [including FGF23]), Supplementary Table S4: Reporting of all hazard ratios of all covariates included in the Cox Regression Analyses for the association between FGF23 as continuous variable and risk of all-cause and cardiovascular death (multivariable excluding erythropoietin [Supplementary Table S2 includes erythropoietin]), Supplementary Table S5: Prevalence of the different etiologies of CKD described in the total cohort of 579 RTRs and across the quartiles of erythropoietin levels.Conceptualization, M.F.E.; Data curation, M.H.D.B. and S.J.L.B.; Formal analysis, M.F.E., M.A.D.J., D.E.L. and I.M.N.; Investigation, S.J.L.B.; Methodology, M.F.E., M.H.D.B., S.J.L.B. and C.A.J.M.G.; Supervision, S.J.L.B. and C.A.J.M.G.; Visualization, M.F.E.; Writing—original draft, M.F.E.; Writing—review and editing, M.A.D.J., D.E.L., I.M.N., M.H.D.B., S.J.L.B. and C.A.J.M.G. All authors have read and agreed to the published version of the manuscript.This research received no external fundingThe current study was based on TransplantLines Insulin Resistance and Inflammation (TxL-IRI) Cohort Study (NCT03272854).The authors declare no conflict of interest.Kaplan–Meier Curves depicting the association between EPO quartiles and risk of all-cause (left panel) and cardiovascular death (right panel). Reported p-values have been calculated with the log-rank test for trend.Hazard ratios and corresponding 95% confidence intervals are depicted for risk of death, both all-cause (A) and cardiovascular death (B), according to quartiles of erythropoietin levels. First, the univariate association is shown. Second, the multivariable adjustment is performed with adjustment for age, sex, body surface area (BSA), eGFR, proteinuria, time since transplantation, presence of diabetes, systolic blood pressure (SBP), total cholesterol, and use of calcineurin inhibitors, proliferation inhibitors, and angiotensin-converting enzyme (ACE)-inhibitors and angiotensin II-receptor blockers (ARBs). Third, adjustment for FGF23 is performed following the multivariable adjustment. The first quartile was chosen as a reference group in all analyses. Significance levels are indicated by numbers of asterisks, i.e., *** <0.001, ** <0.01, * <0.05; Abbreviations—BSA, body surface area; CI, confidence interval; eGFR, estimated glomerular filtration rate.Baseline characteristics of the included 579 renal transplant recipients (RTRs) across erythropoietin (EPO) quartiles.‡ Erythrocytosis defined as hemoglobin level >16 g/dL (F) and >16.5 g/dL (M) * Only available in a subset cohort of 415 RTRs. Values are means ± standard deviation, medians (interquartile range) or proportions (%). Diabetes mellitus was defined as serum glucose > 7 mmol/L or the use of antidiabetic drugs. Abbreviations—ACE-i, angiotensin converting enzyme inhibitors; DBP, diastolic blood pressure; eGFR, estimated glomerular filtration rate; FGF23, fibroblast growth factor 23; hs-CRP, high-sensitivity C-reactive protein; MCV, mean corpuscular volume; SBP, systolic blood pressure; Tx, transplantation.Association between erythropoietin levels and risk of all-cause and cardiovascular death.* Hazard ratios are shown per 1 ln IU/L increase in EPO levels; Model 1: Adjusted for age, sex, body surface area, eGFR, proteinuria, time since transplantation, presence of diabetes, systolic blood pressure, total cholesterol, use of calcineurin inhibitors, proliferation inhibitors, and ACE-inhibitors or ARB; Model 2: Model 1 + adjustment for hemoglobin; Model 3: Model 1 + adjustment for ferritin; Model 4: Model 1 + adjustment for hs-CRP; Model 5: Model 1 + adjustment for FGF23. Ferritin, hs-CRP, and FGF23 were naturally log transformed before adding to the Cox regression analysis due to skewed distribution. Abbreviations—ACE, angiotensin-converting enzyme; ARB, angiotensin-receptor blockers; FGF23, fibroblast growth factor 23; CI, confidence interval; eGFR, estimated glomerular filtration rate; HR, hazard ratio; hs-CRP, high-sensitive C-reactive protein.Association between erythropoietin quartiles and risk of all-cause and cardiovascular death.Model 1: Adjusted for age, sex, body surface area, eGFR, proteinuria, time since transplantation, presence of diabetes, systolic blood pressure, total cholesterol, use of calcineurin inhibitors, proliferation inhibitors, and ACE-inhibitors or ARB. Model 2: Model 1 + adjustment for hemoglobin; Model 3: Model 1 + adjustment for ferritin; Model 4: Model 1 + adjustment for hs-CRP; Model 5: Model 1 + adjustment for FGF23. Ferritin, hs-CRP, and FGF23 were naturally log transformed before adding to the Cox regression analysis due to skewed distribution. Abbreviations—ACE, angiotensin-converting enzyme; ARB, angiotensin-receptor blockers; FGF23, fibroblast growth factor 23; CI, confidence interval; eGFR, estimated glomerular filtration rate; HR, hazard ratio; hs-CRP, high-sensitive C-reactive protein.Mediation analysis of FGF23 on the association between EPO and all-cause and cardiovascular death in renal transplant recipients.* The coefficients of the indirect ab path and the total ab + c’ path are standardized for the standard deviations of EPO, FGF23, all-cause and cardiovascular death. **All coefficients are adjusted for age, sex, body surface area, eGFR, proteinuria, time since transplantation, presence of diabetes, systolic blood pressure, total cholesterol, use of calcineurin inhibitors, proliferation inhibitors, ACE-inhibitors or ARB. *** The size of the significant mediated effect is calculated as the standardized indirect effect divided by the standardized total effect multiplied by 100, e.g., 0.090 divided by 0.124 multiplied by 100 constitutes 72% as percentage of mediation ‡ Odds ratios for risk of outcomes can be calculated by taking the exponent of the unstandardized total effect. For example, the unstandardized coefficient of the direct effect of EPO on all-cause death while adjusting for FGF23 is 0.120, which can be calculated to an OR by taking the exponent of this regression coefficient, i.e., e0.120=1.12, which corresponds to the HR of 1.28 (see Table 2). The discrepancy between the ratios is due to taking into account time-to-event with HR in contrast to OR. †95% CIs for the indirect and total effects were bias-corrected confidence intervals after running 2000 bootstrap samples. Abbreviations—ACE, angiotensin converting enzyme; ARB, angiotensin receptor blockers; Bc, bias corrected; CI, confidence interval; eGFR, estimated glomerular filtration rate; FGF23, fibroblast growth factor 23.

Med-MDPI/jcm_2/jcm-09-06-01738.txt

@@ -0,0 +1 @@
+Contributed equally to this work as first and senior authors, respectively.Cytoreductive Surgery and Hyperthermic Intraperitoneal Chemotherapy (CRS-HIPEC) may be curative for colorectal cancer patients with peritoneal metastases (PMs) but it has a high rate of morbidity. Accurate preoperative patient selection is therefore imperative, but is constrained by the limitations of current imaging techniques. In this pilot study, we explored the feasibility of circulating tumor (ct) DNA analysis to select patients for CRS-HIPEC. Thirty patients eligible for CRS-HIPEC provided blood samples preoperatively and during follow-up if the procedure was completed. Targeted Next-Generation Sequencing (NGS) of DNA from PMs was used to identify bespoke mutations that were subsequently tested in corresponding plasma cell-free (cf) DNA samples using droplet digital (dd) PCR. CtDNA was detected preoperatively in cfDNA samples from 33% of patients and was associated with a reduced disease-free survival (DFS) after CRS-HIPEC (median 6.0 months vs median not reached, p = 0.016). This association could indicate the presence of undiagnosed systemic metastases or an increased metastatic potential of the tumors. We demonstrate the feasibility of ctDNA to serve as a preoperative marker of recurrence in patients with PMs of colorectal cancer using a highly sensitive technique. A more appropriate treatment for patients with preoperative ctDNA detection may be systemic chemotherapy in addition to, or instead of, CRS-HIPEC.Metastatic colorectal cancer (CRC) remains the second most common cause of cancer-related death, despite improvements in treatment over recent decades [1]. Peritoneal metastases (PMs) are diagnosed in 10–25% of CRC patients [2,3,4], either at the time of primary tumor diagnosis or during subsequent investigations. If restricted to the peritoneum, referred to as isolated PMs, treatment with systemic chemotherapy confers a median overall survival (OS) of 12–18 months [5,6,7,8]. Patients with isolated PMs that are limited in their spread throughout the peritoneum may qualify for Cytoreductive Surgery and Hyperthermic Intraperitoneal Chemotherapy (CRS-HIPEC) treatment, which is associated with a disease-free survival (DFS) of 13.1 months and an increased median OS of 35–45 months [9,10,11,12]. Patients with the least extensive peritoneal spread have the potential to experience the greatest benefit from CRS-HIPEC, reflected in a median OS of 56 months [13]. The addition of HIPEC following CRS did not show significant survival benefit in the recent large randomized-controlled PRODIGE-7 trial [14]. However, unlike the patients included in our study, patients in the PRODIGE-7 trial had received neo-adjuvant systemic therapy prior to CRS-HIPEC and had received oxaliplatin rather than mitomycin C during HIPEC [14]. Nevertheless, CRS performed in high-volume expert centers resulted in an OS of 41 months [14]. This demonstrates that a dedicated centralized multimodality approach to treatment offers improved survival for patients with PMs.Unfortunately, shortcomings in modalities of detection and quantification of PMs often result in diagnosis at an advanced stage, when the peritoneal spread is extensive. These patients typically see reduced benefit from CRS-HIPEC because of a higher incidence of recurrence after the procedure [15,16] and because they experience greater treatment-related morbidity from extensive CRS [17]. Timely preoperative detection of PMs could help decide if any potential survival benefit outweighs the treatment-related morbidity. Similarly, recurrence of PMs after CRS-HIPEC is a common phenomenon [13] that also tends to be diagnosed at an advanced stage. Improved detection in this situation could indicate an early start to chemotherapy to reduce disease symptoms, or a repeat CRS-HIPEC with potentially curative intent if the recurrence is isolated and limited [18].If PMs are not detected during resection of the primary tumor, initial detection is by subsequent (PET)-CT imaging, which is limited to a sensitivity of 72% and further decreases to 11% for nodules smaller than 5 mm [19]. Carcinoembryonic antigen (CEA) levels can be elevated in the blood of CRC patients, particularly in those with metastases. However, the sensitivity of the CEA test is limited in patients with isolated PMs [20,21]. If PMs are suspected, diagnostic laparoscopy (DLS) is often performed preoperatively to determine the Peritoneal Cancer Index (PCI), a numeric score ranging from 1–39 that combines lesion size with the number of affected abdomino-pelvic regions [10,22]. Although DLS offers a higher sensitivity and specificity for detection, it is invasive and often still underestimates the extent of PMs [23]. The most reliable measure of PMs is an intraoperative assessment performed immediately before the intended CRS-HIPEC. However, up to 25% of patients are disqualified at this point due to irresectable PMs, reflected by a PCI higher than 20, so undergo an open–close procedure whereby the abdomen is closed without CRS-HIPEC [24]. There is an urgent clinical need for less invasive and more accurate tools to detect and quantify the extent of PMs. Novel approaches include improved imaging techniques such as diffusion-weighted (DW) MRI [25] and liquid biopsy analysis.Recent technical advances have enabled circulating tumor (ct) DNA, the fraction of cell-free (cf) DNA that originates from tumor cells, to be detected in plasma with high sensitivity and specificity. Analysis can serve as a dynamic marker in CRC patients by the quantification of cfDNA levels and the identification of tumor-specific genetic and epigenetic markers including; mutations, structural variations and methylation [26,27,28]. Studies that used digital PCR methods have shown ctDNA to be detectable in up to 100% of CRC patients with systemic metastases and in 73% of those without evidence of systemic metastasis [29]. However, in a study that used a next-generation sequencing (NGS) panel to test patients with resectable PMs of diverse cancer types, ctDNA was only detectable preoperatively in 39% of cases [30]. The lower plasma ctDNA representation in CRC patients with isolated PMs compared to systemic metastases is probably due to inherent biological differences in ctDNA shedding.The detection and quantification of relatively low quantities of ctDNA in plasma from CRC patients with PMs requires a highly sensitive technique. Droplet digital (dd) PCR has emerged as one of the most sensitive and specific methods of ctDNA analysis in the oncology setting [31]. By specifically targeting genomic loci in a tissue-guided manner, ddPCR can precisely quantify DNA fragments that contain either the mutant or wild-type nucleotide in a relatively simple, fast and cost-effective work-flow. Although this approach has previously been applied to colorectal malignancies [32,33], to our knowledge, it has not been used as a biomarker in a well-defined cohort of CRC patients who are candidates for CRS-HIPEC.In this pilot study, we aimed to explore the feasibility of ctDNA analysis in a clinical situation to select patients for CRS-HIPEC. We assessed the capability of ctDNA to quantify the extent of PMs and its suitability as a preoperative prognostic marker of recurrence.Patients diagnosed with synchronous or metachronous PMs of colorectal adenocarcinoma and considered eligible for CRS-HIPEC following standard work-up including imaging and DLS, were initially included in this study at the Cancer Center Amsterdam of Amsterdam University Medical Center (location VUmc) between August 2016 and March 2018 [11,34]. Patients were preoperatively excluded from the study if: the estimated extent of PMs was deemed to be irresectable by subsequent (PET-) CT or DLS; systemic metastases were detected (excluding resectable liver metastases with minimal tumor burden) [35]; or PMs removed during a previous procedure were found by histological assessment to be non-colorectal in origin or were not adenocarcinoma. Patients were excluded from the study on a technical basis if no mutations were identified in PMs. Clinical and pathological data were retrospectively obtained from patient records (Supplementary materials and methods, clinical and pathological data) [10,22]. The mismatch repair status of PMs was not tested in any of the patients included in the study. The outcome of the CRS was determined according to the maximal size of residual tumor tissue and was classified as: R1) when no macroscopically visible tumor remained in situ (complete resection), R2a) when the residual tumor was smaller than 2.5 mm or R2b) when it was larger than 2.5 mm [36]. CRS-HIPEC was performed according to a standard protocol [11,37]. If a complete CRS was achieved, HIPEC was performed using the open coliseum technique with mitomycin C. If residual tumor tissue remained after CRS, HIPEC was not performed [6,34].This study was registered with the Dutch Trial Registry [38] and was conducted in accordance with the Declaration of Helsinki with the approval of the Amsterdam UMC, VU University Medical Ethical Testing Committee (2016.254-NL57226.029.16 and 2017-302(A2018). All patients provided written informed consent to participate in the study.PMs previously removed alongside primary tumor resection were retrieved from the Biobank at Amsterdam University Medical Centers (UMC) -location VUmc, and retrospectively analyzed to identify mutations to test in cfDNA. Preoperative blood samples were collected after patients were placed under general anesthesia but immediately before surgical incision for the intended CRS-HIPEC procedure. Blood was kept at room temperature until plasma separation within 8 h of collection (Supplementary materials and methods, blood processing). At least one postoperative blood sample was taken by venipuncture from all CRS-HIPEC patients, typically within 2–4 weeks, but no later than 3 months after the procedure. Further samples were taken during routine follow-up every 3 months, up to 25 months after CRS-HIPEC. If a recurrence was diagnosed during follow-up by physical assessments and (PET-) CT imaging, an additional blood sample was taken at diagnosis or within 1 month. No postoperative blood samples were obtained from patients who did not undergo the complete CRS-HIPEC procedure because the presence of residual tumor excluded them from postoperative ctDNA analysis.Formalin-fixed paraffin-embedded (FFPE) tumor tissue of PMs was processed as previously described [39,40]. Genomic DNA was subsequently isolated using a Qiagen QIAamp DNA FFPE Tissue kit according to the instructions of the manufacturer (Qiagen, Venlo, the Netherlands). Cell-free DNA was isolated from up to 3 mL aliquots of plasma using the Qiagen QIAsymphony Circulating DNA Kit. A plasma sample of known cfDNA concentration was included in each isolation run to ensure consistent performance of the isolation kit. Genomic DNA from PMs underwent NGS-based mutation analysis using the TruSeq Amplicon Cancer Panel (TSACP; Illumina Inc., San Diego, CA, USA) or a High-Resolution Melting assay followed by Sanger sequencing (HRM-sequencing) (Supplementary materials and methods, DNA isolation and mutation analysis) [40,41]. Only genomic variants of known oncogenic significance identified by HRM-sequencing, or by TSACP sequencing with a Variant Allele Frequency (VAF) of ≥3% in the PMs were included (Table S1A). These mutations were targeted in cfDNA samples using specific mutant and wild-type ddPCR primer and probe combination kits (BioRad, California, USA) (Table S1B). Results for each assay were used to calculate the VAF and estimate the concentration of cfDNA (Table S1C). All cfDNA samples were tested using the KRAS G12/13 screening kit (cat. #1863506, BioRad, Hercules, CA, USA) regardless of the KRAS mutation status of the PMs. The performance of all kits was verified using gBlocks when available (Integrated DNA Technologies, Iowa, USA); 191–230nt fragments of synthetic double-stranded DNA containing the nucleotide change of interest (Supplementary materials and methods, gBlocks and ddPCR; Table S1D).The association between clinico-pathological variables and ctDNA detection was tested using the Fisher’s exact test for two dichotomous variables or the Mann–Whitney U test for a continuous variable combined with a dichotomous variable. Comparison of preoperative cfDNA input to ddPCR reactions was performed using the Mann–Whitney U test. Comparison of preoperative cfDNA input with follow-up samples was tested using the Kruskal–Wallis test for a continuous variable in multiple groups. Statistical significance was defined as a p-value < 0.05 (two-sided test). Univariate associations between DFS and clinico-pathological variables and ctDNA detection were tested using the log-rank test (Kaplan–Meier method). A cox regression analysis was performed to generate hazard ratios and 95% confidence intervals (95% CI). No correlation between OS and clinico-pathological variables was calculated as the number of events within the follow-up period was insufficient for statistical analysis. Dichotomization was performed on the basis of mean values for continuous variables. Statistical analyses were performed using the Statistical Package for Social Sciences (SPSS) version 23 for Windows (IBM Corporation, Armonk, NY, USA) and GraphPad Prism v7.02 (GraphPad Software Inc., San Diego, CA, USA).Thirty patients eligible to receive CRS-HIPEC were included in the study, following the exclusion of fourteen patients for the reasons outlined in Figure 1. Baseline characteristics of all patients are presented in Table 1. Of these patients, 24 underwent CRS-HIPEC after intraoperative assessment determined them to have resectable metastases. None of the patients had received neo-adjuvant chemotherapy prior to CRS-HIPEC. To account for the effect of liver metastases on ctDNA levels, CRS-HIPEC patients were postoperatively sub-classified into those with isolated PMs (n = 22) or PMs with resectable liver metastases (n = 2) (Figure 1). The remaining six patients were intraoperatively disqualified from CRS-HIPEC so instead underwent an open–close procedure due to a PCI higher than 20 (n = 4), or because irresectable liver metastases (n = 1) or para-aortic lymph nodes (n = 1) were discovered.Between one and four mutations (median 1.5) were identified in the tumor tissue of PMs from each of the 30 patients in the study. The three most frequently mutated genes were: KRAS (18/30 patients; 60%); TP53 (13/30 patients; 43%) and APC (8/30 patients; 27%). Mutations in these genes are typically found in 43%, 60% and 81% of non-hypermutated CRC tumors, respectively [42]. A detailed list of mutations is described in Table S1A. By targeting these tissue-guided mutations by ddPCR analysis, ctDNA was detected preoperatively in 10/30 (33%) patients (Figure 2A). In the subgroup of patients who underwent CRS-HIPEC, the rate of detection was the same (8/24; 33%) and was marginally lower when the two patients with liver metastases were excluded (6/22; 27%). To estimate the relative amounts of ctDNA shed into the plasma, a median VAF of 1.8% (range 0.6–10) was determined for the eight patients who had detectable ctDNA. Interestingly, the VAF was marginally higher in patients who were later diagnosed with a systemic recurrence (median 2.8%, range 2.1–10, n = 4) compared to a loco-regional recurrence (median 0.6%, range 0.6–1.4, n = 3, p = 0.057). In comparison, patients in the open–close subgroup had the same detection rate as the overall study group (2/6; 33%), with a median VAF of 1.75% (range 0.9–2.6, n = 2). This suggests that there is no correlation between the preoperative PCI and the likelihood of ctDNA detection in the circulation.Detection of ctDNA did not correlate significantly with any of the tested clinical variables used to assess eligibility for CRS-HIPEC, including PCI (Table S2). When the total cfDNA concentration was calculated for all 30 patients, there was no significant difference between samples that contained detectable ctDNA and those that did not (p = 0.422), or between samples from patients who underwent CRS-HIPEC (median 8.7ng ml−1, range 4.3–70.4) or an open–close procedure (median 7.2 ng ml−1, range 4.4–9.9; p = 0.174) (Figure 2C). If a cfDNA sample contained multiple mutations, their proportions closely mirrored those observed in PMs (Figure 2B), which suggests a faithful representation of tumor DNA in the circulation. All preoperative plasma samples had 100% concordance with the KRAS mutation status in PMs when tested with the KRAS G12/13 screening kit, which further indicates that the PMs were the source of the ctDNA.The association between preoperative ctDNA detection and recurrence was tested in 14/24 patients who were diagnosed with a recurrence during the follow-up period (median DFS 17 months, range 6–25). The proportion of these patients who had detectable ctDNA preoperatively was higher for those with a systemic- (4/5, 80%) compared to a loco-regional recurrence (3/9, 33%). Regardless of the type of recurrence, preoperative ctDNA detection was associated with a median DFS of 6.0 months (95%-CI 1.8–10.2), significantly worse compared to patients without ctDNA detection (median DFS not reached, p = 0.016; HR 3.454, 95% CI 1.145–10.423) (Figure 3A). When the two patients with resectable liver metastases were excluded from the survival analysis, both of whom had detectable ctDNA, a trend was still observed, but the difference was no longer significant (median DFS 7.0 months vs median DFS not reached, p = 0.086; HR 2.673, 95% CI 0.806–8.857) (Figure 3B). A univariate analysis was performed to test the association between other clinically relevant clinico-pathological variables and DFS after CRS-HIPEC. A PCI higher than 10 (5.0 months versus median DFS not reached, p = 0.035) and the presence of liver metastases (2.0 months versus 12.0 months, p < 0.001) were found to have a significant association (Table S3).None of the samples taken initially after CRS-HIPEC had detectable levels of ctDNA, except for patient L-27. In this case, recurrences to the lungs and the spleen were diagnosed 7 months after CRS-HIPEC, which suggests the potential presence of systemic micro-metastases at the time of CRS-HIPEC. Due to the exploratory approach of this study, only 19 of the 24 patients provided additional samples during the follow-up period. Of the five patients who had a systemic recurrence, ctDNA was detected in four out of four patients who provided a follow-up sample (Figure 4). Of the nine patients who had a loco-regional recurrence, ctDNA was detected in one out of eight patients who provided a follow-up sample. Detection of ctDNA in these samples either occurred at or after diagnosis of a recurrence. Circulating tumor DNA was not detectable in any of the seven patients who provided a sample and did not have a recurrence during the follow-up period, which suggests a negative predictive value. Notably, levels of background cfDNA were significantly higher during the 2 months following CRS-HIPEC (p ≤ 0.001) and returned to preoperative levels by 7 months, commensurate with tissue damage associated with the procedure (Figure S1).This pilot study demonstrates the promising feasibility of ctDNA as a prognostic marker of recurrence in CRC patients with PMs who are eligible for CRS-HIPEC. Preoperative detection of ctDNA could influence the decision to undergo CRS-HIPEC but larger studies are required to validate the clinical utility of this approach. In this study, ctDNA was detected preoperatively in 33% of patients and there was a trend between detection and reduced DFS. However, it was not possible to quantify the extent of PMs based on ctDNA detection and cfDNA quantification. This is the first study to apply preoperative tissue-guided ctDNA analysis exclusively to patients selected for CRS-HIPEC to treat PMs of CRC.A comparable study by Baumgartner et al. (2018) used an NGS-based approach to investigate preoperative ctDNA in patients who underwent surgery to treat PMs of various cancer types. This study reported an overall ctDNA detection rate of 39% across all the included cancer types, with a solid-tissue concordance of 35.3% when comparison was possible [30]. In our investigation, the ctDNA detection rate in patients with isolated PMs was similar at 33%. Despite patients having a form of metastasis, these ctDNA detection rates are more in line with those seen in stage I CRC (40%) [43]. There is little evidence that ctDNA detection in early-stage CRC has prognostic value [28], but interestingly in our study it was found to be predictive of a reduced DFS, despite a comparatively low detection rate. PMs are understood to spread to the peritoneum through a local form of dissemination [2,44], rather than through the circulation [45]. The metastatic site is an important determinant of ctDNA detection, as CRC metastases to the lung, for example, have a lower VAF than to liver or lymph nodes [46]. If CRC patients have isolated PMs, the VAF tends to be lower compared to patients with no PMs or PMs with involvement from other metastatic sites [45]. We hypothesize that ctDNA from isolated PMs is usually poorly shed into the plasma compared to primary or systemically-spread CRC, which could explain the comparatively low preoperative ctDNA detection rate and lack of a significant relationship to the PCI score. Preoperative detection of ctDNA could occur because these patient’s PMs shed greater than usual quantities of ctDNA and are more prone to systemic spread. This might also explain why seven out of eight patients who had detectable ctDNA preoperatively were later diagnosed with a recurrence, despite the intended removal of all PMs by CRS-HIPEC. Alternatively, there may have been an additional ctDNA contribution from un-diagnosed systemic metastases that already existed below the detection threshold of imaging techniques at preoperative assessment. CtDNA from these metastases may have been masked in the weeks following CRS-HIPEC, due to increased background cfDNA from tissue damage and inflammation [47], then become detectable again at diagnosis of recurrence. Another reason for postoperative detection could have been the presence of minimal residual disease after CRS-HIPEC. Recent studies have shown patients with localized CRC to be at a higher risk of recurrence if ctDNA was detected after resection [48]. Similarly, locally advanced rectal cancer patients could be stratified into groups at high or low risk of recurrence based on postoperative ctDNA detection [49].If preoperative ctDNA detection is validated as a marker of a more invasive tumor type or of undiagnosed metastases, the use of a local treatment such as CRS-HIPEC as a stand-alone therapy may have limited curative potential in this situation. A more appropriate treatment for these patients would be (neoadjuvant) systemic therapy in addition to CRS-HIPEC, an approach used in the PRODIGE-7 trial [14] and is currently under investigation in the CAIRO-6 study [50]. It may even be appropriate to withhold CRS-HIPEC, as the median DFS of 6 months in patients with detectable preoperative ctDNA is similar to the typical physical and quality of life recovery time [51].Our results with a limited number of available follow-up blood samples hints at the potential of ctDNA analysis to support recurrence monitoring after CRS-HIPEC. Detection of ctDNA in samples from all four patients with a systemic- and one out of eight with a loco-regional recurrence, either at or after the diagnosis of recurrence by standard techniques, suggests a confirmatory role for ctDNA. If validated by a larger study, this finding could lead to an earlier start of palliative treatments to reduce disease-related symptoms, or indicate a repeat HIPEC with curative intent in patients with oligo- and loco-regional recurrence [18].Despite the limited number of tissue-guided mutations tested in cfDNA by ddPCR, the detection rate was comparable to Baumgartner et al. (2018) who screened for mutations with an NGS panel. We identified a median of 1.5 mutations per patient (range 1–4, n = 30) in PMs, with all tested mutations found back in samples that contained ctDNA, except for in one follow-up sample from patient L-18. Baumgartner et al. (2018) detected a median of 2 mutations (range 1–6, n = 7) in ctDNA of CRC patients with PMs [30]. A comparable study of ctDNA in patients with metastatic or recurrent CRC had a similar detection rate (median 2 mutations, range 0–25, n = 74) [52]. An explanation for our slightly lower detection rate could be that the TSACP panel was limited to 48 genes, so may have missed less common mutations. This, plus the low sample numbers, could also explain the difference between the mutations detection in our study and those detected in frequently mutated genes described in the literature. A limitation of our tissue-guided approach is that patients who had not undergone a primary tissue resection before CRS-HIPEC—in this case 5/30 patients—would need a biopsy of PMs to enable ctDNA analysis. However, available biopsied tissue does not guarantee that mutations will be identified, as three patients were excluded from the study due to a lack of targetable mutations in their PMs. If biopsied tissue is not available or mutations are not detected, an NGS panel of commonly mutated genes should be considered to screen for ctDNA, provided a sensitivity comparable to ddPCR can be reached.An alternative approach to non-invasive PM assessment is through improved imaging modalities. Two recent studies showed the feasibility of DW-MRI to improve the detection and quantification of colorectal PMs compared to conventional CT imaging [25,53]. However, DW-MRI could underestimate the extent of metastases, especially of signet ring cell or mucinous adenocarcinomas [25,53]. In our study, 9/30 (30%) patients had these sub-types so may not have benefitted from DW-MRI alone. Interestingly, a study by Vidal et al. 2017 showed patients with mucinous tumors to have a lower than expected ctDNA VAF [45]. This poses an additional challenge for the detection and quantification of tumors with mucinous histology. Molecular characterization of the tumor through a solid or liquid biopsy, as demonstrated in our study, would still be required to enable the use of targeted therapies, for example. Confirmatory studies are necessary before DW-MRI can be implemented as a standard technique for analysis of PMs.The clinical variability of patients in our pilot study supports the feasibility of our approach. Inclusion of patients with a primary tumor in situ or liver metastases reflects the clinical reality of CRS-HIPEC candidates, even though these factors may have increased the chance of ctDNA detection. To the best of our knowledge, this is the first study to analyze the mutation status of PMs and both pre- and postoperative plasma samples in a cohort of CRC patients eligible to receive CRS-HIPEC. Investigations are needed to understand the biological factors affecting ctDNA representation in plasma, particularly in cases of localized dissemination such as PMs. Although a pilot study, the findings presented here are potentially practice changing and should be validated by larger clinical studies.This pilot study demonstrates the feasibility of ctDNA as a prognostic marker in the clinical management of CRC patients with PMs. If ctDNA is detected preoperatively, patients may experience greater benefit from chemotherapy in addition to, or instead of, CRS-HIPEC. Additionally, the approach outlined here could support the detection of recurrences along-side conventional diagnostic methods during follow-up. To allow clinical implementation, these results require confirmation by larger trials and ultimately, by prospective studies in which treatment decisions are based on ctDNA analysis.The following are available online at https://www.mdpi.com/2077-0383/9/6/1738/s1, Figure S1: Mean cfDNA yields of plasma samples, Table S1A: Mutations detected in peritoneal metastasis solid-tissue, Table S1B: ddPCR results, Table S1C: mean cfDNA yields, Table S1D: gBlock sequences, Table S2: Preoperative cfDNA detection and clinical and pathological variables, Table S3: Clinical/pathological characteristics and Disease Free Survival, Supplementary materials and methods.Conceptualization, N.C.T.v.G., B.Y. and J.B.T.; Data Curation, J.J.B. and N.R.S.; Formal analysis, J.J.B. and N.R.S.; Funding acquisition, G.K., N.C.T.v.G., B.Y. and J.B.T.; Investigation, J.J.B., N.R.S., S.B., P.P.E. and D.A.M.H.; Methodology, D.A.M.H., M.K., D.M.P., N.C.T.v.G., B.Y. and J.B.T.; Project administration, J.J.B., N.R.S., B.Y. and J.B.T.; Resources, B.Y.; Supervision, B.Y. and J.B.T.; Validation, J.J.B. and N.R.S.; Visualization, J.J.B., N.R.S. and S.B.; Writing—Original Draft, J.J.B. and N.R.S.; Writing—review & editing, S.B., P.E., S.L.V., D.A.M.H., M.K., D.M.P., G.K., N.C.T.v.G., B.Y. and J.B.T. All authors agreed to the manuscript in its published version and contributed to the Review & Editing. All authors have read and agreed to the published version of the manuscript.This project was supported by the Vrije Universiteit Medical Center, Cancer Center Amsterdam Foundation. The funding body had no role in the study design, execution of the research activities, or the decision to submit the manuscript for publication.DAMH has been on the speakers’ bureau of QIAGEN, serves occasionally on the scientific advisory boards of Pfizer and Bristol-Myers Squibb, and is minority shareholder of Self-screen B.V., a spin-off company of Amsterdam UMC, Vrije Universiteit Amsterdam. No other authors declare a conflict of interest.Flowchart of patient classification after initial inclusion in the study. The reasons for patient exclusion prior to intraoperative assessment are described. After the intraoperative assessment, patients either underwent Cytoreductive Surgery and Hyperthermic Intraperitoneal Chemotherapy (CRS-HIPEC) or received an open–close procedure. PMs: Peritoneal Metastases.(A) An overview of the genes in which a somatic mutation was identified in peritoneal metastases (PMs) and tested for in the corresponding preoperative cfDNA sample. A black square indicates that the mutation was identified in both PMs and cfDNA; a grey square indicates a negative result for the corresponding mutations in the cfDNA. All cfDNA samples were tested using the KRAS G12/13 screening kit. (B) Comparison of the relative proportion of mutations detected in PMs and preoperative cfDNA samples. (C) Concentration of cfDNA samples. A black dot denotes that ctDNA was detected; lines indicate the median and the interquartile range. HRM-sequencing: High-Resolution Melting assay followed by Sanger sequencing.Survival analysis of patients who received CRS-HIPEC. (A) All patients (n = 24). (B) All patients except those diagnosed with preoperative liver metastases (n = 22). DFS: Disease-free Survival.Tissue-guided ctDNA analysis of plasma samples taken from 14/24 patients during follow-up after CRS-HIPEC. Ten patients did not provide a sample. Patients are categorized by the location of metastases before or during CRS-HIPEC and the location of recurrence(s) during follow-up. The blue color is to highlight a positive result.Baseline characteristics of all patients in the study and the selection of patients that underwent Cytoreductive Surgery and Hyperthermic Intraperitoneal Chemotherapy (CRS-HIPEC).SD: Standard Deviation. BMI: Body Mass Index. ASA: American Society of Anesthesiologists Physical Status Classification System. TNM: TNM classification of malignant tumors. PCI: Peritoneal Cancer Index.

Med-MDPI/jcm_2/jcm-09-06-01739.txt

@@ -0,0 +1 @@
+Sleep disordered breathing (SDB), as defined by the Apnea Hypopnea Index (AHI), is a highly prevalent disturbance in end stage kidney disease. SDB improves early on after renal transplantation but long-term changes in AHI in these patients have not been studied. We studied the long-term changes in AHI in a series of 221 renal transplant patients (mean age: 47 ± 12 years; 70% males) over a median follow up of 35 months. Data analysis was made by the generalized estimating equations method (GEE). On longitudinal observation, the median AHI rose from 1.8 (Interquartile range: 0.6–5.0) to 2.9 (IQR: 1.0–6.6) and to 3.6 (IQR: 1.7–10.4) at the second and third visit, respectively (p = 0.009 by the GEE model and the proportion of patients with moderate to severe SDB rose from 8% to 20%. Longitudinal changes in minimum oxygen saturation (minSaO2) mirrored those in the AHI. In adjusted analyses, repeated measurements of BMI (p < 0.009) emerged as the strongest independent longitudinal correlate of AHI and MinSaO2. The AHI worsens over time in renal transplant patients and longitudinal changes of this biomarker are directly related to simultaneous changes in BMI. Overweight/obesity, a potentially modifiable risk factor, is an important factor underlying the risk of SDB in this population.Sleep disordered breathing (SDB) is a complex disorder which impacts on neurocognitive function, metabolism, cardiovascular health and quality of life and significantly increases the risk of cardiovascular events in the general population [1]. In over 3 million US veterans, incident SDB is associated with higher mortality, incident coronary heart disease, stroke, chronic kidney disease (CKD), and faster CKD progression [2]. SDB is increasingly common when CKD progresses to more severe stages [3] and associates with mortality in the pre-dialysis CKD population [4]. In patients maintained on chronic dialysis, the prevalence of this disturbance attains 56% [5] and entails a high risk of cardiovascular mortality [6,7].In dialysis patients, SDB is a multifactorial complication which in part depends on compromised upper airway stability secondary to rostral fluid shift overnight [8,9], chronic metabolic acidosis [10], altered central and peripheral chemosensitivity, and probably on the accumulation of still uncharacterized uremic toxins [11] that reduce airway muscle tone. Kidney transplantation restores renal function, corrects fluid overload, metabolic alterations and uremic toxicity and may therefore reverse SDB. Accordingly, in a case–control study the prevalence of SDB among renal transplant patients was similar to that in the general population [12]. Furthermore, with the exception of the study by Tandukar et al., [13] most studies that compared SDB before and after transplantation [12] [14,15,16,17,18] documented that this disturbance improves after kidney grafting.Risk factors change dramatically after renal transplantation [19,20,21,22]. Among the factors related to SDB, fluid overload [19] and anemia [20] are corrected by renal transplantation early on. On the other hand, after transplantation there is a surge of traditional risk factors [21] of which obesity gradually becomes a prevalent risk factor in this population [22] and fluid overload may re-emerge at later stages [23] in these patients. The long-term dynamics of risk factors for SDB post-transplantation suggests that SDB, after the initial improvement, may not remain stable over time in these patients. In studies performed so far, sleep studies were only performed in the early phase after transplantation. However, there is still no longitudinal study based on repeated sleep studies investigating the long-term evolution of SDB in this population. In particular, the role of overweight and obesity, a major risk factor for SDB which frequently develops after renal transplantation [22], has never been tested in this population in a study with longitudinal design. The issue is important because SDB is a strong predictor of cardiovascular disease events in the general population [24] and entails a high death risk in patients with CKD not on dialysis [4], a category including renal transplant patients [25].In this study, we investigated whether the improvement in SDB achieved after transplantation is maintained in the long term. With this scope, we enlarged the previous survey at our center [12] from 163 to 221 transplant patients and performed 404 polygraphic sleep studies over a median follow up of 35 months and herein we describe the long-term changes over time in SDB in the same cohort.The protocol was in conformity with the local ethical guidelines of our institution and with the Declaration of Helsinki, and informed consent was obtained from each participant.This is a longitudinal study (median follow up 35 months) embedded in the clinical practice at our Renal Transplantation center. The study includes a baseline sleep recording session performed at least 4 months after transplantation, i.e., when patients had a stable condition, and two or more additional recording sessions at various time intervals.The study cohort was composed of 221 renal transplant patients on follow up at the Nephrology, Dialysis and Transplantation Unit of Reggio Calabria, Italy. The flow chart of this cohort is shown in Figure 1.Polygraphic sleep study: All recording sessions were carried out while patients were in a stable condition and without intercurrent clinical problems. Polygraphic recordings were performed with patients sleeping in a quiet single bedroom or in a double room where the room mates were in stable conditions and did not need night care [12]. Bedtime and awakening times were at each subject’s discretion and the start of the sleeping time was declared by the patient, which was checked based on gross movements records and confirmed by a nurse who periodically checked the patient. The study consisted of continuous polygraphic cardiorespiratory recording from surface leads for electrocardiography and from non-invasive sensors for nasal airflow, thoracic and abdominal respiratory effort, oxyhemoglobin level (finger-pulse oximeter) and gross movements records. Until 2009, we used the Pamela Sleep Recorder by MEDATEC, Medigas, Milan, Italy and, as of 2009, the Somté recorder by Compumedics, Victoria, Australia distributed in Italy by MEDIGAS, Assago-Milan, Italy. For nasal airflow, the Pamela recorder used a thermistor, while the Somté recorder uses a flowmeter. The transducers and lead wires permitted normal positional changes during sleep. The polygraphic recording was terminated after final wakening which coincided with the time adopted for the calculation of the total sleep time. Laboratory data: Laboratory data either collected during the same day of the sleep study or during the visit preceding the same recording (in general within 2 weeks) were abstracted from electronic clinical files of our Unit. Serum lipids, glucose, albumin, phosphate, PTH, and hemoglobin were measured by standard methods in the routine clinical laboratory. C-reactive protein (hs-CRP) was measured by a high sensitivity method (Dade Behring, Marburg, Germany). Serum creatinine was measured by an automated technique based on the Jaffe chromogen method (calibrated to the Isotope Dilution Mass Spectrometry standard, IDMS) implemented in an auto-analyzer. The Modification of Diet in Renal Disease (MDRD) equation developed by Levey et al. [26] estimated the glomerular filtration rate (eGFR).Records were then scored for sleep, breathing, oxygenation, and movements in 30-second periods. An abnormal breathing event during sleep was defined as a complete cessation of air flow lasting 10 s or more (apnea) or a discernible reduction in respiratory airflow accompanied by an oxygen saturation drop of >4% (hypopnea) [26]. The average number of episodes of apnea and hypopnea per hour of sleep (the Apnea Hypopnea Index, AHI) was calculated as the summary measure of sleep-disordered breathing. At all study visits, for the categorical analysis, we applied the AHI cut off points adopted in the Wisconsin study [27]: 0–< 5 apnea-hypopnea episodes (normal sleep pattern), >5–< 15 episodes (mild SDB), >15 < 30 episodes (moderate SDB) and ≥30 episodes (severe SDB). Nocturnal hypoxemia was measured by considering the minimal and average O2 saturation as well as the number of O2 desaturation episodes (Oxygen Desaturation Index, ODI) during nighttime. A fall in oxygen saturation was considered significant if the oxygen desaturation during the episode was ≥4% of the surrounding values. Nocturnal apneas were further classified according to recommendations by the American Academy of Sleep Medicine Task Force [28]. Briefly, an apnea episode was classified as ‘central’ when it was associated to a fall in inspiratory muscles activity following an exhalation and as ‘obstructive’ when inspiratory muscle activity was present without airflow. An observer that was kept blind to the scope of the study analyzed all recordings.Data are summarized as mean and standard deviation, median and interquartile range, or as percent frequency as appropriate. Comparisons among AHI categories were performed by One Way ANOVA (for normally distributed continuous variables), Kruskall–Wallis Test (for non-normally distributed continuous variables), or Chi-Square Test, as appropriate. The relationship between non-normally distributed variables was investigated by Spearman Rank Correlation coefficient (rho) and p values. The study had no missing data for the variables included in the multivariable models.The relationship between repeated polygraphic sleep measures (namely: AHI, ODI, minimum O2 and average nocturnal O2 saturation) and time was investigated by generalized estimating equations (GEE) [29]. The correlates of polygraphic sleep data, which significantly changed over time, were analyzed by multiple GEE models including all factors which differed among AHI groups (see Table 1) with p ≤ 0.10. In these models, BMI and eGFR were introduced as repeated measurements. Based on the distribution of the key dependent variables (namely: AHI and minimal nocturnal O2 saturation), GEE models were fitted by adopting a Tweedie distribution [30] with a log10 based link function. In GEE models, data are expressed as regression coefficients, 95% CI and p value. Data analysis was performed by SPSS® 24.0 (IBM Corporation, Armonk, NY, USA) for Windows®.3.The total transplant population on follow up at our institution during the enrolment years (March 2004–June 2015) counted 340 patients. Among these, 47 patients refused to participate in the study and the remaining 72 did not complete polygraphic recordings for technical reasons or polygraphic recordings could not be arranged for logistic reasons (Figure 1). Thus, the final study population was composed of 221 renal transplant patients.As shown in Table 1, the mean age was 47 ± 12 years and 70% were males. Nine percent were diabetics. The causes of CKD were: Glomerulonephritis in 82 cases (37.%), diabetic kidney disease in 18 cases (8%), cystic kidney diseases in 20 cases (9%), interstitial nephropathy in 13 cases (6%), congenital anomalies of the kidney and urinary tract (CAKUT) in 10 cases (4%), hypertensive nephrosclerosis in 2 cases (1%), hereditary nephropathies in 2 cases (1%), acute kidney injury (AKI) in 2 cases (1%), and other causes/unknown in the remaining 72 cases (33%). Most patients received the organ from cadaveric donors (88%) and a minority from living donors (12%). The majority (71%) were on triple immunosuppressive therapy (three drug combinations among cyclosporine, steroids, azathioprine, tacrolimus, sirolimus and mycophenolate) and the remaining patients were on double therapy (28%) or on mono-therapy (1%) with these agents.Twenty-four patients had experienced cardiovascular events. In detail, eighteen had suffered from one event (myocardial infarction in three, stroke in four, transient ischemic attack in three, angina in two, arrhythmia in four, and peripheral vascular disease in two patients) and the other six had had two or more of these events. Ten percent of the patients were active smokers and 41% past smokers. One hundred and ninety-five patients were on anti-hypertensive therapy. Sixty-seven patients were on monotherapy with calcium channel blockers (n =12), beta-blockers (n = 25), ACE inhibitors/angiotensin II receptor blockers (ARB) n = 24), sympatholytic or vasodilatatory agents (n = 5) and diuretics (n = 1). One hundred and twenty-eight patients were on multiple therapies with various combinations of these drugs. Ten percent of the patients were being treated with erythropoietin stimulating agents (ESA) and 42% were on treatment with statins. Estimated glomerular filtration rate (eGFR) was on average 56.1 ± 20.5 mL/min/1.73 m2 (range 7.2–135.9 mL/min/1.73 m2). Office BP was on average 132 ± 16/78 ± 10 mmHg (Table 1).At baseline, the median value of theAHI was 1.8 episodes/h (interquartile range: 0.6–5.0 episodes/h). One hundred and sixty-six patients (75%) had an AHI <5; 37 patients (17%) had an AHI ranging from 5 to <15 and the remaining 18 patients (8%) had an AHI >15. This prevalence of SDB is substantially similar to that observed in the general population matched for age, gender and BMI [29] (Figure 2).The AHI was directly related to age (Spearman rho = 0.24, p < 0.001) and BMI (rho = 0.28, p < 0.001) (Table 1). The eGFR was similar across the three AHI strata (p = 0.10). As expected, the ODI paralleled the AHI while the average O2 saturation and the median minimum O2 saturation gradually reduced across AHI categories denoting SDB of increasing severity (Table 1). In Supplementary Table S1, drug treatments are reported across the same categories.AHI was directly related to the use of statins, calcium antagonists, sympatholytic agents/vasodilators and the total number of anti-hypertensive drugs and inversely to the use of ESA and steroids. However, these associations were no longer significant after simple data adjustment for the BMI (p ranging from 0.09 to 0.89).Over a median follow up of 35 months (whole range 4 to 110 months; interquartile range 24–46 months), 404 polygraphic recordings were performed. In detail, 82 patients were only studied at baseline, 139 patients had at least two polygraphic studies and 44 had three studies. Face to face comparison of patients only studied at baseline (n = 82) did not differ from the remaining patients (n = 139) as to age (47 ± 11 vs. 47 ± 12 years, p = 0.74), 24 h systolic BP (126 ± 14 vs. 124 ± 11 mmHg, p = 0.48), BMI (26.3 ± 3.9 vs. 25.6 ± 3.4 k/m2, p = 0.18), diabetes (9.8% vs. 8.6%, p = 0.78), male sex (69.5% vs. 70.5%, p = 0.88), and background CV comorbidities (13.4% vs. 9.4%, p = 0.36).The median AHI rose from a baseline median value of 1.8 episodes/h (IQR: 0.6–5.0) to 2.9 episodes/h (IQR: 1.0–6.6) at the second visit (time spanning from baseline, median: 32.2 months, IQR: 32.2–41.0 months) and to 3.6 episodes/h (IQR: 1.7–10.4)) at the third visit (time spanning from baseline, median: 52 months, IQR: 36.8–67.3 months) (p = 0.009, calculated by the GEE model). In a sensitivity analysis restricted to the subgroup of patients with three sleep recordings (n = 44) the worsening AHI (baseline: 1.3 episodes/h, IQR 0.50–2.3; first visit, 2.0 episodes/h, IQR 0.7–3.3; second visit, 3.6 episodes/h IQR 1.7–8.5) was similar to that observed in the primary analysis (see above). In a categorical analysis, the prevalence of moderate to severe SDB rose markedly across the follow up, from 8% (18/221 patients) at the first visit to 13% at the second visit (18/139 patients) and 20% (9/44 patients) at the last visit (Figure 3).The time trend of minimal nocturnal O2 saturation mirrored the AHI trend (first visit, 87 ± 7%; second visit, 87 ± 7%; third visit, 85 ± 11%, p = 0.048 by the GEE model). Neither the ODI (p = 0.27), nor the average nocturnal O2 saturation (p = 0.14) changed significantly over time. To identify risk factors for SDB worsening over time, we built a GEE model of longitudinal changes of the AHI (Table 2) including age, gender, diabetes, and repeated measures of the BMI and eGFR, i.e., the variables that were associated with the AHI at baseline with a p value <0.10. In this analysis, age, gender, and repeated measurements of BMI emerged as the strongest correlates of the longitudinal changes of the AHI. In a multivariate GEE model looking at changes in minimal nocturnal O2 saturation, age, gender and repeated measurements of BMI were again independent correlates of this indicator of SDB (Table 2).The recorded change over follow up (Pamela → Somtè, see methods) was associated per se with the change in AHI and minimal O2 saturation over follow up but it did not confound the longitudinal relationship of the BMI and other variables with the same outcome measures (Supplementary Table S2). Accordingly, in a formal effect-modification analysis testing the interaction term sleep recorder change (Pamela → Somté) and study visit in the same model described in Table 2, this interaction term was largely insignificant both for the AHI ((p = 0.62) and the minimal O2 saturation (p = 0.74). The BMI and AHI evolved in parallel (p = 0.002) in patients for whom the BMI increased by >0.5 kg/m2 but were dissociated in those for whom the BMI remained stable or declined (p = 0.48).We found that 8% of the renal transplant patients studied, after having achieved clinical stability after kidney grafting, had moderate to severe SDB and an additional 17% had mild SDB. These figures are substantially less than those typically seen in dialysis patients where moderate to severe SDB has a 56% prevalence [5] and are similar to those in age-, sex- and BMI-matched individuals in the general population [12,31]. However, we also found that SDB worsens in these patients over time and, after a median follow up of 3 years, 20% of the renal transplant patients had moderate to severe SDB. Furthermore, a BMI change over time was the sole modifiable risk factor underlying SDB worsening in this population.Restored renal function after transplantation normalizes several risk factors associated with SDB in dialysis patients. Fluid overload, uremic toxicity and other factors are rapidly corrected after transplantation. However, after the early improvement, fluid overload re-emerges over time and at 5 years post-transplantation the proportion of transplant patients with fluid overload attains 30% [23]. Similarly, body weight gradually increases over time post-transplantation and obesity is recognized as a highly prevalent (30% to 50%) [32,33], high-risk problem [32,33,34] in the renal transplant population. Thus, after the early improvement, renal transplant patients may be exposed to a high risk of SDB recurrence. The initial beneficial effect of renal transplantation on SDB is generally considered to be a stable improvement and perhaps for this reason SDB is not considered as a possible risk factor for adverse cardiovascular outcomes in the transplant population [21,35].Given the strong link between SDB, death [4], and cardiovascular events [2,6,7] in the CKD population, testing the long-term evolution of SDB in renal transplant patients is a relevant issue for our understanding of the potential risk of SDB for these outcomes in the renal transplant population. With this study, we confirm our previous findings in a larger population [12] that, in an early phase, stable renal transplant patients have no excess risk of SDB as compared to well-matched individuals in the general population. However, we also found a progressive worsening of this alteration after long-term longitudinal observation. Indeed, the prevalence of moderate to severe SDB in the baseline study was only 8% but increased to 20% over a median follow up of 3 years, which is more than double that registered at baseline and in the coeval general population [29]. In a large study with 2921 elderly people in the general population, a 20% prevalence in SDB was only observed in patients between 65 and 72 years of age [36] i.e., in people about 20 years older than the renal transplant patients included in the present study. The recurrence of SDB after transplantation may have clinical implications. Indeed an AHI between 4 and 12, an alteration we found in one out of four male and in one out of 11 female transplant patients in this study, entails a 75% excess risk of death and stroke in patients with suspected sleep apnea [37]. Of note, we found that SDB worsening in transplant patients over time associated with simultaneous changes in the BMI. BMI is a robust causal risk factor for SDB in the general population [38]. A link between the BMI and SDB in transplant patients emerged in a cross-sectional study by Molnar et al., [39] and in a survey by Mallamaci et al. [12]. This longitudinal study is the first to focus on the long-term changes of AHI after transplantation. We enrolled a number of patients (n = 221) about 2.8 times higher than the aggregate population (n = 80) of previous studies that measured AHI just before and a relatively short time after renal transplantation [13,16,40,41]. Our observations are based on over 400 polygraphic recordings distributed over a median follow up of 3 years. On longitudinal analysis, we found an independent relationship between the BMI and the AHI. Longitudinal studies provide more robust information than cross sectional studies for the assessment of causality [42]. Thus, the present study supports the hypothesis [12,39] that a high BMI is a causal risk factor for SDB in transplant patients.This study has limitations. One third of the patients were only studied at baseline and only 44 patients had three sleep recording studies. However, the generalized estimating equations model adequately deals with the unequal number of patients at various time points and with the variable time interval between successive recordings [29]. Our population includes only Caucasian patients in a single transplant center. Therefore, our results need to be confirmed by studies in other transplant center enrolling patients of other races and ethnicities. Another limitation is the fact that we used two polygraphic recorders with different airflow sensors (thermistor and airflow pressure detector, respectively). However, the progressive rise of AHI and its relationship with BMI was largely independent of the type of polygraphic sleep recorder. Furthermore, for technical reasons we could not produce an analysis according to apnea type and the time spent with SaO2 < 90%. In addition, we did not measure body fluids. Therefore, we could not test the relevance of fluid overload as a long-term risk factor for SDB. Finally, our data are observational in nature. The randomized clinical trial is the standard for testing causal hypotheses. Trials focusing on renal transplant patients would be important to understand the effects of obesity on health in the specific context of this population. Future trials targeting obesity in transplant patients may include polygraphic recordings to test the hypothesis that the longitudinal link between BMI and SDB observed in the present study is causal in nature.In conclusion, this longitudinal study shows that sleep disordered breathing worsens over time in renal transplant patients. In the context of the present study, the BMI emerged as the sole modifiable risk factor underlying SDB worsening in this population.The following are available online at https://www.mdpi.com/2077-0383/9/6/1739/s1, Table S1: Concomitant therapies in the whole study population and in patients divided according to the AHI., Table S2: Adjusted multiple generalized estimating equations (GEE) of AHI and minimal nocturnal O2 saturation over time.C.Z. conceived this study with F.M. G.T. designed the analytical plan. C.Z., F.M. and G.T. prepared the first draft of the study which was critically reviewed and approved by all co-authors. R.T. was the research technician responsible for all polygraphic recordings and 24 h ABPM measurements and M.C.V. was the second research technician involved in this study. G.D. made the statistical analysis with G.T. G.P. and M.C.S. were responsible for the maintenance of the study database and C.M. carefully checked the quality of the same database. All authors have read and agree to the published version of the manuscript.This research received no external funding.The authors declare no conflict of interest.Flow chart of patients included in the study.AHI stratified by gender, and BMI in transplant patients in the present study (grey circles) and in individuals in the general population (white circles) described in [29].Prevalence of moderate to severe sleep disordered breathing (SDB) across the study in the whole cohort, in patients who had at least 2 visits and in those with at least 3 visits.Main demographic, clinical and biochemical baseline characteristics of patients grouped by baseline apnea-hypopnea index. * ANOVA; ^ Chi Square; # Kruskall–Wallis.Data are expressed as mean ± SD, median and interquartile range or as percent frequency, as appropriate.Adjusted multiple generalized estimating equations (GEE) of Apnea Hypopnea Index (AHI) and minimal nocturnal O2 saturation over time. The model includes all univariate correlates of AHI at baseline with p < 0.10 (age, gender, diabetes as well as the BMI and estimated glomerular filtration rate (eGFR). BMI and eGFR are tested as repeated measures). Data are regression coefficients, 95% CI and p value.* BMI and eGFR are introduced into the model as repeated measurements.

Med-MDPI/jcm_2/jcm-09-06-01740.txt

@@ -0,0 +1 @@
+The objective of this study was to evaluate whether the extent of tumor resection and free flap reconstruction influences functional outcome and complications in patients with solid malignancies of the cheek. Therefore, we retrospectively assessed recipient site complications and functional outcomes in 47 patients with solid malignancies of the cheek who underwent either partial (n = 30; 63.8%) or full-thickness (n = 17; 36.2%) cheek resection with free flap reconstruction. Complications occurred in 12 (70.6%) patients after full thickness resections with creation of through-and-through defects compared to 14 (70.6%) patients with partial defects (p = 0.138). Among those 26 patients (55.3%), major recipient site complications, like development of salivary fistula or free flap loss, were observed in 10 (21.3%) and 2 (4.3%) cases, respectively, while minor complications, like wound dehiscence and local infections, were found in 14 (29.8%) and 9 (19.1%) patients. Complications were noticed particularly after reconstruction of suborbital defects (69.2%; p = 0.268), of which occurrence of salivary fistulae was the most common (46.2%; p = 0.035). Similarly, functional outcomes including oral incompetence, ectropion, and trismus were not affected by the extent of resection (p = 0.766). However, oral incompetence was higher in patients with tumors originating from the oral cavity (p = 0.020) and after the performance of mandibulectomy (p = 0.003). Overall, there was no difference in functional outcome or recipient site morbidity between tumor resections resulting in full-thickness and partial defects.In addition to trauma and burns, resection of malignancy represents the main cause for cheek defects requiring reconstruction [1,2]. Malignancies can arise from the oral mucosa or the skin with infiltration of the submucosa, loose connective tissue, and mimetic muscles of the cheek [3,4]. Occasionally, these malignancies can infiltrate or even arise from neighboring structures, like the nasal cavity or the parotid gland.Squamous cell carcinomas (SCCs) originating from the oral mucosa or skin are the most common tumor entity of the cheek and surgical tumor resection represents the mainstay of therapy [5,6]. Depending on tumor size and level of invasiveness, extended resections may end up in partial or even though-and-through defects, requiring more complex reconstructions with free flaps [4,7]. Recent publications regarding buccal SCCs further indicate that more extensive tumor resections are even associated with improved oncologic outcome [3,8].With respect to functional and aesthetic aspects, the cheek region represents the major part of the lateral facial unit that is essential for mimic, facial expression, and maintenance of oral competence [4]. Large composite defects of the cheek also often result in deficits of the facial nerve, further adding to the aesthetic and functional morbidity of the resection. Hence, cheek reconstruction is challenging and requires thorough knowledge of anatomic subsites and danger zones, corresponding function, and available reconstructive techniques to achieve functionally and aesthetically satisfactory results [4].Although studies have already reported on improved oncological outcome after more extensive resections of tumors of the cheek, functional endpoints, particularly after free flap reconstruction and associated recipient site morbidity, have not been addressed so far. It was therefore the purpose of this retrospective, multicentric study to evaluate the effect of cheek resection and free flap reconstruction on the functional outcomes (I) and associated complications (II).We conducted a retrospective, multicenter chart review of patients with solid malignancies originating in the cheek who underwent tumor resection and free flap reconstruction between 2012 and 2017. Patients were treated at the Department of Otorhinolaryngology, Head and Neck Surgery, Medical University of Vienna, Austria (Center 1), the Institute of Head and Neck Diseases, Evangelical Hospital Vienna (Center 2), the Department of Otolaryngology, Head and Neck Surgery, Icahn School of Medicine at Mount Sinai, New York, USA (Center 3), and the Department of Head and Neck Surgery, Shaukat Khanum Memorial Cancer Hospital and Research Centre, Lahore, Pakistan (Center 4). All surgeries were performed by experienced head and neck surgeons who had completed previously head and neck fellowships.Data of potential patients were provided by attending centers that were further evaluated regarding appropriateness by two authors individually (S.J., B.E.). Patients had to fulfill the following inclusion criteria: carcinomas with infiltration of the cheek, including oral carcinomas, carcinomas of the nasal cavity, skin and parotid gland (I), solid carcinomas (II), radical tumor resection with free flap reconstruction (III), and primary or recurrent carcinomas (IV). Finally, 47 patients were eligible for inclusion and analysis. Ethical approval was obtained from the Ethics Committee of the participating centers prior to enrolment.From the aesthetic point of view, the cheek region can be divided into three overlapping aesthetic zones including the suborbital (I), the preauricular (II), and the bucco-mandibular (III) zone, as illustrated in Figure 1 [9]. Anatomically, the cheek consists (from inside to outside) of buccal mucosa, submucosa, loose connective tissue, muscles, the parotid gland in zone II and III, and the skin. Depending on the extent of resection, and consequently, the depth of defect, patients were dichotomized into patients with partial or through-and-through (full) defects.Depending on the size and depth of the defect, cutaneous, myocutaneous or osteocutaneous flaps were used. In cases with partial defects, harvested skin paddles were used for the inner lining of the oral cavity or for reconstruction of the skin if required. Conversely, in patients with through-and-through defects, split or full thickness skin grafts were used for the inner lining, while harvested skin paddles were used for the outer lining (Figure 2).We assessed recipient site complications and functional outcomes as main endpoints of the study within the first 6 to 12 months after surgery and free flap reconstruction. Recipient site complications were further classified either as minor complications including wound dehiscence and local infection or major complications including salivary fistula and free flap failure. Presence of ectropion, oral incompetence and trismus were used as functional endpoints. All outcomes were rated by treating head and neck surgeons.Statistical analyses were performed using SPSS software (version 22; IBM SPSS Inc., Chicago, IL, USA). Unless otherwise specified, data in the results section are shown as median ± standard deviation. Chi-square test was used to assess associations between nominal variables. In cases with expected cell counts below 5, p-values of Fisher’s exact value were reported. Moreover, the unpaired student’s t-test was used to compare means of normally distributed variables of two independent groups. Forty-seven patients, including 29 men (61.7%) and 18 women (38.3%), with a median age of 64 ± 15.2 years (range: 30y–93y), were included in this retrospective, multicentric analysis, all of whom underwent tumor resection and free flap reconstruction of the cheek. SCC was the predominant tumor histology (n = 38; 80.9%) followed by adenoid cystic carcinoma (ACC; n = 3; 6.4%), sarcoma (n = 2; 4.3%), melanoma (n = 2; 4.3%), merkel cell carcinoma (MCC; n = 1; 2.1%), and malignant adnexal skin tumor (n = 1; 2.1%). Malignancies originated from buccal mucosa, oral cavity, parotid gland, nasal cavity and skin in 22 (46.8%), 10 (21.3%), 4 (8.5%), 3 (6.4%) and 2 (4.3%) cases, respectively. With respect to aesthetic zones, tumors were located predominantly at zone I, II, and III in 13 (27.7%), 11 (23.4%), and 23 (48.9%) cases, respectively (Table 1).Patients had three T1 (6.4%), eleven T2 (23.4%), twelve T3 (25.5%), and twenty-one T4 (44.7%) tumors, respectively, with a median tumor size of 3.9 ± 1.9 cm (range: 1.0–8.8 cm). At initial presentation, 30 (63.8%), 6 (12.8%), and 7 (14.9%) patients had N0, N1 and N2 disease, while cervical lymph node classification (Nx) was unknown in 4 patients (8.5%). Altogether, we had 3 stage I (6.4%), 8 stage II (17.0%), 11 stage III (23.4%), and 25 stage IV (53.2%) malignancies (Table 1).Radical tumor resection created partial and through-and-through defects in 30 (63.8%) and 17 (36.2%) patients, respectively. Socio-demographic characteristics, including male to female ratio (21:9 vs. 8:9), age (65.1 ± 15.8y vs. 60.7 ± 14.0y), and body-mass-index (25.1 ± 4.4 kg/m2 vs. 24.5 ± 4.6 kg/m2) did not significantly differ in patients with partial and through-and-through defects, respectively (p = 0.211; p = 0.388; p = 0.677). Moreover, T-classification (p = 0.901), N-classification (p = 0.372), and AJCC tumor stage (p = 0.492), did also not significantly differ between both groups (Table 1).ND was performed in 80.9% of patients. Level I–III, I–IV, II–IV, and II–III ND were done in 13 (27.7%), 9 (19.1%), 8 (17.0%), and 4 (8.5%) cases, respectively. In two patients (4.2%), the extent of ND was not indicated. Maxillectomy was necessary to perform in 19 out of 47 patients (40.4%), including partial-, hemi-, and total maxillectomy in 5 (10.6%), 9 (19.1%), and 5 (10.6%) cases, respectively. Otherwise, partial and total mandibulectomy was done in 11 (23.4%) and 1 patient (2.1%). Primary tumor resection was further accompanied by partial glossectomy and orbital exenteration in 4 (8.5%) and 3 patients (6.4%).The radial forearm free flap (RFFF) was most commonly used for cheek reconstruction (n = 15; 31.9%) followed by anterolateral thigh (ALT) flap (n = 13; 27.7%), scapular/parascapular free flap (n = 10; 21.3%), FFF (fibula free flap; n = 6; 12.8%), supraclavicular free flap (n = 2; 4.3%), and serratus anterior free flap (SAFF; n = 1; 2.1%), respectively. Altogether, cutaneous, myocutcaneous and osteocuteanous free flaps were harvested in 19 (40.4%), 12 (25.5%) and 16 (34.0%) cases, respectively (Table 2). As indicated in Table 2, RFFF was most commonly used for reconstruction of one-layer skin or mucosal defects (14 out of 19; 73.7%), while the ALT flap was mostly used as myocutaneous flap (10 out of 12; 83.3%), and the scapular/parascapular free flap for bone reconstruction (10 out of 16; 62.5%) (Table 2). In eight patients (17.0%), free flaps were oversized and too bulky. Bulkiness of the free flap occurred particularly in zone I defects (30.8% vs. 11.8%), more likely in through-and-through defects (29.4% vs. 10.0%), and after harvest of free scapular/parascapular free flaps (40.0% vs. 10.8%). However, differences failed to reach statistical significance (p = 0.288; p = 0.118; p = 0.331) and revision surgery with thinning of the free flap was performed in six out of eight patients in order to optimize final cosmetic results.We had two losses of free flaps resulting in a free flap success rate of 95.7%, while flap revision due to venous congestion was necessary in three (6.4%) cases. In those two cases with flap loss, an ALT and a latissimus dorsi flap were used for revision surgery.Donor site complications, which have been of minor concern, were observed in six patients (12.8%) who experienced wound dehiscence.Recipient site complications occurred in 26 (55.3%) patients, which was not statistically significant different between patients with partial compared to those with through-and-through defects (46.7% vs. 70.6%; p = 0.138). Wound dehiscence, formation of salivary fistula, local infections and free flap failure occurred in 14 (29.8%), 10 (21.3%), 9 (19.1%), and 2 (4.3%) cases, respectively, but did not significantly differ between both groups (Table 3).However, with regards to affected aesthetic zones, we observed formation of salivary fistula particularly in 46.2% of zone I defects, which was significantly higher compared to 9.1% and 13.0% in zone II and III defects, respectively (p = 0.035). It is noteworthy to mention that extent of resection (p = 0.136), performance of maxillectomy (p = 0.496) or mandibulectomy (p = 1.000), T-classification (p = 0.751) or size of defect (p = 0.145) had no significant impact on the development of salivary fistulae.Oral incompetence, ectropion, and trismus occurred in nine (19.1%), eight (17.0%), and six patients (12.8%), respectively. Again, the extent of resection had no significant impact on the development of any functional impairment (Table 3). Nonetheless, solely patients with malignancies originating of the buccal mucosa and oral cavity suffered from oral incompetence (22.7% and 66.7%; p = 0.020). Performance of mandibulectomy (p = 0.003), but not maxillectomy (p = 0.064), affected significantly oral competence, which was otherwise significantly associated with the occurrence of trismus (p = 0.009). In addition, the size of used free flaps was 14.7 ± 5.1 cm in patients with oral incompetence, which was significantly larger compared to 7.4 ± 2.7 cm in patients with oral competence (p = 0.008), while primary tumor size did not significantly differ (4.4 ± 1.4 vs. 3.5 ± 1.9 cm; p = 0.206) (Table 3).We have analyzed clinical outcome of 47 patients with solid malignancies of the cheek that underwent radical tumor resection with creation of partial or through-and-through defects and free flap reconstruction. Within our study, SCCs were the predominant histologic subtype (80.9%) and malignancies mainly originated from the oral cavity (80.8%). This is consistent with the literature, reporting mostly on oral carcinomas and rarely on skin carcinomas, requiring cheek reconstruction with free flaps following oncological resections [3,4]. Nonetheless, our data further display the great diversity of tumors affecting the cheek region that may hamper analysis of more homogenous subgroups with large patient numbers.In solid malignancies, and for patients with SCCs in particular, surgical tumor resection with adjuvant therapy in selected cases represents the most frequent treatment modality [5,6,10]. However, despite radical surgical resection, recurrence rates range from 45.0% to 80.0% in patients with buccal SCCs [11,12]. Several authors assume that the absence of “real” anatomic boundaries limiting tumor growth and spread might contribute to the high rate of recurrences [11,13]. This prompted Ren ZH and coworkers (2017) to perform a more extensive resection of functional anatomic buccal units to achieve a better oncologic outcome [3]. In fact, they analyzed data of 127 patients with buccal SCCs reporting on significantly better 2-year overall survival (OS: 83.3% vs. 60.1%) and disease-free survival (disease-free survival (DFS): 76.6% vs. 51.9%) in patients undergoing more extensive unit resection compared to conventional surgery [3].Although oncologic principles must supersede reconstructive desires [7], we were particularly interested in knowing how the extent of resection impacts functional outcome. Oral incompetence represented the main functional complication occurring in nine (19.1%) patients followed by occurrence of ectropion and trismus in eight (17.0%), and six patients (12.8%), respectively. This is in line with the results of other publications, reporting on problems with oral incompetence in 4.8% up to 40% of patients with cheek carcinomas [8,14,15]. It is noteworthy to mention that the extent of resection had no significant impact on functional outcome in our cohort, which is in accordance to the work of Ren ZH et al. (2017). The authors assume that insignificantly changed functional outcomes in patients with conventional surgery compared to more extensive unit resections, have resulted from the loss of function of preserved structures secondary to induction of fibrosis and loss of functional adjacent structures/attachments by tumor resection [3]. However, we found a strong association between oral incompetence and trismus in patients after mandibulectomy. This indicates that functional outcomes more likely depend on the preservation of certain anatomic structures and chosen surgical approach than on the depth of defect.Recipient site complication rate was 55.3% (n = 26), of which wound dehiscence was the most common complication occurring in 29.8% (n = 14) of cases. Although the majority of complications were of minor concern, 10 patients developed salivary fistulae (21.3%) that occurred significantly more often in suborbital zone I defects (p = 0.035). This is in accordance to former studies reporting on fistula rates of 4.3% to 27.3% of patients [8,14,16,17,18]. The development of fistulae is characteristic for maxillary reconstruction and occurs typically near to the medial canthus (zone I) due to breakdown of suture lines [16]. In alignment to that, we observed the highest rate of wound dehiscence (38.5%) in patients with zone I defects compared to 18.2% and 30.4% in zone II and III defects, respectively. We assume that gravity causes drag down of cheek skin by bulky flaps, resulting in a significant scarring, wound break, and occasionally creation of salivary fistula and ectropion with epiphora. Depending on how much the wound break and ectropion have progressed either a lateral tarsorrhaphy with a bone anchored suture, particularly at the orbital rim, or additionally to the tarsorrhaphy, a split or full thickness skin graft should be placed below the lower lid to gain more skin volume. By doing so, additional skin may prevent wound break and subsequently creation of salivary fistula as well as development of ectropion and epiphora. We think this is the best option to prevent significant aesthetic and functional sequelae in suborbital zone I defects.Until now, a number of different free flaps have been described for cheek reconstruction including the ALT [19], RFFF [19], FFF [20] and the scapular/parascapular free flap [21]. Recently, the versatility of the SAFF has been demonstrated for general head and neck reconstruction, and in particular, for cheek and tongue reconstruction [22,23]. In our study, the RFFF was used in 31.9% (n = 15) of cases followed by the ALT free flap and the scapular/parascapular free flap in 27.7% (n = 13) and 21.3% (n = 10) of patients, respectively. Among those, the RFFF was mainly harvested for cutaneous reconstruction (73.7%), the ALT flap for reconstruction of myocutaneous defects (83.3%), and the scapular/parascapular free flap for bone reconstruction (62.5%). Our flap survival rate was 95.7%, which is comparable to 95% to 96% reported in former studies [17,19].We believe that the strength of this study lies in the analysis of functional outcomes as well as complications. We see three limiting factors: first, the retrospective study design bears an inherent risk of information and selection bias. Second, the heterogeneity of our cohort with solid malignancies originating from different parts of the cheek allows only limited conclusions. Third, the lack of standard measures for functional outcomes in head and neck oncology [24] and the fact that functional and aesthetic outcomes have been rated by treating head and neck surgeons as opposed to patient reported outcomes, represent further limitations.More extensive resections with creation of through-and-through defects did not automatically correlate with worse functional outcome or higher recipient site complications. Nonetheless, especially suborbital zone I defects carried an increased risk of wound dehiscence and formation of salivary fistula, and requires, therefore, particular consideration and meticulous reconstruction. However, further prospective studies with homogenous patient cohorts are necessary to define and identify additional factors that may contribute to the functional outcome of patients with carcinomas of the cheek.Conceptualization: S.J. and B.M.E.; Methodology: S.J., S.G. and B.M.E.; Validation: R.S.; M.B.; M.F.; Formal Analysis: S.J.; Investigation: S.J., S.G., E.V., R.E., B.A.M. and M.F.; Writing—Original Draft Preparation: S.J. and B.M.E.; Writing—Review and Editing: B.A.M., R.S., M.B., M.F. and B.M.E.; Visualization: S.J. and B.M.E. All authors have read and agreed to the published version of the manuscript.This research received no external funding.The authors declare no conflict of interest.Aesthetic zones of the cheek. The cheek can be divided into three partially, overlapping aesthetic zones, comprising the suborbital (I), the preauricular (II), and the bucco-mandibular (III) zone [9]. The circle already marks the margins for planed tumor resection of a squamous cell carcinoma that is mainly located in zone III.Reconstruction of a Through-and-Through Defect. A 60-year-old female patient experienced regional failure of a sinonasal carcinoma. Radical tumor resection was performed, resulting in creation of a through-and-through defect with 4.0 × 5.0 cm in size (A–C). A free radial forearm free flap (RFFF) was used for reconstruction. The harvested skin paddle was used for the outer lining (D,F), while full thickness skin graft of the neck (asterisk) was used for inner lining (E). Postoperative (G) and 2 year follow up results are shown (H). The reconstructed region on the left side of the nasal dorsum (D–G) represents the former tumor infiltrated area that has been resected and reconstructed.Patient Characteristics.ACC, adenoid cystic carcinoma; AJCC, American Joint Committee on Cancer; MCC, merkel cell carcinoma; n, number of patients; N—Classification, cervical lymph node classification; Nx, unknown cervical lymph node status; SCC, squamous cell carcinoma; T—Classification, tumor size classification; p, p-value; a chi-square test.Free flap Reconstruction.ALT, anterolateral thigh flap; FFF, fibula free flap; RFFF, radial forearm free flap; SAFF, serratus anterior free flap.Complications and functional outcome according to depth of defect and aesthetic zone.Full, Through-and-Through defects; n, number of patients; p, p-value; a chi-square test.

Med-MDPI/jcm_2/jcm-09-06-01741.txt

@@ -0,0 +1 @@
+Microsatellite instability (MSI) represents one of the major types of genomic instability in human cancers and is most common in colorectal cancer (CRC) and endometrial cancer (EC). MSI develops as a consequence of DNA mismatch repair (MMR) deficiency, which can occur sporadically or in the context of Lynch syndrome (LS), the most common inherited tumor syndrome. MMR deficiency triggers the accumulation of high numbers of somatic mutations in the affected cells, mostly indel mutations at microsatellite sequences. MSI tumors are among the most immunogenic human tumors and are often characterized by pronounced local immune responses. However, so far, little is known about immunological differences between sporadic and hereditary MSI tumors. Therefore, a systematic literature search was conducted to comprehensively collect data on the differences in local T cell infiltration and immune evasion mechanisms between sporadic and LS-associated MSI tumors. The vast majority of collected studies were focusing on CRC and EC. Generally, more pronounced T cell infiltration and a higher frequency of B2M mutations were reported for LS-associated compared to sporadic MSI tumors. In addition, phenotypic features associated with enhanced lymphocyte recruitment were reported to be specifically associated with hereditary MSI CRCs. The quantitative and qualitative differences clearly indicate a distinct biology of sporadic and hereditary MSI tumors. Clinically, these findings underline the need for differentiating sporadic and hereditary tumors in basic science studies and clinical trials, including trials evaluating immune checkpoint blockade therapy in MSI tumors.DNA mismatch repair (MMR) deficiency is a major pathway enabling genomic instability in human cancers. MMR deficiency is characteristic for 15% of colorectal cancers (CRC) and 30% of endometrial cancers (EC) (Figure 1a) [1]. MMR deficiency is caused by the inactivation of one of the MMR genes (MLH1, MSH2, MSH6 and PMS2). As a consequence of non-functional MMR, tumors accumulate insertion/deletion mutations at repetitive microsatellite sequence stretches and therefore present with the microsatellite instability (MSI) phenotype [2].Indel mutations at coding microsatellites (cMS) inactivate critical tumor suppressor genes and promote tumor development. Simultaneously, they can trigger shifts of the translational reading frame and lead to the generation of frameshift peptides (FSPs). FSPs play a major role in the pronounced immunogenicity of MSI CRCs as they often encompass long neopeptide stretches and multiple potential epitopes that can be recognized by the immune system (Figure 1b). FSPs are not only tumor-specific neoantigens but also shared by most MSI tumors, as they are the result of certain driver mutations that are positively selected during tumor evolution, e.g., the cMS tract in exon 3 of the TGFBR2 (transforming growth factor beta receptor 2) gene is mutated in around 90% of MSI CRCs [2]. The immunogenicity of certain FSPs was comprehensively demonstrated in vitro [3,4,5].Likely because of their high load of neoantigens, MSI cancers, typically those in the colorectum, display signs of high immunogenicity, such as pronounced local T cell infiltration [6,7,8,9], elevated counts of tumor-infiltrating lymphocytes (TILs) with cytotoxic potential [9,10,11] and a high frequency of immune evasion phenomena [2,12,13]. Patients with MSI CRCs also show a more favorable prognosis in a stage-wise comparison with microsatellite-stable (MSS) tumor patients, potentially reflecting active anti-tumor immune responses [14,15]. Notably, treatments supporting the anti-tumoral immune response, such as the recently developed immune checkpoint blockade (ICB) therapy, showed great success specifically in patients with metastasized MSI cancers [16,17].Although these clinical and immunological observations have been documented by several independent studies, little attention has been attributed to the influence of the origin of MSI cancers on their immunogenicity: Whereas the majority of MSI CRCs develop sporadically, mostly due to hypermethylation of the MLH1 promoter leading to MLH1 silencing [18,19,20,21,22], about 20–30% of MSI CRCs (Figure 1a) have a hereditary background and are associated with Lynch syndrome (LS), the most common inherited CRC syndrome [23]. LS is caused by an inherited monoallelic germline alteration inactivating one of the MMR gene alleles and leading, upon a second somatic hit, to the molecular phenotype of MMR deficiency in the affected cells [24,25].The overall lifetime cancer risk in LS depends on the affected MMR gene and is estimated to vary between 50% and 80% for the high penetrance genes, MLH1 and MSH2 [26,27,28,29,30,31,32]. Clinically, LS CRC patients typically present with cancer onset at a younger age (∼45 years) compared to sporadic MSI CRC patients [27]. Due to the high lifetime cancer risk associated with LS, carriers often, but not always, fulfill clinical criteria indicative of hereditary cancer such as young age of onset or multiple cancer manifestations (Bethesda criteria) [33,34].There are profound differences in the pathogenesis of sporadic and hereditary MSI CRCs. Whereas sporadic MSI tumors predominantly arise through BRAF-mutated serrated lesions [35,36], LS-associated CRCs have been described to arise either through conventional adenomas [37,38,39] or through MMR-deficient crypt foci (MMR-DCF) [40,41,42,43]. MMR-DCF lack the expression of MMR proteins but, in contrast to adenomas, are morphologically undistinguishable from normal colonic crypts. MMR-DCF are specifically associated with LS and found in abundance in the intestinal mucosa of LS carriers (about 1 per 1 cm2) [41]. Recently, evidence for the true precancerous nature of MMR-DCF has emerged, including the molecular observation of MSI and mutations in cMS (e.g., HT001, AIM2, BAX, TGFBR2) in larger MMR-DCF [41,42]. The ongoing, FSP-associated immune surveillance in LS mutation carriers would potentially enable the elimination of most MMR-deficient cell clones.Based on the differences between LS and sporadic MSI tumorigenesis, it seems plausible that the distinct evolutionary forces shaping the tumors are also reflected in phenotypic differences. However, evidence supporting this hypothesis has so far been scarce and scattered across the scientific literature. To the best of our knowledge, no comprehensive literature analysis comparing immune characteristics of LS and sporadic MSI tumors exists yet. In the present systematic literature review, we aim to comprehensively describe the local immune phenotype of MSI tumors, with emphasis on differences between LS-associated and sporadic MSI tumors. The existence of such differences would provide new insights into the role of immune surveillance in LS-associated and sporadic MSI cancers and guide therapeutic approaches in these molecularly distinct tumor types.A systematic literature search was conducted to identify studies that compared the extent of immune infiltration and immune evasion in sporadic and hereditary MSI CRCs. The inclusion criteria were direct comparison of immune characteristics between sporadic and LS-associated MSI tumors, irrespective of tumor location and evaluation method. Exclusion criteria were comparison between MSS and MSI only, not considering hereditary or sporadic pathogenesis of MSI tumors, as well as lack of clear definition of the comparison groups. An online MEDLINE search (http://www.pubmed.com) was conducted between 29 February and 10 March, 2020 and the following keywords were used: {Lynch} OR {Lynch syndrome} OR {HNPCC} OR {MSI} OR {microsatellite instability} OR {MMR-deficient} OR {MMR-deficiency} OR {hereditary MSI} OR {sporadic MSI} OR {microsatellite instable} OR {RER +} OR {mutator phenotype} AND {immune infiltration} OR {immune microenvironment} OR {T cell infiltration} OR {TIL} OR {tumor infiltration} OR {immune evasion} OR {T cell density} OR {B2M} OR {RFX5} OR {CIITA} OR {HLA} OR {TAP} OR {NLRC5}.The screening of retrieved literature was performed manually. In addition, search of studies cited in the manuscripts that were analyzed on the full text level was performed. Although sample overlap between some of the analyzed studies cannot be formally excluded, sample collection from different sources or substantial temporal separation/analysis of different immune markers in the studies with mutual sources argues against major influence of the sample overlap on the results. A subset of the original search results was analyzed by two observers independently. The inter-rater reliability was measured by the Cohen’s kappa coefficient using Rstudio [44]. The strength of agreement on the title level was found to be substantial (κ = 0.7). The status of the few studies with discrepant evaluation was unanimously agreed on by both observers when abstracts were analyzed. Identified studies were checked for their eligibility by screening title, abstract and, lastly, the full text article. During the screening process, Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) recommendations were followed (Figure 2). The selected papers were analyzed qualitatively and quantitatively whenever possible. Quantitative analysis of the immune infiltration data was not conducted as different antibodies and counting methods (percentage of positive staining vs. cell counts) were used, and different cell categories (e.g., stromal vs. epithelial) were assessed. Additional studies and review articles addressing the immune phenotype of MSI CRCs were identified manually and independently of the used search terms from the authors’ own files.All statistical evaluations were performed in GraphPad Prism (Version 6.07, GraphPad Software Inc., La Jolla, CA USA) and Rstudio (Version 3.6.1, RStudio Inc., Boston, MA USA) [44]. The 95% confidence interval (CI) of the proportion of B2M mutations in the hereditary and sporadic MSI CRC groups was calculated with the modified Wald method in GraphPad Prism. Fisher’s exact test (α = 0.05) was used to check for a statistically significant association between B2M mutations and the origin of the MSI CRC; the analysis was also performed in GraphPad Prism. Odds Ratios (OR) and respective 95% CIs for the hereditary vs. the sporadic MSI CRC group were calculated and illustrated using the Rstudio packages epitools [45] and ggplot2 [46] Subsequently, statistical significance was checked using a Fisher’s exact test (α = 0.05).The literature search with the defined keywords covering immune infiltration and immune evasion (see Methods) revealed 522 articles in total. Following PRISMA recommendations, all articles (n = 522) were screened, and 364 publications were excluded based on the title. From the remaining 158 articles, 88 were excluded based on the abstract and 49 were excluded on the basis of the full text (Supplementary Table S1). 17 eligible articles fulfilled the criteria and were included in the final analysis. The selection process is visualized in Figure 2. The majority of identified studies focused on MSI CRC, whereas only a small proportion of studies analyzed MSI EC.We first compared the immune infiltration determined by the quantitative analysis of different immune markers (CD3, CD4, CD8, PD1, PD-L1, FoxP3, CD45) between hereditary and sporadic MSI cancers. Seven studies [47,48,49,50,51,52,53] were informative for quantitatively evaluating the extent of local immune infiltration in MSI CRCs (Table 1). We looked at the results from the immunohistochemical analysis of different T cell subsets of these studies, where a total of 122 hereditary MSI CRC and 145 sporadic MSI CRC samples were examined. A direct comparison between hereditary and sporadic MSI CRC groups was performed in each study. The remaining ten studies that did not provide exact T cell counts but worked with frequencies or defined categories are not included in Table 1 but will be discussed below.Studies using the pan-T cell marker CD3 for assessing intraepithelial T cell infiltration in sporadic and hereditary MSI CRCs uniformly observed higher T cell counts in the hereditary group [47,48]. Equivalently, Young et al. (2001) reported a higher frequency of T cell infiltration in the hereditary group (73%) compared to the sporadic MSI CRC group (56%) [54]. Further, a study by Jass et al. (2002), which was identified independently of the literature search, also mentions a higher frequency of Crohn’s like reactions and peritumoral T cells in hereditary compared to sporadic MSI CRCs [55]. However, Takemoto et al., who did not use exact numbers for intraepithelial T cells but instead distinguished between poor, moderate and severe intraepithelial T cell infiltration, found no significant difference between the two groups [49]. Similarly, Wang et al., using hematoxylin and eosin slides to quantify T cell infiltration, did not find a significant difference between sporadic and hereditary MSI CRCs [56].In addition to pan-T cell counts, some studies differentiated between T cell subpopulations. For instance, Takemoto et al. reported slightly higher numbers of CD8- and CD4-positive T cells in the stroma of hereditary compared to sporadic MSI tumors [49]. However, other studies observed similar CD8 and CD4 infiltration levels in both MSI CRC groups [50,51,57].In summary, the existing literature provides ample evidence that LS-associated MSI CRCs generally have a more pronounced local T cell infiltration compared to sporadic MSI CRCs (Figure 3). Although the evidence level is lower and further research is warranted, enhanced immune cell reactivity in LS-associated lesions seems to be already present at a premalignant stage.Our literature search identified a few studies analyzing immune infiltration not only in manifest cancers, but also in cancer precursors. Koornstra et al. (2009) reported significantly higher CD8-positive T cell counts in LS-associated compared to sporadic adenomas; however, the MSI status of these adenomas was not specified. Notably, T cell infiltration of LS-associated adenomas was significantly lower than the T cell infiltration of LS-associated CRCs [50]. Fittingly, another study [58] reported a higher frequency of T cell infiltration in LS-associated, MMR-deficient adenomas, compared to sporadic adenomas. However, as sporadic adenomas are mostly MMR-proficient, one should be careful when comparing LS-associated adenomas to this control group. An indication for activation of the immune system in premalignant lesions of LS patients has also been reported by Chang et al. who found increased RNA expression levels of CD4, IFNG, LAG3, PDL1, IL12A and TNF in LS-associated compared to Familial Adenomatous Polyposis (FAP)-associated adenomas suggesting immune activation in LS-associated adenomas. Interestingly, the authors reported that this immune profile was not related to the MMR status and mutational rates [59]. Further experiments are required to validate these findings and to provide visualization of immune cells infiltrating LS-associated adenomas.Alongside with the direct assessment of T cell infiltration, the immune status of a lesion can be elucidated from other indirect markers of immune response, such as expression of immune checkpoint molecules or lymphocyte recruitment markers. The expression of PD-L1, an interferon-gamma-inducible marker of prolonged immune activation, did not display significant differences between LS-associated and sporadic MSI CRCs [60,61]. However, Pfuderer et al. found much higher densities of high endothelial venules (HEV) in peritumoral tissue adjacent to hereditary MSI CRCs than to sporadic MSI CRCs [62]. HEVs are specialized postcapillary venules responsible for lymphocyte trafficking [62,63]. The evidence for enhanced lymphocyte recruitment to LS-associated MSI CRCs may point towards pre-existing immune responses in hereditary MSI patients induced by MMR-DCF [42]. Thus, long before tumor manifestation, MMR-DCF may prime the immune system of LS carriers against specific cMS mutation-induced FSP neoantigens. This is supported by the observation of FSP-specific immune responses in the peripheral blood of tumor-free LS carriers [64]. A primed immune system may facilitate strong immune infiltration of manifest cancers in the setting of LS but not, or only to a lesser degree, in sporadic MSI cancers.The phenomenon of enhanced immune responses associated with the hereditary origin of MSI cancer is not restricted to the colon. A study by Pakish et al. (2017) showed that a trend towards elevated immune cell infiltration can also be observed in LS-associated MSI ECs. The stromal compartments of LS-associated ECs displayed a significantly higher number of CD8-positive T cells, compared to the sporadic controls. In addition, higher CD8-positive intraepithelial T cell counts were observed in LS-associated ECs, but this increase did not reach statistical significance [52]. Similarly, Ramchander et al. reported a significantly elevated number of CD8-positive T cells in the invasive margin of LS-associated MSI ECs compared to sporadic cases. The same tendency was observed in the tumor center but failed to reach statistical significance [53]. Furthermore, another study reported a higher frequency of PD-L1 expression in LS-associated MSI ECs, compared to sporadic MSI ECs [65]. Ultimately, the picture of immune responses in LS-associated and sporadic MSI CRCs may go beyond tumor- and adenoma-related immune infiltration, as systemic immune responses against FSP neoantigens have been detected in tumor-free LS mutation carriers [64]. This finding may point at tumor-independent differences in the immune phenotype between the general population and LS carriers being present long before tumor manifestation and possibly influencing cancer risk in LS carriers. Such systemic immune responses might be linked to LS-associated MMR-deficient lesions, which can be detected not only in colorectal [40,41] but also in the endometrial tissue of LS carriers [66].Pronounced immune infiltration and strong local immune response can exert a substantial selective pressure on emerging precancerous cell clones [67]. This can trigger immune evasion through positive selection of tumor cell clones capable to escape destruction via the immune system. The main immune evasion mechanism in MSI CRCs is impairment of HLA class I-mediated antigen presentation, present in approximately 70% of all MSI CRCs [13]. In MSI CRCs, HLA class I-mediated antigen presentation is most frequently abrogated by mutations in the B2M (Beta-2 microglobulin) gene, which is affected in 30% of MSI CRCs. The B2M gene contains four microsatellites in its coding region, which make this gene susceptible for mutation in MMR-deficient cells [12,13,68,69]. Interestingly, the observed frequency of B2M-inactivating mutations is significantly higher than expected by chance based on the short length of the microsatellites [70], illustrating the tremendous survival advantage of B2M-mutant MSI clones under immune surveillance. Moreover, B2M mutations are closely associated with active local immune responses, represented by high numbers of activated T cells and a high density of HEVs [62,71].Five studies were identified that could be analyzed regarding immune evasion mechanisms affecting HLA class I- and HLA class II-related antigen presentation in hereditary and sporadic MSI CRCs (Table 2).Loss of HLA class I due to B2M mutations was examined in a total of 178 hereditary and 166 sporadic MSI CRCs. The proportion of B2M-mutated LS-associated MSI CRCs varied between 17% and 50%. For sporadic MSI CRCs this proportion ranged between 3% and 29%. The majority of included studies reported a higher frequency of B2M mutations in hereditary MSI CRCs compared to sporadic cases [12,47,73] (Figure 4). Only one study identified through our literature search [72] reported a higher rate of B2M mutations in sporadic compared to LS-associated CRCs. Although the reasons for this discrepancy are not clear, the study by Clendenning et al. confirmed previous reports that B2M mutations are associated with increased immune infiltration and elevated numbers of Crohn’s like lesions in the tumor vicinity, supporting the hypothesis of immunoediting as the cause of outgrowth of tumor cells with impaired B2M [72].In addition, two studies analyzed the frequency of B2M mutations in adenomas: Kloor et al. (2007) reported a 15.8% B2M mutation frequency in 38 MSI adenomas [12]. In contrast, Clendenning et al. did not observe any B2M mutations in 42 MMR-deficient adenomas [72]. Although the hereditary or sporadic origin of the analyzed adenomas was not clearly specified in these two studies, the current literature suggests high likelihood of MSI/MMR-deficient adenomas to be indicative of LS [74].The frequency of B2M mutations also seems to be influenced by the affected MMR gene, as a genomic and transcriptomic characterization of LS-associated CRCs revealed that MLH1-mutated tumors, compared to those having defects in the other MMR genes, may present with a higher rate of B2M mutations [75]. This observation potentially implies that MLH1-deficient tumors may be exposed to more pronounced immune responses than MSH2-deficient tumors. Similar findings have been reported independently by Clendenning et al. (2018) and Echterdiek et al. (2016) [71,72]. The mechanisms underlying this difference between MLH1- and MSH2-related immune phenotypes are not fully understood yet and require further research.Further studies are also necessary to evaluate potential differences related to the HLA class II antigen presentation machinery, which may have an important role in MSI cancer immunology [76]. Mutations of components involved in the HLA class II antigen processing and presentation pathway have so far only been addressed by one study [51], which analyzed 35 hereditary and 34 sporadic MSI CRCs for CIITA and RFX5 mutations. Although not reaching statistical significance, potential differences were observed in the patterns of mutations in HLA class II-regulatory genes, as CIITA mutations were more frequent in hereditary MSI CRCs, whereas RFX5 mutations occurred more frequently in sporadic MSI CRCs [51].Long before the era of ICB treatment, differences in therapy response have been observed between MSI and MSS CRC patients. 5-Fluorouracil (5-FU) treatment demonstrated limited therapeutic effects in MSI CRC patients and has even been reported to be associated with detrimental effects [77,78,79]. Therefore, adjuvant 5-FU-based chemotherapy is not recommended for MSI CRC patients of UICC (Union for International Cancer Control) stage II. However, MSI CRC patients seem to respond well to irinotecan-based therapies [80,81]. The mechanisms responsible for these differences are not fully clear. Besides potential immune modulation by chemotherapy, MMR deficiency may directly influence responsiveness on the cellular level. A functional MMR system may be required for cell death induction after 5-FU integration [80], whereas irinotecan-induced impairment of DNA damage repair may be synthetically lethal in MMR-deficient cells. Only few studies so far analyzed chemotherapy responsiveness differentially between hereditary and sporadic MSI CRCs. Sinicrope et al. (2011) reported a significant reduction of distant recurrences and improved disease-free survival in LS-associated MSI CRC patients compared to sporadic MSI CRC patients, both groups receiving 5-FU-based chemotherapy [82]. In addition, a study by Zaanan et al. reported a significantly improved progression-free survival for patients with LS-associated MSI CRCs under chemotherapy plus anti-EGFR, compared to patients with sporadic MSI CRCs [83].Recently, ICB therapy revolutionized the field of oncology by showing a tremendous success in metastasized (UICC stage IV) MSI tumors. Blockade of the PD-1 receptor, e.g., with Pembrolizumab, shows success in treating metastatic MSI cancers irrespective of the tissue of origin [16,84,85]. A phase II study conducted by Le et al. demonstrated enhanced responsiveness of MMR-deficient CRCs to Pembrolizumab treatment, in comparison to MSS CRCs [16]. However, little is known about potential differences between sporadic and hereditary MSI CRCs under ICB treatment. Differences between the immune phenotype of sporadic and hereditary MSI cancers would be of immediate clinical relevance for ICB therapy, because sporadic MSI cancers can be differentiated from hereditary MSI cancers by molecular tumor analysis. If one of the two groups showed preferential response to ICB, information about hereditary and sporadic origin may guide patient selection for treatment.Based on the data reported above, two major implications of pathogenesis on ICB responsiveness are conceivable.On the one hand, LS-associated MSI CRCs, according to the performed literature search, show an elevated density of local immune cells compared to the sporadic group. Local tumor infiltration with cytotoxic T cells has been reported as a prerequisite for ICB response [86,87]. Therefore, LS patients with MSI cancers may have a higher chance of having sufficient numbers of T cells present at the tumor site, which can be reactivated by ICB to attack tumor cells.On the other hand, most of the studies reported above found a higher rate of impaired HLA class I-mediated antigen presentation, mainly caused by B2M mutations, in hereditary MSI CRCs. Therefore, the opposite constellation is reasonable to assume; LS-associated MSI CRCs may more frequently show upfront resistance towards ICB, which however might be masked by the generally favorable prognosis of patients with B2M-mutant tumors [12,69]. Although the reason for the latter observation is not yet fully clarified, different theories explaining this counter-intuitive association exist, including enhanced NK cell-mediated tumor killing or lower metastatic capacity due decreased platelet binding [2,88,89,90].Le et al. found no significant difference in the objective response rate between the hereditary and sporadic patient group receiving PD-1 blockade therapy, possibly because positive and negative factors offset each other [16,17]. The scarce data on differential responsiveness of MSI cancer patients based on hereditary or sporadic pathogenesis clearly indicates that future studies with systematic classification of patients according to the origin of the tumor will be necessary. Moreover, information about immune infiltration and immune evasion phenomena needs to be recorded for tumors treated by ICB.In addition to the hypothesis of immune stimulation via FSP neoantigens prior to tumor development in LS setting, other explanation for the observed immunological differences should be considered, such as host factors that shape the immune system and respective immune responses. Recently, the importance of the host’s immune status in the LS setting has been highlighted by a study showing that an immune dysfunction due to immunosuppression or autoimmune conditions is associated with a higher rate of multiple cancers [91]. This association underlines the major role of immune surveillance in determining LS cancer risk and indicates the necessity accounting for the immune status of LS carriers when performing cancer risk estimation studies.LS-associated MSI cancers have a younger average age of onset than sporadic MSI cancers [27,92]. As the activity of the immune system and probably its capacity to recognize tumor cells are related to age [93,94,95], one may speculate that the observed elevated immune responses in LS mutation carriers compared to sporadic MSI tumor patients may merely reflect the generally younger age of the former. However, the local immune activation reflected in the HEV density and detected specifically in LS-associated tumors remained significantly higher than in sporadic MSI tumors even after correction for age [62], indicating that age alone cannot explain the enhanced local immune responses against LS-associated compared to sporadic MSI CRCs.Additionally, gender should be taken into account when discussing the differences between hereditary and sporadic MSI CRCs as sporadic MSI CRCs predominantly occur in women. This is also reflected by the gender-specific differences in tumor location, molecular characteristics and mortality of CRCs [96]. Right-sided CRCs occur more frequently in women than in men [97]. In addition, the MSI tumor phenotype is also more common in women and a link to estrogen exposure could be made. Especially older, menopausal women with lowered estrogen levels are at risk to develop sporadic MSI CRCs, and it has been shown that hormone replacement therapy reduces the risk for MSI CRCs [98,99]. Females also present with a higher incidence of BRAF mutations associated with sporadic MSI cancers as a consequence of age-related enhanced levels of DNA methylation and silencing of tumor-suppressor genes responsible for oncogene-induced senescence [96,100].Besides being involved in tumorigenesis, BRAF mutations are also associated with immunosuppression [101,102,103,104]. As the BRAF V600E mutation very rarely occurs in LS-associated CRCs (1.4%) but is displayed by around 64% of sporadic MSI CRCs with MLH1 methylation [105], the proposed immunosuppressive features of this mutation have to be considered when comparing local immune responses in the two MSI CRC groups. However, the majority of studies reporting immunosuppressive effects of BRAF mutations were done in melanoma and some additional evidence points to the absence of BRAF-mediated immunosuppression in the context of MSI CRCs: First, TIL density is not significantly reduced in BRAF-mutated MSI CRCs, compared to BRAF-wild-type MSI CRCs [106]. Second, BRAF-mutated MSI CRCs responded to ICB treatment with Nivolumab plus Ipilimumab as well as BRAF-wild-type MSI CRCs [84]. Third, BRAF mutation in the MSI setting is not associated with poorer prognosis and reduced overall survival, in contrast to MSS CRCs [107].It is worthwhile to note that studies analyzed in this systematic literature review were heterogenous with regard to the definition of LS: whereas some studies only included proven pathogenic MMR variant carriers, others relied on LS definitions based on immunohistochemistry or fragment length analysis suggestive of the MMR-deficient/MSI phenotype, complemented by MLH1 methylation/BRAF mutation analysis and clinical criteria. Omitting such studies would have substantially reduced the comprehensiveness of our survey. As the percentage of tumors wrongly classified using the “suspected LS” classifier is expected to be low based on existing literature data, we decided to include both types of studies. Future studies addressing the immunological differences between sporadic and LS-associated MSI tumors are encouraged to define LS based on pathogenic MMR variants in the germline.Our systematic literature analysis has two major findings: First, MSI cancers developing in the context of LS show more active local immune responses than sporadic MSI cancers. Second, and possibly related to the first finding, immune evasion phenomena in MSI cancers seem to be associated predominantly with a LS background. We found evidence that recurrent encounters with premalignant lesions prime the immune system against MSI-associated tumor antigens in LS individuals (Figure 5). Possibly, elevated immune surveillance is responsible for the incomplete penetrance of LS. However, to verify these hypotheses and gain more confidence in the general character of the findings currently based on CRC and EC, studies analyzing immunological features in other MSI tumors of the LS spectrum are warranted.Our study demonstrates that classification of a tumor as MSI is insufficient to inform about its immunological phenotype and to predict response to immune therapy. With the advent of ICB as a powerful modality to treat advanced-stage MSI cancer patients, however lacking predictors of therapy success among MSI cancer patients, systematic characterization of (1) hereditary or sporadic origin, (2) immune infiltration and (3) immune evasion phenomena has become paramount in order to better delineate the determinants of therapy response. In particular, literature data suggesting differences with regard to immune evasion phenotypes between hereditary and sporadic MSI CRCs may point towards differential responsiveness to ICB. Although potentially being associated with additional costs and logistical hurdles, it is strongly encouraged to account for germline variant status and tumor immune status in future clinical studies addressing ICB in MSI cancer patients.The findings of our study also have implications on tumor prevention. If immune surveillance governs MSI cancer evolution, modulation of immune surveillance may well hold the key towards effective prevention of MSI tumors, possibly by vaccination with shared MSI-induced FSP neoantigens.The following are available online at https://www.mdpi.com/2077-0383/9/6/1741/s1. Table S1: List of papers that were excluded based on the full text during the systematic literature search. Reasons for exclusion are given.Conceptualization and methodology, L.B., M.v.K.D., M.K. and A.A.; formal analysis, L.B.; investigation and validation, M.K. and A.A.; resources and software, L.B.; data curation, L.B.; writing—original draft preparation and review and editing, L.B., M.v.K.D., M.K. and A.A.; visualization, L.B.; supervision, project administration and funding acquisition, M.v.K.D., M.K. and A.A. All authors have read and agreed to the published version of the manuscript.This research study was supported by Else Kröner-Fresenius Foundation, 2018_A44.The authors declare no conflict of interest. The funders had no role in the design of the study; in the collection, analyses or interpretation of data; in the writing of the manuscript or in the decision to publish the results.(a) Proportions of colorectal cancers (CRCs) and endometrial cancers (ECs) that present the microsatellite instability (MSI) phenotype and the respective percentages of MSI tumors with a hereditary background. (b) The generation of frameshift peptides (FSPs) starts with a polymerase slippage event in a microsatellite sequence that results in an insertion/deletion mutation. Due to the impaired mismatch repair (MMR) system, this error is not corrected and leads to a shift of the reading frame. After translation, the FSP is processed by the human leukocyte antigen (HLA) processing and presenting machinery (depicted: HLA class I) and FSP-derived neoantigens are presented on the surface of the MSI cell. Interaction of the HLA-bound antigen can trigger FSP-specific immune responses. The shown example is based on a mutation in the A10 microsatellite of the TGFBR2 gene.Flow diagram illustrating the systematic literature search according to Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) recommendations. The selection process identified 17 studies that were included in the analysis.Graphical summary of the immune infiltration data in hereditary and sporadic MSI tumors. Studies that were assessed quantitatively were divided into three groups: studies reporting significantly higher immune infiltration counts in LS-associated MSI tumors compared to sporadic MSI tumors (“significantly higher in LS”), studies finding no significant differences between hereditary and sporadic MSI tumors (“no significant difference”) and studies reporting significantly higher immune infiltration counts in sporadic MSI tumors (“significantly higher in sporadic MSI”). Study size is proportional to the box area, and color indicates the respective marker. † study only counted immune cells in stromal regions. ‡ data for colorectal adenomas. * studies that analyzed MSI ECs.Forest plot of the calculated Odds Ratios for the B2M mutation frequency for hereditary vs. sporadic MSI CRCs with a 95% CI. Datapoints are proportional to study size. The dashed line indicates the total Odds Ratio.Summary figure showing the relationship between different colonic precursor lesions and the immune phenotype of the manifest tumor.Summary of papers quantitatively evaluated for T cell immune infiltration in hereditary and sporadic MSI CRCs, colorectal adenomas and ECs. The number of patient samples, LS status, applied T cell markers and quantification results are displayed; the respective counting methods are indicated. The T cell counts are given separately for tumor epithelium and stroma. Asterisk represents statistically significant differences between the hereditary and sporadic group. hpf = high-power-field.Summary of publications analyzing immune evasion in LS-associated and sporadic MSI CRCs and adenomas, distinguishing between mutations of the HLA class I and HLA class II pathways. Numbers of used samples are indicated for all groups and percentages of mutated samples are shown. Asterisk represents statistically significant differences between the hereditary and sporadic group.

Med-MDPI/jcm_2/jcm-09-06-01742.txt

@@ -0,0 +1 @@
+Adipokines are powerful agents involved in the development of obesity-dependent cancers. This prospective study aimed to investigate the association between pre-treatment body mass index (BMI) and serum YKL-40, leptin, and adiponectin concentrations as well as the plasma activity of tissue factor (TF) and the future prognosis of early, non-metastatic breast cancer (BrC) subjects. The serum levels of YKL-40, leptin, and adiponectin as well as plasma TF activity, anthropometric parameters, and clinicopathological parameters were analysed in 81 treatment-naïve females with invasive BrC. The predictive value of YKL-40, BMI, leptin, adiponectin, and TF was determined with a 95% confidence interval (CI). Kaplan–Meier plots and log-rank and F Cox tests were used to determine the clinical outcomes of progression-free survival (PFS). The median follow-up duration was 44 months with complete follow-up for the first event. Follow-up revealed a significantly higher incidence of disease relapse in BrC patients with a high baseline concentration of YKL-40 (22.22%) and TF activity (21.43%). Body mass index was an independent predictor of survival, with women who were overweight/obese being less prone to relapse (hazard ratio (HR): 0.75; 95% CI: 0.59 to 0.95). The recurrence rates for normal-weight BrC cases was 21.05% versus 7.14% for their overweight counterparts. The receiver operating characteristic analysis showed the strong ability of the analysed biomarkers to predict disease progression, with an area under the receiver operating characteristics (ROC) curve of 0.84 (95% CI, 0.823 to 0.931). In a prospective cohort of invasive BrC patients, overweight/obesity was associated with improved future outcomes. The combination of a normal BMI with high leptin and low adiponectin levels and high TF activity was associated with an increased risk of recurrence and decreased survival.It is well-established that breast cancer (BrC) and obesity are the most widespread diseases in the female population, all over the world. There is no doubt that obesity is a well-identified risk factor for cervical, ovarian, endometrial, and breast cancer, and it is associated with 88% mortality rates in females [1,2]. Interestingly, one in two breast cancer women are overweight or obese [2]. Obesity leads to increased susceptibility for postmenopausal luminal breast cancers, but premenopausal obese women are more prone to developing a basal-like breast cancer (BLBC). Furthermore, adiposity is associated with a more aggressive breast cancer phenotype, including larger tumours, more often with nodal metastases and worse prognosis [3,4,5,6]. Thus, adiposity is linked with both cancer predisposition and cancer-dependent death [7].Against this background, prognostic indicators in patients with breast cancer are fundamental. Since BrC is a very heterogeneous disease with well-defined morphologies, molecular attributes, prognoses, and treatment options, clinical decisions are mainly made on the basis of the tumour stage, lymph-node status, and the expression of molecular determinants, represented clinically by oestrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor receptor 2 (HER2), and the proliferation marker Ki67 [8]. There is a five-year survival rate of 100%, 93%, 72%, and 26% for stages I, II, III, and IV, respectively [9]. Thus, the early diagnosis of breast cancer has great clinical prominence.Nowadays, the underlying biological mechanisms that link obesity to BrC have largely focused on the excessive exposure of the mammary glands to the abundant secretion of oestrogen, cytokines, pro-inflammatory mediators, and growth factors, along with an altered adipokine profile, all of which are believed to be involved in the promotion, invasion, and metastasis of cancer cells. Normal adipose tissue and mature adipocytes, especially, produce the insulin-sensitising, anti-inflammatory, anti-atherogenic, and anti-proliferative hormone, adiponectin, and a low amount of the pro-angiogenic, pro-mitogenic leptin as well as a low quantity of pro-inflammatory cytokines [3,10]. However, adiposity leads the way in a pivotal shift towards the up-regulation of leptin in parallel to the down-regulation of adiponectin and overproduction of pro-inflammatory M1 macrophages [2,3,11,12].The additional threat of neoplasm is a pro-coagulant state, in which the enhancement of coagulation is linked to malignant transformation via tumour cell growth, migration, angiogenesis, and extravasation. Cancer-associated overexpression of tissue factor (TF) and thrombin has been noted. Interestingly, TF can also be released under the influence of tumour necrosis factor alpha (TNF-α) and interleukin-1 beta (IL-1β) in soluble form. Thus, these pro-inflammatory agents are incident to the hypercoagulability state. On the other hand, tumour cells produce TNF-α and IL-1β, which activate pro-angiogenic growth factors to cause enhanced tumour growth [7,13]. Summarising, the malignant process, inflammation and the pro-coagulant state are inseparably connected with each other to create the “self-powered machine”.Chronic low-grade inflammation is a common attribute of adiposity and cardiovascular complications [7,13]. The contributions of the local immune response and systemic inflammation in cancer initiation and invasion have been confirmed [14]. A promising biomarker for predicting the risk of BrC recurrence is an inflammatory protein named YKL-40, also called human cartilage glycoprotein-39 (HC gp-39) and chitinase 3-like 1 protein (CLP). YKL-40 is secreted by leukocytes, tumour cells, and tumour-related macrophages. Il-6 and hypoxia induce YKL-40 synthesis [14,15,16,17,18]. The serum and plasma levels of YKL-40 are associated with poor clinical outcomes in breast, lung, prostate, liver, bladder, colon, and other types of tumours [14,15,16,17,18]. YKL-40 prevents apoptosis and promotes malignant cell proliferation by the up-regulation of anti-apoptotic or pro-proliferative, pro-angiogenic factors. In severe infections, increases in plasma YKL-40 are noted before rises in the plasma C-reactive protein [18]. Interestingly, Kjaergaard et al. suggest that plasma YKL-40 is not a causal agent of neoplastic changes but rather an inflammatory biomarker [14].It is crucial to recognise the association between adiposity and the activation of TF-dependent coagulation and inflammatory reactions and to elucidate their importance in malignant transformation. Thus, the purpose of the present study was to assess the association between pre-treatment body mass index (BMI), serum YKL-40, leptin, adiponectin concentrations, the plasma activity of TF, and the future prognosis for early, non-metastatic BrC subjects. Written informed consent was received from all patients in accordance with the Declaration of Helsinki. Subjects could withdraw from the study before or at any time during the study. The present research protocol was approved by the local bioethical committee (permission no. KB 547/2015; 16.06.2015).Eighty-one consecutive female subjects with newly diagnosed, histologically-confirmed invasive BrC were enrolled in this study. The median age at inclusion was 54 years (interquartile range, 49–59 years). The participants were enrolled in the study before systemic adjuvant treatment. Since the levels of adipokines and pro-inflammatory markers may vary with advancement of disease, only patients with Stage I and II disease (according to the AJCC classification of 2016) were included in the present series (Stage I, n = 40; Stage II, n = 41). The patients were under the care of the medical staff from the Clinical Ward of Breast Cancer and Reconstructive Surgery, Oncology Center Prof. F. Łukaszczyk Memorial Hospital, Bydgoszcz, Poland.The inclusion criteria of the study were as follows: (1) the diagnosis of primary, unilateral, invasive BrC; (2) complete clinical record and follow-up data; (3) all the peripheral blood samples were obtained within 24 h before surgery; and (4) proper haematological, liver, and renal function. The exclusion criteria for the patients included: (1) bilateral BrC; (2) a tumour larger than 5 cm; (3) Stage IIIA or higher; (4) neoadjuvant chemotherapy, radiotherapy, or endocrine therapy; (5) a previous diagnosis of any cancer type; (6) distant metastasis; (7) chronic inflammatory diseases or autoimmune disease; (8) a carcinoma in situ; (9) morbid obesity (BMI over 40 kg/m2); (10) diabetes mellitus type 2; (11) psychiatric illness; and (12) male gender.All of the study participants underwent a personal interview administered by oncologists in the hospital. Data were collected on the sociodemographic characteristics (such as age and education), menstrual and reproductive history, menopausal status, menopausal hormone therapy, lifestyle behaviours, and medical history as well as the history of breast and other cancers using a standardised lifestyle questionnaire. Postmenopausal status was identified in the current study as the absence of a menstrual cycle for 12 months after the last menstruation. Body mass index (BMI) (weight in kg/height in m2) was calculated from the patient’s height and weight, measured at the initial patient visit, when the individuals were wearing no shoes and few clothes. Tumour histology and size; lymph-node status; tumour staging; and immunohistochemistry (IHC) of the oestrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor receptor 2 (HER-2), Ki67- proliferation marker, and E-cadherin were determined. The Nottingham histological grade of malignancy (the Elston-Ellis modification of the Scarff-Bloom-Richardson grading system) was established, based on three components: (1) the amount of tubule formation, (2) the nuclear grade, and (3) the mitotic rate. The tumour size was defined by the maximum diameter of the sample. The tumor (T)/node (N)/metastasis (M) stage of the disease at the initial diagnosis was confirmed by the American Joint Committee on Cancer (AJCC), 7th Edition. The histological type was classified as ductal carcinoma or lobular carcinoma according to the WHO Classification of Tumours. The intrinsic subtypes of BrC were divided into luminal A, luminal B HER2-positive, luminal B HER2-negative, non-luminal HER2-positive, and basal-like.All the included patients had post-operative inpatient or outpatient follow-up every three months for the first and second years, and every six months for the third and fourth years after surgery. For the progression-free survival analysis, 11 events were recorded, including 3 (3.7%) loco-regional recurrences, 3 (3.7%) distant metastases, and 5 (6.17%) deaths. The median follow-up time after the index date was 44 months with a 13.58% recurrence rate. Follow-up evaluation included laboratory tests (blood biochemical test), physical examination (breast and lymph-node palpation), breast ultrasonography, liver ultrasound, mammography, and other suitable examinations. Relapse was defined as signs of metastatic disease or local recurrence confirmed by PET/CT or death (excluding deaths unrelated to the disease).Venous blood (4.5 mL) for testing TF activity was collected into cooled tubes (Becton Dickinson Vacutainer® System, Plymouth, UK) containing 0.13 mol/L of trisodium citrate (final blood anticoagulant ratio, 9:1). Additionally, the blood was collected into 4.0 mL tubes (Becton Dickinson Vacutainer® System, Plymouth, UK) without anticoagulant in order to determine the concentrations of adiponectin, leptin, and YKL-40. Blood samples from all patients were obtained within 24 h before surgery procedures, when patients had been in a fasting state; after 30 min of rest and after a 12 h overnight fast, blood sampling took place between 7.00 a.m. and 9.00 a.m., to minimise diurnal variability. The samples were processed according to the standard conditions for blood samples. All clinicopathological data were obtained as part of routine care. All the blood samples were stored at −80 °C until analysis. The blood samples were centrifuged at 3000× g for 15 min. There was no variation in the average storage time between the case samples. For all kits, the reaction mixture was added to in a 96-well plate. The laboratory assistants were blinded to the study population.The baseline serum leptin concentration was measured using a commercially available kit—Human Leptin Enzyme-Linked Immunosorbent Assay (ELISA) Clinical Range (BioVendor Research and Diagnostic products, Brno, Czech Republic; catalogue number: RD191001100)—in accordance with the manufacturer’s instructions. The limit of leptin detection was 0.2 ng/mL. The intra-assay coefficient of variation (within-run) was 5.9% with an inter-assay coefficient of variation (run-to-run) of 5.6%. The subjects were separated as having low or high values, dichotomised using a cut-off for leptin of 12.58 ng/mL, based on the median value for the whole study population.The pre-treatment serum adiponectin level was determined by a human adiponectin ELISA high sensitivity ELISA kit, (BioVendor Research and Diagnostic products, Brno, Czech Republic; catalogue number: RD191023100). The limit of adiponectin detection was 0.47 ng/mL and had an intra-assay coefficient of variation (within-run) within 3.9% and an inter-assay coefficient of variation (run-to-run) of 6.0%. Patients were divided as having low or high values, dichotomised using a cut-off for adiponectin of 27.05 ng/mL, based on the median for the whole study population.The baseline serum YKL-40 protein concentration was assessed with a Human Chitinase 3-like 1 ELISA kit (BioVendor Research and Diagnostic products, Brno, Czech Republic; catalogue number: RD193444200CS). The minimal detectable limit for YKL-40 was 10 ng/mL. The intra-assay coefficient of variation (CV) was 5%, and the inter-assay CV was <6%. Patients were grouped as having low or high values, dichotomised using a cut-off for YKL-40 of 1.93 ng/mL, based on the median value for the whole study population. Additionally, the YKL-40 concentration was categorised either as low (<1.60 ng/mL), moderate (1.60–2.50 ng/mL), or high (>2.50 ng/mL) based on the assumption to create three numerically equal subgroups; each group consisted of 27 cases.The plasma activity of TF was measured using chromogenic assays, the ACTICHROME®TF test (Sekisui Diagnostics, LLC, Stamford, CT, USA; catalogue number: 846). The minimal detectable limit for TF activity was 2 pM. The standard range was set at 0–30 pM. Patients were grouped as having low or high values, dichotomised using a cut-off for TF of 14.18 pM, based on the median value for the whole study population. Additionally, TF activity was categorised either as low (<12 pM), moderate (12–24 pM), or high (>24 pM) based on the assumption to create three equal subgroups; the first group consisted of 26 cases, the second was made up of 27 patients, and the third was composed of 28 subjects (there were 28 patients in the third group due to the fact that two patients had the same results).The measurements of hormone receptors including ER and PR as well as HER2 and Ki67 were performed by the immunohistochemistry method according to the manufacturer’s indications. To define ER and PR status, immunostaining using anti-ER (clone SP1) and anti-PR (clone 1E2) pre-diluted monoclonal antibodies (Ventana Medical Systems, Tucson, AZ, USA) was applied. For the demonstration of the Ki67 antigen in the specimens, monoclonal mouse antibody (Autostainer Link 48, Agilent Technologies, Santa Clara, CA, USA) was used. For the semi-quantitative detection of HER2, the rabbit monoclonal primary antibody VENTANA anti-HER2/neu (clone 4B5) was used.The immunostaining for ER, PR, and Ki67 was quantified and expressed as percentages of immunostained neoplastic cells. We used 20% as the threshold value to judge the level of Ki67 expression. For HER2, strong complete membrane staining in more than 10% of tumour cells was defined as positive expression (3+). Scores of 0 and 1 were established as negative expression, and all tumour cells with a 2+ score were further tested by fluorescence in situ hybridisation [19]. Based on the expression of molecular determinants, all subjects were classified according to the St. Gallen 2013-15 recommendations.The Shapiro–Wilk test was used to test normality, and the Mann–Whitney test and the ANOVA Kruskal–Wallis test were used to compare subgroups. The non-normally distributed variables were described by medians and interquartile ranges (IQR). The clinicopathologic categorical variables were presented as numbers and percentages (%). The Kaplan–Meier product limit estimator method was used to determine the survival time of PFS. The log-rank test and the F Cox test were used to check the dependence of patients’ survival on single or on combinations of variables. The multivariate Cox proportional hazards regression model was used to estimate the hazard ratios (HRs) and the 95% confidence intervals (CIs) for the associations between the serum YKL-40, leptin, and adiponectin levels as wells as the plasma TF activity and the progression-free survival. A 95% confidence interval (95% CI) was set as the criterion to establish statistical significance. Finally, to evaluate the biomarkers’ ability to predict disease recurrence, analysis of the receiver operating characteristics curve (ROC) by the area under the curve (AUC) was performed according to the logistic regression model. Differences with p-values of less than 0.05 were considered to be statistically significant.Between November 2015 and January 2018, from a total of 92 newly diagnosed and treatment-naïve BrC subjects, 81 patients fulfilled the inclusion/exclusion criteria and were prospectively included in the study. Table 1, Table 2 and Table 3 summarise the study group according to anthropometric, demographic, and clinicopathological features as well as biochemical parameters. All cases were females, and the median age was 54 years (interquartile range (IQR): 49–59 years). The median BMI was 25.22 kg/m2 (IQR: 23.05–28.90 kg/m2). More than half (53%) of the cases were overweight or obese. The median follow-up for the entire cohort was 44 (IQR: 37–48) months, with complete data for the cause and date of death. All 81 patients underwent surgery, 13 cases underwent mastectomy, and 68 cases underwent breast-conserving surgery. In terms of the histological pathology, 69 patients had invasive ductal carcinoma (IDC) and 12 patients had invasive lobular carcinoma (ILC). Fifty (62%) BrC subjects had a luminal-A intrinsic subtype. The most common histological grades according to Elston–Ellis classification were highly (G1) and moderately differentiated carcinomas G2 (80%). The median tumour diameter was 1.67 cm (IQR: 1.2–2.1 cm). More than the 75% of cases were free of nodal involvement.The age at diagnosis was categorised into two groups (<55/≥55 years) as well as menopausal status (premenopausal/postmenopausal). A lack of significant differences in the concentrations of YKL-40, leptin, and adiponectin as well as the TF activity with respect to age and menopausal status was noted. BMI (kg/m2) was classified as normal (18.5–24.9), overweight (25.0–29.9), and obesity class I and II (30.0–39.9) according to the WHO’s recommendations. Parity was defined as the number of full-term pregnancies. There was a trend towards a higher concentration of leptin and a lower concentration of adiponectin with increasing BMI in our study group. Interestingly, the concentrations of adiponectin and YKL-40 were dependent on parity status. A significantly higher concentration of YKL-40 was observed in BrC subjects who gave birth to one child than in cases who gave birth to two or three children. Patients who gave birth to one child or two children had a significantly higher concentration of adiponectin compared to subjects who gave birth to three children (Table 1).In Table 2, the statistical calculations for the study group are presented based on the proliferation marker (Ki67), human epidermal growth factor receptor 2 (HER2), and the status of hormone receptors with respect to the analysed parameters. The patients were divided into two groups on the basis of the proliferation marker, expressed by the Ki67 protein. The first group was made up of women with expression of the Ki67 antigen lower than 20%, and the second group included patients with Ki67 expression above 20%. A markedly higher concentration of adiponectin was observed in patients with expression of Ki67 lower than 20% (p < 0.01). Other molecular determinants did not differ significantly with regard to YKL-40, leptin, adiponectin, and TF activity.Furthermore, we hypothesised that the concentrations of YKL-40, leptin, and adiponectin and TF activity can vary according to the molecular subtype, tumour diameter, nodal status, staging, grading indexing, and histological subtypes of BrC. A significantly higher concentration of adiponectin in the luminal A-type of BrC compared to patients having a basal-like type was noted (p < 0.01). Additionally, a significantly higher TF activity was observed in the luminal B HER2-negative type of BrC with respect to patients having other molecular subtypes, including luminal A, luminal B HER(+), and HER(+)/non-luminal, basal-like BrC (p < 0.01). Interestingly, we observed significantly higher concentrations of YKL-40 and leptin (both p < 0.01), but a lower concentration of adiponectin (p < 0.001) in the group of patients with invasive ductal carcinoma with respect to their invasive lobular carcinoma counterparts. Furthermore, circulating YKL-40, leptin, and adiponectin concentrations and TF activity have not been associated with other prognostic indicators, such as tumour grade and TNM stage or tumour size and nodal status (Table 3).The multivariate Cox hazards analysis, which takes into account the function of time, revealed that the activity of tissue factor (HR, 1.07; 95% confidence interval (CI), 1.01–1.14; p = 0.0240), leptin concentration (HR, 1.21; 95% CI, 1.02–1.44; p = 0.0327), BMI (HR, 0.75; 95% CI, 0.59–0.95; p = 0.0178), and adiponectin level (HR, 0.86; 95% CI, 0.76–0.97; p = 0.0135) were significant predictive factors for disease relapse in BrC subjects. The multivariate Cox regression showed that BrC subjects with lower BMI values and adiponectin levels but with higher TF activity and leptin concentrations have a significantly higher risk of recurrent events (Table 4).Additionally, we performed receiver operating characteristic (ROC) curve analysis according to logistic regression to describe the ability of the five biomarkers including BMI and TF activity and the concentrations of YKL-40, leptin, and adiponectin to predict disease relapse (Figure 1). The estimated area under the ROC curve for the tested model was AUC = 0.84, with 95% CI = 0.823 to 0.931, R2 = 0.7515, and p value = 0.0075. This model points to strong diagnostic potential for the prediction of disease progression.We analysed the association between the pre-treatment concentrations of YKL-40, leptin, and adiponectin and TF activity as well as body mass index and disease relapse based on the 44 month follow-up of those patients (Figure 2). The data show that, during the follow-up period of more than four years, 11 of the 81 BrC patients (13.58%) had a relapse of disease. Five (6.17%) patients died during the follow-up period due to systemic metastatic disease (bones, liver, and spine metastases). Figure 2A demonstrates the division of the study group according to YKL-40 concentration. In the low YKL-40 subgroup, recurrent events occurred in two subjects out of the total of 27 followed-up subjects (7.41%). In the moderate YKL-40 subgroup, recurrent events happened in three patients out of the total of 27 followed-up subjects (11.11%). Finally, in the high YKL-40 subgroup, recurrent events occurred in six cases out of the total of 27 followed-up subjects (22.22%). Thus, patients with YKL-40 higher than 2.5 ng/mL demonstrated a significantly higher risk of disease recurrence compared to patients with a YKL-40 concentration lower than 1.6 ng/mL (p = 0.0464). Patients with TF activity lower than 12 pM (n = 26) had a significantly lower risk of disease progression (3.85%) with respect to those patients with TF activity higher than 24 pM (21.43%; n = 28); p = 0.0142 (Figure 2B). Furthermore, the study group was divided into three subgroups according to BMI values. There was a significantly higher incidence of disease relapse in BrC patients with a BMI (n = 38) lower than 24.9 kg/m2 compared to those with a BMI of 25–29.9 kg/m2 (n = 28; p = 0.0444). The recurrence rate for normal-weight BrC cases was 21.05% versus 7.14% for overweight cases. Thus, a BMI lower than 25 kg/m2 was associated with a three-fold higher risk of disease relapse (Figure 2C). In those eight normal-weight recurrent cases (the mean value of BMI was 22.4 kg/m2), three of them had luminal A, three had luminal B HER-, and two had a basal-like intrinsic subtype of BrC. Four patients were premenopausal, and four of them were postmenopausal. All of the patients were symptomatically free of other co-morbidities. All of them had moderately differentiated carcinomas (G2). Two out of eight had a positive lymph-node status. Four out of eight had overexpression of the proliferation marker Ki67. Five out of eight had a tumour larger than 2 cm. Five out of eight underwent breast-conserving surgery due to a lack of lymph node metastases. Additionally, five out of eight received radiotherapy, chemotherapy, or hormone therapy. According to additional analysis in those eight normal-weight recurrent women, the highest level of YKL-40 and TF activity but the lowest adiponectin level were observed (Table 5). Due to the limited space in the manuscript, we do not show separate plots for leptin and adiponectin, since neither demonstrated significant associations with disease recurrence.A combination of YKL-40 concentration with TF activity in BrC cases revealed an interesting observation. A worse future prognosis was shown in subjects with a combination of a YKL-40 concentration higher than 1.93 mg/mL and a TF activity higher than 14.18 pM (Figure 3A). Figure 3C and Figure 4A demonstrate a significantly better BrC specific survival for overweight/obese subjects with higher YKL-40 (>1.93 ng/mL) and higher leptin (>12.58 ng/mL) levels with respect to normal-weight cases with YKL-40 higher than 1.93 ng/mL and leptin higher than 12.58 ng/mL (p = 0.0065 and p = 0.0459, respectively). The recurrence of the disease in the group of overweight/obese patients with higher YKL-40 (>1.93 ng/mL) occurred in one out of 22 (4.55%) cases, but in the subgroup with normal weight and higher YKL-40 levels (>1.93 ng/mL), six out of 19 (31.58%) cases had a recurrence of the disease. The recurrence of the disease in the group of overweight/obese patients with a leptin concentration higher than 12.58 ng/mL occurred in three out of 33 (9.09%), but in the subgroup with normal weight and leptin levels higher than 12.58 ng/mL, three out of eight (37.5%) cases had a recurrence of the disease. Furthermore, follow-up revealed a significantly higher incidence of disease relapse in breast cancer patients with a BMI lower than 25 kg/m2 and an adiponectin level lower than 27.05 ng/mL with respect to those with a BMI higher than 25 kg/m2 and an adiponectin level lower than 27.05 ng/mL (p = 0.0069) (Figure 4B). The recurrence of the disease in the group of patients with a BMI lower than 25 kg/m2 and an adiponectin level lower than 27.05 ng/mL occurred in five out of 16 (31.25%), but in the group with a BMI higher than 25 kg/m2 and an adiponectin level lower than 27.05 ng/mL, only one out of 24 (4.17%) cases had a recurrence of the disease.According to our invasive breast cancer cohort, we demonstrated that parity status may predict disease recurrence (p = 0.0317). Interestingly, nulliparous women (33.3%) as well as women who gave birth to five children (100%) had the worst future prognosis (Figure 5). Both nulliparous women were postmenopausal subjects, free of other co-morbidities, with BMI = 24 kg/m2 and 22.41 kg/m2, and age = 55 and 60 years, respectively. The first patient had the following clinicopathological profile: IIA, T2, N0, M0, invasive ductal carcinoma, histological grade = 2, tumour diameter = 2.2 cm, ER+/PR- and HER2-, Ki-67 = 20%, and luminal B HER2 (-). The baseline values were YKL-40 = 1.99 ng/mL, leptin = 34.1 ng/mL, adiponectin = 20.51 ng/mL, and TF activity = 14.10 pM. The patient, after 28 months of follow-up, developed right ovary metastasis. The second patient presented the following clinicopathological profile: IIB, T2, N1, M0, invasive ductal carcinoma, histological grade = 2, tumour diameter = 2.8 cm, ER+/PR+ and HER2-, Ki-67 = 80%, and luminal B HER2 (-). The pre-treatment values were YKL-40 = 2.83 ng/mL, leptin = 14.41 ng/mL, adiponectin = 34.83 ng/mL, and TF activity = 39.16 pM. After 13 months of follow-up, multi-organ metastases were confirmed by PET. Both multiparous women (five children) were premenopausal subjects, free of other co-morbidities with BMI = 21.97 kg/m2 and 24.24 kg/m2, and age = 49 and 53 years, respectively. The first patient demonstrated the following clinicopathological profile: IIA, T2, N0, M0, invasive ductal carcinoma, histological grade = 2, tumour diameter = 2.5 cm, ER+/PR+ and HER2-, Ki-67 = 40%, and luminal B HER2 (-). The baseline values were YKL-40 = 1.36 ng/mL, leptin = 4.1 ng/mL, adiponectin = 14.58 ng/mL, and TF activity = 23.59 pM. The patient, after 34 months of follow-up, developed bone, mediastinal, and supraclavicular lymph-node metastases confirmed by PET. The second patient showed the following clinicopathological profile: IIB, T2, N1, M0, invasive ductal carcinoma, histological grade = 2, tumour diameter = 3.0 cm, ER+/PR+ and HER2-, Ki-67 = 15%, and luminal A. The pre-treatment values were YKL-40 = 2.55 ng/mL, leptin = 16.92 ng/mL, adiponectin = 28.49 ng/mL, and TF activity = 8.05 pM. However, this patient passed away due to lung, liver, and bone metastases within 18 months of diagnosis.BrC is a complex neoplastic disease, comprising the processes of unsuppressed proliferation, the elusion of growth suppressors, abnormalities in cell death, the promotion of a pro-angiogenic phenotype, the initiation of invasion and metastasis, the disruption of cellular homeostasis, and the avoidance of immune destruction with the additional significant possibility of relapse [11,20]. Thus, the complex heterogeneity of breast cancer at the molecular level forces a search for prognostic biomarkers, which can be simply estimated or implemented.Adipokines and coagulation factors as well as pro-inflammatory cytokines, have been proposed as a link between obesity and oncogenesis and are related to future prognosis. Our analyses showed a significantly higher concentration of leptin as well as a lower concentration of adiponectin with increasing BMI in breast cancer subjects. This configuration of circulating adipokines is widely accepted in the literature, since body composition changes with age, especially the amount of body fat, which leads to the attenuation of energy homeostasis and variation in adipose tissue metabolism [2,3,11,21,22]. Leptin and adiponectin in cancer biology work in an inverse manner, since leptin plays a crucial role in the promotion of tumour angiogenesis, cell invasion, and metastasis, whereas adiponectin possesses anti-tumour properties, including anti-proliferative, anti-migratory, and pro-apoptotic actions [2,3,21]. Our findings are in line with previous studies, which demonstrated that leptin positively correlates with breast cancer risk, while a low adiponectin concentration is linked to a higher risk of breast cancer [23]. The overexpression of leptin and down-regulation of adiponectin are most commonly associated with progressing or poor prognosis tumours [21,24].Interestingly, in our study, we observed that the concentration of YKL-40 and adiponectin levels vary with respect to parity. A significant higher concentration of YKL-40 was observed in breast cancer subjects who gave birth to one child than in those cases who gave birth to two or three children. Patients who gave birth to one child or two children had a significantly higher concentration of adiponectin compared to subjects who gave birth to three children. Due to the fact that adiponectin is recognised as an anti-cancer protein, its higher concentration seems to have a protective effect in those patients [3,10]. However, a higher concentration of YKL-40 in those cases can attenuate this positive effect, since YKL-40 is a pro-cancerogenic, pro-angiogenic, pro-inflammatory protein [14,15,16,17,18]. Furthermore, nulliparous as well as multiparous (five children) BrC subjects had the worst future prognosis. According to our results, we speculate that nulliparous and multiparous BrC patients developed a more aggressive phenotype of breast cancer, expressed by a larger tumour size, a higher stage, and a higher expression of the mitotic index. Interestingly, all of those patients had a BMI in the normal range and three out of four had a concentration of YKL-40 higher than 1.93 ng/mL. Three of those four patients demonstrated the luminal B HER2-negative type of BrC.The present study also shows that the adiponectin level is higher in breast cancer cases with a lower mitotic index, expressed by Ki67, compared to cancers with overexpression of Ki67. Ki67 is a proliferation marker in clinical practice as a marker of tumour aggressiveness and a predictor of future outcomes. Additionally, a higher adiponectin level was observed in luminal A breast cancer cases than in their basal-like counterparts. The luminal A breast cancer type is defined as hormone receptor-positive and HER2-negative with an expression of Ki67 lower than 20%. Thus, our study confirmed previous findings that the adiponectin concentration depends on the oestrogen receptor status [25,26]. Those observations indicate an inverse association between the adiponectin concentration and an aggressive phenotype, and the adiponectin level depends more on the molecular status than clinical features, including tumour diameter or lymph-node status. It is well-known that luminal A breast cancer presents a better prognosis than other intrinsic subtypes. Adiponectin diminishes proliferation, stimulates apoptosis, and negatively influences breast carcinogenesis and the ability for invasion and metastasis [27]. This result may also confirm that luminal A breast cancer with low expression of Ki67 presents the possibility of a more favourable outcome for patients in light of the anti-neoplastic properties of adiponectin. However, the findings of Ando et al. and Panno et al. demonstrated that adiponectin in ER-negative BrC cells diminishes cell growth and suppresses proliferation, invasion, and cancer cell motility [25], but divergent effects of adiponectin were noted with respect to luminal-like BrC cells [26]. A scientific consensus with respect to adiponectin’s role in neoplastic diseases is needed; thus, we would recommend analysis of adiponectin levels in the context of hormone receptor status in subsequent studies. Furthermore, significantly higher TF activity was observed in the luminal B HER2-negative type of breast cancer with respect to patients having other molecular subtypes. The overexpression of TF is associated with an invasive character of tumours and may be particularly related to cancer-associated thrombosis. Moreover, TF stimulates an angiogenic switch, leading to new vessel formation and further cancer cell growth [28]. Thus, elevated TF activity is associated with decreased overall survival [28,29]. TF activity and adiponectin levels are sensitive to the tumour’s molecular status and may serve as single biomarkers for future prognosis.Interestingly, we observed significantly higher concentrations of YKL-40 and leptin but a lower concentration of adiponectin in the group of patients with invasive ductal carcinoma with respect to their invasive lobular carcinoma counterparts. This constellation of cytokines seems to be unfavourable for those patients due to the role of leptin in the production of anti-apoptotic agents, pro-inflammatory cytokines (TNF-α, IL-6), growth factors (VEGF), and the hypoxia-inducible factor-1a (HIF-1α), promoting the cancerous behaviour of cells [23]. Additionally, leptin has been shown to be a potential prognostic marker in breast cancer, along with tumour size and lymph-node status, and an independent predictor of a poor outcome [22]. Additionally, YKL-40 is a key pro-inflammatory cytokine associated with the pathology of obesity-linked cancers. A high YKL-40 serum concentration was significantly associated with invasive lobular carcinoma, TMN stage III, lymph-node metastases, and death [17]. However, a low adiponectin concentration is associated with a more aggressive BrC character [2,10].Our analysis indicates that the pre-treatment concentration of YKL-40 and TF activity are suitable prognostic biomarkers of progression-free survival (PFS) with respect to considering them as a single parameter as well as combining them. We reported the main significant effects of circulating TF and YKL-40 in predicting disease relapse, indicating that an elevated YKL-40 level and TF activity demonstrate negative prognostic value. Furthermore, a worse future prognosis was shown for subjects with a combination of a higher YKL-40 concentration (>1.93 mg/mL) with higher TF activity (above 14.18 pM), since both components exert pro-tumorigenic properties. The cut-off values were established according to the median values for the whole study population. As YKL-40 is an inflammation-related molecule, its elevated level was positively associated with poor disease-free survival [14,18]. Likewise, YKL-40 controls key pathways and processes within the tumour microenvironment, including inflammation, angiogenesis, cell proliferation, differentiation, and the remodelling of the extracellular matrix and thus promotes tumour progression [15]. Interestingly, Roslind et al. claimed that serum concentrations of YKL-40 may provide a more consistent biomarker of a specific patient’s disease progression as intratumoral YKL-40 expression can vary across a single breast cancer nodule [16]. Interestingly, TF can also be secreted under pro-inflammatory conditions in a soluble form. TF can be hyperexpressed in response to TNF-α and IL-1β release [21]. Perhaps YKL-40 also stimulates TF shedding, since it is a pro-inflammatory protein. These outcomes support the concept that the inflammatory response is a complex systemic reaction and is associated with a poor outcome in patients with invasive BrC.Undoubtedly, an excess amount of body adipose tissue is a risk factor for breast cancer development and is associated with higher cancer mortality [3]. Surprisingly, in our study, we reported that adiposity positively influences overall survival. Similar to the results for the Kaplan–Meier plots, the Cox’s analysis showed that a low BMI (normal weight) was a predictive factor for a shorter time to disease recurrence in BrC subjects. Undeniably, this clinicopathological characteristic is crucial in disease prognosis; however, our normal-weight cohort were not affected substantially (Figure 2C). We found that patients with a BMI > 25 kg/m2 had a three-fold lower risk for recurrence than patients with a BMI < 25 kg/m2. Pajares et al. showed that obesity (BMI 30.0 to 34.9 kg/m2) is not associated with worse survival outcomes in operable BrC patients; only severely obese patients (BMI ≥ 35.0) had an increased risk of recurrence [6]. Additionally, Widschwendter et al. noted that severely obese subjects demonstrated significantly worse overall survival than underweight or normal-weight cases [5]. However, Playdon et al. noted that for each 5 kg/m2 rise in BMI, there was an approximately 18% increase in the risk of total breast cancer mortality [23]. Since the data are inconsistent, further evaluations are still required.Furthermore, our study has shown increased rates of recurrence, as well as cancer-specific mortality, in the cases with a normal BMI and a lower concentration of adiponectin as well as those with a normal BMI and a higher concentration of leptin and YKL-40, independently of other prognostic covariates. Interestingly, in our study, leptin and adiponectin did not demonstrate predictive value as single biomarkers. Thus, a normal BMI is not equivalent to proper body composition. We speculate that it can be associated with subclinical insulin resistance and hyperinsulinemia independent of adiposity, since an altered adipokine profile (high level of leptin and low level of adiponectin) leads to metabolic derangements. It follows that patients with a low adiponectin concentration develop a biologically more aggressive BrC phenotype independently of hormone receptor status [2,10], since a low serum adiponectin level is associated with elevated levels of vascular endothelial growth factor (VEGF) and insulin-like growth factor (IGF) [30]. Moreover, adiponectin exerts efficient tumour growth-limiting influence by the inhibition of the inflammasomes in breast cancer cells [10]. Adiponectin is also considered an anti-inflammatory cytokine due to its ability to inhibit nuclear factor kB (NF-kB) phosphorylation and exhibits anti-migratory activity by inhibiting the Wnt/β-catenin signalling pathway, fundamental for cancer progression [2]. However, elevated leptin levels are associated with breast cancer aggressiveness and a bad prognosis. Leptin is believed to be involved in the growth and invasion of breast cancer cells by stimulating the conversion of aromatisable androgens (androstenedione and dehydroepiandrosterone) to oestradiol [4,11]. Therefore, the diminution of adiponectin (an anti-inflammatory adipokine) and exacerbation of leptin and YKL-40 (pro-inflammatory cytokines) in normal-weight breast cancer subjects may be associated with a more aggressive tumour phenotype. Additional research is needed to further determine the exact relationship between our observed interaction and the modulation of adipokines in breast cancer.From our study, we suggest that pre-treatment obesity itself is not the cause of cancer-associated sequelae, but the adiposity-related metabolic abnormalities include pro-inflammatory macrophage polarisation, insulin resistance, and disproportionate adipokine secretion by adipose tissue. We herein propose a tentative division of patients into metabolically healthy obese subjects and metabolically unhealthy lean individuals. It suggests a substantial heterogeneity of metabolic features in normal-weight breast cancer females. Our findings potentially reject the hypothesis that adiposity is an independent prognostic factor for developing distant metastases and death as a result of breast cancer. Apparently, obesity is a risk factor for hormone-sensitive cancer development, but it improves future outcomes. It is likely we observed a kind of “obesity paradox phenomenon” in those cases. There are data demonstrating that a moderately increased BMI ameliorates survival and the response to therapy [31]. Sánchez-Jiménez et al. suggest that increased fat mass may provide an energy depot that may be beneficial for a longer survival rate [11]. Murphy W.J. and Longo D.L. have noted a positive impact of adiposity on a better response to immunotherapy in different types of cancer [32].According to the multivariate Cox proportional risk model and the excellent area under the ROC curve for the tested model, we have demonstrated the usefulness of applying pre-treatment YKL-40, BMI, leptin, and adiponectin levels and TF activity together as biomarkers for predicting disease relapse. Thus, our results suggest that a non-obese phenotype with high leptin and YKL-40 and low adiponectin levels in breast cancer leads to cancer cell migration and invasion to promote metastasis and reduced survival. Using the biomarkers proposed here would provide more information about future prognosis and survival rates. Leptin has been identified as a key driver in the progression of breast cancer [27]. Adiponectin was originally defined as a tumour suppressor protein by promoting the inhibition of cell proliferation and sensitisation of cells to apoptosis [27]. YKL-40 is a protein that stimulates cancer cell dissemination and progression. Adipokines could be useful diagnostic, prognostic, and predictive biomarkers, reflecting advanced-stage BrC, adverse prognosis, and an inflammatory state.The strength of this study is that it is based on an early-stage breast cancer cohort with complete information on clinicopathological characteristics and more than four years of follow-up. In addition, the prospective design contributed to a reduction in potential biases. The major limitations of this study include the small sample size and the lack of information related to waist circumference (WC) and waist-to-hip ratio (WHR). The BMI was evaluated only at the beginning of follow-up, and further fluctuations were not examined. Additionally, the analyses in this study were conducted using a single measurement of the YKL-40, adipokines, and TF activity from one pre-treatment blood sample. The patients were recruited from a single institution, predisposing to selection and referral bias and potentially limiting the generalisability to larger populations. We excluded patients with Stage IIIA or higher and metastatic BrC tumours; perhaps this action limited the ability to assess the analysed parameters in more advanced tumours. Finally, since we only studied individuals of Polish descent, our results are, therefore, not necessarily directly applicable to other ethnic groups.Our study concluded that a combination of high baseline serum levels of YKL-40 and leptin with a normal BMI as well as of a normal BMI with a lower adiponectin level confer a poor prognosis. Additionally, a single higher YKL-40 concentration and single elevated TF activity were significantly associated with poorer prognosis of breast cancer over the four-year follow-up. According to our results, we suggest that overweight breast cancer patients have a better prognosis regardless of the YKL-40, leptin, and adiponectin levels since we estimated an approximately three-fold increased risk of disease recurrence or death for normal-weight versus obese women. Further investigations that elucidate the mechanisms concerning adipose tissue derivatives, including adipokines and cytokines, may help to understand and prevent severe outcomes in normal-weight-related breast cancer.Conceptualization, K.B., B.R.-C., P.R.; Data curation, B.R.-C., P.R.; Formal analysis, K.B.; Funding acquisition, K.B., P.R., M.B.; Investigation, K.B., B.R.-C.; Methodology, K.B., B.R.-C.; Project administration, B.R.-C., M.B.; Resources, K.B., P.R., M.B.; Supervision, B.R.-C.; Software, K.B.; Validation, K.B., B.R.-C.; Writing—Original draft, K.B., B.R.-C.; writing—Review & editing, B.R.-C. All authors have read and agreed to the published version of the manuscript.The authors received no specific funding for this work.We would like to thank all of the patients for participating in our study. Many thanks to the biomedical laboratory scientists Barbara Góralczyk, Małgorzata Michalska for their contribution in samples preparation and laboratory assistance.The authors declare no conflict of interest.Receiver operating characteristic (ROC) curve for the tested model in patients with versus without disease progression. The area under the curve (AUC) and the p value are indicated.Kaplan–Meier survival curves showing baseline serum YKL-40 (A), TF activity (B), and BMI (C), divided according to cut-offs for YKL-40: < 1.6 ng/mL, 1.6–2.5 ng/mL, and >2.5 ng/mL; TF activity: <12 pM, 12–24 pM, and >24 pM; and BMI value: <24.9 kg/m2, 25–29.9 kg/m2, and >30 kg/m2, respectively.Combination of the YKL-40 level with TF activity (A), the value of BMI with YKL-40 (B), and BMI with TF activity (C), in disease relapse prediction.Combination of BMI with the leptin concentration (A) and the combination of BMI with the adiponectin concentration (B) with respect to disease relapse prediction.Kaplan–Meier survival analysis according to parity status.The biomarker concentrations according to the demographic and anthropometrical determinants in the study population.Abbreviations: Significant p-values are highlighted: * <0.01, ** <0.001, *** p < 0.0001; TF—tissue factor; hsAdiponectin—high sensitive adiponectin.Baseline median and (IQR) serum levels of YKL-40, leptin, and adiponectin as well as citrate plasma TF activity stratified by Ki67, HER2 and hormone receptor expression.Abbreviations: IQR—interquartile range, Ki67—proliferation marker, HER2—human epidermal growth factor receptor 2, ER—oestrogen receptor, PR—progesterone receptor, TF—tissue factor, hsAdiponectin—high sensitive adiponectin; significant p-values are highlighted: * <0.01.The biomarker concentrations with respect to clinical and pathological characteristics in breast cancer (BrC) subjects.Abbreviations: LN—free of lymph node involvement, LN+—lymph node involvement, TF—tissue factor, hsAdiponectin—high sensitive adiponectin; significant p-values are highlighted: * <0.01; ** <0.001.The multivariate Cox HRs for PFS according to the baseline serum values of BMI, TF activity, YKL-40, leptin, and adiponectin.Abbreviations: HR—hazard ratio, PFS—progression free survival, CI—confidence interval, BMI—body mass index, TF—tissue factor, hsAdiponectin—high sensitive adiponectin; significant p-values are in bold.Distribution of YKL-40, TF activity, leptin, and adiponectin in recurrent patients.Abbreviations: TF—tissue factor, hsAdiponectin—high sensitive adiponectin.

Med-MDPI/jcm_2/jcm-09-06-01743.txt

@@ -0,0 +1 @@
+The relationship between smoking and the risk of pregnancy-induced hypertension (PIH) is not clearly established. Therefore, we conducted an analysis of cigarette smoking in a Polish cohort of women, recruited in the first trimester of a single pregnancy in 2015–2016. We evaluated the women who subsequently developed PIH (n = 137) (gestational hypertension—GH (n = 113) and pre-eclampsia—PE (n = 24)), and the women who remained normotensive (n = 775). The diseases odds ratios (and 95% CI—confidence intervals) were calculated in a multivariate logistic regression. In the PIH cases (vs. normotensive women) we found more smokers (25.6% vs. 17.2%, p = 0.020) including smokers in the first trimester (14.6% vs. 4.8%, p < 0.001). The average number of cigarettes smoked daily per smokers in the first trimester was 11.2 (range 2–30), and the average number of years of smoking was 11.6 (range 2–25). The number of years of smoking was a stronger risk factor for GH and PE than the number of cigarettes/day. Compared to the women who have never smoked, smoking ever before pregnancy was associated with a higher GH risk (AOR = 1.68; p = 0.043), and with no effect on PE risk (OR = 0.97; p = 0.950). Smokers in the first trimester had a higher odds ratio of GH (AOR = 4.75; p < 0.001) and PE (OR = 2.60; p = 0.136). Quitting smoking before pregnancy (ex-smokers) was associated with a lower odds ratio of GH (AOR = 0.83; p = 0.596) and PE (OR = 0.33; p = 0.288). However, quitting smoking during pregnancy was associated with a higher risk of GH (AOR = 11.63; p < 0.0001) and PE (OR = 3.57; p = 0.238). After dissection of the cohort into pre-pregnancy body–mass index (BMI) categories, smoking in the first trimester was associated with the higher hypertension risk in underweight women (OR = 22.00, p = 0.024). Conclusions: The factors that increased the risk of GH and PE were smoking in the first trimester and (paradoxically and more strongly) smoking cessation during pregnancy. Our results suggest that women of childbearing potential should be encouraged to quit smoking before pregnancy.Pregnancy-induced hypertension (PIH) is characterized by an increase in de novo blood pressure after the 20th week of pregnancy [1]. It affects an average of about 7–10% of pregnant women and includes isolated gestational hypertension (GH) and pre-eclampsia (PE) [1,2,3]. PE is accompanied by organ disorders and it is one of the main causes of morbidity and mortality of mothers and fetuses. GH has a milder course (not associated with organ disorders), but also increases the risk of adverse pregnancy outcomes, compared to normotensive women. There remains doubt as to whether these are separate diseases. However, they have been shown to have many common risk factors as well as pathogenesis elements [1,4]. In the pathomechanism of PE, the primary role of trophoblast invasion disorders in the period of placenta development has been demonstrated, which results in ischemia and hypoxia of the placenta, an increase in oxidative stress and the initiation of numerous processes leading ultimately to endothelial damage [5,6,7]. The result is the increase in blood pressure [1]. Numerous studies have found dyslipidemia, inflammation, oxidative stress as the main element of the pathogenesis of GH and PE [1,2,4,6]. It has been previously shown that lower levels of antioxidants in early pregnancy may increase the risk of PIH [8,9].There are significant discrepancies in the relationship between smoking and the development of PIH [10,11,12,13]. Establishing these relationships is important because it is estimated that around a quarter of the population in the European Region smokes [14,15], and maternal smoking is the most prevalent preventable cause of pregnancy complications such as neonatal morbidity, low birthweight, and premature delivery [16]. Many studies have shown that smoking during pregnancy (paradoxically) reduces the risk of PE. However, there are also opposite results, and mechanisms confirming the beneficial or harmful effects of smoking on this risk have not been established [11,12,13]. Tobacco smoke contains more than 7000 chemicals, among them various toxic and carcinogenic and cocarcinogenic substances [8,9,11,12,17,18,19,20]. Smoking has been shown to trigger inflammatory processes and to increase oxidative stress, as well as to reduce placental flows [16,17]. On the other hand, the study by Laule et al. indicates that nicotine can have anti-inflammatory effects through the α7-nAChR (acetylcholine receptor subunit) and inhibition of pro-inflammatory cytokine production, which may attenuate hypertension caused by placental ischemia. At the same time, this beneficial effect can be overshadowed by other effects of nicotine on nicotinic receptors and the ability to increase systemic blood pressure (directly) [11]. Another component of cigarette smoke, carbon monoxide (CO), has the property of reducing vascular tone and inhibiting inflammatory cascades [11].Both systematic reviews by England et al. (covering 48 epidemiologic studies from 1959���2006) and a systematic review by Conde-Agudelo et al. (covering 35 studies from 1966–1998) showed the inverse relationship between smoking during pregnancy, and PE [12,13]. However, several studies have shown that smoking during pregnancy increased the risk of hypertension in pregnancy [10,21,22,23,24]. In a population-based prospective cohort study, Bakker et al. found a slight increase in the risk of PE for smoking in the first trimester [22]. In a prospective study of 508 single-pregnant women in the 16–22th week, Rauchfuss et al. found that women who never smoked and those who reduced the number of cigarettes smoked during pregnancy had the lowest risk of PIH; and the women who quit smoking before or during pregnancy increased the risk of PE [23]. On the contrary, in a registry-based study, Kharkova et al., found negative relationship between smoking and PE risk and the risk was the same in the women who quit smoking during the first trimester [25]. Different designs and methodologies of research, including various risks of the studied populations or various sample size, could have influenced the discrepancy. Studies found in the literature focused mainly on the assessment of PE.Our study is one of the first to comprehensively assess smoking in Polish pregnant women for the risk of GH and PE separately. In a multivariate statistic model we examined the risk of the diseases for several categories of smoking of cigarettes: smokers (ever before pregnancy), the women who quit smoking before pregnancy (ex-smokers), smokers in the first trimester, those who quit or reduced smoking during pregnancy, or smoked unchanged until the end of pregnancy, as well as considering the length of smoking time, number of cigarettes, and pack-years. In addition, we conducted a study after dissection into pre-pregnancy body–mass index (BMI) categories.The study was conducted in accordance with the Helsinki Declaration. It was approved by the Bioethics Committee of the Medical University of Poznan, Poland (approval number 769/15). All the participants signed the Informed Consent Form. The current analysis was conducted in a prospective cohort of pregnant women recruited in the end of the first trimester. The participants were recruited at the Obstetric-gynecological and Neonatological Center (Poznan, Poland); It is a III-grade reference center, with 6000–8000 deliveries a year. The participants were recruited in 2015–2016; Pregnancy outcomes were collected in 2016–2017; Statistical Analyses were conducted in 2017–2019.We recruited white (Caucasian) women of descent from one region of Poland (Wielkopolska) in the 10 (+0) − 14 (+6) th week of a single pregnancy without aneuploidy. The criteria also included delivery of a phenotypically normal child ≥25th gestational week, and mother’s age at conception between 18–45.The criteria for excluding candidates from the original cohort were chronic diseases, e.g., hypertension or/and diabetes mellitus, and kidney or liver diseases, as well as immunological and inflammatory diseases, and thromboembolism.Women’s characteristics were collected using a questionnaire. The participants filled it out themselves (but in the presence of midwives) during recruitment. We collected the information such as age, height, and pre-pregnancy weight. The set of data also included current pregnancy history, obstetrical and gynecological histories, and concurrent diseases, as well as socioeconomic and demographic characteristics (education status, financial status, and place of residence, and marital status). We also collected the information about the use of multivitamins and medications, as well as about family history. Importantly (especially for the current analysis) we collected information about smoking. All women declared consuming no alcohol in pregnancy.After the end of pregnancy, outcomes were taken from medical records. All pregnancy outcomes (among others, birth weight and intrauterine growth restriction (IUGR), gestational age at birth and preterm birth, as well as fetal sex and APGAR results) and pregnancy complications (e.g., hypertension in pregnancy (PIH) and its main forms, and gestational diabetes mellitus (GDM) and its main forms) have been recorded.The recruitment was conducted among women taking typical laboratory tests. Information about the study was available to everyone in the laboratory. Participation in the study was voluntary. It was proposed to each of 1300 women who had the adopted criteria, in 2015–2016 (in the period of 12 months). Missing data concerned 388 (29.9%) women (the participants who did not meet the inclusion criteria after the end of pregnancy, as well as women whose data were incomplete). The original cohort consisted of 912 women.In the current analysis, the study group (n = 137) consisted of the women who subsequently developed hypertension in pregnancy (113 women developed GH and 24 women developed PE), and the control group (n = 775) consisted of the participants who remained normotensive. The aim of the current analysis was to assess the relationship between several categories of smoking, and the risk of hypertension in pregnancy (GH and PE), in the whole cohort and after breaking down into maternal pre-pregnancy BMI categories.Information on cigarette smoking addiction was self-reported. In the current analysis we evaluated the following smoking categories: women who have never smoked, smokers (all smokers before and during pregnancy), ex-smokers (women who quit smoking before pregnancy), smokers in the first trimester, those who quit smoking during pregnancy (in II-III trimester), smokers who reduced smoking during pregnancy (in II-III trimester), those who smoked unchanged until the end of pregnancy, considering the number of years of smoking, the number of cigarettes smoked a day and pack-years. The value of pack-years of smoking for each woman was the result of multiplying the number of years of smoking and the size of a packet of cigarettes (calculated as a ratio of the number of cigarettes smoked per day/20 cigarettes in a packet). We compared smoking categories in the case and control group. Height and pre-pregnancy weight were self-reported. The height was also measured in the hospital (before delivery) and the values from hospital records were included in the analysis. Pre-pregnancy BMI (a ratio of the weight and the height2, kg/m2) was calculated for each woman). Normal BMI was defined as 18.5–24.99 kg/m2. Underweight, overweight and obesity were defined as BMI <18.5 kg/m2, 25.00–29.99 kg/m2 and ≥30.00 kg/m2, respectively.Gestational weight gain (GWG) was calculated as the difference between the weight measured before delivery (data available in the medical records) and the pre-pregnancy weight. The GWG ranges recommended by the US National Academy of Medicine (formerly the Institute of Medicine) were defined as 12.5–18 kg for underweight women, 11.5–16 kg for normal BMI, 7–11.5 kg for overweight women and 5–9 kg for obese women.Pregnancy outcomes were taken from the medical records. PIH was defined in accordance with the Polish guidelines (2015), as “arterial pressure equal to and higher than 140/90 mmHg (on two occasions, at least 4 h apart, in a sitting position, with an oscillometric device) developed de novo after the 20th week of pregnancy, receding up to 12 weeks after delivery.” PIH includes two (main) forms. “Gestational hypertension (GH) was diagnosed if no other disturbance was found; PE was diagnosed when any of the following appeared de novo: Proteinuria (≥300 mg/day or ≥0.3 g/L; protein/creatinine ratio ≥0.3; 1+ in the strip test); thrombocytopenia <100 G/L; worsening of renal function; damage to the liver function; pulmonary edema; symptoms from the central nervous system; blurred vision. Intrauterine growth restriction (IUGR) was not one of the criteria of diagnosis”. Only proteinuria was found in all cases of PE in this cohort (≥0.3 g/L).The mother’s arterial pressure (systolic and diastolic) was measured before and after delivery, and the mother’s arterial pressure from a postpartum ward were included in the analysis. Values of blood pressure before pregnancy were self-reported.Statistical analyses were performed using the Statistica 13 software. Before statistical analyses, the Shapiro–Wilk test was used to assess the normality of the data distribution. The Mann–Whitney U test was applied for comparisons of continuous variables; means and medians were given as description. For binomial categories, the Pearson chi-square test (or Fisher exact test when Cochran assumption was not met) was used (p-value < 0.05 was assumed to be significant). Only available data were taken into consideration.The impact of smoking on the risk of developing PIH and its forms (isolated GH and PE, separately) was assessed in the logistic regression. p-value was calculated using the Wald test (p-value < 0.05 was considered to be significant). Odds ratios of PIH (and 95% confidence intervals CI) were calculated for smoking categories expressed as continuous variables, and for each smoking category expressed as categorical variables with respect to the reference category which was assigned OR = 1.00; the reference category was “women who have never smoked” or “smokers in the first trimester”.Crude odds ratios (OR) of PIH for smoking categories were calculated in the univariate logistic regression, and adjusted odds ratios (AOR) of PIH for smoking categories were calculated in the multivariate logistic regression after adjusting for maternal age, primiparous, pre-pregnancy BMI, GWG outside the range of the recommendations regardless of the BMI category, prior PIH, and infertility treatment.Graphs of PIH risk were made based on a one-and multidimensional logistic model regression models presenting the odds ratios (OR and AOR) and 95% confidence intervals on a logarithmic scale (using the PQStat software).Characteristics of participants in the control and study groups are presented in Table 1 and Table S1.In the whole cohort (Table S1), 775 (85%) women remained normotensive and 137 (15%) women developed PIH, including GH (n = 113) and PE (n = 24). In the cohort, 168 (18.4%) women were smokers, 111 (12.1%) women quit smoking before pregnancy, and 57 (6.3%) smoked in the first trimester (22 women quit smoking during pregnancy, 10 women reduced smoking during pregnancy, and 25 women smoked unchanged until the end of pregnancy). The average number of cigarettes smoked daily per person among the smokers in the first trimester was 11.2 (range 2–30), and the average number of years of smoking was 11.6 (range 2–25) (Table S1).General characteristics of the groups of GH and PE are presented in Table 1. The women in the GH group (compared to the normotensive women) were significantly older, had significantly higher pre-pregnancy BMI, gave birth significantly earlier, and gave birth to newborns with significantly lower weight. In the GH group, we found a statistically significantly higher number of smokers (ever before pregnancy) and smokers in the first trimester. Similar profiles of the results (but not identical) were found for comparison of normotensive controls and cases of PE (Table 1).Figure 1 presents graphic pictures of the risk of GH (A) and PE (B) for studied smoking categories. Compared to the women who have never smoked, women who smoked in the first trimester as well as women who quit smoking during pregnancy and those who reduced smoking in pregnancy had a higher risk of GH and PE.The AOR of both forms of hypertension for smoking categories are presented in Table 2, Table 3 and Table S2. Each analysis covered the cases and normotensive controls.Compared to the women who have never smoked (Table 2), smoking ever before pregnancy was associated with a higher risk of GH (AOR = 1.68; p = 0.043), and with no effect on PE risk. Smokers in the first trimester had a 4.75-fold higher adjusted odds ratio of GH (p < 0.001) and a 2.60-fold higher odds ratio of PE (p = 0.136). Quitting smoking before pregnancy (ex-smokers) was associated with a lower odds ratio of GH and PE (OR/AOR < 1.0). However, quitting smoking during pregnancy was associated with a 11.63-fold higher adjusted risk of GH (p < 0.0001) and a 3.57-fold higher risk of PE (p = 0.238). Women who reduced smoking in pregnancy also had a higher risk of GH and PE, but in a statistically insignificant way.The results for another reference category (smokers in the first trimester) are presented in Table S2.Analyses of continuous variables (Table 3) showed that the number of years of smoking (ever before pregnancy) was a stronger risk factor of GH and PE than the number of cigarettes/day. An increase in the number of years of smoking of 1 year increases the adjusted risk of GH by 5% (AOR = 1.05, p = 0.006). The results for number of cigarettes/day and pack-years were statistically insignificant after adjustment. The profile of the results for PE risk was similar, but the results were statistically insignificant.The odds ratios of PIH (and its forms) after dissection of the cohort into pre-pregnancy BMI categories are presented in Table 4 and Table S3.Table 4 presents the results for all cases of PIH because some subgroups were small and / or included a small number of cases. The highest risk of PIH for smoking in the first trimester was found in women with pre-pregnancy underweight (OR = 22.00; p = 0.024).Table S3 presents the results for GH and PE separately. The highest risk of these forms of hypertension for smoking in the first trimester was also found in women with pre-pregnancy underweight.In our prospective cohort study, we found that smoking in the first trimester increased the risk of isolated GH and PE, compared to the women who have never smoked. Smoking ever before pregnancy was associated with weaker results, and was associated with a higher risk of GH and with a negligible effect on the risk of PE. Compared to the women who have never smoked, quitting smoking before pregnancy (ex-smokers) was associated with a lower odds ratio of GH and PE. However, quitting smoking during pregnancy was associated with a higher risk of GH (much higher) and PE.Our results are consistent with some reports in the literature [12,13,21,22,23]. In a population-based retrospective cohort study (USA) after excluding white-non-Hispanic women and Indian non-Hispanic women, Chang et al. found an increased risk of PE for smoking during pregnancy [21]. In a prospective cohort study of 508 single-pregnant women in the 16–22th week, Rauchfuss et al. found that women who never smoked had the lowest risk of pre-eclampsia; and the women who quit smoking during pregnancy increased the risk of PE [23]. In a population-based prospective cohort study, Bakker et al. found a slight increase in the risk of PE for smoking in the first trimester [22]. In a prospective study covering 605 women without chronic hypertension in the 24–26th week, Luo et al. found no effect of smoking during pregnancy on pre-eclampsia risk, but found an over 6-fold higher risk of PE for “previous smokers”, compared to “nonsmokers” [24].Contrary to our results, many studies (and meta-analyses) showed a reduction in the risk of PE for smokers during pregnancy [10,12,13,25]. However, the strength of the relationship between smoking and PE (it is about 50% reduction in heavy smokers) is not as pronounced as with other smoking-related diseases [12]. The cited studies differed in their structure (retrospective or prospective studies) and methodology. The discrepancies found could be attributed to the risk differences of the surveyed populations, the different numbers of the studied groups, the different degree of matching of maternal features and various risk factors used to adjust the odds ratios.The analysis of our results requires highlighting a few facts. The frequency of all PIH cases in our cohort (n = 137 = 15.0%) is 2–3 times higher than in other Polish studies [26,27]. The probable reason for this is the fact that the study was performed in a third-degree reference center, where women with risk factors report for additional tests. At the same time, women who developed pregnancy complications and those with risk factors (e.g., older or obese women) are more likely to cooperate. In our study, we found a small number of cases of pre-eclampsia (n = 24 = 2.6%), which is consistent with the latest FIGO report [1] in which the average occurrence of PE in the world was estimated at 2–5%. Importantly, in our study, we excluded a priori some risk factors, including multiple pregnancy, pre-existing hypertension, and other chronic diseases.Secondly, in our study, 6.3% of women were smokers during pregnancy (in the first trimester), and 18.4% of participants were smokers ever before pregnancy. In addition, this is in line with the reports of recent meta-analyses presented in the Lancet; the prevalence of smoking during pregnancy was estimated to be 8.1% in the European Region (6.0% in Poland in 2015) [15]. These values are lower than presented in the World Health Organization (WHO) reports in previous years. In Poland, amendments to the Health Protection Act against the effects of tobacco and tobacco products took place in 2010 and 2012; in 2015 tobacco products were used by every fourth citizen (about every fifth woman) [14,15].The mechanisms of the relationship between smoking and PIH are not clear. This is a heterogeneous disease entity. Isolated GH and PE may have common and different elements of pathophysiology. PE also has different phenotypes. In the pathomechanism of PE, the primary role of trophoblast invasion disorders in the period of placenta development and placental ischemia has been demonstrated. In addition, increased oxidative stress and inflammation as well as endothelial damage are main elements of the disease pathogenesis [2,5,6,7,28]. Importantly, different components of tobacco smoke may act on different processes or they may act differently on the same cellular processes, which may cause discrepancies between tests.There are some mechanisms that may explain the increased risk of GH and PE in smokers [29]. Generally, pathomechanisms triggered by smoking (by the components of tobacco smoke) may affect the processes connected with the development of the placenta and cause damage to the structure and function of the vascular endothelium of the mother. It was shown that exposure to tobacco smoke triggers inflammation and oxidative stress in various organs and causes impairment of thrombotic events [17,29,30]. First, it has been previously shown that lower levels of micronutrients with antioxidant properties (e.g., selenium) may increase the risk of PIH [8,9,31,32]. At the same time, smoking has also been associated with lower selenium levels [8,9]. The increase in the level of lead and cadmium in smokers may cause a decrease in the concentration of antioxidant selenium that binds heavy metals [33]. Placental cadmium levels were also associated with increased risk of PE [34]. Secondly, it is possible that tobacco smoke affects the oxidative damage to endothelial cells (these cells synthesize endogenous nitric oxide NO, which has vasodilating properties). This can cause an imbalance between the action of vasoconstricting and vasodilating factors [35]. Thirdly, nicotine increases blood pressure by acting on the cardiovascular system [11]. However, it remains unexplained why smoking cessation during pregnancy increases the risk of hypertension. This requires testing in a large sample.We consider our pilot study of the additive effect of smoking and the pre-pregnancy underweight (BMI < 18.5 kg/m2) on risk of hypertension in pregnancy to be extremely important (Table 4 and Table S3). It is possible that the result is associated with deficiencies of many nutrients in underweight people, including microelements and vitamins involved in oxidative balance, and inflammatory or immunological processes [36]. The sizes of the subgroups were small, but we believe that our result requires further research.Some mechanisms may explain the reduction in the risk of PE in smokers [10]. For example, nicotine can have anti-inflammatory effects through the acetylcholine receptor subunit (α7-nAChR) and inhibition of pro-inflammatory cytokine production, which may attenuate hypertension caused by placental ischemia [11]. Exogenous carbon monoxide (CO) may have immunosuppressive effects and has the property of reducing vascular tone [11,12]. It was found that components of tobacco smoke reduce the dose-dependent level of sFlt-1 (an anti-angiogenic factor released by epithelial cells in the placenta), and previous studies correlated higher levels of sFlt-1 in early pregnancy with a higher risk of PIH/PE [10,37]. Nicotine inhibits the production of thromboxane A2, which may explain the results of a reduced risk of PE in smokers. Nicotine can reduce the volume of plasma by affecting the production of vaso-constrictive prostaglandins [10]. The possibility of permanent adaptation of the mother’s circulatory system is also taken into consideration in smokers [10]. An advantage of this study was its prospective one-center cohort model. We have assessed the most possible smoking categories to get a full picture of the problem. We have taken into account many risk factors; some have been excluded a priori before the recruitment, and others have been used to correct the odds ratios. However, it is possible that there are other confounding variables that affect the results. The women came from one region, which matched them well in terms of the comparable level of prenatal care.We are aware of the limitations of this study. We do not have any information on passive smoking. The participants reported their habits themselves. The subgroup sizes were smaller and met the minimum size requirements, but the limited group sizes account for a larger margin of error. The group of PE cases was not numerous.In our study, the factors that increased the risk of isolated GH (statistically significant) and PE (statistically insignificant) were smoking in the first trimester and (paradoxically more strongly) smoking cessation during pregnancy. Our results were sustained after adjusting for many confounding variables.Our results suggest that women of childbearing potential should be encouraged to quitting smoking before pregnancy. Efforts should be made to make women aware of the negative effects of smoking. Increased risk of hypertension in the women who quit smoking during pregnancy requires further research and clarification.We also found that the risk of PIH for smoking in the first trimester increased more strongly in the women with pre-pregnancy underweight. Although this and some of our other results were obtained by examining small subgroups, further research in this topic is needed.The following are available online at https://www.mdpi.com/2077-0383/9/6/1743/s1, Table S1: Full characteristics of mothers in the control and PIH group; Table S2: The adjusted odds ratios of gestational hypertension (GH) and PE for all smoking categories; Table S3: The odds ratios of gestational hypertension (GH) and pre-eclampsia (PE) for smoking in the first trimester, after dissection into pre-pregnancy BMI categories.Conceptualization, M.L.; methodology, M.L.; software, M.L.; validation, B.W. and M.L.; formal analysis, M.L.; investigation, M.L.; resources, M.L.; data curation, B.W. and M.L.; writing—original draft preparation, M.L.; writing—review and editing, M.L.; visualization, B.W. and M.L.; supervision, M.L.; project administration, M.L. All authors have read and agreed to the published version of the manuscript.This research received no external funding.The authors would like to thank pregnant women at the University Obstetrics and Gynecology Hospital in Poznań, Poland for their participation in this study.The authors declare no conflict of interest.Odds ratios (and 95% confidence intervals) of gestational hypertension (A) and pre-eclampsia (B) for smoking categories on a logarithmic scale. The adjusted odds ratios (AOR) were calculated after adjusting for maternal age, primiparous women, pre-pregnancy body–mass index (BMI), and gestational weight gain outside the range of the recommendations regardless of the BMI category.Maternal characteristics in the control and study groups.* GH: gestational hypertension and PE: pre-eclampsia; ** The Mann–Whitney U test was used for comparisons of continuous variables; For binomial categories the Pearson chi-square test (or Fisher exact test when Cochran assumption was not met) was used (p < 0.05 was assumed to be significant); *** GWG: gestational weight gain.The adjusted odds ratios of gestational hypertension (GH) and pre-eclampsia (PE) for smoking categories.* AOR: adjusted odds ratios (CI—confidence intervals) calculated in the multivariate logistic regression (p < 0.05 was assumed to be significant); ** In these analyses, the adjusted odds ratios were calculated after adjusting for maternal age, primiparous, pre-pregnancy BMI, gestational weight gain outside the range of the recommendations regardless of the BMI category, prior PIH, and infertility treatment; *** In these analyses, the adjusted odds ratios were calculated after adjusting for maternal age, primiparous, pre-pregnancy BMI, gestational weight gain outside the range of the recommendations (the reason for the smaller number of confounders was the small number of cases in the subgroups).Adjusted odds ratios of gestational hypertension (GH) and pre-eclampsia (PE) for additional smoking categories (continuous variables).* AOR: adjusted odds ratios (CI—confidence intervals) calculated in the multivariate logistic regression (p < 0.05 was assumed to be significant); ** In the analyses, the adjusted odds ratios were calculated after adjusting for maternal age, primiparous, pre-pregnancy BMI, gestational weight gain per week, prior PIH, and infertility treatment; *** In the analyses, the adjusted odds ratios were calculated after adjusting for maternal age, primiparous, pre-pregnancy BMI, gestational weight gain outside the range of the recommendations regardless of the BMI category (the reason for the smaller number of confounders was the small number of PE cases).The odds ratios of pregnancy-induced hypertension (PIH) for smoking in the first trimester, after dissection into pre-pregnancy BMI categories.* OR: crude odds ratios (CI—confidence intervals) calculated in the univariate logistic regression (p < 0.05 was assumed to be significant); PIH: pregnancy-induced hypertension; BMI: body–mass index.

Med-MDPI/jcm_2/jcm-09-06-01744.txt

@@ -0,0 +1 @@
+Authors with equal contribution.Background: Due to the higher frequency of ischemic stroke in men compared to women, we aimed to determine if gender differences exist regarding periodontal status and several plasma biomarkers in patients with a recent large artery atherosclerosis ischemic stroke (IS). Material and methods: Patients with their first IS within less than six weeks who were able to undergo periodontal examinations were evaluated. Demographic data, periodontal status, oxidative stress parameters/plasma antioxidant capacity, and C-reactive protein in patients who suffered a recent large artery atherosclerosis ischemic stroke were reccorded. Results: 93 patients were included in the study. More men were smokers (12/57 vs. 3/36) and consumed alcohol (17/57 vs. 3/36), and more women had higher glycemic values (p = 0.023), total cholesterol (p < 0.001), LDL (low-density lipoprotein)-cholesterol (p = 0.010), and HDL (high-density lipoprotein)-cholesterol (p = 0.005) levels. Significantly more men than women had moderate plus severe periodontal disease (p = 0.018), significantly higher levels of nitric oxide (p = 0.034), and significantly lower levels of total antioxidant capacity (p = 0.028). Conclusions: In this pilot study, men seem to be more prone to oxidative stress and to develop more severe forms of periodontitis among patients with stroke, but the results need validation on a larger sample.Ischemic and hemorrhagic stroke are two of the most common causes of mortality and adult disability worldwide [1,2]. Despite the increased risk for stroke in post-menopausal women, men have higher frequency of stroke even in old age, while the cause of this difference is still under evaluation [3,4,5,6]. Almost 80% of strokes are caused by focal cerebral ischemia, while the remaining occur due to hemorrhages [7]. Atherosclerosis of major intracranial arteries remains one of the primary etiological factors for ischemic stroke, [8] but some risk factors, such as age, gender, hypertension, diabetes mellitus, smoking, and severe tooth loss, have been associated with cerebrovascular disease [9].Periodontitis (periodontal disease) is a ubiquitous chronic infectious-inflammatory disease [10,11,12,13,14,15]. Periodontitis has a negative impact on oral function, quality of life, and general health. Moreover, severe periodontitis has been independently associated with increased mortality in several different populations [16].Wu et al. first reported the epidemiological and etiopathogenetic associations between chronic periodontitis and stroke [17], and Straka and Trapezanlidis emphasized the possible relationships between oral infections and stimulation of atherogenic mechanisms [18]. Since then, other studies have reported an association between periodontitis and acute ischemic stroke [17,19,20,21,22,23], while several reviews showed a moderate association between periodontitis and stroke [24,25,26,27,28,29]. However, Khlande et al. found no strong evidence for such an association [30].Several confounding risk factors for periodontitis and cerebrovascular disease have been highlighted, such as diabetes mellitus, smoking, and age [31]. However, an independent association between periodontal disease and incident stroke risk has been reported [32].Leira et al. reported a 2.8-fold higher risk of developing ischemic stroke in patients with periodontitis compared to those without periodontitis [24]. In addition, Aarabi et al. claimed that chronic oral infections are associated with an imaging surrogate of cerebrovascular ischemia beyond acute ischemic stroke (e.g., silent infarcts and brain white matter hyperintensities) related to small vessel disease [33]. A dose-dependent association has also been reported between tooth loss as a measure of periodontitis severity and an increased risk for stroke [27,34,35]. It is not clear whether the stroke-periodontitis relationship occurs prior to the stroke or develops later, but due to the long-standing characteristic of periodontitis in which the tissue destruction develops during many years (or decades), it is highly probable that severe alveolysis develops before stroke [36].Periodontitis could be involved in chronic systemic diseases and the pathogenesis of atherosclerosis via a periodontium-originating low-level bacteremia invading the arterial walls, the release of local inflammatory mediators into the general circulation, autoimmunity to host proteins, and pro-atherogenic effects of bacterial toxins [37,38]. The persistent systemic inflammation triggered by periodontitis induces vascular endothelial dysfunction [25] and may increase inflammation in existing atherosclerotic lesions, thereby increasing the risk of cardiovascular disease and its associated events [37,38]. The concept of the “brain-oral axis” has been introduced to emphasize the impact of oral microbiota on the health of the brain [39,40]. The presence of Porphyromonas gingivalis, an important periodontal pathogen, has been linked to an increased risk of ischemic stroke [41], and a targeted periodontal treatment against this bacteria protects against ischemic stroke [42]. The presence of a proinflammatory status in patients with periodontitis is associated with the increase in systemic pro-inflammatory mediators, such as C-reactive protein (CRP) [43], and stimulation of the innate and adaptive immune responses.One of the essential triggers for inflammation is increased oxidative stress and decreased antioxidant molecular activity. Bacteria present in subgingival areas disrupt this balance and could constitute a significant risk factor for stroke [44,45]. Oxidative stress associated with a lower serum total antioxidant level and salivary antioxidant capacity is involved in the progression of periodontitis [46]. Neutrophils and other phagocytes together with other periodontal cells, such as monocytes, gingival fibroblasts, and periodontal ligament cells, enhance production of reactive oxygen species (ROS) upon stimulation by periodontal pathogens and/or their components in in vitro studies [47,48].Altered redox signaling caused by increased bioavailability of ROS is a significant contributor to the onset and/or progression of atherosclerosis and hypertension [49]. ROS include free radicals, such as superoxide and hydroxyl radical, and non-radicals, such as hydrogen peroxide [50]. Oxidative stress induces inflammation that can further augment ROS production, according to Kumar et al. [44]. Oxidative stress plays a major role in endothelial dysfunction and the onset and progression of atherosclerosis. Superoxide can react with nitric oxide (NO) to form peroxynitrite, which reduces the bioavailability of NO that has vasodilatory function [51].Atherosclerosis and its primary complications, myocardial infarction and stroke, remain a major cause of death and disability worldwide [51]. However, no conclusions can be drawn in the relationship between periodontitis and stroke. If causal, the association would be important because treating periodontitis could potentially reduce the risk for stroke [51]. Well-designed prospective studies should be carried out to provide robust evidence for the association between periodontitis and stroke together with an appropriate adjustment for confounding vascular risk factors and restrictive diagnostic criteria for periodontitis [33]. Considering globally, men continue to have a higher incidence of ischemic stroke than women [6]. In this context, this study evaluated the gender differences regarding periodontal status, oxidative stress parameters/plasma antioxidant capacity, and C-reactive protein in patients who suffered a recent large artery atherosclerosis ischemic stroke.A cohort pilot study was conducted at the Neurology Department, Clinical Rehabilitation Hospital Cluj-Napoca, after receiving ethical approval from the Institutional Ethics Committee (No. 3/2018). Written informed consent was obtained from all participants before they underwent the physical examination, blood sampling, and periodontal examination. This study adhered to principles outlined in the Declaration of Helsinki on experimentation involving human subjects. The study was conducted between May 2018 and February 2019.Patients aged 18–80 years with ischemic stroke, defined according to the World Health Organization criteria as “the rapid development of localized or global signs of brain dysfunction with symptoms lasting more than 24h without any other apparent cause except those of vascular origin” were eligible for the study [52]. The physical examination (I.S. and A.C.B.) was done to assess the degree of neurological deficit and assess general apprerance and orgnas systems (vital signs, deafness, carotid bruit, urinary incontinence or erectile dysfunction, dysarthia, muscle weekness, vertigo, defness, nystagmus and hemiparesis, gait abnormalities or ataxia, cranial nerve abnormalities). Ischemic stroke was confirmed by a cranial computed tomography (CT) examination. All patients with large artery atherosclerosis stroke (sub-type 1, TOAST classification - Trial of Org 10172 in Acute Stroke Treatment) [53] were included in the study.According to the hospital register, investigators consecutively recruited cases each day based on the inclusion/exclusion criteria. The patients were referred from other hospitals within the county as well as from the central-northern part of Romania. All patients included in this study had a complete neurologic and cardiologic evaluation. Patients at their first ischemic stroke were included if ischemic stroke onset was less than 6 weeks, they were able to undergo a dental examination, and they could provide informed consent. The patients included in the study were undergoing antiplatelet therapy initiated during the acute phase of ischemic stroke and antihypertensive therapy was applied for those who had hypertension. Patients with diabetes mellitus were under treatment with oral hypoglycemic drugs or insulin therapy.All patients with a hemorrhagic stroke (confirmed by cranial CT examination) or recent myocardial infarction were excluded. Patients with acute infections (requiring antibiotic treatment), inflammatory disorders, degenerative brain diseases, oncologic diseases (in the past 5 years), on immunosuppressant therapy, with periodontal treatment in the last year or with recent (<1 month) history of tooth extraction were also excluded. The STROBE (StrengThening the Reporting of Observational Studies in Epidemiology) guidelines were used to ensure accurate reporting of this study [54].Demographic data, such as age, gender, BMI (kg/m2), and behavioral risk factors (namely smoking, and alcohol consumption) were recorded. Blood pressure (systolic and diastolic blood pressure), and electrocardiogram were measured in all patients as input data to identify the presence of cardiac co-morbidities.Blood samples were taken from all subjects included in this study in the morning after a 12 h fast. The following parameters were measured: basal glycemia, total cholesterol, LDL and HDL-cholesterol, triglycerides, and CRP (a biomarker for inflammation). Oxidative stress was assessed by measuring the oxidative stress biomarkers, such as MDA (malondialdehyde), NOx (the indirect assessment of NO synthesis), total oxidative stress (TOS), and plasma antioxidant status was assessed by measuring the thiol, catalase and total antioxidant capacity (TAC) plasma level, as described previously [55,56].The plasma parameters of oxidative stress/antioxidant capacity were assessed using a spectrophotometer (Jasco International Co., Ltd., Tokyo, Japan). MDA assessment was made by using thiobarbituric acid reaction, reported as a standard method [57,58,59]. Briefly, the serum sample was mixed with trichloracetic acid and vigorously vortexed, then thiobarbituric acid was added. After the solution was kept for 30 min at 95 °C followed by ice cooling and centrifugation, the absorbance of each sample was determined at 530 nm compared with a blank [55,56]. The NOx was assessed by Griess reaction (reduction of nitrate to nitrite by mixing the plasma sample with vanadium chloride and further mixing with Griess reagent, followed by sample absorbance measurement at 540 nm [55,56]). TOS was determined by Erel method: plasma was added to Erel reagent, vortexed, incubated for 30 min at room temperature and followed by absorbance measurement at 560 nm [60]. TAC assessment was made also by a method described by Erel: plasma was mixed with Erel reactive, kept for 10 min at room temperature, followed by absorbance determination at 444 nm [60]. Catalase was assessed by the method described by Aebi [61] that measure the catalase activity from decomposition of hydrogen peroxide. Thiols assessment was made by Ellman method [62]: a buffer solution was added to plasma sample, followed by mixing with 3,5-dinitrobenzoic acid and incubation at room temperature for 15 min. The sample’s absorbance was measured by spectroscopic method at 417 nm.All chemicals were obtained from Sigma-Aldrich Co. (St. Louis, MO, USA). All substances used were of analytical grade.Periodontal measurements and recordings were performed by two previously calibrated, experienced investigators (I.C.M. and D.G.F.), filling for each patient using the Periodontal Chart (see Supplementary Materials). Further, both investigators attended two training meetings, in the presence of a senior periodontist, where they received oral and written instructions on development of the study examination protocol [63] and were given data compilation sheets and their precise role and responsibilities in the study. A full-mouth periodontal examination in a standard environment using standard methodology and equipment was applied to all patients. The oral health screening was conducted in natural light using a dental mirror and a 1 mm marking periodontal probe (UNC-15 periodontal probe, Hu-Friedy, Chicago, IL, USA). Six sites per tooth were evaluated for probing depth (PD), gingival recession (GR), and clinical attachment loss (CAL) [14,64]. PD, GR, and CAL were evaluated according to standard clinical definitions. All probing measurements were rounded down to the nearest millimeter. The full-mouth Gingival Bleeding Index (GBI) was calculated. The GBI was defined as the total number of sites with gingival bleeding on probing divided by the total number of sites per mouth (four sites at each tooth) multiplied by 100 [65]. The number of missing teeth was also recorded.Periodontal status was defined initially as a five-level categorical variable according to the previously published definition [14,66,67]: severe periodontitis, moderate periodontitis, mild periodontitis, gingivitis, or periodontal health. The Center for Disease Control and the American Academy of Periodontology case definition of periodontitis was based on measures of CAL and PD at interproximal sites [14]. Severe periodontitis was defined as having at least 2 sites with CAL ≥6 mm (not on the same tooth) and at least 1 interproximal site with PD ≥5 mm. Moderate periodontitis was defined as 2 or more interproximal sites with CAL ≥4 mm (not on the same tooth) or 2 or more interproximal sites with PD ≥5 mm, also not on the same tooth. Mild periodontitis was defined as at least 2 interproximal sites with CAL ≥3 mm and at least 2 interproximal sites with PD ≥4 mm (not on the same tooth) or 1 site with PD ≥5 mm. Gingivitis patients were cases that did not meet the preceding definitions and had a GBI >10% and probing depths ≤3 mm. The remaining subjects were considered to have healthy periodontal tissue [67].The five-level periodontal categorical variables were regrouped as healthy periodontium or gingivitis, mild periodontitis, and moderate or severe periodontitis for the statistical analysis. The grouping aimed to reflect the intensity of the inflammatory local burden, which increased with the severity of the periodontal disease. Periodontitis was considered when mild, moderate, or severe periodontitis existed.Quantitative data were tested for normality on sub-groups with the Shapiro–Wilk test (for groups lower than 30 patients) or Kolmogorov-Smirnov test (groups larger than 29 patients) according to the sample size of the group. The values are reported as mean (standard deviation) when they followed a normal distribution. Otherwise, the quantitative data were reported as median (Q1–Q3), where Q is the quartile (first and third, respectively). The categorical data are reported as absolute and relative frequencies. The following statistical tests were used to compare groups: Student’s t-test for independent samples (quantitative data normally distributed, two groups), the Mann–Whitney test (nonnormally distributed quantitative data, two groups), Chi-square or Fisher exact test (categorical data), and the Kruskal–Wallis test (nonnormally distributed quantitative data for more than two groups). The Z-test for proportions was also used to compare categorical data between women and men. The Statistica program (v. 8; StatSoft, Tulsa, OK, USA) was used to analyze the data. A p-value < 0.05 was considered statistically significant whenever two groups were compared, and has been adjusted by the number of groups whenever more than two groups were compared (p-value < 0.013 for the Kruskal–Wallis test).Ninety-three patients with ischemic stroke (age 44–80 years) satisfied the inclusion criteria and agreed to participate in the study (Figure 1). The patients’ demographic and clinical characteristics and behaviors are summarized in Table 1. More frequently, the men were current smokers than women (12/57, 21.1% vs. 3/36, 8.3%) and consumed alcohol (17/57, 39.8% vs. 3/36, 8.3%). Most of the patients were overweight (53.8%) or obese (14.0%), but the frequencies were not significantly different between genders (p = 0.185). Abnormal blood glucose levels were observed in 25 patients (26.9%), without a significant difference between the genders (p = 0.111). Sixteen patients (10 women and 6 men) had total cholesterol levels outside the normal ranges (women:men = 27.8%:10.5%; p = 0.032). In contrast, more men had HDL-cholesterol levels outside the normal ranges compared to women (men:women = 21.1%:2.8%; p = 0.013).The blood levels of all investigated oxidative stress and antioxidants biomarkers were outside the normal ranges except total antioxidant capacity. The total antioxidant capacity values were in normal range of 15.1% of the subjects, without a significant difference between genders (men:women = 12.3%:19.4%, p = 0.347). Higher levels of serum NOx were observed in men and higher serum levels of TAC were detected in women (see Table 2) without other significant differences in the oxidative stress and antioxidant biomarkers.More women (11/36) than men (8/57) were edentulous (p = 0.054), but more men (43/49) than women (21/25) had periodontal disease (p = 0.331). No significant association was observed between periodontal disease and gender (p = 0.079, Table 2), but the frequency of men with ischemic stroke and moderate plus severe periodontal disease (56.1%) was significantly higher than the percentage of women with stroke and moderate or severe periodontal disease (30.6%) (Z-test for two proportions; p = 0.018).No significant differences were observed when the investigated oxidative stress and antioxidant biomarkers were compared according to periodontal disease at an adjusted significance level of 0.013 (Table 3). Comparable distributions of smokers (ranging from 2/19 in the edentulous group to 2/21 in the mild periodontitis group) and alcohol consumption (ranging from 1/10 in the healthy periodontium and gingivitis group to 5/21 in the mild periodontitis group) were observed among the periodontal disease groups. Similar distribution among periodontal disease groups were observed regarding the presence of hypertension (from 17/21 in the mild periodontitis group to 18/19 in the edentulous group) and ischemic heart disease (from 3/10 in the healthy periodontium and gingivitis group to 11/21 in the mild periodontitis group). The majority of patients in the healthy periodontium and gingivitis group had diabetes mellitus (8/10), while the presence of this co-morbidity had similar distribution among those with mild periodontitis (6/21) and moderate and severe periodontitis (13/43). In our group of stroke patients, an association tended to be observed between periodontal disease and gender (p < 0.10), with significantly more men with moderate or severe periodontitis than women (p < 0.02). The oxidative stress and antioxidant parameters were similar among the different stages of periodontal disease. NOx as an oxidative stress parameter and TAC as an antioxidant parameter were significantly different between the genders in patients with ischemic stroke, but the small differences could also be by chance and thus need caution in the interpretation and further validation.The sample was comprised of more men than women (Table 1), which is an expected result associated with the characteristics of the primary disease represented by stroke [68]. The age of the evaluated patients was similar between men and women but women had a tendency to have a significantly higher body mass index (BMI) than men (p < 0.10). Overweight was more frequently observed in our sample (53.8%) compared to the values reported previously in stroke cohorts (44.2% in the community-based Framingham Heart Study) [69]. The higher BMI values in women than men (Table 1) could be attributed to difficulties in controlling weight during pregnancy, and hormonal changes during the postmenopausal period [70] as well as an increased frequency of subclinical hypothyroidism [71]. Perissinotto et al. also reported a significantly higher BMI than men, suggesting that visceral redistribution in old age predominantly affects more women than men [72]. A higher BMI in women than men is associated with a significant increase in all lipid metabolic parameters in women than men [73], and the link between these pathogenetic loops is represented by disturbances in thyroid function that can influence lipid metabolism [71,74].Our results show a significant increase in total cholesterol, LDL-CST, and TGs in women than men and also a significant decrease in HDL-CST in men than women (Table 1). It has been suggested that periodontitis is significantly associated with lower HDL-CST levels and respective increases in LDL-CST and triglyceride levels, supporting the hypothesis that periodontitis is associated with disturbances in the control of lipid metabolism [75].In our study, smoking and alcohol consumption were predominant in men, but the results could not be validated statistically due to the small sample. According to previous studies, 18.7% of stroke survivors are active smokers, despite strong recommendations to stop smoking, and 51.7% are men [76]. The assessment of the association between periodontitis and smoking or alcohol consumption was not statistically evaluated in our study because the sub-groups were very small. Current smokers have an increased prevalence, severity, and progression of periodontitis as well as a negative therapeutic response compared to never or former smokers [77]. Quitting smoking limits the destruction of the periodontal tissues and leads to better clinical results [78]. Alcohol consumption is also more frequent in men who are stroke survivors compared to women [79,80,81], as in our study. The risk for periodontitis among men reporting regular alcohol intake previously proved dose-dependency [82]. Alcohol consumption impairs neutrophil, macrophage, and T-cell functions, increasing the likelihood of connective tissue inflammation and stimulating alveolar bone resorption [83]. No significant differences between genders were found regarding the number of patients with hypertension or ischemic heart disease (Table 1). Women from the study group had a higher frequency of diabetes mellitus than men, which was directly related to significantly higher basal glycemia (Table 1). The frequency of diabetes mellitus between women and men varies according to the global region but is estimated to have a complex relationship with differences in ethnicity, migration, culture, lifestyle, gene-environment interactions, socioeconomic status, and social roles [84]. The reported difference could also be explained by random testing, selection bias, and differences in access to healthcare in some countries [85,86,87]. The difference could also be a result of the insulin resistance as an incipient phase of type 2 diabetes mellitus [88,89]. Our study found the highest frequency of diabetes mellitus among patients in the healthy periodontium and gingivitis group that could be explained by the healthcare awareness of these patients. Diabetes mellitus is a known risk factor for the development, progression, and increasing severity of periodontitis [90]. Immune dysfunction in diabetes due to unresolved infection with opportunistic microorganisms is one of the pathways driving periodontitis. Advanced glycated end products are involved in destroying periodontal tissues through enhanced inflammation, impaired wound repair, and increased oxidative stress [91]. The presence of dyslipidemia in diabetic patients is associated with increased destruction of periodontal tissues and gingival inflammation [92] probably due to the stimulation of pro-inflammatory status, which increases the severity of periodontitis [93], or as a source for ROS production in the periodontium through lipid peroxidation [90].Inflammatory markers (CRP) increased in our study in both men and women (Table 1) with a tendency for significantly higher values in women than men (p < 0.10). Elevated levels of CRP are a risk factor for stroke. High CRP values have been reported in patients who suffer an ischemic stroke, and are correlated with prognosis, emphasizing the role of inflammation in stroke pathogenesis [94]. Moreover, CRP, as a biomarker of systemic inflammation, reflects chronic inflammatory status due to periodontitis, and treating a periodontal infection can significantly lower serum CRP levels [95]. These findings support the hypothesis that chronic oral infections and periodontitis-related inflammation are contributors to the systemic inflammatory response [96]. CRP is an acute-phase protein and a biomarker of systemic inflammation produced by the liver in response to various inflammatory stimuli that can trigger its synthesis due to excessive production of pro-inflammatory cytokines, such as interleukin-6 (IL)-6, IL-1β, and tumor necrosis factor-α [97]. The elevated inflammatory factors increase inflammatory activity in atherosclerotic lesions and potentially increase the risk for cardiovascular and cerebrovascular events [98]. Slade et al. reported that patients with extensive periodontal disease (>10% of sites with periodontal pockets >4 mm) have significantly increased CRP levels compared with healthy people [99]. Chronic periodontitis has been associated with elevated CRP levels, even after controlling for several potential confounders, such as age, education level, gender, smoking, HDL-cholesterol, and diabetes [100]. However, in our study CRP values were not associated with periodontitis (Table 3).Oxidative stress is a mechanism involved equally in the pathogenesis of ischemic stroke and inflammatory diseases, such as periodontitis, which alters the balance between free radical production and plasma antioxidants [101]. Oxidative stress molecules are involved in triggering and enhancing inflammatory reactions [102]. Patients with moderate and severe periodontitis have lower antioxidant capacity and higher oxidative stress markers, such as MDA and NOx [103]. The increase of NOx in patients with periodontitis is explained by increased production of NOx due to activation of inducible nitric oxide synthase (iNOS) by proinflammatory cytokines produced by inflammatory cells at the site of gingival inflammation [104,105]. Increased iNOS has been reported in gingival tissue in patients with periodontitis compared with control subjects [106]. The severity of periodontitisis was also related to salivary nitrite concentration [107]. An increase in serum NOx levels was reported in patients with periodontitis by Menaka et al. [108]. In our study, NOx levels increased significantly in men compared with women (Table 2). This difference could be related to significant increases in smoking among men or to increased alcohol consumption, but this needs validation in a larger cohort.All serum antioxidant biomarker levels were reduced in men compared with women; only TAC was significantly lower in men than women (Table 2, p < 0.03). One of the most critical antioxidant molecules that contribute to the antioxidant defense mechanism is represented by catalase. Despite the fact in our study catalase was not significantly statistically different in men compared to women, catalase was previously demonstrated to be lower in gingival tissues in patients with chronic periodontitis, and a greater decrease occurs within tissues adjacent to pockets deeper than 6 mm [109]. Our study also demonstrated that men suffered from a higher frequency of moderate-to-severe periodontitisthan women (p < 0.02). Duarte et al. [110] reported that gene expression of catalase is upregulated in poorly controlled diabetics with chronic periodontitis. The increased catalase activity in patients with diabetes has been attributed to protective and adaptive mechanisms that develop in tissues in response to oxidative stress [111]. In the present study, women had significantly higher basal glycemia compared with men (Table 1), which supports the hypothesis that the increased catalase in women is the result of an adaptive mechanism to the oxidative stress enhanced by hyperglycemia. In our study, more women who suffered a recent ischemic stroke were edentulous than men. Tooth loss in women is associated with a higher risk of all-cause mortality [112]. Tooth loss lacks specificity as a marker of periodontitis because tooth loss occurs not only in the final stage of periodontitis but for many other reasons [83]. However, tooth loss is considered a surrogate indicator of periodontitis, as it occurs in the final stages of the disease and could reflect the persistence of a severe oral inflammation and of a low-grade systemic inflammation. It appears that missing teeth may be a contributing factor determining cardiovascular risk [113]. A significant association has been found between levels of tooth loss and subclinical atherosclerosis, such as the incidence of carotid plaque [114]. Patients with fewer teeth may have an increased risk of stroke [27,35]. A consistently strong and dose-dependent association has been found between the number of missing teeth and incident ischemic stroke and mortality in a large population-based cohort of 4,440,970 relatively young subjects (mean age, 41.5 years) after multivariate adjustment for cardiovascular risk factors [34]. Other studies show no association between missing teeth and stroke [113]. Tooth loss is also related to a lower socioeconomic status in Europe, as poor oral hygiene is a major contributor to periodontal disease and thus a potentially modifiable stroke risk factor. An increase in tooth brushing frequency decreases the concentration of systemic inflammatory markers in serum [115], and dental prophylaxis or periodontitis treatment reduces the incidence of ischemic stroke [42]. Intensive periodontal treatment improves systemic inflammation, high blood pressure, the lipid profile [116], and endothelial dysfunction [117]. An independent role of regular dental care in the prevention of ischemic stroke in an elderly population was reported by Sen et al. [32]. These results indicate that identifying and treating periodontitis at a large scale population level, particularly in high-risk patients, may reduce stroke risk.This study has some potential limitations. The main limitation is related to sample size and the limited number of patients by gender in each periodontitis group. The small sample size of our pilot study could have hidden some significant differences and also limited the statistical analysis as well as the evaluation of oxidative stress and antioxidant parameters between genders according to the class of periodontal disease. Moreover, no evaluation of the former smoking status, smoking history (e.g., ex-smoker, type: cigarettes, cigars, pipe, chewing, snuff) or smoking quantification (e.g., no. of years, how frequent per day, etc.) was done in the evaluation sample. Smoking history of the patients could bring important information due to the cumulative effect of smoking on both atherosclerosis and periodontitis. Furthermore, the applied experimental design does not allow the assessment of the causality between periodontitis and stroke. However, this pilot study constitutes a starting point for more studies on larger samples with further investigations of NOx, total oxidant status, and total antioxidant capacity. The results reported in this study could be used to calculate the required sample size for further studies carried out of the Romanian population after the evaluation of the parameters on matched controls not affected by ischemic stroke and/or atherosclerosis. Further larger studies investigating the relationships between ischemic stroke subtypes and periodontitis severity are needed to identify the influence of periodontal disease in stroke pathophysiology. However, in the context of actual disastrous epidemiological circumstances, it seems improbable to conduct large-scale clinical studies in the next short time period.This is the first study to address Romanian stroke patients and observe the impact of periodontitis and oxidative stress markers on stroke. Specialists and authorities seem to have no concerns regarding the systemic impact of periodontitis. The present data could contribute to elaborating large-scale screening and preventive plans after validation of the results on larger samples.Serum oxidative stress parameters, such as NOx, constitute important biomarkers indicating the persistence of chronic inflammation in patients with stroke and associated periodontitis, which was significantly higher in men than women in the present study, but the difference was small. This result could be related to the higher frequencies of smokers and chronic alcohol drinkers in men than women. The decreased TAC level in men who suffered an ischemic stroke and associated periodontitis compared with women demonstrates that plasma antioxidative capacity is slightly lower in men than women. The oxidative stress/antioxidant status differences between men and women could contribute to increased severity of periodontitis among men, which could lead to a greater impact on stroke morbidity, but the results need validation on larger samples. Men with severe or moderate periodontitis should alert physicians to conduct extensive examinations, particularly when other stroke risk factors are present. Sex disparities related to stroke and periodontal disease could have an important clinical implication that can indicate resources needed for national public health programs, but the results observed in our study need to be validated on a larger sample and in the context of comparison with a matched control group without ischemic stroke.The following are available online at https://www.mdpi.com/2077-0383/9/6/1744/s1, Figure S1: Periodontal Chart.Conceptualization, A.E.B., A.C.B., I.S., S.D.B., and A.R.; Methodology, S.D.B., G.B., A.P.-W., D.G.F., G.D. and A.R.; Investigation, A.C.B., D.G.F., I.S., and I.C.M.; Formal analysis, G.B., S.D.B., and P.M.B.; Visualization, A.E.B., S.D.B., and A.R.; Supervision, A.C.B. and A.R.; Project administration, A.E.B.; Writing—Original draft preparation, A.E.B., B.G., D.G.F., G.D., I.S., I.C.M., and P.M.B.; Writing—Review and editing, A.C.B., S.D.B., A.P.-W., and A.R. All authors have read and agreed to the published version of the manuscript.This research received no external funding. The authors declare no conflict of interest.Flow chart of the patients.Sample baseline characteristics, grouped by gender.* Continuous data are summarized as mean (standard deviation) or median (*** first to the third quartile). ** Categorical data are summarized as the number of cases (percentages). p-values reflect the comparison between men and women: Student’s t-test for independent samples was used when the results are reported as means (standard deviation); the Mann–Whitney test was used for data reported as medians (first quartile to third quartile); or the chi-square test was used for categorical data. BMI = body mass index, LDL-CST = low-density lipoprotein cholesterol, HDL-CST = high-density lipoprotein cholesterol, TG = triglycerides, CRP = C-reactive proteinOxidative stress biomarkers and periodontal disease-related characteristics.Continuous data are summarized as mean (standard deviation) * or median (first to third quartiles) ** Categorical data are summarized as the number of cases (percentages) *** p-values reflect the comparisons between men and women; Student’s t-test was used for independent samples when results are reported as means (standard deviation) (*); the Mann–Whitney test was used for data reported as medians (first to third quartiles) (**); the chi-square or Fisher exact test was used for categorical data (according to the expected values) (***). PD = periodontal probing depth; CAL = clinical attachment loss; MDA = malondialdehyde; NOx = indirectly assessed nitric oxide; TOS = total oxidative stress; TAC = total antioxidant capacity.Differences among the periodontal disease groups.Continuous data are summarized as median (first to third quartiles) * p-values reflect the comparisons between the groups with the Kruskal-Wallis test. MDA = malondialdehyde; NOx = nitric oxide; TOS = total oxidative stress; TAC = total antioxidant capacity.

Med-MDPI/jcm_2/jcm-09-06-01745.txt

@@ -0,0 +1,3 @@
+The aim of this study was to investigate whether early intermittent electroencephalography (EEG) could be used to predict neurological prognosis of patients who underwent extracorporeal cardiopulmonary resuscitation (ECPR). This was a retrospective and observational study of adult patients who were evaluated by EEG scan within 96 h after ECPR. The primary endpoint was neurological status upon discharge from the hospital assessed with a Cerebral Performance Categories (CPC) scale. Among 69 adult cardiac arrest patients who underwent ECPR, 17 (24.6%) patients had favorable neurological outcomes (CPC score of 1 or 2). Malignant EEG patterns were more common in patients with poor neurological outcomes (CPC score of 3, 4 or 5) than in patients with favorable neurological outcomes (73.1% vs. 5.9%, p < 0.001). All patients with highly malignant EEG patterns (43.5%) had poor neurological outcomes. In multivariable analysis, malignant EEG patterns and duration of cardiopulmonary resuscitation were significantly associated with poor neurological outcomes. In this study, malignant EEG patterns within 96 h after cardiac arrest were significantly associated with poor neurological outcomes. Therefore, an early intermittent EEG scan could be helpful for predicting neurological prognosis of post-cardiac arrest patients after ECPR.Neurological prognosis is one of the most important issues in patients who survive a cardiac arrest [1,2]. It is important to estimate the reversibility of cerebral function in patients after return of spontaneous circulation. It may prevent inappropriate continuation of intensive treatment in patients who are predicted to have poor neurological outcomes [2,3]. Recently, extracorporeal membrane oxygenation (ECMO) has been increasingly used as an adjuvant therapy of conventional cardiopulmonary resuscitation (CPR), providing oxygenated blood and hemodynamic support in the absence of spontaneous cardiac circulation [4,5].Autoregulation of cerebral blood flow may be changed in patients resuscitated from cardiac arrest [6]. It is difficult to predict how highly oxygenated continuous flow by ECMO affects cerebral autoregulation and neurological recovery in the setting of extracorporeal cardiopulmonary resuscitation (ECPR) [2]. In previous studies of ECPR, several predictors of mortality have been reported. However, limited data are available on neurological prognosis after ECPR [7].Among electrophysiologic studies, electroencephalography (EEG) has been most widely used as one of the assessment tools for survivors after cardiac arrest [8]. In the setting of ECPR, whether an early intermittent EEG scan may be helpful for systemically estimating neurological outcomes of survivors has not been reported yet. Therefore, the objective of this study was to investigate whether an early intermittent EEG scan could be used to predict neurological outcomes of patients who underwent ECPR.This was a retrospective, single-center, observational study of adult patients who underwent ECPR during hospitalization between February 2012 and December 2018. This study was approved by the Institutional Review Board of Samsung Medical Center (IRB no. SMC 2019-05-002). The requirement for informed consent was waived due to its retrospective nature. Clinical and laboratory data were collected by a trained study coordinator using a standardized case report form. Inclusion criteria were: (1) those who underwent ECPR during the study period; (2) those who had decreased mentalities (a score of <13 on the Glasgow Coma Scale) on EEG scan after cardiac arrest; and (3) those whose EEG scans were performed within 96 h after ECPR. Exclusion criteria were: (1) those who were under 18 years of age; (2) those with malignancy whose life expectancy was less than 1 year; (3) those with insufficient medical records; (4) those with causes of death verified to be other than brain death; and (5) those with a history of head trauma or a chronic neurological abnormality upon admission to the intensive care unit (ICU). Ultimately, a total of 69 patients with an EEG scan who were resuscitated by veno-arterial ECMO were analyzed in this study (Figure 1).In this study, ECPR was defined as a successful veno-arterial ECMO implantation and pump-on with chest compression for external cardiac massage during index procedure in patients with cardiac arrest. When a return of spontaneous circulation occurs during ECMO cannulation, practitioners typically do not remove the cannula or stop the ECMO pump-on process [7]. Surface cooling and the degree of targeted temperature were determined by each intensivist in the ICU according to the targeted temperature management protocol [9]. The primary endpoint was the neurological status on discharge from the hospital. It was assessed with the Glasgow–Pittsburgh Cerebral Performance Categories (CPC) scale (scores range from 1 to 5) [10]. CPC scores of 1 and 2 were classified as favorable neurological outcomes while CPC scores of 3, 4 and 5 were considered as poor neurological outcomes [11,12]. Medical records were thoroughly reviewed. Patients were graded on the CPC score by two independent neurologists. If the CPC scores did not match between the two neurologists, they discussed it and reached an agreement. A neurointensivist, attending physician or consultant neurologist determined EEG scan. An EEG scan was performed to identify causes of decreased consciousness or to predict neurological outcomes in ECPR patients. However, an EEG scan was not performed for patients who had a rapid recovery of mentality and neurological deficits. An EEG scan was also performed when patients had accompanied seizures or abnormal movements. If sedatives, analgesics or antiepileptic drugs were administrated to patients after ECPR, these drugs were not stopped during the EEG scan. EEG was performed using a 64-channel digital video-EEG system (Nicolet Biomedical, Inc., Madison, WI, USA). Surface electrodes were placed according to the international 10–20 system. Additional electrodes were placed whenever needed [13]. EEG patterns of ECPR patients were defined using the EEG terminology of the American Clinical Neurophysiology Society [14,15]. Malignant EEG patterns were defined as highly malignant EEG patterns and moderate malignant EEG patterns. Highly malignant EEG patterns were defined as suppressed background (amplitude <10 μV, 100% of the recording) without discharges, suppressed background with superimposed continuous periodic discharges, or burst-suppression (periods of suppression with amplitude <10 μV constituting >50% of the recording) with superimposed discharges or without discharges [15] (Figure 2).Moderate malignant EEG patterns were defined as malignant periodic or rhythmic patterns (abundant periodic discharges; abundant rhythmic spikes, polyspikes, sharp waves, spike-and-wave or sharp-and-slow wave; unequivocal electrographic seizure), malignant background (discontinuous background; low voltage background; reversed anterior-posterior gradient) or unreactive EEG (absence of background reactivity or only stimulus-induced discharges) [15]. Benign EEG patterns were defined as an absence of all malignant features stated above. EEG findings were confirmed by three EEG specialists.CPR was led by the CPR team of the hospital. All facts related to the CPR scene were recorded by bedside nurses according to Utstein-Style guidelines [16]. When CPR was performed for more than 10 min or in the event of unstable vital signs or recurrent cardiac arrest, the institutional rapid response team contacted the on-call ECMO team leader, who along with the CPR leader assessed the patient and made a decision about whether to institute ECPR. ECPR was performed when a witnessed arrest was confirmed, when the arrest persisted despite conventional CPR lasting for more than 10 min, and when the event that caused the arrest was considered reversible [4]. Cases in which ECPR was deferred included those with a short life expectancy (<6 months), terminal malignancy, an unwitnessed collapse, limited physical activity or CPR undertaken for more than 60 min at the time of initial contact. Age alone did not constitute a contraindication to ECPR [4].The ECMO team consisted of cardiologists, cardiovascular surgeons, intensivists, special nurses and perfusionists. Either a Capiox Emergency Bypass System (Terumo, Tokyo, Japan) or a Prolonged Life Support System (Maquet Cardiopulmonary, Hirrlingen, Germany) was used in all cases. A crystalloid solution such as normal saline or balanced solution was used for priming. No patient had blood-primed ECMO. A percutaneous vascular approach was tried initially in all cases using the Seldinger technique. When percutaneous cannulation failed, surgical cutdown exposure was performed [4]. Femoral vessels were the most common sites of vascular access using 14 to 17 French arterial cannulas and 20 to 24 French venous cannulas [7]. Cardiac compression was stopped once ECMO pump-on was successful during CPR. Anticoagulation was accomplished by a bolus injection of unfractionated heparin, followed by continuous intravenous heparin infusion to maintain an activated clotting time between 150 and 180 s. The initial number of revolutions per minute of the ECMO device was adjusted to achieve an ideal cardiac index greater than 2.2 L/min/m2 of body surface area, central mixed venous oxygen saturation above 70%, and a mean arterial pressure above 65 mm Hg [7]. Blood pressure was monitored continuously through an arterial catheter. An artery in the right arm was used for arterial blood gas analysis to estimate cerebral oxygenation. After ECMO, necessary steps were taken to treat the cause of the arrest, such as percutaneous coronary intervention, coronary artery bypass grafting, heart transplantation, non-coronary cardiopulmonary surgery or non-cardiopulmonary surgery [7].All data are presented as medians and interquartile ranges (IQRs, Q1~Q3) for continuous variables and as numbers (percentages) for categorical variables. Data were compared using the Mann–Whitney U test for continuous variables and the Chi-squared test or Fisher’s exact test for categorical variables. Variables with p values less than 0.2 in univariate analyses and clinically relevant variables were subjected to a stepwise multiple logistic regression model to obtain statistically meaningful predictor variables. They were EEG groupings by its pattern, age, target temperature management, first monitored rhythm, CPR duration, Glasgow Coma Scale on EEG scan and use of sedative or analgesic. Due to small event rates, we took the caution of the general rule of 10 events per variable before any routine application of statistical methods and applied the Firth’s correction. Adequacy of the prediction model was determined using the Hosmer–Lemeshow test, along with the areas under the curve (AUCs). The predictive performance of malignant EEG patterns assessed using the AUCs of the receiver operating characteristic (ROC) curves for sensitivity vs. 1-specificity. The AUCs were compared using the nonparametric approach published by DeLong et al. [17] for two correlated AUCs. All tests were two-sided and p < 0.05 was considered statistically significant. All data were analyzed using IBM SPSS version 20 (IBM, Armonk, NY, USA) and R Statistical Software (version 3.6.3; R Foundation for Statistical Computing, Vienna, Austria).The median patient age was 56 (IQR: 47–70) years. Of 69 patients included in this study, 52 (75.4%) were males. Hypertension (42.0%) and diabetes mellitus (33.3%) were the most common comorbidities among patients who underwent ECPR. Hypertension was more common in patients with poor neurological outcomes than in patients with favorable neurological outcomes (50.0% vs. 17.6%, p = 0.005). A cardiac cause of arrest was verified in 59 (85.5%) patients. Acute coronary syndrome was the main cause of cardiac arrest in 26 (44.1%) patients. Fourteen (20.3%) patients had a history of ischemic heart disease. Forty-seven (68.1%) patients experienced cardiac arrest in the hospital while 22 (31.9%) patients suffered cardiac arrest in an out-of-hospital setting. Compared with the group with favorable neurological outcomes, the group with poor neurological outcomes had a longer CPR duration (p = 0.005). Baseline characteristics of ECPR patients are presented in Table 1.Among the 69 adult cardiac arrest patients who underwent ECPR, 32 (46.4%) survived until discharge from the hospital. Of these 32 survivors, 17 (24.6%) had favorable neurological outcomes (CPC score of 1 or 2). The entire distribution of CPC scores is shown in Figure 1.Sedatives or analgesics were used in 41 (59.4%) patients who underwent ECPR. These drugs were used more in patients with favorable neurological outcomes than in patients with poor neurological outcomes (88.2% vs. 50.0%, p = 0.012). There was no significant difference in the use of antiepileptic drugs between the two groups of patients (p = 0.999). Characteristics on the EEG scan are presented in Table 2.Malignant EEG patterns were more common in patients with poor neurological outcomes than in patients with favorable neurological outcomes (73.1% vs. 5.9%, p < 0.001, Table 3). All patients with highly malignant EEG patterns (43.5%) had poor neurological outcomes. Moderately malignant EEG patterns were reported in eight (11.6%) patients with poor neurological outcomes and in only one (1.4%) patient with a favorable neurological outcome. Regardless of the interval between ECPR and EEG scans, most patients with malignant EEG patterns had poor neurological outcomes in this study. In addition, all patients with myoclonic status epilepticus had poor neurological outcomes. Benign EEG patterns were more common in patients with favorable neurological outcomes than in patients with poor neurological outcomes (94.1% vs. 26.9%, p < 0.001, Table 3).In multivariable analysis, the only significant indicators were EEG grouping by its pattern and CPR duration. That is, malignant EEG patterns (adjusted odd ratio (OR): 36.43, 95% confidence interval (CI): 4.632–1013.711) and CPR duration (adjusted OR: 1.06 per minute increase, 95% CI: 1.00–1.138) were significantly associated with poor neurological outcomes in patients who underwent ECPR (Hosmer–Lemeshow Chi-squared = 2.32, df = 8, p = 0.969) with the AUCs of 0.946 (95% CI 0.893–0.999) (Table 4).Although there were no differences between the AUCs of malignant EEG patterns and CPR duration, the performance of a composite of these marker was strongly associated with poor neurological outcomes compared with the use of either marker alone (p = 0.008 and p = 0.006, respectively) (Figure 3).In this study, we investigated whether intermittent EEG could be used to predict neurological outcomes of patients who underwent ECPR. The major findings of this study were as follows. First, regardless of sedation, malignant EEG patterns were more common in patients with poor neurological outcomes than in patients with favorable neurological outcomes. Especially, all patients with highly malignant EEG patterns had poor neurological outcomes. In addition, patients with moderate malignant EEG patterns had poor neurological outcome except for one patient. Second, benign EEG patterns alone did not necessarily imply a favorable neurological outcome. Third, in multivariable analysis, malignant EEG patterns and CPR duration were significantly associated with poor neurological outcomes in patients who underwent ECPR. Therefore, early intermittent EEG scans and CPR duration could be helpful for predicting neurological outcomes of post-cardiac arrest patients after ECPR.EEG signals mainly reflect cerebral cortical function and some subcortical function [8]. EEG is very sensitive to ischemia because cortical neurons of the brain need a consistent blood supply to maintain signaling and integrity [8]. Therefore, the EEG scan is a standard and useful tool to predict neurological outcomes after cardiac arrest [3,15]. Especially, malignant EEG patterns such as suppressed background, status epilepticus, burst suppression, periodic patterns and unreactive EEG are associated with poor neurological prognosis after cardiac arrest [8,15,18]. In addition, early continuous waves with normal voltage could be a predictor of favorable neurological outcomes after cardiac arrest [19].Cerebral autoregulation may be changed in survivors after cardiac arrest [6]. Highly oxygenated continuous ECMO flow could affect cerebral autoregulation after ECPR [2]. In addition, neurological outcomes may be affected by functional recovery of native heart and lung, the amount of ECMO support and changed cerebral autoregulation [2]. Altered cerebral hemodynamics by ECMO support may influence neurological outcomes after ECPR. Therefore, it is difficult to predict neurological prognosis by these changed situations after ECPR [2]. Ultimately, the interaction between cerebral autoregulation and ECMO flow may affect neurological recovery and prognosis in ECPR patients through mechanisms of primary ischemic damage and secondary additive injury [2]. Thus, EEG change by this interaction should be studied for neurological prediction after ECPR. However, there has been no report of EEG according to neurological outcomes after ECPR.Sedation may confuse outcome predictions in survivors of cardiac arrests [1,8,20]. Sedatives are commonly used in survivors after cardiac arrest for 72 h as important confounders [1,20]. A motor response to noxious stimuli, corneal reflex, caloric testing and some electrophysiologic studies may also be confounded by sedation [20,21]. Although mild to moderate hypothermia does not significantly affect EEG in patients with induced hypothermia [8,22], a confounder accompanied by induced hypothermia such as analgesics, sedatives or artifacts from shivering, mechanical ventilator or electrical devices may affect the reliability of EEG interpretations [8]. However, a recent study has reported that the predictive performance of EEG after cardiac arrest is similar between patients with ongoing sedation and those without ongoing sedation [8,15]. In this study, sedation or targeted temperature management did not significantly affect the prediction of poor neurological outcomes after ECPR. Regardless of sedation or targeted temperature management, patients with malignant EEG patterns had poor neurological outcomes in this study.Benign EEG patterns may be associated with a favorable neurological outcome in survivors after cardiac arrest [19]. Especially, early continuous waves with normal voltage could be a predictor of favorable neurological outcomes after cardiac arrest [8,15,19]. However, in previous studies, benign EEG patterns are not always associated with a good neurological outcome [15,18]. Additive secondary injury is characterized by an imbalance in post-resuscitation cerebral oxygen delivery and use [23]. This injury is associated with reperfusion injury, impaired autoregulation, fluctuations in oxygen support and arterial carbon dioxide, hyperthermia and concomitant anemia [23]. Early EEG findings may not be shown to be malignant EEG patterns in patients with poor neurological outcomes if the secondary cerebral injury is more serious than the primary cerebral injury. In this study, benign EEG patterns were not always associated with a favorable neurological outcome. In addition, intermittent EEG scans may be less sensitive for predicting favorable neurological outcomes than continuous EEG monitoring in this study.It is not easy to predict neurological prognosis of survivors after cardiac arrest [24,25]. In addition, there is no optimal timing to assess prognosis after cardiac arrest [25]. Although highly malignant EEG patterns may be associated with poor prognosis, other EEG patterns may be ambiguous in predicting outcomes in various conditions after cardiac arrest [15,26]. Therefore, it may be unreasonable to predict the prognosis of survivors after ECPR based on EEG patterns alone. Early EEG scans may help to predict neurological outcomes and reduce uncertainty over coma prognostication after cardiac arrest [26]. Therefore, EEG should be used as one of the multimodal tools rather than an absolute tool for predicting prognosis. Eventually, a multimodal approach, including neurological examination, biomarker, brain imaging, evoked potential and EEG may be needed to predict neurological outcomes in patients after ECPR [24,25].This study has several limitations. First, this was a retrospective review. Thus, CPC score was determined based on medical records. By using two independent specialists’ agreement on the score, any bias may be mitigated to some extent. In addition, although the cause of death had to be accurately verified, its identification was insufficient due to the retrospective nature of this study. Second, the nonrandomized nature of registry data might have resulted in selection bias. EEG scans were not protocol-based in their performance. Particularly, during the study period, EEG scans were not performed in all patients. They were only performed in patients with abnormal consciousness, seizure, abnormal movements or other symptoms. Although EEG scans were performed within 96 h following ECPR, a major limitation of this study might be that EEG scans were performed in different time settings. Lastly, our study has limited statistical power due to its small sample size. To alleviate the small sample issue, we applied the Firth’s correction, which dampens the results somewhat. Although it still provides a valuable insight, prospective large-scale studies are needed to confirm the usefulness of early intermittent EEG scan in predicting neurological outcomes of patients after ECPR to arrive at evidence-based conclusions.In this study, malignant EEG patterns within 96 h after cardiac arrest were significantly associated with poor neurological outcomes in patients who underwent ECPR. A multimodal approach is needed to predict neurological outcomes in patients after ECPR. Especially, an early EEG scan may help to predict neurological outcomes and reduce uncertainty about coma prognostication after ECPR.Conceptualization, Y.O.K. and J.A.R.; methodology, Y.O.K. and J.A.R.; resources, Y.O.K., R.-E.K. and J.A.R.; supervision, C.R.C., J.H.Y., T.K.P., Y.H.C., K.S., G.Y.S. and J.-A.R.; writing—original draft preparation, Y.O.K. and J.A.R. All authors have read and agreed to the published version of the manuscript.This research received no external funding.We would like to thank the excellent statistical support of Keumhee C. Carriere, and Joonghyun Ahn, of the Samsung Biomedical Research Institute. We also would like to thank the nursing director of the neurosurgical intensive care unit, Hye Jung Kim, who provided excellent advice and fruitful discussions. We also thank all nurses of the neurosurgical intensive care unit at Samsung Medical Center.The authors declare no competing interests.Regarding data availability, our data are available on the Harvard Dataverse Network (http://dx.doi.org/10.7910/DVN/KYJNVA).Not applicable. This study does not contain individual or personal data in any form (including individual details, images, or videos).Study flow chart. ECPR, extracorporeal cardiopulmonary resuscitation; EEG, electroencephalography; GCS, Glasgow Coma Scale; CPC, Cerebral Performance Categories.Highly malignant patterns of electroencephalography in patients after extracorporeal cardiopulmonary resuscitation. (A) Suppressed background without discharges; (B) Suppressed background with superimposed continuous periodic discharges; (C) burst-suppression without discharges; and (D) burst-suppression with superimposed discharges.Receiver operating characteristic curves for predicting poor outcomes using malignant patterns of electroencephalography (EEG) and cardiopulmonary resuscitation (CPR) duration. Although there were no differences between the areas under the curve (AUCs) of malignant EEG patterns and CPR duration, the performance of a composite of these marker was strongly associated with poor neurological outcomes compared with the use of either marker alone (p = 0.008 and p = 0.006, respectively). CI, confidence interval; Malig EEG, malignant EEG patterns.Baseline characteristics.IQR: interquartile range; CPR: cardiopulmonary resuscitation; VT: ventricular tachycardia; VF: ventricular fibrillation; ECMO: extracorporeal membrane oxygenation.Characteristics on electroencephalography scan.ECPR: extracorporeal cardiopulmonary resuscitation; EEG: electroencephalography.Findings of electroencephalography.
+              ECPR: extracorporeal cardiopulmonary resuscitation; EEG: electroencephalography.
+            Multivariable logistic regression of clinically relevant variables associated with poor neurological outcomes.OR: odd ratio; CI: confidence interval; EEG: electroencephalography; VT: ventricular tachycardia; VF: ventricular fibrillation; CPR: cardiopulmonary resuscitation. Variables with p values less than 0.2 in univariate analyses and clinically relevant variables were subjected to a stepwise multiple logistic regression model to obtain statistically meaningful predictor variables. They were EEG groupings by its pattern, age, target temperature management, first monitored rhythm, CPR duration, Glasgow Coma Scale on EEG scan, and use of sedative or analgesic. To alleviate the small sample issue, we applied the Firth’s correction.

Med-MDPI/jcm_2/jcm-09-06-01746.txt

@@ -0,0 +1 @@
+These authors contributed equally to this work.Background: Early brain injuries (EBI) are one of the most important causes of morbidity and mortality after subarachnoid hemorrhage. At admission, a third of patients are unconscious (spontaneously or sedated) and EBI consequences are not evaluable. To date, it is unclear who will still be comatose (with severe EBI) and who will recover (with less severe EBI) once the aneurysm is treated and sedation withdrawn. The objective of the present study was to determine the diagnostic accuracy of S100B levels at hospital admission to identify patients with severe neurological consequences of EBI. Methods: Patients were consecutively included in this prospective blinded observational study. A motor component of the Glasgow coma score under 6 on day 3 was used to define patients with severe neurological consequences of EBI. Results: A total of 81 patients were included: 25 patients were unconscious at admission, 68 were treated by coiling. On day 3, 12 patients had severe consequences of EBI. A maximal S100B value between admission and day 1 had an area under the receiver operating characteristic curve (AUC) of 86.7% to predict severe EBI consequences. In patients with impaired consciousness at admission, the AUC was 88.2%. Conclusion: Early S100B seems to have a good diagnostic value to predict severe EBI. Before claiming the usefulness of S100B as a surrogate marker of EBI severity to start earlier multimodal monitoring, these results must be confirmed in an independent validation cohort.The 28-day mortality associated with subarachnoid hemorrhage (SAH) is reported to range from 26% to 40% in Europe [1] and half of those who survive sustain irreversible brain damage [2]. Bleeding consequences, such as cerebral edema, intracerebral hemorrhage, mechanical stress, or raised intracranial pressure, as well as re-bleeding or aneurysm treatment complications, are important determinants of poor outcome [3,4,5,6,7]. These first brain insults have been grouped together and termed early brain injuries (EBI). Physicians usually rate EBI at two critical moments: at the first physician contact using the World Federation of Neurosurgical Surgeons (WFNS) grading scale, which is a good predictor of clinical outcome [8]; and 3 days later, once the aneurysm is secured, potential hydrocephalus is treated, anesthesia is withdrawn, and before the period during which delayed cerebral ischemia (DCI) can occur, as the latter impacts prognosis [9,10]. At day 3, patients without any loss of consciousness or neurological deficit, supposed to have no or mild EBI, can be differentiated from unconscious patients suffering severe EBI [9,10]. Since severe EBI increases the risk of DCI, it is important to identify early on those who require multimodal monitoring to detect delayed injury [11,12,13]. Yet, the consequences of EBI may not be assessable by clinical examination, as many patients require sedation before or at admission to the intensive care unit (ICU; e.g., due to respiratory failure, early seizure or loss of consciousness).Several early markers have been used to rate EBI after SAH. For instance, the presence of a cerebral edema on computed tomography (CT) images, graded using the subarachnoid hemorrhage early brain edema score (SEBES), or the extent of fluid attenuation inversion recovery (FLAIR) hypersignals on magnetic resonance imaging, are predictors of long-term disability and DCI occurrence [6,7,14]. Systemic biomarkers have also been evaluated to rate the burden of brain injury, such as the neuron specific enolase (NSE) following cardiac arrest [15,16]. However, in hemorrhagic stroke, astrocytic markers are better biomarkers of the primary hemorrhagic lesion [17,18,19]. The most studied is S100B, a calcium-binding protein expressed predominantly in astroglial cells. A high serum S100B level during the first days after SAH is also predicative of a poor long-term outcome, and is increased by cerebral infarction [18,20,21]. However, the long-term outcome is the result of both early and delayed injury, and it is unclear whether S100B can be used to rate EBI severity. In the absence of an early biomarker to identify patients who will remain unconscious (i.e., with severe EBI) 3 days after the bleeding, we hypothesized that early S100B serum concentration reflects the extent of EBI caused by the primary hemorrhage.The primary aim of this study was therefore to evaluate the diagnostic accuracy of S100B to predict the severe consequences of the primary hemorrhage. We chose a pragmatic clinical approach and defined nonconscious patients at day 3 as having severe consequences of EBI. We also explored early S100B diagnostic accuracy to predict the outcome at ICU discharge.This prospective blinded single-center observational study was conducted in the neurological department of anesthesiology and intensive care medicine of the Pierre Wertheimer Hospital (Hospices Civils de Lyon, Lyon, France). S100B measurements were not available to the clinical staff and therefore did not influence therapeutic choice or any outcome assessment. Written and oral information was given to the patients or their relatives. This study is part of the single-center Prospective, Observational Registry of Patient with Subarachnoid Hemorrhage in Neurocritical Care Unit (ProReSHA) which was approved by our local ethics committee (Comité de Protection des Personnes Lyon Sud Est II, n°IRB: 00009118), and registered on clinicaltrials.org (NCT02890004).The ProReSHA started in September 2016. Since S100B measurements were available in our hospital on 28 November 2016, the patients of the study reported herein were consecutively included from this date to 23 November 2017. Inclusion criteria were admission for a recent clinical history of SAH with evidence of bleeding on CT and age ≥ 18 years. Exclusion criteria were traumatic SAH, the absence of aneurysm on first angiography, a delay > 36 h between the ictus and ICU admission, and ongoing sedation at day 3.The definition of EBI can vary depending on the method of evaluation (i.e., clinical evaluation, brain imaging, invasive neuro-monitoring, neuroinflammation, electrophysiology) [10]. We chose a pragmatic approach usable at the bedside and tried to differentiate patients without or with mild EBI (e.g., without loss of consciousness) from those with moderate EBI (i.e., with an initial loss of consciousness that will rapidly recover) and those with high EBI who will remain unconscious for several days. Although the WFNS score is used to assess patient severity at first physician contact, it was not designed to monitor clinical changes during the hospital stay. Instead, the Glasgow coma scale (GCS) is used to monitor the neurological state and look for a new neurological deficit after SAH [22]. Therefore, its motor component (M-GCS) on day 3 was chosen as the reference for early outcome, as M-GCS is both widely used and reproducible at the bedside and because, at day 3, patients are expected to have received aneurysm treatment, with or without hydrocephalus treatment, and anesthesia should have been withdrawn. A score < 6 on the M-GCS indicates an unconscious patient who does not respond to an oral command. Based on the M-GCS at ICU admission and day 3, we defined three groups: the EBI-mild group for whom the M-GCS was 6 at arrival and day 3, the EBI-moderate group for patients with an M-GCS < 6 at arrival and 6 on day 3, and the EBI-severe group presenting an M-GCS score < 6 at day 3 regardless of the M-GCS on arrival.The physicians and nurses in charge of patients and the clinical data assessors were blinded to the S100B levels. SAH severity was evaluated using the WFNS at first contact with a physician (emergency room or out of hospital) and the GCS at admission to the ICU [8]. The Fisher grade [23,24], the SEBES  [14], and the Hijdra score [25] were also collected from the first CT. On day 3, the best motor, eye, and verbal GCS sub-scores were collected from the daily evaluation. Other clinical and laboratory parameters, as well as medications, were available hourly from the IntelliSpace Critical Care and Anesthesia software (Philips Informatique Médicale, Suresnes, France), as well as the presence of sedation and hemodynamic or ventilation support. The assessor of the modified Rankin Scale (mRS) at ICU discharge was also blinded to the S100B levels.Patients were managed per international SAH treatment guidelines [2]. Briefly, they received daily oral nimodipine (60 mg every 4 h). Coiling was usually preferred to clipping, except in anatomical conformations that prevented coiling from being used or when intraparenchymal hematoma required surgical evacuation [26]. In the case of hydrocephalus observed on CT, an external ventricular drainage (EVD, Integra Life Science, Saint Priest, France) was inserted. Blood glucose was kept between 8 and 10 mmol/L, systolic arterial pressure between 120 and 160 mmHg before aneurysm repair, SpO2 above 96%, and in the case of mechanical ventilation, carbon dioxide partial pressure was kept between 35 and 40 mmHg. An arterial and a central venous catheter were used if patients needed noradrenaline or mechanical ventilation (with or without sedation). Blood samples were collected at admission and on day 1. They were centrifuged in the biology laboratory and frozen at −20 °C for further analyses. S100B serum levels were measured with an electroluminescent immunoassay kit using a sandwich technique (Cobas, Roche, Mannheim, Germany). This allows the measurement of S100B concentrations between 0.005 and 39 µg/L; the limit of quantification is 0.02 µg/L. The within-assay coefficient of variation (CV) was 2.1% at 0.059 µg/L and 0.9% at 0.372 µg/L mean concentrations. The between-assay CV was 6.2% at 0.200 µg/L and 3.2% at 2.430 µg/L mean concentrations. This study follows the standards for reporting diagnostic accuracy studies (STARD) 2015 recommendations for analyses of biomarker diagnostic accuracy [27]. The necessary number of subjects was calculated using the Obuchowski method with the assumption of a ratio of one case (patients with M-GCS < 6 at day 3) to seven controls (patients with M-GCS = 6 at day 3) [28]. A minimum total population of 64 patients, including eight cases (patients with M-GCS < 6 at day 3) was necessary to detect an area under the receiver operating characteristic (ROC) curve (AUC) > 0.8, with a power of 0.8 and an alpha risk of 0.05. Data were expressed as their median and interquartile range (IQR). Comparisons between the groups were carried out using the Kruskal–Wallis test and post hoc analyses using the Bonferroni–Dunn test. Comparisons of proportions were made using Fisher’s exact test, with Bonferroni correction for multiple comparisons.ROC efficiency functions were calculated using the pROC library [29]. The reference standard (M-GCS at day 3) was chosen before the statistical analyses. The best threshold values were calculated according to the point closest to the ROC curve’s top-left point (i.e., optimization of: min ((1 - sensitivity)² + (1-specificity)²). The 95% confidence intervals (95% CI) were calculated using a bootstrapping method. The AUCs were compared using the Delong test. The effect size of the EBI-severe group on S100B levels was evaluated by the Cliff’s delta statistic because of a non-normal distribution (approaching 0 meaning a small effect size and 1 meaning a large effect size) [30]. The AUCs of a combination of EBI markers was evaluated with a logistic regression model, including the following predictors: S100B, SEBES, Hijdra, the presence of an intracerebral hematoma, and GCS at ICU arrival. The dependent variable was the presence or absence of severe EBI at day 3 (i.e., M-GCS < 6).Biomarkers help to stratify the risk of a condition and guide clinical decision making. Therefore, rather than a single cut-off that dichotomizes the population, another approach is to assess a gray zone with diagnostic uncertainty, allowing for a certain continuum in risk stratification. The first cut-off is chosen to include the diagnosis with near-certainty (i.e., privileging sensitivity); the second is chosen to exclude the diagnosis with near-certainty (i.e., privileging specificity). When the values of the biomarker fall into the gray zone between the two cut-offs, uncertainty exists, and the physician should pursue a diagnosis using additional tools. Therefore, the zone of diagnostic uncertainty, or gray zone, was defined from the first values with a sensitivity or specificity of at least 90% [31].A value of p < 0.05 was considered significant. Statistical analyses were performed using the R software (R Foundation for Statistical Computing, Vienna, Austria, version 3.3.1) [32].A total of 132 patients were potentially eligible, and 81 were included in the analyses (Figure 1). For 84% of them, aneurysm was treated by coiling within a median (IQR) of 14 h [5,6,7,8,9,10,11,12,13,14,15,16,17,18,19,20,21]. At ICU admission, 56 patients had an M-GCS of 6, among whom 53 had an M-GCS of 6 both at admission and on day 3; three patients worsened within 3 days (two had hydrocephalus requiring EVD and the third was intubated for respiratory disorder; none had re-bleeding or aneurysm treatment complications). According to the M-GCS at admission and day 3, 53 patients were classified as EBI-mild (M-GCS = 6 at admission and day 3), 16 as EBI-moderate (M-GCS < 6 at admission and M-GCS = 6 at day 3), and 12 as EBI-severe (M-GCS < 6 at day 3 regardless of the M-GCS on admission). In the EBI-severe group, the median M-GCS at day 3 was 4/6 [4,5] (i.e., M-GCS of 5: n = 4/12; 4: n = 4/12; 2: n = 1/12; 1: n = 3/12). At day 3, no patients were sedated or had seizures, and all had been treated for hydrocephalus.The three groups were significantly different in terms of the simplified gravity index (SAPSS), anti-coagulant treatment, WFNS, Fisher grade, Hijdra score, and length of stay, as well as the frequency of hydrocephalus, re-bleeding, and sedation need. Re-bleeding was a rare event (n = 4) and occurred within 24 h after admission to the ICU. Patients in the EBI-severe and EBI-moderate groups had a worse outcome at ICU discharge compared to the EBI-mild group (Table 1).We were able to assess the S100B time course in 61 patients with both measurements at admission and on day 1 (EBI-severe: n = 11, EBI-moderate: n = 13, EBI-mild: n = 37). There were no significant differences between admission and day 1 (median (IQR), respectively, 0.098 µg/L (0.067–0.185) vs. 0.100 µg/L (0.069–0.172), p = 0.501). Moreover, 46 patients had daily measurements from admission to day 3. The median (IQR) S100B value declined significantly on days 2 (p = 0.003) and 3 (p < 0.001) compared to admission (admission: 0.094 µg/L (0.061–0.176); day 1: 0.092 µg/L (0.068–0.169); day 2: 0.085 µg/L (0.057–0.112); day 3: 0.064 µg/L (0.048–0.09). We included 81 patients with a ≥ 1 S100B measurement at admission or day 1 and used their maximal value to evaluate the S100B diagnostic accuracy to predict a M-GCS < 6 at day 3. The median S100B value was higher in the EBI-severe group (0.467 µg/L, IQR (0.171–1.09)) than in the EBI-moderate (0.134 µg/L, IQR (0.092–0.204); p = 0.022) or the EBI-mild group (0.098 µg/L, IQR (0.068–0.138); p < 0.001). There was a large effect size of the EBI-severe group vs. the others on the maximal S100B value at admission or day 1, as evidenced by a Cliff’s delta of 0.73 (95% CI (0.46;0.88)). The S100B AUC was not significantly different from that of the GCS at ICU admission (86.7% (95% CI (73.6;95.9) vs. 84.1%, 95% CI (73.9–92.6), p = 0.573), and tended to be higher than the AUC of the Hijdra score (68.03% 95% CI (51.4;82.2), p = 0.057), as well as the SEBES (68% 95% CI (52.1;81.7), p = 0.054). The AUC of a logistic regression model, including S100B, the presence of an intraparenchymal hematoma, the SEBES, the Hijdra score, and the GCS at arrival, was not significantly greater than the AUC of S100B alone (p = 0.946).There were 12 patients with severe EBI and 69 controls (EBI-moderate + EBI-mild), allowing us to detect an AUC > 63% with a power of 95% and an alpha risk of 5%. This value is lower than the lower limit of the 95% CI of the S100B AUC.Among unconscious patients at admission (i.e., M-GCS < 6, n = 25), the effect size of the EBI-severe group on S100B was also large; Cliff’s delta of 0.76 (95%CI (0.34;0.93)). The AUC of S100B (88.2% 95% CI (72.2;100)) was significantly greater than that of the GCS at admission (62.3% (53.1;72.0), p = 0.003) and of the Hijdra score (53.8% 95%CI (29.9;77.8), p = 0.012), and tended to be better than the AUC of the SEBES (70.1% 95% CI (50;88.5), p = 0.151). The AUC of a logistic regression model including S100B, the presence of an intraparenchymal hematoma, the SEBES, the Hijdra score, and the GCS at arrival was not significantly greater than the AUC of S100B alone (p = 0.77).The best S100B thresholds and gray zone limits are described in Table 2 and highlighted on the ROC curves in Figure 2.The median maximal serum S100B value at admission or day 1 was significantly different between the seven levels of the mRS (p < 0.001). Post hoc analyses found that median S100B values were significantly higher in patients with severe disability (mRS = 5, 0.302 µg/L, IQR (0.216–0.531), p = 0.003) or dead (mRS = 6, 1.49 µg/L, IQR (0.957–3.1), p = 0.003) compared to those without disability (i.e., mRS < 2, 0.093 µg/L (0.062–0.124)). The maximal serum S100B value at admission or day 1 had an AUC of 95.7% (95% CI (90.52;100)) to predict an mRS > 4 (severe disability or death). A value > 0.262 µg/L (95% CI (0.164;0.494)) predicted an mRS > 4 at ICU discharge with a 94.3% (95% CI (88.6;98.61)) specificity and 81.8% (95% CI (54.6;100)) sensitivity. The gray zone of diagnosis ranged from 0.199 µg/L to 0.262 µg/L. The maximal value of serum S100B was in this gray zone for 12.3% of patients (n = 10/81).Among conscious patients at day 3 (EBI-mild and EBI-moderate groups, n = 69), seven had an early S100B > 0.256 µg/L (upper gray zone threshold supposed to predict an M-GCS < 6 with almost certainty), and were considered false positives. They were all Fisher 4 with a SEBES > 2, and three had an intracerebral hematoma. They had a poor outcome (median (IQR) mRS = 4 (3–4.5)) with frequent DCI occurrence (n = 5/7, 71.4%).In this prospective blinded study, the maximal S100B serum concentration at admission and day 1 was found to have a good diagnostic value to predict severe consequences of EBI leading to an unconscious state at day 3 [31]. The early S100B had a better diagnostic accuracy than other EBI markers (i.e., clinical examination, SEBES, or Hijdra scores), in particular for patients unconscious at admission. To stratify the risk of severe EBI, we used a gray zone approach, defined by two thresholds: one to exclude (lower) and one to predict (upper) severe EBI with almost certainty. The lower threshold of the gray zone was close to the 95th percentile reported in healthy controls [33]. Values below this threshold are considered normal by the laboratory (Cobas, Roche, Mannheim, Germany). In the study population, given a 14.8% (n = 12/81) pre-test probability of being unconscious at day 3, an S100B level below the gray zone gives a 4% post-test probability, while an S100B level over this gray zone gives a 50% post-test probability (Figure 3). If patients were already unconscious at admission, the pre-test probability was 36% (n = 9/25), thus an S100B level below the gray zone gives an 11% post-test probability, while an S100B level above the gray zone gives an 87% post-test probability. Using this approach, patients with an S100B level greater that the upper threshold, but with an M-GCS = 6 at day 3, were considered as false positives. This S100B elevation could, in some patients, be explained by an intraparenchymal hematoma instead of severe EBI. Nevertheless, these patients had a high frequency of DCI and a poor neurological outcome at ICU discharge, suggesting a greater amount of brain injury, despite a reassuring early clinical presentation. Therefore, we propose that patient management strategy and EBI evaluation include an early S100B measurement, in addition to clinical examination and brain imaging. Patients with a low S100B level are expected to have mild EBI with good early and delayed outcomes, while those presenting with moderate or severe EBI are at risk of secondary injury and require close neurological monitoring. Taken together, these results suggest that physicians should adapt their neurological monitoring strategy in patients with moderate or severe EBI without delaying multimodal monitoring (e.g., continuous EEG, transcranial Doppler, or brain oxygen probe, Figure 3).S100B serum concentration has previously been evaluated to predict long-term prognosis. Since the long-term outcome is the result of early and delayed injury, the peak or mean S100B values of the first week was found to have a good diagnostic value to predict poor long-term outcomes (AUC from 80% to 93%) [18,20,21]. Unlike the values of the first week, early S100B has a poorer diagnostic accuracy to predict a poor outcome (i.e., mRS > 2) [18,21,34]. Herein, S100B was elevated only in patients with an mRS > 4 and was a good predictor of catastrophic outcome, such as death or disorder of consciousness. The early S100B values in patients with intermediate outcomes were more heterogeneous and this could explain why other authors found a poor diagnostic accuracy, as they aimed to predict mRS > 2 (i.e., all levels of disability) [18,21,34], whereas herein the cut-off was mRS > 4 (i.e., severe disability or death).As evidenced by the sustained S100B elevation, patients with severe EBI undergo prolonged brain aggression leading to catastrophic outcomes. The pathophysiology of these brain injuries is yet to be elucidated, and S100B protein could by itself play a role in brain injury progression. The ratio between serum and cerebral spinal fluid S100B concentration is close to the one early after SAH [35]. Therefore, brain concentration is expected to reach the micromolar range, at which it can have neurotoxic effects [36]. S100B could thus be both a biomarker of brain damage and a damage-associated molecular pattern (DAMP) molecule. There is now growing evidence that DAMPs released upon the primary hemorrhage are involved in EBI and trigger cellular processes, such as neuro-inflammation, eventually leading to delayed ischemia [37]. DAMPs, such as the S100 family of proteins but also HMGB-1 or extracellular matrix-derived proteins, are released into the blood compartment and have been used as biomarkers of cerebral vasospasm, DCI, or long-term outcome [37,38,39]. Furthermore, HMGB-1 early concentration (i.e., at day 1 after the ictus) has been reported to predict the occurrence of cerebral vasospasm regardless of the initial severity [37]. Unlike HMGB-1, early S100B systemic concentration seems to not be predicative of DCI or cerebral vasospasm [21], and further evaluation of several biomarkers of EBI, including DAMPs and markers of cellular damage, are required to identify patient profiles.Several limitations of this study must be noted. First, there were some missing measurements at admission or day 1. Nevertheless, even if S100B is known to have a short half-life both in the blood and brain extracellular space [40], we and others have found a sustained plasmatic elevation over the first 48 h following the ictus [18,20]; we therefore believe that a single measurement of S100B within the first day is representative of this period. Second, although the number of patients included was higher than that required according to power calculations, the small number of unconscious patients at admission limits the extrapolation of the results to all poor-grade SAH. This preliminary study will need an independent validation cohort to confirm the accuracy of the different thresholds that we suggest to be used at the bedside, in particular among patients with an M-GCS < 6 at admission. Furthermore, we could not separately analyze patients with early complications, such as re-bleeding, in whom the S100B levels did not only reflect brain injury due to the primary hemorrhage. Third, the preference for coiling whenever possible in our institution may also limit the generalizability of the reported thresholds to institutions with a higher frequency of surgical procedures which are known to increase S100B levels [20]. With such a strategy, S100B values are expected to be higher even in those with mild or moderate EBI, and its diagnostic accuracy might thus be poorer with higher thresholds. Finally, since S100B is also expressed in melanocytes and Langerhans cells [41], its serum concentration is known to be higher in dark-skinned people, as well as after sun exposure [42,43]. Although we did not collect patient skin pigmentation, the population admitted to our institution is predominantly Caucasian. This study provides evidence that S100B serum concentration within the first 24 h seems to have a good diagnostic value to predict severe EBI consequences, especially in patients unconscious at admission. Before claiming that S100B could thus be used as a surrogate for EBI severity to start earlier multimodal monitoring, these results must be confirmed in an independent validation cohort.Conceptualization, B.B., T.R., and F.D.; methodology, B.B., T.R., A.P.-L.; S100B data acquisition: C.R., L.D., A.P.-L., patient data acquisition T.R., F.G., C.B., R.C., A.T., F.D.; data analysis, B.B. and T.R.; writing—original draft preparation, B.B., T.R., F.G., A.P.-L., and F.D.; writing—review and editing, all authors; supervision, F.D. and A.P.-L.; project administration, F.D. All authors have read and agreed to the published version of the manuscript.This research received no external funding.The authors wish to acknowledge Philip Robinson, Department of Clinical Research and Innovation of the Hospices Civils de Lyon, for his help in writing the manuscript.The authors declare no conflict of interest.Flow chart. EBI: Early brain injury, GCS: Glasgow coma scale, M-GCS: motor score of GCS, SAH: subarachnoid hemorrhage. Patients’ clinical evolution over the first 3 days and their affiliation to the EBI-mild, EBI-moderate, and EBI-severe groups is presented. Among the three patients who worsened within 3 days, two had hydrocephalus requiring external ventricular drainage (EVD). The third was intubated for respiratory disorder, her M-GCS at day 3 was 5 without sedation and her modified Rankin Scale (mRS) at intensive care unit discharge was 3 (i.e., moderate disability). None had re-bleeding or aneurysm treatment complications.Receiver–operator characteristic (ROC) curves of early brain injury (EBI) biomarkers. (a) ROC analyses of the total population (n = 81). (b) ROC analyses of patients with a motor component of the Glasgow coma score < 6 at admission (n = 25). Several S100B thresholds are plotted on both panels: (black square) the best threshold; (gray line) the gray zone of diagnostic uncertainty with (black dots) as its lower limit (i.e., to exclude severe EBI with almost certainty (Se ≥ 90%)) and upper limit (i.e., to predict severe EBI with almost certainty (Sp ≥ 90%)).Algorithm chart for early brain injury severity prognostication. Early brain injury (EBI) severity prognostication and the proposed resulting management strategies are presented based on early S100B serum concentration. Low S100B: concentrations below the lower threshold of the gray zone rejecting severe EBI; high S100B: concentrations above the upper threshold of the gray zone predicting severe EBI. An S100B concentration within the gray zone has an uncertain diagnostic value but suggests significant brain injury. EEG: electroencephalography; ICH: intracerebral hematoma; TCD: transcranial doppler.Patient characteristics according to EBI group.The three groups were defined according to M-GCS at ICU admission and day 3: EBI-mild M-GCS = 6 at arrival and day 3, EBI-moderate M-GCS < 6 at arrival and =6 on day 3, EBI-severe M-GCS < 6 both at arrival and day 3. Results of post hoc tests: Differences compared to the EBI-severe group, * p < 0.05 ** p < 0.01 *** p < 0.001; differences between EBI-moderate and EBI-mild groups, ¥ p < 0.05 ¥¥ p < 0.01 ¥¥¥ p < 0.001. GCS: Glasgow coma scale, OAD: oral antidiabetic drug, ACA: anterior cerebral artery, ACoA: anterior communicative artery, AChA: choroidal artery, PCA: posterior cerebral artery, PCoA: posterior communicative artery, MCA: mean cerebral artery, DCI: delayed cerebral ischemia, ICH: intracerebral hematoma; mRS: modified Rankin Scale.Accuracy of maximal S100B and GCS at ICU admission to predict an M-GCS < 6 on day 3 in the whole cohort and only in patients with an M-GCS < 6 at ICU admission.The gray zones of diagnostic uncertainty are presented with the percentage of patients inside, and their interval likelihood ratio (inter-LHR). The positive LHR of the highest value of the biomarker in the gray zone is considered to include the diagnosis and the negative LHR of the lowest value to exclude the diagnosis. AUC: area under the curve, 95% CI: 95% confidence interval, Se: sensitivity, Sp: specificity, PLR: positive likelihood ratio, NLR: negative likelihood ratio.

Med-MDPI/jcm_2/jcm-09-06-01747.txt

@@ -0,0 +1 @@
+Equal contribution as ‘first author’.Equal contribution as ‘last/senior author’.Background: Oral and enteral mucositis due to high-dose cytostatic treatment administered during autologous and allogeneic stem-cell transplantation increases mortality. Salivary secretory immunoglobulin A (sIgA) is a basic pillar of local immunity in the first line of defense. Altered salivary sialoglycoprotein carbohydrates are important in the pathologies in the oral cavity including inflammation, infection and neoplasia. Therefore, we assessed whether changes in the salivary and serum IgA glycosylation correlated with development and severity of oral mucositis. Methods: Using capillary electrophoresis, comparative analysis of serum and salivary IgA total N-glycans was conducted in 8 patients with autologous peripheral stem-cell transplantation (APSCT) at four different stages of transplantation (day −3/−7, 0, +7, +14) and in 10 healthy controls. Results: Fourteen out of the 31 structures identified in serum and 6 out of 38 in saliva showed significant changes upon transplantation compared with the control group. Only serum core fucosylated, sialylated bisecting biantennary glycan (FA2BG2S2) showed significant differences between any two stages of transplantation (day −3/−7 and day +14; p = 0.0279). Conclusion: Our results suggest that changes in the serum IgA total N-glycan profile could serve as a disease-specific biomarker in patients undergoing APSCT, while analysis of salivary IgA N-glycan reflects the effect of APSCT on local immunity.Comprehensive analysis of the carbohydrate moiety of glycoproteins offers new paths for biomarker research [1,2]. Glycosylation is essential for the functions of immunoglobulins, such as secretory immunoglobulin IgA (sIgA) dimerization, polymeric Ig receptor-mediated transcytosis, and adhesion of pathogens to the mucosal surface, and is responsible for antibody binding to the mucus layer [3]. Several saliva components protect the mucosa. Salivary sIgA is crucial in immune exclusion via direct interaction with microbial antigens, and eliminates viruses by non-virulent immune complex formation, whereby N-glycan sialic acids of sIgA inhibit sialic acid-binding viruses. It also neutralizes bacterial lipopolysaccharide (LPS), and maintains commensal homeostasis, thereby preventing disseminating pathogens [4]. IgA in the serum also has an anti-inflammatory role [5], especially in cases when systemic immunity weakens. In particular, as a result of chemotherapy, there is decreased sIgA secretion in acute leukemia. In oral mucositis (OM), which is a common complication of the cytostatic treatment, there is a further decrease in IgA secretion [6]. The attenuated antimicrobial activity is partly due to the reduced antioxidant capacity, which increases mucosal sensitivity to infections and tissue injury [7]. High-dose cytostatic therapy administered as part of autologous peripheral stem-cell transplantation (APSCT) in hematological malignancies often causes severe oral and enteral mucosal barrier injury. As a consequence, dissemination of pathogens and fatal sepsis may ensue. In less severe cases, mucositis increases the length of hospitalization, enhances susceptibility to further infections and significantly worsens quality of life. During APSCT, serum IgA, which has a diverse role in mucosal immunity, also decreases [8]. While serum immunoglobulin A (IgA) usually returns to the normal level within six or seven months, salivary sIgA level needs up to five years to recover, probably due to the persistence of weak mucosal immunity [7]. The biochemical and immunochemical properties of serum and secretory IgA are different. Therefore, the sIgA glycosylation pattern could be suitable as a possible biomarker to monitor pathological processes in the oral cavity [3].Our aim was to examine the effect of high-dose intensive immunosuppressive therapy on local immunity in the context of APSCT (Figure 1). We tested whether the N-glycosylation patterns of both serum and salivary IgA at different stages of APSCT could serve as possible biomarkers of OM, even in subclinical cases. This would enable early diagnosis, more effective therapy and disease monitoring.Collection of serum and saliva samples was carried out at the institutional hematopoietic transplantation centre from 8 patients admitted for APSCT due to malignant hematological disease, and at the dental outpatient clinic from 10 healthy controls between 7 and 8 a.m. The use of human subjects followed an approved protocol and satisfied the requirement of the IRB (ethical approval UD 5570-1/2018/EKU). All subjects gave their informed consent for inclusion before they participated in the study. The study was conducted in accordance with the Declaration of Helsinki.Eight patients (5 females and 3 males) with malignant hematological disease who required APSCT were included in the study. Average age was 49.5 ± 14.19 years. Reason for APSCT was non-Hodgkin lymphoma (NHL) in 3 patients, Hodgkin lymphoma (HL) in 2, and multiple myeloma (MM) in 3 cases. Five patients were in complete morphologic remission (CMR), one in very good partial remission (VGPR) and 2 in partial remission (PR) prior to transplantation. Serum and saliva samplings were performed at the same time on specific days of the peritransplantation period as follows: day of hospital admission (day −3/−7), day of transplantation (day 0) and day +7 and day +14 post-transplantation. OM grade was established according to WHO criteria [9]. As a control group, 10 age-and sex-matched patients were selected (average age: 41.9 ± 18.35 years, 9 females and 1 male). Mann–Whitney t-test (p = 0.2645) showed no statistically difference between the control and the transplanted group. For more details of patients’ demographics see Table S1. The conditioning regimen was BEAM (BCNU, etoposide, cytosine arabinoside, melphalan) protocol in Hodgkin and non-Hodgkin lymphoma prior to the transplantation [9], while in MM it was high-dose melphalan (≥200 mg/m2) [9]. Patients with severe chronic disease (diabetes, autoimmune diseases, acute or chronic inflammatory diseases, etc.) and previous malignancy were excluded from the study. Patients in both groups were free of dental foci (dental calculus, radices, etc.) at the time of sampling. Study design was aligned with STROBE recommendations [10] and, using sample size calculator Sampsize (epiGenesys, Sheffield, UK), it was a pilot study [11]. Power values were in the range of 59–99% with median 94% using G-power 3.1.9.2. software (Informer Technologies Inc., Düsseldorf, Germany). Bone marrow biopsy examination, qualitative and quantitative analysis of peripheral blood samples and measurement of serum immunoglobulin levels were performed at admission (day −3/−7). Results were in the normal range in each patient and immunoglobulin A levels in particular were between 0.85 g/L and 3.2 g/L (reference range: 0.7–4.00 g/L). This indicates that the plasma cell repertoire was not affected prior to transplantation. Serum samples were collected using clot activator containing serum tubes (BD Biosciences, Franklin Lakes, NJ, USA). The collected blood samples were centrifuged at 7500× g for 30 min and the serum fractions were stored at −70 °C one hour after collection until further processing.Saliva collection was performed according to the standard methods [12]. Both controls and patients were in a sitting position during the sampling with eyes open and a slightly tilted head. Following oral cavity rinse with 25 mL of physiological saline solution (B. Braun Melsungen AG, Melsungen, Germany) for 30 s, saliva was collected for 5 min in an externally pre-disinfected 15 mL lockable Falcon tube (Sigma-Aldrich, St. Louis, MO, USA). Participants adapted to the test condition for 5 min prior to sample collection. Taking into account the diurnal variation of saliva constituents, samplings were done at a specified time window: between 7 a.m. and 8 a.m., one hour after eating, drinking, or tooth-brushing in order to avoid contamination. Patients in sterile rooms used a gauze plate or DenTips (MDS096502, Medline Industries. Inc., Mundelein, IL, USA), and a disposable oral swab, impregnated with physiological saline solution, in order to maintain optimal oral hygiene during the period of cytopenia. Within one hour of collection, Halt Protease Inhibitor Cocktail (Sigma-Aldrich, St. Louis, MO, USA) was added proportionally to the saliva samples. After homogenization, saliva samples were aliquoted into 1.5 mL Eppendorf tubes and stored at −70 °C until further processing.Venous blood samples (5 mL) were collected into Vacutainer tubes anticoagulated with ethylenediaminetetraacetic acid (EDTA) (Vacutainer Systems, Rutherford, NJ, USA) and serum IgA levels were detected using Sysmex XN-2000 Hematology Analyzer (Sysmex Hungary, Budapest, Hungary).After collection of saliva samples, IgA levels were measured by IDK sIgA ELISA kit (Immundiagnostik, Bensheim, Germany) according to the manufacturer’s instructions. We determined the salivary IgA secretion rate (µg/min), because it is a more stable value than IgA concentration [13].Principal component analysis (PCA) and one-way analysis of variance (ANOVA) were performed with SPSS 22 (IBM, Armonk, NY, USA) using PeakAreas% as input derived from 32 Karat software (SCIEX, Brea, CA, USA). The Shapiro–Wilk test was performed to investigate the normal distribution of data. If it passed the normality test (p > 0.05), ANOVA followed by Tukey post hoc test was used to compare peak intensities between experimental groups (see Supplementary Materials Tables S5 and S6), otherwise the Kruskall–Wallis test followed by Dunn’s multiple comparison was used. Differences between means at p < 0.05 were considered as significant. Spearman correlation analysis was performed to correlate serum or salivary IgA concentrations and flow rate with oral mucositis grades. For analysis of serum and salivary IgA ELISA results and flow rates, Mann–Whitney and Wilcoxon tests were used.Chemicals and reagents, Z(IgA1) antibody production, expression and purification, serum and salivary IgA capturing, N-glycan release and fluorophore labeling, exoglycosidase based carbohydrate sequencing and capillary electrophoresis analysis were applied as described in detail in our recent paper [14].There was a continuous significant decrease in serum IgA levels during APSCT (day 0, day +7, day +14) as compared to the control group (p = 0.024; p = 0.005; p = 0.004) and to the day of admission (p = 0.027; p = 0.028; p = 0.028) (Figure 2a). The IgA secretion rate was lower in the remission stage than in controls at the first sampling (day −3/−7 prior to transplantation). At the further stages of APSCT (day 0, day +7, day +14), significant differences were observed between the controls and patients (p = 0.015; p = 0.001; p < 0.001) (Figure 2b).Contrary to expectations, the amount of UWS did not decrease in the patients in pre-APSCT remission compared to the control group. During APSCT, there was a significant decrease at day 0, day +7 and day +14 in UWS flow rate as compared to the control group (p = 0.008; p = 0.004; p = 0.001) and the day of admission (p = 0.012; p = 0.012; p = 0.012), respectively (Figure 2a).The highest OM grade was variable, with grade 1 (n = 3), 2 (n = 2), 3 (n = 2) and 4 (n = 1). There was no correlation between serum IgA (g/L) or salivary IgA secretion rate (µg/min) and the degree of oral mucositis (p = 0.685; p = 0.1729) (Figure 2d,e). In contrast, there was negative correlation (r = −0.3622; p = 0.0416) between decreased salivary flow rate (mL/min) and increasing severity of OM (Figure 2f).Figure 3a shows the significantly changed serum IgA N-glycan structures in the control group and patient group at four stages of transplantation. The core fucosylated, sialylated bisecting biantennary glycan (FA2BG2S2) was the single significantly different structure between any two specified time points of the peritransplantation period (day −3, −7 and +14; p = 0.0279). Further 14 N-glycan structures showed significant differences (p < 0.05) between controls and any stages of APSCT. The key statistical characteristics (mean and standard errors of the PeakArea%) of the significantly changed serum IgA N-glycan structures are listed in Table S2.Next, principal component analysis was performed on serum IgA N-glycans. In serum samples, the two principal component axes accounted for 11.47% and 53.45% data variance, respectively, representing 64.92% of data variance cumulatively, which was sufficient to resolve the data into two distinct groups (controls and patients). In the patient group, there was no separation of the four transplantation stages into distinct statistical groups (Figure 4a).There were six significantly changed salivary IgA N-glycan structures in the control as well as in the patient group at the four stages of transplantation (Figure 3b). None of the structures changed significantly between any two specified time points of the peritransplantation period. The key statistical characteristics (mean and standard errors of the Peak Area%) of the significantly changed salivary IgA N-glycan structures are listed in Table S5.In saliva, the two principal component axes accounted for 21.14% and 35.55% data variance, respectively, representing 56.69% of data variance cumulatively, which was insufficient to resolve the data either into control and patient groups or into different groups representing the four transplantation stages (Figure 4b). Comparison of serum- and saliva-specific N-glycans revealed higher numbers of neutral and mannosylated structures in the saliva than in the serum.We calculated the ratios of sialylated and neutral structures in all three possible scenarios (present in serum; in saliva; and in both (i.e., ’overlapping structures’) in the control and patient group at four stages of transplantation (Figure 5, Table S6). This ratio was significantly higher in serum in all examined stages of APSCT as compared to the control group (p = 0.002; p = 0.001; p = 0.002; p = 0.043). A significant change of the SF/NF ratio was observed between two specified time points of the transplantation (day −3/−7 and day 0; p = 0.05). This ratio was also significantly higher in saliva samples at the day of admission and day 0 compared to the controls (p = 0.021; p = 0.009). The SF/NF ratio of the overlapping structures in serum was significantly higher in all examined stages of the APSCT compared to the controls (p < 0.001; p < 0.001; p < 0.001; p = 0.006) and significantly lower between day 0 and day +14 (p = 0.036).Mucosal barrier injury (oral and enteral mucositis) developing as a result of high-dose intensive cytostatic treatment administered during autologous peripheral stem cell transplantation is often a life-threatening complication. Oral mucositis is a disease of multifactorial ethiopathogenic origin with several patient and treatment related risk factors, which were considered at the study design and exclusion criteria. However, genetic susceptibility was not assessed and we cannot exclude that this could have affected the incidence and severity of mucositis in our patients [15]. Oral mucositis has neither targeted therapy nor biomarkers [9]. Saliva, which contains more than 1000 mostly glycosylated proteins, is a good indicator of changes in plasma constituents’ (hormones, drugs, etc.) and an active subject of biomarker research [16]. Immunoglobulins are glycoproteins with a wide range of functional diversity. Highly glycosylated IgA has the most prevalent role in the mucosal defense [17]. The connection between altered glycosylation of salivary immunoglobulins and oral disease is a rapidly emerging field of research.Several studies have shown that during APSCT, salivary flow rate (both UWS and stimulated whole saliva) decreases because the cytotoxic drugs damage the salivary glands [6,18]. Our results are consistent with these observations. On the day of admission, there was no difference between the control and patient groups, which suggests that the effect of treatments prior to APSCT was not significant compared to the profound effect of high-dose cytostatic drugs. The reserve and regenerative capacity of salivary glands may also play an important role.Serum IgA level is affected by several factors, e.g., gender, age, infections. In humoral immuno-deficiencies, including hematological diseases, IgA levels are usually lower than that of the normal population [19,20]. During transplantation, we can observe a further decrease in serum IgA levels, which reach the normal range six months after autologous bone marrow transplantation (BMT) and 12–24 months after allogeneic BMT, unless graft-versus-host disease (GVHD) occurs [8]. In our study, there was a significant decrease in serum IgA during APSCT as compared to both controls and admission day (Figure 2a).A significantly reduced level of immunoglobulin subclasses was observed in saliva before APSCT and within five years after [18,21]. Although previous data suggest that sIgA secretion rate decreases in patients with mucositis undergoing chemotherapy compared to patients without mucositis [6], our study revealed no significant correlation between serum IgA and salivary IgA secretion rate and OM degree (Figure 2d,e). Our results indicate that not only the decreased amount but also the altered quality (i.e., glycosylation pattern) of saliva constituents contributes to mucosal barrier injury, increasing the risk and severity of OM.Analysis of the serum IgA N-glycome profile revealed 14 structures with significant differences between the control and patient group at any stage of transplantation. There was one structure, FA2BG2S2, which showed a significant difference between two stages of transplantation (day −3/−7 and day +14) (Figure 3a), a finding worth detailed analysis in the future. In the serum, there was a clear distinction between the control group and the patient group (Figure 4a). This suggests that changes of the serum IgA N-glycan profile can be a disease specific biomarker in APSCT patients. Our results are confirmed by a recent report on altered glycosylation and increased sialylation of IgA1 in the serum of patients with breast cancer, thus stressing its role as a potential biomarker [22].The salivary IgA N-glycome profile revealed six structures with significant differences between the control and patient groups at any stage of transplantation. None of these changes have been reported in APSCT yet. Earlier studies reported only on the physiological analysis of the sIgA H chain, J chain and secretory component (SC), and detailed N-glycan and sIgA1 O-glycan profiling [23] and site-specific analysis of salivary IgA [3]. Salivary cytokines and immunoglobulins are important in oral inflammation in malignant hematological disorders, with biomarker potential regarding the efficacy of chemotherapy [24].Our results suggest that the salivary IgA N-glycome profile reflects the effect of APSCT on local immunity; therefore, it can be suitable for long-term patient follow-up. Serum N-glycome profiles are capable of recognizing differences between control and patient groups. However, this is less pronounced in saliva because it is a sensitively changing open system with great individual variability. It contains approximately 108 viable microorganisms per ml, many of them are capable of producing diverse exoglycosidases (e.g., Streptococcus oralis sialidase), modifying the glycosylation patterns of salivary components [16]. Parallel with the progression of periodontitis, hypogalactosylation of serum and gingival crevicular fluid (GCF) IgG has been observed [25]. Although sIgA is more resistant to bacterial proteases than serum IgA [3], the immunoglobulin-degrading effect of certain periodontal pathogens needs to be considered [26]. Thus, the highly variable cariological and periodontal status and person-specific oral flora can further affect glycosylation of the secreted sIgA [3,9,27,28,29].At least 95% of salivary pool IgA is produced by the plasma cells (PCs) of salivary glands. A minority can be derived from the local PCs of the periodontium as well as serum through the gingival sulcus (crevicular epithelium) via paracellular passive leakage. sIgA- and IgA-producing B-cells have different glycosylation mechanisms and the expression of glycosyltransferases can be altered by several local factors. This explains the differences in regulatory mechanisms in B-cells in different anatomical sites [3].Several studies have examined the histomorphological effects of chemotherapy on salivary glands both in solid tumors and in hematological malignancies, which is accompanied by a decrease in IgA-producing plasma cells and impaired transcellular immunoglobulin transport [6]. Thus, we can assume that the pre-treatments used and the high-dose conditioning treatment itself during APSCT contribute to salivary gland degeneration to a variable degree, leading to quantitative and qualitative differences in the produced sIgA. Due to the cytotoxic effects, mucosal barrier integrity is impaired, allowing IgA to enter saliva from serum, which can modify the glycosylation profile of the saliva. Hormonal effects can also be assumed. While there is no substantive difference between the two genders in terms of the amount of serum IgA, the salivary sIgA of fertile women was higher than in men, with a fluctuation parallel to serum estradiol (E2) levels suggestive of a relationship between sIgA secretion and E2 [30]. Furthermore, B cells express estrogen receptors and estrogens play regulatory roles in the galactosylation of IgG [31]. The role of sexual dimorphism on the amount of sialic acids in the gingival crevicular fluid has also been confirmed [32]. Appearance of tear and nasal secretion in the saliva may further modify glycosylation patterns [26]. Several other lifestyle, psychological and environmental factors (physical activity, emotions, smoking, infections) can affect the amount and quality of salivary IgA [33], offering a scope for further studies.Asialoglycoprotein receptors in the liver eliminate non-sialylated proteins from serum, leading to a higher ratio of sialylated components in the blood than in the saliva [3], in agreement with our findings. Hypersialylation is a glycosylation modification that correlates with tumor genesis, stage, progression and prognosis [34]. Increased free serum sialic acid occurs both in solid tumors including cancers of the oral cavity [35] and in lymphoma and multiple myeloma [36,37,38]. Changes in both free sialic acid levels and N-glycan associated sialylated structures reflect the effectiveness of cytostatic treatment as a useful marker of pathological processes and therapeutic response [38,39,40]. Accordingly, we examined SF/NF ratios. There was a decrease in neutral glycans and an increase in sialylated structures in the patient group compared to controls at any time point of transplantation, except for the overlapping structures in the saliva. Further studies are needed to correlate this with therapeutic response and disease course.Oral mucositis is a frequent and life-threatening complication of the intensive conditioning regimen used for hematopoietic stem cell transplantation. Our results indicate that the serum IgA N-glycan profile can serve as a disease-specific biomarker in APSCT patients, whereas changes in salivary IgA reflect its effect on local immunity, rendering it suitable for monitoring OM progression and therapy response. Considering the lack of reliable biomarkers and proven therapeutic targets in OM, our results are clinically relevant, requiring further studies with more patients and longer follow-up periods. Glycosylation site-specific analysis of IgA isotypes and of the secretory component may also provide further insight into the underlying pathological processes. The relevance of IgA glycosylation patterns in viral and other infections of the oral cavity, enteral and respiratory mucosal surfaces is another very timely topic with an urgent need to elaborate novel therapeutic strategies against aggressive pathogens.The following are available online at https://www.mdpi.com/2077-0383/9/6/1747/s1, Table S1: The key demographic and clinical data of the participants, Table S2: Mean and standard errors of the PeakArea% of the significantly changed IgA N-glycan structures in the serum, Table S3: Comparison of various IgA N-glycan structures in serum in different groups and time points with Tukey’s multiple comparisons statistical test, Table S4: Comparison of various IgA N-glycan structures in saliva in different groups and time points with Tukey’s multiple comparisons statistical test. Table S5: Mean and standard errors of the PeakArea% of the significantly changed IgA N-glycan structures in the saliva, Table S6: The sialoform to neutral (SF/NF) carbohydrate ratio of saliva and serum total IgA-associated glycans (upper half) and of the overlapping (i.e., present both in serum and saliva) glycan structures (lower half).Conceptualization, E.G., T.H.; Methodology, E.G., Z.K., B.M., H.J., F.V., A.G., T.H.; Validation, E.G., T.H., Z.K., B.M., F.T., Á.S., A.G.; Formal Analysis, E.G., Á.S.; Investigation, E.G., T.H., Z.K., B.M., F.T., Á.S., A.G.; Resources, E.G., T.H., A.K., A.G.; Data Curation, E.G., T.H., A.G.; Writing—original draft preparation, E.G.,T.H.; Visualization, E.G., Á.S., T.H., A.G.; Supervision, T.H., A.G.; Project Administration, E.G., T.H., A.G.; Funding Acquisition, E.G., T.H., A.G. All authors critically revised the manuscript for important intellectual content, gave final approval, and agree to be accountable for all aspects of the work. All authors have read and agreed to the published version of the manuscript.The authors disclose receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the Faculty of Dentistry University of Debrecen Research Fund [GINOP-2.3.2.15-2016-0001]; Hungarian Brain Research Programme [2017-1.2.1-NKP-2017-00002]; National Research, Development and Innovation Office [NKFIH-SNN-NN-132999-2019]; [NN-127062]; [2018-2.1.17-TÉT-KR-2018-00010]; and ÚNKP-19-4 New National Excellence Program of The Ministry for Innovation and Technology, Hungary [BIONANO_GINOP-2.3.2-15-2016-00017].We are grateful to Csaba Hegedűs, Ágnes Kinga Bágyi and István Varga for providing the research facilities. The help of dentistry students Miss Réka Ágoston and Miss Helga Sipos in sample collection is thankfully appreciated. This is contribution #172 from the Horváth Csaba Memorial Laboratory of Bioseparation Sciences.The authors declare no conflict of interest. This paper is dedicated to doctors, health professionals and researchers fighting the new coronavirus (COVID-19) pandemic.Study design. Timeline of autologous peripheral stem cell transplantation (APSCT). Saliva and serum sampling were performed at the four defined time-points of APSCT (day −3/−7, day 0, day +7, day +14). Timescale pre-and post APSCT were different among the individual cases (i.e., the thin line is not proportional to the elapsed time). Day −3/−7 is the time of admission and start of the conditional therapy; day 0 is the day of the transplantation (administering the harvested CD34+ stem cells in a stem cell infusion); day +7 is usually the deepest point of cytopenia with most severe oral mucositis; by day +14, neutrophil and thrombocyte engraftment has developed if insertion and proliferation of stem cells were successful.Serum IgA concentration (a), salivary IgA secretion rate (b), salivary flow rate (c) of unstimulated whole saliva (UWS) in the control and patient group at four stages of autologous peripheral stem cell transplantation (APSCT) and its correlation (d–f) with severity of oral mucositis (OM). In (a–c), the values are expressed as sample means; error bars represent the estimates of standard deviations calculated from three parallel measurements (* p ˂ 0.05, ** p ˂ 0.01, *** p ˂ 0.001).Fourteen serum (a) and six salivary (b) IgA glycoforms showed significant changes during autologous peripheral stem cell transplantation (APSCT). The abbreviations of the Oxford nomenclature were applied as referred to in our previous work [14]. The plots show the distribution pattern at various APSCT time points in patients and in controls (* p ˂ 0.05, ** p ˂ 0.01, *** p ˂ 0.001). For detailed results of statistical tests see also Tables S3 and S4.Principal component analysis of serum (a) and salivary (b) IgA N-glycosylation datasets in controls and at four different stages of APSCT (for key to symbols see upper right corner).The sialoform to neutral (SF/NF) carbohydrate ratio (%) of serum (a) and salivary (b) IgA total N-glycans and the overlapping (i.e., present both in serum and saliva) structures (c,d) in the control and patient group at four stages of autologous peripheral stem cell transplantation (APSCT). Values are expressed as sample means; error bars represent the estimates of standard deviations calculated from three parallel measurements (* p ˂ 0.05, ** p ˂ 0.01, *** p ˂ 0.001).

Med-MDPI/jcm_2/jcm-09-06-01748.txt

@@ -0,0 +1 @@
+Co-first authors.Limited studies have examined the intestinal microbiota composition in relation to Crohn’s disease (CD) prognosis. We analyzed the differences in microbial communities and relevant metabolic pathways associated with prognostic variables in patients with CD. We applied 16S rRNA gene sequencing to analyze a cohort of 1110 CD and healthy control (HC) fecal samples. We categorized patients with CD into good (CD-G), intermediate (CD-I) and poor (CD-P) prognosis groups, according to the history of using biologics and intestinal resection. Microbiota α-diversity decreased more in CD-P than CD-G and CD-I. Microbiota ß-diversity in CD-P differed from that in CD-G and CD-I. Thirteen genera and 10 species showed differential abundance between CD-G and CD-P groups. Escherichia coli (p = 0.001) and species Producta (p = 0.01) and genera Lactobacillus (p = 0.003) and Coprococcus (p = 0.01) consistently showed differences between CD-G and CD-P groups after adjusting for confounding variables. Functional profiling suggested that the microbial catabolic pathways and pathways related to enterobacterial common antigen and lipopolysaccharide biosynthesis were better represented in the CD-P group than in the CD-G group, and E. coli were the top contributors to these pathways. CD prognosis is associated with altered microbiota composition and decreased diversity, and E. coli might be causally involved in CD progression, and may have adapted to live in inflammatory environments.Inflammatory bowel disease (IBD), the most prevalent forms of which are Crohn’s disease (CD) and ulcerative colitis, is a chronic inflammatory gastrointestinal disorder. The pathogenesis of IBD involves a complex interplay among a genetically susceptible host, enteric microbiota, environmental factors, and dysregulated immune response [1,2]. As one such factor, the gut microbiota is gaining increasing attention, owing to its influence on IBD. IBD has emerged as one of the most studied diseases linked to the gut microbiota [3].Using fecal and mucosal samples and culture-independent techniques, several studies have consistently reported that patients with IBD, in particular patients with CD, are associated with a lower microbial α-diversity and are enriched in several groups of bacteria when compared to the microbial diversity of healthy controls (HC) [4]. Although no single causative microbe has been identified, previous studies have reported the potential contribution of microbial pathogens, such as Clostridium difficile [5], enterotoxigenic Bacteroides fragilis [6] and Escherichia coli [7] to CD disease pathogenesis.CD has a wide range of clinical symptoms and manifestations with varying phenotypes, and the natural history of CD has shown rates of disease complications ranging from 48% to 52%, five years after diagnosis [8]. Thus, many patients with CD require intensive treatment, such as using various biologics or multiple surgical intervention, and efforts have been undertaken to identify risk factors associated with disease complication. In the past, clinical factors such as disease location at the ileum, perianal disease, early age of diagnosis, smoking, requirement for steroids at first flare, and deep mucosal ulcerations were identified as important risk factors for the development of CD complications [9]. In addition to these clinical factors, a substantial heterogeneity in the disease course of CD suggests the possible involvement of a strong host biological component like the microbiome; there are emerging fields of research focusing on the microbiome, metagenomics and metaproteomics [10].In this study, we assessed whether there are differences in taxonomic and metabolic microbial signatures among patients with CD with a different clinical course.The present study was undertaken in parallel with a prospective multicenter study performed by the IMPACT (Identification of the mechanism of the occurrence and progression of Crohn’s disease through integrated analysis on both genetic and environmental factors) study. In 2017, the IMPACT study team was established in Korea and obtained a national grant to organize a prospective CD patient cohort (aged > 8 years), for identification of the mechanism of the occurrence and progression of Crohn’s disease. A total of 16 university hospitals are now participating in this study, and collect clinical data of patients with CD who were newly diagnosed or followed-up in the institutions and biological specimens (including blood, stool and tissue specimens).To identify the characteristics of the microbiome according to the prognosis of CD, a total of 388 patients with CD who provided stool samples at 14 centers between May 2017 and November 2018 were included in this study. The stool samples were collected on the day of enrollment and if patients were taking antibiotics or probiotics at the time, stool samples were collected after more than 3 months after discontinuation of their medication. Of these, 18 patients were identified to have good prognosis, but since the disease duration was less than 3 years, they were subsequently considered not enough to be categorized as having good prognosis, and were hence excluded. Thus, a total of 370 patients with CD were included in the final analysis.For the healthy control (HC) group, we used the fecal microbiome data from the Kangbuk Samsung Health Study, a cohort study of Korean men and women who undergo comprehensive annual or biennial examinations at the Kangbuk Samsung Hospital Healthcare Screening Center. Among the cohort, 1473 adults agreed to provide stool samples between June 2014 and September 2014, and details were described in previous studies [11]. The stool samples were obtained at the day of comprehensive examination, and participants who use antibiotics within 6 weeks prior to enrollment or probiotics within 4 weeks prior to enrollment (n = 55) were excluded. These HC were matched for sex and age (±5 years), at a matching ratio of 1:2 with CD group, and in total, 370 patients with CD and 740 HC were included (Figure 1).Ethical approval of the present study was provided by the institutional review boards of Kangbuk Samsung Hospital (KBSMC 2016-07-029) and each center. Written consent was obtained from all participants after the nature and possible consequences of the studies were explained. All applicable institutional and governmental regulations concerning the ethical use of human volunteers were followed during this research. The research was carried out in accordance with the Declaration of Helsinki.For patients with CD, data on diagnosis date, disease location, disease behavior, history of medication, history of surgery and disease activity at fecal sampling were collected through a review of electronic medical report at each center.Hospitalization and surgery are well known poor clinical outcome in CD. However, we did not consider hospitalization, as indications for CD related hospitalization might be different among centers and physicians. Instead, we considered using biologics as poor outcome as biologics are covered by National Health Insurance Service of Korea only in moderate–severe active Crohn’s disease, which does not respond to treatment with corticosteroids, immunomodulators, or, have no tolerability, or are contraindicated in these treatments. Thus, we categorized these into three prognosis groups as follows: CD-good prognosis group (CD-G): patients who were using or had a history of using only 5-ASA or immunomodulator (azathioprine or 6-MP); CD-intermediate prognosis group (CD-I): patients who were using or had history of using one biologics or had history of one CD-related intestinal resection; CD-poor prognosis group (CD-P): patients who had a history of using one or more biologics with one or more CD-related intestinal resection, had history of using two or more biologics or two or more intestinal resection. Despite the national insurance coverage on the use of biologics in Korea being within those who do not response to corticosteroids or immunomodulators, there is a tendency to use biologics at an earlier state, as it is well known to improve long-term prognosis on patients with CD. In our study, more stringent criteria for our CD-poor prognosis group were devised, which included: patients on whom the first biologic failed and a second (or additional) biologic agent was introduced or required bowel resection; and patients whom, after an initial bowel resection, the worsening of the disease required them to start biologics or undergo a second (or additional) bowel resection.In both CD and HC groups, fecal samples were collected from participants’ homes by self-sampling within 24 hours, before visiting the hospital and immediately frozen at −20 °C freezer after defecation, as instructed. After collection, the samples were stored in a deep-freezer (−70 °C) in the laboratory, as soon as the participants arrived at the hospital. Details including DNA extraction from stool samples and 16S rRNA gene sequencing in HC were described in previous studies [11]. In patients with CD, DNA from stool samples was extracted within one month after storage using the Stool DNA Isolation Kit (#27600, NIRGEN BIOTEK CORP, Thorold, ON, Canada), according to the manufacturer’s instructions. The V3‒V4 region of the 16S rRNA gene was amplified using the 341F and 805R primers, with added Illumina adaptor overhang sequences, 341F (5′ TCG TCG GCA GCG TCA GAT GTG TAT AAG AGA CAG CCT ACG GGN GGC WGC AG 3′) and 805R (5′ GTC TCG TGG GCT CGG AGA TGT GTA TAA GAG ACA GGA CTA CHV GGG TAT CTA ATC C 3′). Amplicons were purified with a magnetic bead-based clean-up system (Agencourt AMPure XP, Beckman Coulter, Brea, CA, USA). Indexed libraries were prepared by limited-cycle PCR using Nextera technology, further cleaned up, and pooled at equimolar concentrations. The final library was denatured with 0.2 N NaOH and diluted to 6 pM with a 20% PhiX control. Sequencing and demultiplexing were performed on Illumina MiSeq platform using a 2 × 300 bp paired-end protocol, according to the manufacturer’s instructions. DADA2 plugin of the QIIME2 package (version 2018.11, https://qiime2.org) [12] was used to performed the sequence quality control, such as filtering low quality sequences and chimera, and to construct the feature table of amplicon sequence variants (ASVs). The ASVs were generated by denoising with DADA2 and regarded as 100% operational taxonomic units (OTUs). For taxonomic structure analysis, taxonomy was assigned to ASVs using a pre-trained Naïve Bayes classifier and the q2-feature-classifier plugin against the Greengene 13_8 99% OTUs of the 16S rRNA gene sequence database.Analyses for baseline clinical characteristics were performed using SPSS software (SPSS 24.0 for Windows; SPSS, Chicago, IL, USA).For diversity analysis, the feature table was rarefied to 10,000 sequences per sample by random subsampling in QIIME2. To evaluate the alpha diversity, we computed the number of ASVs observed in each sample, Shannon index accounting both evenness and richness, and Faith’s phylogenetic diversity (PD) [13]. The Kruskal–Wallis test as a non-parametric statistical test was used to test difference for all or pairwise groups. For measuring beta diversity, we used UniFrac distance [14] to estimate dissimilarity among group membership, by incorporating the phylogenetic distances between ASVs. The unweighted and weighted UniFrac distance were calculated for the presence/absence and the abundance of ASVs, respectively, and permutational multivariate analysis of variance (PERMANOVA) with 999 random permutations was used to test significance among groups. PERMANOVA pairwise comparisons for post-hoc test were then conducted to test for significant differences between two groups using the Benjamini and Hochberg (FDR) correction (q-value < 0.05).An analysis of the composition of microbiome (ANCOM) [15] test was performed to determine if there were significant differences in the relative abundance of any taxa between two groups (HC vs. CD and CD-G vs. CD-P), across multiple taxonomic levels. ANCOM accounts for the compositional nature of the relative abundance of taxa. Linear discriminant analysis effect size (LEfSe) analysis was used to detect potential CD progression-specific bacterial markers [16]. Finally, to adjust for confounding factors (age, location, behavior, disease activity and bowel resection history), we applied multivariate association with linear models (MaAsLin) [17], which has the capability to deal with covariates, and compared the abundance of taxa between the CD-G and CD-P group.Functional inferences of the microbial community were conducted according to the Phylogenetic Investigation of Communities by Reconstruction of Unobserved States (PICRUSt). We performed PICRUSt2 (v2.2.0-b) [18] with ASVs, according to the instructions (https://github.com/picrust/picrust2/wiki). Phylogenetic placement in PICRUSt2 is based on a sequence of three steps: HMMER (www.hmmer.org) to place ASVs, EPA-NG to determine the best position of these placed ASVs in a reference phylogeny, and GAPPA to output a tree of the most likely ASV placements. This results in a phylogenetic tree containing both reference genomes and environmentally sampled organisms, which is used to predict individual gene family copy numbers for each ASV. PICRUSt2 predictions based on the following gene families are supported by enzyme classification numbers (EC numbers, as of 21 Jan 2016). We generated PICRUSt2 EC gene family predictions and Metabolic Pathway Database (Metacyc) pathway abundance predictions [19]. Results were visualized in statistical analysis of taxonomic and functional profiles (STAMP) version 2.1.3 [20], and tested using Welch’s t-test for two groups, CD-G vs. CD-P. All predictions were adjusted by multiple testing correction (Bonferroni q-value < 0.05).The baseline characteristics of patients with CD, according to the prognosis groups, are presented in Table 1. Among the 370 patients, 132 (35.7%) were in the CD-G group, 151 (40.8%) were in the CD-I group and 87 (23.5%) were in the CD-P group. The CD-P group (n = 87) included the patients with one biologics + one bowel resection (n = 36), one biologics + two bowel resection (n = 9), one biologics + three bowel resection (n = 1), two biologics (n = 16), two biologics + one bowel resection (n = 11), two biologics + two bowel resection (n = 6), three biologics (n = 3), three biologics + one bowel resection (n = 3) and four biologics (n = 1).Age at diagnosis was more likely to be lower in the CD-P group, as A1 was 6.0%, 9.3%, and 12.6% in CD-G, CD-I, and CD-P groups, respectively (p = 0.003). Most of the patients were in remission state (93.3%, 89.2%, and 85.2% in CD-G, CD-I, and CD-P groups, respectively) during sampling. There were no differences in location between groups, but when we classified location as L1, L2, L3, regardless of the presence of L4, L3 tended to show a higher proportion in the CD-P group (76.1%) than in the CD-G (60.2%) and CD-I groups (57.4%) (p = 0.05). In behavior, there was a significantly higher proportion of B3 in CD-P group (6.0%, 14.2% and 28.4% % in CD-G, CD-I and CD-P group, respectively p < 0.001), and when we classified behavior into three groups (B1, B2, and B3) regardless of perianal involvement, B3 was increased up to 47.4% in the CD-P group (7.5% and 20.9% in the CD-G and CD-I group, respectively). In CD-P group, the biologics most commonly used by the study participants were Infliximab followed by adalimumab, vedolizimab, and ustekinumab. Other biologics that were used, such as risankizumab (protocol number 1311.6), etrolizumab (protocol number GA29144), mongersen (protocol number GED-0301-CD-002), and PF-00547659 (anti MAdCAM-1 antibody, protocol number A7281007), JNJ-64304500, (NKG2D receptor blocker, protocol number 64304500CRD2001), were in open label clinical trials at the time of this study. Bowel resection history was present in 75% of people in the CD-P group. The mean age of the HC group was 41.7 ± 7.5 and 73.5% of them were male and their mean BMI was 24.1 ± 3.2.The 16S rRNA gene amplicon sequences after QC ranged from 3865 to 949,266 reads per sample (mean = 45,174) and 14,627 features in 1110 subjects. After rarefying the feature tables to 10,000 sequences per sample, HC (n = 623), CD-G (n = 133), CD-I (n = 148) and CD-P (n = 86) were included in the diversity analysis. The alpha diversity of gut microbiota among the HC, CD-G, CD-I, and CD-P groups showed statistically significant differences in the observed ASVs (p = 8.5 × 10−62), Faith’s PD (p = 7.6 × 10−71), and Shannon’s index (p = 3.4 × 10−77). In the pairwise Kruskal–Wallis test, we confirmed significant differences between the HC and CD-G/CD-I/CD-P in observed ASVs, Faith’s PD, and Shannon index (Figure 2). There were significant differences between the Shannon index of the CD-G and CD-P (q = 1.5 × 10−3) and CD-I and CD-P groups (q = 2.4 × 10−3). The CD-P group had lower richness than the CD-G and CD-I groups, as shown in Figure 2 and Table S1.The phylogenic distance indices in both the weighted- and unweighted-Unifrac distance revealed that microbial communities were significantly different among all four groups (PERMANOVA p = 0.001, pseudo-F = 54.91; p = 0.001, pseudo-F = 65.96, respectively). In pairwise comparisons, the unweighted Unifrac distance used for identifying the presence/absence of ASVs between groups differed between all pairs of groups, except CD-G and CD-I groups. Weighted Unifrac distance also showed significant differences in the gut microbial community composition between all pairs of groups, except for CD-G and CD-I groups (Table S2 for full pairwise results). However, owing to a high sample number and inter individual variation, the fecal microbiota between the CD-G and CD-P groups could not be separated clearly by principal coordinates analysis (Figure 3A,C), although the differences in microbial community composition were significant between the two groups in both beta diversity indices (Figure 3B,D).To better understand how the microbial community composition changed with the prognosis of CD, we investigated which organisms were present at different taxonomic levels and their relative abundance. Overall, 18 phyla, 39 classes, 71 orders, 129 families, 314 genera, and 459 species were detected. Bacterioidetes (27.6%), Firmicutes (52.7%), Proteobacteria (12.1%), and Actinobacteria (6.6%) were the most abundant phyla across all subjects. Other phyla identified, such as Verrucomicrobia, Fusobacteria, and Tenericutes, had relatively low abundance (<1%).First, we compared the phylum through species levels of microbial composition between the HC and CD using ANCOM. The W statistic for the significantly different taxa, relative to more than 90 percent other taxa at each taxon, is shown in Table S3. Differential abundance testing revealed 5 phyla, 11 classes, 9 orders, 22 families, 18 genera, and 12 species that were differentially abundant between HC and CD groups. The phyla Actinobacteria (w = 9) and Firmicutes (w = 9), including their classes Coriobacteria (w = 17), Bacilli (w = 14) and Clostridia (w = 14), had significantly different abundance across the four groups. The “w = 14” of Clostridia indicates that the class was significantly different relative to 14 other classes between the two groups.Secondly, we compared the genera and species levels between the CD-G and CD-P groups within the patients with CD. In Table 2, genera and species that had different abundance between the CD-G and CD-P groups, either by the ANCOM analysis (W statistic for the significantly different taxa relative to more than 90 percent other taxa) or LEfSe analysis (with LDA score > 4), are listed. Among the genera and species that had different abundance between HC and CD (Table S1), genera Enterococcus, Lactobacillus, Blautia, Megasphaera, Veillonella, Fusobacterium, Escherichia, and Klebsiella and the species V. dispar and E. coli had different abundances between the CD-G and CD-P groups by the ANCOM analysis. LEfSe analysis revealed that CD-G had a significantly higher abundance of genera Blautia and Fusobacterium than the CD-P group, and CD-P had a significantly higher abundance of genera Lactobacillus, Megasphaera, Veillonella, Escherichia, and Klebsiella and the species V. dispar and E. coli than the CD-G. Figure 4 depicts the LEfSe results (LDA score > 4) from phylum to genus level. In addition to the ANCOM result, the genera Abiotrophia and Selenomonas had a higher abundance in CD-P than the CD-G group and Faecalibacterium, Coprococcus and Bifidobacterium showed higher abundances in CD-G than in the CD-P group.Thirdly, we used the generalized linear models using MaAsLin packages [17], which have the capability of dealing with covariates, to investigate significantly different taxa between the CD-G and CD-P groups. It is well known that CD prognosis is related to age, CD location, and CD behavior at diagnosis, and these factors showed difference among the CD prognosis group (Table 2). In addition, as stool microbiome might be affected by CD activity at sampling and bowel resection state, we also considered these factors as confounding variables. Among the 13 genera and 10 species that showed different abundance between CD-G and CD-P groups by either ANCOM or LEfSe analysis (Table 2), two species and two genera consistently showed difference after adjusting for confounding variable (Figure 5). An increased level of species producta (Figure 5A) and genus Coprococcus (Figure 5D) was maintained in the CD-G group, relative to those with CD-P groups (both p-value = 0.01). Species E. coli (Figure 5B) and genus Lactobacillus (Figure 5C) were significantly enriched in the CD-P group than in the CD-G group (p = 0.001 and p = 0.002, respectively) and in the CD-I group (p = 0.03 and p = 0.02, respectively).Among other organisms that had a different abundance between CD-G and CD-P groups, Faecalibacterium prausnitzii, a well-known anti-inflammatory organism, that is considered to be a marker of health, showed no difference between CD-G and CD-P groups after adjusting for confounding variables (CD-G and CD-I, p = 0.18, CD-G vs. CD-P, p = 0.24).To understand the gut microbial functions related to the CD prognosis, we used PICRUSt2 to infer putative metagenomes from 16S rRNA gene profiles. STAMP was used to identify microbially relevant functions linked with the CD prognosis. Among the predicted MetaCyc pathways inferred by PICRUSt2 for ASVs, 95 pathways passed the significant thresholds (q < 0.05) (Table S4) and 27 pathways among them satisfied the specified filter of effect size (ratio of mean proportions of sequences between groups >2.5) (Figure S1) in CD-G and CD-P. Microbial catabolic pathways such as aromatic compound degradation (4-hydroxyphenylacetate degradation, 3-phenylpropanoate and 3-propanoate degradation to 2-oxopent-4-enoate, superpathway of phenylethylamine degradation, phenylacetate degradation I, etc.), carboxylate degradation (D-glucarate degradation I, D-galactarate degradation I, etc.), and fatty acid degradation (fatty acid beta oxidation I) were present at higher levels in the CD-P than in the CD-G. The family Enterobacteriaceae or its species E. coli were the top contributors of the catabolic pathways. Sequences for the metabolism and biosynthesis of ubiquinol (ubiguinol-7, 8, 9, 10 biosynthesis), enterobacteria common antigen (enterobacterial common antigen biosynthesis), heme (superpathway of heme biosynthesis from glycine), lipopolysaccharide (LPS) (superpathway of lipopolysaccharide biosynthesis), and ppGpp (ppGpp biosynthesis) were also increased in the CD-P group. Additionally, enzyme classification numbers (EC numbers) were predicted with PICRUSt2 and 103 EC numbers, including that of flavin reductase (NADH), glyoxylate reductase (NADP+), nitrate reductase, and formate dehydrogenase, were significantly different between good (CD-G) and poor groups (CD-P) in CD (all q-value < 0.05) (Table S5).Using a bioinformatics approach and a large multicenter cohort data, this study highlighted the fact that the prognosis of CD is characterized by unique microbial signatures. The CD-P group had lower richness with respect to the α-diversity than the CD-G and CD-I groups. Moreover, ß-diversity indicated significant differences in the composition of the gut microbial community between CD-P vs. CD-G and CD-P vs. CD-I groups. Among the 18 genera and 12 species that were differentially abundant between HC and CD groups, 13 genera and 10 species showed differences in abundance between CD-G and CD-P groups, when analyzed by either ANCOM or LEfSe. E. coli from the phylum, Proteobacteria and species producta, and Lactobacillus and Coprococcus genera from the phylum, Firmicutes consistently showed differences between CD-G and CD-P groups after adjusting for confounding variables. Functional profiling using PICRUSt2 suggested that the microbial catabolic pathways, such as aromatic compound degradation, carboxylate degradation, and fatty acid degradation were better represented in the CD-P group than in the CD-G group.The advance in DNA sequencing technology and analysis have set the stage for investigations into the IBD microbiome. Recent studies consistently report a decrease in diversity along with a decreased representation of several taxa such as Bacteroides, and Clostridia, and an increase in the Gammaproteobacteria, and the presence of E. coli or Fusobacterium when compared to HC [21]. However, beyond the association between microbiome and IBD pathogenesis, the role of the microbiota at various phases of the disease or for the complicated disease course of CD is less well appreciated. Dovrolis et al. [22] reported that the α-diversity of the microbiota was decreased within all 3 CD behavior phenotypes (B1, B2, and B3) vs. HC, with more significant reductions in B2 and B3 compared with B1. Microbial composition was similar in B2 and B3 samples and it was different from those of B1 and HC. An abundance analysis of microbial families identified significant differences between the B2 and B3 and the B1 phenotype. Solok et al. [23] reported that when compared with a non-recurrence setting, endoscopic recurrence is associated with strong changes in the ileal mucosa-associated microbiota, that are highly reminiscent of those observed generally in ileal CD, with a reduction in the alpha diversity, an increase in the abundance of several members of the phylum, Proteobacteria, and a decrease in the abundance of several members of the Lachnospiraceae and the Ruminococcaceae families within the phylum, Firmicutes.In our study, we categorized patients with CD into three groups according to their prognosis, and the CD-P group had lower richness than the CD-G and CD-I groups and significant differences in the gut microbial community composition when compared to the CD-G and CD-I groups. Within the Firmicutes phylum, bacteria from the order, Clostridiales and notably those from the Lachnospiraceae families, including Coprococcus and Blautia producta, have decreased abundance in patients with poor prognosis. In the previous studies, bacteria of the Firmicutes phylum belonging to the Clostridiales order and the Lachnospiraceae families was frequently reported as having decreased abundance in CD compared with that in HC [24], especially in the naïve patients with CD [21]. In addition, as a decreased level of the Lachnospiraceae families was associated with post op recurrence [23], it might be associated with not only pathogenesis, but also the progression of CD. In this study, we could specify the genus (Coprococcus) and species (Blautia producta) level among the Lachnospiraceae families. However, within the Firmicutes phylum, members of the Bacilli class, such as Lactobacillus genera, increased in patients with poor prognosis. This was in contradiction to our expectation, as Lactobacillus is well known to be a protective bacterium in IBD, by the down regulating of inflammatory cytokines in gut mucosa [25]. The increased levels of Lactobacillus were unlikely due to the intake of probiotic, as we enrolled the patients after discontinuing the probiotics, and certain studies have shown an increase in the Lactobacillus in IBD patients [26,27]. As it also increased in B2, and B3 phenotype than B1 [22], the controversies on this bacterium and its role in the pathogenesis of IBD should be further investigated.It is well known that the relative proportions of Proteobacteria, including E. coli, are increased in the mucosa of IBD patients [21,28,29]. Among the E. coli strains, adherent invasive E. coli (AIEC) has the ability to adhere to and invade intestinal epithelial cells (IEC), as well as to replicate within macrophages [30]. Several molecular and culture-based investigations support a putative role of AIEC in CD [31]. In addition, there are some data supporting the role of E. coli infection in regulating the activity and severity of CD. Mylonaki et al. [32] demonstrated an increased number of E. coli in the epithelium and lamina propria of patients with active CD in comparison to their number in inactive CD. Elliott et al. [33] found an association between the abundance of mucosa-associated E. coli and the severity of endoscopic inflammation, and several studies have associated the presence of AIEC with ileal recurrence after surgery for CD. As E. coli was increased in the CD-P group after adjusting these confounding factors, such as disease location, behavior, activity and surgery, the results of the present study corresponded with the results of previous studies and suggested an independent association of E. coli with CD prognosis.Our PICRUSt2 results identify metabolic signatures associated to inflammation-associated dysbiosis and might provide insight about the role of selective growth of E. coli in CD. Firstly, the poor prognosis group was associated with increased microbial catabolic pathways, and the family Enterobacteriaceae or its species E. coli was the top contributor to the catabolic pathways. Inflammation is a heavily energy-consuming process, as the activation, proliferation and recruitment of cells require energy-enriched substrates involved in inflammation and immunity [34]. Bacteria form a cell-wall containing organisms and require large amounts of carbohydrates during growth for the biosynthesis of complex structural polysaccharide. During inflamed status, available carbohydrates are absent and the glyoxylate cycle permits the synthesis of glucose from lipids via acetate generated in fatty acid β-oxidation. We identified that several compound degradation pathways were over-represented in the CD-P group. These pathways ultimately yield precursor metabolites, which is further degraded to pyruvate and acetyl-CoA to enter into the TCA cycle. E. coli is a prototropic facultative anaerobe which can use these compounds as the sole source of carbon for growth, and has the ability to respire oxygen, use alternative anaerobic electron acceptors, or ferment [35]. Thus, one of the potential mechanisms by which E. coli survives, despite the inflammatory environment, is their growth and anaerobic respiration by the utilization of stable reactive oxygen and nitrogen produced as by-products of the host inflammatory response [36]. Second, we found that pathways related to enterobacterial common antigen and LPS biosynthesis were highly represented in the CD-P group in comparison to their representation in the CD-G group. High levels of E. coli colonization in the gut are correlated with high concentrations of its LPS, which activates the host immune system. LPS is recognized by toll-like-receptor (TLR) 4, and TLRs are upregulated at inflamed tissue sites [37]. Additionally, gram-negative bacteria have several strategies to defend themselves from polymyxin antibiotics (polymyxin B and colistin), including a variety of LPS modifications. The polymyxin resistance pathway was highly detected in the CD-P group. Third, the enterobactin pathway was highly detected in the CD-P group, and a previous study showed that enterobactin, a catecholate siderophore secreted by E. coli, inhibits the activity of the neutrophil bactericidal enzyme and myeloperoxidase, both in vitro, hence promoting the survival of E. coli in the inflammatory environment [38].We demonstrated, for the first time, that microbiome associated to the poor prognosis group is different from that associated to the good or intermediate prognosis group in CD by the large multicenter cohort. In addition to the organisms previously implicated in CD, including the Lachnospiraceae and Enterobacteriaceae families, we identified that organisms belonging to the genus Coprococcus, and Blautia of Lachnospiraceae family and E. coli of Enterobacteriaceae family showed difference in abundance between the different CD prognosis groups. Although the analysis dealing with covariates resulted in only a short list of taxa significantly associated with CD prognosis, these taxa might be independently associated with CD prognosis, in addition to clinical factors. In agreement with previous studies on CD cohorts from western countries, the decrease of Lachnospiraceae family and increased of E. coli was consistently observed in our study in Korean patients with CD, although NOD2 and ATG16L1 mutations, which are related with bacterial handling, are not frequently observed in Korean patients with CD [39]. These results can be helpful to understand the pathogenesis of CD in different genetic and epidemiologic backgrounds. Finally, our PICRUSt2 results provide insight about the mechanisms of selective growth of E. coli in patients with CD, that E. coli might not only be causally involved in patients with CD, but may have also merely adapted to live in this affected environment.This study has several limitations. First, the retrospective, cross-sectional nature of this study precludes definite conclusions as to whether the changes of intestinal microbial composition according to the CD prognosis are primary or secondary events. So far, an informative study of pediatric patients with CDconfirms the association between the disease state, alterations to bacterial taxa and reduction in diversity in treatment-naive CD cohort [21]. Second, we collected stool sample only one time at enrollment. However, bacteria, such as E. coli, contributed to a large number of shifts in outcome of IBD [40]. In order to validate these results, future work will be needed to consistently collect samples from patients over a period of extensive disease durations. Third, in a new-onset CD study, microbial balance is less shifted toward a dysbiotic state in the lumen, supporting the need to examine mucosal tissue in addition to stool samples [21]. However, the large-scale collection of stool samples is an even less invasive approach, and some bacteria such as those belonging to the Escherichia/Shigella genus were predominantly observed in both mucosal tissues and stool samples of patients with CDcompared to that in HC [24]. Large studies involving gut microbiome profiling should incorporate mucosal microbiome sampling in addition to fecal sampling in the future, and should compare their effectiveness as a biomarker.In conclusion, CD prognosis is associated with changes in the composition of the microbiota and decreased diversity. Further studies of a prospective nature and a longitudinal design are needed to confirm that particular bacteria can be used as a predictor of disease progression.The following are available online at https://www.mdpi.com/2077-0383/9/6/1748/s1, Table S1: Significantly different taxa between the HC and CD group, Table S2: Comparisons of the alpha diversity between different groups in pairwise Kruskal-Wallis test, Table S3: Beta diversity between groups in pairwise PERMANOVA, Table S4: Predicted MetaCyc pathway abundances significantly associated (based on Welch’s t-test [STAMP]) with severity of Crohn’s disease (good vs. poor), Table S5: Predicted enzyme classification (EC) gene families significantly associated (based on Welch’s t-test [STAMP]) with severity of Crohn’s disease (good vs. poor), Figure S1: Prediction of the correlation of metagenome functional content with Crohn’s disease prognosis using PICRUSt.D.I.P. planned the study, D.I.P., S.-K.P., C.H.C., J.P.I., J.M.C., C.S.E., T.-O.K., S.-B.K., K.B.B., H.G.K., Y.J., H.Y., H.-L.K., D.-S.H. and C.-W.L. collected data, S.-K.P., H.-N.K. and K.A. interpreted the manuscript, S.-K.P. and H.-N.K. drafted the manuscript. All authors have read and agree to the published version of the manuscript.This study was supported by a National Research Foundation (NRF) grant funded by the Korea government (NRF-2017R1A2B2009596).We gratefully thank Eun Sun Kim (Korea University), Jae Jun Park (Yonsei University) and Ja Seol Koo (Korea University) for collecting data.The authors declare no conflict of interest.Enrollment of subjects.Alpha diversity comparisons of gut microbial communities among groups, according to the Crohn’s disease prognosis. (A) Observed ASVs, (B) Faith’s phylogenetic diversity, (C) Shannon index. * q < 0.05 (pairwise Kruskal-Wallis test, Benjamini-Hochberg correction). CD-G = good prognosis group; CD-I = intermediate prognosis group, CD-P = poor prognosis group; HC = health control.Beta diversity comparisons of gut microbial communities among groups, according to the Crohn’s disease prognosis. (A) Unweighted UniFrac distance by 3D principal coordinates analysis (PCoA) plot, (B) Unweighted UniFrac distance by boxplot, (C) Weighted UniFrac distance by PCoA plot, (D) Weighted UniFrac distance by boxplot. The y-axes represent the distance of each group to the CD-P group (baseline). * q < 0.05 (pairwise PERMANOVA); CD-G = good prognosis group; CD-I = intermediate prognosis group; CD-P = poor prognosis group; HC = health control.Differentially abundant bacterial taxa between good prognosis (CD-G) and poor prognosis (CD-P) groups in Crohn’s disease. (A) A forest plot showing the LDA score indicating significant differences in the bacterial taxa between the CD-G (red) and CD-P (green) groups (LDA score >4.0; p < 0.05). (B) Cladogram generated using the LEfSe method, indicating the phylogenetic distribution of microbes associated with the CD-G and CD-P groups. CD-G = good prognosis group; CD-P = poor prognosis group; LDA = linear discriminant analysis.Two species (A,B) and two genera (C,D) consistently showed differences between good prognosis and poor prognosis groups after adjusting for confounding variables, when analyzed by the generalized linear models using MaAsLin packages. * p < 0.05 CD-G = good prognosis group; CD-I = intermediate prognosis group; CD-P = poor prognosis group; HC = health control.Baseline characteristics of the patients with Crohn’s disease (CD) patients.a Age at diagnosis, location and behavior were according to the Montreal classification. b CDAI, Location and behavior at sampling date. CDAI <150: remission, 150‒220: mild, >221: moderate to severe. c Medication included both past and current medication. d Others included vedolizimab, risankizumab, etrolizumab, ustekinumab, mongersen, PF-00547659 (anti MAdCAM-1 antibody) and JNJ-64304500, (NKG2D receptor blocker). ADA = adalimumab; ASA = aminosalicylic acid; BMI = body mass index; CDAI = Crohn’s disease activity index; CD-G = good prognosis group; CD-I = intermediate prognosis group; CD-P = poor prognosis group; IFN = infliximab.Significantly different taxa between the CD-G and CD-P groups.a number of genus: 314, number of species: 459. b W by ANCOM analysis. c The group showing high abundance, by LDA effect size (LEfSe) analysis. d LDA score by LEfSe analysis. c_ = class; CD-G = good prognosis group; CD-P = poor prognosis group; f_ = family; g_ = genus; LDA = linear discriminant analysis; ns = not significant; o_ = order; p_ = phylum; s_ = specie.

Med-MDPI/jcm_2/jcm-09-06-01749.txt

@@ -0,0 +1,3 @@
+The study described in this paper was conducted to assess the short-term outcomes of intra-articular administration of hyaluronic acid in patients with symptoms of temporomandibular joint disorders. A group of 40 patients suffering from temporomandibular joint disorders underwent a series of hyaluronic acid intra-articular injections. Questionnaires and clinical examinations were conducted to assess stress exposure of the subjects and to evaluate short-term treatment outcomes, i.e., reducing joint and muscle pain and increasing the mobility of the mandible. A weak positive correlation between stress exposure and pain was observed. As a result of treatment, 61% of subjects revealed a total reduction of muscle pain, while joint pain completely resolved in 88% of patients. Mandibular mobility increased by 11%, 31%, 9%, and 11% regarding opening, protrusive, and lateral right and left movements, respectively. The study confirms the short-term effectiveness of intra-articular administration of hyaluronic acid on reducing joint and muscle pain in patients with articular disc displacement. The treatment positively affected the mobility of the mandible in all directions. The verification of late treatment effects of hyaluronic acid viscosupplementation requires the continuation of the research.Temporomandibular joints (TMJs) belong to the category of synovial joints in the human body, where joint surfaces are lubricated by the synovial fluid produced by the synovial membrane lining the joint capsule. The synovial fluid consists mainly of hyaluronic acid (HA)—an anionic, nonsulfated glycosaminoglycan and lubricin—a surface-active mucinous glycoprotein.An appropriate concentration of HA in the synovial fluid is crucial for proper tissue elasticity and for reducing friction between the joint surfaces of the bone and the synovial disc. In this way, the viscoelastic properties of the synovial fluid protect the chondrocytes against mechanical damage caused by excessive pressure. HA is also responsible for suppressing stresses generated inside the joint when subjected to trauma. In addition, HA has antioxidant properties, reducing the amount of free radicals present in the joint cavity. HA also facilitates the penetration of nutrients from the blood through joint mucus into the joint cartilage. Finally, by interacting with collagen, HA forms a barrier against microorganisms and toxins.Viscosupplementation, i.e., supplementation with a component of synovial fluid in the form of intra-articular injections, is one of the possible solutions to reduced HA levels in joint cavities in chronic inflammation, already successfully used in the treatment of knee osteoarthritis [1]. Importantly, a detailed summary of 24 reports on 6 HA-based preparations used in the viscosupplementation of temporomandibular joints showed no serious adverse events [2].Temporomandibular disorders (TMD) and related pain are becoming an increasingly common social problem. Patients suffering from TMJ pain seek help from specialists in various fields of medicine, including prosthodontists, orthodontists, neurologists, and physiotherapists. Importantly, the effects of monotherapies are often insufficient due to the interdisciplinary nature of the problem. Given the unsatisfactory results of current medical solutions and the scarcity of research on viscosupplementation with HA (usually on small groups of patients), in this study, we decided to investigate the effects of this therapy in patients with TMD.This study was conducted in accordance with the Declaration of Helsinki. All procedures involving human subjects were approved by the Ethics Committee of the Pomeranian Medical University in Szczecin (Approval No. KB-0012/230/11/18). Written informed consent was obtained from all subjects. The study included 40 adult patients with symptoms of masticatory system disorders, who qualified for intra-articular HA injections in accordance with the guidelines published by Escoda-Francoli et al. [3] and developed on the basis of a meta-analysis of research on viscosupplementation of the temporomandibular joint published between 1966 and 2008. By adopting these criteria, our study included patients who were diagnosed with (i) disc displacement without reduction, (ii) disc displacement with reduction, and (iii) degenerative joint diseases. The study included patients who had previously been excluded from known somatic diseases, whose coexistence may affect the assessment of TMJ [4]. It was also ensured that patients did not take medicines that could conceal TMJ symptoms.All patients who qualified for the study complained of pain, which was diagnosed as associated with TMJ dysfunction. The therapeutic regimen proposed by Abouelhuda et al. [5] was adopted. Patients who had no pain relief in the subsequent stages of the therapeutic regimen were included in the study. These stages were (i) pharmacological treatment, (ii) physiotherapeutic procedures, and (iii) splint therapy. In cases of contraindications or the patient’s disagreement with any of the types of treatment, it was not implemented. In none of the cases prior to the study was any arthrocentesis or arthroscopy performed. During HA injection therapy, the patients were not treated with anything other than painkillers (paracetamol, tramadol). Thus, between HA injections, patients were not treated with night splint, arthrocentesis, or arthroscopy. Patients were strongly advised to use painkillers only when necessary. No regular analgesic pharmacotherapy was performed during the study. Prior to individual follow-up visits, patients were not allowed to take painkillers during periods when their use could affect the examination result.The group of 40 patients consisted of 36 women and 4 men, aged 18–69 years (43 years on average). The duration of symptoms before inclusion in the study ranged from 1 month to 20 years (the median was 11 months). In addition, 95% of respondents reported exposure to psychological stresses of varying intensity. According to the protocol of the International Network for Orofacial Pain and Related Disorders Methodology [6], the appropriate questions from the questionnaire by Dworkin and LeResche [7] were translated into Polish and used to assess the intensity of stress.The questions asked the patients concerning the indicators of stress—somatic and psychological sensations in the last month. According to Dworkin and LeResche [7], somatic sensation includes, among others, tightness in throat and stomach ache. Psychological indicators of stress include feelings of guilt, entrapment, or loneliness. The intensity of each of the 32 sensations was assessed on a five-point scale from 0 to 4. After summing up the points, the number of points is divided by the number of questions and the collective result of stress exposure for a given patient is determined. The results were interpreted as follows: 0–1—no stress, 1–2—minimal stress, 2–3—moderate stress, and 3–4—severe stress. The characteristics of the study group are presented in Table 1.The study group was divided according to the diagnostic criteria for TMD (DC/TMD) proposed by Schiffman et al. in 2014 [8] (Table 2).We also used another questionnaire developed by a team of two orthodontists, two maxillofacial surgeons, and two prosthetic surgeons to evaluate the objective and subjective symptoms of TMD. It consists of manual functional analysis of TMD according to Bumann et al. [9] and palpation of head and neck muscles according to the protocol used by Felicio Festa and described by Tecco et al. in 2011 [10]. For the purpose of this paper, we used the Polish version of the muscular system examination developed by Czerwińska-Niezabitowska and Kulesa-Mrowiecka [11]. These techniques were used to evaluate joint and muscle pain, characterize the mandibular pathway during opening, and determine the amplitudes of mandibular movements in 3 planes. A preliminary examination using our own questionnaire was performed immediately before the first HA injection. During the qualification of patients for the study, the TMJ pain treatment performed so far was taken into account, according to the therapeutic ladder proposed by Abouelhud et al. [5]. Its elements are, sequentially, pharmacotherapy, occlusal splint therapy, intra-articular injections, physiotherapy, arthrocentesis, arthroscopy, and open TMJ surgery. Our study excluded those patients who had received any of these therapies during the three months preceding the study, except for “adhoc” analgesic pharmacotherapy without anti-inflammatory components. We have also excluded from this study the patients with local contraindications to intra-articular injections, according to the results of the review by Soni et al. [12]. These are abscess, inflammation or tumor of the skin, connective tissue or bone of the puncture site, bleeding diathesis or acquired coagulopathies, blood infections, and malignant tumors. Patients were re-examined at the last visit, 5–7 weeks after the first injection, which gave two dated and fully completed questionnaires for each patient.The number and frequency of intra-articular injections were based on the recommendations of other researchers. All publications known to the authors of this paper proposed a series of up to 5 intra-articular HA injections. These observations are consistent with the analysis of literature on the administration of drugs to the temporomandibular joint cavity conducted 10 years ago by Mountziaris et al. [13], who identified two most commonly used HA viscosupplementation schemes: 2 injections at 7–14-day intervals or 5 injections at 7-day intervals. In our study, we adopted the regimen of 5 intra-articular HA injections at intervals of 7 to 10 days. The duration of intervals between these injections depended on the availability of patients.For the purpose of this study, a protocol was developed for each of the 5 visits. On the first one, subjective and physical examinations were carried out on the basis of a questionnaire to qualify the patient for intra-articular injections. Viscosupplementation with HA was performed on each of 5 visits. During the last (fifth) visit, the injection was followed by subjective and physical examinations based on the questionnaire. Those patients who did not come for the fifth appointment—and so were not surveyed twice—were not included in the study. This inclusion criterion was met by 40 patients.Out of the group of 40 patients who were present at the first and fifth appointments, 27 did not miss any of the appointments and received 5 HA injections. Ten patients canceled one of the visits (visits number 2, 3, or 4) and were given four intra-articular injections. The remaining three patients were only present at three appointments, and so received only three HA injections. The total absenteeism rate was 8%. In most cases, it resulted from unexpected events or satisfactory effects of treatment, which reduced the motivation to attend all five procedures.The puncture site was determined in accordance with the protocol proposed by O’Connor et al. [14]. After depressing the mandible, on the line joining the lateral canthus of the eye with the cutaneous point tragus, we established a point 10 mm forward from the latter, and then the puncture site was determined by descending 2 mm below the point, perpendicularly to the described line. This location of the puncture site guaranteed the HA injection to the upper compartment of the joint. The skin was disinfected each time with a propanol-based preparation (Kodan Tinktur Forte) and the injection was performed when it was completely dry. No local anesthesia was used. All the injections were performed by the same maxillofacial surgeon. HA was administered at a dose of 0.4 mL of 2% hyaluronic acid (Synocrom) to one joint. In 19 patients, unilateral viscosupplementation was performed; in 21 patients, a bilateral injection was performed due to bilateral ailments.The pain was determined on the basis of a clinical trial. Its presence was recorded for each of the patients in the preliminary and final questionnaires. Then, according to the protocol proposed by Skeie et al. [15], the pain was divided into muscular pain and joint pain, using a simple YES/NO questionnaire and palpation of head and neck muscles. Joint pain was assessed during the examination of the joint surface, joint capsule, and ligaments. This test was carried out in accordance with the manual functional analysis of masticatory system disorders. The joint surface was examined by dynamic protrusion compression and dynamic medial and lateral translations. The examination of the joint capsule and ligaments consisted of passive compression, stretching, and translation tests.The opening pathway was evaluated in the preliminary test and after viscosupplementation. It is shown in a simplified Farrar’s diagram according to the protocol presented by Gorzałek and Kulesa-Mrowiecka [16]. In our study, anomalies of the opening pathway of less than 4 mm, called deviation by Okeson and Grocholewicz [17], were classified as an S-shaped mandibular opening pathway. Anomalies greater than or equal to 4 mm—permanent alteration, according to Okeson and Grocholewicz [17]—were described as deflection.In the study group, muscle pains were present in 36 patients before viscosupplementation and 14 patients after the treatment, which means a 61% effectiveness of the therapy in this regard (i.e., in 22 patients). Joint pain was present in 25 patients before the HA injections. At the end of the therapy, only 3 patients complained about this type of pain—it was not reported by 22 out of 25 (88%) patients. These data are presented in Table 3.In 23 patients, muscle and joint pain coexisted before injection. The presence of both types of ailments after viscosupplementation was found in one patient. In 22 out of 23 people (96%), at least one of the types of pain disappeared. A person in whom both muscular and joint pain were present after the end of HA therapy was a 34-year-old woman who had reported the presence of pain in the right temporomandibular joint for about a month before the start of HA injections. She had not been treated previously for temporomandibular joint ailments. That patient reported no concurrent diseases, and on the basis of a questionnaire, we determined exposure only to moderate stress (approximately 2.5 points on a scale of 0–4). During the clinical examination, the patient was diagnosed with a dislocation of the joint disc with reduction, characterized by clearly limited mobility of the mandible: the opening amplitude was 32 mm, the protrusive and lateral movements were about 2 mm. HA viscosupplementation did not improve mandible mobility. The patient was qualified for further diagnostics, manual therapy, and possible treatment with the use of a repositioning splint.The analysis of diagrams showing the movement of the mandible in the coronal plane allowed us to assess the influence of HA viscosupplementation on the opening pathway. In 5 patients, the anomalies of mandibular movement in the coronal plane did not exceed 4 mm. During the initial examination, the remaining 35 patients had pathological opening pathways (88% of the study group). Of these 35 subjects, an S-shaped opening pathway occurred in 26 patients and deflection in 9 patients. After completion of the intra-articular HA injection therapy, the mandible mobility in the frontal plane was normalized in 22 out of 35 patients (63%). Out of 26 patients with the S-shaped pathway, 16 (62%) patients showed satisfactory improvement at the end of treatment, and the remaining 10 (38%) patients maintained an alteration of at least 4 mm. In the group of 9 persons diagnosed with deflection, the therapy resulted in 6 (67%) patients obtaining a normal opening pathway during the final examination, while the remaining 3 patients (33%) showed no significant improvement.The treatment resulted in an increase in the amplitude of mandibular movements in all planes. The mean opening movement in the study group was 40.1 mm. After HA injections, it increased by 4.5 to 44.6 mm, i.e., by 11%. The applied therapy had the greatest influence on the protrusion, which increased from 5.4 to 7.2 mm (by 31%). The range of lateral movement was 7.8 mm to the right and 8.0 mm to the left before the start of HA injections. It improved by 0.7 (by 9%) and 0.9 mm (by 11%), respectively, reaching 8.5 and 8.9 mm. Negative correlations were observed between the differences in the initial and final amplitudes and the initial amplitudes of individual extreme movements of the mandible. They indicate a weaker therapeutic effect of HA injections in cases of high initial mandibular mobility. The data discussed are presented in Table 4 and Figure 1, Figure 2, Figure 3 and Figure 4.A correlation was also observed between the presence of pain before treatment with HA and the severity of stress experienced by patients. For the purpose of this report, pain was rated on a scale from 0 to 2, where 0 meant no pain, 1—muscle or joint pain, and 2—concurrent muscle and joint pain. Exposure to stress was determined on the previously mentioned scale from 0 to 4, resulting from an interview based on the questionnaire by Dworkin and LeResche [7]. The patients’ experiences from the period of 30 days preceding the initial examination were taken into account. Pearson’s correlation coefficient for stress and pain was 0.3, indicating a clear tendency for pain to increase in exposure to stress.As indicated by Panek and Maślanka [18], as well as by Kurpiel and Kostrzewa-Janicka [19], the terminology and classification of pathologies concerning temporomandibular joints have significantly evolved since the first written reports on the subject, i.e., since 1920. Despite the fact that the authors of various divisions were convinced about the relevance of their classifications, each of them was eventually replaced by a new one. An ideal classification should take into account anatomical and functional aspects, be clinically useful, support the physician in making therapeutic decisions, and at the same time, be concise and simple enough to be used in everyday practice. Among the many divisions, the closest to this ideal are the International Classification of Orofacial Pain (ICOP) [20] and DC/TMD [8]. As the study was conducted prior to the publication of the ICOP, we followed the guidelines proposed by Schiffman et al. under the DC/TMD protocols.Painful ailments resulting from masticatory system disorders have a very complex nature. Harper et al. [21] attempted to analyze the mechanisms responsible for the pain associated with masticatory system disorders, distinguishing central and peripheral pain, and dividing the latter into nociceptive and neuropathic pain. In their conclusions, they emphasized that the cause of pain is no less important than the mechanism of its manifestation. Depending on the cause, the pain accompanying TMJ disorders can be divided into muscle and joint pain. Gorzałek and Kulesa-Mrowiecka [16] noted that isolated muscle pain restricts the lateral movements of the mandible without affecting its opening. This type of pain results from contractions and increased tension in the chewing muscles. According to Gorzałek and Kulesa-Mrowiecka [16], joint pain is associated with reduced opening and reduced amplitude of lateral movements. During the free opening, the mandible deviates towards the affected side.Meticulous diagnostics of pain etiology allows us to take optimal therapeutic action. The method of treatment of muscle pain was proposed and examined by Pihut et al. [22]. These authors carried out intramuscular injections of botulinum toxin in 42 patients suffering from masseter muscle pain. The therapy resulted in a significant reduction in the incidence and severity of pain, which was also reflected in a reduction in the number of analgesics taken by patients. The intensity of pain of muscular origin was reduced in each of the patients.In our study group, out of 36 people complaining of muscle pain, only 61% reported its total disappearance following HA viscosupplementation. This significantly lower therapeutic efficacy of intra-articular HA injections compared to intramuscular botulinum toxin administration is associated with the fact that intra-articular therapy only has an indirect effect on the masticatory muscle system. Similar effectiveness of HA viscosupplementation in the treatment of muscle pain was observed by Pihut et al. [23], where, out of 24 patients complaining of muscular pain, 71% stated that muscular pain subsided as a result of the therapy. Those authors compared the effectiveness of HA viscosupplementation with the results of intra-articular platelet-rich plasma (PRP) injections. Out of 25 patients treated with PRP, muscle pain disappeared in 68%. Similar results may indicate similar indirect efficacy of intra-articular injections with HA and PRP in the treatment of muscle pain.With regard to joint pain, our results showed 88% effectiveness of HA viscosupplementation. It is difficult to put these results in perspective because, in the available literature, only Pihut et al. [23] identified joint pain as an isolated component of TMD, with pain relief experienced by 17 out of 22 patients (77%). Even given some differences in the method of establishing the presence of pain between Pihut et al. [23] and our study, HA viscosupplementation seems to be more effective in reducing joint pain than muscle pain.In the treatment of pain resulting from TMD, intra-articular HA injections are used as a viable alternative to rinsing the joint cavity. However, the lack of a uniform protocol of management in TMD patients has resulted in various modifications of the treatment. These two methods are proposed to be used separately, and there are also recommendations for combining them. De Riu et al. [24] examined the effectiveness of intra-articular 2-mL HA administration preceded by rinsing the joint cavity with about 350 mL of saline. Pain on the VAS scale dropped from 8.26 to 2.03, which means 75% effectiveness of the therapy in 30 patients. However, that study did not distinguish between muscle and joint pain, in a typically synthetic approach that dominates the research on the effectiveness of TMD treatment with joint rinsing and viscosupplementation. The study by Gurung et al. [25] on the effectiveness of intra-articular HA injections as an additional procedure performed after rinsing the joint cavity is no exception. In the studied group of 10 patients, the pain intensity expressed on the VAS scale decreased from 5.9 to 1.3, i.e., by 78%, compared to a 56% pain reduction following joint rinsing only. In our study, a cumulative assessment of muscle and joint pain following HA therapy showed a 58% reduction in pain.Widening the joint cavity by the mechanical displacement of the condylar processes downwards is also possible to achieve by conservative methods using various types of occlusal splints. However, their therapeutic effect is difficult to evaluate. Raphael and Marbach [26], in a study of 63 patients diagnosed with face and myofascial pain, did not observe any differences in the effectiveness of splint therapy compared to placebo. Pihut et al. [27] proposed intra-articular injections of rich platelet plasma in the case of ineffectiveness of occlusal splints used for joint disc displacement or increased chewing muscle tension. In the study conducted by those authors, good results were achieved by viscosupplementation of platelet-rich plasma in patients in whom splints had proved ineffective.In our own study group, there were 6 patients whose complaints persisted as a result of previous therapies, and the effectiveness of viscosupplementation with HA in terms of pain relief was 83%. In this context, it is worth considering splint therapy as a preliminary treatment in sudden pain and unknown etiology. This approach was used by Shoush et al. [28], who compared the effectiveness of occlusal splints and therapeutic exercises. Those authors examined their effectiveness in terms of pain relief and normalization of mandibular opening amplitude in two 56-person groups of patients treated for 6 weeks. The first group used standard occlusal splints during the day. Patients from the second group took part in a series of 15-min exercises twice a week. The exercise session consisted of two parts. The first one included exercises proposed by Kijak et al. [29], similar to those developed by Gerry and described by Czerwińska-Niezabitowska and Kulesa-Mrowiecka [11]. The second part consisted of stretching the masseter and the medial pterygoid, according to Okeson’s protocol [30]. In the group using exercises, the efficacy of pain relief was improved by 9% and the amplitude normalization by 14%.Akbulut et al. [31] showed much higher effectiveness of occlusal splint therapy, which resulted in pain relief and a significant increase in mandibular opening in 88% patients (n = 25). However, those authors admitted that during the first three months, they did not observe any significant effects of treatment with occlusal splints. It was only after 12 months of observation that the 88% success rate of the therapy was determined.Akbulut et al. [31] defined the term “total healing” of TMD as simultaneous elimination of pain and normalization of opening amplitude. In their evaluation, however, they omitted an important indicator—the opening pathway. Our observations on the effect of HA viscosupplementation on the opening pathway in the coronal plane may be compared with the only study describing this parameter [23]. In our study, it improved in 63% of 35 patients, while Pihut et al. [23] described the normalization of this parameter in as many as 82% of 22 patients following intra-articular HA injections. These results clearly indicate a significant positive influence of HA viscosupplementation on the mandibular opening pathway. The observed differences may have resulted, among others, from different subjective methods of assessment of the opening pathway.The method proposed by Kijak et al. [32]—based on a digital facial arch—offers a possibility to objectify the assessment of mandible mobility. The method includes a detailed analysis of the pathway of articular heads, which, in the future, may become a perfect complement to the imaging and manual functional analysis of TMD that we used in our examination. Using precise digital measurements, Kijak et al. [32] also determined the mean amplitudes of the mandibular opening. For the healthy group, the researchers calculated the average value of 45.6 mm. In 76 patients with diagnosed TMDs, the mean opening amplitude was 37.6 mm. In our group of patients, this value was 40.1 mm before the start of treatment. After the completion of the intra-articular therapy, HA increased to 44.6 mm, i.e., by 11%, which means that the final value was similar to the physiological value calculated by Kijak et al. [32].Lewandowski [33] also showed a beneficial effect of HA viscosupplementation on the mandibular mobility assessed by the amplitude of its opening. However, the following years did not bring many studies on the impact of HA injections in this regard. Fonseca et al. [34] described a group of 10 cases of patients diagnosed with TMD. Mean amplitudes of mandibular opening before and after HA therapy increased from 30 and 37 mm, respectively, which allows us to determine the improvement of mandibular dislocation by 23%. This means that the HA viscosupplementation calculated in the analysis by Fonseca et al. [34] is more than twice as effective as in our study group. This discrepancy can be easily explained by the strong correlation between the difference between the initial and final amplitudes and the final amplitude of the mandibular opening. This correlation was calculated on the basis of data from our own study group and expressed as Pearson’s coefficient, r = −0.68. The significantly higher HA viscosupplementation efficiency for lower initial mandibular opening amplitudes in our study is shown in Figure 5.The trend line shown in Figure 5 above can be represented by the following formula:y = −0.4 × x + 18.7 mm(1)
+      where x is the initial amplitude of the mandibular opening and y is its increase following HA viscosupplementation. The mean final opening calculated using this formula would be 36.7 mm in the group of patients examined by Fonseca et al. [34], which is consistent with the actual posttreatment value of 37 mm presented in their study.The improvement of mandibular mobility expressed as an increase in the amplitude of opening was also determined in groups treated with joint rinsing. In an analysis of 7 publications by the authors listed below, the relationship between the initial value of mandibular excitation and its increase in the course of therapy was also observed for joint cavity rinsing. It may be presented by the following formula:y = −0.4 × x + 23.0 mm(2)
+      where x is the initial amplitude of mandibular opening, and y is an increase in the amplitude of the opening following joint lavage. In the discussed material, the final amplitudes of mandibular opening after joint cavity rinsing were, on average, as much as 5 mm higher than the expected results of treatment of the same group of patients who were treated only with intra-articular HA injections. The results of treatment by rinsing the joint cavities and HA viscosupplementation are presented in Table 5.Gouveia et al. [41] demonstrated a strong correlation between the range of mandibular opening and satisfaction of patients with the quality of mastication. In our study, due to the complexity of mastication, the mobility of the mandible was evaluated by examination in three planes. HA injections improved the function of the mandible not only in terms of the mandibular opening—the amplitudes of protrusive and lateral movements also increased.While the influence of HA viscosupplementation on the mandibular opening has been confirmed by numerous authors, data on the activity of the mandible in the transverse plane are very limited. These are presented by Chandrashekhar et al. [37], based on a study of 50 patients in whom they performed joint cavity rinsing with Ringer’s solution. Before treatment, the mean right lateral motion of the mandible was 7.2 mm. The maximum amplitude of the opposite direction was 7.6 mm on average. The mandibular mobility in patients treated by Chandrashekhar et al. [37] improved by 33% and 23%, taking into account right- and left-hand movements, respectively. In our study group, the improvement of mandibular function was 9% to the right and 11% to the left. Although scarce literature data do not allow us to formulate general conclusions, however, a clear difference in the effectiveness of joint lavage and HA viscosupplementation seems to be in favor of the former method. This would be in line with the analysis of the data already carried out on the increase in the amplitude of the mandibular opening.The possibility of combining joint rinsing with HA viscosupplementation should also be considered. In a unique study by Gurung et al. [25], the highest efficacy in improving chewing function by using this combination of therapies was demonstrated. The amplitude of mandibular opening in the group of patients treated with a combination of joint rinsing and viscosupplementation was 13% higher than in the control group treated with joint rinsing alone. Moreover, the pain present during the protrusive and lateral movements of the mandible disappeared in all patients treated by both methods as early as 6 weeks of therapy. In the case of a group treated only with joint rinsing, the pain was still present in 20% of people after 12 weeks.The influence of HA viscosupplementation on the degenerative changes of the articular fossa and head is also worth mentioning. In our own examination, radiological assessment of joint surfaces was performed on the basis of orthopantomograms and functional radiographs of temporomandibular joints. Degenerative changes on these surfaces were found in 8 patients before the treatment. Among them, remodeling was observed in only one patient who had previously been diagnosed with osteophytes. In the remaining 7 patients who were diagnosed with various resorption lesions before the treatment, no noticeable improvement in the condition of joint surfaces was observed. Similar observations were presented by Sun et al. [42] who injected HA into both compartments of the temporomandibular joints of 51 patients and evaluated the effects of the therapy using cone-beam computed tomography—no improvement in the condition of joint surfaces was found, as well as no slowdown in bone pathology progression.Many authors have emphasized the significant influence of psychological components on the development of TMD. For example, in our study, we found a positive correlation between muscle and joint pains and stress (Pearson’s correlation coefficient r = 0.30), which indicates a tendency for pain to coexist with increased stress. Indirectly, this result may also prove the participation of psychological components in the process of initiation and evolution of TMD.TMD is often discussed in the context of somatic diseases, but also stress and mental disorders. For example, Czerwińska-Niezabitowska and Kulesa-Mrowiecka [11] discussed the etiology of psychogenic defects of posture and educational therapy in the treatment of TMD. According to these authors, stress is a clear causative factor of TMD. Primary emotions such as anger or anger manifest themselves as increased tension of masticatory muscles innervated by the trigeminal nerve. The motor nucleus of this nerve is connected to the limbic system by gamma loops, which predisposes to teeth clenching and may explain such phenomena as clenching, bruxism, and other oral parafunctions. Among the patients with SLE dysfunction covered by this study, 88% of individuals suffered from exposure to severe and moderate stress. A relationship between stress and TMD can also be found in other authors’ publications. For example, Augusto et al. [43], in a study conducted on a group of 586 students of medical universities, showed a statistically significant relationship between TMD and parafunctions, stress, and mental illnesses. Similar observations were made by Ahuja et al. [44] who, on the basis of a study of a group of 450 people, found stress to be an important causative factor of TMD.According to the results of our own research and the literature analysis, the high short-term effectiveness of HA viscosupplementation should be taken into consideration when treating pain and functional limitations related to TMD. Although it was most effective in reducing joint pain, good results were also observed for muscle pain. HA viscosupplementation positively influenced the mandibular opening pathway and amplitude. It also increased mandibular mobility in other directions. The verification of late treatment effects of hyaluronic acid viscosupplementation requires a continuation of the research.Conceptualization: M.S. (Maciej Sikora); data curation: M.S. (Maciej Sikora); investigation: M.S. (Maciej Sikora), B.C.-N., M.A.C., and M.S. (Marcin Sielski); writing—original draft: M.S. (Maciej Sikora), B.C.-N., M.A.C., and M.S. (Marcin Sielski); writing—review and editing: M.S. (Maciej Sikora) and D.C.; supervision: D.C.; funding acquisition: D.C.; overall responsibility: M.S. (Maciej Sikora) and D.C. All authors have read and agreed to the published version of the manuscript.This research received no external funding.The authors declare no conflict of interest.Mandible depression (columns represent values relevant to every patient).Mandible protrusion (columns represent values relevant to every patient).Right lateral movement of the mandible (columns represent values relevant to every patient).Left lateral movement of the mandible (columns represent values relevant to every patient).Initial amplitude of the mandibular opening and its increase following HA viscosupplementation.Characteristics of the study group.Qualification of the examined group according to the Schiffman et al. [8] classification.Presence of muscular and articular pain in the study group before and after intra-articular HA injections.Mean extreme movements of the mandible in three planes.SD—standard deviation. Correlation expressed in Pearson’s correlation coefficient (r).Summary of the initial and final amplitudes of mandibular opening in other authors’ studies.The therapy column uses abbreviations: JL—joint lavage, HA—hyaluronic acid. The expected effect of HA viscosupplementation was calculated for a given test group on the basis of the formula y = x − 0.4 × x + 18.7 mm. The expected effect of arthroscopic lavage was calculated for a given study group on the basis of the formula y = x − 0.4 × x + 23.0 mm.

Med-MDPI/jcm_2/jcm-09-06-01750.txt

@@ -0,0 +1 @@
+Poly-lactic-co-glycolic acid nanoparticles (PLGA-NPs) were approved by the Food and Drug Administration (FDA) for drug delivery in cancer. The enhanced permeability and retention (EPR) effect drives their accumulation minimizing the side effects of chemotherapeutics. Our aim was to develop a new theranostic tool for cancer diagnosis and therapy based on PLGA-NPs and to evaluate the added value of vascular endothelial growth factor (VEGF) for enhanced tumor targeting. In vitro and in vivo properties of PLGA-NPs were tested and compared with VEGF-PLGA-NPs. Dynamic light scattering (DLS) was performed to evaluate the particle size, polydispersity index (PDI), and zeta potential of both preparations. Spectroscopy was used to confirm the absorption spectra in the near-infrared (NIR). In vivo, in BALB/c mice bearing a syngeneic tumor in the right thigh, intravenously injected PLGA-NPs showed a high target-to-muscle ratio (4.2 T/M at 24 h post-injection) that increased over time, with a maximum uptake at 72 h and a retention of the NPs up to 240 h. VEGF-PLGA-NPs accumulated in tumors 1.75 times more than PLGA-NPs with a tumor-to-muscle ratio of 7.90 ± 1.61 (versus 4.49 ± 0.54 of PLGA-NPs). Our study highlights the tumor-targeting potential of PLGA-NPs for diagnostic and therapeutic applications. Such NPs can be conjugated with proteins such as VEGF to increase accumulation in tumor lesions.The latest advances in molecular imaging are closely related to the use of new tools, such as nano- or micro-particles that can be used for several applications, from detection and diagnosis to drug delivery and treatment [1]. Different nanomaterials are used to create particles with a range from 1 to 1000 nm, and so, are defined as nano-particles (NPs) [2]. They offer the advantage to deliver drugs to the target with high efficiency and low systemic toxicity [3,4]. The NPs formulated with organic polymers (polymeric NPs) are generally one of the best choices for clinical or pre-clinical use due to their favorable characteristics such as non-immunogenicity, non-toxicity, biodegradability, and biocompatibility [5]. Indeed, PLGA have been approved by the Food and Drug Administration (FDA) and the European Medicine Agency (EMA) as copolymers to deliver drugs, and today about 16 approved pharmaceuticals are based on the use of these NPs [6,7]. Their in vivo biodistribution is greatly influenced by different physical and chemical characteristics, among which size and glycolic:lactic acid ratio play a key role [8,9]. The NPs accumulate in target lesions with an active or passive mechanism. The passive mechanism is represented by the enhanced permeability and retention (EPR) effect, that allows NPs with a size in the 20–200 nm range to accumulate in cancer lesions with an impaired vasculature [10,11]. Therefore, this mechanism is of great importance when using nanotechnologies in oncology [12,13,14]. Moreover, the flexibility of PLGA-NPs offers the advantage to combine their ability to deliver drugs with the possibility to functionalize them with peptides, proteins, or imaging probes [15]. Since tumor and stromal cells produce several proangiogenic factors, such as proteins from the vascular endothelial growth factor (VEGF) family, they are usually characterized by high and irregular vascularization [16]. Therefore, targeting of either VEGF or VEGF receptor (VEGFR) can be achieved and exploited to increase PLGA accumulation in tumor lesions [17].In the present study, we investigated the possibility of using specifically designed PLGA-NPs as a tool for future theranostic applications. We selected PLGA-NPs (lactic acid:glycolic acid ratio of 50:50, average size of 100–200 nm) conjugated with a near-infrared (NIR) fluorochrome with an excitation wavelength at 780 nm and emission wavelength at 825 nm, allowing a deeper tissue penetration of fluorescence [18]. The target capacity and the pharmacokinetic of native PLGA-NPs was investigated in vivo to evaluate the tumor detection and then the retention of PLGA-NPs up to 240 h.To actively target tumor cells over-expressing VEGFR, the PLGA-NPs were loaded with a recombinant human VEGF-A165 (rhVEGF) analog by the 1-Ethyl-3-[3-dimethylaminopropyl] carbodiimide hydrochloride/N-hydroxysuccinimide (EDC/NHS) covalent coupling method. The successful functionalization of NPs was examined with an in vitro kinetic binding of VEGF-PLGA-NPs with the VEGF Receptor-2 (KDR)/Fc chimera human compared to native PLGA-NPs. Tumor targeting of VEGF-PLGA-NPs was examined in vivo 24 h post-injection (p.i.) and compared with native PLGA-NPs. The T/M showed an increasing of PLGA-NPs capability to target the tumor over-expressing VEGFR.PLGA (D, L-lactide-co-glycolide) nanoparticles with a lactic acid:glycolic acid ratio of 50:50, average size of 100–200 nm, conjugated with a NIR fluorochrome with an excitation/emission wavelength of 780/825 nm, were purchased from Degradex® (Phosphorex Inc., Hopkinton, MA, USA).The conjugation of VEGF was performed by using the 1-Ethyl-3-(3-dimethylaminopropyl) carbodiimide (EDC) and N-hydroxysuccinimide (NHS) coupling protocol. EDC and NHS were purchased from Thermo Scientific (ThermoFisher, Waltham, MA, USA). The recombinant human VEGF-A165 analog with a molecular weight of 38.2 kDa was purchased by Prospec-Tany Technogene Ltd. (Rehovot, Israel). This molecule shares 88% homology with murine VEGF and has been previously used in mice [19,20]. The carboxylate (-COOH) PLGA-NPs react with NHS in the presence of EDC to create a stable crosslinking with the primary amines (-NH2) of the VEGF molecule. The conjugation condition was initially optimized with the use of bovine serum albumin (BSA), evaluating the protein-particle ratio, pH, the choice of buffer, the reaction time, and the purification method. A suspension of 6 mL MES buffer (pH 6.0) containing a concentration of PLGA-NPs (5 mg/mL) were first reacted with 30 mg of EDC (5 mg/mL in phosphate-buffered saline (PBS) pH 7.4). Then, 30 mg of NHS (5 mg/mL in PBS, pH 7.4) were added to the solution and incubated at room temperature with agitation for 15 min. To separate the activated PLGA-NPs from an excess of EDC, EDC-by-products, and NHS, the sample was centrifuged with a high-speed micro-centrifuge (ThermoFisher, Waltham, MA, USA) at 12,000 rpm (9500× g) 4 °C for 20 min and washing 3 times with 1 mL PBS (pH 7.4). The EDC coupling creates an unstable reactive o-acylisourea ester group that is easily substituted by an amine-reactive ester in the presence of NHS. The resulting NHS ester is semi-stable but very reactive towards the amino groups on the VEGF molecule. The carboxyl-amine reaction allows the conjugation of the VEGF onto the PLGA-NPs. RhVEGF (2 mg) was added to the PLGA-NPs suspension and the conjugation proceeded for 2 h at room temperature. The resulting VEGF-PLGA-NPs were collected by 3 times ultracentrifugation at 12,000 rpm (9500× g), 4 °C for 20 min, and was washed with 1 mL of PBS (pH 7.4) to remove unconjugated VEGF.The mean size, the polydispersity index (PDI), and the net surface charges (zeta potential) of native and functionalized PLGA-NPs were measured by dynamic light scattering (DLS), using photon correction spectroscopy, electrophoretic mobility analysis, and potential distribution at 25 °C with water as suspension medium. Reading was performed with a NanoZetaSizer analyzer (Malvern Instruments Ltd., Malvern, UK) equipped with a 5 mW HeNe laser (wavelength λ = 632.8 nm), a digital logarithmic correlator and a non-invasive backscattering (NIBS) optical system. Briefly, 10 µL (100 µg) PLGA-NPs were suspended with 90 µL H2O and loaded in Sarstedt polystyrol/polystyrene cuvettes (10 × 10 × 45 mm) for size and PDI measurements. For zeta potential analysis, 20 µL (200 µg) PLGA-NPs and VEGF-PLGA-NPs were suspended with 980 µL H2O, sonicated to reduce the aggregation and loaded in Malvern folded capillary cells for zeta potential measurements. Absorption spectra were acquired by a Jasco V-630 spectrophotometer. Briefly, 50 µL (500 µg) PLGA-NPs and VEGF-PLGA-NPs were diluted with 400 μL H2O and loaded in J18 Jasco quartz cells (path length = 10 mm). Water solution was measured separately as a blank solution and subtracted by sample spectra. All experiments were performed in triplicate.In vitro binding of native or VEGF-functionalized PLGA-NPs was performed with Nunc MaxiSorp™ 96 well plates (ThermoFisher, Waltham, MA, USA). The binding properties due to the hydrophilic surface of the wells allowed the coating of the VEGF Receptor-2 (KDR)/Fc chimera human (Sigma-Aldrich, St. Louis, MO, USA). Briefly, 50 µL of KDR-Fc in a final concentration of 0.002 µg/µL in bicarbonate/carbonate coating buffer (100 mM) was added to each well and the plate was covered and incubated 48 h at 4 °C. The coating solution was removed and the wells were rinsed two times with PBS (pH 7.4). Then, 150 µL of skimmed milk powder 2% (w/v) in PBS were added per well to block residual binding sites for 1 h at 37 °C. As a negative control, a blocking solution was added to each well that had not been coated with KDR-Fc. The blocking solution was removed by rinsing twice with 1 × PBS, pH 7.4. Then, 100 µL of two-fold dilution of VEGF-PLGA-NPs and native PLGA-NPs were added to each well followed by overnight incubation at 4 °C. KDR-Fc-uncoated wells were used to evaluate non-specific binding to the plastic. After incubation, the plate was washed two times with PBS and imaged with an in vivo FX station (Molecular Imaging Software, Kodak, Sevier County, TN, USA). Regions of interest (ROIs) were drawn for each well and the mean fluorescent intensity (mean IF) was calculated. The mean IF from wells without KDR-Fc (−KDR) was subtracted to the mean IF calculated in well coated with KDR-Fc (+KDR) to obtain PLGA-NPs and VEGF-PLGA-NPs net binding to KDR-Fc. Experiments were performed in triplicate.All animal experiments were carried out in compliance with the local ethics committee and in agreement with the National rules and the EU regulation (Study 204/2018-PR). A syngeneic murine tumor model was used for in vivo studies. The model was obtained by subcutaneous injection in the right thigh of 106 J774a.1 cells (reticulum cell sarcoma) in a medium: Matrigel® (BD-Biosciences, Bergen, NJ, USA) solution (200 μL, 50:50, v:v), in female BALB/c mice (8 weeks). Cells were purchased from American Type Culture Collection (ATCC® TIB-67™, Milan, Italy) and grown in ATCC-formulated Dulbecco’s Modified Eagle’s Medium supplemented with 10% of fetal bovine serum at 37 °Cand in 5% CO2. The 8-week-old female BALB/c mice were purchased from Harlan Laboratories. After about 20 days from the inoculation, the tumors became palpable and the targeting experiments were performed.To evaluate the kinetics and tumor targeting of native PLGA-NPs, 100 μL of NPs (500 μg) diluted with 50 μL NaCl were injected in the tail vein of 22 BALB/c mice, bearing a subcutaneous syngeneic tumor (reticulum cells sarcoma). At 2, 24, 48, and 72 h p.i. 5 mice per time point were anesthetized to acquire whole-body images with a Kodak in vivo FX station. Then, mice were sacrificed to excise major organs (liver, spleen, lungs, kidneys, muscle, tumor) to perform ex vivo optical imaging and quantify the uptake of PLGA-NPs in selected organs. ROIs were drawn over each organ, and the fluorescence signal was calculated as net fluorescence/area of the organ. Whole-body optical imaging only was performed in the last two mice up to 240 h.For tumor targeting experiments, fluorescent PLGA-NPs and VEGF-PLGA-NPs (500 µg in 150 μL 0.9% NaCl solution) were injected in the tail vein of 10 BALB/c mice (5 mice for each compound), bearing a subcutaneous syngeneic tumor (reticulum cells sarcoma). After 24 h, whole-body images were acquired and then mice were sacrificed. Liver, spleen, lungs, kidney, muscle, and tumor were excised for ex vivo optical imaging. On ex vivo images, ROIs were drawn to quantify the uptake of PLGA-NPs and VEGF-PLGA-NPs in selected organs and to calculate the tumor-to-muscle ratio (T/M). The fluorescence signal was calculated as net fluorescence/area.Statistical analysis was performed using SAS v. 9.4 (SAS Institute Inc., Cary, NC, USA).Variables continuous was showed as mean ± standard deviation (SD). Shapiro-Wilk test was used to verify the normality of distribution of continuous variables. We applied the Box-Cox procedure which allowed to identify suitable mathematical functions (log10, quadratic, and inverse) which make the non-normal continuous variables/residuals subsequently distributed according to the Gauss condition. Comparisons between “PLGA-NPs” vs. “VEGF-PLGA-NPs” and continuous variables were analyzed by t-test. We used the Satterthwaite formula when the variances were unequal. Differences between time (2 h, 24 h, 48 h, and 72 h) and the continuous variables were tested by GLM (General Liner Model) test. Homoscedasticity was verified by Levene and Brown-Forsythe tests. Post-hoc analysis was performed by the Tukey test. Mann-Whitney test was used comparing data of net binding of PLGA-NPs and VEGF-PLGA-NPs to KDR. A p < 0.05 was considered statistically significant.Preliminary characterization showed that native PLGA-NPs have a zeta average of 180 ± 17.08 nm with a PDI of 0.25 ± 0.02. The VEGF-PLGA-NPs have a zeta average of 173 ± 7.39 nm with a PDI of 0.17 ± 0.01. The zeta average and the PDI were reported as mean ± standard deviation (SD) of three measurements performed on the same sample (Table 1). Native PLGA-NPs showed a negative zeta potential value of −37.60 ± 0.67 mV, that excludes the presence of aggregates due to the Van der Waal interactions. The VEGF-PLGA-NPs have a zeta potential value of −9.43 mV that indicated a change in the potential difference across the boundaries between liquid and the NPs surface, revealing that the conjugation was successful (Figure 1).The spectroscopy was performed to confirm the absorbance of the sample. The results confirmed the absorbance of the fluorochrome conjugated with PLGA in the near-infrared region, generating an emission peak >700 nm (Figure 2).In vitro binding studies with PLGA-NPs and VEGF-PLGA-NPs are summarized in Figure 3.PLGA-NPs showed low NET binding to KDR-Fc that increased linearly with the concentration, properly due to non-specific interactions.On the other hand, NET binding of VEGF-PLGA-NPs reached a plateau at 1.2 mg/mL due to a VEGFR saturation, demonstrating the specific interaction between VEGF and KDR-Fc.The results indicated the binding specificity of VEGF functionalized PLGA-NPs with the KDR-Fc.In vivo pharmacokinetic studies of PLGA-NPs showed maximum tumor uptake at 72 h p.i., as shown in Figure 4. This result was confirmed by ex vivo imaging of the collected organs and a semi-quantitative analysis of the ROIs (Figure 5, Table 2). After PLGA-NPs injection, the tumor was clearly visible in planar whole body images, with a signal that increased up to 24 h with a high contrast to noise ratio. Tumor accumulation of PLGA-NPs gradually decreased with time over 240 h p.i.Ex vivo studies (Table 2) revealed that the main route of excretion is the liver due to the size of PLGA-NPs that exceed the glomerular filtration cut-off. However, fluorescence from the kidneys was also observed, probably due to the elimination of PLGA metabolites. In the blood circulation, PLGA are cleared by the cells of the mononuclear phagocytic system (MPS), that are also present in lungs, thus explaining their mean IF.The signal from the spleen, lungs, liver, and kidneys decreases from 2 h to 24 h, whereas the signal from the tumor increases with time. Imaging studies with PLGA-NPs and VEGF-PLGA-NPs are reported in Figure 6. Mice injected with VEGF-PLGA-NPs showed increased tumor uptake and higher T/M ratio if compared to PLGA-NPs (Figure 7, Table 3).Recently, biodegradable PLGA-NPs have been intensively investigated as carriers for drugs, peptides, and other molecules to treat cancer with low systemic toxicity [21,22]. However, nanoparticles are versatile molecules that could be also used for diagnostic imaging [23].PLGA-NPs characteristics such as size, surface charge, and polymer composition, can be tuned to modify their in vivo biodistribution and make them suitable tools for different purposes [24,25].For example, they could be even modified to enhance binding and active targeting to specific tumor antigens [26,27]. Given the many reports on the use of PLGA-NPs as a delivery system, we wanted to test the potential of specifically designed (lactic acid:glycolic acid ratio of 50:50, average size of 100–200 nm) NIR-fluorescent PLGA-NPs as theranostic tools for diagnosis and therapy of cancer. Preliminary results obtained by our group and confirmed by this study, showed that PLGA-NPs have suitable characteristics to be used as an in vivo targeting tool due to high accumulation in tumors thanks to the EPR effect. Indeed, high T/B ratios in tumors are achieved within 24 h p.i. of NIR-PLGA-NPs and reach their maximum at 72 h. To increase their accumulation in tumor lesions and reduce uptake in the liver and kidneys, we also developed fluorescent VEGF-conjugated NPs. Indeed, pathological neo-angiogenesis is involved in tumor growth and distant metastatization [28]. The angiogenic cytokines, as the vascular endothelial growth factor A (VEGF-A), are involved in the growth and remodeling of vessels in the tumor microenvironment [29,30,31]. Several targeted therapies based on VEGF/VEGFR signaling have been developed in different oncological diseases [32]. For example, the clinically approved anti-VEGF monoclonal antibody (mAb), bevacizumab, recognizes the free VEGF isoforms blocking their binding with VEGFR [33]. The anti-angiogenic tyrosine kinase inhibitors (TKIs), sorafenib, and sunitinib, were approved to target the VEGFR2, blocking the signaling cascade [34].The clinical implications of VEGF-targeted therapies caused several benefits for the majority of patients, with the exception of a small fraction [35]. This highlighted the importance of angiogenic markers when it comes to theranostic. In literature PLGA-NPs are widely described as a delivery system, encapsulating inside the polymers drugs or molecules usually with the double emulsion-solvent evaporation technique or nanoprecipitation method [36,37,38]. VEGF molecules are usually encapsulated inside the PLGA-NPs for therapeutic purposes as therapeutic angiogenesis or tissue regeneration [39,40]. In the present study, we functionalized the surface of NIR-fluorescent PLGA-NPs with the rhVEGF-A165 analog to enhance their accumulation in tumors.A similar approach has been described by Shi et al. that used recombinant human VEGF-C and achieved successful conjugation of the protein with NPs. However, their particle size was bigger than our (400 nm vs. 150 nm) and no biodistribution in vivo was shown [41].Our results from DLS analysis showed a significant drop in the zeta potential from −37.6 mV (of PLGA-NPs) to −9.4 mV (of VEGF-PLGA-NPs). However, the zeta potential indicates the potential difference across the boundaries between liquid and solid phases. This value should be higher than +25 mV or lower than −25 mV to indicate good stability. In the range between +25mV and −25 mV it indicates poor or no stability. The zeta potential should be evaluated together with the PDI that shows the dispersity of NPs in the liquid suspension and should be closer to 0. This index reveals the degree of dispersion of NPs in suspension (PDI higher than 0.7 indicates polydisperse NPs and aggregates; PDI less than 0.5 indicates monodisperse NPs without aggregates) [42]. Our results showed that VEGF-NPs, despite a suboptimal zeta potential (−9.4 mV), have an excellent value of PDI (0.17) and therefore reasonably stable to be used for in vitro or in vivo studies.In vitro binding studies to KDR-Fc, showed that, despite some non-specific interactions with the plastic surface, the binding of VEGF-PLGA-NPs to VEGF receptors (KDR-Fc) is specific. This result supports the finding of an increased T/M ratio of VEGF-PLGA-NPs if compared to PLGA-NPs. We also observed in vivo a lower uptake in other major organs (e.g., liver and spleen) and higher accumulation in kidneys. From a translational point of view, it would be very useful to have a diagnostic imaging probe that allows us to evaluate the degree of accumulation in tumors prior to administer the same NPs containing an anticancer drug. Fluorescent probes, despite their usefulness in pre-clinical applications, have limited penetration in tissues and are not suitable for human studies [43]. The limited penetration of light can be overcome by the use of radioactive isotopes, such as Copper-64 (T1/2 = 12.7 h) for positron emission tomography (PET) or Technetium-99 m (T1/2 = 6 h) for gamma-camera imaging [44]. Our study, showing high tumor accumulation of PLGA-NPs (with or without VEGF) within 24 h from injection, is preparatory for the development of radiolabeled NPs with diagnostic and/or therapeutic purposes, replacing the NIR-fluorescent probe. The use of radioisotopes, especially alfa or beta- emitters, poses a serious problem of liver and kidneys radiotoxicity and we believe that the added value of VEGF (or other targeting molecules) functionalization might mitigate this issue [45,46]. In this perspective, we selected for targeting studies the time point of 24 h p.i., as it matches with the half-life of most common diagnostic isotopes. Indeed, it would be of great interest to investigate later time points with VEGF-NPs, especially if radiolabeled with a therapeutic isotope, but priority should be given to test NPs radiolabeled with diagnostic isotopes to confirm the results obtained with optical imaging.The use of PLGA as a delivery system for several drugs has already been approved by the FDA and several studies have focused on their design for this purpose. Despite the extensive work with the PLGA in the therapeutic field, they have not been extensively explored as an imaging tool in humans [47,48]. Despite the recent progress in nanomedicine, the imaging depth-limit of fluorescence does not allow the application of these NPs for human diagnostic purposes [49]. For this reason, our strategy was to use fluorescent-PLGA-NPs as screening probes to assess pharmacokinetic, tumor targeting, and T/M ratio of native and functionalized PLGA-NPs. Second step will be to develop radiolabeled NPs with translational potential.In summary, our study confirms the potential of 50:50 100–200 nm PLGA-NPs as a theranostic tool in oncology. Functionalization with targeting molecules, such as VEGF, can increase their T/M ratio in vivo, but the replacement of fluorescent probes is mandatory to translate results in humans.Conceptualization—A.S., G.C. (Gabriela Capriotti), F.P., and M.V. Methodology—M.V., M.M., R.C., and F.G. Writing first draft—M.V., M.M., G.C. (Giuseppe Campagna), R.C., F.G. Final writing, review and editing of paper—A.S., F.P., and G.C. (Gabirela Capriotti). All authors have read and agreed to the published version of the manuscript.This research was funded by Associazione Italiana per la Ricerca sul Cancro, grant number AIRC IG-Grant 20411. The authors declare no conflict of interest.Particle size distribution and zeta measurement profile of native (a,b) and vascular endothelial growth factor-conjugated poly-lactic-co-glycolic acid nanoparticles (VEGF-PLGA-NPs) (c,d). Data are three different measurements made by the instrument on the same sample. (a) Size distribution of PLGA-NPs; (b) zeta potential distribution of PLGA-NPs; (c) size distribution of VEGF-PLGA-NPs; (d) zeta potential distribution of VEGF-PLGA-NPs.Light absorption spectra of poly-lactic-co-glycolic acid nanoparticles (PLGA-NPs) (a), and vascular endothelial growth factor-conjugated poly-lactic-co-glycolic acid nanoparticles (VEGF-PLGA-NPs) (b). Both the nano-formulations generated an emission peak >700 nm. Solvent was bidistilled water, pH 5.0.In vitro binding assay of PLGA-NPs (a) and VEGF-PLGA-NPs (b) to VEGF Receptor-2 (KDR)/Fc chimera human (KDR-Fc). KDR-Fc at concentration of 0.002 µg/µL was coated on the surface of 96-well plates. Two-fold dilutions of PLGA-NPs and VEGF-PLGA-NPs were incubated overnight at 4 °C. The mean fluorescent intensity (IF) was calculated for each well using in vivo FX station Kodak. Net binding was calculated by subtracting the mean IF in - KDR wells (negative control) to the mean IF calculated in +KDR well. One well for each dilution has not been coated with KDR-Fc and it was used as a negative control (-KDR). Results are presented as the means ± S.D (bars) of three separate experiments. Statistical analysis by Mann–Witney test showed significant difference between the two binding curves (p < 0.0001).Whole body optical images of the same mouse bearing a subcutaneous syngeneic tumor at 2, 24, 48, 72, 96, 168, 240 h post-injection of 500 µg of fluorescent PLGA-NPs subcutaneously in the right flank.Biodistribution of PLGA-NPs in BALB/c mice. Data are expressed as average fluorescence (NET/Area) ± SD of five different mice per time point.Whole body images of two mice bearing a syngeneic J744a.1 tumor in the right thigh and acquired 24 h post-injection (p.i.) of native PLGA-NPs (left) and VEGF-PLGA-NPs (right).Comparative distribution of PLGA-NPs and VEGF-PLGA-NPs at 24 h post-injection (p.i.) in collected organs. Data are expressed as average fluorescence (NET/Area) ± SD of five different mice per group.Characterization of PLGA-NPs and VEGF-PLGA-NPs to the DLS. Data are expressed as mean ± SD of three measurements.Bold: statistically significant (p < 0.05).Ex vivo fluorescence (NET/Area) of organs at different time points.* log10 transformed; ** quadratic transformed; Tumor: post-hoc analysis: p (2 h vs. 24 h) < 0.0001; p (2 h vs. 48 h) < 0.0001; p (2 h vs. 72 h) < 0.0001; p (24 h vs. 72 h) = 0.0023; Liver: post-hoc analysis: p (2 h vs. 24 h) = 0.042; p (2 h vs. 48 h) < 0.0001; p (2 h vs. 72 h) < 0.0001; p (24 h vs. 48 h) = 0.0016; p (24 h vs. 72 h) < 0.0001; Spleen: post-hoc analysis: p (2 h vs. 72 h) = 0.026; Lungs: post-hoc analysis: p (2 h vs. 24 h) = 0.0008; p (2 h vs. 48 h) < 0.0001; p (2 h vs. 72 h) < 0.0001; Kidneys: post-hoc analysis: p (2 h vs. 48 h) = 0.015; p (2 h vs. 72 h) = 0.0015; p (24 h vs. 48 h) = 0.039; p (24 h vs. 72 h) = 0.004. Bold: statistically significant (p < 0.05).fluorescence (NET/Area) of different organs at 24 h p.i. of PLGA-NPs and VEGF-PLGA-NPs.* Inverse transformed; ° Satterthwaite correction. Bold: statistically significant (p < 0.05).

Med-MDPI/jcm_2/jcm-09-06-01751.txt

@@ -0,0 +1 @@
+The aim of this study was to evaluate the relationship between serum levels of advanced glycation end products (AGEs) and abdominal aortic calcification (AAC) in patients with type 2 diabetes mellitus (DM2). This was a prospective cross-sectional study. One-hundred and four consecutive patients with DM2 were given lateral lumbar X-rays in order to quantify abdominal aortic calcification (AAC). Circulating levels of AGEs and classical cardiovascular risk factors were determined. Clinical history was also registered. Patients with higher AGEs values had higher grades of aortic calcification and higher numbers of diabetic-related complications. Multivariate logistic regression analysis showed that being older, male and having high levels of AGEs and triglycerides were the independent risk factors associated to moderate-severe AAC when compared to no-mild AAC. Our results suggest that AGEs plays a role in the pathogenesis of aortic calcifications. In addition, the measurement of AGEs levels may be useful for assessing the severity of AAC in the setting of diabetic complications.Cardiovascular (CV) disease is the leading cause of death in patients with type 2 diabetes mellitus (DM2) in Western countries and represents a major burden on morbidity, quality of life and health resources. Thus, the identification of patients with DM2 at a high risk of a CV event is important to develop early interventions and prevention strategies. In clinical practice, patients with worse lipid profiles, higher blood pressure, obesity, hyperglycemia, cardiac hypertrophy or dysfunction, nephropathy, neuropathy and/or retinopathy tend to have a higher prevalence of CVD [1]. In recent years, however, several studies have suggested that abdominal aortic calcification (AAC) increases the risk of major CV events in patients with DM2 [2,3,4]. AAC has been related to severe coronary artery calcification and cardiovascular outcomes specially in patients with DM2 and chronic kidney disease [5,6]. Also, it has been reported that AAC predicts major CV events after an acute coronary syndrome and predicts health care costs in older men independent of CV disease status as well [7,8]. Hence, detection and scoring AAC could improve the clinical identification of patients with DM2 at risk of developing major CV events.AAC is described as the hardening of the medial layer of the artery through deposition of hydroxyapatite crystals into the extracellular matrix [9,10,11]. Calcification occurs, amongst other factors, as a consequence of aging [12], but it can be accelerated by other diseases like renal insufficiency and diabetes [13,14]. Although several factors could accelerate or arrest the natural process of vascular calcification, the pathogenesis of vascular calcification in diabetes is not completely understood. Among the list of inducers, we can find hyperphosphatemia, vitamin D, lipids or inflammatory cytokines, whereas fetuin-A, phytate, pyrophosphate, vitamin K, osteopontin or matrix Gla protein have been described to be inhibitors of vascular calcification [15]. Consequently, further understanding of the mechanisms by which diabetes might favor this process needs to be elucidated. Indeed, high glucose appears to have an important role in the initiation and progression of this complication.Ideally, AAC can be quantified with computed tomography (CT), lumbar X-ray and dual-energy X-ray absorptiometry (DXA) scans [16]; however, these imaging studies are not routinely performed in the clinic. Therefore, circulating biomarkers could be useful to identify those patients more prone to AAC and to develop new treatments. In this regard, low-density lipoproteins (LDL), high-density lipoproteins (HDL), glycated hemoglobin (HbA1c), osteocalcin, phosphate and advanced glycation end products (AGEs) have been identified as potential markers of AAC but literature is ambiguous and further research is needed [2].AGEs are the irreversible products of the reaction between amino groups of proteins and sugar resulting from long-term hyperglycemia [17]. AGEs are formed at an accelerate rate under non-controlled diabetic state and play an important role in the development and progression of vascular complications in diabetes [18]. AGEs can affect cell function via intracellular glycation of proteins [19], provoke arterial stiffness by cross-linking of extracellular matrix proteins in arteries [20], and induce receptor-mediated cell activation by binding to the receptor of AGEs (RAGE) [21]. AGEs are believed to be involved in the onset and progression of atherosclerosis through multiple mechanisms and several studies indicate that AGEs and their receptors (RAGEs) play an important role in vascular calcification [22]. In individuals with chronic kidney disease (CKD), AGEs have been associated with arterial calcification [23,24]. Recently, several studies using different models of vascular smooth muscle cells (VSMC) demonstrated the implication of AGEs in calcification [25,26,27,28]. Indeed, it was shown that exposure of VSMC to AGEs induced cell calcification [29]. However large studies investigating plasma AGEs levels and their relation with AAC are scarce.Therefore, the main objectives of the present study are: (1) to examine the association between serum levels of AGEs with AAC, quantified on lumbar X-ray; and (2) to determine the most important CV risk factors and diabetic complications associated to moderate-severe AAC in T2DM. One-hundred and four patients with DM2 were recruited consecutively at the Outpatient Diabetic Clinic of Son Llàtzer University Hospital (Palma de Mallorca, Spain). The demographic characteristics of the patients according to the AGEs levels are shown in Table 1. Briefly, patients were included if they were older than 40 years, had DM2 duration longer than 1 year, and had had a lumbar X-ray performed in the past 6 months. Exclusion criteria were: life expectancy shorter than 1 year, immunodeficiency, human immunodeficiency virus (HIV) infection, drug abuse or alcohol intake higher than 50 g/day, body mass index (BMI) higher than 40 Kg/m2, participation in a clinical trial in the past year, inability to attend programmed visits, illiteracy, chronic or acute infection in the past 3 months, neoplasia in the past 5 years, vasculitis, resistant hypertension, moderate or severe hepatic insufficiency or glomerular filtration rate (GFR) lower than 30 mL/min/1.73 m2.The study protocol was approved by the Ethics Committee of the Balearic Islands (approval number: CEIC-IB 2523/15PI). Written informed consent was obtained from all study participants.The main outcome measures were serum level of AGEs and AAC scores. Clinical histories were obtained from the electronic medical records. Furthermore, data from anamnesis, laboratory analysis, and physical examinations were prospectively collected during the study. Physical and anthropometric measurements were determined by qualified personnel. Blood samples were collected in the morning (after 12 h of fasting). The samples were left to stand for 30 min at room temperature, and the serum was then separated by centrifugation. Biochemical analyses were performed in an automated analyzer (Cell-Dyn Sapphire and Architect ci16200, Abbott, IL, USA). Insulin was analyzed by chemiluminescent-immunometric assay (Advia Centaur, Siemens, NY, USA). Highly sensitive C-reactive protein (hs-CRP) and lipoprotein (a) (Lp [a]), were analyzed by nephelometry (Immage 8000, Beckman Coulter Inc, CA, USA). All samples were run in duplicate, and the coefficients of intra- and inter-assay variation were below 10%.Blood pressure was measured 3 times consecutively after 5 min of rest while the subject was sitting quietly. The average of the second and third measurements was recorded. Patients using anti-hypertensive drugs and those with systolic blood pressure of 140 mmHg or more and/or diastolic blood pressure of 90 mmHg or more were categorized as having hypertension [30]. Atherosclerosis was diagnosed by having clinically significant carotid and/or femoral plaque burden documented with arterial ultrasonography. Chronic kidney disease (CKD) was diagnosed based on the estimated GFR (stage 2 CKD: 89–60 mL/min/1.73 m2; stage 3a CKD: 59–45 mL/min/1.73 m2; stage 3b: 44–30 mL/min/1.73 m2), calculated as previously described [31].AGEs in serum samples were measured using the OxiSelect™ AGE Competitive enzyme-linked immunosorbent assay (ELISA) Kit (Cell Biolabs Inc, CA, USA), which provides rapid detection and quantification of AGE protein adducts. Quantitation was determined by comparing absorbance with that of a known AGE–bovine serum albumin (BSA) standard curve. First, an AGE conjugate was coated on an ELISA plate. The unknown AGE samples or AGE–BSA standards were added to the AGE conjugate pre-absorbed ELISA plate. After a brief incubation, an anti-AGE polyclonal antibody was added, followed by horseradish peroxidase-conjugated secondary antibody. The content of AGE protein adducts in unknown samples was determined by comparison with a pre-determined AGE–BSA standard. The intra-assay coefficient of variation was 6.9% (three replicates for each sample on the same day). The inter-assay coefficient of variation was 9.2% (3 days at the same time).Lateral lumbar X-rays were performed while patients were standing, using standard radiographic equipment. A minimum of 8 cm of tissue anterior to the lumbar spine, including the abdominal aorta, was visible. The focus-film distance was 100 cm, the tube potential was 94 kV, the tube-current-time product was 33–200 mAs, and the estimated radiation dose was approximately 15 mGy.ACC was assessed using a previously validated 24-point scale [32,33]. For this 24-point scale, calcified deposits along the anterior and posterior longitudinal walls of the abdominal aorta, adjacent to each lumbar vertebra (L1 to L4), were assessed using the midpoint of the intervertebral space above and below the vertebrae as the boundaries. Calcifications were graded as 0 (no aortic deposits), 1 (small scattered deposits less than one-third the length of the vertebral length), 2 (intermediate quantity of deposits, about one-third or more, but less than two-thirds of the vertebral length), or 3 (extensive deposits of two-thirds or more of the corresponding vertebral length). The scores were determined separately for the anterior and posterior walls; the range was 0 to 6 for each vertebral level, and 0 to 24 for the total.All subjects were assessed independently by two graders who were blinded to patient data. To validate the X-ray assessment, double readings were performed in all patients demonstrating an excellent inter-observer agreement (intra-class coefficient of correlation R = 0.948, p < 0.0001).Data are presented as means and standard deviations, medians and interquartile ranges, or numbers and percentages. Patients were divided in three groups according to tertiles of serum AGEs levels: low (L < 6.5 U/mL), intermediate (I: 6.5–10.4 U/mL) and high (H > 10.4 U/mL). Intergroup comparisons of serum levels of AGEs employed one-way analysis of variance (ANOVA) and the independent-samples t-test (as post-hoc test); or Kruskal–Wallis test and Mann–Whitney U test (as a post-hoc test) for continuous variables. A chi-square test and Fisher’s exact test were used for categorical variables. For the second objective, patients were divided in two groups according the median of AAC scores: no-mild (AAC < 6) and moderate–severe (AAC ≥ 6). Intergroup comparisons of aortic abdominal calcification (AAC) scores employed the independent-samples t-test or the Mann–Whitney U test for continuous variables, and the chi-square test or Fisher’s exact test for categorical variables. Receiver operating characteristic (ROC) curves of quantitative risk factors associated to moderate–severe AAC were performed. The optimal cutoff values were determined by the maximum Youden index (J), defined as sensitivity + specificity −1. Binary logistic regression models were used to identify risk factors associated to moderate–severe AAC (AAC ≥ 6), with an AAC less than 6 (no-mild AAC) as the reference (odds ratio [OR] = 1). Analysis was performed using the stepwise backward method. A two-tailed p-value less than 0.05 was considered statistically significant. Statistical analyses were performed using SPSS 23.0 (SPSS Inc., Chicago, IL, USA).Demographic and clinical characteristics of the three groups are shown in Table 1. Patients with higher AGEs levels were older (L: 67.5 ± 7.3; I: 66.9 ± 10.4; H: 73.7 ± 7.9 years; p < 0.001) and the percentage of males was also higher (L: 34.3%; I: 50.0% and H: 60.0%; p = 0.095), although it did not reach statistical significance. Patients with higher AGEs levels presented more diabetic complications, such as diabetic nephropathy, atherosclerosis and chronic kidney disease (p < 0.05). Furthermore, the number of diabetic complications increased with AGEs levels (Figure 1). The percentage of patients with three or more DM complications was 6%, 21% and 34% for low, intermediate and high AGEs groups, respectively (p = 0.009) while the percentage of patients with no DM2 complications were 57%, 38% and 23%, respectively. Regarding medication, a higher percentage of patients with high AGEs levels were taking calcium antagonists and antiplatelets when compared to patients with low AGEs levels (p = 0.025 and p = 0.001, respectively).Table 2 shows the clinical and biochemical characteristics for AGEs groups. As can be seen, GFR was lower for high AGEs group compared to low and intermediate AGEs groups [L:77.8(55.7–88.9); I:83.1(50.1–98.2); H: 65.4(46.0–79.1) mL/min/1.73 m2; p = 0.040] whereas blood levels of hemoglobin, urea, creatinine, urate, albumin: creatinine ratio and glycated hemoglobin significantly increase with AGEs levels (p < 0.05).There was a trend toward more AAC as serum AGEs levels increased (Figure 2). The median (inter-quartile range, IQR) of AAC 24-points score was 2 (0–5), 9.5 (4–13) and 8 (2–13) for low, intermediate and high AGEs groups, respectively (Figure 2a, p = 0.025). There were 23%, 68% and 63% of patients with moderate-severe AAC in the low, intermediate and high groups respectively (Figure 2b, p < 0.001). Furthermore, 21 (20.2%) of all patients presented no AAC and of these, 10 (47.6%), 5 (23.8%) and 6 (28.6%) presented low, intermediate and high AGEs levels, respectively.Table 3 shows the clinical characteristics and laboratory parameters between patients with no-mild AAC (AAC < 6) and moderate-severe AAC (AAC ≥ 6). The group with moderate-severe AAC was older, had a higher percentage of males, more DM2 complications, more prevalence of atherosclerosis, higher percentage of smokers or ex-smokers, lower GFR and higher levels of urinary albumin, albumin: creatinine ratio, creatinine, triglycerides, HbA1c and AGEs compared to those with no-mild AAC.ROC curves and optimal cut off values were calculated for quantitative risk factors associated to moderate-severe AAC (Figure 3). As can be seen, circulating serum AGEs upper than 7 U/mL had a sensitivity and specificity of 81.1% and 56.9%, respectively. The overall accuracy was 68.3% for AGEs above 7 U/mL and it was bigger than those for the rest of the analyzed variables (62.5% for age above 70 years; 61.5% for triglycerides upper than 146 mg/dL; and 64.4% for GFR lower than 74 mL/min/1.73 m2).Univariate and multivariate logistic regression analysis were used to investigate independent factors associated to the presence of moderate–severe AAC (vs. no-mild AAC). All previously listed factors (p < 0.05 in Table 3) were included initially in the model before stepwise and backward elimination. The final model included age older than 70 years, being male, circulating AGEs levels upper than 7 U/mL, and triglycerides levels upper than 146 mg/dL as the independent risk factors associated to moderate–severe AAC in DM2 (Figure 4).Our study reports that serum levels of a pool of circulating AGEs determined by ELISA positively correlated with AAC (based on lateral lumbar X–rays) in subjects with DM2. Previously, Saremi et al. prospectively demonstrated in the VA Diabetes Trial and Follow-up Study that baseline plasma levels of Nε-carboxyethyl lysine (CEL) determined by liquid chromatography-mass spectrometry were strongly associated with the extent of AAC determined by CT scanning after an average of 10 years of follow-up [34]. However, this study did not report AGE measurements at the time of CT acquisition, so the relationship between AGE and AAC at the same time-point was not examined.Secondly, we have also observed that both groups of patients with intermediate and high AGEs levels had a higher prevalence of diabetes-related complications. Several data have linked circulating AGEs to the development and progression of diabetes complications [35,36,37] including diabetic retinopathy [38,39,40,41], nephropathy [42,43], and cardiovascular disease [44,45,46]. Furthermore, an accumulation of AGEs has been detected in most of target tissues of diabetes complications, such as the kidney, retina, and atherosclerotic plaques [47]. Several studies have shown correlations between serum levels of AGEs and the development/severity of vascular disease [48,49,50,51,52]. Our results indicate that circulating levels of AGEs are mainly correlated with CKD and aging. These results are consistent with previous studies which found that serum AGEs are positively correlated with serum creatinine and inversely correlated with GFR [23,24,35,42,53]. Regarding aging and circulating levels of AGEs, our results are in accordance with other authors who indicated that AGEs are linked to the aging process [35,36,47,51,54,55] and are considered a marker of senescence [56].On the other hand, we find an association between circulating AGEs and HbA1c. Similar results have been found in other studies [35] where AGEs levels were higher in patients with poor glycemic control measured by HbA1c [54,57,58,59]. It is important to consider that the extent and duration of hyperglycemia are predicted by increased levels of HbA1c, which is considered an acceptable marker. As mentioned above, prolonged exposure to glucose produces early AGEs and affects different proteins. An important example of early glycated proteins is HbA1c, which is further modified through a series of reactions in Hb-AGE. Under normal conditions, Hb-AGE makes up 0.42% of circulating hemoglobin (Hb) levels, increasing to 0.75% in diabetic subjects [60]. Besides HbA1c correlates with AGEs, AGEs-modified proteins are hardly degraded and remain in the vessels, kidney and heart for a long time, even after glycemic control has improved. Therefore, AGEs are considered to be a better index of cumulative diabetic exposure and one of the main actors of the metabolic memory observed in diabetic patients.Concerning the factors associated to AAC, our results show that moderate-severe AAC is associated to current or ex tobacco use. Similar results were found in other studies, where cigarette smoking was strongly associated with calcification [61]. The independent factors, however, associated with AAC in our patients were age, male gender, AGEs levels and plasma triglycerides. These results are in accordance with previous studies which have reported that age and prior CV disease are strongly associated with AAC [62,63,64]. Triglycerides can be markers of glycemic control and have become an independent predictor of cardiovascular risk, as demonstrated by multiple prospective epidemiological studies [65,66,67,68]. Also, it is well-known that vascular calcification is more common in males than in females [69]. It is worth underlining that we developed ROC curves and searched for the optimal cut-off values associated with AAC. Remarkably, the accuracy for AGEs was higher than the observed for the other factors classically related to AAC (Figure 4). Moreover, an AUC higher than 0.8 showed a good fit for the multivariate model of factors associated to moderate-severe AAC (Figure 4). All these data suggest that AGEs could be useful markers of AAC which could identify those priority patients of diagnostic imaging techniques. The presence of calcification in any arterial wall is associated with a 3- to 4-fold higher risk for mortality and cardiovascular events [4]. Consequently, the early detection of AAC could have clinical importance in order to intensify cardiovascular prevention efforts. Therefore, it would be worth developing further prospective studies to draw definitive conclusions about the application of AGEs in algorithms for AAC prediction.Our results support the notion that AGEs can play a role in the pathogenesis of aortic calcifications and are in accordance with tissue culture studies that indicate that AGEs can act as an inducer of calcification affecting cell function via both, intracellular glycation of proteins and AGE-RAGE signaling [20,22,70].Considering that morbidity and mortality in patients with DM2 are mainly due to cardiovascular disease, the measurement of circulating AGEs could be useful as an early marker of the presence and severity of AAC and the occurrence of future diabetic complications. In addition, an inhibitory therapy of the formation of AGEs or a therapy addressed against the AGE-RAGE system could be useful as a preventive treatment in the development of AAC and diabetic complications Recently, DNA aptamers directed against AGE and RAGE, attenuating their interaction, have been developed and are being studied with promising experimental results [71]. Also, phytate has been described as inhibiting glycation processes by the iron chelation process [72] and is a well-known inhibitor of AAC via absorption over the hydroxyapatite crystals. Consequently, it is possible that this compound can be a strong inhibitor of the formation of AAC via the two aforementioned mechanisms [73,74,75]. Interestingly, a recent study suggests that the antidiabetic therapy with a sodium-glucose cotransporter inhibitor could reduce the levels of AGEs in comparison with a dipeptidyl peptidase−4 inhibitor [76]. Finally, beside drugs or compounds, the avoidance of smoking and the restriction of dietary AGEs consumption must be borne in mind to reduce the AGEs burden [77,78].Our study has several limitations. First, the sample is small and it is a cross-sectional study from a single medical center; therefore, the findings presented here should be interpreted with caution. Even though we found that age, gender, triglycerides and circulating levels of AGEs were associated to AAC, this does not prove causality. Prospective multicentric longitudinal studies are needed to fully clarify how AGEs actually contribute to the genesis and progression of AAC and to assess the possibility that circulating levels of AGEs may serve as biomarkers to predict or diagnose AAC. Another possible limitation is that we measured AAC using lateral abdominal plain X-rays. One could argue that CT is more sensitive. Several studies, however, have positioned lateral lumbar radiography as a tool to estimate the severity of AAC. Furthermore, AAC evaluated by lateral X-rays shows a good correlation with coronary calcium score [4,79]. Therefore, we consider that plain radiography is an appropriate screening method for evaluating AAC in daily practice since it is simpler, cheaper and with lesser radiation than CT.Finally, there are several methods available to measure AGEs and there are no single standardized protocols or normal reference ranges established for them. We measured a pool of AGEs by ELISA. It has been observed that CEL but not methylglyoxal hydroimidazolone, 3-deoxyglucosone hydroimidazolone, 2-aminoadipic acid and methionine sulfoxide was related to AAC [34]. Therefore, it is possible that our results were related to specific AGEs. Thus, further studies aimed at addressing these issues are needed.In conclusion, we provide evidence that serum levels of AGEs positively correlate with AAC in patients with DM2. In addition, our study suggests that age, gender, triglycerides and circulating levels of AGEs are the risk factors associated to moderate–severe AAC. We also have shown that circulating levels of AGEs were positively associated with AAC severity and diabetes-related complications. Therefore, increased levels of AGEs can be contemplated as a biomarker of AAC and other vascular complications of diabetes.All authors make substantial contributions to conception and design, and/or acquisition of data, and/or analysis and interpretation of data: conceptualization, P.S., L.M. and F.G.; methodology, P.S., R.R., M.M.-G. and L.M.; software, M.M.-G., P.S. and R.R.; validation, P.S., R.R. and L.M.; formal analysis, P.S and R.R.; investigation and resources, R.R., P.S., R.F., M.G.-F., M.M. and C.R.; data curation, R.R., and P.S.; writing-original draft preparation, P.S., R.R. and L.M.; writing-review and editing L.M., F.G., R.R., M.G.-F. and P.S.; visualization, L.M., R.R. and P.S.; supervision, L.M., F.G. and P.S.; project administration L.M., F.G. and P.S.; funding acquisition, L.M., F.G. and P.S. All authors have read and agreed to the published version of the manuscript.This research was funded by Balearic Islands Government, grant number AAEE033/2017.MGF acknowledges funding by the Miguel Servet Program (MS19/00201), Instituto de Salud Carlos III (ISCIII), Madrid. All authors acknowledge funding by the LIBERI program, Health Research Institute of the Balearic Islands (IdISBa), Palma de Mallorca.The authors declare no conflict of interest.Diabetic related-complications among serum AGEs levels. The significance of differences between groups were determined using chi-square test (p = 0.009). * p < 0.05 vs. Low AGEs group (<6.5 U/mL).Abdominal aortic calcification (AAC) score (a) and percentage of patient with no-mild AAC and moderate-severe AAC (b) among AGEs levels in patients with DM2. The significance of differences between groups were determined using Kruskal–Wallis and Mann–Whitney U test for (a); and chi-square test for (b). * p < 0.05 vs. low serum levels of AGEs (<6.5 U/mL).Receiver operating characteristic (ROC) curves and optimal cut-off of the following quantitative risk factors associated to moderate-severe AAC (vs. no-mild AAC): age (a), AGEs (b), triglycerides (c) and GFR (d). The optimal cutoff values were determined by the maximum Youden index (J), defined as sensitivity + specificity −1. The table indicate the area under the curve (AUC), sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) and overall accuracy of the optimal cut off values. Abbreviations. AGEs; advanced glycation end products; GFR: glomerular filtration rate.Forest plot (a) and ROC curve (b) of the multivariate logistic regression of risk factors associated to moderate–severe AAC in patients with DM2. Multivariate analysis was performed using stepwise backward method. Crude and adjusted odds ratio (OR) are indicated in the table. Comparison of the expected and observed frequencies by the Hosmer–Lemeshow goodness-of-fit test (p-value = 0.734) and by ROC curve (AUC = 0.802; p < 0.001) indicated a good fit for the model. AUC = the area under the curve, GFR = glomerular filtration rate.Clinical characteristics of patients among circulating advanced glycation end products (AGEs) groups.Each value is given as mean ± standard deviation or frequency (percentage). The significance of differences between groups were determined using one-way analysis of variance (ANOVA) and the independent-samples t-test; or Kruskal–Wallis test and Mann–Whitney U test for quantitative data. Chi-square test and Fisher’s exact test were used for qualitative data. a: p < 0.05 vs. corresponding value of Low AGEs group; b: p < 0.05 vs. corresponding value of Intermediate AGEs group. Abbreviations. DM2: type 2 diabetes mellitus; GFR: glomerular filtration rate calculated by MDRD-4 IDMS equation.Laboratory parameters among AGEs groups.Each value is given as median (interquartile range). The significance of differences between groups were determined using one-way ANOVA and the independent-samples t-test; or Kruskal–Wallis test and Mann–Whitney U test for quantitative data. a: p < 0.05 vs. corresponding value of Low AGEs group; b: p <0.05 vs. corresponding value of Intermediate AGEs group. Abbreviations. HbA1c: glycated hemoglobin; HDL: high-density lipoprotein; LDL: low-density lipoprotein; PTHi: intact parathyroid hormone; GFR: glomerular filtration rate calculated by MDRD-4 IDMS equation; Alb: albumin; Creat: creatinine.Clinical characteristics and laboratory parameters between patients with no-mild AAC and moderate–severe AAC.Each value is given as mean (± standard deviation), frequency (percentage) or median (interquartile range). The significance of differences between groups were determined using independent t-test or Mann–Whitney U test for quantitative data and chi-square test or Fisher’s exact test for qualitative data. Abbreviations. AGEs: advanced glycation end products; HbA1c: glycated hemoglobin; HDL: high-density lipoprotein; LDL: low-density lipoprotein; PTHi: intact parathyroid hormone; GFR: glomerular filtration rate calculated by MDRD-4 IDMS equation; Alb: albumin; Creat: creatinine.

Med-MDPI/jcm_2/jcm-09-06-01752.txt

@@ -0,0 +1 @@
+Dermatofibrosarcoma protuberans (DFSP) is a slow growing, low- to intermediate-grade dermal soft-tissue tumor. It has a high local recurrence rate but low metastatic potential. It is characterized by a uniform spindle cell arrangement, classically with a storiform pattern and CD34 immunoreactivity. The histomorphology and immunophenotype overlap with a broad range of other neoplasms. The standard treatment is complete surgical excision. The surgical procedures include wide local excision (WLE) with tumor free margins, Mohs micrographic surgery (MMS) and amputation. Unresectable DFSPs are treated with radiation therapy and/or targeted therapy. DFSP has characteristic t(17; 22) (q22; q13), resulting in a COL1A1- PDGFB fusion transcripts in more than 90% of DFSPs. Molecular detection of the gene rearrangement or fusion transcripts is helpful for the diagnosis of patients with atypical morphology and for screening candidates for targeted therapy with tyrosine kinase inhibitors. The aims of the present review are to update the clinical presentation, tumorigenesis and histopathology of DFSP and its variants for diagnosis and differential diagnosis from other benign and malignant tumors, to compare the advantages and drawbacks of WLE and MMS, to propose the baseline for selecting surgical procedure based on tumor’s location, size, stage and relationship with surrounding soft tissue and bone structures, and to provide a biologic rationale for the systemic therapy. We further propose a modified clinical staging system of DFSP and a surveillance program for the patients after surgical excision.Dermatofibrosarcoma protuberans (DFSP), derived from dermal fibroblasts, was initially characterized as keloid sarcoma. It was named as DFSP by Hoffman in 1925 [1]. Clinically, it is a slow-growing, low- to intermediate-grade malignant sarcoma and frequently occurs in middle-aged adults. Histopathologically, it is composed of uniform spindle cell fascicles growing in a storiform pattern with multiple variants and with strong and diffuse CD34 immunoreactivity. However, its spindle cell morphology and CD34 immunostaining pattern are overlapped with other benign and malignant lesions, which must be differentiated from. Ultrastructurally, DFSP is characterized by stellate or spindle cells, with long, slender, ramified cell processes joined by primitive junctions, which are similar to dermal dendrocytes [2]. Cytogenetically, more than 90% of DFSP have a t(17; 22) (q22; q13), leading to the formation of COL1A1-PDGFB fusion transcripts. Molecular detection of gene rearrangements or fusion transcripts is beneficial not only for the diagnosis in cases without typical morphology, but also for screening patients who are candidates for using imatinib mesylate (a tyrosine kinase inhibitor that affects PDGFβR). This provides neoadjuvant targeted therapy for the patients with unresectable, recurrent or metastatic DFSP [3]. The standard treatment of resectable DFSPs is complete surgical excision with either wide local excision with tumor free margins or Mohs micrographic surgery, or, rarely, amputation. Each procedure has advantages and drawbacks. Unresectable DFSPs are treated with radiation therapy and/or targeted therapy. Accordingly, we update the DFSP’s clinical manifestations, histological features with its variants, diagnosis, differential diagnosis and tumorigenesis. We discuss the advantages and drawbacks of different surgical procedures for resectable tumors and the width of margin resection. We review the biologic rationales of radiation therapy and/or targeted therapy for unresectable tumors. We also propose a modified staging system of DFSP for clinical practice and a surveillance program to monitor local recurrence and metastasis after surgical excision.DFSP is an uncommon, indolent dermal soft-tissue sarcoma that accounts for less than 0.1% of all malignancies and less than 1% of all soft-tissue sarcomas [4,5,6]. Two large epidemiological studies in the United States demonstrated that the annual incidence of DFSP was 4.2 per million people studied in a 30-year period from 1973 to 2002 [7], and 4.1 per million people in a 10-year period between 2000 and 2010 [8]. The incidence of DFSP is higher in women than men, and higher in African-American than white patients [4,5]. It most frequently occurs in young and middle-aged patients, between 25 and 45 years of age, with a mean age between 40 to 43 years [9,10]. However, patient’s age ranged widely from infancy to the elderly [11,12,13].Cytogenetic and molecular studies have demonstrated that more than 90% of DFSPs are characterized by either supernumerary ring chromosomes derived from chromosomes 17 and 22 or chromosomal translocation t(17; 22) (q22; q13), resulting in the fusion of collagen type 1-alpha 1(COL1A1 at 17q22) and platelet-derived growth factor beta (PDGFB at 22q13) genes. The gene fusion places the PDGFB gene under the control of the COL1A1 promoter [14,15,16], leading to PDGFβ overexpression and dimerization, and subsequently resulting in continuous activation of the PDGF receptor β protein-tyrosine kinase [17]. Interaction of PDGFβ and PDGF receptor β is involved in multiple signaling pathways including Ras mitogen-activated protein kinases (RAS-MARK) and phosphatidylinositol 3-kinase-akt-rapamycin (mTOR) (PI3K-AKT-mTOR) [18,19,20,21]. Correspondingly, increased expression of the phosphorylated Akt-mTOR pathway proteins including Akt, mTOR, 4EBP1, and S6RP and phosphor-PDGFRα/β have been demonstrated in about half of DFSP tissues by immunoperoxidase studies [21], suggesting that Akt-mTOR pathways are involved in the tumorigenesis of DFSP. Gene fusion transcript of COL1A1−PDGFB can be detected by either fluorescent in situ hybridization (FISH) or multiplex reverse transcriptase-polymerase chain reaction (RT-PCR) in formalin fixed, paraffin embedded tissues [22,23]. These are helpful for the diagnosis, differential diagnosis and guiding treatment of DFSP [24], especially when the tumor’s histomorphology is not typical or when the tyrosine kinase inhibitors are considered for treatment. The COL1A1-PDGFB fusion transcript cannot be detected in about 8% of DFSPs [15]. FISH analysis revealed genetic translocations involving the CSPG2 gene at 5q14.3 and PTK2B gene at 8p21.2 in a patient of DFSP without the COL1A1-PDGFB fusion transcript [16]. Other reported genetic translocations in DFSP include COL1A2-PDGFB [3], COL6A3-PDGFD [25] and elastin microfibril interface 2 (EMILIN2)-PDGFD [26]. Moreover, p53 mutation and overexpression, murine double minute 2 (MDM2) overexpression were reported in fibrosarcomatous variant of DFSP [17,27]. Furthermore, DFSPs were reported in patients with immunodeficiency disorders, including X-linked agammaglobulinemia [28], adenosine deaminase-deficient severe combined immune deficiency [29], ataxia telangiectasia syndrome [30] and HIV infection [31]. In addition, pregnancy may increase the risk for the development of DFSP [32]. These data suggest that multiple factors including oncogenes, tumor suppressor genes and immunodeficiency are involved in the development of DFSP. Further investigation is required to understand the relationship of these risk factors to the development of DFSP.In early stages of DFSP, the patient typically notices a slow-growing, small, firm, painless, skin-colored dermal plaque (Figure 1A,B), subcutaneous thickening or atrophic non-protuberant lesion [33,34]. Early pediatric DFSPs were classified into four variants: (1). Small palpable nodules slowly forming a confluent sclerotic plaque; (2). Keloid-like homogenous cutaneous thickening plaque; (3). Tumor ab initio and (4). Atrophic plaque [35]. Congenital DFSPs were reported as either erythematous atrophic plaque or irregular shaped, pigmented macula resembling melanocytic nevus after birth [34,36]. These congenital DFSPs carried the COL1A1-PDGFB fusion detected by either RT-PCR or FISH [34,36]. The early non-protuberant lesions gradually enlarge to form protuberant, indurated, reddish-blue or violaceous nodules in protuberant stages (Figure 2A–C). If untreated, the tumors can locally invade more deeply into the fascia, muscle, periosteum, bone and occasionally metastasize to other organs in advanced stages [37]. The tumor cells most frequently metastasize to lung, brain, bone, visceral organs, lymph nodes and soft tissues [13,38,39,40]. This insidious growing process takes several months or years to as long as 60 years [13,33,34]. The reported tumor sizes vary in a wide range from 0.5 to >10 cm in diameter, with a mean of 2–3.5 cm [41,42]. Most DFSPs are localized on the trunk (40–50%), followed by proximal extremities (30–40%), and then the head and neck (10–15%) [43,44,45]. Infrequently documented cases have been reported on the toes [22,46,47,48,49], scalp [50], breast [51,52,53] and vulva [54].In addition to a thorough history and physical examination, diagnosis of suspected DFSP ideally requires a generous biopsy (punch biopsy or excisional biopsy) for a pathologic diagnosis. The pathology report should note the presence of fibrosarcomatous change or other high-risk features. Examination of lymph nodes and imaging studies are important for staging and surgical planning.Magnetic resonance imaging (MRI) delineates tumor’s size and extent, and its relationship with adjacent neuromuscular structures and bone (Figure 3, Figure 4 and Figure 5A,B). MRI is therefore recommended for pre-operative evaluation, surgical planning and follow-up for recurrence [55,56]. MR T1-weighted images exhibit well-defined homogeneous isointense lesions, while T2-weighted images show a well-defined subcutaneous soft tissue nodules or mass with intermediate-to-marked homogeneous hyperintensity to the surrounding muscular tissues (Figure 4A,B) [57]. Poorly defined irregular margins can be observed in some cases (Figure 4B) [58]. High frequency ultrasound can be used to evaluate the extent of the tumor involvement as well as provide biopsy guidance. The ultrasound of DFSP often appears as a hypoechoic superficial nodular mass [59,60]. Computed tomography (CT) reveals a solitary, subcutaneous lobular or nodular architecture and soft tissue attenuation and post-contrast enhancement (Figure 5C,D) [61]. Intratumoral non-enhancement areas in large tumors (>5 cm) may suggest necrosis and cystic degeneration. CT is useful to evaluate distant metastatic disease. The technique of 18F-fluorodeoxyglucose–positron emission tomography/computed tomography (18F-FDG PET/CT) has shown potential value in both identifying metastatic disease and evaluating treatment response [61,62,63]. X-rays have no role in the imaging of the primary DFSP.Grossly, DFSP is commonly a white to yellow color, poorly circumscribed, soft-tissue mass without a smooth outer surface. The cut surface is white to yellow, poorly encapsulated, solid and a fish flesh-like texture (Figure 2D and Figure 6). Hemorrhagic and/or cystic changes can be observed in larger tumors (>5 cm). All sides of margins need to be labeled and grossed to examine if any part of the margins contains tumor cells or not (Figure 6).Histologically, DFSP is derived from fibroblasts in the dermis and subsequently it can infiltrate into the subcutaneous tissues or it can develop directly from subcutaneous tissues [64]. In non-protuberant stage I with dermal plaque or subcutaneous thickening, the elongated spindle cells are loosely scattered in the upper dermis without involving grenz zone (Figure 7A). In stage II and later with protuberant lesions, DFSP is typically featured with uniformly monomorphous spindle cells, with little atypia and mitotic activity, arranged in a storiform pattern, in the subcutaneous and dermal layers (Figure 7C). The cellular nuclei are elongated with mild hyperchromasia, small to inconspicuous nucleoli and low to moderate quantities of cytoplasm. The neoplastic cells often infiltrate into subcutaneous adipose tissue in a honeycomb pattern (Figure 7D). This poses challenge to determine the true extent of the tumor tissue. All margins need to be carefully grossed and examined for residual tumor cells (Figure 6). Immunohistochemically, spindle cells typically show strong and diffuse cytoplasmic expression of CD34 (Figure 7B), but negative expression for other immunohistochemical stains, such as alpha-smooth muscle actin, factor XIIIa, S-100 and melan-A. It should be noted that CD34 expression could be reduced or even lost in up to 45% of the fibrosarcomatous DFSP (Figure 7I) [13,65]. CD34 expression is not unique to DFSP. Other tumors, including spindle cell lipomas, fibromas, fibromyxomas and Kaposi sarcomas, also express CD34.DFSPs have multiple histological variants including myxoid (Figure 7E), pigmented, giant cell (Figure 7F), giant cell fibroblastoma, granular cell, sclerotic (Figure 7G) and fibrosarcomatous (FS) component (Figure 7H, Table 1). These variants reflect the morphologic heterogeneity which is associated with the spindle cell differentiation during tumor development. They do not bear significant clinical manifestations and outcomes, except for the FS variant with increased risk of local recurrence and metastatic potential [28].DFSP without FS component is categorized as classic or conventional DFSP, accounting for 80–90% of all DFSPs [66]. It is considered a low-grade malignancy with a propensity of local recurrences following resection, but almost no metastatic potential [67,68]. FS-DFSP is featured with atypical spindle cells with significantly increased mitotic activity arranged in a herringbone pattern (Figure 7H), often with reduced or even loss of CD34 expression (Figure 7I), but increased Ki-67 expression (a marker of cellular proliferation) [13]. FS-DFSPs consisting of 10–20% of DFSPs are considered intermediate-grade sarcoma with a higher metastatic risk (5–15%) [13,67]. Analysis of 24 reports containing 1422 patients with DFSP and 225 with FS-DFSP revealed that FS-DFSPs, compared with classic DFSPs, had a significantly higher risks of local recurrence (29.8% vs 13.7%, risk ratio 2.2 (95% confidence interval 1.7–2.9)); metastasis (14.4% vs 1.1%, risk ratio 5.5 (95% confidence interval 4.3–7.0)); and death from disease (14.7% vs 0.8%, risk ratio 6.2 (95% confidence interval 5.0–7.8)) [66].Similarities in clinical manifestations and overlaps in histopathologic and CD34 immunostaining profiles with other tumors require DFSP to be differentiated from other benign and malignant lesions including dermatofibroma, schwannoma, cutaneous neurofibroma, solitary fibrous tumor, intradermal spindle cell lipoma and spindle cell or desmoplastic melanoma. Detailed evaluation of the clinical presentation and morphologic features with immunohistochemistry are needed to make an accurate diagnosis (Table 2). Difficult cases can be further tested by molecular techniques including FISH and RT-PCR to detect gene rearrangements and gene fusion transcripts in formalin fixed, paraffin embedded tumor tissues [79].No standard staging system of DFSP is available [86]. We propose a modified staging system of DFSP based on European consensus-based interdisciplinary guideline [86], the progression of DFSPs’ tumorigenesis and clinical presentation, as shown in Table 3. This staging system is useful for treatment.Surgical excision is the standard treatment of DFSP including stage I and II, even III and IV whenever feasible. Wide undermining following surgical excision is not advisable as it may seed tumor in incomplete resections and also may cause difficulty in interpreting subsequent re-excisions. Initial resected tumors with positive margins or relapsed/recurrent tumors need to be further resected to achieve wide clear margins whenever possible [41]. Surgical reconstruction should be delayed until all margins are confirmed negative by complete peripheral and deep margin examination. If concern exists for positive surgical margins following wide excision, a split thickness skin graft may be placed to facilitate monitoring for recurrence.Surgical techniques include wide local excision (WLE) with tumor-free margins, Mohs micrographic surgery (MMS) (Figure 1B–D and Figure 9B–D), partial or total amputation if the tumor is located on the upper or lower digits [28,87,88]. Both WLE and MMS are used in the clinical practice and each has advantages and drawbacks, as summarized in Table 4.Multiple studies have shown that MMS can significantly lower the risk of recurrence of DFSP, compared with WLE [88,90,91]. A comprehensive retrospective meta-analysis involving 684 patients of DFSP published on Medline from 2008 to 2018 revealed that the recurrence rates of DFSP treated with WLE and MMS were 9.10% and 2.72%, respectively, with mean follow-ups of 5.32 years for both groups [92]. Lowe et al. reported the Mayo Clinic experience illustrating a 30.8% recurrence rate following WLE and 3.0% with Mohs surgery; primary closure was performed following MMS in 73% of cases, vs. flaps, graft and other closures in 52% of the WLE cases [90]. Multidisciplinary management is advantageous with infiltrative DFSP of the head and neck as well as large tumors on the trunk, where the Mohs surgeon does the tumor mapping and histologic examination of all tumor margins in concert with another ablative surgeon (Figure 9) [93].The width of the tumor free margins is an important factor to be considered for complete excision for both WLE and MMS. However, no agreement on optimal width of margins is available. NCCN guidelines (version 1.2020) suggest 2–4 cm lateral margins from the tumor and the excision of the investing fascia to remove any infiltrating tumor in WLE [94]. Ratner et al analyzed records of 58 patients with primary and recurrent DFSP treated with Mohs surgery and reported that 70% had positive margins with a 1 cm margin, 39.7% with 2 cm, 15.5% with 3 cm and 5.2% with 5 cm margins [95]. The reported local recurrence (LR) rates with width margins varied widely. Monnier et al. reported a LR of 47% in 66 patients with width margins less than 3 cm at a mean follow-up of 32 months [96], whereas Farma et al. reported LR of 0.9% in 206 patients using 2 cm margins at a follow-up of 64 months [97]. Snow at al recently reported a LR of 1% and a distal recurrence of 1% in 98 patients at a follow-up of 53 months [98]. Among them, 44 patients with microscopically incompletely excised DFSP were treated with conservative re-excision with a mean width of 1.54 cm, and 54 patients with primary tumors were excised with a mean margin of 2.4 cm [98]. Harati analyzed 68 patients with DFSP and observed that 2 cm width margins of normal tissues in primary tumors and in incompletely resected tissues around the scar yielded median negative margins of 0.35 cm and 0.8 cm [99], respectively. These data suggest that narrower margins may be good enough to prevent local recurrence. Mullen suggests to choose WLE with a 1.0 to 1.5 cm safe margins from tumor boundary for most DFSPs on the trunk or extremities since the tumor can be excised in a single stage to achieve excellent cosmetic and functional outcomes, whereas MMS should be selected for relatively small DFSPs in cosmetically sensitive regions including face, scalp (Figure 1B–D) and neck, for best tissue preservation, cosmetic and functional outcomes [89].Dissection of the tumor bed should be based on the infiltrating depth of the tumor. Deep tumors (Stage IIB) should be excised to include the underlying investing fascia of muscle or pericranium whereas superficial tumors (Stage I + IIA) may be directly excised without dissecting underlying fascia. Since WLE usually requires wide and deep excision from the periphery of the tumor, this makes it difficult to treat the DFSP on digits due to the limited space and the complex structures surrounding the toes and fingers. Mohs surgery may allow tissue sparing for DFSP of the digits, however if tumor extends to periosteum, partial or total amputation of the involved digit will be necessary to obtain tumor-free margins and allows for primary closure with faster return to function [22]. When the patient does not agree with amputation due to concern of functional and cosmetic impairments, tumor can be directly enucleated as demonstrated in Figure 2C,D. The patient needs to be referred to oncologists for further radiation therapy and/or targeted therapy. In surgical practice, selection of which procedure for individual patient must be based on tumor location, size, stage, relationship with surrounding neuro-muscular and bone structures, cosmetic and functional requirements, cost to the patient and the medical resources.The FS-DFSP variant is a much more aggressive tumor with local recurrence in more than 50% of patients and metastasis in 10% to 15% of patients [13,100]. FS-DFSP treatment needs consultation with a multidisciplinary specialized soft-tissue sarcoma tumor board [86]. Aggressive treatments for FS-DFSP include adequate WLE with clear surgical margins or MMS (Figure 1B–D). Adjuvant radiotherapy and targeted therapy may be used to reduce the incidence of both local recurrence and metastasis [101].Unresectable DFSPs include advanced stage tumors, recurrent tumors without any possibility for further resection due to the size and/or location, or tumors in which further resection is likely to cause severe functional or cosmetic defects as shown in Figure 2C,D, and multiple organ metastases. These tumors should be treated with adjuvant radiation and/or targeted therapy in consultation with a multidisciplinary specialized soft-tissue sarcoma tumor board. Metastatic DFSP may also be treated with single or multiple agent chemotherapy regimens that are used for sarcomas.Multiple studies have shown that DFSP is a radio-responsive tumor and adjuvant radiation therapy is effective to control tumor growth and reduce the incidence of postoperative recurrence [102,103,104,105,106]. A retrospective analysis of 53 DFSP patients treated with surgery and either preoperative or postoperative radiation therapy in the MD Anderson Cancer Center showed disease-free survival rates of 98% and 93% at 5 and 10 years, respectively [105]. A total of 60 Gy for indeterminate or microscopic positive margins and up to 70 Gy for macroscopic positive margins or primary gross tumor should be given [86,94]. The radiation field should extend 3–5 cm beyond the surgical margins or primary tumor boundary whenever feasible. An individual dose can be given at 2 Gy daily, 5 times weekly [86,94].Imatinib mesylate (IM) is a potent and specific protein tyrosine kinase inhibitor interfering with the phosphorylation and activation of the PDGF receptor β which is constitutively activated due to translocation and fusion between PDGFB and COL1A1 genes as discussed in Pathogenesis section. Recent in vitro and in vivo studies demonstrated that IM had growth-inhibitory effects on DFSP [60,61]. The effectiveness was evidenced by decreased tumor cellularity and formation of strong hyalinic fibrosis in tissues responded to IM treatment [107]. Treatment with IM in FS-DFSP revealed significant upregulation of the cell-cycle inhibitor p21Cip1 and β-galactosidase (a marker of cellular senescence) but decreased Ki-67 [108]. These data suggest that IM therapy is involved in modulation of tumor cell senescent and proliferative activities. IM may also play a role in immune modulation in tumor tissues. IM treated FS-DFSP tumor tissues compared with untreated tissues exhibited increased infiltration of the CD4 and CD8 T-cells in accompany with increased upregulation of cytokines and chemokines including IL-6, transforming growth factor-β1, CXCL-1, IL-1β and IL-8 and CD163+CD14+ myeloid cells transforming to express CD209 [108]. CD163+ macrophages are known to downregulate immune response whereas CD209+ myeloid cells promote T-cell-mediated antitumor responses. These data suggest that IM may exert direct antitumor effects via targeting the PDGFR pathway and indirect antitumor effects via T-cell mediated immune-modulation.The efficacy of IM was observed in localized and metastatic DFSP with t(17; 22), but not in FS-DFSP lacking t(17; 22) [39,109,110,111]. A systemic review of IM treatment of 152 older patients (mean age: 49.3 years) with locally advanced (mean size: 9.9 cm in diameter) or metastatic DFSPs revealed 5.2% of patients with complete response (CR), 55.2% with partial response (PR), 27.6% in stable condition (SC) and 9.2% with progression. There were no differences in response rate using 400-mg or 800-mg daily doses for those with complete or partial responses [112]. A multicenter phase II trial of neoadjuvant IM therapy in advanced primary or locally recurrent DFSP resulted in 7.1% patients with CR, 50.0% with PR, 35.7% with SC and 7.1% with progression [107].Further studies showed that IM used as a preoperative adjuvant therapy in DFSP lead to median tumor volume shrinkages from 20% to 31.5% [39,107,113,114], which transformed very large, unresectable tumors into resectable ones. Wang et al. reported four of 10 patients with primarily unresectable DFSP received complete surgical resection following IM treatment [38]. These studies provide new treatment options for patients with unresectable, recurrent, advanced and metastatic DFSP.It should be noted that about 10% of DFSPs do not respond to IM treatment [38,107,112,115]. Some patients initially responding to IM treatment develop secondary resistance to IM rapidly [115,116]. Therefore, surgical excision for advanced primary tumor following IM therapy should be performed expeditiously during the “shrinkage window”.The mechanism of resistance to IM is not clear yet. A low PDGFR phosphorylation level observed in IM resistant tumor tissues may not respond to IM [107,115]. Sunitinib, with its binding capacity of 10 times greater than that of IM, was demonstrated to be effective in IM resistant DFSP patients [107,115]. However, this effect may not be solely due to inhibiting PDFGR since Sunitinib is a multikinase inhibitor targeting not only PDGFR, but also vascular endothelial growth factor receptors 1–3, KIT, colony stimulating factor-1 receptor and FMS-like tyrosine-kinase-3 (FLT3) [117]. Other gene mutations or signal pathways may play a role in IM resistance. Whole-genomic sequencing in a patient with resistance to IM identified 8 nonsynonymous somatic gene mutations, including ACAP2, CARD10, KIAA0556, PAAQR7, PPP1R39, SAFB2, STARD9 and ZFYVE9 [118]. Single nucleotide polymorphism array and sequencing analysis of DFSP105 (an imatinib-resistant human cell line established from a FS-DFSP) showed a localized 9p21 homozygous deletion, encompassing CDKN2A and CDKN2B, which encode p14ARF, p15INK4b and p16INK4a [119]. p16 inhibits CDK4/6 activity. Loss of p16 expression leads to unrestricted cell cycle progression. Selection of new clonal cells after treatment with either immunotherapy or radiotherapy may also take a part in IM resistant treatment [120].In the cases of IM resistance, other multikinase inhibitors, including sunitinib [107], sorafenib [121] and pazopanib [122], can be considered since treatment with these inhibitors showed effectiveness in IM resistant DFSP patients. In vitro and in vivo studies demonstrated the effectiveness of CDK4/6 inhibitors PD-0332991 and LEE011 in inhibiting DFSP105 proliferation, suggesting that CDK4/6 inhibitors can be potential drugs in p16 negative FS-DFSP [119]. Loss of p16 expression was demonstrated in 2 of 12 classic DFSP and 2 of 6 FS-DFSP [119]. Programmed cell death 1 ligand (PD-L1) expression was detected in metastatic FS-DFSP, but not in the primary tumor, suggesting a role of PD-L1 in the metastasis of FS-DFSP [123]. Since PD-L1 expression in the tumor cells is involved in immune escape from T cell attack [124], PD-L1 signal pathway may be a potential target for metastatic FS-DFSP.The prognoses of the patients with DFSP, after surgical resection with negative and sometimes even positive microscopic margins, are generally good. The five- and ten-year recurrence-free survival rates of DFSP are 86% and 76% [44], respectively. Increased age, high mitotic index, positive margins and increased cellularity are predictors of poor clinical outcome [43]. Local recurrence is a major concern after surgical excision. The frequencies of local recurrences ranged from 20% to 50% [41,44], especially with positive margins. The median time from excision to local recurrence was reported from 32 to 38 months [43,44]. Therefore, mandatory long-term surveillance, at an interval of 6 to 12 months, is recommended [43]. Several studies suggest patients should be reevaluated every 6 months for the first five years and then yearly thereafter [86,99]. A thorough history and clinical examination of the primary sites and draining lymph nodes should be performed at each visit. Further imaging examinations should be considered based on patient’s tumor’s clinical manifestations (size, site, location, rate of growth), surgical procedures and histopathology (presence of high-risk features and margin status). Biopsy should be done on suspected local or distant recurrences or lymph node metastasis.DFSP is a low- to intermediate-grade malignancy frequently occurring in the young to middle-aged population. It is histologically characterized by bland spindle cells in a storiform pattern with multiple variants. This needs to be differentiated from other benign and malignant lesions. A vast majority harbor t(17; 22) (q22; q13) resulting in the formation of COL1A1-PDGFB fusion gene transcript, which holds not only diagnostic value, but also therapeutic significance. WLE with a safe margin should be selected for most DFSPs located on the trunk and extremities. MMS should be considered for relatively small DFSPs in cosmetically sensitive regions to achieve the best tissue preservation with more appealing cosmetic and functional recoveries and can also be used for medium to large DFSP of the trunk and extremities with high cure rates. Subtotal or total digital amputation should be considered for DFSPs located on the digits where complete surgical excision is impractical. Adjuvant therapies—including radiation and targeted therapy—should be chosen for the patients who are unsuitable for surgical excision. Periodic surveillance, at an interval of 6 months in the first five years and then yearly post-excision, is recommended to monitor potential recurrence and metastasis.X.H.: study design, literature review, data collection, analysis and interpretation, writing and revising the manuscript. S.D.B., A.T.V., F.W., T.W.S. and G.W.P.: literature review, data collection, analysis and interpretation, manuscript revision. G.M.: study design, literature review, manuscript revision. All authors have read and agreed to the published version of the manuscript.This research received no external funding.The authors declare no conflict of interest.Clinical presentation and Mohs micrographic surgery (MMS) of a female patient with a primary fibrosarcomatous–dermatofibrosarcoma protuberans. (A) Asymptomatic, ill-defined plaque on her left frontal scalp, with scouting biopsies at inferior aspect; preoperative magnetic resonance imaging showed no bony invasion (Figure 3), PET CT showed no local or distant metastases; (B) surgical marks for MMS. The inner dashed lines indicate the palpable and pathologically positive tumor boundaries while the solid outer lines indicate the surgical incision with the first Mohs layer; (C) wound after first stage of Mohs surgery, all deep and peripheral margins were clear; (D) postoperative at 12 months following a free flap repair. Currently 9 years postop with no evidence of recurrence.Clinical presentation of dermatofibrosarcoma protuberans. (A) Multiple, raised, erythematous, confluent nodules on the vertex of the scalp from an adult male. The inner circle indicates the palpable tumor boundary and the outer line indicates the surgical excision during Mohs surgery; (B) recurrent, skin colored, raised mass around the scar in the right lower abdomen from an adult female; (C) skin-colored, raised tumor on the lateral side of the right hallux from an adult male; (D) white to yellow, solid, fish flesh-like soft tissue mass enucleated from figure C with patient’s agreement. Tumor cells were observed on all margins. The patient was referred to an oncologist for radiation therapy since any further invasive surgery was declined.Magnetic resonance imaging of fibrosarcomatous-dermatofibrosarcoma protuberans on the left forehead from patient in Figure 1. (A) Left scalp enhancement (arrow) in axial post-gadolinium thin slice image; (B) left scalp hyperintensive lesion without bony invasion (arrow) on coronal fat saturated T2-weighted image; (C) left scalp isointense lesion (arrow) before gadolinium on T1-weighted image; (D) left scalp hyperintense lesion (arrow) after gadolinium on T1-weighted image.Magnetic resonance imaging (MRI) of dermatofibrosarcoma protuberans on the left thigh of an 8-month-old girl. (A) MR T1-weighted image exhibiting well-defined homogeneous isointense lesion; (B) T2-weighted image showing a well-defined mass with intermediate-to-marked homogeneous hyperintensity with infiltration of the adjacent fat plane and encasement of the gracilis muscle at the level of this image.Magnetic resonance imaging (MRI) and computed tomography (CT) of a locally recurrent dermatofibrosarcoma protuberans (DFSP) of the right hip. (A) axial and (B) coronal MR T1-weighted fat suppressed post-contrast images of the right hip show enhancing tumor (arrows) diagnosed as a recurrent DFSP; (C) axial and (D) coronal corresponding images from non-contrast CT demonstrate soft tissue attenuation of the tumor (arrows).Photograph of pathologic grossing and mapping of the deep tumor margins during Mohs surgery for a dermatofibrosarcoma protuberans. All of the tumor margins were labeled and grossed to examine the residual tumor cells.Histopathology of dermatofibrosarcoma protuberans. (A) Elongated spindle cells loosely scattered in the upper dermis without involving grenz zone in early stage (dermal plaque) (HE, 40 × 1); (B) diffuse and strong CD34 immunostaining in spindle cells (same case as A, DAB, 40 × 1); (C) dense spindle cells in a storiform (HE, 100 × 1); (D) spindle cells infiltrating into surrounding fatty tissues forming a honeycomb-like structure (HE, 40 × 1); (E) myxoid variant: Spindle cells in the myxoid stroma (HE, 100 × 1); (F) giant variant: Polymorphic and giant cells admixed with spindle cells (HE, 100 × 1); (G) sclerotic variant: Less than 50% of spindle cells in the hypocellular collagenous stroma (HE, 100 × 1); (H) fibrosarcomatous component (right lower part) with increased atypia of cellularity, hyperchromasia and mitosis in a transition from classic DFSP (upper left part) (HE, 100 × 1); (I) loss of CD34 expression in fibrosarcomatous components compared with classic DFSP part where CD34 was strongly expressed (upper left part, same case as H) (DAB, 100 × 1).Histopathology of different tumors differentiated from dermatofibrosarcoma protuberans. (A) Dermatofibroma: Interlacing fascicles of spindle shaped fibroblasts and histiocytic cells mixed with collagens (HE, 100 × 1); (B) factor XIIIa expression in dermatofibroma (same tissue as A, DAB, 100 × 1); (C) schwannoma: wavy hyperchromatic spindle cells arranged in palisades (Antoni A, upper part) and myxoid hypocellular components (Antoni B, lower part) (HE, 100 × 1); (D) S-100 expression in schwannoma (same tissue as C, alkaline phosphatase red, 100 × 1); (E) neurofibroma: bland serpentine spindle shaped cells and shredded carrot collagens (HE, 100 × 1); (F) neurofibroma: mast cells and lymphocytes interspersed among the spindle cells and tumor stroma (same tissue as E, HE, 100 × 1); (G) S-100 expression in neurofibroma (same tissue as E, alkaline phosphatase red, 100 × 1); (H) CD34 expression in neurofibroma (same tissue as E, DAB, 100 × 1); (I) spindle cell lipoma: bland spindle cells without matured lipocytes embedded in ropey/refractile collagen bundles (HE, 100 × 1); (J) CD34 expression in spindle cell lipoma (same tissue as I, DAB, 100 × 1); (K) melanoma: spindle shaped cells with light scattered pigments (HE, 100 × 1); (L) S-100 expression in spindle cell melanoma (DAB, 200 × 1).Clinical presentation and Mohs map in a patient with a primary dermatofibrosarcoma protuberans on the back. (A) Several irregular, firm confluent nodules on the left upper back; (B) Mohs map of A for the first stage dissection; (C) surgical surface after the primary tumor was dissected out; (D) Mohs map of the second stage dissection.Histopathologic features of different dermatofibrosarcoma protuberans variants.Differential diagnosis of dermatofibrosarcoma protuberans.Staging system of dermatofibrosarcoma protuberans.Comparison between wide local excision (WLE) and Mohs micrographic surgery (MMS).

Med-MDPI/jcm_2/jcm-09-06-01753.txt

@@ -0,0 +1 @@
+Coronavirus disease 2019 (COVID-19), due to the severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2), has become an epidemiological threat and a worldwide concern. SARS-CoV-2 has spread to 210 countries worldwide and more than 6,500,000 confirmed cases and 384,643 deaths have been reported, while the number of both confirmed and fatal cases is continually increasing. COVID-19 is a viral disease that can affect every age group—from infants to the elderly—resulting in a wide spectrum of various clinical manifestations. COVID-19 might present different degrees of severity—from mild or even asymptomatic carriers, even to fatal cases. The most common complications include pneumonia and acute respiratory distress syndrome. Fever, dry cough, muscle weakness, and chest pain are the most prevalent and typical symptoms of COVID-19. However, patients might also present atypical symptoms that can occur alone, which might indicate the possible SARS-CoV-2 infection. The aim of this paper is to review and summarize all of the findings regarding clinical manifestations of COVID-19 patients, which include respiratory, neurological, olfactory and gustatory, gastrointestinal, ophthalmic, dermatological, cardiac, and rheumatologic manifestations, as well as specific symptoms in pediatric patients.The first reported case of a severe acute respiratory syndrome coronavirus 2 (SARS-Cov-2) infection (Wuhan, Hubei Province, China), in December 2019, began the outbreak of a novel coronavirus disease (COVID-19), immediately becoming a huge global health concern. On 30 January 2020, COVID-19 was registered as the sixth Public Health Emergency of International Concern (PHEIC) by the World Health Organization (WHO), which was officially declared as a pandemic on 11 March 2020 [1,2]. Currently, there are approximately 6,500,000 confirmed cases of COVID-19 and more than 384,000 deaths, which were reported in more than 200 countries worldwide [3]. So far, the fatality rate due to COVID-19 varies from 1% to more than 7%, and the main causation remains a respiratory failure; however, the complete course of the disease is still not yet understood [4]. To compare, the mortality rates of the major previous epidemics—a severe acute respiratory syndrome (SARS) and the Middle East respiratory syndrome (MERS)—were estimated at 9.6% and up to 34.5%, respectively (Table 1) [5].Several risk factors are associated with the complications of COVID-19, and these include older age (>65), chronic respiratory diseases, cardiovascular diseases, hypertension, diabetes, and obesity. Acute respiratory distress syndrome (ARDS) is reported to be the most common complication [24,25]. Other severe or fatal complications include pneumonia, type I respiratory failure, sepsis, metabolic acidosis, septic shock, arrhythmia, acute cardiac injury, heart failure, acute kidney injury, bleeding, or hypoxic encephalopathy [26,27,28,29]. So far, males are registered to be infected with a higher prevalence compared to females and the reason is yet undiscovered [30,31]. COVID-19 lasts approximately 6 weeks and the duration, as well as the severity of the disease, depends primarily on the age and the immune system of an infected individual. Clinical manifestations can range from being mild to severe and patients can present as either symptomatic or asymptomatic, but a majority of COVID-19 cases are symptomatic with a moderate case-fatality rate (Figure 1) [25,32].The majority of confirmed cases are aged 30–79 (86.6%), and the highest fatality rate is in a group of patients aged >80 years old [33]. Since an effective antiviral treatment is yet unavailable, clinicians worldwide make a significant effort to develop a vaccine and potential therapeutical drugs to minimize the fatal cases and alleviate the symptoms of COVID-19. Even though COVID-19 has a lower percentage of severity and mortality rates compared to SARS or MERS, it is much more transmissive and contagious and can affect everyone—from infants to the elderly—resulting in a wide spectrum of clinical manifestations [32,34,35].The majority of patients with COVID-19 present common symptoms that include fever, shortness of breath, cough (either with or without sputum), sore throat, nasal congestion, dizziness, chills, muscle ache, arthralgia, weakness, fatigue or myalgia, chest tightness, excessive mucus production with expectoration, hemoptysis, and dyspnea [36,37,38,39,40,41,42]. Even though fever is not the only initial clinical manifestation of SARS-CoV-2 infection, it is considered to be critical [43,44]. Fever, cough, and fatigue are the three most prevalent symptoms in COVID-19 patients [25,45]. Other less characteristic symptoms include headache, diarrhea, abdominal pain, vomiting, chest pain, rhinorrhoea, or pharyngalgia [46,47,48,49]. Approximately 90% of the patients present more than one symptom [50,51]. An approximate proportion of severe versus common cases of COVID-19 is estimated to 1:4 [52]. It is suggested that an early onset of shortness of breath constitutes a poor prognostic factor for patients. Among 81 fatal cases of patients from Wuhan, the most common cause of death was a respiratory failure (46.91%), followed by septic shock (19.75%), multiple organ failure (16.05%), and cardiac arrest (8.64%). Rarer death causes were acute coronary syndrome, malignant arrhythmia, or disseminated intravascular coagulation (DIC) [53]. Zhou et al. reported a case of a COVID-19 patient with a spontaneous pneumomediastinum and subcutaneous emphysema [54]. Clinical characteristics might differ between critically ill and non-critically ill patients [55,56].The majority of patients show bilateral pneumonia and only a small percentage of COVID-19 patients show unilateral pneumonia. The most frequent computed tomography (CT) findings are bilateral patchy shadows and ground-glass opacities (GGO); multilobe involvement and focal lesions (patches, stripes, or nodules) are also very characteristic [57,58,59,60]. Less characteristic CT findings include centrilobular nodules, tree-in-bud sign, cystic change, pleural effusion, interstitial fibrosis, or lymphadenopathy. CT examinations show that lesions are more likely to be localized in the periphery than in the center of the lungs and the lesions are more patchy than oval [61,62]. Other CT findings include either pure GGO or GGO with reticular and/or interlobular septal thickening, GGO with consolidation, or pure consolidation [63,64]. Less common, but still characteristic, CT findings include ground-glass followed by irregular or halo sign, air bronchogram, bronchovascular bundle thickening, grid-form shadow, and hydrothorax [57]. Ground glass-like shadows, fibrous stripes, patchy shadow, and pleural thickening are observed both in common-type and severe or critical-type patients, independent to the severity of the COVID-19 course [65]. Single or multiple lobes of a single lung or both lungs (without a characteristic pattern) can be affected; interestingly, some of the studies showed that severe critical-type patients exhibit lesions primarily in the right lung [66].Generally, COVID-19 patients tend to have normal or decreased white blood cell counts, lymphopenia, or thrombocytopenia [67,68]. Zhang et al. showed that patients with high leukocyte count (>10 × 109/L), higher neutrophil count (>7 × 109/L), and lower lymphocyte count (<0.4 × 109/L) are much more prone to severe COVID-19 pneumonia and composite endpoint (which was the admission to an intensive care unit, mechanical ventilation, or death) [69]. Besides, higher levels of C-reactive protein (>150 mg/L) and increased D-dimer levels (>1 mg/L) are also strongly associated with an increased risk of COVID-19 pneumonia and the composite endpoint. Additional laboratory indicators of increased risk are higher alanine aminotransferase (ALT) activity (>80 U/L), higher aspartate aminotransferase (AST) activity (>80 U/L), higher α—hydroxybutyrate dehydrogenase activity (>540 U/L), higher lactate dehydrogenase activity (>720 U/L), higher creatine kinase activity (>600 U/L), and lower total protein level (<60 g/L). So far, researchers have not observed a significant statistical association between altered platelet counts and creatinine levels with an increased risk of COVID-19-related pneumonia. As opposed to numerous studies, Zhang et al. showed that COVID-19 pneumonia and composite endpoint are associated with leukocytosis rather than leukopenia [69]. However, the abovementioned results differ among COVID-19 patients. Du et al. observed that the majority of COVID-19 patients (81.2%) had lowered eosinophil count and many patients had decreased hemoglobin and hematocrit, as well as decreased activated partial prothrombin time (APTT) and increased prothrombin time (PT) [53]. Among studied patients, 22.4% had increased procalcitonin levels and elevated levels of blood urea nitrogen or serum creatinine. It is still speculated whether eosinophilopenia might constitute a prognostic factor for COVID-19 patients. Some patients present progressive lymphopenia with a concurrent progressive neutrophilia [70]. However, among the most common reported laboratory findings, those of the highest prevalence include elevated levels of C-reactive protein and erythrocyte number, as well as increased myohemoglobin, liver enzymes, and muscle enzymes [25]. Additionally, patients with a severe course of COVID-19 usually have elevated D-dimer levels, increased procalcitonin, increased leukocyte number, and lymphocytopenia [29,71]. In some cases, lymphocytes and white blood cell levels might remain within physiological ranges. The decrease in the number of lymphocytes is generally observed in the CD4+ subpopulation. No significant changes are stated in the case of CD8+ and B cell subpopulations [72]. Further, interleukin 10 (IL-10), interleukin 6 (IL-6), interleukin 1 (IL-1), interleukin 2R (IL-2R), and tumor necrosis factor alpha (TNF-α) levels might exceed the upper limit in COVID-19 patients [73,74,75,76]. Chemokines, such as interferon gamma-induced protein 10 (IP-10) and monocyte chemoattractant protein 1 (MCP1), are also overexpressed during the course of COVID-19 [74].Besides severe clinical manifestations, primarily of the respiratory system, SARS-CoV-2 presents neurotropic properties [75]. Autopsies have revealed the presence of SARS-CoV-2 nucleic acid in both cerebrospinal fluid and brain tissue of infected patients [77,78]. The entering of SARS-CoV-2 into the central nervous system is possible, either via hematogenous, lymphatic, synapse-connected, or retrograde neuronal routes [79,80]. Neuroinvasion of SARS-CoV-2 and the presence of neurological manifestations might be an explanation of the presence of neurological impairments without other typical symptoms of infection, especially in asymptomatic patients.Neurological manifestations might occur in both symptomatic and asymptomatic patients. Neurologic manifestations are commonly described in COVID-19 patients, and these might involve the central nervous system, peripheral nervous system, and skeletal muscles [81]. Patients with a severe course of COVID-19 are more likely to develop neurological dysfunctions, among which acute cerebrovascular disease, conscious disturbance, and skeletal muscle injury are highly prevalent [82]. Helms et al. reported that patients with ARDS due to SARS-CoV-2 infection also presented encephalopathy, prominent agitation and confusion, acute ischemic strokes, or corticospinal tract signs [83]. Some patients manifest only neurological symptoms, including headache, languidness, malaise, cerebral hemorrhage, or cerebral infarction [84,85]. Cases of encephalitis, necrotizing hemorrhagic encephalopathy, strokes, epileptic seizures, or rhabdomyolysis associated with SARS-CoV-2 infection have also been described [86]. Similar to adults, neurological findings might appear in the case of infected infants, and this is so far reported as the observed upward gaze, dystonic bilateral leg extension, and alterations in a child’s responsiveness [87,88].Duong et al. reported a case of a female with meningoencephalitis with concurrent hallucinations and disorientation without respiratory manifestations [89]. Other studies reported a possibility of the occurrence of the Miller Fisher syndrome, polyneuritis cranialis, or encephalopathy in COVID-19 patients [90,91]. Another neurological disease associated with COVID-19 is the Guillain–Barré Syndrome, reported as a neurological complication due to SARS-Cov-2 infection in several patients so far [92,93,94,95]. Detailed clinical, neurological, and electrophysiological examinations are crucial to assess neurological symptoms of COVID-19 patients. Additionally, the abovementioned examinations are highly important, since neurological manifestations could appear alone and might present as non-specific symptoms in patients infected by SARS-CoV-2.Isolated sudden-onset anosmia is reported to be the fourth the most common symptom of SARS-Cov-2 infection [96,97,98,99]. Further, patients who present sudden olfactory and/or gustatory dysfunctions irrespective of co-existing symptoms should be suspected of SARS-Cov-2 infection [100,101,102]. However, the pathogenesis of olfactory and gustatory dysfunctions in COVID-19 is still undiscovered. Approximately 79.7% of COVID-19 patients without nasal obstruction or rhinorrhea report hyposmia or anosmia [103]. The occurrence of fever is highly associated with the olfactory dysfunctions, and these may appear before, during, or after the general symptoms. The severity of olfactory dysfunctions differs from complete anosmia to severe, moderate, or mild microsmia or normosmia. Olfactory dysfunctions might persist even in up to 56% of patients who were reported as recovered from COVID-19 [103]. After the recovery, some of the olfactory dysfunctions might persist and gustatory dysfunctions might be resolved, and vice versa. There are also cases of complete losses of olfactory functions [104]. Anosmia might constitute the only symptom of COVID-19 [96,105,106]. The mean duration of smell and taste disorders due to SARS-Cov-2 is estimated at 7.5 days [107]. It was suggested that the presence of olfactory dysfunctions might constitute a potential (but limited) marker of SARS-Cov-2 infection [108,109]. Furthermore, loss of smell in COVID-19 patients might be associated with a milder clinical course of the disease [110].A significant number of studies indicate that SARS-CoV-2 actively infects and replicates within the gastrointestinal tract, inducing digestive symptoms primarily via overexpression of viral receptor angiotensin-converting enzyme 2 (ACE2), found in gastrointestinal epithelial cells [111]. SARS-CoV-2 can be detected in the esophagus, stomach, duodenum, and rectum. It can also be found in the fecal samples [22,112,113,114]. Furthermore, it was shown that negative results from the nasopharyngeal swabs do not exclude viral infection, since the virus might be detected only in the rectal swabs [115]. The most common digestive symptoms in COVID-19 patients include nausea and/or vomiting, diarrhea, anorexia, or loss of appetite [116,117,118,119,120]. Rarer digestive symptoms include abdominal pain, abdominal distension, tenesmus, dysgeusia, gastrointestinal bleeding, or hematochezia [121,122,123,124]. Findings of Nobel et al. indicate that the presence of gastrointestinal dysfunctions might be associated with a more indolent form of COVID-19; such patients might present longer duration of the disease course [125]. Likewise, the severity of COVID-19 is associated with the more pronounced gastrointestinal manifestations. Gastrointestinal manifestations might constitute the only symptoms of SARS-CoV-2 infection without the impairments from the respiratory system or fever [126,127]. Further, COVID-19 patients might present gastrointestinal dysfunctions before the occurrence of other symptoms [128]. Despite prolonged prothrombin time and lowered monocyte counts, no significant differences were found in the complete blood count, electrolytes, or kidney functions in COVID-19 patients with gastrointestinal impairments [129].Apart from the gastrointestinal manifestations, SARS-CoV-2 infection might involve liver impairments of a wide spectrum of a severity degree [125,130,131]. COVID-19 patients show increased levels of ALT and AST. Furthermore, serum bilirubin and gamma-glutamyl transferase (GGT) might also be elevated during the course of the disease [131,132,133,134,135]. Elevated levels of ALT and AST might be observed both in severe and non-severe cases of COVID-19 [136]. However, so far, it has been reported that liver injury due to SARS-CoV-2 infection occurs more prevalently in severe cases rather than mild cases of COVID-19. Lagana et al. reported a case of hepatitis associated with COVID-19 [137]. It must be mentioned that the pathological mechanism of liver injury is not yet understood; however, the possible mechanisms include direct viral infection of hepatocytes, drug hepatotoxicity, binding to cholangiocytes via ACE2 receptors, or immune-related injuries [132,138,139].Coronaviruses are capable of inducing a wide spectrum of ophthalmic manifestations, such as conjunctivitis, anterior uveitis, retinitis, or optic neuritis [140]. SARS-CoV-2 presents its ability of the ocular transmission, which might result in ocular manifestations; however, the prevalence of such incidents is extremely low [141,142]. Similarly, to other symptoms not related to the respiratory system, ophthalmic manifestations might appear as the first symptom without any other impairments. Besides, ocular impairments are rather more prevalent in patients with a severe course of the disease. The prevalence of ocular manifestations varies from 2% to 32% [143]. Compared to standard nasopharyngeal samples, the sensitivity of ocular swabs in SARS-CoV-2 detection is very low. It was reported that SARS-CoV-2 RNA can be detectable in ocular swabs days after being undetectable in the nasal swabs [144]. Ocular manifestations might occur relatively early during the COVID-19 course. It was reported that ophthalmic manifestations might be associated with the severity of the COVID-19 course [145]. Wu et al. showed that, among 38 infected patients, 12 presented ophthalmic manifestations, such as conjunctivitis, conjunctival hyperemia, chemosis, epiphora, or increased secretions [146]. Ocular manifestations primarily include the onset of conjunctivitis, keratoconjunctivitis, or ocular irritation symptoms [147,148,149,150,151]. Daruich et al. reported a case of an infected patient with unilateral eyelid edema and moderate conjunctival hyperemia [152]. Conjunctivitis can be the first symptom of SARS-Cov-2 infection [153].The SARS-CoV-2 infection has been reported to manifest in the form of cutaneous symptoms. The first report of skin involvement in COVID-19 patients was observed in the form of an erythematous rash, widespread urticaria, and chickenpox-like vesicles, especially occupying the trunk [154]. Mahé et al. reported a case of an infected patient with a distinctive skin rash [155]. It was suggested that, in some cases, skin lesions might constitute a late manifestation of COVID-19, especially in young healthy individuals, and might appear due to the immunological reactions [156]. Nevertheless, there are cases in which dermatological manifestations (e.g., acute urticaria with pyrexia) might occur first before other (more or less) characteristic symptoms [157]. COVID-19 patients might appear with the herpetiform lesions primarily located on the trunk; these lesions can be characterized by vesicles surrounded by erythematous halos with mild pruritus; vesicles might also form crusts [158,159,160]. Other researchers showed that skin lesions might appear as intensely pruritic, in form of a petechial rash, or even as an urticarial eruption [161,162,163]. Zulfiqar et al. reported a case of a COVID-19 patient with immune thrombocytopenic purpura [164]. Cutaneous manifestations associated with SARS-CoV-2 infection also include maculopapular exanthem, papulovesicular rash, urticaria, livedo reticularis lesions, or petechiae [165,166]. Since skin lesions in COVID-19 patients might be similar to those during the course of dengue, some patients might be easily misdiagnosed [167].Although cardiovascular diseases might significantly worsen the clinical outcome of COVID-19 patients, SARS-CoV-2 infection might also induce cardiac complications de novo [168,169]. Cardiac impairments might occur even without any symptoms or signs of pneumonia. The pathophysiological mechanisms probably involve ACE2 receptors, a cytokine storm induced by the imbalanced response between type 1 and 2 T-helper cells or strong interferon-mediated immunopathological events [43,170,171,172]. Further, atrial fibrillation, which is the most prevalent causation of arrhythmias, might be triggered by COVID-19-related hypoxia and the complications can persist even after pulmonary recovery [173]. Cardiac impairments might also be associated with pharmacological interventions (drugs currently used during COVID-19 treatment might prolong the QT interval or can be proarrhythmic). The most prevalent cardiovascular complication of COVID-19 is an acute myocardial injury (usually defined as an increase in cardiac troponin I above the 99th percentile upper reference limit), with a prevalence of 8%–12% [174,175,176]. An elevation of troponin levels is rather observed in a severe course of COVID-19, compared to mild or moderate courses. Increased levels of creatine kinase myocardial band (CK-MB), myohemoglobin, cardiac troponin I, and N-terminal pro-brain natriuretic peptide are associated with the severity of COVID-19 [177,178,179]. Other most prevalent complications include either brady- or tachyarrhythmias, with an estimated incidence of 16.7%, acute pericarditis, left ventricular dysfunctions, heart failure, cardiogenic shock, blood pressure abnormalities, or myocarditis [180,181,182,183,184]. It is very prevalent among COVID-19 patients that cardiac manifestations coexist with respiratory impairments [185]. Myocardial injury associated with SARS-CoV-2 infection impairs cardiac functions and induces ventricular tachyarrhythmias [186]. Inciardi et al. reported a case of a COVID-19 patient with an acute myopericarditis [187]. Zeng et al. described the first case of fulminant myocarditis as a COVID-19 complication [188]. The Kawasaki-like disease has been recently described as a post-infectious inflammatory syndrome that might constitute a complication of the COVID-19 disease, especially among pediatric patients [189,190]. Furthermore, acute pulmonary embolism and aortic thrombosis might be non-characteristic presentations in COVID-19 patients [191,192,193,194]. Similar to other viral types of pneumonia, patients infected by SARS-CoV-2 are at a higher risk of an acute pulmonary embolism. COVID-19 patients with pulmonary embolus have higher D-dimer levels compared to infected patients without pulmonary embolism [195,196]. Thus, a potential association between COVID-19 and pulmonary embolism should be taken into consideration, especially among patients with high D-dimer levels and without other clinical manifestations typical for COVID-19. Additionally, arterial and venous thromboembolic events are quite common cardiovascular manifestations among COVID-19 patients, which indicates a crucial role of COVID-19-associated coagulopathy [197,198,199]. A prominent elevation of D-dimer levels and higher levels of fibrin/fibrinogen degradation products are the most prevalent presentations of COVID-19-associated coagulopathy during the initial stages; altered coagulation parameters might be associated with poorer clinical outcomes of patients [200,201]. It was reported that antiphospholipid antibodies might be tested positive in COVID-19 patients and their presence might (rarely) induce thrombotic events [194]. The mortality rate of patients with cardiac injury due to SARS-CoV-2 infection is much higher compared to those without cardiovascular complications [202,203]. It was estimated that coagulation dysfunctions constitute the major cause of death in severely ill COVID-19 patients [204]. An endomyocardial biopsy of a COVID-19 patient with a cardiogenic shock showed that viral particles are not detected in myocytes specifically, but they can be detected in the interstitial cytopathic macrophages and their surroundings; myocytes were only characterized by focal myofibrillar lysis [205]. COVID-19 patients might also develop endothelial cell infection or endotheliitis, as it was observed that viral particles can be detected within endothelial cells, causing diffuse endothelial inflammation [206]. Cui et al. reported a case of an infected 55-days-old infant who, despite pneumonia, had liver injury and heart damage due to SARS-CoV-2 infection [207]. Hua et al. reported the first case of a COVID-19 patient with cardiac tamponade [208].So far, data on the rheumatic manifestations in COVID-19 patients is still limited. It was reported that arthralgia might be an initial presentation of COVID-19 and this manifestation might be easily missed, especially in regions where the viral arthropod-borne disease is relatively common [209]. According to studies, lupus patients are more susceptible to SARS-CoV-2 infections and the course of COVID-19 might be very complicated in such patients [210]. Similarly, patients with rheumatoid arthritis are much more susceptible to SARS-CoV-2 compared to the general population [211].The amount of literature regarding clinical manifestations of COVID-19 in pediatric patients is continually increasing. Children usually present mild symptoms or might remain asymptomatic [212]. Pediatric patients with other comorbid diseases are much more vulnerable to SARS-CoV-2 infection and a more severe course of COVID-19 [213]. Generally, the majority of pediatric patients have mild symptoms, without fever or pneumonia, and the recovery time is estimated to be 1–2 weeks after the onset of the disease [214,215]. Such a mild course of COVID-19 in pediatric patients might be due to several reasons, including more effective immune responses in children, differences in the expression of the ACE2 receptor, or a simultaneous presence of other viruses in the respiratory tract of children, which might limit SARS-CoV-2 infection [216,217,218]. Children with COVID-19 present longer incubation periods compared to adults, estimated for 6.5 days in children and 5.4 days in adult patients [213]. The most common clinical manifestations in children include fever and cough; in some cases, additional symptoms, such as fatigue, myalgia, nasal congestion, runny nose, sneezing, sore throat, headache, vomiting, dizziness, or abdominal pain might be present [219,220,221,222,223,224]. There are also incidents of asymptomatic pediatric patients or those who only exhibit cough or diarrhea [225,226]. In some cases, infected infants of children might present typical symptoms, such as gastrointestinal manifestations, asthma, or shortness of breath alone [227]. Viner and Whittaker have reported that the Kawasaki-like disease might constitute a complication of the COVID-19 course, primarily in a population of pediatric patients [228]. Those atypical symptoms might significantly delay the diagnosis of COVID-19 in pediatric patients, potentially worsening their clinical outcomes [229]. Pediatric patients might have common clinical manifestations, such as fever or mild pneumonia, but a lot of reported patients have neither obvious symptoms nor abnormal CT findings [230,231]. However, the severe cases might progress to acute respiratory distress syndrome, septic shock, refractory metabolic acidosis, and coagulation dysfunctions [232,233].The prevalence of SARS-CoV-2 infection among newborns and infants is very low but still possible and can manifest as asymptomatic, mild, or severe infection. So far, there is no evidence for the vertical transmission of the virus from mother to newborn; however, an infection is possible, primarily due to the close contact with an infected mother [234]. Newborns and infants infected by SARS-CoV-2 might present such manifestations as fever or mild upper respiratory symptoms alone but can also remain asymptomatic [235]. Generally, newborns and infants remain asymptomatic and present mild, non-specific symptoms, including cough, headache, runny nose, nasal congestion, expectoration, tachypnea, apnea, tachycardia, lethargy, vomiting, or diarrhea [236,237]. Gastrointestinal symptoms (diarrhea, food aversion, abdominal distension) are primarily observed among smaller infants [238]. Acute respiratory distress syndrome and temperature instability, as well as gastrointestinal and cardiovascular dysfunctions, are non-specific clinical features of infected (in particular preterm) infants and newborns [236]. Atypical clinical features of infants with COVID-19 also include neurological symptoms, such as axial hypotonia, drowsiness, or moaning sounds [88].Regarding laboratory findings in pediatric patients, the white blood cell counts might be normal or decreased. Some patients might have leukopenia and only a small percentage present with lymphocytopenia. Severe or critical cases of pediatric COVID-19 patients might present elevated hepatic and muscular enzymes, as well as increased D-dimer levels [239,240]. Imaging findings usually present ground-glass opacities and segmental consolidation with surrounding halo signs, which is considered to be a typical sign in pediatric patients [239,241,242]. Regarding immunocompromised pediatric patients, and those with chronic diseases, current data does not suggest that such populations are at higher risk of severe infection [243].The outbreak of SARS-CoV-2 infection, which started in Wuhan, China, in December 2019, has now become a global concern, being reported in more than 200 countries. The understanding of COVID-19, its diagnosis, transmission routes, molecular mechanisms of infection, prevention, and treatment strategies are rapidly evolving. Compared to previous infections of the severe acute respiratory syndrome-related coronavirus (SARS-CoV) or the Middle East respiratory syndrome-related coronavirus (MERS-CoV), SARS-CoV-2 is much more transmissive and dangerous and might affect nearly everyone, resulting in a wide spectrum of clinical manifestations. SARS-CoV-2 does not only affect the respiratory tract, resulting in pneumonia, but can affect the gastrointestinal, nervous, or cardiovascular systems. Less typical manifestations include dermatologic or ophthalmic manifestations. However, the pathomechanisms of the abovementioned manifestations are yet undiscovered in the majority of cases. The course of COVID-19 can be mild, moderate, severe, or critical; the number of asymptomatic carriers is also very high, worsening the epidemiological situation. Even though the long-term complications are unknown, pneumonia, acute respiratory failure, acute respiratory distress syndrome, acute liver or kidney injury, cardiac complications, septic shock, or coagulopathy are described, so far, as being the most prevalent. Some of the clinical manifestations that are not typical might appear first, predicting COVID-19; therefore, the knowledge about them is inherent.Conceptualization, J.B., H.K.-J., G.T. and R.M.; methodology, M.C., G.B. and R.M.; formal analysis, A.F. and E.S.; writing—original draft preparation, A.F., K.K., and W.F.; writing—review and editing, P.P., H.K.-J., J.B. and R.M.; visualization, J.B.; supervision, R.M. All authors have read and agreed to the published version of the manuscript.This research received no external funding.The authors declare no conflict of interest.The severity of SARS-CoV-2 infection with typical characteristics.Clinical features of severe acute respiratory syndrome coronavirus (SARS-CoV), Middle East respiratory syndrome coronavirus (MERS-CoV), and severe acute respiratory syndrome coronavirus-2 (SARS-CoV-2 infection).Ground-glass opacitiesLung consolidation: focal, multifocal, or diffuse (primarily peripheral)Lung involvement: unilateral (two-thirds of patients) or bilateralLesions: distributed within the lower lobes of the lungs [9]Ground-glass opacitiesLung consolidationLung involvement: bilateral (80%) or unilateral (20%)Pleural effusionIntralobular septal thickening [10]Ground-glass opacities: single or multiple focalLung consolidationPatchy consolidative opacitiesPulmonary nodulesInterlobular septal thickeningBronchial wall thickeningLesions: usually bilateral, peripheral, and distributed within the lower lobes of the lungs [11,12]Close (droplets) contact with symptomatic patients [16]Contaminated surfaces [17]Contact with infected camels or consumption of contaminated milk or meat [18]Limited human-to-human transmission (via droplets) [19]Close (droplets) or distant (aerosol particles) contact with symptomatic or asymptomatic patients [20]Contaminated surfaces [21]Fecal transmission [22]

Med-MDPI/jcm_2/jcm-09-06-01754.txt

@@ -0,0 +1 @@
+These authors share co-first authorship.These authors share co-last authorship.Early diagnosis of oral cancer through visual inspection followed by histopathological confirmation is a pivotal step for reducing rates of morbidity and mortality. There are several auxiliary devices used to improve oral examination. The purpose of this cross-sectional pilot study is to evaluate the sensitivity and specificity of the Visually Enhance Lesion Scope (VelScope) system when it is used by the general dentist after a yearly oral medicine training. Thirty-five patients with oral lesions were evaluated with clinical and VelScope examination by two general dentists, one of which trained with a specific course. A comparison of the histopathological results, clinical examination, and VelScope made by both dentists was performed through statistical analysis. Sensitivity, specificity, positive predictive value (PPV), and negative predictive value (NPV) for detecting oral potentially malignant disorders (OPMDs) are 53.3%, 65%, 53.3%, 76.5% for unskilled dentist, 73.3%, 65%, 61.1%, 76.5% for skilled clinician. When both examiners use VelScope the values are 53.3%, 70%, 57.1%, 66.7% for unskilled general dentist (u-GD), 86.7%, 90%, 86.7%, 90% for skilled general dentist (s-GD). Improvement of a skilled general dentist for detecting malignancies is higher than inexperienced examiner when using VelScope. VelScope alone is unable to improve the general dentist’s ability to detect malignancies, but it could be a useful adjunctive device for clinicians when a focused training program is performed.Cancer is one of the most common causes of morbidity and mortality, with over 10 million new cases and over six million deaths each year in the world, and it is also responsible for 20% of all deaths in high-income countries and 10% in low-income ones [1]. Oral squamous cell carcinoma (OSCC) is a pathology still scarcely known by the general population, who is often unaware that the oral cavity may represent the place of the onset for malignant modifications. The symptoms are not interpreted correctly and are often ignored in the early stages of the pathological process. Despite the improvements in medical and surgical techniques, the prognostic data have not changed significantly over the last 10 years and are usually linked to the size of neoplasia and lymph node involvement: almost 85% of five-year survival for stages I/II, 25% at 5 years for stages III/IV, whereas the total five-year survival rate is around 50% [2]. It is clear that early diagnosis, therefore secondary prevention, is the only means to improve the prognostic pathway through the clinical oral examination (COE), where chromatic alteration, more or less accompanied by alterations of consistency and thickness of the tissue need to be detected. The oral cavity is an easily accessible area for COE with a reported sensitivity equal to 93% and a specificity equal to only 31% [3]. Still today, COE, together with histological examination, represents the “gold standard” for the definitive diagnosis. In literature, there are several auxiliary devices to improve COE, which should simplify the diagnostic process, such as Identafi, Narrow Band Imaging, Oral Cdx, ViziLite, and Visually Enhance Lesion Scope (VelScope) [4]. VelScope works through the emission of light with a wavelength of between 400–460 nm, which arouses the fluorophores in the oral epithelium. Normal tissue should show a normal green aspect on the VelScope examination (Vel-E), while dysplastic tissue should show darker aspects in comparison with the surrounding areas due to the loss of fluorescence (LAF) [5,6]. In the presence of LAF, the Vel-E is defined positive; it is due to several alterations, particularly: (i) break-up of the collagen cross-links, (ii) increase in blood supply due to micro-vascularization and inflammation, and (iii) reduction in flavine–adenine–dinucleotide (FAD) and nicotinamide-adenine-dinucleotide (NADH) (fluorochromes). These alterations could be present both in tissues with dysplasia and in tissues with ongoing inflammatory phenomena, thereby reducing the reliability of the test [7].Many studies have highlighted several deficiencies in the knowledge and attitudes of primary health care dentists regarding OSCC [8,9]; therefore, strategies to improve diagnostic skills of general dentists are needed. In a new scenario, a cross-sectional pilot project was planned to explore the knowledge and clinical skills of general dentists (GDs) on the use of Vel-E, in cooperation with an oral medicine team. A one-year period of post-graduate oral medicine training was scheduled in order to improve clinical competence in patients with suspected oral lesions and discriminate benign oral diseases from oral potentially malignant disorders (OPMDs)/OSCC. The purpose of the project is to evaluate whether the VelScope system improves the diagnostic ability to detect OPMDs when it is used by inexperienced examiners.From December 2018 to November 2019, we asked two GDs to perform COE and Vel-E on consecutively selected patients during the first visit, one of which after a 1-year training program in oral medicine at the Oral Medicine Unit, Federico II University of Naples, Italy. All patients signed an informed consent before undergoing the project. The study was conducted in accordance with the Declaration of Helsinki, and the protocol was approved by the Institutional Ethics Committee of the University of Bari (N° 1442). The reporting of data followed the guidelines of the STARD statement.The patients were first subjected to a complete clinical interview, and COE encompasses a manual and visual examination of intra- and extra-oral sites. Based on the clinical characteristics found, both the examiners classified the lesions into benign or malignant and performed Vel-E on the basis of clinical information and in the absence of histological data. The device was used, maintaining a distance of approximatively 5 cm from the oral cavity to optimize the visualization of the natural tissue fluorescence and, if necessary, darken the operating theatre to reduce natural light and bring it to an acceptable level. Subsequently, incisional or excisional biopsy was carried out by an oral medicine specialist with a histological examination by a pathologist at the Interdisciplinary Department of Medicine, University of Bari Aldo Moro, Italy. The margins of the incision are guided by the result of the Vel-E. All specimens were placed in 4% buffered formalin solution for fixation, embedded in paraffin and were cut into 4 μm thick sections, and stained with hematoxylin/eosin. The oral pathologist who examined the biopsy specimens was blind to the Vel-E. The time interval between the Vel-E and the biopsy examination was an average of 15 days; during this period, the patients did not perform any therapy. The results of the histological examination have been compared with those of the COE and VelScope.The clinical photographic documents were obtained by using a Canon® Eos 750D (Tokyo, Japan) camera with Canon® EF-S 60 mm f/2.8 Macro lens, while the VelScope photographic images were produced with a digital camera (Canon) with a specific adaptor.Potentially eligible participants were identified on the basis of the following inclusion and exclusion criteria (Figure 1). The inclusion criteria were: 19–90-year-old patients of both genders, Caucasian, affected by OPMDs as leukoplakia and erythroplakia, oral lichen planus, lichenoid lesions, unspecified keratotic lesions (KUS). The exclusion criteria were patients affected by benign lesions and para-physiological anomalies (such us Fordyce granules, thickened Linea alba, white sponge nevus, hypertrophy of lingual tonsils and of the lymphoid tissue of Waldeyer’s ring, exogenous and endogenous pigmentations, prominent varices of the ventral lingual, morsicatio buccarum, benign alveolar ridge keratosis, geographic tongue, leukoedema).A descriptive statistical analysis was produced through the assessment of sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV) of the Vel-E performed on oral mucosal lesions. Sensitivity and specificity, PPV and NPV of COE, and Vel-E were calculated using 2 × 2 contingency tables. Biopsy and histological examination were used as the gold standard.The clinical and demographic characteristics of the patients are shown in Table 1. Thirty-five patients, 17 males (48.6%) and 18 females (51.4%), with an average age of 58.5 years, were enrolled. The unskilled general dentist examiner (u-GD) classified as benign 20 oral lesions and as malignant 15 oral lesions, instead the skilled general dentist examiner (s-GD) 17 and 18, respectively.The Vel-E was interpreted by u-GD as positive in 14 patients (40%) and negative in 21 patients (60%); instead, VelScope data from s-GD were positive in 15 patients (42.9%) and negative in 20 patients (57.1%) (Figure 2a–h). LAF showed a range of colors from dark green to black, while one case showed up orange/red due to bacterial contamination (Figure 3).On histological examination, 15 (42.9%) of 35 biopsied lesions were malignant, and the other 20 (57.1%) were benign. The suspected clinical diagnosis (benign/malignant) made by both GDs and the VelScope data were compared with the histological results. As seen in Table 2, sensitivity, specificity, PPV and NPV of u-GD for detecting OPMDs are 53.3%, 65%, 53.3%, 76.5% respectively, while the same parameters for s-GD are 73.3%, 65%, 61.1% and 76.5% respectively. Differently, when both examiners use an adjunct device the values of sensitivity, specificity, PPV and NPV change in the following way: for u-GD they were 53.3%, 70%, 57.1%, 66.7%, conversely for s-GD they were 86.7%, 90%, 86.7% and 90% (Table 2).We observed only two cases of false positives (5.71%) in s-GD and 6 (17.14%) in u-GD and two false-negative cases with minimal and poorly detectable LAF in s-GD, and 7 in u-GD.Benign and malignant oral lesions clinically appear with great variability in terms of heterogeneous morphology, onset, and clinical course; this is why histopathological examination remains the gold standard for final diagnosis. Up to now, there are unresolved medical issues in the methodology of published reports, due to selection bias of patients with different OPMDs and different prognosis and to the absence of uniformity in histological classification and description of the grade of dysplasia [10]. The heterogeneity of clinical and histological aspects among experts clearly reflects the diagnostic difficulty to screen oral lesions and does not allow data to be uniformly compared. Awan et al., in a previous study on 164 patients analyzed with Vel-E both lesions clinically suspected for malignancies and clinically benign showed a good sensitivity of 84% with a very low specificity equal to 15%, suggesting that data on specificity could be very different based on the inclusion criteria of the study group [11].To deal with the issue of OSCC diagnostic delay, it is necessary to focus on early diagnosis and prevention systems in the general population [12]. Identification of high-risk individuals and screening systems are part of secondary prevention. An effective oncological screening system has the main goal to find cancer early and to reduce the possibility that patients screened regularly will die from cancer; it is also a preventive measure for a high-risk population. Not all types of cancer currently have an effective screening method, and there is no single approach that fits all situations [13]. There is currently no standard or routine screening test for OSCC; however, a visual inspection should be a routine step in oral health practitioner daily activity [14]. COE is characterized by a simple, non-invasive procedure, that is easy, effective, and cheap [15]. Nevertheless, the lack of evidence does not support COE as part of a population-based screening program, while it reduces the mortality rate of OSCC in “high-risk” patients [16]. The use of adjunctive tests or devices, such as toluidine blue, brush biopsy and fluorescence, can be useful in the evaluation of oral mucosal lesions [17], but it is not evidence-based as well as a useful tool to reduce OSCC mortality in the absence of specialist training [18]. Among the different screening systems described in the literature, VelScope is the one supported by the majority of published studies. The efficacy of VelScope in detecting OPMDs when used by professionals who are experts in the oral diagnosis and/or on lesions clinically suspected for malignancies is widely documented [19]. This is a key point of criticism because the use of VelScope now is “targeted”, usually applied on “high-risk” patients with OPMDs/OSCC and/or entrusted to trained clinicians [20]. Therefore, the high values of sensitivity and specificity reported have not been tested in the general dentistry field; for this reason, VelScope is not supported by robust evidence for its use in primary care, but only in selected specialist clinics where trained physicians may discriminate among OPMDs [21]. However, an effective screening system must be applicable in the context of primary care by all the physicians who take care of oral health (general/specialist dentists, otorhinolaryngologists, general physicians, maxillofacial surgeons).Huff et al. reported in a retrospective pilot study the routine utilization of Vel-E in general dental practice on “low-risk” adults were showing that it could be useful in identifying OPMDs also in this different population setting [22]. We have tried to understand if a training focused course of continuing education on OSCC in general dental practice based on a one-year program with the specific clinical session can improve early diagnosis on high-risk patients when Vel-E is added to COE. Our analysis showed an increase of 13.4% and 25% for sensitivity and specificity, respectively, when s-GD used Vel-E in adjunction to COE. Instead, for u-GD, the sensitivity value was not changed, while the specificity value was increased by 5%. Thus, the improvement of s-GD’s efficacy for detecting OPMDs was more significant with the aid of fluorescence compared to u-GD’s. In our study, the use of autofluorescence increases sensitivity, specificity, PPV, and NPV for the s-GD in detecting suspected oral lesions compared to COE. These results are consistent with a similar project by Simonato et al., in which it proved a similar increase in the efficacy of Vel-E to detect early oral suspected lesions, but in this case, the main actors are represented by dental students versus expert professional in oral medicine [23].One of the major problems using Vel-E is that it does not provide an objective result in evaluating color change of oral epithelium. Therefore, the threshold between loss of auto-fluorescence and mucosa with normal auto-fluorescence can be arbitrary, and it can often depend on the ability and experience of the operator. This issue limits the effectiveness of this device as cancer screening adjunct if used by inexperienced operators and reinforces the need for training courses [24]. Another critical aspect of the auto-fluorescence observed in literature is the presence of false positives. The percentage of false positives is much higher in groups of patients with reactive or not dysplastic alterations because many inflammatory lesions may show LAF due to the presence of vascular inflammatory modifications (hyperemia and vasodilatation) [25]. In our study, we observed only two cases of false positives (5.71%) in s-GD and 6 (17.14%) in u-GD. On the contrary, false negative are rarely reported in the literature; in our study, we had two false-negative cases with minimal and poorly detectable LAF in s-GD, and 7 in u-GD. The false negatives could be due to an unclear autofluorescence pattern considered negative by the examiners.Therefore, the VelScope could be an adjunctive device for GDs in the screening of OSCC only when a focused training program is performed. From these data, we confirm the pivotal role of continuing oral medicine education to improve diagnostic skills of GDs for detecting oral “high-risk” lesions, and we highlight the need of free mandatory yearly post-graduate training courses in order to give more and more expertise and clinical skills to use screening system devices [26].Indeed, the targeted-education is the reason for improving oral diagnosis by GDs and increases their awareness in referring patients to the specialists for early diagnosis, which remains the best tool to reduce OSCC mortality.The study has, however, different limitations. First, the relatively small sample size both of the examiners and patients; second, the absence of objective autofluorescence visualization including an image analysis software to obtain semi-quantitative or quantitative values.Data from this pilot study indicate that VelScope alone is unable to improve GDs’ ability to detect OPMDs, but it could be an aid for skilled oral practitioners so as to refer patients to oral medicine centers at an early stage of the disease. It would be necessary to educate non-experts on the correct use of adjunct devices whose interpretation of the results requires appropriate and specific training in oral medicine. Further studies with extended samples are necessary to assess the efficacy of VelScope screening on the general population and to determine if it might be useful to improve early diagnosis of OPMDs and OSCC by general dentists after adequate training.Conceptualization, S.L. and M.D.M.; methodology, S.L.; software, A.T.; validation, M.D.M., G.S. and N.C.; formal analysis, A.T. and M.E.B.; investigation, G.S; resources, G.F.; data curation, G.F.; writing—original draft preparation, S.L. and N.C.; writing—review and editing, M.D.M. and G.S.; visualization, G.S.; supervision, S.L. and M.D.M. All authors gave final approval and agree to be accountable for all aspects of the work and in ensuring that questions relating to the accuracy or integrity of any part of the work are appropriately investigated and resolved. All authors have read and agreed to the published version of the manuscript.This research received no external funding.We would like to acknowledge with great appreciation the crucial role of the nursing staff, who provided the support needed to handle the cases.The authors declare no conflicts of interest.Flow chart of the study. 1 u-GD: unskilled general dentist; 2 s-GD: skilled general dentist.The image shows cases of loss of autofluorescence (LAF) on the VelScope examination on suspicious lesions that have been confirmed malignant on histological examination. (a,b) Clinical appearance and VelScope examination of nodular lesion of the left ventral tongue in a 58-year-old male smoker patient; (c,d) Clinical appearance and VelScope examination of single keratotic lesion of the left ventral tongue in a 47-year-old male patient; (e,f) clinical appearance and VelScope examination of keratotic and erythematous lesion of the left ventral tongue in a 52-year-old male patient; (g,h) clinical appearance and VelScope examination of erythematous speckled lesion of the buccal mucosa in a 65-year-old female patient.Clinical appearance of the painful suspected lesion on the right buccal mucosa in a 63-year-old female smoker patient and appearance of the VelScope exam of the same lesion characterized by LAF and associated red/orange coloring typical of secondary bacterial contamination.Baseline demographic and clinical characteristics of participants.1 COE: conventional oral examination; 2 u-GD: unskilled general dentist; 3 s-GD: skilled general dentist; 4 VEL-E: VelScope examination; 5 OLP: oral lichen planus; 6 OLL: oral lichenoid lesion; 7 OL: oral leukoplakia; 8 CIS: carcinoma in situ; 9 OSCC: oral squamous cell carcinoma.Sensibility, specificity, positive predictive value, negative predictive value for the conventional oral examination, and autofluorescence visualization for unskilled-general dentist and skilled-general dentist.1 u-GD: unskilled general dentist; 2 COE: conventional oral examination; 3 VEL-E: VelScope examination; 4 s-GD: skilled general dentist; 5 PPV: positive predictive value; 6 NPV: negative predictive value.

Med-MDPI/jcm_2/jcm-09-06-01755.txt

@@ -0,0 +1 @@
+Cardiovascular disease (CVD) is the principal cause of death in women. Walking speed (WS) is strongly related with mortality and CVD. The rate of all-cause hospitalization or death was assessed in 290 female outpatients with CVD after participation in a cardiac rehabilitation/secondary prevention program (CR/SP) and associated with the WS maintained during a moderate 1 km treadmill-walk. Three-year mortality rates were 57%, 44%, and 29% for the slow (2.1 ± 0.4 km/h), moderate (3.1 ± 0.3 km/h), and fast (4.3 ± 0.6 km/h) walkers, respectively, with adjusted hazard ratios (HRs) of 0.78 (p = 0.24) and 0.55 (p = 0.03) for moderate and fast walkers compared to the slow walkers. In addition, hospitalization or death was examined four to six years after enrollment as a function of the change in the WS of 176 patients re-assessed during the third year after baseline. The rates of hospitalization or death were higher across tertiles of reduced WS, with 35%, 50%, and 53% for the high (1.5 ± 0.3 km/h), intermediate (0.7 ± 0.2 km/h), and low tertiles (0.2 ± 0.2 km/h). Adjusted HRs were 0.79 (p = 0.38) for the intermediate and 0.47 (p = 0.02) for the high tertile compared to the low improvement tertile. Improved walking speed was associated with a graded decrease in hospitalization or death from any cause in women undergoing CR/SP.Despite the current decrease in cardiovascular mortality in women, cardiovascular disease (CVD) continues to be the main cause of death in women [1,2]. Over the last two decades, the American Heart Association, the American College of Cardiology, and the National Heart, Lung and Blood Institute have invested significantly in actions to increase mindfulness of the importance of CVD among women. CVD as the major cause of death among women has been reported to increase from 30% in 1997 to 54% in 2009 [3,4]. Approximately 500,000 women are hospitalized annually for an acute coronary event in the US [3]. Moreover, women have twice the probability of having an adverse outcome after coronary artery bypass graft surgery and are more likely to die within a year after an acute myocardial infarction compared to men [3,4].The absolute numbers of women living with, and dying of, CVD exceed those of men, as does the number of hospital discharges for heart failure and stroke [5]. In the US, the yearly medical costs of CVD are estimated to increase from USD 396 billion in 2012 to USD 918 billion in 2030. Of these, 60.5% is due to the cost of hospitalization [6]. The prioritization of healthy behaviors (including physical activity), in addition to the optimization of medical treatment of established CVD, is a leading objective of several health organizations, in order to ameliorate cardiovascular health and lessen healthcare costs related to CVD [6].Referral to cardiac rehabilitation/secondary prevention (CR/SP) is a class I recommendation for women with CVD [3]. Nevertheless, only 15% to 20% of eligible women participate in these programs [3]. A cornerstone of CR/SP is physical activity, with the goal to increase cardiorespiratory fitness and functional ability [3,7,8,9]. Cardiorespiratory fitness (CRF) is a powerful predictor of cardiovascular and all-cause morbidity in subjects with and without CVD [10,11]. CRF has been demonstrated to be strictly related to walking capacity [12,13]. The capability to walk reproduces the combined performance of many organ systems. Walking speed (WS) has been inversely associated with hospitalization, morbidity, and mortality in men and women with CVD [14,15,16]. Slow WS has also been associated with various health-related issues including disability, morbidity, and mortality [17]. Thus, the evaluation of WS is a marker of health and function in aging and disease and has been advocated as a potential additional “vital sign” [18].WS is a suitable measure of functional capacity, consistent with the daily activities of most adults, and the preferred physical activity by insufficiently active individuals [19]. The WS kept during a moderate and perceptually regulated treadmill-walking test has been shown to be a valid and simple tool for the estimation of CRF [20,21,22,23] and is inversely related to all-cause mortality [24,25,26]. Nevertheless, little is known about the relationship between WS and fatal and non-fatal events in female patients with CVD. In addition, the usefulness and clinical significance of changes in walking capacity over time as predictors of health outcomes is uncertain since the predictive value of WS is mostly founded on a single measure at baseline. Although physical activity behaviors can be modified during a certain follow-up period, interpretations founded on a single measure at baseline can potentially be misleading [25,26].This study was conducted among women with CVD to examine whether the WS during a moderate 1 km treadmill-walking test is related to hospitalization or death from any cause, and whether variations in WS are accompanied by variations in hospitalization or death.A total of 290 consecutive patients, aged 25–88 years, were referred by their GP to an exercise-based CR/SP program at the Public Health Department and Department of Medicine, Division of Cardiology, AUSL Ferrara, Italy. The main objective of the program was the promotion of a long-term physically active lifestyle in order to improve cardiorespiratory fitness and functional capacity. Patients firstly underwent an inclusive clinical evaluation. Hypertension was defined as blood pressure ≥140/90 mm Hg, or current treatment with antihypertensive drugs. Recent blood chemistry analyses—along with left ventricular ejection fraction derived from an echocardiographic evaluation—were registered. The study protocol has been approved by the Ethics Committee of the University of Ferrara, no. 22–13. An approved written informed consent was obtained from all patients at the time of enrollment.Functional assessment was performed at baseline and habitually during follow-up by administering the 1 km treadmill-walking test (1 km TWT). Details of the protocol have been previously described [20]. Based on the results of the 1 km TWT, patients were educated on how to replicate similar walking sessions at home, progressively increasing the duration, from 20 min up to 60 min per session, at least 2–3 times/week but preferably daily. Participants were identified by the regional Health Service Registry, providing data on date of death and hospitalization. The primary outcome was the composite occurrence of all-cause hospitalization and death ascertained throughout the first three years after baseline. The second outcome was the composite occurrence of hospitalization and death during the subsequent fourth, fifth, and sixth year. Any hospitalization was considered an event. For patients undergoing >1 admission, only the first event was considered in the analysis.Medical history was determined from the hospital discharge letter. If >1 cardiovascular diagnosis was recorded during a given re-hospitalization, we defined the diagnosis as follows: A coronary artery bypass graft (CABG) replaced other reasons such as myocardial infarction (MI) or valvular repair/replacement. If the diagnosis was an MI followed by a percutaneous transluminal coronary angioplasty (PTCA), it was classified as an MI. If PTCA was completed without an MI, it was classified as a PTCA. If valvular repair/replacement was performed without an MI, it was classified as a valvular repair/replacement. Other cardiovascular diagnoses such as coronary artery anomalies, heart transplantation, or cardiac tumors were classified as others. Then, patients were grouped into tertiles based on 1) the WS maintained during the 1 km TWT at baseline and 2) the change in WS maintained during the 1 km TWT performed three years later. Differences in continuous variables (age, BMI, LVEF, total and HDL cholesterol, triglycerides, fasting blood glucose) and categorical variables across tertiles were assessed using a one-way analysis of variance and a χ2 test for trend, respectively. The covariates considered as potential confounders were age, BMI, family history, fasting glucose, LVEF, hypertension, medical history, serum creatinine, serum triglycerides, smoking status, total and HDL cholesterol, and use of ACE inhibitors, angiotensin receptor blockers, aspirin, β-blockers, calcium antagonists, diuretics, and statins.To evaluate the association among WS and the occurrence of hospitalization and death over time, we constructed Kaplan–Meier curves. Significantly correlated variables were entered for the fully adjusted multivariable regression model. The composite risk of hospitalization and death was considered independently for each variable, including WS (using increases of 1 km/h); adjustments were made for age. In addition, formal tests of interaction were performed between WS and WS change and all the covariates included in the multivariable models. Patients in the lowest WS tertile at baseline and those in the lowest improvement in the WS tertile three years later were considered the reference groups. The assumption of proportionality for the variables considered in the models was evaluated by the analysis of Schoenfeld residuals. The proportional hazard assumption was met for all models. The level of statistical significance was set at p < 0.05. Statistical analyses were performed using MedCalc 17.6 (Ostende, Belgium).A total of 290 subjects completed the 1 km TWT without complications. The average WS resulted in 3.1 ± 1.0 km/h. Table 1 shows the baseline demographics and the clinical characteristics of the study population grouped into tertiles of WS. Comparison between categories shows significant differences for age, triglycerides, history of CABG, MI, use of antihypertensive agents, and diuretics.During the 3 years following baseline, 124 subjects (42.7% of the sample) were hospitalized for all causes, and nine died. Among the subjects who died, one, three, and five subjects were in the high, moderate, and low tertiles at baseline, respectively. Because of the limited number of hospitalizations and deaths, the analysis was conducted using a unique all-cause composite outcome [27]. Figure 1 shows the cumulative risk of the composite outcome by tertiles of WS (log rank, p < 0.001). Hospitalization or death decreased across increasing tertiles of WS (slowest, n = 55, 57%; moderate, n = 42, 44%; fastest, n = 28, 29%), without interactions between WS and the covariates. The fully adjusted risk for hospitalization or death was lower in the moderate group (hazard ratio (HR) 0.78, 95% CI 0.51 to 1.18, p = 0.24) and in the fastest group (HR 0.55, 95% CI 0.32 to 0.93, p = 0.03) compared with the slowest. Every 1 km/h increase in walking speed was associated with a fully adjusted 26% reduction in the risk of hospitalization or death (HR 0.74, 95% CI 0.59 to 0.92, p < 0.01).During the third-year follow-up, 176 patients were re-evaluated. Among the 105 patients who missed the second test, 36, 33, and 36 subjects were in tertiles one, two, and three at baseline. Compared with the re-evaluated patients, those who missed the second 1 km TWT were less likely to have a history of CABG (35% vs. 45%), were more likely to have a history of PTCA (15% vs. 5%), and had more common use of diuretics (36% vs. 24%). The clinical characteristics of the patients re-evaluated 3 years after baseline are presented in Table 2. WS improved from 3.2 (0.9) to 3.9 (1.0) km/h for the total population. The improvements across tertiles resulted from 3.4 (1.0) to 3.5 (1.0) km/h (p = 0.15), from 3.1 (0.9) to 3.8 (0.9) km/h (p < 0.0001), and from 3.0 (0.8) to 4.5 (0.9) km/h (p < 0.0001) for the low, moderate, and high tertiles, respectively. During 4 and 6 years after baseline, 76 (43% of the sample) subjects were hospitalized and nine died from all causes. The survival curves for hospitalization and death stratified according to tertiles of WS improvement are presented in Figure 2. Hospitalization or death risk was increasingly lower across tertiles of WS improvement and resulted in 53%, 50%, and 35% in the low, intermediate, and high improvement groups (p for trend 0.06). There were no significant interactions between WS change and the covariates examined.Compared with patients in the low improvement tertile, the HRs for those in the moderate improvement and high improvement tertiles were 0.79 (95% CI 0.46 to 1.34, p = 0.38) and 0.47 (95% CI 0.25 to 0.88, p = 0.02), respectively. Every 1 km/h increase in walking speed was associated with a fully adjusted 42% reduction in the risk of hospitalization or death (HR 0.58, 95% CI 0.39 to 0.87, p < 0.01).In 290 female outpatients with CVD, WS at baseline was inversely associated with the risk of all-cause hospitalization or death, independently from age, clinical history, and established risk factors. During the three years follow-up, patients in the fastest group showed a full adjusted 45% reduction in the risk of hospitalization or death compared to the patients in the slowest group. Correction for baseline WS resulted in a minimal change in these risk estimates. A 26% lower rate of all-cause hospitalization or death was associated with every 1 km/h increase in WS. Similar results have been observed in men and women post-myocardial infarction [28], among men and women with stable ischemic heart disease followed for 8 years [29], among men and women with chronic heart failure followed for one year [30], and in men and women undergoing cardiac surgery [15]. Overall, these studies included 1333 men and 266 women.The next relevant finding of this study was the inverse association between the improvement in WS and the risk of hospitalization or death documented in 176 women with CVD re-evaluated three years after baseline. Compared to the 59 subjects with a slight change in WS, among the 60 and 61 patients who increased WS by 23% and 50%, reductions of 31% and 53% in hospitalization or death resulted during the fourth to sixth year of follow-up. This strong association persisted after adjustment for confounders. These findings further emphasize the importance of walking capability assessment and counseling, recently recommended as a potential ”vital sign” [18].Why might being able to walk faster influence hospitalization or death in patients with CVD? Walking is a composite activity, with several factors affecting pace or speed (i.e., cardiorespiratory fitness, neuro-muscular and skeletal condition, habitual physical activity, cognition, sensory and perceptual function, motivation, and mental health) [31,32,33,34]. Therefore, walking speed is considered an indicator of physical fitness and overall health [35].WS is strongly related to cardiorespiratory fitness [12,13] as well as mortality from various causes [36]. Thus, it is not surprising that the greatest benefits were found in the fastest walkers, perhaps reflecting overall higher levels of physical activity. In fact, regular physical activity favorably influences numerous established CVD risk factors, including fibrinolysis and coagulability, inflammation, and autonomic function, as well as an age-related decline in myocardial blood flow and endothelium-dependent vasodilatation [37,38]. All these factors may contribute to WS and help explain the association between change in WS and all-cause hospitalization or death.The current study included a group of women with CVD over a wide range of ages and functional capacities, over a long follow-up period. Secondly, the ease of WS makes it simple to be used in clinical practice. Thirdly, in comparison to other common walking tests (often performed at a near maximum exercise intensity), the 1 km TWT is carried out at a moderate effort and is therefore more agreeable to patients and is likely to be safer. Finally, the large inclusion criteria used in this study is likely to reflect real-world clinical practice.This study was conducted in women with CVD who willingly participated in an exercise-based secondary prevention program; thus, they may not represent the general population. Since the adherence rate was not determined, a causal relationship between the change in physically active behavior and the change in WS cannot be established. Cognitive decline or social, behavioral or psychological factors that could independently affect WS [39,40,41] were not considered. In addition, environmental and cultural factors that could influence WS [31] were not examined. Finally, considering the observational nature of the study, the grade to which walking faster may causally influence hospitalization or death cannot be addressed.The present results support the concept that in women with CVD, those being able to walk faster have a lower rate of hospitalization and death, which represent relevant endpoints in clinical and research settings. The change in survival and hospitalization rates is related to the extent of the change in WS. These results propose that the walking speed maintained during a moderate endurance test can be used by health professionals seeking simple tools for the promotion and maintenance of physically active lifestyles in women with cardiovascular disease.Conceptualization, G.G. and G.C.; methodology, G.G., G.C., and J.M.; validation, G.G., G.C. and G.M.; formal analysis, G.C., G.G., A.R., and L.C.; investigation, G.G.; resources, N.N., F.G., M.F., M.P., and G.C.; data curation, G.G., G.C., and L.C.; writing—original draft preparation, G.G.; writing—review and editing, G.M., J.M., B.S., G.P., V.Z., S.M. (Sabrina Masotti), S.M. (Simona Mandini) and G.C. All authors have read and agreed to the published version of the manuscript.This research received no external funding.The authors thank Cinzia Nanini and Stefano Palazzi for their support in exercise testing sessions.The authors declare no conflict of interest.Kaplan–Meier curve showing the rate of hospitalization or death during 36 months after enrolment as a function of walking speed at baseline.Kaplan–Meier curve showing the rate of hospitalization or death 36 to 72 months after enrolment as a function of walking speed improvement.Demographic and clinical characteristics of the 290 subjects by baseline tertiles of walking speed.Data are presented as mean (standard deviation) or percentage. ACE, angiotensin-converting enzyme; ARB, angiotensin receptor blocker; BMI, Body Mass Index; CABG, Coronary Artery Bypass Graft; LV, Left Ventricular; PTCA, Percutaneous Transluminal Coronary Angioplasty, stenting or both.Clinical characteristics of the subjects 3 years after baseline subdivided by tertiles of improvement of walking speed.Values are presented as mean (standard deviation, SD) or %. Abbreviations: ACE, angiotensin-converting enzyme; ARB, angiotensin receptor blocker; BMI, Body Mass Index; CABG, Coronary Artery Bypass Graft; HDL, high-density lipoproteins; LV, Left Ventricular; PTCA, Percutaneous Transluminal Coronary Angioplasty, stenting or both. The values of the variables considered (except walking speed) are baseline values.

Med-MDPI/jcm_2/jcm-09-06-01756.txt

@@ -0,0 +1 @@
+In patients with knee osteoarthritis, when only medial or lateral compartment of the knee is involved, unicompartimental knee arthroplasty (UKA) is a reliable option for addressing the symptoms and restore function. The main aim of the present review is to systematically collect the available evidence concerning the return to sport activity in the elderly patients after UKA. An electronic search was carried out on the following databases; Pubmed-Medline, Cochrane central, and Scopus, searching for randomized controlled trials, prospective cohort studies, retrospective case-control studies, and case series. Data concerning the evaluation of the return to sport (RTS) and of functional outcomes in the elderly patients after UKA surgery. MINORS score was used to assess the risk of methodological biases. Odds ratios and raw proportions were used to report the pooled effect of UKA on the return to sport in comparative and non-comparative studies, respectively. Same level RTS in elderly patients was of 86% (pooled return proportion 0.86, 95%CI 0.78, 0.94), showing also better relative RTS and time to RTS of patients undergoing UKA, in comparison to those undergoing TKA. Sport-specific RTS showed that higher return rates were observed for low-impact sports, whereas high-impact sports prevented a full return to activities. UKA is a valid and reliable option for elderly patients to satisfactorily resume their sport practice, especially for low impact activities. The rate of return to sports following UKA is higher than TKA.In the present social scenario, the needs of elderly people are changing. It is not infrequent that patients want to stay active and be able to perform physical exercises and sport activities even in an advanced age [1,2]. However, these requests are often undermined by chronic painful conditions, such as osteoarthritis (OA), that do not allow the patient perform all desired activities [3]. In particular, knee OA is a common, debilitating condition that is increasingly widespread accordingly with the aging of the general population [1,2,4,5,6]. It is widely accepted that the definitive treatment for the end stage knee OA is the joint arthroplasty [7]. When only medial or lateral compartment of the knee is involved, unicompartimental knee arthroplasty (UKA) is a reliable option that is raising in popularity [8,9]. Indications to UKA have been widely discussed, but it is well known that this implant provides some advantages: lower invasiveness, shorter rehabilitation time, restoration of a wider range of motion, and physiological proprioception of the knee due to cruciate ligaments retention [10]. Several studies in literature reported the benefits of patients who underwent UKA in terms of pain relief and quality of life, with a good to excellent return to activities [11]. The opportunity to move and walk without pain, with also a good recovery of the motion, allows individuals to perform physical activity and sport, which is particularly important to prevent systemic diseases associated to sedentary life such as obesity, diabetes, cardiovascular accidents, and cancer [12,13,14]. One of the principal expectation for active patients before undergoing UKA surgery, is about their chances to perform physical activity and sport after surgery. Moreover, active patient is mostly interested in type and level of sport activity [15,16]. The current scientific literature answers those questions mainly with recommendations based on expert opinions and surgical society guidelines [17,18], but still lacks high level evidence-based guidelines, especially regarding the elderly population. The main aim of the present manuscript is therefore to systematically collect the available evidence concerning the return to sport activity in the elderly patients after UKA, with a special concern to the type of activity. A secondary endpoint of this investigation is to assess the functional outcomes in the same population.A systematic review and meta-analysis was carried out using the Preferred Reporting Items for Systematic Reviews and Meta-analysis (PRISMA) guidelines [19]. The review was planned and conducted following the PRISMA checklist. According to PICO, the following elements have been used to frame the study question.Population: Elderly patients.Intervention: Unicompartmental Knee Arthroplasty (UKA).Comparison: Total Knee Arthroplasty (TKA) or no comparison.Outcomes: Return to sport activity and functional outcomes.The studies considered for inclusion were randomized controlled trials (CRT), prospective cohort studies (PCS), retrospective case–control studies (RCS), and case series (CS). The main topic of the papers had to be the evaluation of the return to sport activity and of functional outcomes in the elderly patients after UKA surgery. Case Reports, Reviews, and Meta-analyses were not eligible for inclusion. Moreover, in vitro studies and cadaver studies were excluded from the review analysis. Given the specific focus on a selected population, only studies reporting outcomes of patients aged 65 or older were considered for inclusion (average population age > 65). The absolute numbers and proportions of patients returned to same level sport activities (RTS) was considered as the primary outcome measure and assessed throughout the included studies. Sport-specific return was extracted from the studies, to stratify results according to the type of activities.The secondary endpoint was achieved by evaluating the following measures; Oxford Knee Score (OKS), Knee injury and Osteoarthritis Outcome Score (KOOS), Knee Society Score (KSS) and American Knee Society Score (AKSS), University of California Los Angeles (UCLA), Tegner and Lysholm scales, and the Western Ontario McMaster universities osteoarthritis index (WOMAC). These measures were evaluated across the included studies. The online search was carried out on the following databases; Pubmed-Medline, Cochrane central, and Scopus. The following search string was used; (Arthroplasty, Replacement [MeSH Terms]) AND joint, knee [MeSH Terms]) AND sports [MeSH Terms] and (arthroplasties, knee replacement [MeSH Terms]) AND sports [MeSH Terms]. The bibliography of the included studies and of recent review articles was screened for further relevant articles, potentially missed at the electronic search. After duplicates removal, all the retrieved studies were firstly screened by title to find studies dealing with UKA. Two independent reviewers (B.Z. and G.T.) evaluated the abstract of each of the papers considered for inclusion. Discordant opinions concerning study inclusion were discussed with a third experienced reviewer (R.P.). If abstract was not sufficient to define inclusion of a paper, the full-text was retrieved and evaluated. Articles included for review process were retrieved in full-text and read. The search process was summarized in Figure 1.Data were extracted independently by two reviewers (B.Z. and L.A.D.B.) and tabulated according to primary and secondary outcomes of this review. Discordant opinions in data extraction were solved by discussion with a third reviewer (R.P.).The quality of the included non-randomized studies was independently evaluated by two reviewers (L.D.B. and B.Z.) using the Methodological Index for Non-randomized Studies (MINORS) score. [20] The following domains were assessed; a clearly stated purpose, inclusion of consecutive subjects, prospective data collection, endpoints appropriate to the purpose of the study, unbiased assessment of the study endpoints, follow-up period appropriate for the study, loss to follow-up of less than 5%, prospective calculation of the study size, adequate control group, contemporary group, baseline group equivalence, and adequate statistical analysis. The last four items are specific to comparative studies. Each item was scored from 0 to 2 points, with a global ideal score of 16 points for non-comparative studies and 24 points for comparative studies. Meta-analysis was carried out to investigate the effect of UKA on the return to sport activity either in comparison with TKA or in non-comparative studies. Furthermore, return to specific sport activities was pooled if at least three studies reported the same sport. Odds ratios (ORs) and raw, i.e. untransformed, proportions were used to report the pooled effect of UKA on the return to sport probabilities in comparative (vs. TKA) and non-comparative studies, respectively. Heterogeneity was evaluated using Q statistic, expressed as the p value for the χ test under the null hypothesis that the between-study variance (τ) equals 0, and I2 test. All the conducted meta-analyses evidenced the presence of significant heterogeneity, defined as a I2 > 55% and/or a Q statistic p value below 0.05. Accordingly, random effect models were applied. Finally, the likelihood of methodological bias among included studies was estimated with the visual inspection of the funnel plot. Analyses were conducted using metafor and meta packages in R 3.6.1 (R Foundation for Statistical Computing, Vienna, Austria). Electronic search identified 447 papers, and of these 287 scientific products were screened for analysis. Full text of 49 papers was accurately analyzed and 28 were excluded for following reasons; absence of postoperative sport-related outcomes, cohort mean age lower than 65 years old, duplicated papers and no UKA patients. Finally, 10 [10,21,22,23,24,25,26,27,28,29] articles were included according PRISMA selection process (Figure 1).Among 10 studies, only 2 study were prospective [24,26], 6 evaluated the cohort retrospectively (LOE III) [21,22,23,27,28,29], and 2 were case series (LOE IV) [10,25]. Five studies out of 10 compared clinical outcome of UKA and TKA cohorts [22,23,24,28,29]. Six studies reported specific RTS outcome like preoperative and postoperative sport participation, RTS rate, time to RTS and pre and postoperative sport-specific participation [10,21,22,27,28,29]. Canetti et al. compared two different cohort of lateral UKA performed with and without robotic assistance [21].Overall number of patients analyzed in the present review was 5220 with 2930 UKA and 2447 TKA implanted. Mean age of UKA’s was 66.3, whereas mean age of TKA cohort was 74 years old. Mean follow-up was 2.1 years. In three studies type of prosthesis was not specified [22,24,29] while 4 papers reported outcome of mobile bearing UKA [25,26,27,28]. Demographic parameters of included study are summarized in Table 1.Sport-specific return rates were analyzed in 50% of the studies included [10,21,22,27,28,29]. Mean RTS rate for UKA was 89.5%, ranging from 75% [22,29] to 100% [28]. Mean preoperative sport participation rate was 71.8% of the patients, ranging from 36% [29] to 100% [21], and mean postoperative sport participation rate was 70.2% of the patients, ranging from 27% [29] to 100% [21]. Mean time to RTS was 6.2 months. Results of the study published by Canetti et al. [21], showed for UKA robotic assisted group a statistically significant difference in terms of time to return to sport compared to conventional UKA with a similar RTS rate (100% vs. 94%). The cohort of medial UKA of Pietschmann et al. [27] had an 88% of RTS rate with 80.1% of patients that returned to preoperative activity level. Naal et al. [10] reported RTS rate of 95%; moreover, the activity frequency (session per week) was maintained in postoperative assessment (2.9 vs. 2.8) with a slight decrease in terms of session length (66 vs. 55 minutes). Overall, by meta-analyzing available studies, we evidenced a good proportional RTS (0.86 95% CI 0.78, 0.94), with sport-specific RTS favoring those sports with low-impact. Meta-analysis results were showed in Figure 2, Figure 3, Figure 4, Figure 5, Figure 6, Figure 7 and Figure 8.Three papers compared RTS outcome of UKA and TKA patients’ cohort [22,28,29]. Harbourne et al. [22] at 12 months of follow-up recorded a higher rate of return to activity in patients with UKA than TKA (75% vs. 59% p < 0.001). According to results of Walton et al. [28], the UKA group had a better percentage of patients that increased or maintained sport activity compared to the TKA group (P_.0003); moreover, TKA patients significantly reduced postoperative sport activity compared to UKA’s (P_.0001). Wylde et al. [29] investigated return to sport after different type of implant (THA, Hip Resurfacing, TKA, UKA) and no significant difference differences were detected in postoperative sport participation between UKA and TKA (75% vs. 73.1%). Meta-analysis study pooling showed a better RTS in patients undergoing UKA (Odds Ratio 2.14 95% CI 1.29, 3.55). Results are shown in Figure 9.Several clinical outcome assessment measure and sport specific questionnaires were utilized. Oxford Knee score were used in six studies [22,24,25,26,27,28], Knee Society Score (KSS) in five studies [10,21,24,25,26,27], WOMAC in three studies [21,24,27,29], Forgotten Joint Score (FJS) and UCLA in two studies [21,24], Tegner Activity Score in two studies [25,26], KOOS [23] and Lysholm Knee Scale [21] in one study. Clinical results of Canetti et al. [21] showed a higher IKSS-Objective (97.2 ± 5.9 vs. 91.2 ± 6.5; p < 0.05) and a higher IKSS-Objective improvement (+ 30.9 ± 7.7 vs. + 22.8 ± 12.2; p < 0.05) compared to conventional group of lateral UKA. Naal et al. [10] obtained an improvement on KSS score in their postoperative assessment (from 129.9 ± 24.8, vs. to 186.9 ± 18.3) with a good result in terms of quality of life stated with SF-36. Pandit et al. [26] obtained a good postoperative results with the first 1000 cases of Oxford phase 3 medial UKA at 5 years: mean OKS was 41.3 (SD 7.2), mean AKS Objective Score 86.4 (SD 13.4), mean AKS Functional Score 86.1 (SD 16.6), and mean Tegner activity score 2.8 (SD 1.1). The same group, in 2015 [25], published results of cementless fixation for the same implant with similar postoperative clinical and functional results: OKS 43 (SD 7), AKSS (objective) 81 (SD 13), AKSS (functional) 86 (SD 17), and Tegner activity score of 3 (1–8). Pietschmann et al. [27] had a higher postoperative OKS, KSS, WOMAC and UCLA score. Active patients in sport preoperatively, except for KSS knee objective score, obtained statistically significant higher score than inactive patients group (OKS < 0.01, UCLA < 0.0001, KSS function < 0.01, KSS knee subjective < 0.01, KSS overall < 0.01, WOMAC < 0.05, WOMAC stiffness < 0.05, WOMAC ADL < 0.01, WOMAC overall < 0.01). Results are summarized in Table 2. Four studies compared clinical outcome in patients aged more than 65 years old and underwent to UKA and TKA [23,24,28,29]. In the patients’ cohort of Lygre et al. [23], UKA had a statistically significant superiority over TKA in terms of KOOS “Symptoms” (adjust mean diff 2.7 p = 0.04), KOOS “Function in Daily Living” (ADL) (adjust mean diff 4.1 p = 0.01) and KOOS "Function in Sport and Recreation" (adjust mean diff 5.4 p = 0.006). A prospective study designed by Matthews et al. [24] showed no statistical difference according to satisfaction (89 vs. 87 p = 0.41) and perception of knee normality (69 vs. 68 p = 0.99) scores between UKA and TKA; nevertheless, UKA reached a statistically significant higher WOMAC (p = 0.003), SF-36 (physical p < 0.001; mental p = 0.25), Oxford knee (p < 0.001), American Knee Society (clinical p = 0.002; function p < 0.001) and Total Knee Function Questionnaire scores (ADL p = 0.002; sport and exercise p = 0.02; movement and lifestyle p = 0.02). Walton et al. [28] compared in their study Mini-Incision Unicompartmental Knee Arthroplasty versus TKA and reported better results in terms of OKS (p = 0.0426) and mean modified Grimby score (3.89 SD:1.27 vs. 2.76 (SD:1.12). In the last comparative study, performed by Wylde et al. [29], no clinical difference between UKA and TKA in terms of WOMAC pain (81.5 SD = 20.8 vs. 81.6 SD = 19.3) and function (76.3 SD = 21.4 vs. 79.1 SD = 20.5).The MINORS score ranged from 7 [27] to 12 [25] for non-comparative studies and from 12 [28] to 17 [21] for the comparative ones (Table 1). The mean value was 10 for non-comparative studies and 15 for comparative studies. The funnel plot of studies evaluating RTS after UKA showed a symmetrical distribution, while a rather poor precision of observations, suggesting an overall low-moderate risk of methodological bias (Figure 10).Elderly patients represent a selected population, which is changing in activity needs in recent years, according to general lifestyle modifications of the society. The main findings of the present investigation suggest a good proportional same level RTS in elderly patients after UKA (return proportion 0.86, 95%CI 0.78, 0.94), showing also better relative RTS and time to RTS of patients undergoing UKA, in comparison to those undergoing TKA. Moreover, patients undergoing TKA were more likely to reduce their activity level after the surgery [28]. Meta-analysis of the sport-specific RTS showed that higher return rates were observed for low-impact sports (e.g., swimming, fitness, hiking), whereas high-impact sports (e.g., tennis and alpine ski) prevented a full return to activities. The proportion of RTS for cohorts of patients undergoing UKA is in line with authors’ experience and with literature-reported rates [10,21,30]. However, given the high heterogeneity of the studies concerning this outcome (90%), the result should be carefully interpreted. A first consideration concerns the average age of the cohorts, which was higher in those study reporting a lower RTS proportion. Furthermore, the differences in type of UKA implants could determine the activity level, with possible implications in polyethylene wearing [31]. Another major concern in general study heterogeneity is the absence, in almost all the studies, of a description of rehabilitation protocols and of surgical incision. In an era of wide differences in rehabilitation (i.e., fast-track, aquatic rehabilitation), understanding the post-operative protocols may be the key to evaluate the postoperative outcomes [32]. Moreover, the knee extension apparatus, thus the surgical approach plays a major role in return to activity and its timing.Clinical outcomes reported in the included studies were filtered to collect those relative to sport participation and activity level. Although heterogeneous, an overall evaluation of the scores showed that either objective and subjective assessment improved significantly after UKA, suggesting that patient perception of the beneficial effects of the surgery reflects a standardized clinical examination and ROM assessment. Similarly, comparison of clinical outcomes after UKA with those after TKR favored the patients undergoing UKA. However, given the absence of control groups in most of the included studies, the meta-analysis evaluation was not possible, given the impossibility to calculate the standardized mean difference.Concerning comparison of RTS and clinical outcomes between cohorts undergoing UKA and TKA, a potential confounding factor is age, as average age of UKA cohorts was 66.3 years, while mean age of TKA cohorts was 74 years. A 10-year difference is relevant by the observation that younger patients are more likely to continue in sport participation. This is especially true, given the higher and higher percentage of elderly population involved in sport activities in recent years [33]. Focusing on elderly population underlined some differences with available evidence on general population (non-elderly) [34,35]. First of all, as reported in previous literature reviews [11], the return to high-impact sport in patients that underwent either TKA and UKA was high, and not excessively different from those of low-impact sports [36,37]. Common experience leads the orthopedic surgeon to suggest caution in sport participation after joint arthroplasty, for the risk of component migration, loosening, and periprosthetic fractures. However, no specific evidence advices about long term results of sport involvement after UKA [30].The UKA, a joint preserving arthroplasty, has been designed for those patients with localized osteoarthritis and was found especially beneficial in active individuals. However, the follow-up length reported in available literature is too short to assess failure and revision rates. It is opinion of the authors that until clear data will be available, the general attitude for patients and surgeon in regard to RTS will go toward a careful approach.Another concern, in regards of the available literature, is the lack of information about reasons of not to return to sport activities. None of the articles included in this review reported the clinical and functional reasons that prevented the patients to return to activity, except a few reported surgical complications (i.e., common peroneal paralysis [21]). The issue of motivational causes that prevent RTS is relevant and only some studies in the gerontology field address the question [6,14]. An interesting point to improve future research would be to introduce motivational and depression scales next to functional patient reported outcome measures.The overall LOE of included studies was low, as most of the evaluations were retrospective or were case series. This is reflected in the relatively high risk of bias which MINORS score showed. In particular, the worst item was the blinding of participants. However, given the type of therapeutic intervention considered, blinding was impossible.The funnel plot evaluation was limited to those studies assessing the proportional RTS of patients after UKA. A balanced funnel was observed at visual inspection, but a relatively low precision was found for some studies, suggesting an overall low–moderate risk of methodological bias.This is the first literature review and meta-analysis that selected age of the cohorts undergoing UKA, focusing the research questions on the elderly population. Meta-analysis had a two-fold aim: to investigate proportional RTS in UKA-only cohorts and to compare UKA and TKA patients. However, the study was not free from limitations. First of all, the limited number of studies reporting sport-specific outcomes after UKA and the low LOE prevented the authors to gather a sufficient evidence to finally answer to the research questions. Furthermore, the differences in study design, age of the cohorts and effect sizes yielded to high heterogeneity and prevented to draw a robust overall meta-analysis.UKA is a valid and reliable option for elderly patients to satisfactorily resume their sport practice, especially for low impact activities. The rate of return to sports following UKA is higher than TKA. The most practiced sports after surgery are low contact activities such as swimming, fitness, and hiking. Unfortunately, there is a lack of consistent clinical data on the functional improvement before and after surgery in elderly patients, thus a standardized evaluation of patient after surgery is prevented. More, prospective, comparative studies are needed to determine the standardized functional improvement of elderly patients after UKA.Conceptualization, R.P. and B.Z.; methodology, A.D.V. and G.T.; formal analysis, A.D.V.; data curation, L.A.D.B. and B.Z.; writing—original draft preparation, G.T. and B.Z.; writing—review and editing, G.P. and S.V.; supervision, R.P. and V.D. All authors have read and agreed to the published version of the manuscript.This research received no external funding.The authors declare no conflicts of interest.Study selection flowchart (UKA: Unicompartimental Knee Arthroplasty).Overall return to sport after UKA (C.I.: Confidence Intervals).Return to hiking after UKA (C.I.: Confidence Intervals).Return to cycling after UKA (C.I.: Confidence Intervals).Return to swimming after UKA (C.I.: Confidence Intervals).Return to alpine ski after UKA (C.I.: Confidence Intervals).Return to fitness after UKA (C.I.: Confidence Intervals).Return to tennis after UKA (C.I.: Confidence Intervals).Return to sport after UKA vs. TKA (C.I.: Confidence Intervals, O.R.: Odds Ratio).Funnel plot showing studies evaluating RTS after UKA.Demographic studies details.UKA: Unicompartmental Knee Arthroplasty TKA: Total Knee Arthroplasty (TKA), LOE: Level: of Evidence, M: Males, F: Females.Functional outcomes of included studies.S.D.: Standard Deviation, N/A: Not Available, OKS: Oxford Knee Score, KOOS: Knee injury and Osteoarthritis Outcome Score, KSS: Knee Society Score, AKSS: American Knee Society Score, UCLA: University of California Los Angeles, WOMAC: Western Ontario McMaster universities osteoarthritis index.

Med-MDPI/jcm_2/jcm-09-06-01757.txt

@@ -0,0 +1,3 @@
+In the past decade, fecal microbiota transplantation (FMT) has rapidly spread worldwide in clinical practice as a highly effective treatment option against recurrent Clostridioides difficile infection. Moreover, new evidence also supports a role for FMT in other conditions, such as inflammatory bowel disease, functional gastrointestinal disorders, or metabolic disorders. Recently, some studies have identified specific microbial characteristics associated with clinical improvement after FMT, in different disorders, paving the way for a microbiota-based precision medicine approach. Moreover, donor screening has become increasingly more complex over years, along with standardization of FMT and the increasing number of stool banks. In this narrative review, we discuss most recent evidence on the screening and selection of the stool donor, with reference to recent studies that have identified specific microbiological features for clinical conditions such as Clostridioides difficile infection, irritable bowel syndrome, inflammatory bowel disease, and metabolic disorders.In the last decade, multiple studies have expanded knowledge in the field of gut microbiota, including pathogenesis, diagnosis, and therapeutics [1]. To date, the therapeutic modulation of the intestinal microbiota is performed with traditional approaches such as antibiotics and probiotics, or increasingly through fecal microbiota transplantation (FMT), which is defined as the transfer of fecal material from a healthy donor into the gastrointestinal tract of a recipient [2].Fecal material has been used in medicine since almost two thousand years. The first description of the use of fecal material for medical purposes dates back to about 1700 years ago; traditional Chinese medicine in particular had perceived the potential role of this biological material and used it for several clinical indications such as gastrointestinal, nervous system, skin, and gynecological diseases [3]. In Western countries, the first description of ancestral FMT dates to the 17th century, when Fabricius Acquapendente reported the transplantation of feces for the cure of animals unable to ruminate [4]. More recently, anecdotal use has been reported during the Second World War. German soldiers residing in North Africa suffered from recurrent episodes of diarrhea that they treated by eating camel stool, being inspired from the local practice of the Bedouins [5]. Western medicine began to study the potential role of FMT only in the second half of the 20th century. Firstly in 1958, Ben Eiseman reported the successfully treatment of four patients with pseudomembranous colitis using fecal enemas [6], and over 20 year later, Schwan et al. reported new evidence supporting the efficacy of FMT in C. difficile infection (CDI) [7]. In the following years, several other reports came out, and a growing body of evidence showed the efficacy of FMT in the treatment of recurrent CDI, and, furthermore, the feasibility of FMT was gradually suggested for other clinical indications.The first randomized controlled trial that investigated the role of FMT for recurrent CDI was published by van Nood et al. in 2013. They reported that a single infusion of fecal material by nasoduodenal route was superior to standard therapy with vancomycin [8]. In further years, other routes of administration were successfully tested in clinical trials, demonstrating the efficacy of FMT by lower route through colonoscopy [9] or upper administration with capsule [10]. Therefore, the growing interest of the scientific community towards FMT has meant that a large amount of data has been published in the last decade; for this reason, a panel of European experts met in Rome in 2017 to release the first evidence-based consensus report for the use of FMT in clinical practice [11].Over the years, further issues have emerged, which are still not clarified to date. In particular, in view of the growing number of patients who could benefit from FMT, it is necessary to identify innovative ways to storing fecal material to be used if necessary. Indeed, in the early experiences, FMT was performed only with fresh material from occasional healthy donors, but this approach is not feasible for large-scale use of FMT. To solve this problem, the possibility to create structures to bank the feces after manipulation was suggested, and this approach is supported by the evidence of the effectiveness of FMT performed with frozen material [12]. In consideration of the increasing interest of the scientific community on this topic, a panel of international experts met in Rome in 2019 to define the general guidelines for the creation of stool banks [13]. Despite these efforts, many problems remain to be solved. Above all, the identification of the optimal donor is a fundamental clinical issue of rising relevance. Indeed, the increasing number of clinical indications suggests the need to identify the ideal donor for each disease or patient that cannot be treated indiscriminately with the same fecal biomass. The fascinating idea of identifying the “perfect” intestinal microbiota has motivated the scientific community for at least one century—ever since in the early 20 century Elie Metchnikoff suggested the role of intestinal bacteria in the development of many pathological conditions and health in the homeostasis of the microbial species [14], generating the concepts of “eubiosis” and “dysbiosis,” which for years were considered only fascinating hypotheses without strong scientific bases. However, in recent years, the molecular techniques of genomic sequencing have allowed to understand the link between gut microbiota and several diseases [15], giving evidence to this old intuition. In particular, we refer to “eubiosis” as considering a status characterized by a preponderance of potentially beneficial species, while “dysbiosis” is a condition characterized by the loss of homeostasis and by the proliferation of microbial species considered potentially pathogenic and, moreover, favor a “milieu” triggering the hyper-inflammatory state [16]. To date, an increasing number of studies confirm these hypotheses, in particular the reduced diversity of gut microbiota, simply defined as the variety and abundance of species in a defined microbial ecosystem [17,18], which is known to characterize several chronic diseases compared to a control group [19,20]. Therefore, in this narrative review, we report the most recent evidences on the screening and selection of the stool donor, with special efforts to describe findings that may lead to the optimal donor in several disease looking for an “optimal microbiota” to be transplanted (CDI, inflammatory bowel disease (IBD), irritable bowel syndrome (IBS), and other emerging pathological conditions).To date, the only recommendation for FMT in clinical practice is the treatment of recurrent CDI, although a large number of emerging indications are being experienced in several studies [21].CDI is a burdensome clinical issue and represent the most relevant cause of antibiotic-associated diarrhea; its incidence has evolved in recent years and the risk of recurrence after standard antibiotic therapy has widely increased [22,23]. The standard treatment for the first occurrence of CDI is still represented by antibiotic therapy, mainly with metronidazole or vancomycin [24]. However, the clinical success rate of antibiotics in the recurrence of CDI is dramatically decreased, consequently, more effective therapies have been proposed, including FMT [25,26]. The clinical success of FMT, in contrast to the loss of efficacy of standard antibiotic therapy, could be explained by understanding the mechanism of action. In fact, FMT is a restorative treatment of gut microbiota alterations, unlike antibiotics, which is a disruptive treatment; accordingly, the administration of FMT results in a prompt and sustained normalization of microbial community structure and then metabolic activity of gut microbiota [27]. Indeed, CDI develops only in subjects with disruption of gut microbiota [28]; supporting this idea, it was demonstrated that the feces of patients with recurrent CDI have a higher relative abundance of several bacterial family as Enterobacteriaceae, Veillonellaceae, and Lactobacillaceae, and lower relative abundance of Ruminococcaceae, Bacteroidaceae, and Lachnospiraceae [29]. Furthermore, FMT recipients have shown changes in microbial profiles and shifts in the gut microbiota composition towards a profile similar to that of the healthy donor; this finding is obtained in a few days and is observed for at least six months [30].To date, several systematic review and meta-analyses have shown an overall cure rate of FMT of up to 90% in preventing further CDI recurrence [31,32]. Moreover, a recent meta-analysis has shown that both the upper and the lower route are effective, with a slight advantage of colonoscopy over other techniques [32]. Based on these positive evidences, scientific societies have included FMT among the recommended treatment for recurrent CDI. Already in 2014, FMT was strongly recommended in recurrent CDI by the European Society of Clinical Microbiology and Infectious Disease (ESCMID) [33], while the American College of Gastroenterology (ACG) stated that FMT can be considered after the third recurrence [34]; more recently, the Infectious Disease Society of America (IDSA) confirmed the indication in the treatment of recurrent CDI with FMT [24].Alongside well-established indications such as CDI, several studies have found emerging clinical conditions for which FMT may represent a promising alternative to standard therapies. Most evidence comes from inflammatory bowel disease (IBD) studies. Several alterations of gut microbiota has been proposed as factors contributing to the development of the aberrant immunological response in IBD [35], but it is still unclear if the perturbations of microbiota are the cause or consequence of the mucosal inflammation associated to IBD [36]. In particular, ulcerative colitis (UC) is the most suitable IBD model for the study of FMT, considering the characteristics of inflammation of the mucosa and the established role of the microbiota in pathogenesis [37]. To date, a little number of clinical trials have reported promising results, but several concerns suggest to better investigate this potential clinical application [38]. According to a Cochrane systematic review of four clinical trials, the overall remission rate at week 8 was 37% (52/140 UC patients) in patients receiving FMT, compared with 18% (24/137 patients) in those receiving placebo; additionally, clinical response and endoscopic remission improved in patients treated with FMT [39]. However, several factors appear to influence the clinical response in UC patients, as the condition during the manipulation of the feces or the donor selection. For instance, anaerobic conditions during the manipulation of stool were associated with better performance considering clinical remission or steroid free response [40]. Donor selection might be a relevant factor considering that a study reported higher success rates with one particular donor compared with other donors [41]. Furthermore, an emerging relevant indication for FMT was represented by the flare of UC associated with concurrent C. difficile over infection. A recent clinical trial, including patients affected by UC or Crohn disease with recurrent CDI, reported that FMT has a curative effect on the recurrence of CDI, but has no apparent beneficial effect on the IBD course [42].Gut microbiota disturbance was also involved in other gastrointestinal diseases such as irritable bowel syndrome (IBS). A systematic reviews with meta-analysis showed that FMT may be beneficial in IBS [43], but this finding is limited by the small number of patients included and by the relevant differences in the design of the studies. In particular, IBS is triggered by multiple factors, and furthermore, is a heterogeneous condition that may require a selection of the donor in each case. For instance, El-Salhy et al. have recently reported that FMT administered through gastroscope was highly effective in IBS if a well-defined donor was chosen with a normal disbyosis index and favorable specific microbial signature [44].Furthermore, metabolic and hepatic diseases are also considered emerging indications for FMT. There is great interest towards the modulation of the gut microbiota in metabolic syndrome, as two studies reported promising results in improving peripheral insulin sensitivity [45,46]. Unfortunately, the improvement of metabolic profile was not maintained in the long term, and a recent systematic review including three studies reported the absence of significant benefits from FMT in metabolic syndrome [47]. Thus, further studies to clarify the feasibility of this approach in metabolic disorders are needed. Furthermore, FMT was able to reverse encephalopathy derived from disturbed gut-brain axis in patients with liver chronic disease, two clinical studies shown promising results in this field of application [48,49].FMT was also proposed in the treatment of several other clinical conditions, but evidence is limited and results were reported by small studies; thus, the application is limited to clinical studies and selected cases. For instance, FMT was reported as effective in the decolonization of patients carrier of multi-drug resistant organism [50], in reducing symptoms in autism spectrum disorders [51], or in reliving symptoms and increasing progression free survival in graft versus host disease after hematopoietic stem cell transplant [52].Donor selection represents a fundamental challenge in view of the implementation of FMT programs worldwide. To date, there is a broad debate regarding the preference of donor selection, whether the stool donor should be known to the patient or whether it is preferable to use feces from unrelated donor. Moreover, in the case of non-related donor, fecal material could be banked at dedicated structures that provide support to the hospital that will perform FMT [53].In particular, the ideal stool donor should be a healthy volunteer, without risk factors for infectious or other chronic diseases, and who is willing to “donate” frequently if needed. Unfortunately, although the conditions do not seem too selective, it is not always easy to identify an adequate number of donors to meet the needs of the FMT program. Indeed, data from large stool bank suggest high rates of donor drop out due to high commitment required [54]; furthermore, physicians often give up FMT because of the complexity and costs of screening [55]. Consequently, to solve these problems, it would be appropriate to implement the undirected donor selection program. Hence, the related donors should be only limited in cases of patient preference. Indeed, undirected donors reduced the likelihood of confidentiality concerns, and then, they are essential for the implementation of stool banking in consideration of easy availability, traceability, and reduction of screening expenses [56].The screening of potential donors consist in two key landmarks, the preliminary interview and the laboratory testing [13]. A preliminary interview is usually performed by a structured questionnaire that investigated several risk factors to minimize the risk of transferring infections or adverse gut microbiota profile. In particular, the medical interview screen potential donors inquiring about the use of drugs that can alter gut microbiota, known history or risk behaviors for infectious disease, and for disorders potentially associated with the disruption of gut microbiota. The schedule of questions reported in this review (Table 1) includes the most frequently investigated features in leading FMT centers. Obviously, this draft of interview is not mandatory, but can be adapted to the socio-cultural context of potential donors. For example, it would be advisable to carefully investigate the eating habits of potential donors from country where the consumption of raw meat and fish is widespread, thereby increasing the risk of transmission of enteric pathogens, or who eat exotic animals that are potential carriers of unknown pathogens; or seasonal habits that increase the risk to get infected with intestinal pathogens (e.g., summer holidays and risk of sea food of poor quality). These examples allow to understand how the aim of the interview is to early intercept potential risks of pathogen transmission; thus, each center should adapt the medical interview to its socio-cultural context to make it more efficient.The optimal donor correspond at young individual (preferably < 50 years, as suggested by a panel of experts [13] taking into account that increasing age has been associated with altered gut microbiota composition [57]; moreover, aged microbiota could have a negative effect contributing to the inflammatory state of the recipient [58]), although is important to exclude candidates with personal history of malignancies or autoimmune disease [13]. Moreover, there are concerns regarding the exclusion of healthcare workers considering the supposed increased risk of colonization by antibiotic-resistant bacteria; however, available data suggest a low prevalence in this population [59].Potential donors who have a permissive medical history must undergo to blood and fecal examination to exclude infective disease transmittable trough fecal transfer [13]. The tests may change between the various protocols, but there are some mandatory examinations (Table 2).In fact, blood testing should include complete blood cell count, liver enzyme, creatinine, and C-reactive protein to check overall clinical condition, serology for Hepatitis virus, and Human immunodeficiency virus (HIV). Furthermore, blood tests can be considered in case of anomalies of the first round of laboratory tests, endemic spread of some pathogens, emergence of new pathogens or selected cases of recipients (e.g., immunosuppressed). In particular, there is debate about the usefulness of serology for EBV and CMV, as the high prevalence of prior exposure among adult individuals weakens the diagnostic power of this approach, limiting the clinical utility to IgM CMV in donors dedicated to immunosuppressed recipients. Of course, it is not appropriate to exclude subjects with prior exposure to EBV or CMV from the donation because of the unlikely risk of transmission, unless clinical or laboratory suspicion of reactivation. Finally, the candidates could be considered for testing the serology for nematodes, based on social and geographical features and tests availability [13].Stool testing should include common enteric pathogens, Clostridium difficile, fecal parasites, and Helicobacter pylori antigen (this last exam only for upper route of FMT delivery). Enteric pathogens, which must also be investigated in asymptomatic subjects, should be detected with conventional methods (culture, microscopy, or antigen test) and/or with molecular diagnosis (PCR-based panels) that have shown a high specificity and sensitivity compared to conventional methods in rapid detection of pathogens [60]. Furthermore, it is mandatory to test all fecal samples for antibiotic-resistant bacteria (including meticillin-resistant Staphylococcus aureus (MRSA), vancomycin-resistant Enterococci (VRE), extended-spectrum β-lactamase-producing Enterobacteriaceae, and carbapenem-resistant Enterobacteriaceae/carbapenemase-producing Enterobacteriaceae), considering the burden of the gastrointestinal carriage in asymptomatic subjects [61,62] and the reporting of some serious adverse events associated to sepsis after FMT [13]. Nowadays, due to the emerging Covid-19 pandemic, a panel of international experts has suggested to include in the tests for Sars-CoV-2 a thorough nasopharyngeal swab and/or RNA detection in stool [63].Finally, if all blood and fecal tests are negative, the candidate is accepted to become a stool donor. Especially in the fecal bank program, the donor should be available to donate on many occasions over time. For this reason, it is advisable to repeat the screening tests every 8–12 weeks and administer a short questionnaire on the same day of the donation to check for any recent-onset harmful events.In this paragraph we have reported the general rules to select and to screen potential donor for FMT, mainly to treat CDI that is cured by the restorative effect of fecal transfer on gut microbiota. However, for other clinical indications, which find their rationale in the modification of metabolic and inflammatory pathways mediated by gut microbiota, it would be appropriate to identify a specific donor for each case. This issue will be discussed later.The correct recruitment of healthy donors is essential for a standardized and safe FMT procedure [11,13]. FMT is considered a safe procedure; however, mild adverse effects attributable to FMT are reported in about one third of the recipients, such as self-limiting abdominal discomfort or changes of bowel habits, and unfortunately, about 2–6% of patients experienced serious adverse events, such as infection, relapse of pre-existing disease, or death [64]. Moreover, the difficulty of selecting the appropriate candidates is increasing due to emerging concerns, as the possibility of transmission of putative procarcinogenic bacteria [65] or the potential risk of serious life threatening infections with multi-drug resistant organisms after FMT [66]. Moreover, recent evidences showed that the efficacy of FMT in recurrent CDI treatment, in clinical trials and in other healthcare settings seems to be linked to different variables, such as the delivery methods of fecal infusate, the bowel preparation, the number of infusion, the disease severity, and in particular to the microbial diversity and composition of the transplanted stools [32,44,67]. Since the idea that the success rate of FMT could be related to the gut microbiota or other features of the donor, the term “super-donors” has been introduced to indicate the ideal individuals whose stools could ensure a better outcome for recipients compared to others fecal donations [68]. Therefore, assuming that dysbiosis-related disorders have been associated to different imbalanced microbial signatures [15], in order to restore the eubiosis, it is reasonable to assume that reaching the correct donor-recipient match with targeted FMT based on specific microbial disturbances might be the key to improve FMT response. Accumulating evidence strengthens this hypothesis, leading to discard the concept of “one stool fits all” and to search an optimal donor [68], as in other organ transplantation procedures [69].The research of the ideal donor in recurrent CDI is obviously a widely debated topic of study. For example, one study identified the optimal donor among nine healthy vegetarian or vegan candidates, selecting the candidate who had a balanced Bacteroidetes/Firmicutes ratio, the highest alpha diversity among screened individuals, and high butyrate concentration. After 10 weeks from a single or multiple FMT, none of the 10 patients experienced CDI recurrence [70]. Of interest, the gut virome may also play a role in CDI treatment [71]. Indeed, enteric virome alterations marked by an increase in the abundance of Caudovirales, together with a decreased Caudovirales diversity, richness, and evenness, have been reported in patients with CDI. Moreover, CD eradication was associated with the colonization of a higher abundance of donor-derived Caudovirales contigs detected during follow up. These findings could possibly explain why bacterial fecal filtrate infusion resulted in effective treatment of CDI [72], and shifted the attention on the importance of the bacteriophages and on the potential role of selecting donors on the basis of their gut virome. Finally, some authors reported that selecting specific enteric bacterial strains with bacterial cultures from healthy donors to prepare a stool substitute blend might be a winning strategy to cure recurrent and antibiotic-resistant C. difficile colitis [73,74]. However, it is likely that the relevant impact on FMT success in CDI depends on the transfer of a complete fecal microbiome rather than specific bacterial strains; moreover, the promising results reported by the study that transfer the fecal filtrate alone suggest a predominant role for bacteriophages rather than for the specific relative abundance pattern of the gut microbiota of donor, shifting the central role from bacteria to viruses in the therapeutic challenge of FMT in CDI; however, these data are still preliminary and need to be confirmed by further studies.Many studies analyzed the microbial profile of donors and tried to relate it with clinical and laboratory outcomes in patients with IBD. Clinical outcomes and immunological changes after FMT in patients with IBD were significantly related to the variations of several specific strains in recipients of fecal microbiota [75]. For instance, intensive FMT in UC patients were associated with negative outcomes in case of abundance of Fusobacterium spp and Sutterella spp in recipients’ fecal microbiota after the FMT [76]. Furthermore, a study that involved refractory UC patients reported that pre-treatment with antibiotic plus repeated FMTs using fecal material from donor with a high bacterial richness and high relative abundance of Akkermansia muciniphila, unclassified Ruminococcaceae, and Ruminococcus spp. was more likely to induce remission compared to antibiotics alone [67]. As also described in other studies [41,77,78], it is plausible that choosing donors based on their taxonomic composition, in particular low or high abundance of specific strains, might reflect the possibility for future trials in IBD. For this purpose, methods aimed at preventing an inflammatory response of the recipient’s intestinal immune system by selecting compatible donors on their microbial profiles are under study [79]. Furthermore, the gut virome could represent a potential marker for FMT response in UC patients. In particular, results from a small case series reported that FMT responders already presented, before undergoing to FMT, a significantly lower eukaryotic viral richness than non-responders. Moreover, the richness of donor virome was not associated with the FMT outcome, as instead proposed for bacteria [80].Several preclinical and clinical studies supported the rationale for donor selection based on gut microbial profile in other disorders associated to gut dysbiosis. Indeed, in the field of anti-cancer treatment, it has been reported that microbiota can influence chemotherapy response [81]. Preclinical studies found a clinical improvement in mouse models of melanoma on anti-PD-1 therapy that received FMT from donors with a melanoma “responder-like” microbial signature (with high alpha diversity and abundance of Ruminococcaceae, Faecalibacterium, Bifidobacterium longum, Collinsella aerofaciens, and Enterococcus faecium) when compared to mice that received “non responder-like” microbiome (characterized by low microbial diversity and high relative abundance of Bacteroidales) [82,83]. Nevertheless, trials on humans, testing the effect of FMT in increasing the response to cancer therapies, are still in progress [84].Recently, a randomized placebo-controlled trial of FMT in IBS reported that the abundance of Streptococcus, Dorea, Lactobacillus, and Ruminococcaceae spp in the donor microbiota was associated with efficacy in relieving IBS symptoms [44]. Interestingly, a small open-label clinical trial evaluated the impact of prolonged FMT with antibiotic pre-treatment in children with autism; authors reported a decrease of gastrointestinal symptoms and an improvement of behavior, together with specific genera increase in recipients (Bifidobacterium, Prevotella, and Desulfovibrio). Conversely, Prevotella, and Desulfovibrio were more represented in recipients after FMT than in the donor samples, suggesting that unknown factors changed the intestinal ecosystem, making it more hospitable to these strains [85].Within the context of metabolic diseases, the effect of allogenic FMT post-Roux-en-Y gastric bypass donors was compared with metabolic syndrome donors on glucose metabolism and other parameters in treatment-naïve patients with metabolic syndrome. The authors assessed a decrease of insulin sensitivity in recipients who received FMT from donors with metabolic syndrome compared with using post-surgical donors. Moreover, they identified several microbial OTUs possibly predictive of metabolic response, suggesting a microbiota-related transmissible mechanism of insulin resistance [86]. Similarly, another study reported a significant increase in insulin sensitivity, together with altered microbiota composition, in patients with metabolic syndrome who received allogenic FMT from lean donors compared to those who underwent autologous FMT [46].To date, these results appear promising but partially controversial; thus, findings need to be confirmed with stronger evidence and by standardized clinical trial. Further research is needed to identify the favorable microbial signature of donor or other ideal features in disease-specific settings.In this review, the stool donor screening process has been described, and recent evidence has been reported that try to identify the optimal donor for each clinical condition (Figure S1).To date, the clinical characteristics of the donor are well defined; in particular, they are recommended to be a healthy volunteer with a balanced lifestyle, without chronic diseases or family history of metabolic diseases or cancer, and defined laboratory exams must certify the current absence of disease. However, identification of the ideal donor through the microbiological typing of the stool is currently not suitable. First of all, understanding the role of the intestinal microbiota in each chronic disease is an indispensable condition before hypothesizing a personalized approach through FMT. In fact, while the restorative mechanism of FMT in recurrent CDI is now understood, many aspects still need to be understood regarding the treatment of other chronic conditions. Interesting evidence has been reported regarding dysbiosis in IBD or in other chronic conditions, but the contrasting results reported in clinical trials of FMT could be justified by the choice of unsuitable donors. The identification of the microbiological characteristics of the ideal donor for each disease appears to be an achievable goal but still far from being accomplished due to the lack of clinical studies. The current evidence is still limited and insufficient for explaining and resolving the complexity of the interaction between the intestinal barrier and its role in gut-related chronic diseases. However, further studies need to be designed to confirm the encouraging results that have been reported in recent years. In particular, it will be necessary to type the fecal microbiota of the donor and the recipient, and to understand how environmental factors, such as diet, or individual features may benefit (or not) the clinical response to FMT. Understanding the microbial characteristics of the optimal donor, in particular if they are modifiable through lifestyle changes or pharmacological measures, could increase the therapeutic potential of FMT.The following are available online at https://www.mdpi.com/2077-0383/9/6/1757/s1, Figure S1: Optimal Stool Donor to Treat Specific Disorders.Conceptualization, S.B. and A.G.; data curation, S.B., C.R.S. and S.P.; writing—original draft preparation, S.B. and C.R.S.; writing—review and editing, S.B., C.R.S., S.P., E.B., G.I., G.C. and A.G. All authors have read and agreed to the published version of the manuscript.This research received no external funding.The authors declare no conflict of interest.Preliminary interview to select donors.Antimicrobial drugsImmunosuppressant agentsChemotherapyProton pump inhibitorsPersonal history of chronic gastrointestinal disease, including functional gastrointestinal disorders; inflammatory bowel disease; celiac disease; other chronic gastroenterological diseases or recent abnormal gastrointestinal symptoms (e.g., diarrhea, hematochezia, etc.)Personal history of cancer, including gastrointestinal cancers or polyposis syndrome, and first-degree family history of premature colon cancer Personal history of systemic autoimmune disordersObesity (body mass index > 30) and/or metabolic syndrome/diabetesPersonal history of neurological/neurodegenerative disordersPersonal history of psychiatric/neurodevelopmental conditionsHistory of HIV, hepatitis B or C viruses, syphilis, human T-lymphotropic virus I and IICurrent systemic infectionUse of illegal drugsHigh-risk sexual behaviorPrevious tissue/organ transplantRecent hospitalization or discharge from long-term care facilitiesHigh-risk travelNeedle stick accident in the last six months Body tattoo, piercing, earring, acupuncture in the last six monthsEnteric pathogen infection in the last two months Acute gastroenteritis with or without confirmatory test in the last two monthsHistory of vaccination with a live attenuated virus in the last two monthsDonor blood and stool testing.Complete blood cell countLiver enzyme (Aminotransferases)Bilirubin CreatinineC-reactive proteinSerology for Hepatitis virus (HAV, HBV, HCV, HEV) and Human immunodeficiency virus (HIV)
+                      Clostridium difficile
+                    Giardia lamblia, Cryptosporidium spp, Isospora and MicrosporidiaProtozoa and helminths and parasites (including Blastocystis hominis and Dientamoeba fragilis)Antibiotic-resistant bacteria Common enteric pathogens, including Salmonella, Shigella, Campylobacter, shiga toxin-producing Escherichia coli, Yersinia, and Vibrio choleraeNorovirus, rotavirus, adenovirusHelicobacter pylori fecal antigen

Med-MDPI/jcm_2/jcm-09-06-01758.txt

@@ -0,0 +1 @@
+M.A.S. and A.C.N. share first authorship of this study.In recent years there has been an increasing percentage of cochlear implant (CI) users who have usable residual hearing in the contralateral, nonimplanted ear, typically aided by acoustic amplification. This raises the issue of the extent to which the signal presented through the cochlear implant may influence how listeners process information in the acoustically stimulated ear. This multicenter retrospective study examined pre- to postoperative changes in speech perception in the nonimplanted ear, the implanted ear, and both together. Results in the latter two conditions showed the expected increases, but speech perception in the nonimplanted ear showed a modest yet meaningful decrease that could not be completely explained by changes in unaided thresholds, hearing aid malfunction, or several other demographic variables. Decreases in speech perception in the nonimplanted ear were more likely in individuals who had better levels of speech perception in the implanted ear, and in those who had better speech perception in the implanted than in the nonimplanted ear. This raises the possibility that, in some cases, bimodal listeners may rely on the higher quality signal provided by the implant and may disregard or even neglect the input provided by the nonimplanted ear.Indications for cochlear implantation have changed significantly since these devices were introduced to clinical practice. Initially, only patients with profound to total hearing loss were implanted, but as outcomes of cochlear implantation improved with the development of newer devices, electrodes, and stimulation strategies, it became more common for patients with residual hearing to receive a cochlear implant. In 2014, the FDA expanded cochlear implantation criteria and gave premarket approval for implantation in ears that have useable acoustic hearing in the low frequencies [1]. Most recently, FDA approval was granted for cochlear implantation for single-sided deaf patients, defined as those who have “profound sensorineural hearing loss in one ear and normal hearing or mild sensorineural hearing loss in the other ear” [2]. The new audiometric criteria for cochlear implantation have yielded a quickly expanding group of persons with cochlear implants and most of them are “bimodal patients”: those who have usable acoustic hearing in the contralateral ear, typically aided by a hearing aid. As these criteria continue to evolve, it is likely that an even larger proportion of future CI patients will have usable hearing in the contralateral ear. The public health significance of this new population is evident when we compare the numbers: more than 800,000 Americans report being unable to hear normal conversation even while using a HA, in addition to the 180,000 classified as “deaf” [3]. In contrast, only 58,000 adults were reported to have received CIs in the United States as of 2012 [4].Initial research on bimodal hearing indicated that even persons with severe to profound hearing loss in the nonimplanted ear could obtain bimodal benefit (e.g., [5,6,7,8,9,10,11,12,13,14]). More recent research has focused on bimodal benefit obtained by persons with a unilateral cochlear implant who have mild to severe hearing loss in the nonimplanted ear. In these cases, bimodal benefit is obtained from amplification to the low frequency region, even when speech recognition performance with the HA alone is very poor (e.g., [15,16,17,18,19,20,21]). Bimodal benefit is also observed in patients with asymmetric hearing loss who receive a cochlear implant in their worse ear [18,19,22,23,24,25,26,27].Bimodal benefit on speech recognition tasks is characterized by a great deal of variability among individuals. For example, in a report by Gifford et al. (2010) [28] 14 of 18 bimodal patients showed higher Consonant–Nucleus–Consonant (CNC) word scores in the bimodal condition than in the CI-only condition. Benefit ranged from 6% to 45%. The mean improvement for the group was 15%. Dorman et al. (2015) [29] reported results of CNC tests from 105 bimodal patients with varying amounts of acoustic hearing in the nonimplanted ear. Mean bimodal benefit on the CNC test material was related to the amount of low frequency acoustic hearing (125, 250 and 500 Hz). Those with the best low-frequency acoustic hearing obtained a mean benefit of 16%, while those with mean low frequency thresholds exceeding 61 dB hearing level (HL) obtained a mean benefit of 4%. But benefit of individual participants ranged from −15% to 55%. In general, the benefit was greater for speech recognition in noise. For example, on the AZBio test material (Auditory Potential, LLC, Goodyear, AZ 85338) presented at +10 dB signal-to-noise ratio (SNR) [10,16,17] mean benefit was 26% for those with the highest level of acoustic hearing, but only 8% for those with the least low-frequency acoustic hearing. Even though many derive some benefit from the simultaneous use of a CI and a contralateral HA, for some, use of the HA results in decreased performance (e.g., [12,30,31,32,33]). Taken together, these results suggest that many patients obtain levels of bimodal benefit that are clinically relevant whereas others simply use “best ear listening” (with the best ear being the one with the cochlear implant, typically) or may even experience interference from the weaker ear signal (usually the HA ear).Some of these speech perception results may be driven by asymmetric hearing, defined as a situation where one ear receives (and conveys to the brain) much more information than the other one. Auditory asymmetries are known to have consequences that have been demonstrated in animal models and human listeners [21,34,35]. This can happen when the input asymmetry is due to actual physiological asymmetries, as in the examples just referenced, or even when it is imposed by unilateral acoustic aiding in the face of bilateral, symmetric hearing impairment. The latter phenomenon has been called “late onset auditory deprivation”, and it was first reported by Silman et al. (1984) [36] (for a review, see Neuman, 1986 [37]). The study observed that many monaurally aided adults with bilateral symmetric sensorineural hearing impairment exhibited decreases in speech-recognition ability in the unaided ear, but not in the aided ear. This happened even though the unaided thresholds remained similar in the aided and the unaided ear.We first reported in 2011 [38] an interesting pattern in bimodal cochlear implant users that was reminiscent of Silman et al.’s (1984) results. A retrospective study of bimodal CI users found statistically significant speech perception decreases in the nonimplanted ear in 11 out of 32 patients after cochlear implantation. It is notable that this analysis excluded any patients whose unaided thresholds in the nonimplanted ear changed by more than 10 dB after implantation, suggesting the possibility that those changes in speech perception could not be completely explained by changes in hearing sensitivity. These results prompted the organization of the multicenter study we report here, aimed at tracking possible changes in speech perception in the acoustically aided ear of bimodal patients after cochlear implantation. Importantly, this study only includes data from patients where there was some indication of correct hearing aid functioning. The main goals were: first, to determine whether the initial observation would be replicated in a larger, multicenter sample, and second, to start teasing out possible explanations for the phenomenon.Participants were adults with post-lingual hearing loss who used a cochlear implant (CI) in one ear and a hearing aid (HA) in the contralateral ear (bimodal users). To be included in the study, documentation of hearing pre- and post-implant surgery was required for the implanted ear and the non-implanted ear. These longitudinal measures included documentation of unaided pure tone thresholds in the nonimplanted ear, as well as speech recognition performance pre- and post-implantation in the CI ear, the HA ear, and in the bimodal condition. An additional requirement was documentation that showed the hearing aid was functioning properly during testing. This requirement could be met in one of four possible ways: 1) in-the ear measures showing that hearing aid gain was adequate; 2) aided and unaided thresholds in the HA ear; 3) unaided speech perception scores in the HA ear pre- and post-implant (if a drop in speech perception scores was not accompanied by a similar drop in unaided scores in the same ear, we assumed that this was due to hearing aid malfunction or improper fitting and the data were excluded); or 4) indication from an audiologist or hearing aid technician that the HA was operating correctly during the evaluation.There were 132 subjects: 68 male and 64 female. The average age at implantation was 56.8 years (minimum 15, maximum 89). Table 1 and Table 2 show how the 132 subjects are classified by etiology, device, electrode, and speech processor. The number of post-implant sessions ranged from 1 to 11, with an average of 2.5 sessions. The average duration of post-implant follow-up for the hearing aid ear was 3.09 years, ranging from just one month to as much as 11 years. Figure 1 shows the average preoperative and postoperative unaided audiograms in the nonimplanted ear. The difference in pure-tone average is 2.5 dB. Given the average age of the subjects, the likelihood that some of them had progressive hearing loss, and the fact that some hearing loss is expected due to presbycusis (as subjects were followed for over three years, on average), this 2.5 dB difference does not seem unusual. Verification of hearing aid status was done by comparing aided and unaided thresholds for 81 subjects and by audiologist report for 41 subjects. In-the-ear verification or comparison of aided and unaided speech scores was not used for any of the subjects. Lastly, it was not possible to obtain information about hearing aid status for ten of the subjects, so their data were excluded from several of the analyses, as reported below.Data was collected at eight different testing sites. Speech testing consisted of words or sentences in quiet presented via audition alone (no visual cues) at 65 dB HL (unless otherwise noted) and zero degrees azimuth. See Appendix A Table A1 for a summary of speech testing materials and sample size at each site.Speech perception scores before implantation were compared to post-implantation speech scores. This was done separately for three conditions: implanted ear only (in the case of pre-implantation data, this means the ear to be implanted, using acoustic amplification when appropriate), nonimplanted ear only, and bilateral. These comparisons were made in two different ways: comparing pre-implant to post-implant scores averaged over all postoperative sessions, or just to the latest available postoperative score. These two ways of comparing preoperative to postoperative data yielded similar results. The statistical significance of pre- to postoperative differences was assessed at the group level using paired t-tests, or the Wilcoxon signed rank test when data failed to follow a normal distribution. Follow-up tests were conducted on subsets of the data to help tease out possible reasons for preoperative to postoperative changes in speech perception in the nonimplanted ear. Specifically, comparisons were made for the original dataset of 132 subjects who had pre- and postoperative scores in the nonimplanted ear and for the same group after excluding any evaluation session when correct hearing aid operation could not be ascertained (N = 122). By design, all datapoints obtained postoperatively in the nonimplanted ear were included if the pure-tone average (PTA) did not differ from the preoperative PTA by more than 25 dB. To better rule out the effect of progressive hearing loss, analyses of speech scores in the nonimplanted ear were conducted after removing datapoints showing an increase of more than 20 dB PTA. The remaining N was 120 for the preoperative to average postoperative comparison, and N = 118 for the preoperative to latest postoperative comparison. The process of elimination was repeated for datapoints showing a PTA increase of more than 10 dB, and the resulting numbers of the remaining subjects were N = 107 and N = 104, when comparing preoperative scores to the average postoperative and latest postoperative scores, respectively. Finally, this was also done after excluding datapoints showing a change of more than 5 dB in PTA. The resulting numbers of subjects in this case were N = 92 and N = 85 for the two types of comparison.Pre- to postoperative comparisons were also conducted for each subject and condition (implanted ear, nonimplanted ear, bilateral) by generating critical difference tables as described by Carney and Schlauch [39]. The critical difference tables were generated using a Matlab program that works in the general case when different numbers of items are used in the two scores to be compared, which can happen, for example, when comparing a score obtained with one 50-word list to a score obtained as the average of three 50-word lists (the Matlab code can be downloaded from https://osf.io/fha47/?view_only=3d5c2e239a64496481ea657215913469). This was done separately for each test. To determine whether two scores are significantly different, computer simulation was used to generate confidence intervals based on the number of observations used to obtain each score. Let n1 and n2 represent the number of observations from which those two scores were obtained, respectively. Next, consider all possible scores (as proportions) that can be obtained for these numbers of observations. For example, if n1 = 50, then the possible scores are 0, 0.02, 0.04, …, etc., and if n2 = 25, then the possible scores are 0, 0.04, 0.08, …, etc. Let p1(i) represent the set of possible scores for n1 where i = 1, 2, …, n1 + 1, and p2(j) represent the set of possible scores for n2 where j = 1, 2, …, n2 + 1. For every pair of possible scores from p1(i) and p2(j), one calculates a mean absolute difference score using computer-generated binomially distributed pseudorandom numbers for p1(i) and p2(j), repeated over many iterations. That is, on one iteration, two binomially distributed numbers are generated, one based on p1(i) and n1 and the other based on p2(j) and n2. The numbers are then converted to proportions by dividing by the respective number of observations, n1 and n2. The absolute value of the difference between these two scores is then recorded. This process is repeated for any number of desired iterations, e.g., 40,000. For this set of 40,000 absolute difference scores, one can calculate a z-score equal to the mean absolute difference divided by the standard deviation estimate across absolute difference scores. Having generated a z-score for each p1(i) and p2(j), one can find the values of p1(i) and p2(j) that produce z-scores exceeding a critical value (e.g., 1.96 for a normal distribution and a two-tailed test using a p-value of 0.05). For p1(i), the values of p2(j), whose z-scores fall within +1.96 and −1.96, define the critical range.Some data from two of the sites were obtained using scores of correct keywords in sentences. The TAM test has between 114 and 125 scored words per list and the CUNY sentence test has 102 words per list, on average. In those cases, critical significance tables are used N = 40, which have been shown to be an appropriately conservative assumption for a sentence test like AzBio (which has 142 words per list) when scoring it using the percentage of correct keywords [40].The change in speech scores in the nonimplanted ear (which was typically a decrease) was correlated with three variables of interest: the preoperative to postoperative change in pure-tone average, the speech score in the implanted ear, and the implanted ear advantage (difference in speech score between the implanted and the nonimplanted ear). The latter correlation was also calculated after partialling out the effect of the initial score in the nonimplanted ear. It was expected that the pattern of correlations might help tease out the effect of some possible causes of speech perception drops in the nonimplanted ear. A significant correlation between drops in speech scores and in pure-tone average in the nonimplanted ear would suggest that the progression of hearing loss might be at least partly responsible for decreases in speech scores. Conversely, a significant correlation between drops in speech scores in the nonimplanted ear and either speech scores in the implanted ear or the implanted ear advantage would suggest that the speech perception drops in question are at least partly due to a more central mechanism: namely, neglect of the poorer ear.Lastly, and for the sake of completeness, the change in speech scores in the nonimplanted ear was also correlated with several preoperative demographic variables: age at cochlear implantation, age at moderate hearing loss, age at profound hearing loss, preoperative aided and unaided PTA in both the implanted and the nonimplanted ear, whether cochlear implantation was done using a cochleostomy or a round window approach, whether the left or the right ear was implanted, whether a hearing was used in the ear to be implanted, years of education, and gender.This study was approved by the New York University Medical Center Institutional Review Board (S12-02951).Figure 2 shows a comparison of preoperative vs. postoperative speech perception scores. The top panels compare preoperative scores to the average of all postoperative scores for a given subject. The bottom panels compare preoperative scores to the latest available postoperative score. As we will see, the conclusions are unaffected whether we examine scatterplots in the top or the bottom panel. Sets of panels, left to right, indicate speech perception scores obtained with the cochlear implant ear, both ears at the same time, or the acoustically stimulated ear. Each symbol represents one subject and color is used to indicate whether the postoperative scores are significantly different (p < 0.05) from preoperative scores for each subject. Green indicates that the postoperative score was significantly higher, red indicates that the postoperative score was significantly lower, and blue indicates no significant difference between preoperative and postoperative scores.The left panels, not surprisingly, show that speech perception in the implanted ear improved significantly for the vast majority of subjects (94 out of 102, regardless of whether we examine average postoperative performance or the latest available score). This expected result is more impressive when we consider that an additional 29 subjects could not be included because their preoperative score was listed as “could not test”, which almost always means that the subject in question did not have sufficient residual hearing in the ear to be implanted to understand any words without lipreading, so the score would have been zero. The same happened with speech perception in the binaural condition shown in the center panels: 62 out of 82 subjects showed statistically significant improvement when comparing to the latest datapoint, and this was the case for 61 subjects when comparing against the average postoperative score. For many of the remaining 19 subjects, it would have been impossible or at least very difficult to show a statistically significant improvement due to ceiling effects. Again, 41 subjects were not evaluated preoperatively in the binaural condition.A very different picture is seen when examining preoperative to postoperative changes in the nonimplanted ear (right panels). Here, we see that 41 subjects scored significantly lower (p < 0.05) in their latest postoperative evaluation than they did preoperatively (right bottom panel) and 38 subjects did so when comparing their preoperative score to their average postoperative score. Twelve subjects obtained significantly higher scores in their latest postoperative evaluation compared to the preoperative one, and 11 showed a similar change when comparing their preoperative score to their average postoperative score. Remember that some of these “statistically significant” differences would have been observed even in the absence of any real change. Because 132 subjects were evaluated and a p-value of 0.05 was used as a threshold, we would expect an average of three subjects (2.5% of 132, rounded up to the nearest integer) to show significant increases and another three to show significant decreases. Thus, the number of subjects with postoperatively decreased scores in the hearing aid ear far exceeds what would be expected due to random variation in the absence of a real effect. This was true both at the site with the highest number of subjects (NYU, where 18 out of 53 subjects had significantly lower scores in the latest postoperative evaluation than in the preoperative evaluation) and in the rest of the sites, where this was true for 23 out of 79 subjects.All preoperative-to-postoperative group differences were highly significant (p < 0.001). Postoperative scores were significantly higher for the implanted ear and for the binaural condition, and lower for the nonimplanted ear. Because the results in the nonimplanted ear were somewhat unexpected, we also examined the preoperative–postoperative differences using increasingly strict criteria. First, we excluded ten subjects for whom proper hearing aid operation could not be confirmed. Then, we excluded subjects who showed more than 20, 10, or 5 dB of hearing loss in the nonimplanted ear during the course of the study. Figure 3 and Table 3 show the results of these additional analyses. As can be observed, after cochlear implantation there is a robust decrease in speech perception in the nonimplanted ear. The effect is highly significant whether we consider the latest available point for each subject or the average of all postoperative scores, and whether we include or exclude data from subjects whose hearing loss became more pronounced during the study. We do observe that this group effect is somewhat less pronounced (but still present) when we only include data from subjects whose PTA did not change by more than 5 dB.To help tease out some of the reasons for the decrease in speech perception in the nonimplanted ear, correlations between that decrease and two relevant variables were examined. For this analysis we only included datapoints when there was evidence that the hearing aid was functioning properly. The left panel of Figure 4 shows the decrease in speech perception in the nonimplanted ear speech score decrease as a function of change in pure-tone average for the same ear. As the legend indicates, there was a correlation in the expected direction (increase in PTA is associated with a drop in speech perception score in the nonimplanted ear) but the correlation is weak (r = +0.157) and fails to reach statistical significance (p = 0.08). The middle panel shows the speech perception drop in the nonimplanted ear as a function of the implanted ear advantage (that is, the extent to which speech perception is better in the implanted than in the nonimplanted ear). In other words, we examine the extent to which asymmetry between ears (with the implanted ear providing more information than the nonimplanted ear) may be associated with speech perception drops in the nonimplanted ear. That association seems to be supported by the correlation in the middle panel (r = +0.507, p < 0.001), but that may be a partially spurious effect driven by the fact that both variables are related to postoperative hearing aid scores. We address this issue in two ways. First, we found that the correlation becomes weaker but remains significant when partialling out the effect of postoperative hearing aid scores (r = +0.247, p = 0.0067). Second, there was also a positive correlation between average postoperative CI speech scores and the speech perception drop in the nonimplanted ear (r = +0.217, p = 0.0174; the corresponding scatterplot is shown in the right panel of Figure 4).None of the other preoperative demographic variables were significantly correlated with the drop in speech perception in the nonimplanted ear, although one of them came close. The drop in speech perception between the preoperative evaluation and the average postoperative score showed a weak (r = +0.106) nonsignificant correlation (p = 0.059, NS) with age at cochlear implantation.The most interesting result we have presented is in the right panels of Figure 2 and Figure 4. Average speech perception scores in the nonimplanted ear showed an unexpected and statistically significant decrease after cochlear implantation. In principle, we do not expect any changes in the nonimplanted ear, particularly in the absence of any major changes in hearing sensitivity in that ear.Results for the other two conditions (CI-only and bimodal) gave the expected pre- vs. postoperative improvements in speech perception scores (left and center panels of Figure 2), whether we consider the average postoperative score for each individual (top row of panels in Figure 2) or the latest datapoint (bottom row of panels). The vast majority of the subjects showed improvements that were both statistically significant and meaningful in a practical sense.There are several potential factors that may underlie these unexpected changes in speech perception in quiet in the nonimplanted ear. Determining the precise combination will likely require a prospective follow-up study, but we can speculate about various possibilities. One of them is hearing aid malfunction, or at least improper hearing aid fitting. Recall, however, that the present study attempted to minimize this factor by requiring documentation that the hearing aid was functioning properly during testing, and that this requirement was met by each datapoint that was included in the study. This makes it less likely that the drops in speech perception were due to hearing aid malfunction or improper fitting. One caveat, however, is that many of the datapoints were confirmed simply by an audiologist’s indication that the hearing aid was functioning properly. A prospective study should require more reliable confirmation of proper hearing aid function, ideally by using in-the-ear measures showing that the fitting targets have been achieved.A second possibility is that of progressive hearing loss in the nonimplanted ear, which could conceivably result in speech perception drops even with a properly fit hearing aid. However, it seems very unlikely that this factor could explain all of the observed change in speech perception in the nonimplanted ear, for three reasons. First, Figure 1 shows that the average group change in pure tone average in the nonimplanted ear was quite small: only 2.5 dB. Second, the scatterplot in the left panel of Figure 4 shows that very few individuals experienced pure-tone average drops of 10 dB or more, and even those who did were not more likely to also experience decreases in speech perception than those who did not. The correlation between change in pure-tone average and speech perception in the nonimplanted ear failed to reach statistical significance. Third, Figure 3 shows that gradually tightening the requirements for data inclusion did not substantially change the main outcome. We first excluded 10 patients for whom hearing aid functionality was not confirmed, and this left the drop in hearing aid speech scores largely unaffected (for example, the drop from preoperative to latest postoperative data point only changed by a small fraction of a percentage point). It is interesting to note that as we then remove patients whose PTA increased by more than 20, 10, or even 5 dB over time after cochlear implantation, the group average drop in speech perception in the nonimplanted ear goes from 9.9% to 8.6%, 8.0%, and 6.7% (bottom panel of Figure 3). Two observations relate to this result. First, even though drops in speech perception in the nonimplanted ear were not significantly correlated with increases in PTA, it seems quite possible that hearing loss progression underlies a fraction of the observed result. The second observation is that hearing loss progression is unlikely to explain the totality of the observed effect. There must be other mechanisms at play.A third possibility is that this phenomenon might be at least conceptually related to that of late onset auditory deprivation, first reported by Silman et al. (1984) [36]. This refers to decreases in speech perception after a prolonged lack of amplification in the unaided ears of some unilaterally aided patients with bilateral hearing loss. It is at least possible that some of the patients in the present study stopped using their hearing aids or started using them less consistently after implantation. This may have happened after speech perception with the cochlear implant became much better than with their acoustically stimulated ear.Alternatively (and this is a fourth possibility), it may be that speech perception decreases happened despite the continued use of acoustic input in the nonimplanted ear. One possible mechanism has to do with the adaptation that is required from postlingually deafened cochlear implant patients who have acoustic hearing in the nonimplanted ear. The frequency-place function imposed by a cochlear implant is typically different from the normal hearing frequency-place [41], and it is possible that some listeners may have difficulty adapting to different functions in each ear, the one with the cochlear implant and the one with acoustic amplification. In their case, the adaptation process that allows them to understand speech better with the cochlear implant may be maladaptive for the purpose of processing acoustic input.The finding that drops in speech perception in the nonimplanted ear were correlated with implanted ear advantage (that is, the extent to which perception in quiet is better in the implanted ear than in the nonimplanted, acoustically aided ear) or even with the raw speech perception scores in the implanted ear suggests that the third or fourth possibilities discussed above (and which could be described as bimodal auditory neglect) may underlie some of the observed changes. In other words, to the extent that speech perception in the implanted ear may be much better than in the acoustically aided ear, this may result in a central auditory process wherein the brain relies on the higher quality signal provided by the implant and disregards the input provided by the nonimplanted ear.A number of important caveats and suggestions for future research must be listed. Ideally, future studies should obtain more precise determinations of hearing aid fitting, particularly the gold standard of in-the-ear verification. Additionally, and given that both modern cochlear implants and hearing aids can log data about device use, it would be good to determine the extent to which the observed speech perception drops in the nonimplanted ear may or may not be related to the daily duration of hearing aid use, or the difference in daily duration of cochlear implant and hearing aid use.Taken together, the present results suggest that there are considerable individual differences in the way acoustic and electrical auditory input interact. Better understanding of this interaction may help guide future clinical practice in cochlear implant patients who have usable residual hearing.All authors have approved this manuscript. The following are author contributions and acknowledgements: Patient testing: E.T., F.S., R.H.; Data retrieval: A.T.d.M.M., M.V.S.G.-G., E.T., F.S., R.H., M.M., R.C.D., R.S., P.S., P.H.S., K.B.C., H.S., F.B., M.H., J.D.N., K.A.A., A.L., M.K.M., P.G., G.D.C.; Conceptualization, Methodology, and Analysis: M.A.S., A.C.N.; Software: M.A.S.; Writing-Original Draft Preparation: M.A.S., J.D.N.This research was funded by NIH grants R01-DC011329 (Principal Investigators: Svirsky and Neuman) and R01-DC03937 (Principal Investigator: Svirsky).The authors declare no conflict of interest.Summary of testing materials. * presented at 65 dB HL unless noted otherwise.Group mean of preoperative and average postoperative unaided thresholds.Postoperative vs. preoperative speech perception scores with the implanted ear only (left), both ears (bimodal condition, center), and the nonimplanted ear only (right). The top panels present average postoperative scores in the y-axis whereas the bottom panels show the latest available postoperative datapoint.Postoperative speech perception scores in the nonimplanted ear are lower than preoperative scores even after excluding subjects whose hearing aid functioning could not be verified, or after excluding datapoints from subjects who showed decreases in pure-tone average greater than 20, 20, or 5 dB. The top panel compares preoperative data to average postoperative data and the bottom panel compares it to the latest available postoperative data.Drop in postoperative speech perception scores in the nonimplanted ear as a function of: change in pure-tone average (left panel), implanted ear advantage (or degree to which speech perception is better in the implanted than in the nonimplanted ear, center panel), and speech perception score in the implanted ear (right panel).Etiology.Cochlear implant brand, electrode, and speech processor.Preoperative and postoperative speech perception scores in the nonimplanted ear (% correct).Left columns show comparison to average postoperative scores, right columns show comparison to latest available score. Rows show comparisons for progressively restricted subsets of subjects.

Med-MDPI/jcm_2/jcm-09-06-01759.txt

@@ -0,0 +1 @@
+The aim of this study was to investigate the management and outcome in the post-laser twin anemia polycythemia sequence (TAPS). Data of the international TAPS Registry, collected between 2014 and 2019, were used for this study. The primary outcomes were perinatal mortality and severe neonatal morbidity. Secondary outcomes included a risk factor analysis for perinatal mortality and severe neonatal morbidity. A total of 164 post-laser TAPS pregnancies were included, of which 92% (151/164) were diagnosed antenatally and 8% (13/164) postnatally. The median number of days between laser for TTTS and detection of TAPS was 14 (IQR: 7–28, range: 1–119). Antenatal management included expectant management in 43% (62/151), intrauterine transfusion with or without partial exchange transfusion in 29% (44/151), repeated laser surgery in 15% (24/151), selective feticide in 7% (11/151), delivery in 6% (9/151), and termination of pregnancy in 1% (1/151). The median gestational age (GA) at birth was 31.7 weeks (IQR: 28.6–33.7; range: 19.0–41.3). The perinatal mortality rate was 25% (83/327) for the total group, 37% (61/164) for donors, and 14% (22/163) for recipients (p < 0.001). Severe neonatal morbidity was detected in 40% (105/263) of the cohort and was similar for donors (43%; 51/118) and recipients (37%; 54/145), p = 0.568. Independent risk factors for spontaneous perinatal mortality were antenatal TAPS Stage 4 (OR = 3.4, 95%CI 1.4-26.0, p = 0.015), TAPS donor status (OR = 4.2, 95%CI 2.1–8.3, p < 0.001), and GA at birth (OR = 0.8, 95%CI 0.7–0.9, p = 0.001). Severe neonatal morbidity was significantly associated with GA at birth (OR = 1.5, 95%CI 1.3–1.7, p < 0.001). In conclusion, post-laser TAPS most often occurs within one month after laser for TTTS, but may develop up to 17 weeks after initial surgery. Management is mostly expectant, but varies greatly, highlighting the lack of consensus on the optimal treatment and heterogeneity of the condition. Perinatal outcome is poor, particularly due to the high rate of perinatal mortality in donor twins.The twin anemia polycythemia sequence (TAPS) can arise from chronic unbalanced feto-fetal transfusion through minuscule placental anastomoses in monochorionic twins, leading to anemia in the TAPS donor and polycythemia in the TAPS recipient [1]. Unlike twin-twin transfusion syndrome (TTTS), TAPS develops in the absence of the twin oligohydramnios-polyhydramnios sequence. The iatrogenic form of TAPS, post-laser TAPS, can develop in 2–16% after laser surgery for TTTS due to the presence of minuscule residual anastomoses [2,3,4]. The rate of post-laser TAPS can be reduced by using the Solomon technique instead of the selective laser technique (3% vs. 16%, respectively) [2]. With this approach, the entire placental vascular equator is laser photocoagulated, thereby blocking all anastomoses, even the miniscule ones that may not be visualized. Although the rate of residual anastomoses has significantly dropped after the implementation of the Solomon technique, post-laser TAPS may still occur [2,5,6]. Options to manage post-laser TAPS include expectant management, preterm delivery, intrauterine transfusion (IUT) with or without a partial exchange transfusion (PET), fetoscopic laser surgery, and selective feticide [7]. The best treatment has not been established. The technical feasibility of a second intrauterine intervention may be limited due to complications that have arisen from the initial laser procedure for TTTS, such as amnion-chorion separation or preterm premature rupture of the membranes (PPROM). For informed decision making regarding the preferred intervention, it is crucial to have insight into perinatal outcome of post-laser TAPS twins. Due to the rarity of the complication, perinatal outcome is insufficiently investigated, and available information is based on small cohort studies. To expand our knowledge on TAPS, we set up the TAPS Registry, an international collaboration aimed at collecting data on diagnosis, treatment, and outcome in TAPS twins.In the current study, the data from this TAPS Registry were used to (1) characterize diagnosis, treatment modalities, and outcome in post-laser TAPS twins, (2) to compare perinatal outcome between TAPS donors and recipients, and (3) to investigate potential risk factors for adverse perinatal outcome.The TAPS Registry, established in 2014, was a web-based registry for anonymous data collection (www.tapsregistry.org). Fetal therapy centers across the world were invited to participate. Participating centers were provided with personal credentials to enter the data of their TAPS cases into the online registry. Between 2014 and 2019, seventeen specialized fetal therapy centers contributed to data collection (see Appendix A).All monochorionic twin pregnancies diagnosed with post-laser TAPS were considered eligible for this study. Pregnancies with spontaneous TAPS were excluded and were described in a separate study [8].Antenatal diagnosis of TAPS was based on discordant middle cerebral artery peak systolic velocity (MCA-PSV) measures, with an increased MCA-PSV (>1.5 multiples of the median (MoM)) in the TAPS donor, indicative of fetal anemia, combined with a decreased MCA-PSV measure (<1.0 MoM) in the TAPS recipient, indicative of fetal polycythemia [9]. Postnatal diagnosis was reached by the presence of an inter-twin hemoglobin difference > 8.0 g/dL combined with at least one of the following: a reticulocyte count ratio > 1.7 or the presence of only minuscule vascular anastomoses (diameter < 1 mm) detected through color dye injection of the placenta [10,11]. Cases that were diagnosed with TAPS within one week after laser surgery for TTTS were excluded from the study, unless TAPS persisted. We did this as a large MCA-PSV discrepancy shortly after laser is likely to be a result of hemodynamic re-equilibration and is not based on TAPS [12].The following information was obtained from local medical records: gravidity, parity, location of the placenta, moment of diagnosis (antenatal or postnatal), gestational age (GA) at diagnosis, TAPS stage at diagnosis, and the presence of additional ultrasound findings including "starry-sky liver” in the recipient and/or a difference in placental echogenicity. The severity of antenatal TAPS was determined according to the previously published staging system by Slaghekke et al. [13]. For post-laser TAPS specifically, the following data regarding preceding TTTS were collected: Quintero stage, GA at laser, laser technique, operator’s opinion on completeness of the laser procedure, and TTTS-donor-recipient role. In addition, the antenatal management for TAPS was recorded, including expectant management, delivery (defined as a delivery within 7 days after diagnosis), IUT (± PET), fetoscopic laser surgery, selective feticide, and termination of pregnancy (TOP). Furthermore, information on placental color dye injection was collected, including classification of the type (arterio-venous (AV), veno-arterial (VA), arterio-arterial (AA), veno-venous (VV)), and the number and size of anastomoses. Perinatal outcome measures included: TAPS donor/recipient status, birth weight, hemoglobin and reticulocyte values, treatment with blood transfusion or partial exchange transfusion on Day 1, the presence of severe neonatal morbidity and/or severe cerebral injury, and the occurrence of perinatal mortality.Primary outcomes were perinatal mortality and severe neonatal morbidity. Perinatal mortality was defined as fetal demise or neonatal death within 28 days after birth. In the context of selective feticide or TOP, a distinction was made between spontaneous fetal demise and intended fetal demise. Severe neonatal morbidity was defined as the presence of at least one of the following, diagnosed within 28 days after birth or prior to discharge: respiratory distress syndrome requiring mechanical ventilation and surfactant, patent ductus arteriosus requiring treatment, necrotizing enterocolitis ≥ Stage 2 [14], retinopathy of prematurity ≥ Stage 3 [15], amniotic band syndrome, ischemic limb injury, or severe cerebral injury. Severe cerebral injury was diagnosed in case of one of the following abnormalities was identified on cerebral imaging: intraventricular hemorrhage ≥ Stage 3 [16], ventricular dilatation (including post-hemorrhagic ventricular dilatation) [17], cystic periventricular leukomalacia ≥ Grade 2 [18], porencephalic or parenchymal cysts, arterial infarction, or other severe cerebral lesions associated with an adverse outcome.Secondary outcomes included diagnosis- and therapy-related characteristics, hematological and placental characteristics, and a risk factor analysis for spontaneous perinatal mortality and severe neonatal morbidity. Cases with intended fetal demise in the context of selective feticide or termination of pregnancy were excluded for the risk factor analysis for spontaneous perinatal mortality. Since TAPS cases may be managed according to different strategies in one pregnancy, management-group assignment was based on the first treatment that was performed. The following parameters were investigated in the univariate risk analysis: GA at diagnosis of TAPS, GA at laser for TTTS, days between laser for TTTS and development of TAPS, Quintero stage, antenatal TAPS stage, persistence of TTTS-TAPS donor-recipient status, type of antenatal management, and GA at birth (in weeks). For antenatal TAPS stage, the highest recorded antenatal TAPS stage was selected. In case of TAPS Stage 5, the highest TAPS stage before Stage 5 was used. For the risk factor analysis for severe neonatal morbidity, two more parameters were added: severe growth restriction defined as birth weight < 3rd centile and the presence of postnatal TAPS.The following additional outcomes were determined: inter-twin hemoglobin difference (highest hemoglobin value–lowest hemoglobin value), reticulocyte count ratio (highest reticulocyte value (‰)/lowest reticulocyte value (‰)), the presence of severe growth restriction (defined as a birth weight < 3rd centile according to Hoftiezer [19]), postnatal TAPS stage (according to Slaghekke [7]), and the configuration of anastomosis type per TAPS placenta.Statistical analyses were carried out using SPSS Version 25.0 (IBM, Armonk, NY, USA). Data are presented as the mean ± standard deviation (SD) or as median and interquartile range (IQR) and/or range (minimum-maximum), as appropriate. A p-value < 0.05 was considered statistically significant. Differences between donors and recipients were calculated using the paired t-test for normally distributed continuous outcomes. To account for the fact that observations between co-twins are not independent, the Generalized Estimated Equation module was executed for analyses per fetus or neonate. Potential risk factors were checked for correlation using Spearman’s rank test (R). A correlation coefficient R > (−) 0.7 was considered to indicate a strong relationship between the factors. Potential risk factors for perinatal mortality and severe neonatal morbidity were assessed in a univariate logistic regression model. A multivariate logistic regression model was applied to the variables that showed significant association in the univariate analysis. Results are expressed as odds ratios (OR) with 95% confidence intervals (CI).Of the 422 TAPS cases entered into the TAPS Registry, two-hundred forty-nine (59%) were spontaneous TAPS and were excluded from the study. The remaining 173 (41%) were post-laser TAPS and eligible for the study. In eight post-laser TAPS cases, TAPS was diagnosed within one week after laser for TTTS. As TAPS did not persist one week after laser surgery for TTTS, these cases were excluded. One case was excluded based on TAPS being diagnosed at Stage 5. A total of 164 post-laser TAPS cases were included in the analysis for the current study.TAPS was diagnosed antenatally in 92% (151/164) and only postnatally in 8% (13/164) of the group (Table 1). Laser for TTTS (prior to post-laser TAPS) was performed using the Solomon technique in 37% (60/164) and the selective technique in 63% (104/164). The operating surgeon assumed that the laser for TTTS was complete in 81% (126/156) of post-laser TAPS cases, in 74% (43/58) of cases treated with the Solomon technique, and in 85% (83/98) of cases treated with the selective technique. Reasons for incomplete laser were: poor visibility (n = 11), fetal position (n = 7), placental position (n = 5), bleeding (n = 2), velamentous anastomoses (n = 1), vomiting of patient (n = 1), and not specified (n = 3). In half of the antenatally detected TAPS cases, TAPS developed within 14 days (IQR: 7–28; range 1–119) after laser treatment for TTTS. Figure 1 depicts time between laser for TTTS and the diagnosis of post-laser TAPS for all antenatally diagnosed cases. In 45% (73/161) of cases, TTTS donors became TAPS donors. Antenatal management included expectant management in 43% (62/151), IUT (± PET) in 29% (44/151), laser reintervention in 15% (24/151), selective feticide in 7% (11/151), delivery in 6% (9/151), and TOP in 1% (1/151) (Table 2).In total, 51% (84/164) of post-laser TAPS placentas were injected with color dye (Table 3). Of the 84 injected placentas, ten cases were treated with laser reintervention, and 74 cases were not treated with laser reintervention. In placentas not treated with laser reintervention, the median total number of anastomoses was one (IQR: 0–1), and 55% (40/74) had only AV anastomoses running in one direction. AA and VV anastomoses were detected in 8% (6/74) and 7% (5/74) of the group, respectively. In ten TAPS placentas, no residual anastomoses were found after color dye injection: five placentas belonged to cases with spontaneous resolution of TAPS after expectant management or treatment with IUT (±PET), and the other five belonged to cases with TAPS confirmed both antenatally and postnatally. Residual anastomoses were found in 40% (4/10) of post-laser TAPS cases treated with laser reintervention. Three cases with residual anastomoses had only one minuscule AV anastomosis, and the fourth had one single minuscule AA anastomosis. Out of the four cases with residual anastomoses, two cases had confirmed postnatal TAPS, one case had a double fetal demise after reintervention with laser, and one case showed an inter-twin hemoglobin difference of 8 g/dL (borderline TAPS) and had a low reticulocyte count ratio, but required a blood transfusion and partial exchange transfusion on day 1.Table 4 provides information on the perinatal outcome of post-laser TAPS twins, separated for TAPS donors and recipients. Median GA at birth was 31.7 weeks (IQR: 28.6–33.7, range 19.0–41.3). TAPS donors had significantly lower mean birth weights than TAPS recipients, 1346 g ± 525 g vs. 1533 g ± 588 g p < 0.001. Fetal demise occurred in 17% (56/327) of the group, either spontaneously in 10% (33/327) or intentionally in 7% (23/327). TAPS donors had a higher risk of fetal demise than TAPS recipients, both for spontaneous fetal demise (15% (25/164) vs. 5% (8/163); p < 0.001) and intended fetal demise (11% (18/164) vs, 3% (5/163); p = 0.007). The rate of neonatal mortality was 10% (27/271) and was higher in TAPS donors than in TAPS recipients, 15% (18/121) vs. 6% (9/150), respectively (p = 0.008). Overall, perinatal mortality (including intended demise) occurred in 25% (83/327) of TAPS twins, in 37% (61/164) of TAPS donors, and 14% (22/163) of TAPS recipients, respectively (p < 0.001). Severe neonatal morbidity was diagnosed in 40% (105/263) of liveborn twins and was similar for TAPS donors (43%; 51/118) and TAPS recipients (37%; 54/145), p = 0.568. Severe cerebral injury was identified in 11% (28/263) of liveborn twins, in 11% (13/118) of TAPS donors, and in 10% (15/145) of TAPS recipients (p = 0.916). Severe cerebral injury was diagnosed based on intraventricular hemorrhage ≥ Grade 3 (n = 15), cystic periventricular leukomalacia ≥ Grade 2 (n = 5), (post-hemorrhagic) ventricular dilatation (n = 6), porencephalic or parenchymal cysts (n = 3), arterial infarction (n = 1), and other severe lesions associated with adverse outcome (n = 1). Ischemic limb injury did not occur in this cohort of post-laser TAPS twins.In cases that were diagnosed with TAPS at birth (72%; 76/106), the median inter-twin hemoglobin difference was 12.6 g/dL (IQR: 10.3–15.1, range: 8.2–21.7), and the median reticulocyte count ratio was 2.7 (IQR: 2.3–3.9). At birth, 68% (52/76) of donors needed a blood transfusion to treat anemia, and 49% (37/76) of recipients needed a PET to treat polycythemia. In total, 33% (25/76) had Stage 1 postnatal TAPS, 34% (26/76) Stage 2, 20% (15/76) Stage 3, 11% (8/76) Stage 4, and 1% (1/76) Stage 5 postnatal TAPS.Univariate risk factor analysis showed that spontaneous perinatal mortality was significantly associated with Quintero Stage 3 (OR = 3.6, 95%CI 1.1–12.0, p = 0.039), antenatal TAPS Stage 4 (OR = 6.1, 95%CI 1.8–20.4, p = 0.003), TAPS donor status (OR = 3.7, 95%CI 2.2–6.3, p < 0.001), and GA at birth (OR = 0.8, 95%CI 0.7-1.0, p = 0.006). There was no strong correlation between the risk factors (Quintero stage–TAPS donor status (R < 0.001, p = 1.000); Quintero stage–antenatal TAPS stage (R = 0.078, p = 0.187); TAPS donor status–antenatal TAPS stage (R = −0.005, p = 0.931); Quintero stage–GA at birth (R = −0.022, p = 0.701); TAPS donor status–GA at birth (R < 0.001, p = 0.998); antenatal TAPS stage–GA at birth (R = −0.005, p = 0.931)), so all parameters were included in the multivariate analysis. In the multivariate analysis, antenatal TAPS Stage 4 (OR = 6.1, 95%CI 1.4–26.0, p = 0.015), TAPS donor status (OR = 4.2, 95%CI 2.1–8.3, p < 0.001), and GA at birth (OR = 0.8, 95%CI 0.7–0.9, p = 0.001) were identified as independent risk factors for spontaneous perinatal mortality. Univariate risk factor analysis showed that severe neonatal morbidity was significantly associated with GA at birth (OR = 1.5, 95%CI 1.3–1.8, p < 0.001) and Quintero Stage 2 (OR = 3.2 95%CI 1.1–9.7, p = 0.038). Both parameters were included in multivariate risk factor analysis as no correlation was found between the two (R = −0.022, p = 0.701). Multivariate risk factor analysis revealed that only GA at birth was an independent risk factor for severe neonatal morbidity (OR = 1.5, 95%CI 1.3–1.7, p < 0.001). More details on the risk analyses for perinatal mortality and severe neonatal morbidity are presented in Table A2 and Table A3 of Appendix B.This was the first large international study investigating management and outcome in post-laser TAPS twins. Our study showed that post-laser TAPS generally developed within one month after laser for TTTS, but could be detected up to 17 weeks after laser intervention. Management for post-laser TAPS was mostly expectant, but varied considerably, highlighting the lack of consensus for optimal treatment. In this cohort, perinatal outcome was poor, particularly due to high perinatal mortality rates in TAPS donors. This study provided important information for clinicians involved in the care for TTTS twins treated with laser surgery and might contribute to a better understanding of post-laser TAPS.This is the first study that gives a clear overview of the time of onset of post-laser TAPS and and shows that there is a wide range in timing of presentation. This variation might be attributed to two factors. First, reversal of the donor-recipient role could result in a slower development of post-laser TAPS. Possibly, former TTTS recipients that became TAPS donors may be protected against anemia for a longer period of time due to the excess of blood they received during TTTS. In contrast, former TTTS donors that became TAPS donors might suffer sooner from anemia due to their relatively hypovolemic state. Alternatively, TAPS might develop later in cases with compensating blood supply, allowed by VA, AA, or VV anastomoses. Our results showed that half of the post-laser TAPS cases presented within two weeks after laser surgery for TTTS. Although all cases in this study had signs of ongoing TAPS after the first week, spontaneous normalization of a large MCA-PSV discordancy after laser surgery has also been reported in previous literature [12,20]. Importantly, a large MCA-PSV difference shortly after laser might in some cases be the result of fetal hemodynamic re-equilibration after intervention, rather than the onset of post-laser TAPS due to the presence of a patent anastomosis. Consequently, intervening directly within 1-2 weeks after laser surgery in these cases could lead to unnecessary treatment since there is no ongoing transfusion. Therefore, close follow-up ultrasound examination to identify persistence or progression of a MCA-PSV discrepancy after the first weeks after laser surgery is recommended to confirm the diagnosis of post-laser TAPS and prevent unnecessary intervention and exposure to iatrogenic risks.Our study demonstrated that 81% of the surgeons initially thought that the laser for TTTS was complete. This “low index of suspicion” causes TAPS to often occur unexpectedly and shows that operator-reported completeness cannot be relied upon. Interestingly, approximately a third of post-laser TAPS twins were treated for TTTS with the Solomon technique. This illustrated that, although the Solomon technique has been proven to decrease the incidence of post-laser TAPS [2], clinicians should remain vigilant for the development of this complication even after a complete Solomon line was thought to be achieved. In agreement with the current recommendations of the twin guideline of the International Society of Ultrasound in Obstetrics and Gynecology [21], we strongly underline the importance of strict routine MCA-PSV Doppler follow-up examination in TTTS twins treated with laser surgery during the entire pregnancy to check for the presence of post-laser TAPS.We found high rates of perinatal mortality, particularly in the TAPS donor, reflecting the detrimental impact of fetal anemia on perinatal survival. Remarkably, TAPS donors only showed increased risk for perinatal mortality. After birth, donors and recipients had similar rates of severe neonatal morbidity, suggesting that neonatal health is more strongly related to the degree of prematurity than TAPS donor-recipient status. Importantly, this study demonstrated that GA at birth is a strong risk factor for both perinatal mortality and severe neonatal morbidity in post-laser TAPS twins. Compared to post-laser TAPS survivors that were previously investigated in the Solomon trial [22], we reported similar rates of severe neonatal morbidity (38% vs. 39%) and higher rates of perinatal mortality (18% vs. 26%). Of note, long-term outcome was not investigated in this study, and therefore, a difference in long-term neurodevelopmental impairment between donors and recipients cannot be precluded.Post-laser TAPS twins showed an overall more detrimental outcome than spontaneous TAPS twins [8]. The exact cause of this difference in outcome is not entirely clear, but it is likely to be multifactorial. The first and most obvious explanation is that post-laser TAPS twins have previously experienced TTTS, a severe condition itself [23]. Therefore, their fetal condition might already be compromised when they start developing TAPS. Given the fact that half of the post-laser TAPS cases occur within the first two weeks after laser, twins have had only limited time to recover, making them more prone to decompensation. A second explanation could be found in the angioarchitecture of placentas of post-laser TAPS twins. This study represents the biggest cohort of injected post-laser TAPS placentas and confirms previous findings that post-laser TAPS placentas often show only one or a few placental anastomoses [24]. Additionally, we found that most cases only had unidirectional AV anastomoses, without compensating flow from VA, AA, or VV anastomoses in the opposite direction. This might lead to an accelerated deterioration of TAPS, resulting more rapidly in abnormal Doppler blood flows, hydrops, and fetal death. Interestingly, a minority of post-laser TAPS placentas did not show residual anastomoses, in spite of the presence of confirmed postnatal TAPS. This could be explained by suboptimal color dye or by the presence of deep, hidden anastomoses [4,25]. Third, the choice of antenatal management might have also influenced the condition of post-laser TAPS twins. Our results showed that treatment for post-laser TAPS was diverse, but that the majority of the group was managed expectantly or received IUT (± PET); two treatment strategies that are not definitive in nature and allow the condition to progress. Possibly, laser was considered more challenging or not feasible in cases that already underwent laser, due to expected reduced visibility, membrane separation, iatrogenic monoamnionicity, PPROM, or because of the same reasons that caused the laser to be incomplete in the first place, such as the position of the placenta. Illustratively, we found a high rate (40%) of residual anastomoses in twins treated with laser reintervention. A detailed evaluation of differences in perinatal outcome between the various management strategies will be presented in a separate study [26].As with all registries, this study was fully dependent on local registrations of post-laser TAPS cases. In many countries, TTTS cases are sent back to the referring hospital after laser procedure, leaving care in less experienced hands. As the diagnosis of post-laser TAPS is only reached by adequate MCA-PSV Doppler examination, hemoglobin and reticulocyte measures, and placental-injection studies, it is likely that some post-laser TAPS cases have been missed. Nonetheless, this study represented the largest cohort of post-laser TAPS twins to date and was able to provide valuable insights into management and outcome in post-laser TAPS.To conclude, post-laser TAPS could occur at any time after laser for TTTS, is managed heterogeneously, and is associated with poor outcome, particularly in donor twins. Our findings necessitate further research into the best treatment option for TAPS. To investigate the best treatment for TAPS adequately, an international randomized controlled trial is needed.L.S.A.T., E.L., F.S., J.A., and D.O. were involved with the conceptualization and methodology of this study, J.A. designed the online database, L.S.A.T., E.L., S.F., M.L., J.S., Y.V., L.L., R.D., A.S.W., R.F., S.R.H., G.R., C.R., S.A., P.K., P.G., K.H., M.T.d.S., A.K., B.T., E.P.B., R.P., G.J.G., A.C., E.B., V.A.S., K.V.K., M.O.B., A.W., M.D.K., E.T., D.O., J.M.M., M.C.H., F.J.C.M.K., J.A. and F.S. provided data of eligible patients, L.S.A.T. performed data curation, L.S.A.T. and E.L. conducted formal analysis of the study data.; L.S.A.T. wrote the original draft of the manuscript; Reviewing and editing of the manuscript was done by L.S.A.T., E.L., S.F., L.L., A.W., G.R., C.R., P.K., M.T.d.S., A.K., E.P.B., G.J.G., A.C., K.V.K., M.O.B., M.D.K., E.T., J.A., and F.S.; F.S., and E.L. supervised the project; L.S.A.T., F.S., and E.L. were involved with the projects’ administration; All authors read and agreed to the published version of the manuscript. This research received no external funding.The authors declare no conflict of interest.Participating fetal therapy centers and their number of contributed post-laser TAPS cases.Univariate and multivariate risk analysis for spontaneous perinatal mortality in the post-laser twin anemia polycythemia sequence.Values are the odds ratios (OR) (95% confidence intervals (CI)), the standard error (SE), and the p-value. * Set to zero as the reference. † Twenty-three of 327 cases were excluded based on selective feticide or termination of pregnancy. ‡ Five cases had a missing Quintero stage value.§ In 3 cases, persistence of TTS-TAPS donor-role was unknown. ¶ SPSS was not able to calculate OR for groups in which the event (spontaneous mortality) did not occur; therefore, a correction for non-occurring events was applied, and with this correction, an unaffected child was changed into an affected child, for all groups. TAPS, twin anemia polycythemia sequence; GA, gestational age; TTTS, twin-twin transfusion syndrome; IUT, intrauterine transfusion; PET, partial exchange transfusion.Univariate and multivariate risk analysis for spontaneous perinatal mortality in the post-laser twin anemia polycythemia sequence.Values are the odds ratios (OR) (95% confidence intervals (CI)), the standard error (SE), and the p-value. * Set to zero as the reference. † Nine missing values (6 cases with unknown neonatal outcome and 3 neonates that died shortly after birth). SNM, severe neonatal morbidity; TAPS, twin anemia polycythemia sequence; GA, gestational age; IUT, intrauterine transfusion; PET, partial exchange transfusion.Weeks between fetoscopic laser surgery for twin-twin transfusion syndrome and diagnosis of post-laser twin anemia polycythemia sequence.Baseline characteristics for the post-laser twin anemia polycythemia sequence.Data are presented as n/N (%) or the median (IQR). † Other types of placental position included: partly anterior and partly posterior (n = 1) and partly fundal (n = 1) lateral left (n = 2) and lateral right (n = 2). ‡ Five missing values. § Three missing values. ¶ Eight missing values. TAPS, twin anemia polycythemia sequence; TTTS, twin-twin transfusion syndrome; Q, Quintero; GA, gestational age.Diagnosis- and management-related characteristics for the post-laser twin anemia polycythemia sequence.Data are the median (IQR) or n/N (%). Except for “female” and “cesarean”, all other parameters are presented for antenatally diagnosed post-laser TAPS cases only (N = 151).† In 10 cases, the presence of a starry-sky liver was not assessed. ‡ Twenty missing values. § Two missing values.TAPS, twin anemia polycythemia; GA, gestational age; IUT, intrauterine transfusion; PET, partial exchange transfusion.Characteristics for placentas not treated with laser reintervention for the post-laser twin anemia polycythemia sequence.Data are the median (IQR) or n/N (%). TAPS, twin anemia polycythemia; TTTS, twin-twin transfusion syndrome; AV, arterio-venous; VA, veno-arterial; AA, arterio-arterial; VV, veno-venous.Perinatal outcome in the post-laser twin anemia polycythemia sequence.Data are presented as the means ± SD, medians (IQR), or n/N (%). † One missing value (1 post-laser TAPS recipient with unknown perinatal outcome). ‡ Nine missing values (same as “†” plus 4 cases with unknown neonatal morbidity info and 4 cases that died shortly after birth). § Two missing values (same as “†” plus one case with missing birth weight). TAPS, twin anemia polycythemia sequence; GA, gestational age; bw, birth weight.

Med-MDPI/jcm_2/jcm-09-06-01760.txt

@@ -0,0 +1 @@
+Background: Thoracolumbar spine fractures in multiple-injured patients are a common injury pattern. The appropriate timing for the surgical stabilization of vertebral fractures is still controversial. The purpose of this study was to analyse the impact of the timing of spinal surgery in multiple-injured patients both in general and in respect to spinal injury severity. Methods: A retrospective analysis of multiple-injured patients with an associated spinal trauma within the thoracic or lumbar spine (injury severity score (ISS) >16, age >16 years) was performed from January 2012 to December 2016 in two Level I trauma centres. Demographic data, circumstances of the accident, and ISS, as well as time to spinal surgery were documented. The evaluated outcome parameters were length of stay in the intensive care unit (ICU) (iLOS) and length of stay (LOS) in the hospital, duration of mechanical ventilation, onset of sepsis, and multiple organ dysfunction syndrome (MODS), as well as mortality. Statistical analysis was performed using SPSS. Results: A total of 113 multiple-injured patients with spinal stabilization and a complete dataset were included in the study. Of these, 71 multiple-injured patients (63%) presented with an AOSpine A-type spinal injury, whereas 42 (37%) had an AOSpine B-/C-type spinal injury. Forty-nine multiple-injured patients (43.4%) were surgically treated for their spinal injury within 24 h after trauma, and showed a significantly reduced length of stay in the ICU (7.31 vs. 14.56 days; p < 0.001) and hospital stay (23.85 vs. 33.95 days; p = 0.048), as well as a significantly reduced prevalence of sepsis compared to those surgically treated later than 24 h (3 vs. 7; p = 0.023). These adverse effects were even more pronounced in the case where cutoffs were increased to either 72 h or 96 h. Independent risk factors for a delay in spinal surgery were a higher ISS (p = 0.036), a thoracic spine injury (p = 0.001), an AOSpine A-type spinal injury (p = 0.048), and an intact neurological status (p < 0.001). In multiple-injured patients with AOSpine A-type spinal injuries, an increased time to spinal surgery was only an independent risk factor for an increased LOS; however, in multiple-injured patients with B-/C-type spinal injuries, an increased time to spinal surgery was an independent risk factor for increased iLOS, LOS, and the development of sepsis. Conclusion: Our data support the concept of early spinal stabilization in multiple-injured patients with AOSpine B-/C-type injuries, especially of the thoracic spine. However, in multiple-injured patients with AOSpine A-type injuries, the beneficial impact of early spinal stabilization has been overemphasized in former studies, and the benefit should be weighed out against the risk of patients’ deterioration during early spinal stabilization.The patient with multiple injuries is a unique challenge for the trauma care team, even more so in association with spinal fractures. Traumatic spinal injuries in multiple-injured patients are less common in comparison to extremity injuries but present with an increasing incidence over the last decades [1,2]. Most of these injuries are located within the thoracic and lumbar spine and are accompanied by lung contusion and/or abdominal trauma [3,4]. Beside worse outcomes, a cost analysis showed that apart from injuries of the head, associated spinal lesions in multiple-injured patients result in higher hospitalisation costs and prolonged length of stay (LOS) [5].Controversy still arises concerning the optimal timing for spinal surgery in the multiple-injured patient. In the literature, the discussion has ultimately focused on the question of early versus late surgical treatment and its consequences on mortality and the patients’ outcome [6,7]. Recent register studies generally advocate that early spinal stabilization is beneficial for multiple-injured patients, mainly due to the increased possibility of early patient mobilisation in the ICU [8]; however, these register studies do not differentiate the wide variety of different spinal injuries. For example, although AOSpine A-type fractures are managed surgically in some countries to maintain sagittal spinal balance, they are—from a biomechanical viewpoint—stable enough to tolerate early mobilisation in the ICU without the risk of neurological compromise. Therefore, a general recommendation to perform early spinal stabilization, which is definitely a surgical burden and second hit in this scenario, may not be justified, and may even jeopardize multiple-injured patients with biomechanically stable spinal injuries.For this reason, the aim of the present study is to analyse the impact of the timing of spinal surgery in multiple-injured patients both in general and in respect to spinal injury severity.This retrospective two-centre study was conducted on all multiple-injured patients with spinal injuries who were admitted to one of the two level I trauma hospitals between January 2012 and December 2016. Patients were eligible for further analysis if they met the following inclusion criteria: age >16 years; injury severity score (ISS) >16; associated spinal injury within the thoracic or lumbar spine; and a fully available computed tomography scan for spinal injury classification, according to the established AOSpine Classification [9]. Demographic data, circumstances of the accident, overall injury severity score (ISS), and time to spinal surgery (24, 72, and 96 h) were documented. Measured outcome parameters were duration of mechanical ventilation, intensive care unit (ICU) length of stay (iLOS), length of hospital stay (LOS), development of sepsis or multiple organ dysfunction syndrome (MODS), and death. Organ failure was defined according to the sequential organ failure assessment (SOFA) score: a value in one or more organ systems of ≥3 points was defined as single or multiple organ failure. Sepsis was defined as systemic inflammatory response syndrome (SIRS) combined with bacteraemia.Data were collected from the inhouse clinical documentation system. Counts and frequencies were used to describe the sample. Correlation analysis (Spearman) was used to determine any dependence between the variables. For evaluation of influencing factors on the timing of surgery, a chi-square test and a one-way ANOVA was performed. Fisher’s exact test was used for categorical variables. Multivariable logistic regression analysis was performed using time to surgery, LOS, iLOS, and sepsis as dependent variables. Results are presented as odds ratios (ORs) with 95% confidence intervals. Significance was set at p < 0.05 for all statistical tests. A power analysis was carried out with G * Power (3.1.9.3, IBM) and resulted in the following statistical power for each statistical test: 0.998 for the ANOVA, 0.963 for the t-test, and 0.999 for both the correlation analysis and the regression analysis.All testing procedures were performed exploratively, so no adjustment for multiple testing has been made. Statistical analyses were performed with IBM SPSS software (version 23; IBM Corp., Armonk, New York, NY, USA). The study complies with the principles of the Declaration of Helsinki (2013) and was approved by the local ethical committee (#EK056/17).A total of 250 multiple-injured patients with vertebral fractures within the thoracolumbar spine were identified (Table 1). The population consisted of 72 (29%) female and 178 male (71%) patients with a mean age of 46 ± 19 years. The mean ISS was 24.8 ± 12.6 points. The most common injury pattern was a fall of >3 m (n = 102, 41%), followed by car accident (n = 39, 16%) and motorcycle accident (n = 38, 15%). In 153 patients (61%) the lumbar spine and in 97 patients (38%) the thoracic spine was involved. In 127 patients (51%) spinal stabilization was performed, of whom 14 patients had to be excluded due to missing data (Figure 1). Thus, 113 patients were included for further statistical analysis.Within the group of the operatively treated patients, 71 fractures (63%) were classified as AOSpine A-type injuries, whereas 42 fractures (37%) were classified as AOSpine B-/C-type injuries (Figure 2). Seventeen thoracic and 18 lumbar spinal injuries were associated with neurological impairment.Time to spinal surgery for the 113 multiple-injured patients included in the study is presented in Figure 3. Setting a cutoff of 24 h for spinal surgery, 49 patients (43.4%) were treated within 24 h after trauma. Twenty-one of these patients presented with an AOSpine B-/C-type injury. Increasing the cutoff to 72 h for spinal surgery, 75 patients (66.4%) were treated within 72 h after trauma, of whom 32 patients presented with an AOSpine B-/C-type injury. Further increasing the cutoff to 96 h, 81 patients (71.7%) were treated within 96 h after trauma, of whom 33 patients presented with an AOSpine B-/C-type injury. Regression analysis identified a higher ISS, the thoracic spine injury location, A-type injuries, and the absence of an evident neurological deficit as independent risk factors for a delay in spinal surgery (Table 2).Comparison of multiple-injured patients surgically treated for their spinal injury within 24 h as compared to those surgically treated later than 24 h is shown in Table 3. Multiple-injured patients with their spinal injuries surgically treated within 24 h showed a significantly reduced length of ICU stay by 7 days (7.31 days vs. 14.56 days; p < 0.001) as compared to those were operated on later than 24 h while having a comparable overall injury severity (ISS: 23.69 vs. 24.80; p = 0.672). Furthermore, the length of hospital stay was significantly reduced by 10 days (23.85 days vs. 33.95 days; p = 0.048). The prevalence of sepsis was significantly higher in multiple-injured patients surgically treated for their spinal injury later than 24 h (9 vs. 1, respectively; p = 0.023). Mechanical ventilation time was higher in patients surgically treated for their spinal injury later than 24 h, although this did not reach statistical significance (150.94 h vs. 184.05 h, respectively; p = 0.525). The prevalence of MODS or death showed no statistical difference between both groups.Comparison of multiple-injured patients surgically treated for their spinal injury within 72 h as compared to those surgically treated later than 72 h is shown in Table 4. Multiple-injured patients with their spinal injuries surgically addressed later than 72 h showed a significant and twofold longer stay in the ICU (17.5 days vs. 8.46 days, respectively; p < 0.001) and in the hospital (44.34 days vs. 22.49 days, respectively; p < 0.001). Mechanical ventilation time was higher in patients surgically treated for their spinal injury later than 72 h, although this did not reach statistical significance (229.94 h vs. 141.56 h, respectively; p = 0.114). However, patients surgically treated for their spinal injury later than 72 h had a significantly higher ISS as compared to those patients treated within 72 h (28.05 vs. 22.49, respectively; p = 0.03). The prevalence of sepsis, MODS, or death showed no statistical difference between both groups.Comparison of multiple-injured patients surgically treated for their spinal injury within 96 h as compared to those surgically treated later than 96 h is shown in Table 5. Multiple-injured patients for whom surgical treatment of their spinal injuries was postponed for more than 96 h showed a significant and twofold longer stay in the ICU (8.55 days vs. 18.41 days respectively; p < 0.001) and in the hospital (23.01 days vs. 45.84 days, respectively; p < 0.001). Although the overall injury severity was higher in the group surgically treated for their spinal injuries after 96 h, it did not reach statistical significance (27.72 vs. 22.94; p = 0.075). Furthermore, mechanical ventilation time was significantly and almost twofold increased from 137.24 h to 250.81 h (p = 0.046). The prevalence of sepsis was also significantly higher in the group surgically treated for their spinal injuries later than 96 h (7 vs. 3, respectively; p < 0.005). However, the prevalence of MODS or death was not higher in the group of multiple-injured patients treated for their spinal injuries later than 96 h.Correlation analysis revealed that there was a significant correlation between time to spinal surgery and outcome parameters (Table 6). Lengths of ICU and hospital stays showed a significant correlation with increasing the time to spinal surgery. The development of sepsis further showed a significant correlation with increasing time to spinal surgery. However, there was no correlation between an increased time to spinal surgery and the development of MODS or death. Regression analysis revealed that increased time to spinal surgery is an independent risk factor for an increased ICU and hospital stay, and spinal stabilization later than 72 h an independent risk factor for the development of sepsis (Table 7).Subgroup analysis showed that the adverse effect of delayed spinal stabilization is mainly attributable to multiple-injured patients with AOSpine B-/C-type injuries. Regression analysis revealed that in patients with AOSpine A-type spinal injuries, an increased time to spinal surgery was only an independent risk factor for an increased LOS; however, in multiple-injured patients with B-/C-type spinal injuries, an increased time to spinal surgery is an independent risk factor for increased iLOS, LOS, and the development of sepsis (Table 8).Comparison of multiple-injured patients with AOSpine A-type injuries compared to those multiple-injured patients with AOSpine B-/C-type injuries is shown in Table 9.Multiple-injured patients with A-type spinal injuries showed a trend towards a lower overall injury severity compared to multiple-injured patients with B-/C-type spinal injuries (ISS: 22.75 vs. 26.36, respectively; p = 0.138). Length of ICU stay was on average 4 days shorter, and time of mechanical ventilation roughly 90 h shorter in multiple-injured patients with A-type spinal injuries compared to those with B-/C-type spinal injuries; however, neither observations reached statistical significance (iLOS: 10.05 days vs. 14.60 days, respectively; p = 0.076; mechanical ventilation: 137.69 h vs. 227.85 h, respectively; p = 0.104). Time to spinal surgery was non-significantly prolonged in multiple-injured patients with A-type spinal injuries as compared to those multiple-injured patients with B-/C-type injuries (113.33 h vs. 70.90 h, respectively; p = 0.206). Multiple-injured patients with B-/C-type spinal injuries, compared to multiple-injured patients with A-type spinal injuries, showed a non-significant, higher prevalence of MODS (4 vs. 2, respectively; p = 0.076), sepsis (6 vs. 4, respectively; p = 0.071), and death (2 vs. 1, respectively; p = 0.310).Correlation analysis showed a significant correlation between higher spinal injury severity and thoracic spinal injury location; however, there was no correlation between spinal injury severity and any measured outcome parameters, and consequently, regression analysis identified that spinal injury severity was not an independent risk factor for adverse outcome.A comparison of multiple-injured patients with thoracic spinal injury location compared to those with lumbar spinal injury location is shown in Table 10.Multiple-injured patients with a thoracic spine injury showed a similar overall injury severity compared to multiple-injured patients with a lumbar spine injury (ISS: 25.04 vs. 24.19, respectively; p = 0.728). Length of ICU stay was significantly increased by an average of 5 days in multiple-injured patients with a thoracic spine injury compared to multiple-injured patients with a lumbar spine injury (14.58 days vs. 9.48 days, respectively; p = 0.032), whereas length of hospital stay showed no significant difference—although it was shorter in multiple-injured patients with a thoracic spine injury (25.92 days vs. 32.77 days, respectively; p = 0.180). Mechanical ventilation time was increased by an average of almost 90 h in multiple-injured patients with a thoracic spine injury compared to multiple-injured patients with a lumbar spine injury; however, this did not reach statistical significance (223.41 h vs. 136.92 h; p = 0.103). Time to spinal surgery was prolonged in multiple-injured patients with thoracic spine injury compared to multiple-injured patients with a lumbar spine surgery (119.24 h vs. 74.24 h; p = 0.152). The prevalence of MODS and death was comparable in both groups, whereas the prevalence of sepsis was significantly higher in multiple-injured patients with a thoracic spine injury compared to multiple-injured patients with a lumbar spine injury (8 vs. 2; p = 0.011).Correlation analysis revealed that spinal injury location significantly correlated with patients’ age, the prevalence of sepsis, the time to spinal surgery, and the spinal injury severity. Regression analysis identified spinal injury location to be an independent risk factor for an increased length of ICU stay (p = 0.04, Table 7).The clinical course for multiple-injured patients is determined by the initial trauma (first hit), the initial surgical burden (second hit), and the resulting systemic inflammatory response [10]. In contrast to the established damage control orthopedics (DCO) concept for long-bone fractures, which recommends performing a definite long-bone fracture stabilization later in the course of treatment, early definite spinal stabilization in multiple-injured patients is gaining increasing popularity in recent years [11,12]. This “spine damage control” concept advocates immediate posterior spinal stabilization in multiple-injured patients, with delayed anterior spinal stabilization, if required, later in the course [11,13]. Support of this concept is provided by several studies showing that early spinal stabilization in multiple-injured patients appears to be associated with a beneficial medical and socioeconomic outcome [5,8,14,15].In general, our data support the concept of early posterior spinal stabilization in multiple-injured patients. Our data show that increasing time to spinal surgery in multiple-injured patients is an independent risk factor for an increased ICU and hospital stay, and that spinal stabilization later than 72 h is an independent risk factor for the development of sepsis. We consider immediate spinal stabilization, as advocated above, as spinal surgery within 24 h after trauma. Applying a 24 h cutoff to our patient population, we are able to show that multiple-injured patients with their spinal stabilization later than 24 h had, on average, a significantly longer stay of 7 days in the ICU, and a significantly longer length of hospital stay of 10 days, while having a comparable overall injury severity, as defined by the ISS, compared to those multiple-injured patients with spinal surgery within 24 h. Furthermore, the duration of mechanical ventilation was increased by approximately 30 h, and the prevalence of sepsis was significantly increased in the case of spinal stabilization later than 24 h. These adverse effects were even more pronounced for 72 h or 96 h as time cutoffs for spinal surgery. Several studies have attributed the beneficial effect of early spinal stabilization mainly to a better patient handling in the ICU, which allows, for example, for prone positioning for the improvement of pulmonary function, or even earlier in- or out-of-bed mobilization [8]. Most of these studies, however, do not consider spinal injury severity from a biomechanical point of view. Thus, the above-mentioned reasons for a beneficial effect of early spinal stabilization may only be true for instable spinal injuries (AOSpine B-/C-type), because stable spinal injuries (AOSpine A-type) may be considered for early mobilization anyway.Therefore, the “spine damage control” concept has to be interpreted with caution and should not result in the recommendation that in multiple-injured patients, every spinal injury requiring surgery should be stabilized as early as possible at almost any cost. Further subgroup analyzation of our patient population supports a more individualized treatment concept, taking into account the degree of spinal injury stability, as described with the AOSpine classification. In multiple-injured patients with AOSpine B-/C-type spinal injuries, which from a biomechanical view are unstable spinal injuries, stabilization as early as possible appears to be beneficial. In these patients, prolonged time to spinal surgery is an independent risk factor for increased length of stay in the ICU and in hospital, as well as for the development of sepsis. In contrast to these findings, the benefit of an early spinal stabilization for multiple-injured patients with AOSpine A-type spinal injuries appears to be less pronounced. In these biomechanically stable spinal injuries, a prolonged time to spinal stabilization is only an independent risk factor for an increased length of hospital stay. Although we could not observe an adverse effect of the stabilization of AOSpine A-type spinal injuries in our multiple-injured patients, this observation reduces the pressure on the trauma team to deal with AOSpine A-type spinal injuries in the acute phase. Since AOSpine A-type injuries are the predominant spinal injury pattern in our multiple-injured patient population, this may have an important impact on trauma care in the future. Although a differentiation of the spinal injury, according to the AOSpine classification, is important for decisions with regard to surgical timing, the injury type in and of itself is not an independent risk factor for adverse outcomes.Thoracic spinal injury location has been identified by other studies to adversely influence patients’ outcome [16,17]. Our data reveals that thoracic spinal injury location is an independent risk factor for prolonged ICU stay. This may be explained by two considerations. First, thoracic spinal injury location in our population is associated with delayed spinal stabilization, probably due to a higher grade of required surgical expertise. Secondly, thoracic spinal injuries are probably associated with a higher rate of thoracic injury (for example, lung contusions). In our population, multiple-injured patients with thoracic spinal injuries on average required 90 h more of mechanical ventilation than those multiple-injured patients with lumbar spinal injuries.Our data show that in our patient population, almost half of the patients (47.8%) were surgically treated for their spinal injury within 24 h, and almost ¾ of the patients (71.7%) within 72 h. Within 24 h, 50% (21/42 patients) of the total AOSpine B-/C-type spinal injuries and 39% (28/71) of the total AOSpine A-type spinal injuries were surgically addressed; thus, the surgeons already recognized the higher surgical urgency of AOSpine B-/C-type injuries. Reasons for a delayed spinal stabilization appear to be multifactorial. Regression analysis identified, in particular, a higher ISS, thoracic spine injury location, A-type injuries, and the absence of an evident neurological deficit as independent risk factors for a delay in spinal surgery.In our population, 31% of the multiple-injured patients (n = 35) presented with neurological impairment due to their spinal trauma. In the literature, urgent surgical spinal treatment is widely accepted in the event of any neurological impairment focusing on isolated spine trauma. The main target is the decompression of the affected nerve structures and the restoration of the correct spinal alignment. The analysis of the STASCI (Surgical Timing in Acute Spinal Cord Injury) study showed that a general improvement of at least two ASIA (American Spinal Injury Association) grades is observable after early decompression within 24 h [18,19]. Our data show that even in multiple-injured patients with different priority surgeries, neurological impairment resulted in an earlier spinal surgery; however, we were not able to assess whether this had a beneficial impact on neurological outcome.There are some limitations of our study, which merit further comments. Endpoint selection may not be meticulous enough to identify further differences between the treatment groups, and no in-depth information concerning patients’ comorbidities were available for analysis. The clinical course and outcome of multiple-injured patients are influenced by several factors, which might not all be included in our analysis. In particular, spinal trauma with neurological compromise might have a severe impact on outcome parameters of these multiple-injured patients, which might not be valued enough in our data due to early out-of-hospital transfer to neurological rehabilitation units.Furthermore, the population size is limited as compared to register studies. However, this is the first study to consider spinal injury severity, according to the AOSpine classification, and to correlate this to the impact of surgical timing of spinal injuries. There are several existing fracture classification systems to evaluate the fracture pattern of spinal fractures. Advantages of the AOSpine fracture classification are better reliability and feasibility in clinical practice, in comparison to the thoracolumbar injury classification and severity score (TLICS). Several studies confirmed that the AOSpine classification is superior to the TLICS with regard to reliability for the identification of fracture morphology [20,21].In conclusion, our data support the concept of early spinal stabilization within 24 h in multiple-injured patients with AOSpine B-/C-type injuries, especially of the thoracic spine. However, in multiple-injured patients with AOSpine A-type injuries, the beneficial impact of early spinal stabilization has been overemphasized in former studies, and the benefit should be weighed against the risk of patients’ deterioration, due to the unneglectable surgical burden of spinal surgery.Conceptualization, P.K. (Philipp Kobbe) and C.H.; Data curation, P.K. (Patrick Krug), H.A., M.H., K.H., C.M., M.J.S., C.F. and G.S.; Formal analysis, H.A., K.H., M.J.S., C.F., G.S. and C.H.; Investigation, M.P., M.H., C.M., M.J.S. and C.H.; Methodology, P.K. (Patrick Krug) and M.P.; Project administration, P.K. (Philipp Kobbe); Supervision, P.K. (Philipp Kobbe), F.H. and C.H.; Validation, F.H. and C.H.; Writing—original draft, P.K. (Philipp Kobbe); Writing—review & editing, C.H. All authors have read and agreed to the published version of the manuscript.This research received no external funding.The authors declare no conflicts of interest.Flowsheet of the included population.Distribution of the fracture type within the thoracic and lumbar spine of the surgically treated patients.Distribution of the time to surgery related to three defined time points: 24, 72, and 96 h.Baseline demographics for spinal fractures in multiple-injured patients.ISS: injury severity score; AIS: abbreviated injury scale.Regression analysis for time to surgery (dependent variable: >24 h).Bold marked values indicate significant parameters. OR: odds ratio.Stratification analysis (ANOVA and chi-square test) for the time to surgery (<24 vs. >24 h).Significant p-values are marked in bold. LOS: length of stay in the hospital; ICU: intensive care unit; MODS: multiple organ dysfunction syndrome. Stratification analysis (ANOVA and chi square test) for the time to surgery (<72 vs. >72 h).Significant p-values are marked in bold.Stratification analysis (ANOVA and chi-square test) for the time to surgery (<96 vs. >96 h).Significant p-values are marked in bold.Correlation analysis (Spearman) for time to surgery.Significant p-values are marked in bold.Regression analysis with the dependent variables: LOS (days), length of stay in intensive care (iLOS; days), and sepsis.Bold marked values indicate significant parameters.Regression analysis for A and B-/C-type fractures with the dependent variables: LOS (days), iLOS (days), and sepsis.Bold marked values indicate significant parameters.Stratification analysis (ANOVA and chi-square test) of the fracture type (A-type vs. B-/C-type fractures).Stratification analysis (ANOVA and chi-square test) of the fracture localisation.Significant p-values are marked bold.

Med-MDPI/jcm_2/jcm-09-06-01761.txt

@@ -0,0 +1 @@
+Only scarce data pertaining to interleukin 8 (IL-8) and monocyte chemoattractant protein-1 (MCP-1) chemokines in human aneurysm can be found in the current literature. Therefore, the aim of this study was the evaluation of cerebrospinal fluid (CSF) and serum IL-8 and MCP-1 concentration in unruptured intracranial aneurysm (UIA) patients (n = 25) compared to the control group (n = 20). IL-8 and MCP-1 concentrations were measured with ELISA method. We demonstrated that CSF IL-8 concentration of UIA patients is significantly higher (p < 0.001) than that presented in the serum, which can indicate its local synthesis within central nervous system. CSF IL-8 concentration was also significantly related to aneurysm size, which may reflect the participation of IL-8 in the formation and development of brain aneurysms. IL-8 Quotient (CSF IL-8 divided by serum IL-8) in UIA patients was statistically higher compared to control individuals (p = 0.045). However, the diagnostic utility analysis did not equivocally indicate the diagnostic usefulness of the IL-8 Quotient evaluation in brain aneurysm patients. Nevertheless, this aspect requires further study.Intracranial cerebral aneurysms (IAs) occur in approximately 5% of the population [1,2]. Risk factors include: hypertension, atherosclerosis, smoking, alcohol consumption, female sex, and genetic predisposition [3,4,5,6,7]. Cerebral aneurysms are usually asymptomatic but associated with the risk of a life-threatening intracranial hemorrhage [8]. IAs are most often diagnosed only incidentally during magnetic resonance imaging (MRI) or computed tomography (CT) performed for a variety of reasons, like headaches or head injuries. However, it must be stressed, that a good portion of small aneurysms can escape visualization with standard MRI/CT, while still retaining potential for rupture and bleeding [9]. Unruptured intracranial aneurysms (UIAs) with a high estimated risk of rupture (i.e., of unfavorable location, size and structure) are secured by surgical clipping or by intravascular embolization. The UIAs with low estimated probability of rupture are often not qualified for obliteration because of the risk inherent to surgical intervention [10,11,12,13].At present, both primary diagnosis of IAs and follow up of those treated conservatively rely exclusively on their visualization. Nevertheless, the imaging either with MRI or with CT is hampered by high costs, limited availability, and a degree of biohazard related to irradiation and the use of a contrast agent [14]. Therefore, any method of screening for the IA’s in population and/or for assessing the dynamics of aneurysm growth would be of paramount importance. Unfortunately, such methods today are in short supply (if at all available), thus giving incentive for the prompt identification of biological markers of IA’s development.Chemokines are a heterogeneous group of soluble, short-acting proteins and peptides that are strong mediators of the inflammatory response [15,16]. Animal studies indicate two chemokines—C-X-C motif chemokine–ligand–8 (CXCL8/IL-8) and monocyte chemoattractant protein-1 (CCL2/MCP-1)—especially important in the formation and rupture of cerebral aneurysms [17,18,19]. IL-8 belongs to the CXC family of chemokines possessing the ELR motif (Glu-Leu-Arg), attracting mainly neutrophils [20,21]. It is secreted by various cell types—monocytes, macrophages, fibroblasts, epithelial cells—but also vascular endothelial cells [22,23]. MCP-1 is a potent chemoattractant for macrophages, T lymphocytes, NK cells and basophils [24,25]. It is expressed by the endothelial cells of the aneurysm under high wall shear stress, which promote macrophage infiltration at an early stage of brain aneurysm formation [17,26].In contrast to the animal models, there are only scarce data pertaining to aneurysms in humans [27,28]. It should be also noted that studies carried out on clinical material of aneurysms after subarachnoid hemorrhage (SAH) may result in rather confusing conclusions as to the problem of primary aneurysm formation and growth [29]. The hitherto published studies were based on sampling blood for the presence of the cytokines. A question needing answering is whether local chemokines’ concentration around the aneurysm is reliably represented by their concentration in blood. An answer can be brought forth only by evaluation of the concentrations in the cerebrospinal fluid (CSF) and in the peripheral blood.Taking into consideration the above, the aim of the study was the evaluation of CSF and serum IL-8 and MCP-1 concentration in patients with unruptured intracranial aneurysm. To exclude influence of the blood–brain barrier and the blood–CSF barrier on mutual relationships of protein levels in these two compartments [30,31,32], we also calculated IL-8 and MCP-1 Quotients by referring their CSF values to the serum values. In our study we tested if the evaluation of IL-8 and MCP-1 concentration would allow to diagnose brain aneurysm before its rupture. We also attempted to check if IL-8 and MCP-1 levels or their Quotients are related to aneurysm size, number and shape, as well as aneurysm risk factors (age, gender, smoking, hypertension, obesity).The study was conducted in agreement with the Helsinki-II-declaration and the protocol was approved by the Bioethics Human Research Committee of the Medical University of Bialystok (Permission No. R-I-002/383/2015). All subjects gave their informed consent for inclusion before they participated in the study.The study group consisted of 25 patients: five males/20 females, mean age 56 years, range 30–70 years. All of them were subjected to craniotomy and direct surgical clipping of unruptured intracranial aneurysm and were operated on by one surgeon (Z.M.). In all patients, their aneurysms were asymptomatic and located at the anterior part of the Willis circle. A decision regarding the need for intervention was undertaken after careful consideration of the risk of aneurysm rupture versus inherited risk of surgical procedure. The commonly accepted guidelines (e.g., size > 5 mm and/or irregular shape, representing increased risk of rupture [12] and the opinion of the patient) were taken into consideration when making the decision. The presence of UIA was established with CT angiography in 14 patients and with MR angiography in 11 patients. In eight individuals, a precise definition of the location, size and shape of the aneurysm required confirmation with contrast digital angiography.Exclusion criteria included: neurodegenerative conditions like multiple sclerosis, neuroinfection and brain tumor in medical history, surgery or major trauma in the previous months, anti-inflammatory, antibiotics, or corticosteroids administration in the previous months.The comparative group was composed of 20 subjects (six males/14 females, mean age 57 years, range 25–78 years), suffering from trigeminal neuralgia due to anatomical conflict between the trigeminal nerve and a cerebellar artery. All patients belonging to this group revealed refractory to conservative treatment and were qualified for posterior fossa craniotomy and microvascular decompression. The procedure involved exposition of the conflict at the cerebro-pontine angle, desinsertion of the compressing artery from the trigeminal nerve, displacement of the artery, and finally its fixation to the cerebellar tentorium with a sticking material (Tachosil®, Takeda, Linz, Austria). The criteria of exclusion from the control group were exactly the same as observed in the study group.All patients before surgery had fasting basic laboratory tests done between 6:00 and 7:00 a.m. Table 1 and Table 2 present the characteristics of the study and control subjects. Tobacco intake was defined when patient smokes more than 20 cigarettes a day for over 12 months and more. People with body mass index (BMI) > 30.0 (kg/m2) were classified as obese.CSF was obtained during craniotomy from the subarachnoid space of the brain. Surgical procedures were performed in a standard manner: under general anesthesia and with the patient’s head fixed in a three-pin Mayfield headholder. Skin incision preceded the lifting of the bone flap and lancing of dura mater, which allowed visualization of the arachnoid membrane and subarachnoid space. With the aid of an operating microscope, the subarachnoid space was carefully opened and inflowing CSF aspirated with single-use, sterile syringe, and soft venous catheter. Particular care was taken to prevent any contamination of the CSF with blood and with warm saline solution used as irrigation. Therefore, all the aforementioned steps were taken in the very beginning of each procedure, before any bleeding may occur.Patients’ blood samples (2.7 mL) were collected into test tubes without anticoagulant (S-Monovette, SARSTEDT, Nümbrecht, Germany). Within 0.5 h after the venipuncture, blood was centrifuged for 20 min at 1000× g to obtain serum. CSF samples were centrifuged for 20 min at 1000× g. Obtained serum and CSF supernatant were aliquoted and stored at −75 °C until further analysis.Cerebrospinal fluid (CSF) and serum IL-8 and MCP-1 concentrations were measured using enzyme-linked immunosorbent assay (ELISA) tests: ELISA Quantikine® Human CXCL-8/IL-8 Immunoassay kit (Catalogue number: D8000C) and ELISA Quantikine® Human CCL2/MCP-1 Immunoassay kit (Catalogue number: DCP00), both from R&D Systems Europe Ltd., Abingdon, England. Experiment was conducted in accordance with the manufacturer’s instruction. CSF and serum samples for IL-8 evaluation were not diluted prior to analysis. The manufacturer of the assay kit referred to the intra-assay coefficient of variation (CV%) as 5.6% at IL-8 mean concentration of 168 pg/mL, SD = 9.4 pg/mL. CSF and serum samples for MCP-1 evaluation were diluted 2-fold prior analysis. The manufacturer of the assay kit referred to the intra-assay coefficient of variation (CV%) as 7.8% at MCP-1 mean concentration of 76.7 pg/mL, SD = 6.0 pg/mL.To calculate IL-8 and MCP-1 Quotients, we divided CSF chemokine result by its concentration in serum, as described elsewhere [30,31,32]. Such an approach allowed to exclude possible influence of the blood–CSF barrier impairment and/or blood–brain barrier functions on IL-8 and MCP-1 levels.The obtained results were analyzed with the use of the STATISTICA 12.0 PL software (StatSoft Inc., Tulsa, OK, USA). The distribution of the studied concentrations did not follow the normal distribution (X2-test), thus nonparametric statistical analyses were applied. The Mann–Whitney test was used in order to compare two independent samples. The X2-test with Yates correction was used to investigate whether patients were gender-matched and whether there were differences regarding risk factors present between them. Correlation coefficients were obtained by applying Spearman’s rank correlation. Values for continuous variables are presented as median with the 25th and 75th percentiles. For the variables CSF and serum chemokines concentration, chemokines Quotients, age, sex, and brain aneurysm risk factors, a logistic regression model was sought. The same covariables were used in the linear regression analysis of predictor variables influencing aneurysm size as the ones in the logistic regression analysis. Differences were considered statistically significant at p < 0.05. Receiver operator characteristic (ROC) curve was generated in order to determine the performance of cytokine evaluation in a task of discrimination between the study and control groups. The Youden index, a function of sensitivity and specificity, indicated an optimal trade-off between these two (cut-off point) for the parameters tested.CSF IL-8 concentration was higher, while serum IL-8 concentration lower in UIA patients compared to the control group, but differences were not significant (see Supplementary Materials: Table S1) In UIA patients, CSF IL-8 median concentration was 3-fold higher compared to median serum concentration (p < 0.001). In the control group, CSF IL-8 was 1.6-fold higher compared to median serum concentration, but the obtained difference was not significant (p = 0.212) (Figure 1).IL-8 Quotient in UIA patients was statistically higher compared to control individuals (Figure 2, see Supplementary Materials: Table S1). The area under the ROC curve (AUC) for IL-8 Quotient was statistically higher than the value of 0.5 (Figure 3, Table 3).The CSF MCP-1 concentration was similar, while serum MCP-1 concentration was somewhat lower in UIA group compared to the control individuals, but both differences were not statistically significant (see Supplementary Materials: Table S1). In UIA patients as well as in the control group, CSF MCP-1 median concentration was statistically higher (1.8-fold, 1.6-fold, respectively) compared to median serum concentration (p < 0.001, p = 0.011, respectively) (Figure 4). The MCP-1 Quotient showed a tendency to be higher in UIA patients compared to control group, but again, the difference was not significant (Figure 5, see Supplementary Materials: Table S1).For the variables CSF IL-8 concentration, serum IL-8 concentration, IL-8 Quotient, CSF MCP-1 concentration, serum MCP-1 concentration, MCP-1 Quotient, age, sex, and brain aneurysm risk factors (obesity, systolic blood pressure, diastolic blood pressure, and smoking) a multivariate logistic regression model was sought, but nothing of significance was discovered. Only a univariate linear regression model was obtained. We showed that if the IL-8 Quotient increases by 1, the chance of having unruptured brain aneurysm increases by 1.84 times (increases by 84%) (Table 4).In UIA individuals, correlation coefficient analysis revealed: positive correlation between CSF IL-8 concentration and aneurysm size (R = 0.41, p = 0.04), and positive correlation between both CSF and serum MCP-1 concentrations and the aneurysms number (R = 0.43, p = 0.032 and R = 0.40, p = 0.049, respectively).Logarithmic transformation of aneurysm’s size was necessary for the linear regression model assumptions to be met. The same covariables were used as the ones in the logistic regression analysis. Univariate linear regression analysis revealed that: (1) with an increase in IL-8 concentration in CSF by 10 pg/mL, the aneurysm size increases by 1.14 times (rises by 14%); (2) with an increase in BMI by 1, the aneurysm size increases by 1.035 times (rises by 3.5%) (Table 5).In the model of multivariate linear regression analysis, predictor variables influencing aneurysm size included: CSF IL-8 concentration and BMI. Adjusted R square (R2) for the created model equals 0.39, which indicates that this model explains 39% of the variance in dependent variable. Multivariate linear regression analysis results for brain aneurysm size revealed that: (1) with an increase in IL-8 concentration in CSF by 10 pg/mL, the aneurysm size increases by 1.15 times (rises by 15%); (2) with an increase in BMI by 1, the aneurysm size increases by 1.03 times (rises by 3%) (Table 5).Patients with aneurysm size ≥ median value (5.4 mm) had statistically higher CSF IL-8 levels compared to those with aneurysm size < median value (Mann–Whitney test, p < 0.05). Interestingly, patients with multiple brain aneurysm had higher CSF MCP-1 levels compared to individuals with single aneurysm (Mann–Whitney test, p < 0.05, Table 6).To indicate the optimal cut-off points of chemokines levels to predict the aneurysm size (<5.4 mm versus ≥5.4 mm) and to predict the number of aneurysms (single versus multiple) we performed a ROC curve analysis in the next step. The optimal cut-off point for CSF IL-8 concentration to predict a larger brain aneurysm size (≥5.4 mm) was 36.9 pg/mL (Table 7, Figure 6). The optimal cut-off point for CSF MCP-1 concentration to predict a number of aneurysms was 489.8 pg/mL (Table 8, Figure 7).As to the “risk factors”, our results suggest that patients with arterial hypertension had significantly lower CSF MCP-1 concentration compared to individuals without hypertension (Mann–Whitney test, p < 0.01, see Supplementary Materials: Table S2).Our study is the first to demonstrate that IL-8 concentration in the cerebrospinal fluid of patients with unruptured intracranial aneurysm is significantly higher than this present in serum (almost 3-fold). In the control group, we did not observe significant differences between these two compartments. These results indicate a local IL-8 synthesis in brain aneurysm patients [33,34] and may reflect the participation of IL-8 in formation and development of brain aneurysm. Especially, we found a correlation between CSF concentration of IL-8 and aneurysm size. This notion is strengthened by multivariate linear regression analysis results indicating that aneurysm size increases with an increase in IL-8 concentration in CSF.Moreover, we found a weak but positive correlation between MCP-1 concentration in the CSF and the number of aneurysms. Our research shows that the assessment of these chemokines concentration in CSF and serum may be also useful in stratifying the risk of a “dangerous” brain aneurysm presence.Simultaneous evaluation of the chemokines both in CSF (which abuts the aneurysm), and serum is the advantage of our study. The calculation of protein Quotient eliminates the influence of biological factors like age, gender, infection, which can change the content of blood cytokines. Additionally, such approach minimizes the impact of methodological factors because chemokines are measured in one analytical process [33,34].We are the first to calculate Quotients for the IL-8 and MCP-1 chemokines in patients with unruptured brain aneurysm. We observed that the Quotient for IL-8 was significantly higher in patients with an unruptured intracranial aneurysm compared to those without vascular lesions in the brain. It is also worth mentioning that IL-8 Quotient tended to increase in patients with larger, polycyclic aneurysms and in patients over 60 years of age, smokers and obese. Univariate logistic regression analysis showed that if the IL-8 Quotient increases by 1, the chance of having unruptured brain aneurysm increases by 84%. Unfortunately, we did not obtain a significant multivariate logistic regression model for IL-8 Quotient.To assess diagnostic utility of the IL-8 Quotient calculation, we constructed an ROC curve for ex post discrimination between our patients with and without cerebral aneurysm. The area under the curve (amounting to 0.72) corresponds to a moderate overall diagnostic accuracy of the test [35]. The positive predictive value (PPV) and negative predictive value (NPV) of IL-8 Quotient equaled 86% and 50%, respectively. Thus, the NPV of the IL-8 Quotient is not satisfactory enough. However, the PPV and NPV are not intrinsic to the test, but are affected by the prevalence of the disease [36]. To conclude, the results obtained do not equivocally indicate the diagnostic usefulness of the IL-8 Quotient evaluation in the diagnosis of patients with unruptured brain aneurysm.An intriguing question pertains to a possible mechanism by which cytokines leak to CSF. Obviously, our results gave no unequivocal basis for concluding on this matter, whereas the number of sources in the literature is scarce. Chalouhi et al. [27] evaluated a number of different chemokines and interleukins in blood taken directly from the lumen of unruptured cerebral aneurysm compared to concentration in blood taken from the femoral vessel. The authors observed a significantly higher concentration of selected cytokines, among them IL-8 and MCP-1. As to the source of the cytokines, they suggest that IL-8 and MCP-1 may be released by endothelial cells and by leukocytes (macrophages) infiltrating the foci of endothelial damage [27]. The passive diffusion from the blood seems not to be very probable, considering the relatively large size of these molecules (8–10 kDa) [37]. On the other hand, this cannot be excluded, taking into account the abnormal structure of the aneurismal wall [38].However, some presumptive evidence indicates also the astroglia and microglia as a local source of chemokines present in the CSF. Both astrocytes and microglia were shown to release IL-8 and MCP-1 when triggered by immune activation, stress and nociceptive input [39,40]. Similar to the brain glia, pericytes can also respond to a range of immunogenic stimuli to induce pro-inflammatory molecules, including tested chemokines [41]. Therefore, the presence of higher levels of the chemokines in the CSF can be alternatively regarded as evidence of an adaptive cerebral immunological reaction in response to ongoing inflammation at a site of aneurysm. At present, this is, however, only a hypothesis, obviously worth of future exploration.Chalouhi et al. [27] found increased concentration of IL-8 and MCP-1 in blood taken directly from the lumen of cerebral aneurysm [27]. Zhang et al. [28] revealed a significantly higher concentration of MCP-1 chemokine in the blood (plasma) of patients with brain aneurysm compared to healthy people as well as higher MCP-1 levels in patients with multiple aneurysms than in patients with a single aneurysm. The last observation is indirectly consistent with our results, because we also observed a positive correlation between serum MCP-1 and the number of aneurysms. In addition to that, we also found this correlation in the cerebrospinal fluid.Though general conclusions from the work of Chalouhi et al. and Zhang et al. are rather convergent with ours, some differences in the set-up must also be indicated. First, the two groups of authors conducted their study in mixed populations of patients, containing both ruptured and unruptured aneurysms. This seems important to indicate, as the SAH is a separate pathological entity with serious consequences, also including an aspect of cytokine release, such as cerebral vasospasm [29,42,43]. Admittedly, both groups of authors have separated a subgroup of unruptured aneurysms, but after this maneuver the resulting number of their material became smaller than ours (15 and 10 patients, respectively, in relation to our 25 patients). Other differences include the lack of vertebrobasilar aneurysms in our material and somewhat smaller average size of our aneurisms (5.4 versus 7 mm, median value) compared to the study of Chalouhi et al. Nevertheless, demonstrating the presence of the chemokines as also accompanying smaller aneurysms seems to strengthen the hypothesis of their role in brain aneurysm formation and development.Chalouhi et al. and Zhang et al. analyzed plasma IL-8 and MCP-1 levels by means of the multiplex method; in our study, the concentration of CSF and serum chemokines was analyzed with the use of the ELISA method. Both the ELISA and multiplex technique use an immobilized antibody to capture the soluble ligand and then detect the captured ligand by a second “reporter” antibody. However, there are significant differences between these two techniques, which include a reporter system, solid phase, or suspension in which the reaction occurs. Determination of the concentration of many ligands simultaneously in the same sample can lead to a cross-reaction (often called “matrix effect”), while ELISA only tests one analyte at a time, thus avoiding such a situation. Moreover, artifacts related to the analysis of many analytes simultaneously can result from: antibodies that can cross-react with other proteins, interspecies antibodies and the presence of other interfering substances [44]. The differences in the level of detection of the multiplex assays probably may translate to discrepancies, e.g., Zhang et al. did not observe significant difference for IL-8.The average size of aneurysms was in our patients was smaller than those in the hitherto published studies. It must be also admitted that patients belonging to our “control group” were not entirely healthy people, as they suffered from persistent facial pain caused by trigeminal nerve irritation. Nevertheless, in spite of these limitations, we still were able to demonstrate statistically significant differences.CSF samples were not obtained via the lumbar puncture but during craniotomy, so it is not clear if these results are reproducible with CSF lumbar puncture samples. This aspect could also be recognized as a study limitation. However, both unruptured intracranial aneurysm as well as trigeminal neuralgia are not indications for CSF collection for diagnostics purposes, thus the local Bioethics Committee did not give permission for CSF collection via the lumbar puncture. The available literature indicates that the total protein concentration in CSF obtained by lumbar puncture is higher than in CSF obtained from lateral ventricles of the brain. In the case of specific proteins, it is difficult to predict the concentration difference between different space of CSF collection [45,46,47]. Therefore, we suggest that for the interpretation of chemokines concentrations in the CSF, the source of the sample is of crucial importance.Finally, the setup of our study does not enable us to address the problem of the cellular sources of chemokines elevation in CSF and these specific mechanisms still need to be determined.We demonstrated that IL-8 concentration in the cerebrospinal fluid of patients with unruptured intracranial aneurysm is significantly higher than that presented in serum, which can indicate its local synthesis within the central nervous system. CSF IL-8 concentration was also significantly related to aneurysm size, which may reflect the participation of IL-8 in the formation and development of brain aneurysm. We also observed that the Quotient for IL-8 was significantly higher in patients with unruptured intracranial aneurysm compared to those without vascular lesions in the brain. However, further statistical analysis did not clearly indicate the diagnostic usefulness of the IL-8 Quotient evaluation in brain aneurysm patients. Nevertheless, this aspect requires further study.The following are available online at https://www.mdpi.com/2077-0383/9/6/1761/s1, Table S1. IL-8 and MCP-1 cerebrospinal fluid (CSF) and serum concentrations and their Quotients in unruptured intracranial aneurysm (UIA) patients compared to control group. Table S2. CSF and serum IL-8 and MCP-1 concentrations and Quotient values depending on aneurysm risk factors: median age value, gender, smoking, hypertension and obesity.Conceptualization, J.K. and O.M.K.-L.; Data Curation, J.K., T.L., R.C., K.S., Z.M. and O.M.K.-L.; Formal Analysis, J.K. and A.J.M.; Investigation, J.K., M.T., J.Z., V.D.-P. and O.M.K.-L.; Methodology, J.K., T.L., R.C., K.S., Z.M. and O.M.K.-L.; Supervision, J.M.-K., V.D.-P. and Z.M.; Visualization, J.K.; Writing—Original Draft Preparation, J.K., Z.M. and O.M.K.-L.; Writing—Review & Editing, J.K., T.L., R.C., K.S., A.J.M., M.T., J.Z., J.M.-K.; V.D.-P., Z.M. and O.M.K.-L.; All authors have read and agreed to the published version of the manuscript.This research received no external funding.We are grateful to Martin Lenkiewicz for his language assistance.The authors declare no conflict of interest.IL-8 cerebrospinal fluid (CSF) and serum concentrations in unruptured intracranial aneurysm (UIA) patients compared to control group (C).IL-8 Quotient in unruptured intracranial aneurysm (UIA) patients compared to control group (C).IL-8 Quotient receiver operator characteristic (ROC) curve for differentiating unruptured intracranial aneurysm patients from individuals without brain aneurysm.MCP-1 cerebrospinal fluid (CSF) and serum concentrations in unruptured intracranial aneurysm (UIA) patients compared to control group (C).MCP-1 Quotient in unruptured intracranial aneurysm (UIA) patients compared to control group (C).CSF IL-8 receiver operator characteristic (ROC) curve for predicting a larger brain aneurysm size (≥5.4 mm).CSF MCP-1 receiver operator characteristic (ROC) curve for predicting a number of aneurysms.Characteristics of unruptured intracranial aneurysm (UIA) patients and control individuals. Results are presented as median with 25th and 75th percentiles. The p-value of < 0.05 is considered statistically significant.APTT: Activated Partial Thromboplastin, HCT: hematocrit, HGB: hemoglobin, INR: International Normalized Ratio, Time, K+: potassium, MPV: mean platelet volume, n: number of individuals, Na+: sodium, P-LCR: platelet-large cell ratio, PLT: platelet count, PT: prothrombin time, RBC: red blood cell count, SAH: subarachnoid hemorrhage, UIAs: unruptured intracranial aneurysms, WBC: white blood cell count. Conversion factors to SI units are as follows: for WBC: 1.0, for RBC: 1.0, for HGB: 10.0, for PLT: 1.0, for glucose: 0.0555, for creatinine: 88.4.Aneurysm geometry characteristics.# The aneurysm’s size is present for the biggest one. MCA: middle cerebral artery, ICA: internal carotid artery, AcomA: anterior communicating artery, L: left, R: right.Diagnostic parameters of IL-8 Quotient in differentiating unruptured intracranial aneurysm patients from subjects without intracranial aneurysm.Cut-off (based on the highest Youden index); AUC: area under the ROC curve, SE: Standard Error, PPV: positive predictive value, NPV: negative predictive value, ACC: diagnostic accuracy, Se: diagnostic sensitivity, Sp: diagnostic specificity.Univariate logistic regression analysis results for UIA diagnosis.OR: odds ratio; CI: confidence interval.Univariate and multivariate linear regression analysis results for logarithm of the brain aneurysm size.β: model coefficient; CI: confidence interval. 3.6. IL-8 and MCP-1 versus aneurysm geometry and risk factors of aneurysm formation.CSF and serum IL-8 and MCP-1 concentrations and Quotient values depending on aneurysm’s size median value and aneurysm number and shape.The optimal cut-off point of CSF IL-8 to predict a larger brain aneurysm size (≥5.4 mm).Cut-off (based on the highest Youden index), AUC: area under the ROC curve, SE: Standard Error, PPV: positive predictive value, NPV: negative predictive value, ACC: diagnostic accuracy, Se: diagnostic sensitivity, Sp: diagnostic specificity.The optimal cut-off point of CSF MCP-1 to predict the number of aneurysms.Cut-off (based on the highest Youden index), AUC: area under the ROC curve, SE: Standard Error, PPV: positive predictive value, NPV: negative predictive value, ACC: diagnostic accuracy, Se: diagnostic sensitivity, Sp: diagnostic specificity.

Med-MDPI/jcm_2/jcm-09-06-01762.txt

@@ -0,0 +1 @@
+Contributed equally.Epidermal growth factor receptor tyrosine kinase inhibitors (EGFR-TKIs) are used for treating EGFR-mutated lung cancer, and osimertinib is effective in cases that acquired T790M mutations after treatment with the first- and second-generation EGFR-TKIs. However, no study has evaluated its safety and efficacy in older patients. This phase II trial (jRCTs071180002) evaluated osimertinib in T790M mutation-positive Japanese patients who were ≥75 years old and had experienced relapse or progression after previous EGFR-TKI treatment. Our previous report that enrolled 36 patients showed the overall response rate (58.3%) and disease control rate (97.2%), while this report describes the results for the progression-free survival (PFS), overall survival (OS), and safety analyses. The median PFS was 11.9 months (95% confidence interval (CI): 7.9–17.5), and the median OS was 22.0 months (95% CI: 16.0 months–not reached). The most frequent adverse events were anemia/hypoalbuminemia (27 patients, 75.0%), thrombocytopenia (21 patients, 58.3%), and paronychia/anorexia/diarrhea/neutropenia (15 patients, 41.7%). Pneumonitis was observed in four patients (11.1%), including two patients (5.6%) with Grade 3–4 pneumonitis. These results suggest that osimertinib was relatively safe and effective for non-small cell lung cancer that acquired T790M mutations after previous EGFR-TKI treatment, even among patients who were ≥75 years old.Treatment for epidermal growth factor receptor (EGFR)-mutated non-small cell lung cancer (NSCLC) typically involves EGFR tyrosine kinase inhibitors (EGFR-TKIs). Gefitinib and erlotinib are the first-generation EGFR-TKIs that provide significant survival benefits compared with platinum-based chemotherapy in clinical trials [1,2,3,4,5,6]. Afatinib and dacomitinib are the second-generation EGFR-TKIs that provide significantly longer progression-free survival (PFS) compared to that of platinum-based chemotherapy and first-generation EGFR-TKIs, although the second-generation EGFR-TKIs did not significantly improve overall survival (OS) [7,8,9,10,11]. In addition, these drugs are associated with more severe toxicity profiles, such as skin disorders, relative to the first-generation EGFR-TKIs.Various mechanisms are responsible for resistance to the first-generation and second-generation EGFR-TKIs, with more than one-half of the cases involving the EGFR exon 20 T790M mutation [12]. Osimertinib is a third-generation EGFR-TKI that was developed to address this issue [12], and the AURA3 study revealed that it provided significantly longer PFS compared to platinum-based chemotherapy among patients with T790M-mutated lung cancer [13]. Moreover, the FLAURA trial conducted on first-line treatment revealed that osimertinib administered as an initial treatment for EGFR-mutated cases significantly prolonged PFS and OS compared with the first-generation EGFR-TKIs, with a median OS of >3 years [14,15]. Furthermore, osimertinib is expected to have good central nervous system translocation and a limited inhibition of the wild-type EGFR, which may make it less toxic, and therefore, the first choice for EGFR-mutated NSCLC [16,17,18]. Nevertheless, additional evidence is needed to support this application based on various patient populations. We have performed a phase II study to investigate the efficacy and safety of osimertinib in elderly Japanese patients (≥75 years old) with NSCLC containing the T790M mutation who progressed or experienced a relapse while receiving the first- and second- generations of EGFR-TKI treatment. In our previous report, the response rate was the primary endpoint, and the disease control rate was the secondary endpoint [19]. This report presents the results from our final analyses of PFS, OS, and safety events, which were the additional secondary endpoints in that trial.The study eligibility and exclusion criteria have been previously reported [19,20]. Patients were enrolled in this study between July 2016 and May 2018 if they met the following eligibility criteria: recurrence of NSCLC after achieving stable disease or better as their best overall response after treatment with the first- and second-generation of EGFR-TKIs; harboring an EGFR mutation (activating) and being T790M-positive; aged over 75 years; performance status of ≤1 based on the Eastern Cooperative Oncology Group (ECOG) scale; adequate bone marrow function (leukocyte count 3000–12,000/µL, platelet count ≥100,000/µL, and hemoglobin level ≥9.0 g/dL), adequate hepatic function (bilirubin level ≤1.5 mg/dL, aspartate aminotransferase of ≤100 IU/L, alanine aminotransferase of ≤100 IU/L), and adequate renal function (serum creatinine ≤2.0 mg/dL); a measurable lesion according to the Response Evaluation Criteria in Solid Tumors (RECIST) guidelines version 1.1; and provision of written informed consent. The exclusion criteria were pulmonary disorders; including idiopathic pulmonary fibrosis; interstitial pneumonia; pneumoconiosis; active radiation pneumonitis and drug-induced pneumonia, active infection; symptomatic brain metastasis; uncontrollable diabetes mellitus or severe comorbidities such as heart disease or renal disease; watery diarrhea; active concomitant malignancy; pregnancy or other medical problems that could prevent compliance with the protocol. The trial protocol was registered at Japan Registry of Clinical Trials (jRCTs071180002) and was approved by the ethical review board of Clinical Research Network Fukuoka Certified Review Board (CRB7180004). All patients provided written informed consent before enrollment.This single-arm-multicenter study involved daily oral administration of osimertinib (80 mg/day). Osimertinib had to be started at 80 mg/day, and if adverse events (AEs) occurred, dose reduction was performed according to the dose reduction criteria. Administration of osimertinib was continued until the patient met the discontinuation criteria or disease progression. Tumor assessments were performed at baseline, every 6 weeks (± 2 weeks) for 6 months, and then every 9 weeks (± 2 weeks) until disease progression. Baseline brain imaging was performed on a similar schedule. Among patients with T790M mutations, the objective response rate (ORR) was 62% (95% confidence interval [CI]: 54–68) in the AURA extension study (201 patients). In the AURA2 study (210 patients) the ORR was 70% (95% CI: 64–77) and the median PFS was 9.9 months (95% CI: 9.5–12.3) [21,22,23]. Docetaxel is the standard treatment for elderly patients based on the Japanese guidelines, as it provided an ORR of 22.7% in a study that compared docetaxel to vinorelbine [24]. Another recent study evaluated carboplatin plus pemetrexed for elderly Japanese patients and revealed an ORR of 41.2% [25]. Based on these findings, a required sample size of 31 patients was calculated according to the normal approximation method, with an expected response rate of 60%, a threshold response rate of 35%, two-sided alpha = 0.05, and 1 – beta = 0.8. However, the target sample size was increased to 35 patients to account for potential dropout cases. The primary endpoint for the trial was the overall response rate (ORR), while the secondary endpoints were PFS, OS, disease control rate (DCR), and safety events.The ORR was calculated as the proportion of subjects with complete response or partial response as their best treatment responses. The DCR was calculated as the proportion of subjects who achieved stable disease (or better) as their best treatment response. The PFS interval was calculated from the date of enrollment to the first instance of disease progression, death from any cause, or the last follow-up without evidence of progression (for surviving patients with no evidence of progression). The OS interval was calculated from the date of enrollment to the date of death from any cause. Adverse events were evaluated from the first drug administration to 30 days after the last drug administration and were graded based on the Japanese JCOG translation of version 4.0 of the Common Terminology Criteria for Adverse Events.The Wilson method was used to estimate the ORR and DCR with their two-sided 95% CIs. Statistical significance was considered present when the lower limit of the estimated 95% CI was above the threshold of 35% for ORR. The Kaplan–Meier method was used to evaluate the survival curves for PFS and OS, as well as the median and annual values. The Brookmeyer and Crowley method was used to estimate the CI values for median values, and Greenwood’s formula was used to estimate the standard error for annual values.The study enrolled 36 patients between July 2016 and May 2018, with 23 female patients (63.9%) with a median age of 80 years, and 19 patients (52.8%) who were ≥80 years old. The histological types were adenocarcinoma in 35 patients (97.2%) and a mixed type with small cell lung cancer in only 1 patient. Based on the 7th edition of the AJCC system for staging lung cancer, 25 cases (69.4%) were considered stage IV, 10 cases (27.8%) involved relapse after surgery, and 1 case (2.8%) was considered stage IIIB. Among the enrolled patients, 30.6% were former smokers. The EGFR gene mutations involved the exon 20 T790M mutation in all cases, as well as exon 19 deletion in 22 cases (61.1%) and the exon 21 L858R point mutation in 11 cases (30.6%). Brain metastasis was detected in 15 patients (41.7%) (Table 1).The ORR from our previous report was 58.3% (95% CI: 42.2–72.9), which included a complete response rate of 2.8% and a partial response rate of 55.6%. The stable disease rate was 38.9%, and the DCR was 97.2%. The median response duration was 54.9 weeks (95% CI: 26.9–69.1), and a waterfall plot revealed that 33 patients (91.6%) experienced tumor shrinkage, which indicated favorable antitumor activity. Sixteen patients (44.4%) continued treatment beyond progression.The median PFS was 11.9 months (95% CI: 7.9–17.5), with 1-year PFS rate of 50.0% and 2-year PFS rate of 18.3% (Figure 1). The median OS was 22.0 months (95% CI: 16.0–not reached), with 1-year OS rate of 77.8% and 2-year OS rate of 49.5% (Figure 2).Adverse events occurred in 31 cases (86.1%), with Grade 3 or higher adverse events observed in 10 cases (27.8%). Seven patients (19.4%) required dose reductions, 10 patients (27.8%) discontinued treatment because of adverse events, and 1 patient died (2.8%). The adverse event leading to death was a pulmonary infection, although this was judged unlikely to have been caused by the osimertinib treatment. There were no death events caused by drug-induced lung injury. The most frequent adverse event was anemia/hypoalbuminemia (27 patients, 75.0%), which was followed by thrombocytopenia (21 patients, 58.3%), paronychia/anorexia/diarrhea/neutropenia (15 patients, 41.7%), leukopenia/aspartate aminotransferase increase (14 patients, 38.9%), fatigue/acneiform eruption (13 patients, 36.1%), and alanine aminotransferase increase/alkaline phosphatase increase/creatinine increase (11 patients, 30.6%). The Grade 3–4 adverse events included fatigue, anorexia, diarrhea, cardiac ejection fraction decreased, prolonged QT, leukopenia, neutropenia, and aspartate aminotransferase increase. The cases of cardiac ejection fraction were decreased and the cases of prolonged QT were different cases, and delirium and hallucinations were observed in the same patient. Pneumonitis was observed in four patients (11.1%), including two patients (5.6%) with Grade 3–4 pneumonitis (Table 2).Treatment of NSCLC has advanced dramatically after the introduction of molecularly targeted drugs, such as EGFR-TKIs for EGFR-mutated cases. The first- and second-generation of EGFR-TKIs proved to be highly effective in several studies, although the effects tended to only last for approximately 1 year [1,2,3,4,5,6,7,8,9,10,11]. Approximately one-half of the resistant cases involved a gatekeeper mutation in exon 20 (T790M), and osimertinib was developed and approved for the treatment of these cases [12,13]. The results of the FLAURA trials positioned osimertinib as a standard treatment option, and even as an initial treatment option [14,15]. However, many cases still involve treatment in the second line or later, as the T790M mutation was identified via re-biopsy in patients who received first-generation or second-generation EGFR-TKIs as their initial treatment. When the T790M mutation was identified in these cases, patients typically received osimertinib.Aging populations are becoming increasingly common worldwide, and many lung cancer cases involve older patients [26,27]. There are concerns that older patients have a higher risk of developing adverse events, which may necessitate dose reduction or treatment discontinuation, and subsequently result in decreased efficacy. Thus, this phase II study aimed to evaluate the safety and efficacy of osimertinib in elderly patients with EGFR-mutated lung cancer involving the T790M mutation. The primary endpoint was the ORR, and our previous report found that the ORR was 58.3% (95% CI: 42.2–72.9), which fulfilled the efficacy criterion (the lower limit of the CI exceeded the threshold response rate of 35%) [19]. This report describes the secondary endpoints, which include the DCR (97.2%), median PFS (11.9 months), and median OS (22.0 months). In terms of efficacy, the pooled results from the AURA expansion and AURA2 studies revealed an ORR of 66%, a DCR of 91%, a median PFS of 9.9 months, and a median OS of 26.8 months [23]. In addition, phase 3 AURA3 studies revealed an ORR of 70.6%, a DCR of 93.2%, a median PFS of 10.1 months, and a median OS of 26.8 months [13,28]. Thus, while our ORR was lower than that shown in the previous studies, it agrees with the slightly lower ORR (61.1%) that was retrospectively observed in another sample of elderly Japanese patients [29]. Furthermore, our findings regarding PFS and OS do not appear inferior to the results from previous studies, thereby suggesting that osimertinib was effective in elderly Japanese patients. Regarding the effects based on the PS, the ORR of PS0 and PS1 was 75% and 53.6%, respectively, and the PFS was 13.7 months and 11.9 months, respectively. Since there were few cases, it was impossible to discuss the significant differences, but the PS0 group tended to be superior.It is also important to compare the results from osimertinib treatment to those from cytotoxic anticancer drugs, which are the alternative options if osimertinib is not used for T790M-positive cases. For example, the control group for the AURA3 study received platinum plus pemetrexed, which provided an ORR of 31%, a DCR of 74%, a median PFS of 4.4 months, and a median OS of 22.5 months [13,28]. A subgroup analysis of ≥70-year-old Japanese patients from the JACAL study evaluated carboplatin plus pemetrexed and revealed an ORR of 24%, a DCR of 68%, a median PFS of 5.2 months, and a median OS of 16.8 months [30,31]. Thus, our OS findings may be comparable to the results from the entire AURA3 population, although our ORR, DCR, and PFS outcomes are comparable or even slightly better. Interestingly, 71% of the patients in the group that received platinum plus pemetrexed subsequently received additional treatment, with 60% experiencing a greater effect after crossing over to osimertinib treatment. Therefore, while the JACAL study had only included EGFR-mutated cases and did not specifically consider older patients, we believe that osimertinib may provide good outcomes among older patients with EGFR-mutated (T790M) NSCLC.Safety is also an important consideration in this setting, given the concerns regarding the potentially higher risk of adverse events among older patients. In the AURA3 study, it appears that Japanese patients had a higher risk of paronychia, diarrhea, and skin pruritus, although no clear increase was observed among elderly patients. However, elderly patients had a clearly increased frequency of myelosuppression events, such as anemia (75% in this study vs. 8% in AURA3 study), leukopenia (38.9% in this study vs. 8% in AURA3 study), neutropenia (41.7% in this study vs. 8% in AURA3 study), and thrombocytopenia (58.3% in this study vs. 10% in AURA3 study), although the frequencies of Grade 3–4 adverse events were generally comparable. Osimertinib has also been reported to be more frequently myelosuppressed than in other EGFR-TKI in a pivotal study [13,14]. In addition, myelosuppression was reported to be stronger in the analysis of the Japanese population [32]. Although the obvious mechanism was unclear, it was suggested that racial differences might be involved. Since myelosuppression was observed more frequently in the present study than in the aforementioned analysis of the Japanese population, caution should be exercised in the elderly Japanese. Fiala et al. reported that pre-treatment hypoalbuminemia correlated with poor prognosis in advanced NSCLC patients treated with erlotinib [33]. The present study also revealed that anorexia and exhaustion were common (30–40% of cases vs. 16-18% of cases in AURA3 study, including some Grade 3–4 cases), as well as hypoalbuminemia (75% of cases vs. N/A in AURA3 study). Therefore, careful follow-up is needed for elderly patients who are receiving osimertinib. Elevated alkaline phosphatase and creatinine values were also observed, albeit not serious cases, and related follow-up testing is also important. Cardiac adverse events, such as decreased left heart ejection fraction and QT prolongation, were observed in some cases, although only one patient experienced a Grade 3–4 cardiac adverse event. Central nervous system events, such as delirium and hallucination, may be explained by the large proportion of cases with brain metastasis (41.7%), although caution should be exercised if these events present in conjunction with sinusitis and pulmonary infection. Regarding AE by PS, no clear difference was observed between PS0 and PS1.All-grade pneumonitis was observed in 11.1% of cases, and Grade 3–4 pneumonitis was observed in 5.6% of cases. The rates after conventional EGFR-TKI treatment were 4% in the AURA3 study and 7.3% in the Japanese subset of patients, which suggests that Japanese patients may have a higher rate of pneumonitis [13,34]. The difference between our findings and the previous findings may be related to differences in the proportions of patients with a history of smoking (69.4% for the present study, 32.2% for the AURA3 study, and 31.7% for the Japanese subset of the AURA3 population). In addition, the Japanese subset of the FLAURA study population had a higher frequency of pulmonary disorders (all grades: 12%, Grade 3 or higher: 2%); it should be noted that this is a first-line trial. Other reports have also suggested that osimertinib may be associated with an increased incidence of pulmonary disorders relative to other EGFR-TKIs [32]. Nevertheless, the odds ratio for pulmonary disorders after gefitinib treatment was 1.92-fold higher among Japanese patients who were ≥55 years old, which suggests that careful follow-up is required for patients who are ≥75 years old [35].The present study revealed all-grade AEs in 86.1%, Grade 3 or worse AEs in 27.8%, and fatal AEs in 2.8% of the patients. These rates did not appear to be substantially elevated among elderly patients, based on results from the AURA3 study and its Japanese subgroup (all-grade: 97.8% and 100%, Grade 3 or higher: 22.6% and 31.7%, and fatal AEs: 1.4% and 0%). However, AEs leading to treatment discontinuation occurred in 12 patients (33.3%) in our study, which was more common than the rates of 6.8% in the AURA3 study and 7.3% in the Japanese subgroup. For example, we observed drug-induced lung injury in four patients (11.1%), and these patients needed to stop treatment. In addition, three patients (8.3%) discontinued treatment because of Grade 4 AEs (pulmonary infection, hallucinations, and hepatic dysfunction), although those events were judged unlikely to be associated with their treatment. One patient (2.8%) required a two-step dose reduction, and two patients (5.6%) were unable to continue the treatment protocol because of a ≥4-week treatment disruption. Treatment was also stopped in one case involving Grade 3 aspiration pneumonia, one case at the attending physician’s discretion, and one case because the patient refused to continue treatment. Thus, although the safety of osimertinib outside the study protocol has not been evaluated, most of these AEs and treatment discontinuations were likely not to have been caused by a drug-induced pulmonary injury.Most all-grade adverse events involved anorexia, fatigue, myelosuppression, and gastrointestinal symptoms. These complications were generally not serious and could be addressed using conventional management strategies. However, it is important to note that the frequency of drug-induced lung injury may increase, which highlights the importance of a careful follow-up in this population. Despite the potential need for a careful follow-up and the small sample size, which was the limitation in this study, it appears that osimertinib can be a standard treatment even for the elderly patients harboring T790M mutation.While the present study provided encouraging data, we are conducting an additional phase II study (SPIRAL-0) to confirm the safety and efficacy of osimertinib in ≥75-year-old patients with untreated NSCLC harboring EGFR-activating mutations [36]. This may provide further information to guide the increasing use of osimertinib treatment in this setting.Conceptualization, J.U.; validation, J.U., A.N., and K.T.; formal analysis, K.Y.; investigation, O.H., C.S., T.A., N.H., T.I., T.T., M.K., Y.G., H.I., N.H., K.N., H.U., K.U., M.F. (Minoru Fukuda), Y.U., T.Y. (Toshihide Yokoyama), M.A., T.M. (Tadashi Mio), S.N., Y.C., N.T., Y.K., T.M. (Takako Mouri), T.Y. (Tadaaki Yamada) and M.F. (Masaki Fujita); data curation, K.Y.; writing—original draft preparation, A.N.; writing—review and editing, J.U.; supervision, K.T.; project administration, J.U.; funding acquisition, J.U. All authors have read and agreed to the published version of the manuscript.This research was funded by AstraZeneca. K.K., grant number ESR-15-11419.We thank all of the patients who participated in this study, as well as their families. We also thank the Clinical Research Support Center Kyushu for managing the study.K. Takayama received grants from Chugai-Roche Co., Ono Pharmaceutical Co. and personal fees from AstraZeneca Co., MSD-Merck Co., Eli Lilly Co., Boehringer-Ingelheim Co., Daiichi Sankyo Co. and Chugai-Roche Co. outside of the submitted work. K. Yoshimura received personal fees from AstraZeneca Co. and Chugai-Roche Co. outside of the submitted work. T. Ishizuka received grants from Boehringer-Ingelheim Co., Bayer Co, MSD Co., Astellas Co., Ono Pharmaceutical Co., Pfizer Co., Eli Lilly Co, Novartis Pharma K.K., Mochida Pharmaceutical Co and personal fees from AstraZeneca Co., Boehringer-Ingelheim Co., Novartis Pharma Co., and GlaxoSmithKline Co. outside of the submitted work. M. Fukuda received grants from AstraZeneca Co. and Eli Lilly Co. outside of the submitted work. M. Fujita received grants and personal fees from AstraZeneca Co. T. Yokoyama received personal fees from AstraZeneca Co., MSD-Merck Co., Eli Lilly Co., Boehringer-Ingelheim Co., Chugai-Roche Co. outside of the submitted work. J. Uchino received grants from Eli Lilly Japan K.K. and AstraZeneca Co. that are outside of the submitted work. The other authors have no conflicts of interest. The funders had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript; or in the decision to publish the results.Progression-free survival.Overall survival.Patient characteristics.PS: performance status, SCLC: small cell lung cancer.Adverse events.

Med-MDPI/jcm_2/jcm-09-06-01763.txt

@@ -0,0 +1 @@
+Differences in fibromyalgia impact on functioning exist and appear to be influenced by numerous factors, including symptomatology severity, as well as the cognitive profile of the individual. The contribution of these elements, however, tends to be explored in a fragmented manner. To address this issue, we tested a comprehensive structural equation model in which associations of cognitive fusion and pain catastrophizing with function limitations are investigated through fibromyalgia symptomatology (i.e., fatigue, pain severity, and depression) in 231 women with fibromyalgia. In the model, cognitive fusion and two catastrophizing components (magnification and helplessness) were associated with poorer functioning indirectly through fibromyalgia symptomatology. Only the rumination component of catastrophizing had a direct association with functional limitations. All fibromyalgia symptoms were linked to increased functional limitations. A parsimonious model with significant associations only obtained an excellent fit (S-B χ2 = 774.191, df = 543, p < 0.001; CFI = 0.943; RMSEA = 0.043; CAIC = −2724.04) and accounted for 50% of the variance of functional limitations. These results suggest that the relationship between psychological cognitive processes, fibromyalgia symptomatology, and functional limitations is complex and support the need for comprehensive models such as the present. The findings are discussed in the context of personalized psychological treatments (i.e., the need to address certain cognitive processes according to the problematic symptomatology or outcome).Functioning is significantly impaired in people with fibromyalgia (FM). It is widely accepted that a moderate-to-strong association exists between functional limitation and pain severity [1,2,3]. However, pain severity is not the only factor associated with the impact of the syndrome and functional limitations are known to vary across patients reporting similar pain severity levels [4,5]. Thus, it has been argued that functional limitation in patients with FM and other pain conditions should be conceptualized within a more comprehensive biopsychosocial perspective [6]. In this line, fatigue and depression have been repeatedly associated with poorer functioning in people with FM [7,8], which explains why both were added as part of the newer diagnostic criteria published in 2010 by the American College of Rheumatology [9].While acknowledging the important role of FM symptomatology (e.g., pain severity, fatigue, and depression) in daily functioning, in the past years, researchers have been encouraged to investigate the mechanisms that explain individual differences in this symptomatology and, ultimately, in overall functioning (i.e., mediators), as these would be the actual targets during interventions [10]. In this sense, several models of pain have pointed to cognitive factors as key elements explaining individual differences in symptomatology and functionality in chronic pain populations, such as FM [11]. For example, according to the fear-avoidance model of pain, aggravation of symptomatology and disability in the presence of pain occurs when individuals catastrophize about their pain experience [12]. Specifically, because catastrophizers tend to magnify the threat associated with pain; feel helpless in the face of pain; and show higher pain levels, poorer functionality, and increased mental distress [13].Another well-established model of pain that has contributed to the understanding of functioning in people with chronic pain is the psychological flexibility model. Different to the fear-avoidance model, the psychological flexibility model of pain shifts the emphasis from thought-content restructuring to changing the functional relationship with thoughts without challenging their content [14]. In this model, the focus is shifted from pain catastrophizing (a thought content) to cognitive fusion, which is conceptualized as the tendency to being trapped by the meaning ascribed to internal events, such as thoughts, emotions, and images (functional relationship with thoughts, irrespective of their content). In this sense, the psychological flexibility model suggests that cognitive fusion would explain the inability of some individuals to implement adaptive coping methods in the presence of pain or related symptomatology [15]. Indeed, there is evidence to support the influence of cognitive fusion on outcomes in patients with FM [16,17]. Research supports the idea that both processes are distinct and that they could potentially interact and influence each other [18].In sum, the reviewed literature suggests that FM symptomatology (e.g., pain, fatigue, and depression) is closely interrelated and associated with functionality. Additionally, individual differences in symptomatology and functionality are influenced by other biopsychosocial components, such as cognitive factors. Several models of pain have been proposed to explain such complex associations between variables. However, to the best of our knowledge a comprehensive model that includes both the fear-avoidance and the psychological flexibility model and the combination of symptoms and functioning in a structural equation model is missing. This is important to investigate the complex associations between involved factors hypothesized in the literature.In the light of the previous literature [1,7,8,13,17], our goals is to test the hypothesized associations of pain catastrophizing, cognitive fusion, and FM symptom severity (pain, fatigue, and depression) in a theoretically-sound model that accounts for the contribution of these elements altogether in relation to functional limitation. We hypothesize that catastrophizing components and cognitive fusion will be associated with greater disability (see Figure 1). We also anticipate that pain, fatigue, and depression will mediate the aforementioned relationship between cognitive factors and functional limitations. Because we expect that study variables will be intercorrelated and will share common variance in relation to functional limitation, we anticipate that only a reduced number of independent variables will be uniquely associated with functional limitation when all variables are considered altogether in the same model. Because such complex models remain unexplored in this population, anticipating which paths will remain significant is difficult at this stage, and this will be investigated in an exploratory manner. A more parsimonious model will be proposed as a post hoc to the exploratory analyses.The sample comprised 231 women with FM (mean age = 56.91 years, SD = 8.9 years, range 30–78 years). Their sociodemographic characteristics are shown in Table 1. The eligibility criteria to participate in the present study included having a diagnosis of FM according to the American College of Rheumatology (ACR) criteria [9,19], being over 18 years of age, and providing a written consent to participate in the investigation. Patients were recruited from different associations in Spain (Madrid, Ciudad Real, Albacete, Guadalajara, and Toledo).The study followed the ethical principles for research with human participants (Helsinki Declaration) and was approved by the University Ethics Committee (Ref. PI17/00858).The Spanish version of the Pain Catastrophizing Scale was used to evaluate the three components of pain catastrophizing, understood as an exaggerated negative orientation towards actual or anticipated pain experiences, namely rumination, magnification, and helplessness [20,21]. Rumination includes ruminative thoughts, worry, and an inability to inhibit pain-related thoughts; magnification relates to an exaggeration of the unpleasantness of pain situations and expectancies for negative outcomes; and helplessness refers to negative appraisals related to the ability to deal effectively with painful stimuli. This 13-item scale with a 4-point Likert response format contains items such as “I keep thinking about how badly I want the pain to stop”, “I become afraid that the pain may get worse”, or “there is nothing I can do to reduce the intensity of the pain”, for rumination, magnification and helplessness, respectively. Higher scores on the scales represent higher catastrophizing. In the present study, the internal consistency of the three scales was adequate (0.88, 0.78, and 0.89 for rumination, magnification, and helplessness, respectively).The Spanish version of the Cognitive Fusion Questionnaire was used to assess cognitive fusion, which is conceived as the degree to which a person is psychologically tangled with and dominated by the form or content of its thoughts [22,23]. High scores in this 7-item scale with a 7-point Likert response format indicate high cognitive fusion. The scale contains items like “I get so caught up in my thoughts that I am unable to do the things that I most want to do”. The internal consistency obtained in this study was excellent (0.91).To assess pain severity, we used the mean score of the four pain severity items from the Brief Pain Inventory [24]: maximum, minimum, and overall pain intensity during the last 7 days and pain intensity at the current time. Each rating is evaluated using an 11-point numerical scale (0 = “no pain” and 10 = “the worst pain you can imagine”). This procedure to measure pain severity has been widely used in the pain literature [25]. In this study, the internal consistency of this scale was high (0.86).The Depression subscale of the Spanish version of the Hospital Anxiety and Depression Scale (HADS) was used to evaluate the severity and presence of depressive symptoms over the preceding week [26]. The HADS is a brief and widely used instrument to measure the possible presence of anxiety and depressive states in medical, non-psychiatric outpatient clinic settings. High scores in this 7-item subscale with a 4-point Likert response format denote high depressive symptomatology. The scale is composed of items such as “I have lost interest in my appearance”. The internal consistency of this subscale was high (0.85) in the present study.The Physical Fatigue dimension of the Spanish version of the Multidimensional Fatigue Inventory (MFI) was used [27]. This questionnaire is a 20-item assessment tool with a 5-point Likert response format that evaluates five domains of fatigue (general fatigue, physical fatigue, mental fatigue, reduced motivation, and reduced activity). High scores in this 4-item dimension indicate a high degree of fatigue symptoms that are physical in nature. This subscale contains items like “physically, I feel only able to do a little”. The selection of this dimension was motivated by its conceptualization of fatigue as a physical symptom, in agreement with the proposed theoretical model and consistent with previous literature on the relationship between fatigue and functionality [28]. Although the rest of the MFI dimensions contribute to assess fatigue in all its complexity, they were discarded because they reflect slightly different (although related) conceptualizations from those intended in this work, such as cognitive (mental fatigue dimension) or affective-emotional (reduced motivation dimension) aspects. Furthermore, the conceptual overlap of both the reduced activity dimension with functionality measures [29] and the general fatigue dimension with pain measures [30] led to discard their use to avoid overlaps with other variables used in the study. In the present work, the internal consistency of this subscale was good (0.73).We used the “overall impact” dimension of the Spanish adaptation of the Revised Fibromyalgia Impact Questionnaire (FIQ-R) to evaluate the impact of FM on functioning [31,32]. In the FIQ-R, the two items that compose the “overall impact” dimension are answered on an 11-point numerical rating scale from 0 to 10, with different verbal anchors depending on the item. Higher scores represent higher impact perception. Note that this dimension has two items only (“fibromyalgia prevented me from accomplishing goals for the week” and “I was completely overwhelmed by my fibromyalgia symptoms”) and three are usually recommended for scale development. However, the use of reduced, even single-item measures is frequent and has shown to be psychometrically valid in the pain literature [33]. This was the case of the “overall impact” scale in the present study, which obtained a high internal consistency score (0.81).The Statistical Package for Social Sciences IBM SPSS for Windows, version 22.0 (Armonk, NY, USA) [34] was used for descriptive analyses, and EQS for Windows, version 6.2 (Encino, CA, USA) [35] was used to conduct structural equation modeling analyses.Preliminarily, both an analysis of missing values and an evaluation of data normality were performed. The variables of interest contained some missing values (range 0.4% to 1.7%) that did not exceed 5% of the initial sample (n = 234), so data missingness did not compromise the reliability of scores [36]. As suggested by the Little’s test (χ2 = 274.01, df = 257, p = 0.22), missing data appeared to be missing completely at random (MCAR). Because all missing data provided from 3 participants only, we eliminated them from the analyses (listwise deletion), as recommended in the literature [37]. This resulted in a final sample size of 231 patients.Convergent validity and reliability of the variables included in the structural equation model were determined assessing the average variance extracted (AVE) and composite reliability (CR), respectively. AVE over 0.50 and CR scores over 0.70 are considered good [38], while AVE values slightly below 0.50 are also acceptable as long as CR is over 0.60 [39]. An alpha level of 0.05 was selected in the present study despite the number of tests conducted due to the exploratory nature of the analyses. Using more restrictive alpha levels, for example by implementing a Bonferroni–Holm correction, is less advisable in exploratory studies as they increase the risk of false negatives by attempting to reduce the risk of false positives [40]. The present study findings should serve guide future confirmatory models in which such corrections would be more sensible.A model development strategy was followed in this study, using the robust maximum likelihood estimation method. The goodness of fit of the model was assessed by: (1) the Satorra–Bentler scaled χ2 statistic (S-B χ2), its degrees of freedom (df), and p values; (2) the Comparative Fix Index (CFI), as an incremental fit index; (3) the Root Mean Square Error of Approximation (RMSEA) with its 90% confidence interval (CI); and (4) the Consistent Akaike Information Criterion (CAIC), as a parsimony index to compare non-nested models. Considering the sample size and the number of observed variables, adequate model fit was determined using the following cutoffs: CFI > 0.92, RMSEA < 0.08, and CAIC of less magnitude indicating greater parsimony [37,38]. Regarding S-B χ2, although non-significant p values indicate good fit, with complex models such as the current one, a model can be considered adequate if the rest of the fit indices reach the cut-off values [37]. Finally, the Harman’s single-factor test [41] was employed in order to test whether common method variance was an issue.This was a cross-sectional survey performed in people with FM belonging to patients’ associations. Of them, 268 agreed to participate in the study and met our inclusion criteria. Finally, a total of 231 patients constituted the final sample (25 did not attend the scheduled assessment appointment, nine questionnaires were left blank, and 3 questionnaires contained missing data that could not be retrieved because participants could no longer be reached).Regarding the measurement model, all items were included as observed variables. However, two of them (one on the depressive symptomatology scale and another on the physical fatigue scale) yielded reduced factor loadings and were eliminated. The measurement model consisted of 35 observed variables grouped into eight latent variables or latent factors. In all cases, we obtained adequate scores both in the factor loadings of the observed variables and in the AVE and the CR of the factors (see Table 2). Finally, the model showed good fit (S-B χ2 = 764.331, df = 532, p < 0.001; CFI = 0.942; RMSEA = 0.044 (90% CI from 0.036 to 0.050)).With regard to the structural model, the hypothesized model (M1) obtained an adequate fit (S-B χ2 = 772.023, df = 533, p < 0.001; CFI = 0.940; RMSEA = 0.044 (90% CI from 0.037 to 0.051); CAIC = −2661.786) and accounted for 50% of the impact of FM. However, this first model could not confirm many of the initially hypothesized relationships between variables (see Figure 2 and Table 3). Specifically, the non-significant pathways were from rumination and magnification to depressive symptoms, pain severity, and FM impact; from rumination to physical fatigue; from helplessness to FM impact; from cognitive fusion to both pain and FM impact; from pain severity to both depressive symptoms and physical fatigue; from depressive symptoms to FM impact; and from physical fatigue to depressive symptoms and FM impact. While the implications of these findings will be discussed in more detail in the next section, the fact that hypothesized contributions were not confirmed is likely to be explained by the multivariate approach used in the present study, in which the independent variables had to compete between them to account for unique variance of outcomes.The hypothesized model was thus re-specified removing the non-significant pathways to test the adequacy of this modified and arguably more parsimonious model (M2). Figure 3 depicts that this second model achieved a good fit (S-B χ2 = 774.191, df = 543, p < 0.001; CFI = 0.943; RMSEA = 0.043 (90% CI from 0.036 to 0.050); CAIC = −2724.04). Different to the first model, all the relationships in this second model were significant (see Table 3), and the same amount of variance of FM impact was accounted for by the independent variables (i.e., 50%). After analyzing the modification indices, the results did not suggest changing any of the relationships in this re-specified model. Thus, the results of comparing the fit indices and the parsimony index of both structural models suggested the preference for the second model M2.To conclude, the results provided by the Harman’s single-factor test (S-B χ2 = 2289.813, df = 560, p < 0.001; CFI = 0.571; RMSEA = 0.116 (90% CI from 0.111 to 0.121)) revealed that a single factor could not account for the variance in the present data, implying that common method variance bias was not a critical problem in this study.Popular models of pain, such as the fear-avoidance and the psychological model of pain, have hypothesized that functional limitations in patients with chronic pain are influenced by a complex interplay between symptomatology (e.g., pain severity, fatigue, and depression) and psychological mechanisms (e.g., catastrophizing and cognitive fusion). Research, however, tends to explore the relationship between the aforementioned variables in a fragmented way (e.g., by including a reduced number of symptoms or psychological mechanisms only or by exploring their relationship with one outcome at the same time without accounting for complex direct and indirect associations). The present study contributes to the existing literature by proposing a more comprehensive model of functional limitations in FM patients that includes a set of important symptoms and psychological mechanisms from different psychological perspectives. Overall, our model was satisfactory in terms of explained variance (50%). Another important finding was that a large number of significant bivariate associations became non-significant when explored in a complex multivariate model. This supports the utility of such complex models that account for shared variance.Regarding the relationship between FM symptoms and functioning, our structural equations model evidenced an association of symptoms (pain severity, fatigue, and depression) with functional limitation, supporting the extensive research in this regard [42,43]. New to the literature is that all symptoms were uniquely and independently associated with functional limitations in the structural equations model after controlling for the role of the other symptoms and the psychological mechanisms (i.e., catastrophizing and fusion). This is important as it suggests that all these symptoms should be considered in a comprehensive evaluation of FM functionality, which is consistent with their inclusion in the 2010 diagnostic criteria of the American College of Rheumatology [9]. These results are also consistent with past research using an integrative model similar to the present one, in which both physical fatigue and depression symptoms were associated with the quality of life of patients with FM. Thus, even though the study outcomes are different, these results support the idea that the relationship between fibromyalgia symptoms and outcomes is relatively robust [44].Additionally, our findings evidenced that, while important psychological factors like catastrophizing and cognitive fusion might indeed impact on functioning, symptoms might as well uniquely contribute to functioning even after controlling for these cognitive factors. Indeed and as noted in past research, pain and other symptoms appear to have a relatively inescapable limiting nature when there are very intense, even in the presence of adaptive psychological thinking [45]. While acknowledging the important role of symptom severity on functioning, our model also showed that pain catastrophizing and cognitive fusion might be important factors uniquely associated both with FM symptomatology and functional limitation, thus supporting a vast amount of literature in this regard [46,47]. Interestingly, the contribution of catastrophizing components and cognitive fusion on FM impact on functioning was mostly indirect through FM symptomatology, and each component showed a particular association with each symptom. Specifically, what our model suggests is that the only catastrophizing component that is directly associated with FM impact is rumination. Conversely, magnification would be indirectly associated with FM impact via increased physical fatigue, while helplessness would be indirectly related to FM impact through pain severity, depressive symptoms, and physical fatigue. As noted in past research, “whether these components of catastrophizing work interactively to explain behavior has not been examined in a systematic fashion, but certainly would represent an exciting avenue for future research” [43].The present study has some limitations. First, the data is based on a cross-sectional design, so it is not possible to establish cause-and-effect inferences. Second, while FM patients were recruited from different associations in Spain, the sample is not representative of the whole population of FM patients in this country. Moreover, we did not include males because of the very small prevalence of this disease in this population, so the results might not be generalizable to males. Similarly, the model was tested with patients reporting very different pain durations. Therefore, the extent to which the relationship between factors differs as a function of pain duration or any other covariate (i.e., moderators) is unclear. Additionally, while we included important psychological mechanisms in pain research, the list is not complete. We expect that the present study will inspire future similar investigations including these and other important psychological processes, such as neuroticism, self-efficacy, coping, and acceptance, to name some examples. Note that future investigations should have a more confirmatory nature if attempting to replicate the present study findings, so paths should be constrained to the ones indicated in the present investigation, and alpha levels should be corrected for multiple testing. Another aspect that could be considered as a limitation refers to the utilization of the outcome variable “overall impact” as opposed to the total FIQ-R score. Note that the FIQ-R is composed by three dimensions, namely, overall functioning, a physical functioning scale that mainly focuses on household chorus, and a measure of symptom severity that includes pain severity and depression. The preference for the “overall impact” dimension was motivated by the study focus on overall functioning as opposed to functioning in specific situations (e.g., household chorus). Additionally, as indicated in past similar research [48], the inclusion of the “symptom severity” scale would have contaminated the relationship between at least two independent variables (i.e., pain severity and depression) and the dependent variable. As a final remark, it is important to note that all measures were obtained through self-report, which means that perceived functioning might differ from objectively-measured functioning (e.g., with accelerometers or performance-based physical fitness tests) [44,49]. Similar to the majority of existing literature, our focus in the present study was on subjective functioning and symptomatology only due to the inherently important nature of the subjective experience on outcomes [50]. However, the debate on whether objective or subjective measures of pain and related outcomes should be used in people with chronic pain has been a matter of concern for decades and multimodal measurement (i.e., a combination of both) appears to be preferable when possible [51].While acknowledging the aforementioned shortcomings, the present study results might as well have a number of clinical implications. In particular, our findings support the idea that psychological cognitive processes (i.e., catastrophizing and cognitive fusion) are associated with poor outcomes in FM patients, which justifies their inclusion in interdisciplinary treatment programs for FM [52]. Similarly, the results indicated that FM symptomatology (e.g., pain severity, fatigue, and depression) significantly and uniquely contributes to functional limitations, which is again consistent with the literature [42,43]. A novel contribution of the present study was the combination of all these elements in a single structural model to explore the unique association between study variables when accounting for the shared variance between independent variables. This is important as it might guide interventions in a more effective manner (i.e., personalization) [53]. For example, according to our findings, in the presence of a patient with FM with impaired functioning because of increased depression, helplessness and cognitive fusion (due to their associations with depressed mood) and rumination (because of its direct contribution on FM impact) might be potentially important treatment targets. Alternatively, if an individual with FM is largely impaired due to severe pain levels, our results support the idea that a treatment focus on reducing helplessness (because of its unique association with pain severity) and rumination (again due to direct relationship with FM impact) might be appropriate.Conceptualization, C.É., O.L.-R., C.S.-R., P.C., and C.P.; methodology, C.É., O.L.-R., C.S.-R., and C.P.; software, O.L-R.; validation, O.L.-R.; formal analysis, O.L.-R.; investigation, O.L.-R.; resources, C.É., O.L.-R., P.C., and C.P.; data curation, C.É., O.L.-R., P.C., and C.P.; writing—original draft preparation, C.É., O.L.-R., C.S.-R., P.C., and C.P.; writing—review and editing, C.É., O.L.-R., C.S.-R., and C.P.; visualization, C.É., C.S.-R., and C.P.; supervision, C.P.; project administration, C.P.; funding acquisition, C.P. All authors have read and agreed to the published version of the manuscript.This work was funded by the Health Research Fund (Fondo de Investigación en Salud), grant number PI17/00858 from the Instituto de Salud Carlos III (Spain).The authors gratefully acknowledge all the participants for their collaboration and enthusiasm.The authors declare that they have no conflict of interest.Hypothesized model of cognitive fusion and pain catastrophizing associations with fibromyalgia impact through physical fatigue, pain severity, and depression. The “+” symbol represents an expected positive association.Standardized coefficients of the hypothesized model (M1). Dashed lines represent non-significant paths. The latent factors of the model are shown, while the observed indicators and measurement errors have been omitted for space and clarity reasons. The squared multiple regression coefficients (R2) reflect the amount of factor variance associated with variance of its independent variables.Standardized coefficients of the re-specified model (M2). Only significant paths and the latent factors of the model are shown; the observed indicators and measurement errors have been omitted for space and clarity reasons. The squared multiple regression coefficients (R2) reflect the amount of factor variance associated with variance of its independent variables.Sociodemographic characteristics of the participants.SD, Standard Deviation.Validity, reliability, and estimates of the measurement model.AVE = average variance extracted; CR = composite reliability; R2 = squared multiple correlation. B = unstandardized coefficient; β = standardized coefficient; SE = standard error. In the model, all β were significant at p < 0.001.Direct association estimates of the structural models.*** p < 0.001, ** p < 0.01, * p < 0.05. B = unstandardized coefficient; β = standardized coefficient; SE = standard error.

Med-MDPI/jcm_2/jcm-09-06-01764.txt

@@ -0,0 +1,34 @@
+Self-rated health is a strong health marker. Migrants have been suggested to have poorer self-rated health than non-migrants (i.e., native-born). However, little is known about whether there are disparities in self-reported health in relation to pregnancy. Therefore, the aim of the current study was to examine the odds of poor self-rated health before, during and after pregnancy in migrant women as compared to women born in Sweden. We utilized population-based data from the Swedish Pregnancy Register containing 0.5 million women born in Sweden (i.e., non-migrant women) and migrant women between 2010 and 2018. Self-rated health was reported on a 5-point scale (from very poor to very good). Very poor and poor health were categorized as poor self-rated health. Logistic regression was utilized to calculate odds ratios (ORs) that were unadjusted and adjusted for covariates (age, parity, educational attainment and body mass index). The results demonstrate disparities in self-rated health across birth regions. In comparison to women born in Sweden, women born in Latin America and the Caribbean, South Asia as well as North Africa and the Middle East had consistently higher odds of poor self-rated health before, during and after pregnancy (ORs ranging from 1.14 to 1.96 in both unadjusted and adjusted models). Although women born in Sub-Saharan Africa did have comparable self-rated health as to women born in Sweden before pregnancy, after accounting for covariates, they had lower odds of poor self-rated health during and after pregnancy (ORs: 0.71 and 0.80 respectively). Therefore, additional measures and support may be needed to tackle disparities in health between migrant and non-migrant women before, during and after pregnancy. Self-rated health refers to the overall perception an individual has about their health and it is often measured by a one-item question, e.g., ‘How would you rate your general state of health?’ [1]. Albeit subjective, self-rated health by a one-item question has shown to be a useful marker of objectively measured health status [1,2,3,4,5,6,7]. Indeed, poor self-rated health has been associated with higher risk of mortality [1,2,3,4], cardiovascular disease [3,5], mental illness [4,6] and pregnancy complications [7]. For instance, a meta-analysis showed that individuals with poor self-rated health had twice the mortality risk as compared to persons with excellent self-rated health irrespectively of sex and follow-up time [1]. Self-rated health is typically used as a trustworthy indicator of public health and has been used to examine health disparities in various populations, e.g., [8,9,10,11]. Migrant health including reproductive health is a public health priority and an increasing proportion of the European adult population are first-generation migrants [12,13]. Pregnancy and the postpartum period are sensitive periods in time in which women may experience poor self-rated health [14], mental illness [15] as well as pregnancy complications [16,17]. Research has shown that migrant populations may have poorer self-rated health than the majority population (i.e., non-migrants) in European countries [11]. However, only a few studies have examined self-rated health in migrant and non-migrant women in relation to pregnancy [8,10]. To the best of our knowledge, no previous study has examined differences in self-reported health among migrant and non-migrant women at multiple time points before, during and after pregnancy in a large nationwide sample. The aim of this study was therefore to examine the odds of poor self-rated health before, during and after pregnancy in migrant women as compared to women born in Sweden. We utilized population-based data from the Swedish Pregnancy Register containing 0.5 million women born in Sweden (i.e., non-migrant women) and migrant women born in all seven so called super-regions [18]: (1) Central Europe, Eastern Europe and Central Asia; (2) high income countries (Sweden not included); (3) Latin America and the Caribbean; (4) North Africa and the Middle East; (5) South Asia; (6) Southeast Asia and East Asia; and (7) Sub-Saharan Africa. We used population-based data from the Swedish Pregnancy Register with a coverage of approximately 90% of all births in Sweden [19]. Data from between 2010 and 2018 was utilized and encompassed 841,503 singleton pregnancies with mothers between 15 and 55 years of age. Of these pregnancies, there was missing data regarding maternal country of birth for 86,590 pregnancies and 125,005 pregnancies had missing data for covariates (i.e., educational attainment, parity and body mass index, BMI). Finally, there was missing data regarding self-rated health before (n = 52,143), during (n = 196,660) and after pregnancy (n = 203,484). Thus, the final analytic sample consisted of 577,765, 433,248 and 426,424 pregnancies before, during and after pregnancy, respectively. This study received ethical approval by the Regional Ethical Review Board, Stockholm, Sweden (2018/656-31). Since this is a registry study, data cannot be shared publicly due to legal reasons. Data could be made available to researchers after ethical approval (see https://etikprovningsmyndigheten.se and https://medscinet.com/gr/engelska.aspx).Data regarding the women’s own birth country, age, parity, occupation and educational attainment (no education/elementary school, high school or university degree) was reported by the women at their first visit in antenatal care which is usually scheduled in the first trimester (around gestational week 8). Body height and weight were measured during this visit and BMI (kg/m2) was calculated as the body weight (kg) divided by height squared (m2). Data on the birth country of the women was categorized as follows: (1) Sweden; (2) Central Europe, Eastern Europe and Central Asia; (3) high income countries (Sweden not included); (4) Latin America and the Caribbean; (5) North Africa and the Middle East; (6) South Asia; (7) Southeast Asia and East Asia; and (8) Sub-Saharan Africa. Super-regions and other regions were categorized as described in the Global Burden of Disease study [18], which considers geographic closeness and epidemiological similarity for the creation of regions. Self-rated health before pregnancy was reported during the first visit in antenatal care whereas self-rated health during and after pregnancy was reported by the women at postpartum visit in antenatal care, 4 to 16 weeks after delivery. Women reported their self-rated health using a 5-point scale (i.e., very poor, poor, neither good or poor, good and very good). Very poor and poor health were categorized as poor self-rated health in the analysis. We utilized logistic regression to calculate odds ratios (ORs), with corresponding 95% confidence intervals (CIs), for poor self-rated health before, during and after pregnancy by birth regions. Two logistic regression models were fitted: an unadjusted model and a model adjusting for maternal age, parity, educational attainment and BMI. Women born in Sweden were the reference group in all analyses. The statistical analysis was conducted using SPSS Statistics 22 (IBM, Arnmonk, NY, USA) and R version 3.6.1.Table 1 presents descriptive data of the women included in this study. The data showed variation in self-rated health across birth-regions. There was also a clear trend that self-rated health decreased during pregnancy and returned to similar levels as before pregnancy, postpartum. Data for 102 individual countries with ≥100 pregnancies is presented in Table S1. We also analyzed the proportion of women with data on self-rated health according to birth region (Table S2). Women born in Sweden had generally higher proportions of available data regarding self-rated health before (88.0%), during (69.8%) and after pregnancy (68.3%) whereas women born in Sub-Saharan Africa had generally lower proportions of available data before, during and after pregnancy (81.5%, 48.8% and 48.9%, respectively). Figure 1A shows the odds of poor self-rated health before pregnancy by birth regions (detailed data is presented in Table S3). In the unadjusted model, women born in Latin America and the Caribbean (OR, 1.68 [95% CI, 1.43 to 1.96]), North Africa and the Middle East (OR, 1.96 [95% CI, 1.87 to 2.05]) and South Asia (OR, 1.34 [95% CI, 1.16 to 1.56]) had considerably higher odds of having poor self-rated health before pregnancy as compared to women born in Sweden. Although the estimates generally were attenuated in the adjusted model, women born in Latin America and the Caribbean (adjusted OR, 1.52 [95% CI, 1.30 to 1.76]), North Africa and the Middle East (adjusted OR, 1.52 [95% CI, 1.44 to 1.60]) and South Asia (adjusted OR, 1.28 [95% CI, 1.10 to 1.49]) still had higher odds of poor self-rated health than women born in Sweden. Women born in Sub-Saharan Africa had higher odds of poor self-rated health than women born in Sweden in the unadjusted model (OR, 1.44 [95% CI, 1.33 to 1.56]). However, after adjustments for maternal age, parity, educational attainment and BMI, women born in Sub-Saharan Africa had slightly lower odds of poor self-rated health (adjusted OR, 0.92 [95% CI, 0.84 to 0.99]) as compared to women born in Sweden.During pregnancy, women born in Latin America and the Caribbean (adjusted OR, 1.33 [95% CI, 1.17 to 1.51]), South Asia (adjusted OR, 1.22 [95% CI, 1.09 to 1.37]) and North Africa and the Middle East (adjusted OR, 1.14 [95% CI, 1.10 to 1.20]) had higher odds of poor self-rated health than women born in Sweden, both in the unadjusted and adjusted model (Figure 1B) (detailed data in Table S3). Conversely, women born in Sub-Saharan Africa had lower odds of poor self-rated heath both in the unadjusted and adjusted model (adjusted OR, 0.71 [95% CI, 0.66 to 0.77]).The ORs of poor self-rated health after pregnancy by birth regions are presented in Figure 1C (detailed data in Table S3). Both in the unadjusted and adjusted model, women born in Latin America and the Caribbean (adjusted OR, 1.70 [95% CI, 1.40 to 2.07]), South Asia (adjusted OR, 1.52 [95% CI, 1.27 to 1.83]) and North Africa and the Middle East (adjusted OR, 1.53 [95% CI, 1.42 to 1.64]) had higher odds of poor self-rated health after pregnancy as compared to women that were born in Sweden. On the contrary, women born in Sub-Saharan Africa had lower odds of poor self-rated heath (adjusted OR, 0.80 [95% CI, 0.70 to 0.91]).We also performed a series of sensitivity analyses to assess the robustness of our findings. First, we examined whether the ORs of poor self-rated health before pregnancy would change if we only utilized data for pregnancies with existing data of self-rated health for all three time points (before, during and after pregnancy). As shown in Figure S1, the ORs were comparable to our main analysis of poor self-rated health before pregnancy. Second, as shown in Table S4, we also included an additional adjustment for maternal occupation in the adjusted model for poor self-rated health before, during and after pregnancy, and estimates remained virtually the same as in the adjusted models. Third, we estimated a generalized linear mixed model with a time by region interaction to corroborate our findings and to analyze changes in self-rated health by birth regions (Table S5). The results showed that the ORs for poor self-rated health before pregnancy were similar across birth regions as compared to our main analyses (i.e., Figure 1A). Regarding changes during and after pregnancy, women born in North Africa and the Middle East as well as Sub-Saharan Africa both had a decrease in the ORs for poor self-rated health which also agrees with the findings presented in Figure 1 (please see Table S5 for all results). Forth, we examined whether the prevalence of poor self-rated health changed during the study years. Our analyses showed that the prevalence of poor self-rated health before (3.0% vs. 2.8%), during (8.6% vs. 8.6%) and after pregnancy (2.7% vs. 2.4%) was comparable between 2010 and 2014 as compared to between 2015 and 2018, respectively. Furthermore, associations of birth regions with self-rated health were generally very comparable between 2010 and 2014 versus between 2015 and 2018. Fifth, we also adjusted associations for premature birth (delivery < 37 gestational weeks) but results remained virtually identical. Sixth, we performed an additional adjustment for the use of an interpreter during the visits in antenatal care (yes vs. no), which may serve as a surrogate measure for acculturation, but results and conclusions remained the same. Finally, we also adjusted our results for the county of the maternity care and results remained similar to our main results. This population-based study of over 0.5 million pregnancies in Sweden demonstrates disparities in self-rated health across birth regions. In comparison to women born in Sweden, women born in Latin America and the Caribbean, South Asia as well as North Africa and the Middle East had consistently higher odds of poor self-rated health before, during and after pregnancy. Interestingly, although women born in Sub-Saharan Africa did have comparable self-rated health as to women born in Sweden before pregnancy, after adjusting for covariates, they had lower odds of poor self-rated health during and after. Only a few previous studies have examined self-rated health in migrant and non-migrant women in relation to pregnancy [8,10]. These studies have only utilized single time points which complicates comparison to our findings before, during and after pregnancy. Liu et al. reported that migrant women born in the five most prevalent countries of emigration (i.e., Syria, Iraq, Somalia, Eritrea or Afghanistan) had greater risk of poor self-reported health before pregnancy than women born in Sweden [10]. These findings are in line with our results although our results (including the country-specific results in Table S1) indicate that women born in Syria, Iraq and Afghanistan (i.e., the Middle East) have worse self-rated health than women born in Somalia and Eritrea (i.e., Sub-Saharan Africa). Furthermore, a French study by El-Khory et al. found that women born in North Africa and Turkey, but not Sub-Saharan Africa, had statistically significant higher odds of poor self-rated health two months postpartum as compared to women born in France or the EU [8,10]. These results can be reconciled with our findings showing that women born in North Africa and the Middle East have considerably greater odds of poor self-rated health postpartum compared to women born in Sweden, whereas women born in Sub-Saharan Africa had comparable or slightly lower odds. Furthermore, we found no previous studies that reported data for self-reported health of women born in South Asia or Latin America which is of importance considering that women born in these regions consistently had greater odds of poor self-reported health before, during and after pregnancy. Finally, we also extend the literature by examining differences in self-rated health before, during and after pregnancy for all super-regions and 102 countries of birth. Our results may also be compared to European surveys of women in working age. Such studies have found that migrant women born in North Africa and the Middle East [20] including specific countries such as Iraq [21,22], Iran [22,23], Turkey [22,23] and Lebanon [22] to have worse self-reported health than non-migrants. There is less data regarding women born in other regions than North Africa and the Middle East even though a previous Swedish study reported that women born in Latin America, but not Sub-Saharan Africa had greater odds of poor self-rated health than women born in Sweden [20] which confirms our findings. Although our observed inequalities in self-reported health were generally attenuated after adjustments for covariates including educational attainment, they remained higher than for women born in Sweden. This observation is in line with previous studies conducted in non-pregnant women suggesting that differences in self-reported health between migrants and non-migrants may be attenuated but not diminished after adjusting for socioeconomic status indicators [20,22]. We also found birth regions that had associations with better self-reported health. For instance, women born in Southeast Asia and East Asia had consistently slightly lower odds of poor self-reported health before, during and after pregnancy. Interestingly, women born in Sub-Saharan Africa had higher or comparable odds of poor self-rated health before pregnancy compared to women born in Sweden, but comparable or lower odds during and after pregnancy. A potential reason why women born in Sub-Saharan Africa may have worse self-rated health before pregnancy but better self-rated health during and after pregnancy is due to attrition, i.e., that women born in Sub-Saharan Africa with poor self-reported health are lost during follow-up. However, the results were identical in our sensitivity analyses which only included women with complete data including the follow-ups. Another reason for the interesting results could be that pregnancy and childbirth may be a protective factor given the cultural importance of childbirth and family with the accompanying social support given to the pregnant women, e.g., [24,25,26] although further studies are needed. The main strength of this study was the population-based sample of women from the Swedish Pregnancy Register which covers approximately 90% of all birth in Sweden [19]. Furthermore, the large study sample of over 0.5 million women in a multiethnic country allowed analysis of all super-regions with large samples sizes for all regions (all n > 3000).This study also has some limitations that need to be acknowledged. First, self-rated health before, during and after pregnancy was examined using a one-item question. However, previous studies of self-rated health have showed strong associations between self-rated health and health outcomes in different populations and ethnic groups [1,27,28] using a one-item question. Second, as for all studies examining potentially vulnerable groups there is a risk of attrition, i.e., that migrant women with poor self-rated health are not included in the data which may underestimate the prevalence of poor self-reported health. Our analyses (Table S2) showed that the registry contained generally more complete data on self-reported health in women that were born in Sweden as compared to other birth regions such as Sub-Saharan Africa. Our sample was, however, population-based and the sensitivity analyses showed that the results for the self-reported health before pregnancy were similar even when only those who had data at all three time points were included. However, further studies should address the role of attrition for the results. Third, we only had data on first-generation migrants and did not have any data regarding the reason for migration. Thus, future studies could consider self-rated health among first and second-generation migrants, reasons for migration and how these factors potentially relate to self-reported health. Forth, although we adjusted for premature birth in a sensitivity analyses, future studies should consider other adverse pregnancy outcomes in relation to self-rated health. Finally, we did not have any data regarding acculturation or length of residency although results remained virtually the same after adjusting for the use of a professional interpreter in maternity care (as a crude proxy for acculturation). Our results demonstrate distinct disparities in self-rated health. In particular, women born in Latin America and the Caribbean, South Asia as well as North Africa and the Middle East had consistently poorer self-reported health before, during and after pregnancy. These disparities may be of both clinical and public health importance considering that poor self-reported health is a strong risk factor for later mortality and morbidity [1,2,3,4,5,6,7]. Thus, the findings indicate that migrant women experience poor self-rated health before, during and after pregnancy which motivate additional efforts to promote health but may also provide an opportunity to intervene considering the regular appointments in antenatal care. Previous literature has identified some probable causes of poor self-rated health in migrant non-pregnant populations such as discrimination, low social support and unhealthy health behaviors, e.g., [29,30,31]. However, further studies of the determinants of self-rated health before, during and after pregnancy may be needed in order to provide optimal support and care during pregnancy as well as to form a basis for future health promotion. Furthermore, additional studies are warranted considering that associations of birth regions and self-rated health in migrants may differ due to the country of residence and its integration policies [32,33]. This population-based study of over 0.5 million pregnancies from the Swedish Pregnancy Register demonstrates disparities in self-rated health across birth regions. For instance, women born in Latin America and the Caribbean, South Asia as well as North Africa and the Middle East had consistently higher odds of poor self-rated health before, during and after pregnancy as compared to women born in Sweden. Our results indicate that additional measures and support may be needed to tackle disparities in health between migrant and non-migrant women before, during and after pregnancy.The following are available online at https://www.mdpi.com/2077-0383/9/6/1764/s1, Table S1: Poor self-rated health before, during and after pregnancy according to birth country (countries/regions with ≥100 pregnancies). Table S2: Proportion of women with data of self-rated health according to birth regions. Table S3: Odds ratios with 95% confidences intervals for poor self-rated health before, during and after pregnancy by birth regions. Figure S1: Sensitivity analysis showing the odds ratios of poor self-rated health before pregnancy by birth regions for women with complete data of self-rated health at all time points, i.e., before, during and after pregnancy (n = 398,025). Table S4: Odds ratios with 95% confidences intervals for poor self-rated health before (n = 576,491), during (n = 432,266) and after pregnancy (n = 425,468) by birth regions with adjustments for maternal age, parity, educational attainment, body mass index as well as occupation. Table S5: Generalized linear mixed model with a time by region interaction for self-rated health before, during and after pregnancy.Conceptualization, P.H., M.B. (Marie Blomberg), P.N., K.P., M.L.; Methodology, P.H., E.S., M.B. (Marie Blomberg), P.N., K.P., M.L.; Formal Analysis, P.H., M.B. (Marcus Bendtsen); Investigation, P.H., M.B. (Marie Blomberg), K.P., P.N., M.L.; Writing—Original Draft Preparation, P.H., E.S., M.L.; Writing—Review & Editing, M.B. (Marie Blomberg), P.N., K.P., K.T., M.B. (Marcus Bendtsen), F.E.-L.; Visualization, P.H.; Funding Acquisition, P.H., M.L. All authors approved the final version of the manuscript.This study was supported by grants from the Swedish Research Council for Health, Working Life and Welfare (Forte, 2017-00088 to P.H.; 2018-01410 to M.L.) as well as the Swedish Research Council (Vetenskapsrådet, 2016-01147 to M.L.). F.E.L. was supported by a grant from the European Union’s Horizon 2020 research and innovation programme under the Marie Skłodowska-Curie grant agreement no. 707404.The authors thank all midwives involved in the data registration as well as the Swedish Pregnancy Registry for excellent data management.M.B. (Marcus Bendtsen) owns a private company that develops and distributes lifestyle interventions to be used in health care settings although this company has no link to this study. All remaining authors declare no competing interests.Odds of poor self-rated health (A) before pregnancy before pregnancy (n = 577,765), (B) during pregnancy (n = 433,248), and (C) after pregnancy (n = 426,424) by birth regions. Logistic regression was utilized to calculate odds ratios with 95% confidence intervals (women born in Sweden was the reference group). Adjusted models included maternal age, parity, educational attainment and body mass index as covariates.Descriptive data of the women in this study.
+              
+                
+                  Data is presented as the mean ± standard deviation or as %. BMI, body mass index.
+                
+                
+                  1 Included regions/countries were: Central Asia (Armenia, Azerbaijan, Georgia, Kazakhstan, Kyrgyzstan, Mongolia, Tajikistan, Turkmenistan and Uzbekistan); Central Europe (Albania, Bosnia and Herzegovina, Bulgaria, Croatia, Czech Republic, Hungary, Macedonia, Montenegro, Poland, Romania, Serbia, Slovakia and Slovenia); Eastern Europe (Belarus, Estonia, Latvia, Lithuania, Moldova, Russia and Ukraine).
+                
+                
+                  2 Included countries were: Australasia (Australia and New Zealand); high-income Asia Pacific (Brunei, Japan, Singapore and South Korea); high-income North America (Canada, Greenland and USA); Southern Latin America (Argentina, Chile and Uruguay); Western Europe (Andorra, Austria, Belgium, Cyprus, Denmark, Finland, France, Germany, Greece, Iceland, Ireland, Israel, Italy, Luxembourg, Malta, Netherlands, Norway, Portugal, Spain, Switzerland and UK).
+                
+                
+                  3 Included regions/countries were: Andean Latin America (Bolivia, Ecuador, Peru); the Caribbean (Antigua and Barbuda, The Bahamas, Barbados, Belize, Bermuda, Cuba, Dominica, Dominican Republic, Grenada, Guyana, Haiti, Jamaica, Puerto Rico, Saint Lucia, Saint Vincent and the Grenadines, Suriname, Trinidad and Tobago and Virgin Islands); Central Latin America (Colombia, Costa Rica, El Salvador, Guatemala, Honduras, Mexico, Nicaragua, Panama and Venezuela); Tropical Latin America (Brazil and Paraguay).
+                
+                
+                  4 Included regions/countries were: Afghanistan, Algeria, Bahrain, Egypt, Iran, Iraq, Jordan, Kuwait, Lebanon, Libya, Morocco, Oman, Palestine, Qatar, Saudi Arabia, Sudan, Syria, Tunisia, Turkey, United Arab Emirates and Yemen.
+                
+                
+                  5 Included regions/countries were: Bangladesh, Bhutan, India, Nepal and Pakistan.
+                
+                
+                  6 Included regions/countries were: East Asia (China, North Korea and Taiwan (province of China)); Southeast Asia (Cambodia, Indonesia, Laos, Malaysia, Maldives, Mauritius, Myanmar, Philippines, Sri Lanka, Seychelles, Thailand, Timor-Leste and Vietnam). Oceania was not included due too few observations.
+                
+                
+                  7 Included regions/countries were: Central Sub-Saharan Africa (Angola, Central African Republic, Democratic Republic of the Congo, Equatorial Guinea and Gabon); Eastern Sub-Saharan Africa (Burundi, Comoros, Djibouti, Eritrea, Ethiopia, Kenya, Madagascar, Malawi, Mozambique, Rwanda, Somalia, South Sudan, Tanzania, Uganda and Zambia); Southern Sub-Saharan Africa (Botswana, Lesotho, Namibia, South Africa, Swaziland (eSwatini) and Zimbabwe); Western Sub-Saharan Africa (Benin, Burkina Faso, Cameroon, Cape Verde, Chad, Côte d’Ivoire, The Gambia, Ghana, Guinea, Guinea-Bissau, Liberia, Mali, Mauritania, Niger, Nigeria, São Tomé and Príncipe, Senegal, Sierra Leone and Togo).
+                
+                
+                  8 Number of pregnancies with data on self-reported health during pregnancy were; Sweden (n = 354,242); Central Europe, Eastern Europe and Central Asia (n = 18,859); high income countries (n = 11,652), Latin America and the Caribbean (n = 2456), North Africa and the Middle East (n = 25,257), South Asia (n = 3206), Southeast Asia and East Asia (n = 6830), Sub-Saharan Africa (n = 10,746).
+                
+                
+                  9 Number of pregnancies with data self-reported health after pregnancy were; Sweden (n = 347,562), Central Europe, Eastern Europe and Central Asia (n = 18,783), high income countries (n = 11,587), Latin America and the Caribbean (n = 2448), North Africa and the Middle East (n = 25,253), South Asia (n = 3200), Southeast Asia and East Asia (n = 6812), Sub-Saharan Africa (n = 10,779).
+                
+              
+            Data is presented as the mean ± standard deviation or as %. BMI, body mass index.1 Included regions/countries were: Central Asia (Armenia, Azerbaijan, Georgia, Kazakhstan, Kyrgyzstan, Mongolia, Tajikistan, Turkmenistan and Uzbekistan); Central Europe (Albania, Bosnia and Herzegovina, Bulgaria, Croatia, Czech Republic, Hungary, Macedonia, Montenegro, Poland, Romania, Serbia, Slovakia and Slovenia); Eastern Europe (Belarus, Estonia, Latvia, Lithuania, Moldova, Russia and Ukraine).2 Included countries were: Australasia (Australia and New Zealand); high-income Asia Pacific (Brunei, Japan, Singapore and South Korea); high-income North America (Canada, Greenland and USA); Southern Latin America (Argentina, Chile and Uruguay); Western Europe (Andorra, Austria, Belgium, Cyprus, Denmark, Finland, France, Germany, Greece, Iceland, Ireland, Israel, Italy, Luxembourg, Malta, Netherlands, Norway, Portugal, Spain, Switzerland and UK).3 Included regions/countries were: Andean Latin America (Bolivia, Ecuador, Peru); the Caribbean (Antigua and Barbuda, The Bahamas, Barbados, Belize, Bermuda, Cuba, Dominica, Dominican Republic, Grenada, Guyana, Haiti, Jamaica, Puerto Rico, Saint Lucia, Saint Vincent and the Grenadines, Suriname, Trinidad and Tobago and Virgin Islands); Central Latin America (Colombia, Costa Rica, El Salvador, Guatemala, Honduras, Mexico, Nicaragua, Panama and Venezuela); Tropical Latin America (Brazil and Paraguay).4 Included regions/countries were: Afghanistan, Algeria, Bahrain, Egypt, Iran, Iraq, Jordan, Kuwait, Lebanon, Libya, Morocco, Oman, Palestine, Qatar, Saudi Arabia, Sudan, Syria, Tunisia, Turkey, United Arab Emirates and Yemen.5 Included regions/countries were: Bangladesh, Bhutan, India, Nepal and Pakistan.6 Included regions/countries were: East Asia (China, North Korea and Taiwan (province of China)); Southeast Asia (Cambodia, Indonesia, Laos, Malaysia, Maldives, Mauritius, Myanmar, Philippines, Sri Lanka, Seychelles, Thailand, Timor-Leste and Vietnam). Oceania was not included due too few observations.7 Included regions/countries were: Central Sub-Saharan Africa (Angola, Central African Republic, Democratic Republic of the Congo, Equatorial Guinea and Gabon); Eastern Sub-Saharan Africa (Burundi, Comoros, Djibouti, Eritrea, Ethiopia, Kenya, Madagascar, Malawi, Mozambique, Rwanda, Somalia, South Sudan, Tanzania, Uganda and Zambia); Southern Sub-Saharan Africa (Botswana, Lesotho, Namibia, South Africa, Swaziland (eSwatini) and Zimbabwe); Western Sub-Saharan Africa (Benin, Burkina Faso, Cameroon, Cape Verde, Chad, Côte d’Ivoire, The Gambia, Ghana, Guinea, Guinea-Bissau, Liberia, Mali, Mauritania, Niger, Nigeria, São Tomé and Príncipe, Senegal, Sierra Leone and Togo).8 Number of pregnancies with data on self-reported health during pregnancy were; Sweden (n = 354,242); Central Europe, Eastern Europe and Central Asia (n = 18,859); high income countries (n = 11,652), Latin America and the Caribbean (n = 2456), North Africa and the Middle East (n = 25,257), South Asia (n = 3206), Southeast Asia and East Asia (n = 6830), Sub-Saharan Africa (n = 10,746).9 Number of pregnancies with data self-reported health after pregnancy were; Sweden (n = 347,562), Central Europe, Eastern Europe and Central Asia (n = 18,783), high income countries (n = 11,587), Latin America and the Caribbean (n = 2448), North Africa and the Middle East (n = 25,253), South Asia (n = 3200), Southeast Asia and East Asia (n = 6812), Sub-Saharan Africa (n = 10,779).

Med-MDPI/jcm_2/jcm-09-06-01765.txt

@@ -0,0 +1 @@
+The interaction between the bone marrow microenvironment and plasma cells plays an essential role in multiple myeloma progression and drug resistance. The vascular endothelial growth factor (VEGF)/VEGF receptor (VEGFR) pathway in vascular endothelial cells activates and promotes angiogenesis. Moreover, VEGF activates and promotes vasculogenesis and vasculogenic mimicry when it interacts with VEGF receptors expressed in precursor cells and inflammatory cells, respectively. In myeloma bone marrow, VEGF and VEGF receptor expression are upregulated and hyperactive in the stromal and tumor cells. It has been demonstrated that several antiangiogenic agents can effectively target VEGF-related pathways in the preclinical phase. However, they are not successful in treating multiple myeloma, probably due to the vicarious action of other cytokines and signaling pathways. Thus, the simultaneous blocking of multiple cytokine pathways, including the VEGF/VEGFR pathway, may represent a valid strategy to treat multiple myeloma. This review aims to summarize recent advances in understanding the role of the VEGF/VEGFR pathway in multiple myeloma, and mainly focuses on the transcription pathway and on strategies that target this pathway.Multiple myeloma (MM), a hematological cancer, accounts for about 1% of all human tumors, and is characterized by the infiltration of rich bone marrow (BM) by mature plasma cells (PCs) that produce monoclonal immunoglobulins [1,2,3]. The clonal PCs produce and release cytokines that are responsible for the typical clinical manifestation of the disease: (i) bone resorption (lytic lesions, hypercalcemia, bone pain) caused by alteration in the activity of osteoclasts/osteoblasts; (ii) anemia caused by modification of the maturation and differentiation of erythroblasts; (iii) renal insufficiency due to Ig light chain deposition; (iv) hypercalcemia and hyperuricemia; and (v) hyper-viscosity syndrome caused by high circulating protein levels [4].MM progression is accompanied by and strictly dependent on changes in the microenvironment of the BM [5,6]. These modifications of the microenvironment induce a permissive environment that protects and stimulates plasma cell survival and proliferation [5,6,7,8]. The interaction of MM PCs with BM stromal cells (SCs) and extracellular matrix (ECM) components in the BM microenvironment is mediated by a plethora of autocrine and paracrine cytokine loops, as well as direct cell–cell and cell–ECM interactions. These direct and indirect interactions result in the activation of multiple signaling pathways that are responsible for modifications in the microenvironment during MM progression [9] and that are responsible for MM plasma cell apoptosis inhibition, survival, proliferation, and invasion. The expansion of neoplastic PCs in BM causes bone lysis and promotes microenvironment modulation and neovessel formation [10,11,12].MM-associated microenvironment modifications include BM neovessel formation for assembling the “vascular niche” and bone cell activation for the constitution of the “osteoblastic niche” [13]. In these two specialized niches, myelomatous PCs grow, survive, and are protected from external attacks [13,14]. The alterations occurring in these niches represent predisposing events that facilitate the survival and expansion of neoplastic PCs. Moreover, the cells that comprise these specialized niches contribute to protecting MM PCs from the aggression of chemotherapy and immunological cells. The elucidation of key niche-associated pathways, including the primary driver of mutations in BM stromal cells, the role of hypoxia, angiogenesis, and inflammation can increase our knowledge of immune evasion and activation of survival pathways, and could indicate ways to improve modern therapeutic approaches [14].The development of a new vascular tree in the BM of MM patients is a pathologic process in which angiogenesis (the formation of new vessels from existing ones) [9], vasculogenesis (the formation of new vessels from endothelial precursors) [15], and vasculogenic mimicry (the completion of neovessels by other non-endothelial cells [ECs]) [16] work simultaneously for the constitution of the vascular niche [9]. BM neovascularization is related to the MM stage, disease progression, and patients’ response to therapy and survival [15,16,17,18,19,20].Taken together, these processes lead to modifications in the BM microenvironment and its controllers (activated cells, cytokines, and their autocrine and paracrine loops, signaling pathways), which are useful targets in the treatment of MM, for example, the direct targeting of MM PCs [5,21,22].The components of the BM microenvironment (SCs and ECM) surround MM PCs and support them by direct cell–cell and cell–ECM interaction, and by the production of cytokines and growth factors [23,24].BM ECs express adhesion molecules and receptors on their cell surface, which is characteristic of a typical “activated” phenotype [25]. This activated phenotype is related to a specific genotype of MM BM ECs [26], unlike those of monoclonal gammopathy of undetermined significance (MGUS) or normal resting ECs. Phenotypic and genotypic activation causes the quick and uncontrolled proliferation of ECs, and angiogenesis self-maintenance [27,28,29,30,31]. Activated MM ECs modulate the expression of receptors, increasing VEGF receptor (VEGFR)-2, tyrosine-protein kinase Met (cMet, also called hepatocyte growth factor receptor), fibroblast growth factor receptor (FGFR), and Tie2/Tek density, integrins, and other adhesion molecules responsible for adhesion to the ECM components and cell motility. Moreover, integrin-activated signaling, particularly the β3-integrin pathway, sustains cell survival, proliferation, migration, and capillarogenesis. The interaction of MM PCs and activated ECs is mediated by endoglin and favors PCs entrance into neovessels. Finally, MM ECs express aquaporin 1 on their membrane, which is a water transporter that is responsible for plasma extravasation and increases interstitial pressure by increasing vascular permeability [25,32,33]. Aquaporin 1 also upregulates hypoxia-inducible Factor-1 alpha (HIF-1α) and vascular endothelial growth factor (VEGF) because of hypoxia induction [25,32,33].The recruitment of progenitor cells to differentiate into a subset of mature ECs is demonstrated by the expression on their surface of the CD133 staminal antigen, which shows active vasculogenesis [7,29,30,31,32,33,34,35,36,37,38,39]. The down or upregulation of some genes like tumor cells and cells with a well-defined proteomic signature render MM ECs different from MGUS and other resting ECs, characterizing their over-angiogenic, and transformed phenotype [34].The presence of fibroblasts, also called “activated myofibroblasts” or “cancer-associated fibroblasts”(CAFs) that are derived from cells undergoing the endothelial-mesenchymal or mesenchymal transition has been demonstrated in the BM of patients with active MM [6,40,41,42,43]. The precursors of CAFs are resident fibroblasts and progenitor cells. BM infiltrating fibroblasts promote neovessel formation by the secretion of high amounts of growth factors and angiogenic cytokines, such as VEGF-A, FGF-2, tumor necrosis factor-alpha (TNF-α), urokinase, and matrix metalloproteinases–MMPs [7,9,13,40,41,42,43]. Moreover, fibroblasts synthesize inducible nitric oxide synthase, which increases blood flow [41].Immune changes during MM progression play an essential role in the BM microenvironment. The distribution and phenotypic and behavioral features of the natural killer cells, cytotoxic and other T cells, monocytes, and macrophages play an essential role in the progression from premalignant MM to active disease. A precise understanding of these modifications will help to develop new therapeutic strategies as well as accurate patient stratification [44].Circulating monocytes are recruited to differentiate into active macrophages in the BM of MM patients through the activation of VEGF-A/VEGFR-1 and FGF-2/FGFR-1, -2, and -3 pathways. Here they acquire phenotypic and functional adaptations that contribute to the completion of neovessel walls through vasculogenic mimicry [16]. They also secrete VEGF and FGF-2, which play a role in promoting and auto-maintaining angiogenesis, vasculogenesis, and vasculogenic mimicry. Mast cells are also recruited in the BM by stem cell factor (SCF), FGF-2, VEGF-A, and platelet-derived growth factor (PDGF) secreted by MM PCs and other SCs [18,38,39]. Mast cells release angiogenic factors (VEGF-A, FGF-2, transforming growth factor (TGF)-β, TNF-α, interleukin (IL)-8, heparin, histamine, tryptase, and chymase) by degranulation [38,45]. Moreover, like macrophages, mast cells contribute to vasculogenesis mimicry by phenotypic adaptation [38,45,46]. They maintain typical tryptase-positivity and acquire endothelial cell-like properties. In the BM of MM patients, mast cells interact physically with ECs participating in the completion of the vascular wall [38,45,46].MM PCs unbalance the osteoblasts/osteoclasts equilibrium and this induces bone disease [13,47,48,49]. In the BM niches, there is an interplay between MM PCs, osteoclasts, and vascular ECs [13]. This close link induces a vicious circle between angiogenesis, bone lysis, and favors BM myeloma cells survival and expansion [13,48,49].Hematopoietic stem cells (HSCs–CD133+) and endothelial progenitor cells (EPCs-CD146+ CD105+ CD34+) residing in the BM and recruited from other sites, are induced to differentiate into mature ECs by angiogenic cytokines (VEGF-A, FGF-2, insulin-like growth factor (IGF-1)) released in the BM. This biological process, known as vasculogenesis, is an ancestral process that is active during prenatal life, and has been demonstrated to be reactivated in the BM of MM patients with active disease [50,51,52,53]. Mesenchymal stem cells (MSCs) are one of the main components of BM niches [15,54,55,56,57]. Their origin remains unclear, but MSCs can differentiate into fibroblasts, adipocytes, chondrocytes, and osteoblasts [54,56,57,58]. MSCs support MM plasma cell growth, survival, and immune system evasion [54,55,56]. Lastly, high amounts of tumor-infiltrating adipocytes have been found in the BM of MM patients [59,60]. BM adipocytes sustain angiogenesis by secretion of VEGF, Angiopoietin (Ang)-1 and -2, leptin, adiponectin, TNF-α, FGF, TGF-β, hepatocyte growth factor (HGF), IL-6, and IL-8 [59,60]. The hypoxic environment of MM BM further stimulates tumor-infiltrating adipocytes to produce angiogenic factors, particularly VEGF-A, which have been demonstrated to promote disease progression through paracrine and autocrine loops in the BM of MM patients [57,58,59,60].BM neovessels in MM patients have an impaired structure [15]. The vascular tree is formed by highly permeable and fenestrated thin, tortuous, and arborized blood vessels. The ECs that line newly formed vessels contain cytoplasmic vesicles and inclusions, broaden intercellular junctions, and are supported by an incomplete and highly permeable basement membrane [15]. These alterations are a result of microenvironment activation and stimulation induced by myeloma PCs, which promotes three different processes: (i) angiogenesis, (ii) vasculogenesis, and (iii) vasculogenic mimicry [9].BM microvessel density is remarkably higher in patients with active MM compared to those with MGUS or healthy subjects and neovessel density correlates with the disease stage, shrinking during the remission/plateau phase, increasing in density in the diagnosis, relapse/refractory phase, and reaching maximum expansion in the leukemic phase [9,11,32,58]. Angiogenesis, that is, the rise of neovessels from already formed and stable ones, is the primary, active biological process in the BM of MM patients [9,11,32,58]. Angiogenesis is essential for cancer cell proliferation, spread, and diffusion. Angiogenesis starts when the preneoplastic disease (MGUS), which is characterized by plasma cell growth equilibrium and poor vascularization (avascular phase), and progresses to active proliferation and spread of PCs with uncontrolled and unlimited neovessel formation (the vascular phase). MM progression is related to the activation of many oncogenes (i.e., c-myc, c-fos, c-jun, ets-1, etc.), which is induced by immunoglobulin translocations and genetic instability [12,32,57]. The activation of angiogenesis in this phase is due to the release of angiogenic cytokines, such as VEGF-A, FGF-2, HGF, etc., by transformed PCs [12,32,58].Like angiogenesis, vasculogenesis contributes to neovascularization in the BM of MM patients [28,29,30]. This biological process is active during prenatal life, where it induces the primary development of the yolk sac vasculature, the heart, and the dorsal aortae during embryogenesis [27]. The aggregation of EPCs (angioblasts) in a primitive capillary plexus stored in the mesoderm and their differentiation into mature ECs are the basis for the formation of the new vascular system [27]. During neovascularization in the BM of MM patients, this ancestral biological process is reactivated to release angiogenic cytokines into the microenvironment. It has been demonstrated that peripheral blood-purified CD34+ VEGFR-2+ cells from MM patients exposed to VEGF-A, FGF-2, and IGF, in appropriate culture conditions, acquire a typical morphology (spindle-shaped) and change their phenotype to acquire typical EC markers such as factor VIII-related antigen (FVIII-RA), vascular endothelial-cadherin (VE-cadherin), CD34, CD31, VEGFR-2, Tie-2/Tek, and E-selectin [29]. These in vitro results have been confirmed in the BM of MM patients where some neovessel forming ECs express the CD133 staminal antigen on their surface, like the typical EC markers [29]. The expression of the CD133 staminal antigen was absent in MGUS patients. Moreover, several studies have demonstrated the presence of circulating EPCs in MM patients [28,29]. Vasculogenesis mimicry is another pathological process of neovascularization that is found in aggressive solid and hematologic tumors. In this particular process, malignant cells and/or other non-ECs, such as inflammatory cells, acquire the capacity to assemble a vessel-like network [16,30,38,39,59]. The Vasculogenic mimicry is a process in which epithelial-mesenchymal transition plays a very important role. Cancer cells, cancer stem cells, inflammatory cells, tumor-associated fibroblasts, under the stimulus of angiogenic cytokines (FGF-2, VEGF-A, HGF, IGF), inflammatory cytokines (IL-6 and -8, TGF-α and -β, TNF-α, CCL-2), and other signaling pathways (stromal cell-derived factor 1 (SDF1), cyclooxygenase-2 (COX-2), Twist, E-cadherin, EPH receptor A2 (EphA2)-phosphatidylinositol 3-kinase (PI3K), Wnt, and Notch signaling pathway) undergo an epithelial–mesenchymal transition by downregulation of E-cadherin, zonula occludins-1, and α-catenin, and upregulation of VE-cadherin, fibronectin, cadherin-2, and vimentin. HIFs and hypoxia-responsive elements (HREs) play an important role in this context. In MM, inflammatory cells (i.e., macrophages and mast cells) isolated from the BM of patients with active disease contribute to BM neovascularization by vasculogenic mimicry. In appropriate culture conditions, when these cells are exposed to angiogenic cytokines (VEGF and FGF-2), they modify their phenotype and acquire biologic properties similar to those of ECs producing functional capillary-like structures in vitro [16,30,38,39,59]. Vasculogenic mimicry mediated by inflammatory cells has been demonstrated in the BM biopsies of MM patients [16,38].VEGF is a member of a family of six structurally related proteins, namely, VEGF-A, -B, -C, -D, -E (viral factor), and PDGF [61,62]. Their activity is mediated by interaction with the relative receptor and mediate angiogenesis (VEGF-A, -E/VEGFR-2-neuropilin (NRP)-1, -2), or lymphangiogenesis (VEGF-C, -D/VEGFR-2, -3) [58,59]. The role of the bind VEGF-B/VEGFR-1 is unclear (Figure 1).Ferrara and Henzel first purified VEGF, a glycoprotein with mitogenic activity for vascular ECs, from bovine pituitary folliculo-stellate cells [62,63]. The VEGF family is composed of six members: VEGF-A, -B, -C, -D, -E, and placental growth factor (PlGF) that recognize and specifically bind three different transmembrane receptor tyrosine kinases, which are VEGFR-1, -2 and -3 (Figure 1). The interaction between VEGFs and VERFRs promotes EC activation and endothelium regeneration (Figure 2). Moreover, VEGF-A, -B, and PlGF/VEGFR-1 interaction increase vascular permeability (Figure 2). The activation of VEGFR-2 (KDR or Flk-1 is predominantly expressed in vascular ECs) by interaction with its ligands (VEGF-A, -C, -D) mediates VEGF-dependent angiogenesis and increases vessel permeability [63,64]. VEGFR-3 is restricted to lymphatic ECs, and the binding of VEGF-C and VEGF-D mediates the regulation of lymphangiogenesis (Figure 2) [64,65].The tyrosine kinases’ VEGFRs are differentially expressed in various cell types [66,67,68,69]. The main member of the family, VEGFR-2, is prevalently expressed in ECs, neuronal, megakaryocytes, and hematopoietic stem cells [68], VEGFR-1 is expressed on the hematopoietic cell surface, monocytes/macrophages and smooth muscle cells [66,67]. The heterodimerization of the VEGFR-1 and -2 subunit mediates the binding of VEGF-B and PlGF (Figure 2) [67]. The activation of VEGFR-2 by VEGF-A on ECs activates a tyrosine kinase pathway, via protein kinase C (PKC) or the Ras protein, that terminally activates the mitogen-activated protein kinase (MAPK) system [69]. Differently from other cytokines (i.e., fibroblast growth factor–FGF-2, and PDGF), VEGF-A shows specific activity in vascular ECs, which are a specific growth factor for angiogenesis. Other cytokines act on different cytotypes and have several effects on biological activity that is different from angiogenesis. Other activities modulated by VEGF-A include the stimulation/differentiation of hemopoietic stem cells, ECM remodeling, and modulation of inflammatory cytokines [30,69,71]. VEGF-A secretion has been observed in several leukemic cell lines as well as in primary cancer cells, including MM cells [25,72,73]. Moreover, in these cells, paracrine and autocrine loops have been reported and shown to modulate the malignant behavior of hematological cancer cells [70,71,72,73]. In MM cells, as in other leukemic cells, VEGF-A activates its specific receptor expressed by EC, inducing the production of growth factors that stimulate leukemic cells to proliferate and induce drug resistance [74,75,76,77]. Moreover, the constitutive activation of VEGFR-2 on ECs of BM-derived from MM patients has been described previously [25,26,72,73].Some metabolic factors, such as hypoxia and hypoglycemia, regulate the production and release of VEGF-A [78]. HIF-1 is a crucial angiogenesis triggering factor because it induces the expression of VEGF mRNA [79,80]. The nuclear-stabilized HIF-1α forms an active complex with p300 and phosphorylated-STAT3. This active complex induces the activation of the VEGF promoter upregulating RNA polymerase II-dependent VEGF transcription [78,79,80,81]. An increase in VEGF-A production and secretion by lysophosphatidic acid (LPA) through the activation of c-Jun N-terminal kinase (JNK) and nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) has been reported [82]. Several studies indicate that angiogenesis is strictly dependent on Akt/NF-κB activation [83,84]. In hematopoietic cancer cells, the inhibitors of the NF-κB pathway induce VEGF secretion inhibition, thus reducing EC proangiogenic activities [85]. NF-κB is also activated by the PI3K/Akt signaling pathway, suggesting that PI3K/Akt activation plays a role in angiogenesis and leukemia progression [86].VEGF expression is also modulated by several transcription factors, such as activator protein-1 (AP-1), NF-κB, and stimulatory protein-1 (SP-1) [87,88,89]. Among those, AP-1 is a critical factor in the modulation of VEGF-A gene transcription in several hematopoietic and solid cancer cells [90,91,92].Published data indicate that VEGF-A directly stimulates the proliferation of cancer cells. In MM, VEGF mainly stimulates plasma cell migration, proliferation, and survival via autocrine and paracrine VEGF-A/VEGFR-2 loops (Figure 2) [9,93]. Also, leukemic cell resistance to apoptosis induced by serum deprivation is caused by VEGF-A by means of the expression of heat shock protein 90 (Hsp90), which binds B-cell lymphoma 2 (Bcl-2) and apoptotic protease activating factor-1 (Apaf-1) [94]. Moreover, EC exposure to VEGF-A induces the production and release of several hematopoietic growth factors including granulocyte-colony stimulating factor (G-CSF), granulocyte-macrophage colony-stimulating factor (GM-CSF), stem cell factor (SCF) and IL-6, which act as growth factors for hematopoietic cancers including MM [94,95]. It has been reported that VEGF-A expression is related to disease stage and tumor burden [96,97,98].The migration of MM PCs during disease progression as well as of ECs during angiogenesis is related to ECM lytic enzymes, including matrix metalloproteinase (MMP)-2 and -9, and urokinase-type plasminogen activator (uPA) [98]. The production of ECM lytic enzymes by PCs and BM SCs is modulated by VEGF-A/VEGFR-2 engagement and activation of relative tyrosine kinase pathway [75,98,99].VEGF-C is implicated in tumor lymphangiogenesis by activation of its specific receptor VEGF-R3 expressed in lymphatic ECs (Figure 2) [65]. VEGF-C acts by c-Jun binding and activation of the promoter of the cyclic adenosine 3′,5′-monophosphate-response element of the cyclooxygenase-2 (COX-2) inducing COX-2 expression [100,101,102]. COX-2 enhances cancer cell survival and proliferation and inhibits anti-tumor immunity. COX-2 upregulation is also due to VEGF-R2 activation, which induces p38 MAPK/JNK signaling pathways in human vascular ECs [101]. On the other hand, VEGF-A may also influence the upregulation of COX-2 in cancer cells. So, cancer tumor angiogenesis may be increased through the induction of COX-2/prostanoids produced by VEGFs (Figure 2) [102].Improved outcomes have been obtained for MM patients thanks to the addition of “target” biological drugs active in the microenvironment to the therapeutic strategies of MM [9,103,104,105].Antiangiogenic activity, as well as the inhibition of BM SCs, has been demonstrated for proteasome inhibitors, immunomodulators (IMIDs), bisphosphonates, monoclonal antibodies, tyrosine kinase inhibitors, PI3K/Akt-MEK/ERK pathway inhibitors, focal adhesion kinase (FAK) inhibitors, interleukin inhibitors, farnesyltransferase inhibitors, and marine cartilage extract [106,107,108,109,110].Several preclinical studies have described the antiangiogenetic activity of proteasome inhibitors (PIs). The first-generation PI, bortezomib inhibits the proliferation and migration of human umbilical venous endothelial cells (HUVEC), EC lines, and primary isolated BM ECs from MM patients. It downregulates several angiogenic cytokines, such as VEGFs, FGF-2, IGF-1, interleukins, and the expression of angiopoietins by MM PCs, ECs, and other SCs (fibroblasts and inflammatory cells) [111,112]. Moreover, bortezomib induces EC apoptosis, reduces the proliferation and motility of MM ECs, and inhibits neovessel formation in vitro [113,114]. The antiangiogenic activities of bortezomib are mediated by the inhibition of IKB degradation and the consequent block of NF-κB activity (Figure 3) [115,116]. Moreover, the block of NFκB causes inhibition of the expression and release of VEGF-A mediated by the stabilization of HIF-1α [117] and the CXC chemokines-mediated signaling [118]. In experimental in vivo systems, bortezomib inhibits neovascularization in both chick embryo chorioallantoic membrane in a dose-dependent fashion and PC-tumors in human plasmacytoma xenograft mouse [119,120,121]. In bortezomib-treated MM patients, the reduction in BM microvessel density and the reduction in circulating angiogenic cytokines levels have been demonstrated [122,123,124].The primary target of PIs is the ubiquitin-proteasome system involved in the regulation of cellular homeostasis, cell death, and angiogenesis, both carfilzomib, a second-generation PI and ixazomib, the first orally available PI, have shown the same antiangiogenic activity [125,126].Thalidomide, the first immunomodulatory drug, has been demonstrated to have antiangiogenic properties that increase its efficacy as an anticancer drug. Various metabolites of thalidomide have antiangiogenic activities [127,128,129]. They inhibit tubulin cytoskeletal rearrangement and filopodial formation, thus causing a reduction in the proliferation, migration, and tube formation of stimulated ECs [130]. Thalidomide and its derivatives, cause microtubules depolymerization by binding the same site of vinblastine on tubulin. Moreover, they inhibit the reassembly of microtubules, alter the dynamics of individual microtubules, decrease the growth rates, and shorten the excursions. These are the bases of the antimitotic effect of thalidomide and its derivatives because the reduction in the dynamicity of microtubules blocks cell mitosis. [131,132]. In chick and rabbit embryo limbs as well as in zebrafish embryos, thalidomide causes loss of FGF and Shh signaling [133,134,135]. Thalidomide’s antiangiogenic action is based on the destruction of vessels without smooth muscle, which typically creates a newly formed cancer vascular tree [133]. Moreover, thalidomide-induced vessel loss is due to nitric oxide expression inhibition [136,137,138], VEGF receptor depletion [139], and destruction of FGF-induced blood vessels in rodent and rabbit corneal assays [140,141]. Finally, thalidomide modulates the expression and activity of actin and tubulin, integrins, vascular endothelial growth factor, PDGFβ, nitric oxide, ceramide, angiopoietins, Notch, HIF, Slit2/Robo signaling and ROS, all molecules that are fundamental in neovessel development [142,143,144,145,146,147].The second-generation IMiD, lenalidomide, inhibits the VEGF-induced PI3K-Akt signaling pathway (Figure 3) and HIF-1 expression [148], induces apoptosis of tumor cells, blocks the activity of TNF, and modulates T cells and NK cells activities [149,150,151,152,153]. Moreover, it inhibits MM PCs/stromal cell interaction by blocking cell adhesion [149,150,151,152,153]. It has been demonstrated that lenalidomide inhibits the proliferation and migration of ECs in patients with active MM by downregulating angiogenesis-related key genes and proteins [154]. Finally, lenalidomide inhibits in vitro neovessel formation in the Matrigel assay and in vivo PCs-induced angiogenesis in the chorioallantoic membrane (CAM) assay [154]. The anti-angiogenic effect of lenalidomide has been suggested in treated MM patients, but no significant reduction in BM neovascularization has been demonstrated [155]. Pomalidomide, like thalidomide and lenalidomide, inhibits experimental angiogenesis in MM ECs by targeting VEGF and HIF-1 [156]. Bisphosphonates, are antiresorptive drugs used in bone disease treatment, that have antiangiogenic activity [157]. Zoledronic acid in therapeutic doses, inhibits ECs proliferation, chemotaxis, tube forming in vitro and MM plasma cell and EC-induced angiogenesis in the in vivo CAM assay [157,158,159]. Synergistic activities in BM macrophages from MM patients have been demonstrated for bortezomib and zoledronic acid [158]. The association of these two drugs inhibits the in vitro macrophage expression of FVIII-RA, Tie2/Tek, and VEGFR-2/VE-cadherin, proliferation, adhesion, migration, and tube formation in the Matrigel assay. Moreover, bortezomib and zoledronic acid synergistically inhibit the production and release of angiogenic cytokines, VEGFR-2 expression, and phospho-activation, ERK1/2 phosphorylation, and NF-κB activation induced by VEGF-2/FGF-2 stimulation [158]. The inhibition of these biological activities blocks macrophage trans-differentiation into endothelial-like cells and inhibits the vasculogenic mimicry process [158,159].Evidence of the role of VEGF and its receptors in cancer progression has resulted in the development of monoclonal antibodies targeting the VEGF/VEGFR complex. In preclinical studies, the inhibition of VEGF-mediated proangiogenic activity in BM ECs derived from patients with MM has been described [72,73]. In vivo, neutralizing anti-VEGF-A antibody (Figure 3), and more efficaciously, anti-VEGFR-2 antibody (Figure 3) inhibits the constitutive autophosphorylation of both VEGFR-2 and the associated extracellular signal-regulated kinase-2 (ERK-2), proliferation, and capillarogenesis ability in MM ECs [72,73].PTK787/ZK 222584 (PTK787), a molecule specifically designed to block the tyrosine kinase domain of VEGFR, inhibits constitutive and VEGF-induced receptor phosphorylation (Figure 3). In vitro, at clinically achievable concentrations, PTK787 can inhibit VEGF-induced cell proliferation, migration, growth, and survival, and overcome drug resistance of MM cells (cell lines and patient-derived) in the presence of Dex and IL-6 and/or cultured with BMSCs [160].Aplidin (originally isolated from the Mediterranean tunicate Aplidium Albicans) and its analogs inhibit cell proliferation, angiogenic sprouting, and neovessel formation in vitro of human ECs [161]. Moreover, they block VEGF-induced neovascularization in the in vivo CAM system [161].There is some evidence that VEGF/VEGFR and HGF/cMet signaling are both dysregulated in MM BM ECs and have a synergistic effect in the progression of the disease [22,162,163]. Following VEGF and HGF binding to their respective receptors, dimerization and autophosphorylation of VEGFR and cMet induce the recruitment of signaling proteins to the docking site. This leads to the activation of downstream pathways such as PI3K/AKT and Ras/ERK and translates into biological responses such as cell survival, growth, migration, proliferation, metabolism, and, finally, angiogenesis [22,77,164]. These observations support the evidence that blocking a single growth factor cannot suppress the entire process of angiogenesis and may be ineffective [165]. The dual inhibition of VEGF/VEGFR and HGF/c-Met signaling may produce more satisfactory results (Figure 3). Pazopanib (GW786034) is a second-generation multi-targeted tyrosine kinase inhibitor against vascular endothelial growth factor receptor-1, -2, and -3, platelet-derived growth factor receptor-alpha, platelet-derived growth factor receptor-beta, and c-kit (Figure 3). In vitro pazopanib modulates the expression of surface adhesion molecule (intercellular adhesion molecule-1 and vascular cell adhesion molecule-1). This inhibits MM plasma cell and BM SCs adhesion which causes a reduction in plasma cell proliferation and migration and an increase of apoptosis. Moreover, pazopanib inhibits neovessel formation in the Matrigel assay, increases HUVEC apoptosis, and sensitizes both MM PCs to melphalan and bortezomib [166]. Sorafenib, a multi-kinase inhibitor that acts predominantly through inhibition of Raf-kinase and VEGF receptor 2 (Figure 3), affects the BM microenvironment and its interaction with myeloma cells by inhibition of VEGF-induced tubule formation and downregulation of VEGF and IL-6 secretion [162,163,167].Derivatives of quinolone and quinazoline (i.e., gefitinib, erlotinib, foretinib, golvatinib, and cabozantinib) have long attracted attention because of their ability to inhibit a variety of tyrosine kinase, including c-Kit, c-Met, VEGFRs, epidermal growth factor receptor (EGFR), PDGFR, FGFR and so on [50,51] (Figure 3). So, quinolones and quinazolines are a series of promising anti-tumor compounds, especially in targeting dual c-Met and VEGFR tyrosine kinase. These small molecules are potent inhibitors of EC tubule formation, c-Met and VEGFR-2 phosphorylation, cellular invasion, and migration-disrupted tumor vasculature and promote tumor and EC death [168]. A new class of drugs includes a multi-domain designed ankyrin repeat protein (DARPin®), which simultaneously binds more target molecules. MP0250, a DARPin that simultaneously binds VEGF and HGF (Figure 3) has been investigated for its antiangiogenic activity in MM [169]. MPO250 reduces the ligand-mediated VEGFR-2 and cMET phosphorylation and inhibits the downstream signaling cascades of both receptors. In MM EC, MPO250 modulates the cytokine secretion profile and inhibits all their angiogenic activities [169]. Finally, MPO250, in combination with bortezomib, decreases the tumor burden and microvessel density in the syngeneic 5T33MM tumor model [169].Bortezomib is active in relapsed/refractory MM patients, and in those with poor-risk cytogenetics because of its dual activity on PCs and the BM stromal compartment [170]. In this MM setting, an increase in progression-free and overall survival has been obtained when compared with dexamethasone treatment alone. No significant adverse event has been found except for reversible peripheral neuropathy and thrombocytopenia. A synergistic effect, in combination with other anti-MM drugs, is the primary goal of bortezomib treatment [170].The major strength of thalidomide in MM patients treatment, and also for bortezomib, is dual activity on MM PCs (direct anti-MM activity) and the BM stromal compartment (immunomodulation, anti-angiogenic activity, and inhibition of the secretion of angiogenic cytokines) [146,171]. In the first trial (single-agent compassionate-use) which included 84 patients, 32% responded [172]. This led to several clinical trials in both relapsed/refractory and newly diagnosed patients [172,173,174,175,176,177], and thalidomide became the standard of care in MM patients. Also, the use of thalidomide as maintenance therapy following autologous stem cell transplantation (ASCT) has been investigated [178], and the results demonstrate a progression-free survival (PFS) benefit but no consistent overall survival (OS) benefit. Moreover, the prolonged use of thalidomide was generally limited due to toxicities [178].Multiple different agents have been combined with thalidomide (i.e., melphalan, cyclophosphamide, liposomal doxorubicin, bortezomib, carfilzomib, elotuzumab, intensive chemotherapy regimens such as D-PACE-dexamethasone, cisplatin, doxorubicin, cyclophosphamide, etoposide) with good results in term of PFS and OS, particularly with thalidomide-containing triplets [179,180,181,182,183,184,185,186,187,188,189,190,191,192,193,194,195,196,197,198,199]. Lenalidomide, the second-generation IMiD, has demonstrated few of the typical thalidomide side effects [200]. Its association with dexamethasone achieves overall response rates (ORR) of 48–61% in patients previously treated with thalidomide [201,202,203]. In two studies of newly diagnosed patients, lenalidomide with dexamethasone, had an ORR of 68–91% [204,205]. Moreover, the reduced dose (“low-dose”) of dexamethasone, in combination with lenalidomide, revealed an improved side effect profile with a better one-year overall survival rate [205]. Finally, in newly diagnosed patients, a significant PFS benefit with lenalidomide in combination with bortezomib and dexamethasone has been obtained [206]. More recently, the use of lenalidomide as maintenance therapy following ASCT has also been investigated, and showed a significant OS benefit [207,208,209]. Multiple different agents including cyclophosphamide [210,211,212], bendamustine [213,214], melphalan [215], liposomal doxorubicin [216,217], proteasome inhibitors (bortezomib, carfilzomib, and ixazomib) [218,219,220,221,222,223,224,225], histone deacetylases (HDAC) inhibitors (panobinostat and ricolinostat) [226,227], and monoclonal antibodies (elotuzumab, daratumumab, and pembrolizumab) [228,229,230,231,232] have been combined with lenalidomide with convincing results. Other trials evaluating the association of lenalidomide with new molecules are ongoing. The third member of the IMID class, pomalidomide, in combination with dexamethasone, has reported a 35–63% ORR with an ORR of up to 40% in lenalidomide-refractory patients and a 37% ORR in thalidomide-refractory patients [232,233,234,235,236]. As for other IMIDs, the association of pomalidomide with other classes of anti-MM drugs has been shown promising results [237,238,239,240,241,242,243,244].Bevacizumab is a humanized monoclonal anti-VEGF antibody with a high affinity for all human VEGF isoforms [245]. Two Phase II randomized trial with bevacizumab alone or in combination with anti-MM drugs have been conducted in the relapsed/refractory MM setting. In the first trial [246], which aimed to test the efficacy and safety of bevacizumab alone and in combination with thalidomide, only 12 patients, six per arm, were enrolled. Bevacizumab-associated adverse events were reported in three patients including grade 3 hypertension, fatigue, and hyponatremia. Regarding the hematological toxicity, only one grade 3 neutropenia was reported. The addition of thalidomide did not result in more toxicity with a grade 4 pulmonary hypertension in one patient, one grade 3 fatigue, and one grade 3 lymphopenia.The AMBER trial [247] was designed to evaluate the clinical benefit and tolerability of bevacizumab with bortezomib versus bortezomib alone and 102 patients were enrolled (49 and 53 patients, respectively). Toxicity was approximately equal in both treatment arms. Diarrhea and dehydration were more frequent in the bortezomib alone group, whereas in the bortezomib/bevacizumab group, more anemia, neutropenia, fatigue, upper respiratory tract infection, neuralgia, and hypertension were reported. No differences in peripheral neuropathy were observed.In both these studies, the addition of bevacizumab to anti-MM therapies did not result in a significant improvement in the outcome of patients. Two partial responses (PR) and three stable diseases (SD) have been achieved in the California Cancer Consortium trial [246].No significant differences in ORR were observed in the two treatment arms of the AMBER study [247], with 51% in the bortezomib/bevacizumab arm and 43.4% in the bortezomib alone arm. One patient per arm achieved a complete response (CR) with no significant difference in PR percentage across the two groups. The only very good partial response (VGPR) was seen in the bortezomib/bevacizumab arm (16.3% vs. 7.5% of patients). The median duration of response (DOR) was 6.9 months in the bortezomib/bevacizumab and 6.0 months in the bortezomib alone arm.These disappointing results are related to the role played by hypoxia and other active pro-angiogenic pathways in the BM microenvironment.In a phase II trial, sorafenib taken orally at a dose of 2 × 200 mg twice daily was administered to eleven relapsed/refractory MM patients up to the completion of 13 cycles or progression [248]. One patient achieved a PR, one patient an SD, seven patients experienced disease progression (DP), one patient died early because of sepsis, and one was lost to follow-up. The median PFS was 2.6 months. The specific drug-related adverse events were mild or moderate, with only one grade 4 toxicity (cardiac infarction) [248]. Twenty-three patients were enrolled in a more extensive phase II study of sorafenib for relapsed/refractory MM patients [249]. Patients received 400 mg sorafenib twice daily for 28-day treatment cycles until progression of the disease or unacceptable toxicity. No responses were observed in this study. Only two patients with MM have been enrolled in phase I/II trial of sorafenib combined with everolimus [250] No response was observed in the two enrolled MM patients.Vandetanib (ZD6474) is a vascular endothelial growth factor receptor (VEGFR)-2, epidermal growth factor receptor (EGFR), and rearranged during transfection (RET) tyrosine kinases inhibitor [251,252]. In a phase II study [246], vandetanib was well tolerated with only Grade 1–2 adverse events, except for rare cases of Grade 3 anemia. Despite its in vitro activity, and reduction of plasma levels of VEGF in treated MM patients, no responses or clinical benefits have been achieved.MPO250, a bispecific VEGF/HGF-targeting DARPin®, was administered with bortezomib and dexamethasone to 20 highly pretreated relapsed/refractory MM patients in the MiRRoR Study [253]. All 20 patients had prior exposure to IMiDs, and PIs and nine patients received PI-based regimens as their last treatment. Preliminary results showed an ORR of 45% with one CR, three VGPR, and five PR. The median duration of response was five months (range 2–24 months).In MM patients, the expression levels of VEGF in BM plasma and peripheral blood are strictly related to BM neovessel density and plasma cell infiltration. Moreover, these two parameters are correlated with the disease stage and the clinical outcome of patients. The importance of angiogenesis in MM is unquestionable, as well as the central role of VEGF in the survival, proliferation, and diffusion of PCs with paracrine and autocrine mechanisms. The interference of VEGF signaling represents a useful antiangiogenic approach in the treatment of MM. The employment of monoclonal antibody against VEGF/VEGFR and small molecule tyrosine kinase inhibitors plays a pivotal role in antiangiogenic therapy.Although impressive preclinical results in vitro and in vivo have been obtained, the inhibition of a single proangiogenic cytokine, and in particular the use of anti-VEGF drugs alone have not performed nearly as well in MM patients. Thus, the disparity between preclinical results and the clinical ones are a significant obstacle to the development of effective anti-VEGF therapy for MM patients. These disappointing results are probably due to the vicarious action of other cytokines and signaling pathways that are active in the BM microenvironment and can bypass the block of circulating VEGF or the path started by this cytokines. The constitutive activation of VEGFR may also contribute to making the anti-VEGF antibody ineffective in humans.Although possible modalities of resistance to blocking the VEGF/VEGFR pathway have been shown, the specific, direct molecular consequences of VEGF depletion on MM BM stromal cells, MM cell lines and primary MM PCs have also been demonstrated. This indicates that VEGF inhibition may be possible in MM patients by combiningantiangiogenic activity with the inhibition of VEGF/VEGFR signaling in the other BM stromal cells and on myeloma PCs.Although the clinical benefits of anti-VEGF monoclonal antibodies and other anti-VEGF target molecules in MM alone do not seem to be as good as was hoped, all the reported pieces of evidence justify further research into the potential of VEGF/VEGFR inhibition for the treatment of MM patients. Future studies should address strategies for multi-target inhibitors (e.g., dual inhibitors, bispecific antibodies) as well as the combination of anti-VEGF/VEGFR inhibitors with currently used anti-MM drugs such as PIs, IMIDs, and monoclonal antibodies. Greater efficacy is emerging with drugs that simultaneously block multiple cytokines. The improved outcome of MM patients treated with the new biological drugs is also related to the critical activity in tumor microenvironment, including antiangiogenesis activity, which has an anticancer effect. So, the inhibition of cytokines pathways, which mediate the interaction between cancer cells and their microenvironment near the anti-MM activity, represents one of the primary goals of modern therapeutic approaches, and better results with regard to ORR, PFS, and OS, have been achieved in the last few years. Moreover, the evidence from preliminary results on the efficacy of the aforementioned new strategies in MM relapsed/refractory MM patients support the assumption that VEGF/VEGFR inhibition represents a useful strategy in the treatment of MM patients.Finally, we need further studies to better elucidate the pathways involved in the BM neovascularization biological process, and the mechanisms underlying the development of microenvironment-mediated PCs resistance. Innovative approaches to bypass the vicariate pathways active after the inhibition of a single molecule, such as the use of multi-target agents or combined treatment strategies are required. The combination of multiple targeted agents could achieve the goal of maximizing the efficacy of biological treatment for MM patients. Finally, we need validated biomarkers that are different from the traditional ones (e.g., β2 microglobulin, albumin, serum, and urine protein electrophoresis, etc.) to establish personalized treatments and select responsive patient sub-populations. These markers will be identified and validated thanks to continued research in the field of microenvironment activity in cancer.Conceptualization, data curation, writing—review and editing, R.R., A.M., V.R., A.V.; funding acquisition, R.R., V.R. All authors have read and agreed to the published version of the manuscript.This work was supported by the Associazione Italiana per la Ricerca sul Cancro (AIRC) Investigator Grant (no. 20441 to V.R.), and grants from MIUR PRIN 2009WCNS5C_004 (to R.R.).In this section you can acknowledge any support given which is not covered by the author contribution or funding sections. This may include administrative and technical support, or donations in kind (e.g., materials used for experiments).The authors declare no conflict of interest.Schematic representation of VEGFRs and relative ligands. VEGFR-1 is expressed in a subset of hematopoietic bone marrow progenitor cells, monocytes/macrophages, smooth muscle cells, and several cancer cells. VEGFR-2 is expressed prevalently in vascular endothelial cells and progenitors, neurons, and megakaryocytes, but it can be expressed on the surface of several cancer cells. VEGFR-3 is restricted to lymphatic endothelial cells. Neuropilins are expressed prevalently by cells of the nervous system, and they work as co-receptors in the endothelial cell during the angiogenic process. VEGF: vascular endothelial growth factor; VEGFR: VEGF receptor; PLGF: placental growth factor; Flt-1: Fms-like tyrosine kinase-1; KDR: Kinase insert domain receptor; Flk-1: Fetal Liver Kinase-1; FLT-4: Fms-like tyrosine kinase-4; NRP: neuropilin.VEGFs and PlGF are is produced and released by all cells in the bone marrow microenvironment and displays several biological activities correlated to multiple myeloma (MM) progression. VEGF: vascular endothelial growth factor; VEGFR: VEGF receptor; PlGF: placental growth factor; NRP: neuropilin.VEGFR and cMet pathways in MM microenvironment. Sites of inhibition of anti-VEGF drugs. VEGF: vascular endothelial growth factor; VEGFR: VEGF receptor; KDR: kinase insert domain receptor; Flk-1: fetal liver kinase-1; HGF: hepatocyte growth factor; cMet: tyrosine-protein kinase Met or HGF receptor; GAB1: GRB2-associated-binding protein 1; FAK: focal adhesion kinase; PI3K: phosphatidylinositol 3-kinase; AKT: also know as protein kinase B (PKB); IKK: inhibitor of nuclear factor kappa-B kinase subunits; IKB: inhibitor of nuclear factor kappa-B kinase complex; mTOR: mammalian target of rapamycin; MEK: Mitogen-activated protein kinase kinase; ERK: extracellular signal-regulated kinase; RAC1: Ras-related C3 botulinum toxin substrate 1; NF-kB: nuclear factor kappa-light-chain-enhancer of activated B cells; PI: proteasome inhibitors; IMIDs: immunomodulators; TK: tyrosine kinase.

Med-MDPI/jcm_2/jcm-09-06-01766.txt

@@ -0,0 +1 @@
+The diagnosis of hereditary hemorrhagic telangiectasia (HHT) is based on the Curaçao criteria: epistaxis, telangiectases, arteriovenous malformations in internal organs, and family history. Genetically speaking, more than 90% of HHT patients show mutations in ENG or ACVRL1/ALK1 genes, both belonging to the TGF-β/BMP9 signaling pathway. Despite clear knowledge of the symptoms and genes of the disease, we still lack a definite cure for HHT, having just palliative measures and pharmacological trials. Among the former, two strategies are: intervention at “ground zero” to minimize by iron and blood transfusions in order to counteract anemia. Among the later, along the last 15 years, three different strategies have been tested: (1) To favor coagulation with antifibrinolytic agents (tranexamic acid); (2) to increase transcription of ENG and ALK1 with specific estrogen-receptor modulators (bazedoxifene or raloxifene), antioxidants (N-acetylcysteine, resveratrol), or immunosuppressants (tacrolimus); and (3) to impair the abnormal angiogenic process with antibodies (bevacizumab) or blocking drugs like etamsylate, and propranolol. This manuscript reviews the main strategies and sums up the clinical trials developed with drugs alleviating HHT.Hereditary hemorrhagic telangiectasia (HHT) or Rendu-Osler-Weber syndrome is a genetic dominant autosomal multisystemic vascular disease, whose penetrance increases with age. The Curaçao criteria were designed to diagnose HHT, and include its clinical symptoms which are spontaneous and recurrent epistaxis (nose bleeds), mucocutaneous telangiectases, visceral localization (gastrointestinal telangiectases and/or arteriovenous malformations (AVMs), mainly in lung, brain or liver), and a first degree family member with a definite diagnosis of HHT (Figure 1A) [1,2,3]. The prevalence of HHT varies between 1:5000 and 1:8000 on average, although because of the “founder effect” and “insulation effect,” the prevalence is higher in some regions such as the Jura region in France, Funen Island in Denmark and the Caribbean Dutch Antilles [3,4,5]. Heterozygous mutations in either ENDOGLIN (ENG) or ACVRL1/ALK1 genes trigger the pathogenesis of HHT in over 90% of HHT patients [6,7]. Less common mutations, responsible for 2% of HHT cases, appear in the SMAD4 gene, leading to a combined syndrome of Juvenile Polyposis HHT (JPHT) [8] consisting of HHT symptoms, colon polyps and thoracic aneurysms [9]. Furthermore, chromosomes 5 and 7 have been described to possess two loci with unknown genes, that cause HHT3 [10] and HHT4, respectively [11]. An HHT-like syndrome called HHT5 has been linked to mutations in BMP9 [12]. All mutations leading to HHT are found in genes belonging to the family of BMP9/TGF-β signaling pathway (Figure 1B).Moreover, the capillary malformation (CM)/AVM syndrome is phenotypically similar to HHT, and is characterized by the appearance of multiple CMs that are small and red, round to oval shaped with a peripheral white halo and randomly distributed. These are linked to heterozygous pathogenic variants in EPHB4 or RASA1 identified by molecular genetic testing [13].This review will focus on the pharmacological treatment for bleeding in HHT patients. With 93% of patients suffering light to moderate bleedings, epistaxis presents as the most frequent clinical manifestation of HHT [14,15]. It affects over 90% of patients before the age of 21, normally interfering with their quality of life [16]. Epistaxis are due to the telangiectases of the nasal mucosa, focally dilated venules, often connected directly with dilated arterioles [17]. Directly related to epistaxis is gastrointestinal (GI) bleeding, because of telangiectases in the digestive tract and observed in up to 80% of HHT patients [18]. However, GI bleeding becomes more frequent with age [19]. Although currently there is no optimal available treatment for either epistaxis or GI bleeding, the systemic pharmacological treatments that are used for epistaxis might also be useful to manage GI bleedings.The pharmaceutical therapies which that are discussed in the following sections address therapies wherein the disease is due to heterozygous germ-line mutations in all cells of the HHT patient. These therapies may not be effective for some cutaneous telangiectases, wherein endothelial cells (EC) may have homozygous mutants for ALK1/ENG according to a recent publication of Snellings et al. [20].To prevent crusting and allow the nasal mucosa to be correctly hydrated in HHT patients, local moisturizing treatments such as humidification, nasal cleaning with a saline solution and lipid-based topical ointments are used [18]. Despite these options, it is challenging to completely avoid nasal or GI bleeding in HHT, often leading to iron deficiency and anemia in these patients. For this reason, the first line treatment of HHT is focused on managing the anemia resulting from bleeding. Iron-enriched diets and iron supplements are cost-effective steps that significantly reduce the need of blood transfusions although the latter may be necessary in severely affected patients [2,21].The following section focuses on reviewing the pharmacological treatments, from a preclinical perspective. Robert et al. have also recently reviewed this topic [22].Options to control nose and GI bleeding could be used, according to the following strategies (Table 1). It should be commented that the drugs were included into each group, according to the main mechanism observed in vitro and/or in vivo experiments, yet in some cases other additional mechanisms maybe contributing to the therapeutic effect.Strategy 1. Although HHT does not result from a clotting failure, the use of antifibrinolytics to restore the balance between coagulation versus fibrinolysis would help to promote a quicker coagulation and to stabilize the fibrin network. Among the antifibrinolytics used for epistaxis treatment, tranexamic acid (TA) and ε-aminocaproic acid (AC) stand out [23,24].Strategy 2. HHT is associated with haploinsufficiency in ENG or ALK1 genes, therefore stimulating their protein expression is thought to revert the HHT phenotype. At this point, raloxifene hydrochloride and bazedoxifene acetate, two specific estrogen receptor modulators (SERMs), have proven efficiency and safety, and have been designated as orphan drugs for HHT (2010 EU/3/10/730 and 2014 EU/3/14/1367; respectively) [25,26].Strategy 3. Antiangiogenic therapies tackle the excess of abnormal vasculature present on the nasal mucosa in HHT. Therefore, bevacizumab (BZ) (Avastin®), a humanized monoclonal antibody against the main angiogenic factor, the vascular endothelial growth factor (VEGF), has been widely used and tested on HHT. Its systemic administration has improved hepatic function, delaying the liver transplant [27] but it has not shown consistent results when tested to decrease epistaxis events by topical spray administration [28,29].Following the same antiangiogenic strategy, pazopanib, thalidomide, and more recently pomalidomide, have been used to inhibit the VEGF pathway.Similarly, other cardiovascular drugs such as propranolol and timolol (non-specific β-blockers) have shown their benefits in nose bleeding when administered topically (both) and systemically (propranolol) [30] in HHT patients. Recently, the use of the fibroblast growth factor receptor (FGFR) blocker etamsylate, by local spray administration has been proven effective and has been designated as orphan drug for HHT in 2018 (EU/3/18/2087) [31].Table 2 and Table 3, respectively, summarize the ongoing and finished clinical trials, conducted in HHT with the different drugs, most of them mentioned in this review. In addition, some other recent candidates like Vitamin D, itraconazole and doxycyclin are in ongoing clinical trials.All the above-mentioned drugs used were repurposed medicines, which have the added value of an immediate use in clinical trials since their safety is already confirmed from their first indication. As stated by Masoudi et al. (2020) [43], “Drug repurposing is a powerful strategy in the discovery scope because of the time and cost savings”. Furthermore, it is an appropriate method for finding therapies for orphan and rare diseases.Antifibrinolytics block the plasminogen to plasmin conversion by inhibition of its enzymatic disaggregation and consequently, stabilization of the fibrin clot. Accordingly, these drugs are expected to target the wall of the telangiectases where fibrinolysis is activated [44,45].TA and AC are antifibrinolytic agents used for HHT epistaxis. Both may be administered topically (with the agent embedded in gauze) or systemically by oral intake (500 mg/8 h or even up to 2 g/day) or intravenous administration. AC was the first antifibrinolytic used but showed thrombosis as a side effect and is therefore not recommended in patients prone to thrombosis [46,47,48,49]. In addition, TA has shown longer half-life and higher potency (10-fold) than AC [23,24].In addition to several case reports with successful results, a study with a total of 14 patients with low risk of thrombosis and for whom their quality of life was poor due to epistaxis, were selected to take TA (500 mg/8 h) orally. TA treatment showed a decrease of nose bleedings and increase of hemoglobin levels in all patients, almost avoiding the transfusion necessity, indicating an overall improvement and no side effects of TA treatment [24]. Although the study was not a formal clinical trial, TA was safe and effective at the doses applied. To highlight, TA administered up to 3 g/day was successful in controlling a massive and life-threatening hemorrhage in an HHT patient [23]. Moreover, some published data from in vitro experiments demonstrated on ECs that TA led to increased mRNA and protein levels of endoglin and ALK1, and improved endothelial functions as tubulogenesis and migration [24]. Thus, elevated endoglin or ALK1 expression may act concomitantly to the main antifibrinolytic action, although the proposed mechanism is only based on in vitro evidence.Nevertheless, some concern must be taken in HHT patients with the elevated levels of the coagulation protein factor VIII (FVIII) and factor V Leiden, since some reports show an HHT-related increment of these protein levels that favor thrombotic risk in these patients. Another putative risk factor is the presence of high levels of factor V Leiden. Therefore, personalized risk-benefit considerations are demanded for HHT management [50].Generally, clotting factors levels are not altered in HHT patients (excluding patients with altered factor V or VIII expression). However, one way of shortening the time and frequency of bleeding is by displacing the balance between coagulation and anticoagulation process toward a more quick and stable clotting when the abnormal vessels (mucosa telangiectases) break. This is enhanced by the antifibrinolytics, which prompt the clotting and delay its fibrinolysis. Moreover, telangiectases have been reported to have high fibrinolytic activity by Sabbà et al. [51].In addition, two clinical trials were performed in HHT centers to assess the benefits of TA in HHT patients with reports published in 2014. In the French ATERO assay, TA was shown effective in a multinational center study [34], while Geisthoff et al. [32], demonstrated efficacy in a double-blind clinical trial phase III-B. However, while TA has demonstrated efficiency by systemic use, when topically used by a nasal spray, it did not significantly decrease nose bleeds when compared to placebo in a clinical trial (NOSE study) conducted by Cure HHT in 2016 [28].The incidence of epistaxis has been observed to be increased in women after they have reached menopausal age, suggesting that estrogens might play a protective role in HHT-derived bleeding in women. Post-menopausal women are also affected by osteoporosis, the imbalance in the rate of bone remodeling/resorption that predisposes elders to higher chances of bone fracture. NFκB, RANK and its ligand RANKL, as well as osteoprotegerin (OPG) play major roles in this pathogenesis [52] (Figure 2). Based on the observation that pre-menopausal HHT women had fewer epistaxis, a study developed by the Yale University’s Vascular Malformation Centre attempted to treat GI bleeding of 40 transfusion-dependent HHT patients, with a mean age of 57 years, by means of estradiol treatment. Men, to avoid estradiol feminizing effects, were also treated with ethinylestradiol/norethindrone and danazol. The results were satisfactory as most of the 40 patients showed an improvement in hemoglobin levels and needed fewer blood transfusions [53].The use of hormones to treat HHT-induced bleedings was published later in the form of case reports, but mostly without controls [55]. The main conclusion obtained from these studies was that estrogen-progesterone administered at the doses typically used for oral contraception might reduce bleedings in symptomatic HHT women, becoming a reasonable option for fertile HHT female patients. Zacharski et al. published in 2001 a case report in which the use of tamoxifen had ceased epistaxis in the long term in a post-menopausal patient, concluding for the first time that SERM was properly used to treat epistaxis in an HHT patient [56]. According to this, tamoxifen was used in two clinical trials where it again successfully decreased epistaxis. The first of these two consisted of a placebo controlled clinical trial that included both men and women, while the second comprised a long-term monitored clinical trial in which patients were administered 20 mg tamoxifen [33,54].In the line of using SERMs to decrease HHT-related bleedings, the safety and efficiency of raloxifene hydrochloride was tested by a Spanish HHT reference unit IDIVAL (Sierrallana/Valdecilla). Raloxifene hydrochloride shows similarities with tamoxifen, also presenting beneficial effects on bone mineralization and on prevention of cardiovascular and gynecological cancer. This study included 19 post-menopausal women, previously diagnosed with osteoporosis, and compared the amount of bleeding before and after 6 months of treatment (no placebo was included in the study). Oral intake of raloxifene (60 mg/day) showed a significant reduction in both the frequency and the amount of epistaxis after 6 months of treatment, also revealing an increase in hemoglobin levels [25]. Raloxifene has been shown to be a transcriptional activator of ENG and ACVRL1/ALK1 promoters, binding to their proximal regions and subsequently increasing these genes’ transcription rate in a context of in vitro experiments on ECs [25]. As a consequence, the protein levels of endoglin and ALK1 increased, thus compensating partially the haploinsufficiency suffered by HHT patients in this study [25]. In 2010, these studies resulted in the European Medicine Agency (EMA) and Food and Drug Administration (FDA) designation of raloxifene hydrochloride as the first orphan drug to treat bleedings in HHT patients (EU/3/10/730). Bazedoxifene acetate, another SERM, significantly decreased the frequency and intensity of epistaxis, while also improving hemoglobin levels as early as one month after treatment with 20 mg/day [26]. In this case, the increase of ENG and ALK1 was not only observed in experiments in vitro with ECs treated with bazedoxifene, but also, in vivo, by measuring ENG and ALK1 levels in macrophages derived from patients before and after bazedoxifene treatment. Bazedoxifene was also designed as orphan drug for HHT in 2014 by the EMA (EU/3/14/1367).Of note, estrogens and SERMs, as hormonal receptor ligands, increase the transcription of different promoters, among them, those of coagulation factor genes. Thus, ENG and ACVRL1 are among the stimulated genes, but are not the only targets. In relation to this fact, especially when the treatment with SERMs may upregulate coagulation factors’ genes, blood tests should be performed periodically for HHT patients under SERM treatment in order to screen for prothrombotic markers and prevent thrombotic events [57].Finally, phytoestrogens, compounds of plant origin with structural similarities with the natural estrogen 17β-estradiol, deserve some words in this section as natural plant estrogen related products. Among them, the isoflavone genistein and the coumestan resveratrol are the most relevant in studies related to HHT. Genistein is found in numerous plant species such as soy and red clover and resveratrol in grape skin and in dried fruits and nuts. These phytoestrogens show high affinity for estrogenic receptors. Genistein and resveratrol are involved in reducing inflammation, stimulating apoptosis and inhibiting angiogenesis [58,59], and might present therapeutic benefits in HHT patients as natural analogues to SERMs and estrogens.Albiñana et al. reported the efficacy of tacrolimus (FK506) in increasing endoglin and ALK1 expression [60,61]. The reason to test this drug came from a case report of an HHT patient who was administered the immunosuppressor FK506 in low doses, in combination with Aspirin and sirolimus to avoid rejection of a liver transplant. One month after the start of this treatment, it was observed that his telangiectases (both internal and external), epistaxis and anemia had all been cured [62]. Based on this report, cultured ECs were treated with tacrolimus and an increase on the protein and mRNA expression of endoglin and ALK1 and enhancement of the TGF-β1/ALK1 signaling pathway and EC functions like tubulogenesis and cell migration were observed [60,61]. These results would explain the improvement in the above-mentioned patient, by means of a partial compensation of endoglin and ALK1 haploinsufficiency. Supporting this view, five years later, Ruiz et al. reported increased ALK1 signaling pathway in HHT patient-derived EC. In an HHT animal model, tacrolimus also inhibited VEGF signaling, decreasing hypervascularization [63].In a more clinical context, Sommer et al. published in 2019 that low doses of FK506/Advagraf decreased bleeding in an HHT patient presenting also pulmonary arterial hypertension [64]. This case report points to low doses of tacrolimus (0.5–1.5 mg/day) as the optimal range for patients with nose or GI refractory bleeding, rather than high doses (5–10 mg/day) normally used for immunosuppression in transplants [64]. An additional report by Hosman et al., including two patients dependent on HHT transfusions due to severe bleeding, demonstrates improvement after treatment with low-dose tacrolimus [65]. Currently, and according to the HHT European Federation, around 24 HHT patients are being treated “off label” with low tacrolimus doses, prescribed by HHT reference doctors, to control epistaxis and GI bleeding.Finally, the results regarding efficacy and safety of 0.1% tacrolimus topically applied as nasal ointment of the clinical trial named TACRO have just been published. Tacrolimus nasal ointment did not result in improvement 6 weeks after finishing treatment, but the good tolerance and the significant improvement in epistaxis duration during treatment invited the researchers for a phase 3 trial on a larger patient population and a longer treatment time, with a main outcome of epistaxis duration during treatment [42].Based on the premise that free O2-radicals might cause precapillary sphincter abnormalities, resulting in epistaxis, Gussem et al. in 2009 wondered whether antioxidants like N-acetylcysteine (NAC) could neutralize those free O2-radicals and reduce or avoid nose bleedings. Thus, 43 HHT patients were followed-up for frequency, severity, and duration of epistaxis after a daily treatment of 600 mg NAC for 12 weeks. There was a reduction in frequency and severity of nosebleed during the day. Male patients with ENG mutations experienced a significant improvement. Only an improvement trend was found in women and patients with an ALK1 mutation [66].Based on these results, Albiñana et al. studied the in vitro effects of NAC on endoglin and ALK1 expression levels in ECs. After NAC incubation, mRNA and protein levels of endoglin increased up to 1.5–2 folds, although there were no changes on ALK1 levels [67]. These data could suggest that the improvement experienced with their symptoms in HHT1 patients might be due, in part, due to the increase of endoglin levels after NAC treatment, which could be normalizing the nasal mucosa [66].Antiangiogenic strategies on HHT act on the mucosa to decrease or normalize its abnormal excessive vasculature. Two key angiogenic pathways in ECs are those triggered by VEGF and FGF.VEGFs specifically act on vascular EC and are a key stone in the angiogenic and lymphangiogenic process in both physiological and pathological conditions such as tumors or wound healing [68]. VEGFs play an important role in HHT since high protein levels have been reported in HHT patients [69,70,71,72].Avastin ranks the first antiangiogenic therapeutic agent approved for advanced colorectal cancer [73]. Since then, BZ has been widely administered in other pathologies such as non-small cell lung cancer diabetic retinopathy or age-related macular degeneration [74].The antiangiogenic properties of Avastin were successfully tested in isolated cases of HHT patients. BZ reverted the need for transplantation in a patient with HHT1 [75]. It also decreased the transfusion requirements and cardiac output in a patient with GI [76]. Thus, the French HHT Network designed a single center phase II clinical trial to address the delay in liver transplantation on HHT patients with serious liver complication [27]. BZ significantly decreased the cardiac output and reduced episodes of epistaxis. Nevertheless, symptoms did not disappear after withdrawal of the drug, making Avastin unsuitable as a surrogated alternative for orthotopic liver transplant (OLT) in HHT. These results supported the designation of BZ as orphan drug for HHT in 2014 (EU/3/14/1390).It is very difficult to determine the optimum time to perform OLT in severe complicated liver venous malformations (VMs) in HHT. OLT is a radical cure for liver VMs and it should be the therapeutic choice in patients under the age of 65 years due to its excellent outcomes, being BZ probably a better option for patients over 65, much more susceptible to higher risk derived from surgery [77].BZ has been also administered for very severe epistaxis. Despite of its good results decreasing epistaxis and GI after intravenous administration, it has also serious side effects. Therefore, in those patients where no other option is available, an individualized BZ treatment should be supplied; since HHT is a chronic disease [27,77,78]. This individualized re-dosing strategy will not compromise the patient’s safety or quality of life and will significantly reduce the costs [79].BZ has also been assayed topically as nasal spray for epistaxis. The NOSE assay (under the Cure HHT support in USA, 2011–2015) and the French Ellipse clinical trial (2011–2012) ran in parallel. Unfortunately, no significant improvement could be demonstrated for the BZ vs. saline solution spray [28,29].An alternative therapeutic approach for reducing epistaxis and GI bleeding caused by the over activated VEGF pathway is the blockade of the tyrosine-kinase activity of the VEGF receptor. As a proof of concept, the therapeutic effect of the TKI GW771806 (a pazopanib analogue), was tested in an Alk1-inducible knockout (iKO) murine HHT2 model. The oral administration of GW771806 significantly improved anemia and GI bleeding in HHT2 mice [80]. Unfortunately, its corresponding phase II human trial designed to follow-up adult HHT patients with significant epistaxis, anemia, or with transfusions could not be completed. A prospective, multi-center, open-label, dose-escalating study on pazopanib has also been published and the results show promising improvements in hemoglobin levels and epistaxis in treated patients [81].Related to pazopanib and in a further step, nintedanib, a TKI targeting growth factor receptors involved in angiogenesis: platelet-derived growth factor receptor (PDGFR), FGFR and VEGFR, will be assayed as a therapeutic drug in a clinical trial. Nintedanib treatment in combination with rapamycin, synergizes to completely block the AVMs in HHT mice pre-clinical models [82]. This has been the main rationale to continue with nintedanib in clinical trials. In that sense, Epicure, a randomized, multicenter, phase II, double-blind placebo-controlled study promoted by the French Hospices Civils de Lyon, has started its recruitment in 2019. Epicure will test the antiangiogenic benefits of the TKI nintedanib in epistaxis. Initially, patients will be monitored for 24 weeks: 12-weeks of oral treatment plus 12-weeks of follow-up. Theoretically, nintedanib, as a non-specific/wide range TKI, should allow a reduction of epistaxis in HHT [83].Lastly, in this section dealing with drugs interfering with the VEGF signaling pathway, thalidomide deserves special mention, since it has been used for epistaxis and GI bleeding in HHT. Its mechanism of action was published in 2010 [35] and since then, several case reports and clinical trials have been performed (Table 2 and Table 3) prior to its designation as an orphan drug (Table 4). Evidence supporting thalidomide compares favorably in cases of serious and refractory GI and nosebleeds. However, risk vs. benefit should be carefully studied, because of the reported side effects for thalidomide [78]. A derivative of thalidomide with potentially fewer side effects, pomalidomide, has been promoted for a multicenter randomized double-blind placebo controlled clinical trial currently ongoing sponsored by the NIH (Table 2).Non-selective adrenergic β-blockers of receptors β1 and β2 such as the known propranolol and timolol, have shown antiangiogenic properties related to vasoconstriction, inhibition of EC migration and proliferation and reduced VEGF expression [30,84,85,86]. Given that an excessive activation of the VEGF pathway is involved in the development of abnormal telangiectases, the properties of these non-selective adrenergic β-blockers may be useful topically. Propranolol and timolol have been used as effective therapies to treat superficial infantile hemangiomas [84,85,86] and could be considered as a potential treatment option for HHT patients.The use of topical timolol has been described in some case reports as well as in clinical trials/studies of HHT. According to several reports, topical timolol (0.5% ophthalmic solution) clearly improved the frequency and severity of epistaxis [87,88]. Those studies were done in individual patients as well as in larger groups of HHT patients, with clear positive results in all cases. Even though no secondary adverse effects were observed in these cases, there are some contraindications for its applications which should be kept in mind when prescribing, because β-blockers can cause respiratory and cardiovascular problems. In fact, β-blockers are metabolized in the liver by CYP2D6 and a decreased expression of this enzyme, either by concomitant treatment with CYP2D6 inhibitors or by genetic variants, may lead to strong sinus bradycardia [89,90]. Timolol has also been proved to be an efficient treatment at lower doses (0.1% ophthalmic solution-timogel) in decreasing the extension and appearance of mucocutaneous telangiectases in an HHT clinical study, showing 100% and 75% improvement in HHT2 and HHT1 patients, respectively [91].Propranolol has also been administred topically in gel formulation and systemically in tablets with very promising results, although the last method could be associated with the above-mentioned side effects, which must be considered before prescribing.There are several studies supporting the improvement in HHT patients after propranolol topical administration. In a pilot study done with six patients, in which 1.5% propranolol gel was applied on the nasal mucosa (0.5 mL/day per nostril) the severity of epistaxis and the number of blood transfusions pre and post-administration were reduced fast and significantly [92]. As continuation of this study, the same group recently finished a double-blind placebo-controlled study to assess the efficacy and safety of topical propranolol for moderate–severe epistaxis in 24 patients with HHT [93]. The combination of sclerotherapy with polydocanol 1% and the use of propranolol cream at 0.5% prepared in a hospital pharmacy was evaluated in a cross-sectional study of 38 HHT patients. This combined therapy significantly reduced the frequency and severity of epistaxis, with an Epistaxis Severity Score (ESS) improvement of 5 points (from 6.9  ±  2.6 at the beginning to 1.9  ±  1.3 after the therapy, p <  0.05); and increased the quality of life of these patients (in a 5D scale, from 0.66  ±  0.27 before therapy to 0.93  ±  0.12 after the therapy p <  0.05) [94].On the other hand, systemic uptake of propranolol has some side effects such as bradycardia and hypotension. Therefore, propranolol could be used in a systemic way only in HHT patients who do not have a low blood pressure. A clinical study with oral propranolol (40–120 mg/day) was done in seven hypertensive HHT patients. Among them, HHT epistaxis disappeared completely in five out of seven, while in the other two the bleeding reduction was highly significant [95].The FGF family is considered one of the largest families of polypeptide growth factors. FGFs interact with membrane tyrosine kinase receptors (FGF Receptors, FGFRs) through which they signal and exert their diverse biological functions. FGF-1 and FGF-2 were the first two pure polypeptides discovered among all FGFs and are considered one of the most potent factors promoting angiogenesis [96,97]. Later on, more factors promoting angiogenesis were described. FGFs constitutively cooperate with VEGF promoting the proliferation of EC [98] and inducing angiogenesis [99]. Consequently, FGF inhibition constitutes an alternative to the antiangiogenic effect of VEGF, becoming an interesting new therapeutic approach to treat diseases caused by uncontrolled angiogenesis. Many of the characteristics summarized previously suggest that the inhibition of the FGF signaling pathways could be an appropriate treatment to inhibit the abnormal vascularization of HHT patients [100].FGF signaling may be inhibited in vitro and in vivo with chemical compounds. One of these products is the anion dihydroxy benzene sulfonate (2,5, DHBS), also known as dobesilate. Dobesilate is commercially available as tablets of its calcium and N-ethylene ethanamine salts (Doxium and Dicynone, respectively), or as injectable solution (Dicynone or etamsylate). Experiments done in vitro with primary cultures from healthy and HHT derived EC, show that etamsylate acts as an antiangiogenic factor, inhibiting wound healing and EC tubulogenesis [31]. A pilot clinical trial (EudraCT: 2016–003982–24, see Table 3) was performed with 12 HHT patients treated with a topical spray of etamsylate twice a day for 4 weeks. The HHT-ESS and other pertinent parameters were analyzed and registered. The significant reduction in the HHT-ESS scale (pre-treatment 4.1 vs. post-treatment 2.8), together with the lack of significant side effects, allowed the designation of topical etamsylate as a new orphan drug for epistaxis in HHT patients in 2018 (EU/3/10/18/2087) [31].The currently available pharmacological treatments for HHT are summarized according to their mechanism of action on Figure 3. The drugs here included may be used on an individual level; treatment for rare diseases should follow the premises of personalized medicine, relying on the reference doctor expertise. A first-line treatment to avoid or decrease epistaxis or GI bleeding is to reinforce the speed and stability of coagulation with antifibrinolytics, preferentially TA. The only contraindication may apply to patients with risk of thrombosis. Doses vary from 1.5 g/day up to 3 g/day in episodes of severe bleeding. In countries where TA is not commercially available, AC could be used instead.For bleedings interfering with normal life, currently the best treatment option for women in fertile age are feminine hormones used for contraception. Women in peri- or post-menopausal age can benefit from the use of SERMs. Although SERMs are primarily used to prevent or treat osteoporosis, they have shown to be effective in HHT compensating partially, the haploinsufficiency present in HHT patients by increasing the protein levels of endoglin and ALK1. Raloxifene hydrochloride was designated as orphan drug by the EMA and by the FDA. Bazedoxifene acetate was also designated as orphan drug by the EMA but it is not commercialized in United States.Among the list of drugs that increase the expression levels of endoglin and ACVRL1/ALK1, tacrolimus was demonstrated to activate the signaling of endoglin/ALK1 in ECs. It would be the treatment of choice for immunosuppression in transplanted patients. However, recent case reports have opened the possibility of using systemic tacrolimus at low, non-immunosuppressive doses, to treat HHT bleeding. Currently, around 25 patients are using tacrolimus as an “off-label” treatment and it will be very interesting to hear about the outcome of these patients. Other treatment includes NAC which increases the endoglin RNA and protein levels and can be used as an anti-inflammatory and antioxidant drug without any side effects.BZ is used (off-label) for antiangiogenic strategy in HHT patients with severe bleeding or symptomatic liver AVMs, in order to reduce bleedings and excessive number of abnormal mucosa vessels. TKIs (pazopanib, nintenadib, sunitinib or buparlisib) and thalidomide have also been used or are planned to be used in clinical trials [22,83,101,102] (Table 3).Another group of antiangiogenic drugs comprises the non-selective adrenergic β-blockers like propranolol and timolol. These drugs are preferentially used topically, as creams or gels. They reduce nose bleedings by decreasing and delaying the formation of telangiectases on the mucosa. Propranolol can also be used systemically, but special attention is indicated with regards to blood pressure and heart rate.Lastly, as emerging treatments, the orphan drug designation of etamsylate for topical treatment of epistaxis opens a new perspective following the antiangiogenic strategy.This review has delved into the various treatment strategies of HHT and the clinical trials that support these treatments. The repurposing strategy has led to the formal approval of several orphan drugs for HHT (Table 4).L.M.B. designed the project. L.M.B. wrote the manuscript with the contribution of V.A., A.M.C., E.G.-V., I.d.R.-P., and L.R.-P., V.A. and A.M.C. designed the figures. V.A., A.M.C., E.G.-V., I.d.R.-P., L.R.-P., proofread the text. C.B. revised the manuscript and provided funding. L.M.B. supervised the project and provided funding. All authors discussed the results and contributed to the final manuscript. All authors have read and agreed to the published version of the manuscript.This research was funded by Ministry of Economy and Competitivity, grant number SAF2017-83351-R to L.M.B.In memoriam of José Sánchez Andújar, who passed away on 22 May 2020. He was a founder member and the first President of the Spanish Association of HHT patients.The authors declare no conflict of interest.Hereditary Hemorrhagic Telangiectasia. (A). Clinical manifestations of HHT, Curaçao criteria. Telangiectasias in ear, hands, tongue, and lips; arteriovenous malformations in internal organs, epistaxis and family history. (B). TGF-β/BMP9/10 signaling pathway in endothelial cells. Once the ligand binds to its receptor complex formed by the kinase receptors I and II, and the auxiliary receptor III (endoglin), the signaling cascade leads to the phosphorylation of Smad proteins. The translocation of the Smad protein complex into the nucleus results in transcriptional regulation on target genes. Endothelial cells (EC) express two types of type I kinase Receptors: ALK1 and ALK5.Scheme of Bone formation vs. Bone resorption (left). Bone remodeling is promoted by the activity of osteoclasts and osteoblasts. RANKL is a protein expressed by osteoblasts which, upon binding to its RANK receptor on the cell surface of osteoclasts and their precursors, causes RANK to stimulate and promote the adhesion of osteoclasts to bone, thus activating their function and preventing apoptosis. OPG is synthesized by the osteoblasts and acts as a decoy receptor, preventing the binding of RANKL to RANK, therefore decreasing the activity of the osteoclast and its survival [52]. For this reason, for years post-menopausal patients with diagnosed osteoporosis have been receiving estrogenic treatment, which, while correcting this imbalance in bone remodeling, also reduces the activity of the osteoclast and activates the expression of OPG. This is also capable of reducing the epistaxis of these patients [25,33,53,54]. Scheme of Estrogen Receptor (ER) signaling (right). Mechanism of action of ER on EC, in the case of the SERMs raloxifene and bazedoxifene. Upon ligand binding, the ER dimer interacts with different transcription factors (TF), promoting gene expression by binding to the TFRE (Transcription Factor Regulatory Element) in the promoter of its target genes. The expression of endoglin and ALK1 increases as a result of the interaction of the ER with Sp1, an essential factor for the expression of both genes. In women, SERMs promote a decrease in the frequency and amount of epistaxis, normalizing the nasal vasculature with concomitant improvement in osteoporosis [25,26].Therapeutic strategies of pharmacological treatments in HHT. (1) Antifibrinolytic strategy; prevents the conversion of plasminogen to plasmin, thus delaying the lysis of the fibrin clot, and therefore the bleeding. (2) Upregulation strategy. The drug works by increasing the expression of the ENG or ALK1 genes, and thus resulting in increased amount of proteins, improving the BMP/TGF-β signaling and normalizing the formation of new vessels. (3) Antiangiogenesis strategy; makes disappear the excess of existing abnormal vasculature or normalizes it.Therapeutic strategies to decrease epistaxis in HHT.Tranexamic acidƐ-aminocaproicacidSERMs: raloxifene, bazedoxifeneTacrolimusBy inactivating the VEGF signaling pathwayBevacizumab, propranolol, timolol, thalidomide or pazopanibBy blocking the FGF-R signaling pathwayEtamsylateCompilation of the HHT interventional clinical trials (Recruiting, Ongoing, or Unknown status).Table 2. Compilation of the Recruiting, Ongoing, or Unknown HHT interventional clinical trials registered at the EU Clinical Trials Register (EudraCT) (https://www.clinicaltrialsregister.eu), the U.S. National Library of Medicine (NCT) (https://clinicaltrials.gov), and the Australian New Zealand Clinical Trials Registry (ACTRN) (http://www.anzctr.org.au/Default.aspx). Only interventional trials where a therapeutic drug was tested are listed. Abbreviations: Countries: AR (Argentina); AU (Australia); CA (Canada); FR (France); IT (Italy); NL (Netherlands); US (United States of America). Phase: NA (Not Applicable). Status: R (Recruiting); O (Ongoing); U (Unknown). Outcome: ESS (Epistaxis Severity Score).Compilation of the HHT interventional clinical trials (Completed or Terminated status).Table 3. Compilation of the Completed or Terminated HHT interventional clinical trials registered at the EU Clinical Trials Register (EudraCT) (https://www.clinicaltrialsregister.eu), the U.S. National Library of Medicine (NCT) (https://clinicaltrials.gov). Only interventional trials where a therapeutic drug was tested are listed. Abbreviations: Countries: AT (Austria); CA (Canada); ES (Spain); FR (France); IL (Israel); IT (Italy); NL (Netherlands); UK (United Kingdom); US (United States of America). Phase: NA (Not Applicable). Status: C (Completed) and T (Terminated). Outcome: ESS (Epistaxis Severity Score). * The Study Record Detail from the website refers to the number of patients in parenthesis.Orphan drugs approved by the EMA for HHT.

Med-MDPI/jcm_2/jcm-09-06-01767.txt

@@ -0,0 +1 @@
+Cardiac surgery-associated AKI (CSA-AKI) is common after cardiac surgery and has an adverse impact on short- and long-term mortality. Early identification of patients at high risk of CSA-AKI by applying risk prediction models allows clinicians to closely monitor these patients and initiate effective preventive and therapeutic approaches to lessen the incidence of AKI. Several risk prediction models and risk assessment scores have been developed for CSA-AKI. However, the definition of AKI and the variables utilized in these risk scores differ, making general utility complex. Recently, the utility of artificial intelligence coupled with machine learning, has generated much interest and many studies in clinical medicine, including CSA-AKI. In this article, we discussed the evolution of models established by machine learning approaches to predict CSA-AKI.Acute kidney injury (AKI) is a common and serious complication after cardiac surgery related to a complex set of exposures, including cardiopulmonary bypass, tissue damage, cardiac dysfunction, and hemolysis [1,2,3,4,5]. Depending on the definition, patient characteristics, and the type of cardiac surgery, the incidence of cardiac surgery-associated AKI (CSA-AKI) varies between 7% and 40%, Table 1 [6,7,8,9,10,11,12]. The incremental index hospitalization cost associated with CSA-AKI is higher than USD 1 billion in the United States [13]. The development of CSA-AKI has a dramatic impact on intensive care unit (ICU) and hospital length of stay as well as on short- and long-term mortality [14,15,16,17,18,19]. In patients with CSA-AKI requiring dialysis, mortality (at hospital discharge or 30-day mortality) can be as high as 60% to 70% [20]. Risk of end-stage kidney disease (ESKD) after cardiac surgery is also substantial, especially in patients with Acute Kidney Injury Network (AKIN) stage 2 or 3 AKI, with a hazard ratio of 3.8 to develop ESKD compared with all patients [21].Several risk factors have been identified that are associated with an increased risk to develop CSA-AKI, including female sex, advanced age, left ventricular ejection fraction less than 35%, comorbidities (diabetes, hypertension, hypercholesterolemia, peripheral vascular disease, chronic obstructive pulmonary disease, congestive heart failure), preexisting chronic kidney disease (CKD), previous cardiac surgery, intraoperative (use of an intra-aortic balloon pump, more extended cardiopulmonary bypass (CPB) and prolonged aortic cross-clamping), severe bleeding requiring transfusion of blood products, a requirement for potent vasopressors, prolonged hypotension, and low cardiac output syndrome, systemic inflammatory response syndrome, more complex cardiac disease such as left main coronary disease, complex cardiac operations, and emergency surgery. Perioperative administration of nephrotoxic agents, such as angiotensin-converting enzyme inhibitors, aminoglycoside antibiotics, loop diuretics, or contrast media, may increase the risk of developing a CSA-AKI, Table 2 [20,45,46,47]. Attempts to improve AKI’s clinical outcomes have centered on early diagnosis and customized treatment [52]. Early identification of patients at high risk of CSA-AKI by applying risk prediction models allows clinicians to closely monitor these patients and start effective preventive and therapeutic approaches to lessen the incidence of AKI. Several AKI risk prediction models have been developed [14,16,17,53,54,55,56,57,58,59]. Risk assessment scores have been developed for CSA-AKI [14,16,17,53,54,55,56,57,58,59,60]. The Thakar Score [53], Mehta score [14], and Simplified Renal Index score [16] have been validated for predicting severe AKI that requires renal replacement therapy. The Thakar model has been examined comprehensively and found to have great discriminations. Nonetheless, dialysis events are uncommon, restricting the utility of these risk scores to prognosticate patients who do not need renal replacement therapy. Moreover, the definition of AKI and the variables utilized in these risk scores differ, making general utility complex [55]. Most of these risk scores are based on clinical factors that are accessible in the preoperative setting. These are helpful for risk stratification and counseling of patients. Nonetheless, perioperative events, including prolonged bypass, blood loss, and transfusion, can negatively influence the risk for CSA-AKI.The model established by machine learning approaches can effectuate early dynamic monitoring based on the actual objective data of all patients and conserve the time of clinicians. [61,62,63,64]. The rise of machine learning is driven by the ability to process “big data” and the need to deliver the best possible value- and evidence-based care. The utility of artificial intelligence (AI) coupled with machine learning, has generated much interest and many studies in clinical medicine [61,65,66,67,68,69,70,71,72,73,74,75,76,77,78,79]. The machine learning approach has been developed recently for advantages in performance and extensibility and has become indispensable for solving complex problems in most sciences [80,81,82]. This method is used to examine postoperative outcomes [83,84,85,86] and predict hypotension [87,88] and the depth of anesthesia [89,90,91,92,93,94]. Machine learning has also been applied in the fields of intensive care unit medicine [95], emergency medicine [96], and neuroimaging [97]. With the notable extension of the application of electronic health records (EHRs) in the area of big data [98,99,100], a substantial amount of EHR data and machine learning algorithms have advanced to fulfill an essential role in the clinical study of AKI. It is presently a relevant tool for AKI diagnosis and prediction [64]. The establishment of an AI-based clinical decision support systems (CDSS) based on a self-learning predictive model may be utilized for monitoring AKI among hospitalized patients in prospective clinical practice [101]. Compared with conventional analysis methods, recent studies have suggested that some machine learning algorithms may reach greater accuracy than the conventional logistic regression models [75,102,103]. Studies have shown that machine learning can predict AKI after general surgery, liver transplant, cardiac surgery, hepatectomy, severe burns, sepsis, and percutaneous coronary intervention [75,104,105,106,107,108,109,110]. Utilizing data from more than 700,000 subjects from multi-centers and stratified by an interval window of 6 hours, a recurrent neural network-based risk prediction model for AKI (AUC of 0.92] was verified [111]. AKI episodes were prognosticated within a 48-h window. Nevertheless, the area under the precision-recall curve was only 30%, which depicts a ratio of two false alerts for each actual alert [111]. Some machine learning algorithms have also been labeled as a “black box”, where there is limited insight into how the model is basing its prediction [112]. This draws into inquiry how clinicians can mitigate certain risk factors in patients to make them a more suitable candidate for treatment without knowing what is influencing their outcome. Nevertheless, there are some machine learning algorithms, such as XGBoost (eXtreme Gradient Boosting), where the relative magnitude of variables in prognosticating a particular outcome can be computed and envisioned. This renders a level of insight comparable to a logistic regression model about individual risk factors and their prognostic significance [113]. A gradient boosting machine (GBM) is currently a widespread approach for predicting AKI onset [75,105,107,114]. Huang et al. [107] presented a hazard prediction model for AKI following a percutaneous coronary intervention (PCI) based on GBM. The investigation involved a substantial amount of data from 947,091 cases that underwent PCI to set a baseline model. Besides, temporal validation was conveyed with data from greater than 900,000 hospitalized patients. The AUC of the GBM model was 79% greater than the baseline linear regression model. Recently, Lee et al. [75,105] presented a prediction model for AKI following liver transplantation and cardiac surgery by several machine learning algorithms. GBM demonstrated the most reliable performance in both investigations [75,105]. In the setting of cardiac surgery, Lee et al. [75] recently used machine learning techniques to predict CSA-AKI among 2010 patients undergoing cardiac surgery based on data obtained from EHR and developed an internet-based risk estimator.  In comparison with logistic regression analysis, decision tree, random forest, and support vector machine displayed comparable performance with regards to AUC. GBM method exhibited the best performance with the highest AUC (C-index 0.78, compared with the logistic regression model that had a C-index of 0.69) [75]. These patterns can accurately distinguish groups of cases with different risks, and their incorporation into clinical practice can reduce intricacies and improve outcomes of CSA-AKI.With the additional deepening of the investigation, machine learning-assisted monitoring may yield valuable upshots to AKI and lessen mortality and morbidity-associated CSA-AKI.  The principal benefit of machine learning is in its capability to distribute with many features with multiple interactions and its specific focus on maximizing predictive performance. Nonetheless, the emphasis on data-driven prediction might dismiss mechanistic perception. Future studies are required to assess whether a machine learning model that combines AKI biomarkers (such as IL-18, NGAL, and KIM-1) [115,116,117,118] and EHR data perform better in predicting CSA-AKI than other commonly used models. Essential prerequisites are comprehensive databases with high-quality data and the evaluation and integration of AI into pragmatic clinical settings; hence, the understanding of AI and its applications in our profession is important for the present and prospective advancement of Nephrology.In summary, CSA-AKI is a complex and multifaceted syndrome associated with significant morbidity and mortality. In the present era of using big data, the application of machine learning in Nephrology clinical practice to predict AKI, including CSA-AKI, holds great future promise. C.T., P.H., and W.C. drafted the manuscript. All authors gave comments on the earlier versions of the manuscript. All authors edited the manuscript and approved the final version. All authors have read and agreed to the published version of the manuscript.The authors declare no conflict of interest.Incidence of acute kidney injury following cardiac procedures and surgeries.CABG, coronary artery bypass graft; ECMO, extracorporeal membrane oxygenation; IABP, intra-aortic balloon pump; LVAD, left-ventricular assist device; MVR, mitral valve replacement; SAVR, surgical aortic valve replacement; TAVR, transcatheter aortic valve replacement.Risk predictors for acute kidney injury following cardiac procedures and surgeries from multivariate analysis.BMI, body mass index; CABG, coronary artery bypass graft; CKD, chronic kidney disease; CPB, cardiopulmonary bypass; eGFR, estimated glomerular filtration rate; IABP, intra-aortic balloon pump; NYHA, New York Heart Association; PVD, peripheral vascular disease; RBC, red blood cell; SAVR, surgical aortic valve replacement; TAVR, transcatheter aortic valve replacement.

Med-MDPI/jcm_2/jcm-09-06-01768.txt

@@ -0,0 +1 @@
+Background: Idiopathic epiretinal membrane (iERM) is a common eye disease, which can be treated by surgical removal of the fibrotic tissue. Morphological outcome is generally evaluated by optical coherence tomography (OCT). Here, we investigate the impact of surgery on hemodynamics of the posterior pole, using OCT angiography (OCTA). Methods: Patients with unilateral iERM and indication for treatment were included. OCT and OCTA images of the posterior pole were obtained preoperatively and 3-months after 23G vitrectomy with membrane peeling. Parameters of interest included full retinal perfusion, choriocapillaris perfusion (CCP), Sattler’s layer perfusion (SLP), and Haller’s layer perfusion, which were evaluated longitudinally and also compared to unaffected fellow eyes. Using these parameters, multiple regression analyses were used to predict visual outcomes. Results: Sixty-three iERM eyes were recruited, which initially showed a significant bigger central retinal thickness (p < 0.001) and total macular volume (TMV) (p < 0.001) compared to fellow eyes, while perfusion parameters were alike. Three months later, treated eyes had a statistically significant thicker subfoveal choroid (p = 0.006) and showed an increase of CCP (p = 0.003), while SLP decreased (p = 0.014). Lower preoperative TMV and higher SLP were associated with better postoperative visual outcome. Conclusion: In this OCTA study, iERM itself does not affect the choroidal circulation. However, preoperative choroidal circulation will be a biomarker to know the influence on the choroidal circulation after ERM surgery and may be considered as a predictor for visual outcome.Idiopathic epiretinal membrane (iERM) is a common macular disease in elderly people, characterized by fibrocellular tissue proliferation along the surface of the internal limiting membrane (ILM) [1]. The structural changes of the macular architecture lead to variable loss of visual acuity and metamorphopsia by causing vertical traction with macular thickening as well as tangential forces dragging the retina from its original position and displacing the vessels [2]. Pars plana vitrectomy with epiretinal membrane peeling is the standard of care to release the tension and restore the normal structure of the macula, commonly leading to visual acuity improvement and reduction of metamorphopsia [3]. Since studies suggest that ERM recurrence arises from remnant ERM components using the ILM as a scaffold, additional staining and ILM peeling is carried out by most surgeons [4]. However, in some cases, improvement of visual acuity is lacking, although ERM removal was successful; thus, an overall prognosis prediction model for postoperative best-corrected visual acuity (BCVA) is still needed [5].With the recent development of optical coherence tomography angiography (OCTA), the retinal and choroidal vascular network can be assessed in vivo and in real time by creating slab-segmented angiograms, without the need for dye injection [6,7,8]. Only a few studies have evaluated vascular changes due to ERM formation and ERM removal by using OCTA, focusing especially on retinal microvascular impairment [9,10,11,12,13]. However, the role of the choroid in ERM pathogenesis is not well-known, although recent OCT- and OCTA-based studies suggest choroidal alterations from ERM and following ERM-ILM peeling [14,15,16].The present study aimed to investigate structural and functional vascular alterations of the retina and the choroidal sublayers in patients with ERM. Furthermore, changes of OCT and OCTA parameters following pars plana vitrectomy with ERM-ILM peeling were analyzed and evaluated as predictive prognostic markers for the postoperative visual outcome.The participants for this prospective observational study were enrolled at the Department of Ophthalmology of the University of Lübeck. The study was approved by the institutional review board (vote reference number 18-255) and adhered to the tenets of the Declaration of Helsinki. All participants received detailed information about the study and written informed consent was obtained individually by each participant before enrolment. Only subjects with unilateral, idiopathic, and most importantly, symptomatic ERM and presence of metamorphopsia, which were scheduled for vitrectomy with ERM and ILM peeling on the next day, were included. Exclusion criteria were: (1) ERM secondary to other pathologies such as retinal detachment, laser photocoagulation, cryotherapy, diabetic retinopathy, uveitis, trauma or vascular occlusion; (2) any history of previous ocular surgery, except for uncomplicated cataract surgery; (3) evidence or history of systemic disorders, including cardiovascular diseases, antihypertensive drug use, as well as diabetes mellitus; (4) history of ocular diseases in the healthy fellow eyes.At baseline, all participants underwent a thorough examination including refraction, best-corrected visual acuity (BCVA) in Snellen, intraocular pressure (IOP), axial length (AL) measurement, slit-lamp biomicroscopy, macular enhanced-depth imaging (EDI) OCT, as well as OCTA. The maximum permissible spherical and cylindrical aberration was ±3 and ±1 diopters respectively. Imaging was performed on all subjects using the Zeiss Cirrus HD-OCT (AngioPlex, CIRRUS HD-OCT model 5000, Carl Zeiss Meditec, Inc., Dublin, CA, USA) by a single, trained operator (F.R.). Each imaging session included EDI-OCT scans (10 × 10 mm2) and OCTA (3 × 3 mm2) volumetric scans of the posterior pole. Only scans with a signal strength ≥7, centered on the fovea, and without motion, as well as segmentation and projection artifacts, were considered to guarantee standardized analysis [6,17].All iERM eyes underwent 23G pars plana vitrectomy with ERM and ILM peeling after staining with MembraneBlue-Dual® (DORC Int., Zuidland, The Netherlands) under general anesthesia. The procedures were performed by a single, experienced vitreoretinal surgeon (M.R.). The periphery was checked with scleral indentation and a fluid-air exchange was performed at the end of the surgery. In cases of phakic eyes, vitrectomy was combined with phacoemulsification and lens implantation in the capsular bag. Three months after treatment, all participants underwent another thorough examination of both eyes including BCVA, slit-lamp biomicroscopy, macular EDI-OCT as well as OCTA. All OCT/OCTA imaging sessions were performed around noon to avoid the influence of physiological diurnal variations of perfusion metrics.Total macular volume (TMV) and central retinal thickness (CRT) were automatically acquired by the device, as defined by the Early Treatment Diabetic Retinopathy Study in the OCT scans [18]. SFCT was measured manually in EDI-OCT scans just below the fovea, extending perpendicularly from the hyperreflective Bruch’s membrane layer to the inner scleral border. Manual measurements were performed by two experienced graders (J.A.M.S. and M.P.) who were blinded to the clinical information of the examined eyes and results were averaged. On baseline examination, the iERM was graded (M.P.B.) according to a previous published ERM staging system and alterations of the central foveal bouquet (CB) were described as published by Govetto et al. [19,20].After acquisition, OCTA data were manually segmented (F.R.) in all B-Scans to get retinal and choroidal sublayer slabs. To avoid artifacts, the retinal vessels were analyzed as full retinal slabs (ILM to retinal pigment epithelium), since subjects with ERM are prone to segmentation errors within the retinal layers (Figure 1A) [17]. The choroidal sublayers were manually segmented to get 20 µm slabs of choriocapillaris (CC) (Figure 1B), Sattler’s layer (SL) (Figure 1C), and Haller’s layer (HL) (Figure 1D) according to previous published protocols [8,15,21,22]. All acquired en face images were exported into ImageJ (NIH, Version 1.48b, Bethesda, USA) for thresholding. Binarization was done by the Otsu method, which is an automatic threshold selection from grey-level histograms [23]. As suggested by Nicolò et al., CC perfusion (CCP) was calculated by scoring the percentage of white pixels, while for SL perfusion (SLP) and HL perfusion (HLP) the values of black pixels were taken into account [24].Statistical calculations were performed using IBM SPSS (Version 24.0, Chicago, IL, USA) and Prism GraphPad (Version 8.0, La Jolla, CA, USA). Snellen BCVA was converted to the logarithm of the minimum angle of resolution (logMAR). Quantitative variables were summarized as mean and standard deviation (SD) and qualitative variables as frequency and percentage. The Shapiro–Wilk test was used to check for normality of the obtained data. Baseline and follow-up values of the same eye were compared using a paired t-test. Differences between iERM eyes and healthy fellow eyes were assessed by an unpaired t-test. A multiple regression was carried out to investigate whether retinal and choroidal OCT/-A baseline parameter could significantly predict the visual outcome after surgery. For all tests, values of p < 0.05 were considered statistically significant.A total of 63 eyes with unilateral iERM and 63 healthy fellow eyes were included in the analysis. Table 1 reports clinical and demographic characteristics of the enrolled patients. Twenty-eight (44.4%) female and 35 (55.6%) male participants were included, with a mean age of 70.79 ± 7.86 years.Anatomical and functional parameters of the iERM eyes and the fellow eyes at baseline examination are reported in Table 2. iERM eyes showed a statistically significantly worse BCVA and a higher TMV, as well as CRT, compared to the healthy fellow eyes. Perfusion values of the retina and the choroidal sublayers did not differ between both eyes.Table 3 presents the anatomical and functional findings of the iERM eyes before and 3 months after treatment. The follow up examination revealed a significantly improved BCVA (0.46 ± 0.27 logMAR to 0.19 ± 0.19 logMAR; p < 0.001), while CRT (470.4 ± 81.02 µm to 405.2 ± 51.83 µm; p < 0.001) and TMV (10.43 ± 1.08 mm3 to 9.46 ± 0.61 mm3; p < 0.001) decreased. Furthermore, CCP rose from 37.86 ± 5.55% to 39.4 ± 5.77% (p = 0.003) and SLP dropped from 68.54 ± 8.92% to 67.76 ± 8.65% (p = 0.014) 3 months after surgery.The untreated healthy fellow eyes did not show any statistically significant differences between baseline and follow up examination (Table 4). Comparison between affected eyes and healthy fellow eyes 3 months after treatment still showed significant differences in BCVA, CRT and TMV (Table 5). Subgroup analysis of phakic and pseudophakic patients revealed similar changes in terms of retinal and choroidal parameters due to combined phaco-vitrectomy or sole vitrectomy, respectively. The visual outcome 3 months after surgery in regard to BCVA improvement did not differ between both groups (phakic: −0.258 ± 0.183 logMAR, pseudophakic: −0.278 ± 0.193 logMAR, p = 0.96).Multiple regression analyses were carried out to investigate preoperative retinal and choroidal parameters as predictors of the postoperative BCVA in iERM eyes. The results of the regression analysis with retinal parameters indicated that the model explained 33.6% of the variance and that the model was a significant predictor of postoperative BCVA, F(3,33), p = 0.003. While TMV (B = 0.128, p = 0.004) contributed significantly to the model, CRT (B = 0.0003, p = 0.592) and FRP (B = −0.001, p = 0.860) did not (Figure 2). The regression analysis with choroidal baseline parameters indicated that the model explained 18.9% of the variance and that the model was a significant predictor of postoperative BCVA, F(4,55), p = 0.019. While SLP (B = −0.019, p = 0.006) contributed significantly to the model, CCP (B = −0.009, p = 0.214) and HLP (B = 0.005, p = 0.214) as well as SFCT (B = 0.001, p = 0.77) did not (Figure 3). Statistical significance of both regression models was independent of the lens status.ERM eyes were grouped by the stage of the disease and baseline SLP values were compared across the four groups. Analysis of variance did not reveal any statistically significant difference between the groups. Another multiple regression analysis was carried out to investigate ERM stage and baseline SLP as predictors of the postoperative BCVA in the ERM eyes. The results of the regression indicated that the model explained 26.2% of the variance and that the model was a significant predictor of postoperative BCVA, F(2,58), p < 0.001. Both, ERM stage (B = 0.057, p = 0.004) and SLP (B = −0.007, p = 0.004) contributed significantly to the model.In this prospective OCTA-based study, we investigated retinal and choroidal vascular alterations in eyes with iERM as well as their reversibility after pars plana vitrectomy with ERM-ILM peeling. Furthermore, regression models, based on retinal and choroidal parameters, were carried out to identify predictive prognostic markers for the postoperative visual outcome. To the best of our knowledge, this is the first study demonstrating significant changes of SLP after iERM-surgery and the prognostic value of preoperative SLP as a predictor for the postoperative BCVA.Idiopathic ERM is a frequent cause of reduced visual acuity with metamorphopsia and distinctly impairs patient quality of life [25]. ERM formation commonly exposes the macular to anteroposterior forces with macular thickening as well as tangential forces with vessel displacement [26,27]. Previous OCTA studies investigated alterations of the foveal avascular zone (FAZ) in eyes with ERM, demonstrating a smaller FAZ area in the superficial capillary plexus (SCP) and deep capillary plexus (DCP) [10,13,28]. They supposed that centripetal retinal displacement and the deformation of macular vascular integrity results in decreases of the FAZ area and parafoveal vessel density (VD) with a central vessel crowding. However, reported retinal perfusion values in eyes affected by ERM are quite different. Kim et al., as well as Chen et al., investigated the parafoveal VD in SCP and DCP, demonstrating reduced VD in both plexuses compared to unaffected fellow eyes [10,11]. Taking into account that mechanical stress due to ERM directly affects the inner retinal layers, it would be plausible that the SCP would be primary affected. However, Lin et al. found lower flow signals only in the DCP, while SCP only showed tortuous alterations [29]. Conversely, Nelis et al. demonstrated a significant increase of the foveal and parafoveal VD of iERM eyes in both retinal plexus. These overall controversial results in vessel quantification measurements could be due to false retinal layer segmentation. In ERM patients, the foveal architecture is distorted and slab segmentation is a serious issue, with segmentation errors occurring in 69.2% of ERM eyes [17,30]. Primarily the inner plexiform layer is prone to inaccurate segmentation; however, it is the most important layer for correct visualization and quantification of the SCP and DCP [6,7]. In order to avoid the potential bias of misalignment in the inner retina, we only investigated perfusion of the full-thickness retina slab in this study. ERM eyes showed a slightly lower mean FRP compared to unaffected fellow eyes at baseline examination, but without statistical significance (p = 0.383). Three months after surgery, overall retinal perfusion of the iERM group did not show any significant difference compared to preoperative values. Mastropasqua et al. evaluated SCP changes occurring after ERM-ILM peeling using OCTA [9]. Likewise, the SCP VD was stable throughout the follow-up examinations at 1 week and 1 month. However, analyzing individual sectors of SCP, they found a significant reduction of VD. In contrast to this study, Mastropasqua et al. evaluated automatically splatted 6 × 6 mm2 OCTA images; hence we may have missed out on those sectoral changes by analyzing the whole 3 × 3 mm2 en face image. Interestingly, another study group, who also investigated SCP perfusion changes after surgery, did not find significant perfusion changes after 1 month follow-up [12]. However, at 6 months follow-up, perfusion parameters significantly improved compared to baseline values. They suggested that forces exerted from ERM could lead to a temporary occlusion of small vessels and, after vitrectomy, the microvasculature needs a longer time to recover completely. We may have missed out on postoperative FRP changes by choosing a follow-up period of 3 months. However, previous studies demonstrated that the decisive retinal alterations occur within the first 3 months after surgery [31].So far, inner-retinal layer alterations have gained attention in eyes with ERM, because ERM is primarily an inner retinal disease. However, with the introduction of EDI-OCT, as well as OCTA, choroidal involvement of ERM can be assessed more accurately. In this study, we analyzed SFCT and perfusion values of choroidal substructures at baseline and 3 months after ERM-ILM peeling. Compared to the unaffected fellow eyes, iERM eyes showed slightly reduced SFCT and CCP without reaching statistical significance. However, 3 months after surgery, the choroid thickened (p = 0.006), while CCP increased (p = 0.003) and SLP decreased (p = 0.014). The influence of ERM on choroidal thickness values has been described by Michalewska et al., who demonstrated a decrease of SFCT 3 months after vitrectomy with ERM removal and ILM peeling, contrary to our data [32]. Kang et al., who also investigated SFCT in patients after vitrectomy for ERM, did not find any significant difference in SFCT values after surgery [33]. In our study, the thickening of the SFCT could be related to the significant increase in CCP, since alterations in choroidal thickness reflect the status of the vasculature within [34]. Yu et al. support these findings as they reported a reduced CCP in ERM eyes compared to unaffected fellow eyes, which was reversible by surgery [14].So far it is unclear, to what extent ERM-associated traction influences the microvasculature in SL and HL, since the vessel size increases from CC to HL. To the best of our knowledge, this is the first study reporting a significant decrease of SLP 3 months after surgery, while HLP stayed steady. An influence of ERM on deeper choroidal sublayers has been reported in terms of changes in the usual diurnal pattern of SLP an HLP [15]. ERM might lead to choroidal vascular involvement through a variety of possible mechanisms. The anteroposterior forces can affect the retinal pigment epithelium (RPE) as well as the choroid directly by focal dilation of choroidal vessel [16,35]. Through the release of mechanical stretching of the RPE after ERM-ILM removal, the local level of vascular endothelial growth factor (VEGF) could change, which might lead to alterations in choroidal permeability and vascular rearrangement [36]. The remodeling of the physiological conditions after surgery could possibly explain the increase of CCP and decrease of SLP as a shift of blood flow from the deeper choroid to the CC. Another factor that must be taken into account is the status of vitreous oxygenation after vitrectomy. The retina as well as the choroid is exposed to a higher oxygenated environment after removal of the oxygen-consuming vitreous, what may result in alteration of perfusion. It should be mentioned, that unaffected fellow eyes did not show any changes in terms of anatomical or functional parameters 3 months after baseline examination.Furthermore, we carried out multiple regression models to investigate preoperative image parameters as predictors of the postoperative visual outcome in iERM eyes. TMV contributed significantly to the model and showed positive correlation with the postoperative BCVA (p = 0.004); thus a higher preoperative TMV led to worse visual outcome. The increase in TMV seen in eyes with ERM generally occurs from strong anteroposterior forces. Patients with lower TMV and less CRT are likely to suffer from an ERM with little traction. Therefore, these patients with low tractive ERM are unlikely to benefit from surgery.In the regression analysis of preoperative choroidal parameters, SLP showed a statistically significant negative correlation with postoperative BCVA. In other words, higher preoperative SLP led to a better visual outcome 3 months after surgery. As we have suggested before, the release of mechanical tension from the ERM might lead to a remodeling of the physiological choroidal conditions with a perfusion shift from SL to CC. Therefore, the preoperative SLP could reflect the capacity of choroidal blood flow, which is able to shift into the smaller vessels of the CC. Interestingly, baseline SLP seems to be independent from the ERM stage. Therefore, it can be used as a universal biomarker. To the best of our knowledge, this is the first study reporting choroidal perfusion metrics as predictors of visual outcome after surgery for iERM.The present study has some limitations, additional to those already discussed. The follow-up period after surgery was relatively short, since morphological changes could last up to 12 months after ERM peeling [31]. However, it has been reported that the decisive retinal alterations occur within the first 3 months after surgery. Furthermore, we included all stages of ERM, in order to avoid selection bias. However, severe retinal distortion due to thick and hyperreflective ERM could confound OCTA signals and measurements depending on the ERM stage. Finally, the manual record of SFCT may represent a potential bias. To reduce this possible confounding factor, all manual measurements were performed by two experienced graders in a masked fashion and the average values were used for statistical analysis.In conclusion, we were able to demonstrate significant choroidal sublayer perfusion changes after surgery for iERM with an increase of CCP and a decrease of SLP, by using OCTA. Preoperative SLP could be a useful predictive marker for functional results, since a significant correlation with postoperative visual acuity was found.Conceptualization, F.B. and M.R.; Data curation, F.R. and M.P.B.; Formal analysis, F.R., M.P.B. and M.P.; Investigation, F.R., M.P.B., J.A.M.S. and M.P.; Methodology, F.R. and M.R.; Supervision, S.G. and M.R.; Validation, F.R., M.P.B., M.P. and M.R.; Writing—original draft, F.R. and M.R.; Writing—review & editing, F.R., S.G. and M.R. All authors have read and agreed to the published version of the manuscript.This work was supported by the University of Lübeck (J01-2020).The authors declare no conflict of interest.Optical coherence tomography angiography (OCTA)-Imaging of the posterior pole in a subject with idiopathic epiretinal membrane (iERM). En face angiogram and corresponding B-scan at the level of the full retina (A), choriocapillaris (B), Sattler’s layer (C), and Haller’s layer (D).Linear regression model showing the 3-months (3M) best corrected visual acuity (BCVA) dependency of each central retinal thickness (CRT) (A), total macular volume (TMV) (B), and full retinal perfusion (FRP) (C) at baseline (BL). Black line represents the linear regression line; dotted line shows the upper and lower bound of the 95% confidence interval. * signifies p < 0.05.Linear regression model showing the 3-months (3M) BCVA dependency of each SFCT (A), CCP (B), SLP (C), and HLP (D) at baseline (BL). Black line represents the linear regression line; dotted line shows the upper and lower bound of the 95% confidence interval. * signifies p < 0.05.Demographic and clinical characteristics of enrolled patients with iERM.iERM: idiopathic epiretinal membrane; SD: standard deviation; CB: central foveal bouquet.Anatomical and functional parameters of iERM and fellow eyes at baseline examination.1 unpaired t-test; all data are presented as mean ± standard deviation; iERM: idiopathic epiretinal membrane; AL: axial length; BCVA: best corrected visual acuity; CRT: central retinal thickness; TMV: total macular volume; SFCT: subfoveal choroidal; FRP: full retinal perfusion; CCP: choriocapillaris perfusion; SLP: Sattler’s layer perfusion; HLP: Haller’s layer perfusion.Anatomical and functional parameters of iERM eyes before and 3 months after surgery.1 paired t-test; all data are presented as mean ± standard deviation; iERM: idiopathic epiretinal membrane; BCVA: best corrected visual acuity; CRT: central retinal thickness; TMV: total macular volume; SFCT: subfoveal choroidal; FRP: full retinal perfusion; CCP: choriocapillaris perfusion; SLP: Sattler’s layer perfusion; HLP: Haller’s layer perfusion.Anatomical and functional parameters of healthy fellow eyes at baseline and 3 months follow up examination.1 paired t-test; all data are presented as mean ± standard deviation; iERM: idiopathic epiretinal membrane; BCVA: best corrected visual acuity; CRT: central retinal thickness; TMV: total macular volume; SFCT: subfoveal choroidal; FRP: full retinal perfusion; CCP: choriocapillaris perfusion; SLP: Sattler’s layer perfusion; HLP: Haller’s layer perfusion.Anatomical and functional parameters of iERM and fellow eyes at 3 months follow up examination.1 unpaired t-test; all data are presented as mean ± standard deviation; iERM: idiopathic epiretinal membrane; BCVA: best corrected visual acuity; CRT: central retinal thickness; TMV: total macular volume; SFCT: subfoveal choroidal; FRP: full retinal perfusion; CCP: choriocapillaris perfusion; SLP: Sattler’s layer perfusion; HLP: Haller’s layer perfusion.

Med-MDPI/jcm_2/jcm-09-06-01769.txt

@@ -0,0 +1 @@
+Pembrolizumab, an anti-programmed cell death (PD)-1 monoclonal antibody, is an anticancer agent showing substantial benefit in lung cancer and melanoma treatment. Biliary tract cancer (BTC) has been shown to respond to pembrolizumab; however, no credible data of such treatment outcomes exist. Therefore, we assessed the clinical outcomes and safety of pembrolizumab in patients with gemcitabine/cisplatin-refractory BTC. In this multicenter study, we retrospectively analyzed 51 patients with programmed cell death 1-ligand 1 (PD-L1)-positive gemcitabine/cisplatin-refractory BTC treated with pembrolizumab in four tertiary hospitals in Korea. PD-L1 positivity was defined as the expression of PD-L1 in ≥1% of tumor cells based on immunohistochemical staining (22C3, SP263, and E1L3N assays). The median age of the patients was 66 (range, 43–83) years and 29 (56.9%) were male. Extrahepatic cholangiocarcinoma was the most common cancer type (n = 30, 58.8%). Partial response and stable disease were achieved in 5 (9.8%) and 13 (25.5%) patients, respectively. Median progression-free survival and overall survival were 2.1 (95% CI, 1.7–2.4) and 6.9 (95% CI, 5.4–8.3) months, respectively. Overall, 30 (58.8%) patients experienced treatment-related adverse events (AEs). Only four (7.8%) patients experienced grades 3 and 4 AEs. In PD-L1-positive gemcitabine/cisplatin-refractory BTC, pembrolizumab presented durable efficacy, with a 9.8% response rate and manageable toxicity.Biliary tract cancers (BTCs) consist of all tumors originating in the epithelial cells of the bile duct and they are classified into intrahepatic cholangiocarcinoma (CCA), extrahepatic CCA (including perihilar and distal CCA), and gallbladder cancer, according to their anatomical location within the biliary tree. More than 90% of BTCs are adenocarcinoma. Although the incidence of BTC is the highest in East and South Asia and parts of South America, its global incidence has been increasing significantly [1,2]. Surgical resection is regarded as the curative treatment option for patients with BTC but most patients are not considered suitable candidates for surgery because of an advanced disease status at diagnosis [3,4]. Patients with advanced BTC have a poor prognosis with a dismal survival time of <1 year [5,6]. Recent updated guidelines recommend gemcitabine plus cisplatin as the first-line chemotherapy for patients with advanced BTC [4,7]. However, effective treatment options for patients with gemcitabine/cisplatin-refractory BTC are currently limited [8].Pembrolizumab, an anti-programmed cell death (PD)-1 monoclonal antibody, is an anticancer agent that has shown substantial benefit in patients with lung cancer and melanoma [9,10,11]. Following the discovery of the PD-1/PD-1-ligand 1 (PD-L1) pathway as an immunological target to control cancer, immune checkpoint inhibitors, including pembrolizumab, have been designed to overcome the inhibitory responses of cytotoxic T lymphocytes to promote antitumor immune responses [12,13,14]. The first related study showed a notable clinical effect of pembrolizumab in patients with non-colorectal cancer, including four with BTC with mismatch repair (MMR) deficiency, with a 71% objective response rate (ORR) [13]. In a few subsequent studies, pembrolizumab was reported to be effective against advanced BTC, targeting a subpopulation with more than 1% PD-L1 expression in tumors [15,16]. However, there are no credible data of the treatment outcomes of pembrolizumab against BTC. In the present multicenter retrospective study, we investigated the clinical outcomes and safety profile of pembrolizumab in patients with gemcitabine/cisplatin-refractory BTC using PD-L1 positivity as a response biomarker and attempted to identify a subpopulation that would potentially benefit from this treatment.From August 2017 to April 2019, 65 patients with PD-L1-positive gemcitabine/cisplatin-refractory BTC who received pembrolizumab were included in this multicenter study with the following inclusion criteria: (1) ≥20 years of age; (2) histologically or cytologically confirmed locally advanced or metastatic BTC; (3) at least one measurable or evaluable lesion according to Response Evaluation Criteria in Solid Tumors (RECIST) version 1.1 [17]; (4) radiologically confirmed presence of progressive disease or intolerance to gemcitabine/cisplatin chemotherapy; (5) ≥1% PD-L1-positive tumor cells as assessed using immunohistochemical (IHC) staining; and (6) receipt of at least one cycle of pembrolizumab.Fourteen patients were excluded according to the following criteria: (1) diagnosis of ampulla of Vater cancer (n = 1); (2) inability to assess tumor response (n = 6); (3) no gemcitabine/cisplatin chemotherapy before pembrolizumab injection (n = 4); and (4) no PD-L1 IHC-staining examination (n = 3). Fifty-one patients (41 from Severance Hospital, Seoul, Korea; 4 each from Seoul National University Hospital, Seoul, Korea, and National Cancer Center, Goyang, Korea; and 2 from Pusan National University Hospital, Busan, Korea) were included in the final analysis. Among them, 44 patients (86.3%) were radiologically confirmed to have progressive disease, and another 7 (13.7%) were intolerant to gemcitabine/cisplatin chemotherapy. All the clinical, laboratory and radiologic data were collected from electronic medical records and were retrospectively reviewed without obtaining informed consent. The study was approved by the institutional review board of each institution (Severance Hospital Institutional Review Board, Seoul National University Hospital Institutional Review Board, National Cancer Center Institutional Review Board, and Institutional Review Board of Pusan National University Hospital).PD-L1 expression was assessed by conducting an IHC staining of achieved tumor tissues before systemic treatment, using the E1L3N (Cell Signaling Technology, Danvers, MA, USA), 22C3 (Agilent Technologies, Santa Clara, CA, USA), and SP263 (Ventana Benchmark Ultra, Tuscon, AZ, USA) assays. PD-L1-positive tumor cells were considered if the viable tumor cells exhibited any perceptible, partial or complete, membranous or cytoplasmic staining, as previously described [18]. PD-L1-positive status was defined based on a 1% threshold in immunostained tumor cells in the entire tumor section by any IHC method. The frequency of use of the E1L3N, 22C3, and SP263 IHC assays was as follows: n = 33 (57.9%), 27 (47.4%), and 9 (15.8%), respectively. PD-L1 expression was categorized into three subgroups based on the proportion of immunostained tumor cells using additional cutoff values of 5% and 50%.All patients received 200 mg pembrolizumab intravenously, every 3 weeks after a 17-day median interval from the last prior treatment. Dose reduction, administration delay, or both were performed if serious treatment-related adverse events (AEs) developed, making treatment intolerable. Pembrolizumab administration was interrupted when disease progression or life-threatening AEs were identified. To evaluate treatment efficacy, we routinely evaluated the tumor response every three cycles using abdominal or chest computed tomography or both according to RECIST version 1.1 [17].To monitor treatment-related AEs, physicians and registered nurses meticulously evaluated the occurrence of AEs at each visit during treatment. The category and severity grade of the AEs were accurately recorded in the medical records of the patients. Treatment-related AEs were evaluated according to National Cancer Institute Common Toxicity Criteria version 4.0. Treatment delays or discontinuations associated with the AEs were also documented with the reason.The primary endpoint of this study was response rate and the secondary endpoints were AEs, progression-free survival (PFS), and overall survival (OS). Tumor responses included complete response (CR), partial response (PR), stable disease (SD), and progressive disease (PD). DCR was defined as the summation of CR, PR, and SD. PFS was defined as the time from the initiation of pembrolizumab treatment to disease progression or date of death, and OS was defined as the time from the initiation of pembrolizumab treatment to the date of death due to any cause or the last follow-up visit. The last follow-up date was 31 September 2019, and the median follow-up duration was 3.8 (range, 0.6–18.4) months.Data are expressed as the mean ± standard deviation, median (range), or n (%), as appropriate. A survival analysis for PFS and OS was performed using the Kaplan-Meier method and compared using the log-rank test. To identify independent risk factors for progression, we performed a multivariate Cox proportional hazard regression analysis using the significant variables in the univariate analysis. Hazard ratios (HRs) and the corresponding 95% confidence intervals (CI) were also calculated. A two-tailed p-value of 0.05 was considered statistically significant, and the statistical analysis was performed using SPSS version 25.0 (PASW Statistics Inc., Chicago, IL, USA).The baseline characteristics of the 51 enrolled patients are presented in Table 1. The median age was 66 (range, 43–83) years and 29 (56.9%) patients were male. Extrahepatic CCA was the most common type of cancer (n = 30, 58.8%), followed by intrahepatic CCA (n = 12, 23.5%) and gallbladder cancer (n = 9, 17.6%). All patients were diagnosed with adenocarcinoma, histologically or cytologically. Most patients were diagnosed at a metastatic stage (n = 45, 88.2%), including 28 with post-operative recurrence. Common organs of distant metastasis were as follows: liver (n = 24, 47.1%), intra-abdominal lymph node (n = 24, 47.1%), peritoneal seeding (n = 17, 33.3%), bone (n = 4, 7.8%), and lungs (n = 2, 3.9%). Histological grading was moderately and poorly differentiated in 28 (54.9%) and 14 (27.5%) patients, respectively. Subgrouping based on the proportion of PD-L1-positive tumor cells with cutoff values of 5% and 50% revealed that low (1–5%), moderate (5–50%), and high (≥50%) PD-L1 subgroups accounted for 35 (68.6%), 11 (21.6%), and 5 (9.8%) patients, respectively. The median carbohydrate antigen (CA) 19-9 level was 276.6 U/mL (range, 1.1–24,253.0). Pembrolizumab was administered as the second, third, and fourth or greater line chemotherapy in 33 (64.7%), 14 (27.5%), and 4 (7.8%) patients, respectively. A median of three cycles (range 1–15) of pembrolizumab was administered during a median of 3.8 (range, 0.6–18.4) months of follow-up.In the 51 patients enrolled, PR and SD were achieved in 5 (9.8%) and 14 (25.5%) patients, respectively (Table 2), whereas the DCR was 35.2%. In five patients with PR, the median time to response was 2.5 (range, 2.1–8.3) months and the median duration of treatment was 9.1 months. Five patients with PR exhibited 1, 2, 5, 10, and 50% of PD-L1 expression level. Overall, the median PFS and OS were 2.1 (95% CI, 1.7–2.4) and 6.9 (95% CI, 5.4–8.3) months, respectively (Figure 1). Figure 2 illustrates the changes in tumor size from the baseline to the best response in 42 patients. When the study population was divided into three subgroups based on PD-L1 expression, this did not appear to be significantly associated with the best tumor response (Figure 2) Overall, 30 (58.8%) patients experienced treatment-related AEs, with fatigue (n = 11, 21.6%, Table 3) being the most common. Four (7.8%) patients experienced grade 3 and 4 AEs (consisting of grade 3 nausea, fatigue, poor oral intake, diarrhea, and drug-induced pneumonitis) and two patients (3.9%) discontinued pembrolizumab treatment owing to AEs. There was no case of treatment-related death.Univariate analysis showed that the number of prior therapies (≥2), hemoglobin level (<10 g/dL), and CA 19-9 level (>500 U/mL) were significantly associated with progression (Table 4). Subgroups based on PD-L1 expression could not be used to predict progression after pembrolizumab treatment. Although the subsequent multivariate analysis did not identify statistically significant predictors of progression, patients receiving pembrolizumab as second-line chemotherapy with a normal hemoglobin level tended to exhibit favorable PFS (p = 0.069 and p = 0.062, respectively).In this multicenter study, pembrolizumab showed durable antitumor activity, with a 9.8% response rate and manageable toxicity in patients with PD-L1-positive gemcitabine/cisplatin-refractory BTCs. The first BTC cohort from the KEYNOTE-028 open-label, phase I clinical trial, which included patients with 20 different types of solid cancers with PD-L1 expression of >1% (determined by IHC-based assay), revealed an ORR of 17% (one CR and three PR) and a DCR of 34% (four additional SD) among 23 patients with BTC [15,19]. The results of another prospective cohort study of 40 patients with BTC with similar characteristics in a single medical institution showed an OCR of 10.0% (four patients with PR) and a DCR of 47.5% (additional 15 patients with SD) [16]. Our study had a larger population scale than these two studies but showed a similar overall response rate (Table 5). Intriguingly, some patients who responded showed a long duration of treatment: >6 months (median 9.1 months in our study; 6.3 months [16] and 2-year duration of treatment of 67% [15] and 50% [20] patients, respectively, in other studies).The interim report of KEYNOTE-158, the largest ongoing phase II, single-arm, open-label cohort study of pembrolizumab in patients with various types of advanced cancers, including 104 patients with BTC, was published in 2018 [20]. The 104 patients with BTC showed an ORR of only 5.8%, including 0, 6, and 17 patients with CR, PR, and SD, respectively (DCR was 22.1%) [20]. This cohort study included 61 and 31 patients with PD-L1-positive and PD-L1-negative expression, respectively. Interestingly, PD-L1 expression did not predict treatment response and this is similar to the findings of our study. Although the ORR was slightly higher in the PD-L1-positive group than in the negative group (6.6% vs. 2.9%), there were no significant differences in the median PFS (1.9 vs. 2.1 months) or OS (7.2 vs. 9.6 months) [20]. This result, which showed the limited clinical efficacy of pembrolizumab monotherapy in advanced BTC, encouraged us to explore biomarkers to identify a subpopulation likely to respond and a new therapeutic strategy to overcome the limited antitumor response in BTC.The IHC analysis of PD-L1 expression predicts the clinical efficacy of immune checkpoint inhibitors in several cancers [16,21,22,23]. In particular, in non-small-cell lung cancer, the cutoff value of PD-L1 was lowered from 50% to 1%, and immune checkpoint inhibitors are now considered as the first-line treatment for advanced stage tumor with or without conventional chemotherapy [21,23,24]. Recently, Kang et al. suggested that a high cutoff value of ≥50% could predict tumor response when evaluated by immune-modified RECIST criteria [16]. In addition, a Japanese phase I study of another PD-L1 inhibitor, nivolumab, reported that PD-L1-positive patients (≥1%) had longer median PFS (2.8 vs. 1.4 months) and OS (11.6 vs. 5.2 months) than those of PD-L1-negative patients among 30 patients with gemcitabine-refractory advanced BTC [25]. These findings, however, resulted from preliminary studies with relatively small-sized study populations and did not correspond with the results of KEYNOTE-158 or our study [20]. Furthermore, our analysis of several clinical factors to validate another response predictor of pembrolizumab revealed few prior therapies and normal hemoglobin level as predictors of a favorable tendency to respond, without statistical significance in the multivariate analysis. Hence, the usefulness of these factors, including tumor PD-L1 status, should be verified in large-scale, prospective clinical trials in the future.An established biomarker for patients’ response to immune checkpoint inhibitors is the neo-antigen tumor burden resulting from microsatellite instability (MSI) owing to MMR deficiency (MSI-high/MMR-deficiency) [26]. The genomic spectra of BTC have shown the presence of MSI in a small portion of approximately 1–2% [26,27,28,29]. The clinical results of pembrolizumab monotherapy in advanced MSI-high/MMR-deficient cancers, including 4 and 11 patients with BTC, demonstrated ORR values of 25% and 27%, respectively, for a subset of patients with BTC with a response duration of >6 months [13,30]. However, we could retrospectively assess the MSI status in seven (13.7%) patients in our study population, none with MSI-H/MMR-deficiency.To evaluate the significant response biomarker, the KEYNOTE-028 (NCT02054806) study analyzed multiple biomarkers including expression of an 18-gene T cell-inflamed profile, PD-L1 expression, and tumor mutational burden. All three factors (alone or combined) were correlated with high response rate to pembrolizumab in the overall study population of multiple tumor types [19]. However, biomarker data specific for BTC has not been indubitably presented until now.The safety profile of pembrolizumab in this study was generally consistent with that in previous studies [9,13,15,16]. Four (7.8%) patients experienced grade 3 AEs with 50% (n = 2, 3.9%) treatment discontinuation in our study. Two patients with pneumonitis associated with immune checkpoint inhibition were conservatively managed with one week of hospitalization. There are some limitations in our study. Although all patients in the study were enrolled in a number of tertiary care hospitals, the possibility of selection bias cannot be excluded due to its retrospective nature. The relatively small study population may also be insufficient to perform a statistically robust analysis. Additionally, inconsistent IHC assay methods could serve as a limitation for further investigation.In conclusion, pembrolizumab monotherapy in patients with gemcitabine/cisplatin-refractory BTC and PD-L1 positivity showed long-lasting anticancer effects in approximately 10% of the study population with an overall manageable safety profile. Intensity of PD-L1 expression was not supposed to an effective surrogate marker for the efficacy of pembrolizumab; therefore, further investigation is needed to confirm an appropriate biomarker for selecting the subpopulation of advanced BTC patients who might be sensitive to immune checkpoint inhibitors.Conceptualization, S.B.; methodology, S.H.L. (Sang Hoon Lee) and H.S.L.; validation, S.H.L. (Sang Hyub Lee), S.M.W. and D.U.K.; formal analysis, S.H.L. (Sang Hoon Lee) and H.S.L.; investigation, S.H.L. (Sang Hoon Lee); data curation, S.H.L. (Sang Hoon Lee), H.S.L., S.H.L. (Sang Hyub Lee), S.M.W. and D.U.K.; writing—original draft preparation, S.H.L. (Sang Hoon Lee); writing—review and editing, S.B.; supervision, H.S.L. and S.B.; project administration, S.B. All authors have read and agreed to the published version of the manuscript.This research received no external funding.The authors declare no conflicts of interest.Kaplan-Meier estimates of the (A) progression-free survival (PFS); and (B) overall survival (OS) of the entire study population (n = 51). Median PFS and OS were 2.1 (95% confidence interval (CI), 1.7–2.4) and 6.9 (95% CI, 5.4–8.3) months.Tumor response and programmed death-ligand 1 (PD-L1) subgroups of evaluable patients (n = 42). The highest percentage changes in the size of tumor target lesions from the baseline, measured using Response Evaluation Criteria in Solid Tumors (RECIST) 1.1 for the 42 evaluable patients, are shown by PD-L1 subgroups. * Eleven patients with progressive disease based on RECIST 1.1 due to the unequivocal progression of non-target lesions despite <20% growth in target lesions.Baseline characteristics of patients.Patients, n = 51, † including 28 with post-operative recurrence. ‡ One patient who concurrently received palliative radiation therapy for cervical LN metastasis. Abbreviations: ECOG, Eastern Cooperative Oncology Group; PD-L1, programmed cell death 1-ligand 1; AST, aspartate aminotransferase; ALT, alanine aminotransferase; CA 19-9, carbohydrate antigen 19-9.Clinical outcomes.Patients, n = 51, † Evaluation in five patients with a partial response. Adverse events (AEs).Patients, n = 51; AST, aspartate aminotransferase; ALT, alanine aminotransferase.Predictors of progression after pembrolizumab treatment.Patients, n = 51; † excluding six cases of unknown histological grading. Abbreviations: HR, hazard ratio; CI, confidence interval; NS, not significant; CCA, cholangiocarcinoma; PD-L1, programmed cell death 1-ligand 1; CA 19-9, carbohydrate antigen 19-9.Reported data of pembrolizumab treatment in patients with gemcitabine/cisplatin-refractory biliary tract cancer (BTC).Abbreviations: BTC, biliary tract cancer; ORR, objective response rate; DCR, disease control rate; PFS, progression-free survival; OS, overall survival; AE, adverse event.

Med-MDPI/jcm_2/jcm-09-06-01770.txt

@@ -0,0 +1 @@
+COVID-19, the illness caused by infection with the novel coronavirus SARS-CoV-2, is a rapidly spreading global pandemic in urgent need of effective treatments. Here we present a comprehensive examination of the host- and virus-targeted functions of the flavonolignan silibinin, a potential drug candidate against COVID-19/SARS-CoV-2. As a direct inhibitor of STAT3—a master checkpoint regulator of inflammatory cytokine signaling and immune response—silibinin might be expected to phenotypically integrate the mechanisms of action of IL-6-targeted monoclonal antibodies and pan-JAK1/2 inhibitors to limit the cytokine storm and T-cell lymphopenia in the clinical setting of severe COVID-19. As a computationally predicted, remdesivir-like inhibitor of RNA-dependent RNA polymerase (RdRp)—the central component of the replication/transcription machinery of SARS-CoV-2—silibinin is expected to reduce viral load and impede delayed interferon responses. The dual ability of silibinin to target both the host cytokine storm and the virus replication machinery provides a strong rationale for the clinical testing of silibinin against the COVID-19 global public health emergency. A randomized, open-label, phase II multicentric clinical trial (SIL-COVID19) will evaluate the therapeutic efficacy of silibinin in the prevention of acute respiratory distress syndrome in moderate-to-severe COVID-19-positive onco-hematological patients at the Catalan Institute of Oncology in Catalonia, Spain.The World Health Organization (WHO) has declared coronavirus disease 2019 (COVID-19) a public health emergency of international concern [1]. The causative agent of the COVID-19 outbreak is the severe acute respiratory syndrome (SARS)-associated coronavirus 2 (SARS-CoV-2), a novel enveloped RNA betacoronavirus [2]. No antiviral drugs are yet available with proven efficacy for SARS-CoV-2 treatment or prophylactic methods to successfully prevent the progression of SARS-CoV-2-driven acute respiratory distress syndrome (ARDS), one of the leading causes of mortality in patients with severe COVID-19. From the early clinical experience in China, we learned that the vast majority of COVID-19 patients (>90%) received a diagnosis of pneumonia during hospital admission, followed by ARDS in the more severe cases [3]. A majority of diagnosed patients had peripheral lymphocytopenia, thrombocytopenia, and leukopenia. Patients with severe disease had also elevated levels of serum C-reactive protein and, less commonly, augmented levels of liver transaminases. Laboratory analysis aiming to distinguish severe from mild disease suggested that circulatory inflammatory markers including interleukin (IL)-6, ferritin, and D-Dimer were closely related to severe COVID-19 in adults. In fact, their combined detection had the highest specificity and sensitivity for early prediction of disease severity in patients [4]. These initial findings were consistent with a clinical scenario in which the presence of hypercytokinemia (cytokine storm syndrome (CSS)) and lymphophenia have a main causal role during the transition from first COVID-19 symptoms to viral sepsis and inflammation-induced lung injury, ultimately to pneumonitis, ARDS, respiratory/multiple organ failure, shock, and potentially death. In the ensuing months, it has been confirmed that a subgroup of patients with severe disease have CSS that rages 7–10 days after disease onset when the ARDS peaks [5,6]. Immune dysregulation, rather than the level of peak viremia, appears to be one of the major mechanisms through which SARS-CoV-2 infection drives an insufficient (too-little-too-late) type I-interferon (IFN) innate immune response, later accompanied by aberrant proinflammatory cytokine secretion from alveolar macrophages that ultimately causes lung damage and reduces survival. Not surprisingly, selective (e.g., IL-6-targeted monoclonal antibodies such as tocilizumab) and non-selective blockade of pro-inflammatory cytokines (e.g., using pan-JAK1/2 inhibitors such as baricitinib) are under evaluation in ongoing clinical trials aimed to reduce the SARS-CoV-2-driven CSS, ameliorate pulmonary inflammation and respiratory distress, and hopefully improve mortality [7,8,9,10,11,12,13,14,15]. An optimal therapeutic approach to manage COVID-19 would involve a drug capable of preventing the CSS that dampens adaptive immunity against SARS-CoV-2 while at the same time directly targeting the key molecular machinery driving the virus lifecycle. We here propose that the flavonolignan silibinin—the major bioactive component of the silymarin extract obtained from the seeds of the milk thistle herb (Silybum marianum)—[16,17,18] might fulfill such requirements by reducing STAT3-related lung and systemic inflammation in the infected host and directly targeting the RNA replication machinery in the virus. Originally described as a remedy for the bites of poisonous snakes more than 2000 years ago, the use of silibinin-containing nutraceuticals to treat liver toxicity (e.g., alcoholic and non-alcoholic liver disease), drug-induced liver injury, cirrhosis, mushroom poisoning, and viral hepatitis has been well documented over the last 40 years [19,20]. Since 2013, there has been an ever-growing (molecular) understanding and (clinical) evaluation of the capacity of silibinin to inhibit cell growth of cultured cancer cells and tumor xenografts, to enhance the efficacy of other anti-cancer agents, to reduce the toxicity of cancer treatments, and to prevent and overcome the emergence of cancer drug resistance [21,22,23,24,25,26]. When used with more bioavailable nutraceutical formulations [27], silibinin has proved successful in advanced systemic cancer, a therapeutic activity that was particularly notable in the central nervous system where it provided greater than 4-fold survival benefit in patients with established brain metastases [28]. Importantly, the groundbreaking clinical activity of silibinin was accompanied by low toxicity and reversible secondary effects, and was compatible with the standard-of-care in oncology patients. Investigations into the molecular mechanisms involved in the antioxidant, immunomodulatory, antifibrotic, anticancer, and antiviral activities of silibinin have consistently suggested its ability to function as a natural down-modulator of the signal transducer and activator of transcription (STAT3) [29,30,31,32,33,34,35]. Using computational and experimental approaches, we recently delineated the molecular bases of the silibinin-STAT3 interaction. Silibinin synergistically works against STAT3 function by directly blocking the STAT3 Src homology-2 (SH2) domain, which is crucial for both STAT3 phospho-activation and nuclear translocation, and directly targeting also the STAT3 DNA-binding domain (DBD) to prevent the transcriptional activity of STAT3 irrespective of its activation/dimerization status [36]. Silibinin is multi-faceted and impedes the activation, dimerization, nuclear translocation, DNA-binding, and transcriptional activity of STAT3, thereby circumventing both the intrinsic difficulty of efficaciously disrupting protein–protein interactions over a large surface such as those involving SH2-mediated STAT3 dimerization and the previously thought undruggable nature of the STAT3 DBD [36]. Accordingly, cells engineered to overexpress a constitutively active form of STAT3 that dimerizes, binds to DNA and activates transcription spontaneously remain largely unresponsive to the transcriptional and phenotypic effects of silibinin (28, and unpublished observations). Importantly, the unique behavior of silibinin as a bimodal SH2- and DBD-STAT3 transcription factor inhibitor in vitro, in situ, and in vivo [28,36] largely explains its ability to fully prevent the hyper-activation of STAT3 imposed by the excessive production of IL-6, a key driver of the dysregulated inflammation in patients suffering from severe COVID-19. Early studies in lung injury models suggested that STAT3 activation plays a central role during the acute lung inflammatory response to lung tissue damage in a macrophage- and neutrophil-dependent manner [36,37]. Increased alveolar epithelial cell death and phosphorylated STAT3 were common phenotypic traits in patients with ARDS, suggesting the feasibility of targeting STAT3 to modulate pulmonary inflammatory responses. In support of this notion, attenuation of STAT3 signaling suppressed the excessive production of pro-inflammatory cytokines in macrophages and inflammatory cells in several pre-clinical and clinical models, including hyperoxia-, lipopolysaccharide (LPS)-, and sepsis-induced acute lung injury, while additionally promoting lung repair [38,39,40,41,42,43,44,45]. Although scarce, evidence from the phylogenetically-related virus SARS-CoV suggested that STAT3 could operate as a key signaling mediator of the link between virus replication and lung fibrotic responses [46].Silibinin has proven therapeutic capacity to protect damaged tissues by regulating the reactive intensity of cells tasked with establishing a repair program (e.g., macrophages, T-cells, and astrocytes) (Figure 1). Silibinin pre-treatment in mice significantly inhibits LPS-induced recruitment of airway inflammatory cells (macrophages, T-cells, and neutrophils) as well as the production of specific pro-inflammatory cytokines (i.e., IL-1β, TNFα), thereby protecting against lung injury [47,48]. In a mouse model of radiation therapy for lung cancer treatment, which partially mimics the late-phase inflammation and end-stage pulmonary fibrosis related to ARDS in severe COVID-19, silibinin was found to reduce inflammatory cell infiltration in the respiratory tract, to ameliorate inflammation and fibrosis, and to increase survival [49]. The efficacy of silibinin in localized lung tumors was originally found to involve the inhibition of the production and secretion of cytokines from tumor-associated macrophages in a STAT3-related manner [30]. In brain metastases, the suppressive effects of silibinin can be explained in terms of its ability to halt the pro-metastatic program driven by STAT3 in a subpopulation of reactive astrocytes surrounding metastastic lesions [28] (Figure 1). Accordingly, silibinin-inhibited STAT3 signaling in reactive (astrocytes) cells to damaged (brain metastatic cancer) cells blocks the cytokine secretome of the former to influence immunity responses against metastatic cells, including changes in the activation of CD8+ T-cells [28]. Early mouse models of SARS-CoV infection predicted that, in patients infected with pathogenic coronaviruses and perhaps other respiratory viruses, the rapid kinetics of viral replication accompanied by delayed type I IFN-I signaling would lead to the pathogenic accumulation of inflammatory monocyte-macrophages (IMMs), resulting in elevated levels of lung cytokines/chemokines (cytokine storm) and impaired virus-specific T-cell responses [50,51]. Such a link between dysregulated inflammatory responses, lung immunopathology, and diminished survival has been confirmed in the present pandemic of COVID-19. Accordingly, the intensity level of the interaction occurring between damaged lung epithelial cells and reactive IMMs (which switch their phenotype from suppressive/protective to stimulatory/destructive) during the course of SARS-CoV-2-driven systemic inflammation appears to determine, at least in part, the degree of the disease severity in patients [5,6] (Figure 1). In moderate COVID-19 cases, bronchoalveolar macrophage-epithelial interactions promote an increase in IL-6 and a decrease in the counts of total T-cells, particularly CD4+ and CD8+ T-cells. In severe COVID-19 cases, the macrophage-epithelial interaction promotes a further augmentation of IL-6 (and IL-2R, IL-10, and TNFα), whereas CD4+ (including IFNγ-expressing CD4+ T cells) and CD8+ T-cells markedly decrease in number [5,6]. While the type of macrophage ultimately driving the cytokine storm in severe COVID-19 remains to be unambiguously identified [52], a recently described landscape of lung bronchoalveolar immune cells in COVID-19 using single-cell RNA sequencing revealed that monocyte-derived ficolin 1-positive macrophages, which are highly inflammatory and potent cytokine producers, likewise overwhelm the severely damaged lungs in COVID-19 patients with ARDS [53].From a mechanistic and therapeutic perspective, the fact that IMMs promote a late and lethal SARS-CoV-2 infection irrespective of the viral load immediately implies that targeted antagonism of such dysregulated response would improve outcomes in patients with severe SARS-CoV-2 infection (Figure 2). Silibinin might be expected to phenotypically integrate the mechanism of action of IL-6-targeted monoclonal antibodies and pan-JAK1/2 inhibitors by directly modulating downstream STAT3 activity in the futile cycle of SARS-CoV-2-damaged lung tissues that orchestrate a reactive inflammatory monocyte/macrophage response and sensitize T-cells to apoptosis, resulting in a further dysregulated inflammatory response. Silibinin would thus operate as an immune therapeutic to alleviate the cytokine storm and T-cell lymphopenia in the clinical scenario of a subgroup of severe COVID-19 patients fully meeting ARDS criteria. The IFN-I-dependent immunopathological events, including the recruitment of immunopathogenic immune cells by inflammatory mediators and impairment of T cell responses, which are largely independent of virus replication, are key promoters of SARS-CoV-2 morbidity and mortality. In the delicate balance of virus-host interactions in COVID-19, however, it is also true that it is the extremely rapid and robust replication of SARS-CoV-2—while IFN-I expression is delayed—that initially stimulates lung inflammation, underscoring the relevance of reducing initial viral load through anti-viral interventions (Figure 2). In the latter regard, most of the current therapeutic efforts are directed to pharmacologically target the host-virus interface linking the viral S protein to the angiotensin-converting enzyme 2 receptor in host cells, main 3C-like protease- and papain-like protease-dependent viral replication (a proposed mechanism for lopinavir/ritonavir, ASC09/TMC-310911, or darunavir/cobicistat), and the RdRp/nsp12-driven viral RNA synthesis (a proposed mechanism for remdesivir, favipiravir, emtricitabine/tenofovir alafenamide, ribavirin, or more recently sofosbuvir) [54,55,56,57,58,59,60,61,62,63,64,65,66,67]. Intriguingly, a recent target-based virtual ligand screening study predicted that silibinin might target RdRp/nsp12 [68]. RdRp/nsp12 is a central component of a multi-subunit RNA-synthesis complex that additionally requires the assistance of the co-factors nsp7 and nsp8 to ensure the fidelity of faithfully replicating the largest known RNA genome among all the RNA viruses [69,70,71,72,73,74]. The architecture of RdRp/nsp12 appears to be common to all viral polymerases; however, the catalytic mechanism by which SARS-CoV polymerase performs de novo RNA initiation is likely to be distinct from other viruses. Thus, while the nsp12/nsp7/nsp8 complex possesses de novo initiation capacity is dependent on the nsp12 polymerase-active catalytic site, it is noteworthy that the SARS-CoV nsp12 double-stranded RNA exit tunnel has no obstruction to act as a platform for priming nucleotides [74]. Nsp7 and nsp8 heterodimers appear to stabilize nsp12 regions involved in RNA binding while a second nsp8 subunit can play a crucial role in polymerase activity via extension of the template RNA-binding surface. To computationally validate the recent prediction of silibinin targeting the coronaviral RdRp, we took advantage of the 3.1 Å resolution of the SARS-CoV nsp12 polymerase bound to its essential co-factors, nsp7 and nsp8 [74]. We employed the SARS-CoV 6NUR crystal structure as a template to generate a homology model of the SARS-CoV-2 RdRp (Figure 3, top). Docking simulations of silibinin over the SARS-CoV-2 nsp12/nsp7/nsp8 model complex produced eleven clusters of docking poses, three of which were predicted to occupy the RNA template tunnel of the coronaviral polymerase. In the absence of the nsp7 and nsp8 cofactors, we similarly predicted the occurrence of three clusters of docking poses occupying the RNA template tunnel (Figure 3, top). Based on these findings, we performed a more detailed comparative docking analysis of silibinin versus currently employed (or predicted) SARS-CoV-2 RdRp-targeted nucleotide analog anti-virals (i.e., remdesivir, ribavirin, and sofosbuvir) in a computational grid centered on the RNA template tunnel (Figure 3, bottom). When the docking results were ranked according to the ascent of the binding energies (up to-9.4 kcal/mol), the silibinin clusters exhibiting the highest affinity were in the low (~100 nmol/L) nanomolar range (Table S1), which was similar to that predicted for remdesivir (Table S2) but notably lower than the ranges predicted for ribavirin and sofosbuvir (Tables S3 and S4, respectively). To verify further these findings, we performed an additional comparative docking analysis of silibinin with remdesivir, ribavirin, and sofosbuvir using the recently reported cryo-electron microscopy analysis of the SARS-CoV-2 full-length nsp12 in complex with cofactors nsp7 and nsp8 at 2.9 Å resolution [75]. The overall architecture of the SARS-CoV-based nsp12-nsp7-nsp8 homology model was similar to that of SARS-CoV-2 with root mean square deviation values of 0.565 (for 6346 atoms; non-reduced form 6M71) and 0.534 (for 6876 atoms; reduced form 7BTF), respectively (Figure 4, top). Noteworthy, the predicted energy bindings of the best clusters of silibinin to SARS-CoV-2 nsp12 (Figure 4, bottom) remained in the submicromolar/low-micromolar range (Tables S5 and S6), which was similar or even slightly lower than those predicted for remdesivir and sofosbuvir (Tables S7–S10), and inferior to that in high micromolar range of ribavirin (Tables S11 and S12). Nonetheless, silibinin was predicted to interact with well-characterized catalytic residues (e.g., Asp618, Asp623, Asp760, Asp761) as well as with other key residues involved in the nsp12 interaction with the entry path of RNA template/nucleoside triphosphate (e.g., Arg555, Val557, Thr680, Ser682, Asn691) when using either the SARS-CoV-based homology model or the newly characterized SARS-CoV-2 nsp12 structure.To add protein flexibility and provide additional information about intra- and inter-molecular movements, we performed molecular dynamics (MD) simulations over the course of 100 ns to confirm the kinetic stability of the poses obtained by docking (Figure S1); the root mean square deviation (RMSD) of silibinin versus remdesivir heavy atoms measured after superimposing either nsp12 alone or the nsp12-nsp7-nsp8 hetero-complex on their references structures during the MD simulation were prepared in parallel. From the RMSD simulations, it was easily observed that a majority of the silibinin-containing systems rapidly equilibrated (less than 10 ns) and remained stable until the end of the simulation. In the case of remdesivir, equilibration was obtained later (at around 20 ns) and more fluctuations were visible over the course of simulation. Molecular mechanics Poisson-Boltzmann surface area (MM-PBSA) parameters, which estimate the free energy of the binding of silibinin/remdesivir to SARS-CoV-2 RdRp and are known to show good correlation with the experimentally obtained values, were finally calculated for the entire MD simulation trajectory of 100 ns (or the last 30 ns). MM-PBSA values > 20 kcal/mol and up to 36 kcal/mol were obtained in the case of silibinin, which were in a similar range of those obtained for remdesivir, thereby predicting the strong binding behavior of silibinin to SARS-CoV-2 RdRp. The RdRp coronaviral replication/transcription machinery is considered a primary target for new antiviral therapeutics. Therefore, the in silico predicted ability of silibinin to share, at least in part, the ability of nucleotide analogs to occupy the catalytic site of nsp12 [75,76] might support the consideration of silibinin as a new antiviral targeting SARS-CoV-2 RdRp, thereby adding a new therapeutic dimension to the previously recognized ability of silibinin to function as a broad-spectrum antiviral. The induction of the JAK/STAT signaling pathway by IFN is known to upregulate a significant number of so-called IFN stimulated genes (ISGs), some of them capable of rapidly kill viruses within infected cells [77]. Therefore, as viral-encoded factors known to antagonize JAK signaling are crucial determinants of virulence, it could be argued that pharmacological blockade of this pathway with available JAK inhibitors (JAKinhibs) or direct STAT3 inhibitors (e.g., silibinin) might produce an impairment of IFN-mediated antiviral response, with a potential facilitating effect on the early-stage evolution of SARS-CoV-2 infection [13,77,78]. Using a JAKinhib to treat a viral infection may play a double role because both type I IFN (IFN-α/β) and type II IFN (IFN-γ) employ the JAK-STAT signaling pathway. Indeed, such a mechanism might explain the increased risk of herpes zoster and simplex infection reported during the development of several JAK inhibitors including baricitinib, upadacitinib, and filgotinib [13,15]. As mentioned above, animal model studies on COVID19-related MERS (Middle East Respiratory Syndrome) and SARS diseases have shown that IFN-α/β therapy might be beneficial in the early inflammatory phase of both diseases, whereas the same therapy can be harmful in the late phase of the diseases [50,51] (Figure 2). As nearly 80% of patients with COVID-19 appear to eradicate the SARS-CoV-2 virus via antiviral immune responses (e.g., IFN-α/β), JAK/STAT3 inhibitors may be a better approach when hospital care is needed for COVID-19 patients. Indeed, the peak of SARS-CoV-2 load takes places ~7 days after symptoms onset in those patients with mild disease requiring hospital care. Later, as the viral titer might even decrease in some patients, hyper-inflammation drives the severe phase of the disease, which is accompanied by increased levels of cytokines—all signaling through the JAK signaling pathway. In the scenario of late severe COVID-19, in which the accumulation of IMMs in the lungs as the main source of pro-inflammatory cytokines is associated with fatal disease, targeted antagonism of the JAK pathway with JAKinhibs such as baricitinib (Olumiant®)—a dual JAK1/JAK2 inhibitor approved by the European Medicines Agency for moderate-to-severe active rheumatoid arthritis in adults [79,80,81,82]—would improve outcomes given their non-selective capacity to inhibit cytokine release and block cytokine signaling. Using the BenevolentAI’s knowledge graph, a large repository of structured medical information that includes numerous connections extracted from scientific literature by machine learning [83], baricitinib is predicted to interrupt both viral entry and intracellular assembly of viral particles by targeting cyclin G-associated kinase and AP2-associated protein kinase (AAK1), two well-known regulators of endocytosis. Accordingly, baricitinib is being evaluated in clinical trials for the treatment of moderate-to-severe COVID-19 in hospitalized patients either as a single agent (2 mg/day/orally for 10 days; NCT04321993) or combined with antiviral therapy (4 mg/day/orally combined to antiviral therapy ritonavir for 2 weeks, NCT04320277). As STAT3 is a latent cytoplasmic transcription factor that can couple with multiple cytokines as an ultimate effector of the JAK/STAT pathway, an important question that remains unexplored is whether STAT3 might be a better target than its upstream JAK activators to clinically manage COVID-19. We recently reasoned that the systemic manipulation of STAT3 signaling with silibinin and its putative, direct anti-viral activity can be trialed as a safe therapy to reduce inflammation and viral replication in a population of SARS-CoV-2-infected oncology patients. The Spanish Agency for Medicine and Health Products (Agencia Española de Medicamentos y Productos Sanitarios; AEMPS)—the regulatory agency that oversees the quality, safety, and efficacy of pharmaceutical and medical devices in Spain—has prioritized the approval of a randomized (1:1), open-label, phase II multicentric clinical trial to evaluate the efficacy of silibinin supplementation in the prevention of ARDS in moderate-to-severe COVID-19-positive onco-hematological patients undergoing systemic treatment or having completed treatment <1 year ago (Figure 5). The so-called SIL-COVID19 trial will be conducted in two phases: a non-randomized safety phase and a randomized phase. During the clinical study design, we acknowledged that the achievement of a bona fide, clinically relevant activity of silibinin remains controversial in human trials. Then, we decided to take advantage of the so-called Eurosil85 formulation, a patent extraction process that augments the oral absorption and bioavailability of silibinin, exhibiting the highest permeability rate in models of human intestinal absorption, that has been shown to exhibit significant clinical activity in cancer patients with advanced systemic disease [24,27,28]. Although milk thistle extracts such as Eurosil85 are known to be well-tolerated, with minimal toxic or adverse effects being observed in clinical trials, and co-administration of silibinin with multiple antiretrovirals (e.g., darunavir-ritonavir in HIV-infected patients) has been shown to be safe without a need for dose adjustment of anti-viral therapy [86], the SIL-COVID19 trial will evaluate oral bioavailability, single and multidose pharmacokinetics, and safety of the Eurosil85 formulation in a lead-in safety cohort in which to test whether silibinin supplementation combined with best supportive care according to physician’s choice exceeds a safety threshold, and will provide the phase II-recommended dose. The phase II will then assess the impact of silibinin on the severity and progression of COVID-19 when combined with best supportive care according to physician’s choice. The primary endpoint of the study will be efficacy in terms of the clinical status at a given day using a scale of death, need for mechanical ventilation or intensive care unit (ICU) admission, non-invasive ventilation or high-flow oxygen devices, low-flow supplemental oxygen, not requiring supplemental oxygen/no longer requiring ongoing medical care, and not hospitalized. Clinical responses in the experimental and control arms will be compared for statistical difference. Secondary outcome measures will include toxicity/tolerability profiles as well as circulating markers of systemic cytokine release syndrome, vascular permeability and leakage, thrombosis, pulmonary embolism, and coagulopathy with the aim of uncover early biomarkers of poor evolution or response of COVID-19 patients to silibinin-based treatment. The infection rates, severity, and lethality of SARS-CoV-2/COVID-19 are substantially higher in the population aged 60 and older. Despite the limited data available, statistics from the COVID-19 pandemic strongly support the notion that SARS-CoV-2 is a gerolavic (from Greek, géros “old man,” and epilavís, “harmful”) virus that disproportionately affects the elderly [87]. Animal models of SARS-CoV pathogenesis corroborating the age-related susceptibility to COVID-19 disease have demonstrated that the age-driven remodeling of the immune response (i.e., immunosenescence) could play a central role in the enhanced susceptibility of the elderly to the severity and lethality of SARS-CoV-2 infection [88,89]. In addition to immunosenescence, other cellular senescence phenomena in lung-epithelial and -mesenchymal cells might promote chronic airway remodeling and sustain chronic airway inflammation through the well-known capacity of senescent cells to produce large amounts of inflammatory cytokines (e.g., IL-6) as a result of the senescence-associated secretory phenotype (SASP) [90,91]. As the SASP could augment cellular senescence in surrounding cells, such a vicious feedback loop certainly generates a pathogenic condition to infections with viruses that can cause a very brisk cytokine reaction such as SARS-CoV-2. Importantly, however, if an exaggerated lung-associated cell senescence is a key co-morbidity impacting on the age-dependent prognosis of COVID-19 patients [92], it then follows that breaking this chronic cycle may help restrain the risk of COVID-19 aggressiveness and lethality in elderly individuals (Figure 5). It is therefore plausible that the immunomodulatory and senoremediative actions of silibinin could go beyond treatment and may provide a preventative measure before elderly individuals are exposed to SARS-CoV-2. Extensive JAK/STAT pathway studies in aging have demonstrated that this pathway plays a major role in regulating cytokine production as part of the SASP [93] and that STAT3 activation is relevant to the lung-cell senescence program [94,95,96]. Although there is currently no evidence for a senolytic activity of silibinin [97], it should be noted that other silibinin-related plant flavonoids such as quercetin have proven beneficial at reducing senescent cell burden and chronic sterile inflammation [98]; a systematic review of the efficacy of dietary flavonoids on upper respiratory tract infections and immune function in healthy adults have suggested their prophylactic value with no apparent adverse effects [99]. It might be argued that because STAT3 appears to direct seemingly contradictory pro- and antiviral responses during the acute phase response of viral infection [100,101], STAT3 inhibition might be exploited by SARS-CoV-2 to evade IFN-related host innate immunity [102]. Intriguingly, however, it has recently been discovered that ACE2, the entry receptor of SARS-CoV-2, is a previously unrecognized human ISG in airway epithelial cells, thereby suggesting that SARS-CoV-2 might exploit species-specific IFN/STAT-driven upregulation of ACE2—a tissue-protective mediator during lung injury—to enhance infection [78]. Moreover, SARS-CoV-2 infection via ACE2 can activate STAT3, which in turn can activate the so-called IL-6 amplifier [103], a mechanism for the hyperactivation of STAT3 that might trigger a positive feedback capable of self-amplifying the IL-6/STAT3 signaling in lung alveolar epithelial cells, thereby accelerating the transition from virus replication/spreading to COVID-19 clinical stages. In this scenario, it might therefore be possible to conduct clinical trials on the protective activity of silibinin for therapeutic development during the initial phases of SARS-CoV-2 infection in residential care nursing homes and elderly day-care centers where old, frail individuals exhibit an elevated pre-disposition for delayed IFN-I signaling (Figure 5), leading to dysregulated IMM response, lung immunopathology, and lethal pneumonia. Although generally overlooked or underestimated, there is strong clinical evidence on the safety and efficacy of intravenous and oral silibinin to induce potent antiviral responses capable of rescuing patients chronically infected with hepatitis C virus (HCV)—an enveloped RNA virus belonging to the Flaviviridae family—that fail to respond to combinations of IFN and anti-virals such as ribavirin [104,105,106,107,108,109,110,111]. The Food and Drug Administration (FDA)-approved drug sofosbuvir has recently been proposed as an antiviral for the SARS-CoV-2 based on the similarity of the replication mechanisms of HCV and coronaviruses [67,75]. Following this reasoning, the capacity of silibinin to reduce viral loads in non-IFN responders along with its recognized ability to exert direct anti-viral effects against flaviviruses (HCV and dengue virus), togaviruses (Chikungunya virus and Mayaro virus), influenza virus, human immunodeficiency virus, and HBV [112], might be considered to develop prophylactic interventions in individuals facing significant risk of developing COVID-19 (e.g., close contacts, households, and healthcare workers). In the latter regard, it should be noted that the anti-COVID-19/SARS-CoV-2 activity of silibinin might involve additional (STAT3- and viral RNA polymerase-independent) mechanisms. For example, clathrin-dependent cellular trafficking is a key mechanism for SARS-CoV-2 entry into host cells. Accordingly, the FDA-approved drug chlorpromazine, an inhibitor for clathrin-dependent endocytosis, has been found to exert inhibitory effects on the entry of coronaviruses including MERS-CoV and SARS-CoV [113]. Of note, silibinin has been shown to potently hinder the entry of HCV, reovirus, vesicular stomatitis, and influenza viruses by slowing-down trafficking through clathrin-coated pits and vesicles [114] (Figure 6). The capacity to inhibit early steps of viral infection by widely affecting endosomal trafficking of virions supports silibinin as a potential broad-spectrum antiviral therapy. In another example, inhibition of endoplasmic reticulum (ER)-resident α-glucosidases, which sequentially trim the terminal glucose moieties on the N-linked glycans attached to nascent glycoproteins [115], has been shown to impair the S protein-mediated entry of SARS-CoV by altering the glycan processing of ACE2 and, consequently, post-receptor binding mechanisms [116]. Thus, suppression of ER α-glucosidases not only can disrupt the morphogenesis of a broad spectrum of enveloped viruses including SARS-CoV, but also impede their entry by altering the glycan structures of their cellular receptors. Indeed, ER α-glucosidase inhibitors have been proposed as host function-targeting broad-spectrum antiviral agents that might be particularly attractive for treatment of respiratory tract viral infections (e.g., COVID-19) caused by enveloped RNA viruses (e.g., SARS-CoV-2), with a short window for medical intervention [117,118]. Although preliminary, when evaluating the mechanistic insights underlying the immune-sensitizing effects of silibinin, we recently observed its capacity to modulate the N-linked glycan decoration of nascent immunosuppressive molecules at the ER of tumor cells (unpublished observations). Moreover, such N-linked glycosylation-targeted activity of silibinin was compatible with in silico models predicting its capacity to directly target the catalytic site of human ER α-glucosidase (unpublished observations). The 2019 SARS-CoV-2 coronavirus outbreak is causing a global pandemic with hundreds of thousands of infections and deaths worldwide. Whereas numerous pharma and biotech companies and academic institutions are racing to develop vaccine candidates for effective COVID-19 prevention, the rapid global spread of COVID-19 has stressed the need for new therapeutics. Currently, three broad groups of anti-coronavirus treatments are being investigated for the prevention and treatment of the life-threatening SARS-CoV-2/COVID-19, namely: (1) those aimed at dampening the exaggerated host immune response; (2) those aimed at blocking viral replication and survival in host cells, and (3) those aimed at halting viral entry into host cells. Here, we present a comprehensive review of the evidence-based research into the multi-faceted capacity of silibinin to target the host cytokine storm and the virus lifecycle to clinically manage COVID-19/SARS-CoV-2 infection (Figure 6). We acknowledge that measurement of the therapeutic efficacy requires randomized trials of silibinin therapy. Accordingly, a clinico-translational research experience will be conducted at the Catalan Institute of Oncology in Catalonia (Spain) for testing the therapeutic efficacy of silibinin in oncology patients hospitalized with COVID-19. Nevertheless, our present work aims to provide a basis for the design of new silibinin-based antiviral therapeutics or supportive care approaches against the COVID-19, a global public health emergency. The following are available online at https://www.mdpi.com/2077-0383/9/6/1770/s1, Figure S1: Incorporation models of predicted RdRp-targeted silibinin and remdesivir clusters, Table S1: Details of the interaction of silibinin docked to SARS-CoV-2 nsp12 (6NUR-based homology model), Table S2: Details of the interaction of remdesivir docked to SARS-CoV-2 nsp12 (6NUR-based homology model), Table S3: Details of the interaction of ribavirin docked to SARS-CoV-2 nsp12 (6NUR-based homology model), Table S4: Details of the interaction of sofosbuvir docked to SARS-CoV-2 nsp12(6NUR-based homology model), Table S5: Details of the interaction of silibinin docked to SARS-CoV-2 nsp12 (6M71), Table S6: Details of the interaction of silibinin docked to SARS-CoV-2 nsp12 (7BTF), Table S7: Details of the interaction of remdesivir docked to SARS-CoV-2 nsp12 (6M71), Table S8: Details of the interaction of remdesivir docked to SARS-CoV-2 nsp12 (7BTF), Table S9: Details of the interaction of ribavirin docked to SARS-CoV-2 nsp12 (6M71), Table S10: Details of the interaction of ribavirin docked to SARS-CoV2- nsp12 (7BTF), Table S11: Details of the interaction of sofosbuvir docked to SARS-CoV-2 nsp12 (6M71), Table S12: Details of the interaction of sofosbuvir docked to SARS-CoV-2 nsp12 (7BTF). Conceptualization, J.B.-B., J.A.E., and J.A.M.; methodology, M.B. and B.M.-C.; formal analysis, J.A.E. and J.A.M.; investigation, J.B.-B., B.M.-C., M.B., J.A.E., J.A.M.; resources, J.B.-B., J.B., J.A.E., and J.A.M.; data curation, J.A.E., and J.A.M.; writing—original draft preparation, J.B.-B. and J.A.M.; writing—review and editing, J.A.E. and J.A.M.; visualization, J.A.E. and J.A.M.; supervision, J.B.-B. and J.A.M.; funding acquisition, J.B.-B., J.A.E., and J.A.M. All authors have read and agreed to the published version of the manuscript.Work in the Menendez laboratory is supported by the Spanish Ministry of Science and Innovation (Grant SAF2016-80639-P, Plan Nacional de I+D+I, founded by the European Regional Development Fund, Spain) and by an unrestricted research grant from the Fundació Oncolliga Girona (Lliga catalana d’ajuda al malalt de càncer, Girona). The Spanish Ministry of Economy and Competitiveness (MINECO, RTI2018-096724-B-C21) and the Generalitat Valenciana (PROMETEO/2016/006) supports work in the Encinar laboratory. Joaquim Bosch-Barrera is the recipient of Research Grants from La Marató de TV3 foundation (201906) and the Health Research and Innovation Strategic Plan (SLT006/17/114; PERIS 2016-2020; Pla stratègic de recerca i innovació en salut; Departament de Salut, Generalitat de Catalunya).We are grateful to the High Performance Computing Center Stuttgart (HLRS) for letting us to take advantage of the HAWK computer cluster (https://hlrs.de/systems/hpe-apollo-hawk/), which has provided a significant part of the computational calculation time necessary to prepare this manuscript. We wish to specially thank Thomas Beisel for his kind assistance and cooperation. We are also grateful to the Cluster of Scientific Computing (http://ccc.umh.es/) of the Miguel Hernández University (UMH) for providing computing facilities. This manuscript was written over the course of the second and third week of the lockdown in Spain due to the SARS-CoV-2 coronavirus pandemic in March-April 2020. This paper is dedicated to the memory of all those who have succumbed to COVID-19 during those hard days. The authors would like to thank Kenneth McCreath for editorial support. The authors declare that the research was conducted in the absence of any commercial or financial relationships that could be construed as a potential conflict of interest.Silibinin as a direct STAT3 inhibitor: regulating the response intensity of reactive reparative cells to damaged tissues. STAT3 is a master checkpoint regulator of the interface between cytokines, inflammation, and immune response against various types of tissue damage including viral infections. By operating as a direct inhibitor of STAT3, silibinin regulates the response intensity of reparative cells (e.g., macrophages, astrocytes) to damaged tissues (e.g., radiation-induced lung injury, primary lung tumor, brain metastasis). In the ongoing scenario of the 2019—2020 SARS-CoV-2 pandemic, in which the futile accumulation of inflammatory macrophages in lungs as the main source of pro-inflammatory cytokines associates with fatal disease in a subgroup of severe COVID-19 patients fully meeting acute respiratory distress syndrome (ARDS) criteria, targeted antagonism of STAT3 with silibinin might ameliorate COVID-19 morbidity and mortality.Silibinin: a putative regulator of the too-little-too-late interferon response in COVID-19. Whereas it is well known that type I-interferon (IFN)-induced anti-viral response is among the earliest and most potent of the innate responses to fight viral infection, the timing of IFN-I response relative to virus replication might be key for SARS-CoV-2 infection outcome. (1) The extremely rapid and robust replication of SARS-CoV2—while IFN-I expression is delayed—is one of the initial triggers of lung inflammation, therefore highlighting the relevance of reducing initial viral load through direct anti-viral interventions. By targeting the central component of the replication/transcription machinery of SARS-CoV-2, namely the viral polymerase RdRp, silibinin is expected to reduce viral load and/or impede delayed interferon responses. The early antagonism of several SARS-CoV-2 proteins to the IFN response delays or prevents the innate immune response. Delayed IFN signaling, however, further orchestrates inflammatory monocyte/macrophage (IMM) responses and sensitize T-cells to apoptosis, which results in a further dysregulated inflammatory response, cytokine-driven acute respiratory distress syndrome (ARDS), and severe/fatal disease in a subgroup of COVID-19 patients. (2) Silibinin might be expected to phenotypically integrate the mechanisms of action of IL-6-targeted monoclonal antibodies and pan-JAK1/2 inhibitors to alleviate the pathological accumulation of inflammatory macrophages as the source of the cytokine storm and T-cell lymphopenia associated with fatal COVID-19 disease. We acknowledge that the proposed viral RdRp-targeted (1) and host STAT3-targeted (2) mechanisms of silibinin are highly intertwined. On the one hand, the (pro-/anti-) responses of STAT3 signaling to virus infection are complex both in a virus-specific and in a stage-specific manner in the virus lifecycle. On the other hand, the host IFN response could promote the ability of SARS-CoV-2 to maintain cellular targets in neighboring human upper airway epithelial cells via up-regulation of the SARS-CoV-2 receptor ACE2 in a STAT1-related manner. As the competition phenomena and distinct dynamics of the STAT3/STAT1 duo in pro- (e.g., IFN-I) and anti-inflammatory pathways might constitute an evolutionary-conserved mechanism of the host to tightly control the immune response while avoiding tissue damage, further studies should clarify whether STAT3 participates in the net protective or detrimental role of type I IFN depending on the stage of SARS-CoV-2 infection. In animal models, IFN-I administration shortly after SARS-CoV infection has positive effects and protects mice from lethal infection, whereas later administration IFN-I fails to inhibit viral replication and has side-effects (i.e., severe COVID-19-like increased infiltration and activation of monocytes/macrophages/neutrophils in the lungs accompanied by enhanced pro-inflammatory cytokine expression) that result in fatal pneumonia from an otherwise sublethal infection. Accordingly, although promising findings have been observed upon IFN-I treatment or IFN-based combination therapy with lopinavir/ritonavir, ribavirin, or remdesivir in pre-clinical animal models, mixed results have been found in humans. Understanding if STAT3 is one of the host restriction/promoter factors targeting SARS-CoV-2 lifecycle in a IFN-I/ACE2-dependent/independent manner may provide better strategies to dissociate the dual roles of IFN-I in SARS-CoV-2 infection.Silibinin is predicted to target SARS-CoV-2 viral polymerase RdRp (I). Representations of a SARS-CoV-1 6NUR-based homology model of the SARS-CoV-2 viral polymerase RdRp showing the computationally predicted locations of silibinin, remdesivir, ribavirin, and sofosbuvir clusters in the nsp12 catalytic subunit of the nsp12-nsp7-nsp8 complex. The protein has been represented as a function of the hydrophobicity of its surface amino acids and the Na+ and Cl- ions have been eliminated to facilitate visualization. Figures were prepared using PyMol 2.3 software.Silibinin is predicted to target SARS-CoV-2 viral polymerase RdRp (II). Representations of the SARS-CoV-2 viral polymerase RdRp in non-reduced (6M71) and reduced (7BTF) conditions showing the computationally predicted locations of silibinin, remdesivir, ribavirin, and sofosbuvir clusters in the nsp12 catalytic subunit of the nsp12-nsp7-nsp8 complex. The protein has been represented as a function of the hydrophobicity of its surface amino acids and the Na+ and Cl- ions have been eliminated to facilitate visualization. Figures were prepared using PyMol 2.3 software.Silibinin in the prevention and treatment of COVID-19. COVID-19 prevention. Severe COVID-19 illness and death is largely more common in the population aged 60 and older. The gerophilic and gerolavic traits of SARS-CoV-2 infection might rely on immunosenescence and other lung-associated cellular senescence phenomena promoting chronic airway remodeling and sustain chronic airway inflammation. The immunomodulatory and senoremediative activities of silibinin might ameliorate the co-morbid features of aging lungs (e.g., chronic inflammation, airway remodeling, function decline) thereby reducing the risk of COVID-19 aggressiveness and lethality in elderly individuals. In frail, elderly individuals with an elevated pre-disposition to experience delayed IFN-I signaling leading to dysregulated inflammatory monocyte/macrophage responses, lung immunopathology, and lethal pneumonia, the direct antiviral actions of silibinin might also be explored as a prophylactic intervention. COVID-19 treatment. An ongoing clinico-translational research study is being conducted at the Catalan Institute of Oncology in Catalonia (Spain) for testing the therapeutic efficacy of silibinin in oncology patients hospitalized with moderate/severe COVID-19. A Jung’s two-stage design for randomized phase II trials with a prospective control was used to estimate the sample size [84]. To keep the sample size small and the study period short, we employed a relatively large type I error (α = 15%) and short-term outcome variable (i.e., the number of responses as primary endpoint), which will allow for early termination of the study if the silibinin-containing arm fails to show efficacy at the interim analysis. The combination of silibinin plus best supportive care will be considered worthwhile if there is a response rate of 90% in the experimental arm (versus the expected 70% in the control arm [85]). By setting an α level of 0.15, a power of 0.80, a balanced allocation (1:1), and an expected drop-out rate of 10%, the sample size will be 16 patients for both arms to ensure the assessment of response rates in 14 patients in each arm at the first stage. Only if at least two more patients achieve the desirable clinical response in the silibinin-containing arm than in the reference arm, and provided no safety issues are identified, would the clinical trial proceed to the second stage. In such a case, an additional recruitment of 25 patients for both arms (to ensure the assessment of response rates in 22 patients in each arm), will proceed. The silibinin-containing arm will be considered effective if four or more additional patients achieve a clinical response in comparison with the reference arm at the end of the study (n = 82 patients at the planned final sample size).Silibinin: A multi-faceted approach to SARS-CoV-2/COVID-19. The SARS-CoV-2/COVID-19 pandemic is a current threat to human health. As the application of antiviral drugs is expected to provide an immediate and direct control of SARS-CoV-2 infection, a major strategy for managing COVID-19 patients might involve targeting conserved viral proteins critical for viral replication such as the viral polymerase RdRp. A second approach might involve the targeting of key virus-host interactions such as the ACE2-dependent binding and clathrin-dependent endocytosis processes enabling the entry of SARS-CoV-2 into the host cell. Finally, the targeting of host cellular mechanisms that are triggered to defend against SARS-CoV-2 infection (e.g., STAT3-driven reactive immune-inflammation) might impede not only the viral exploitation of cellular responses (e.g., host type I interferon) in support of its efficient replication but, perhaps more importantly, the overproduction of pro-inflammatory cytokines and the overactivation of immune cells that ultimately drives an acute respiratory distress syndrome (ARDS) as one of the leading causes of mortality in a subgroup of patients with severe COVID-19. The multi-faceted ability of silibinin to target all the therapeutically relevant clinico-molecular traits of SARS-CoV-2 infection provides a strong rationale for the clinical testing of silibinin against the COVID-19 global public health emergency.

Med-MDPI/jcm_2/jcm-09-06-01771.txt

@@ -0,0 +1 @@
+Background: Tacrolimus (TAC) is used for the management of ulcerative colitis (UC). However, there are few reports on the effectiveness of its long-term administration. TAC is also known to cause renal toxicity. The aim of this study was to evaluate long-term effectiveness and monitor changes in renal function during prolonged TAC use in patients with UC. Methods: Medical records of 50 UC patients treated with TAC were retrospectively reviewed. Clinical outcomes were assessed at 6, 12, 24, and 36 months after initiating TAC. We also monitored chronological changes in renal function. Results: Thirty-nine patients were treated with TAC for more than 3 months. Relapse-free survival among these patients at 6, 12, 24, and 36 months was 82%, 69%, 41%, and 23%, respectively. On the other hand, renal function was reduced in 35.9% of patients. We found that irreversible renal dysfunction was more likely to occur in cases in which the estimated glomerular-filtration rate (eGFR) was reduced by more than 30%. Conclusion: This study demonstrated the potential use of TAC as an effective option in the long-term medical management of UC, although it tended to increase the risk of nephrotoxicity. There is a need for the careful monitoring of renal function during TAC administration.Ulcerative colitis (UC) is a chronic relapsing and remitting inflammatory disorder of the large intestine. In the treatment of UC, corticosteroids are still important basic therapies [1]. However, it was reported that 20% of patients with UC become steroid-dependent after initiating corticosteroid treatment [2,3]. Various treatments have become available in recent years, and calcineurin inhibitors are important in the treatment of steroid-resistant or -dependent cases as well as antitumor necrosis factor-α (TNF-α) antibody preparations [4,5,6]. Tacrolimus (TAC) is a calcineurin inhibitor isolated from soil bacterium Streptomyces tsukubaensis [7]. TAC has immunosuppressive effects by suppressing cellular functions such as cytokine production through T-cell activation, and it is well known that T-cell dysfunction plays a crucial role in the pathogenesis of inflammatory bowel disease (IBD) [8,9]. TAC possesses potent immunosuppressive properties, and it has been commonly used as an immunosuppressant to prevent kidney- or liver-transplant rejection [10]. Effectiveness against UC has been suggested since the late 1990s [11]. In Japan, TAC was approved for clinical use in UC patients in 2009. TAC has been demonstrated to have remarkable short-term therapeutic efficacy [8,12,13,14]. The administration period is up to 3 months for UC, but reductions in dosages have resulted in worsening in many cases, and in some cases, long-term administration is unavoidable. Several retrospective studies showed the long-term effectiveness of TAC therapy [8,15,16,17,18,19,20]. However, evidence of the efficacy and safety of long-term administration is still inadequate. Despite well-documented safety records, patients have been known to develop side effects, namely, infection, hypertension, nephrotoxicity and various neuropsychiatric problems [21]. Nephrotoxicity is one of the important side effects of TAC administration, along with infection [22]. Therefore, regular safety monitoring, including the measurement of plasma trough TAC levels, is required. However, few reports have focused on renal damage caused by TAC in UC patients. The aim of this study was to evaluate the long-term effectiveness of TAC, and monitor changes in renal function during its prolonged use in patients with UC.Data were compiled from 50 moderate-to-severe active UC patients treated with TAC in Juntendo University between April 2010 and December 2018. Their medical records were reviewed retrospectively for diagnosis, clinical course, treatment, and renal function. Patient data were registered into an electronic database after a de-identification process. The protocol for this retrospective investigation was reviewed and approved by the Juntendo University Hospital Ethics Committee (IRB no. 20-014). This study adhered to the principles of the Declaration of Helsinki.UC diagnosis was based on established standardized criteria by prior clinical assessment, radiology, endoscopy, and histology. Patients with prior steroid treatment were classified as having steroid-resistant or -dependent UC. Steroid-resistant UC is defined as an active disease despite a prednisolone dose of up to 0.75 mg/kg/d over a period of 4 weeks. Steroid-dependent UC is defined as a lack of ability to reduce steroids below the equivalent of 10 mg/d prednisolone within 3 months of starting steroids, without recurrence of disease activity or relapse within 3 months of stopping steroids [23]. TAC was administered in its oral formulation. In all cases, TAC was initiated at a dosage of 0.05–0.1 mg/kg/day, aiming for a target trough level of 10–15 ng/mL for the first 2 weeks to induce remission. After inducing clinical remission, TAC whole-blood levels were maintained at a range of 5–10 ng/mL [5]. Clinical outcomes were assessed at 3, 6, 12, 24, and 36 months after initiating TAC. Clinical disease activity was determined using the Lichtiger clinical-activity index (CAI) [24]. The Lichtiger CAI is composed of the following items: number of daily bowel movements, abdominal pain and tenderness, use of antidiarrheics, blood in stools, general well-being, fecal incontinence, and nocturnal diarrhea. A higher score indicates a more severe disease (score range 0–21). CAI ≥ 10, 7–9, and ≤6 were defined as severe, moderate, and mild, respectively [24,25,26]. Endoscopic severity was determined using the Mayo endoscopic-subscore (Mayo ES) classification (0, normal or inactive disease; 1, mild disease with erythema, decreased vascular pattern, mild friability; 2, moderate disease with marked erythema, absence of vascular patterns, friability, erosions; 3, severe disease with spontaneous bleeding, ulceration) [27] and the ulcerative-colitis endoscopic index of severity (UCEIS) [28]. The UCEIS is scored using the three following items: vascular pattern, bleeding, and erosions and ulcers. A score of 0 to 3 is given for each item, with a higher score indicating more severe endoscopic activity (score range 0 to 8). Mucosal healing was defined as a Mayo ES of 0 or 1 [29]. Clinical response was defined as a decrease in the CAI of 3 or more points, and clinical remission was defined as a score of 3 or fewer points. Maintenance of efficacy was defined as no exacerbation of CAI and no need for the increased intensification of treatments [30]. We monitored chronological changes in renal function by following the estimated glomerular-filtration rate (eGFR) and the serum creatinine level during TAC administration, and compared changes in eGFR and serum creatinine after TAC discontinuation. Renal failure was defined as a 30% decrease in the eGFR relative to baseline. Associations of age, sex, plasma trough TAC level, and dose with renal function were examined.All data were analyzed using GraphPad Prism (version 6, GraphPad, La Jolla, CA, USA). Differences between groups were analyzed using Mann–Whitney’s U test. Relapse-free survival was assessed using the Kaplan–Meier method. Significance was defined at p < 0.05. Characteristics of the 50 patients included in this study are shown in Table 1. There were 29 males and 21 females, with a median age of 37.5 years (range 18–68 years). The median duration of the disease was 6 years (range 1–33 years). Nineteen and 31 patients were classified as having left-sided or extensive colitis, respectively. The median Lichtiger CAI at baseline was 12 (range 8–16). Forty-five patients were treated with aminosalicylate prior to TAC treatment, and 4 patients switched medication from infliximab to TAC. Fourteen patients (28%) were steroid-refractory, 34 (68%) were steroid-dependent, and 2 (4%) were steroid-naïve. The median duration of TAC therapy was 13.5 months (range 1–64 months) (Table 1). No patient had a history of renal disease. Five patients had family histories of inflammatory bowel disease, and one patient of renal disease.Among patients who achieved a clinical response (CR) at 3 months after initiating TAC, 39 (78%) maintained the CR while continuing TAC. These included patients who did not have other treatment options or whose condition worsened upon reducing the dose of TAC. Drug withdrawal was observed in 11 patients at 3 months; 8 nonresponders, 1 patient by self-determination and 2 patients due to adverse events such as acute nephrotoxicity and nausea (Figure 1). Relapse-free survival among these patients at 6, 12, 24, and 36 months was 82%, 69%, 41%, and 23%, respectively (Figure 2). Three patients needed total colectomy during the observation period (Figure 1). Median Lichtiger CAIs at 3, 6, 12, 24, and 36 months were 2 (range 0–11), 3 (range 0–6), 3 (range 0–8), 3 (range 0–11), and 3 (range 0–8), respectively. We next assessed endoscopic improvement in cases that underwent colonoscopy. Median UCEIS was 5 (range 3–8), and median Mayo ES was 3 (range 2–3) at baseline. Mucosal healing (Mayo ES 0 or 1) was achieved by 24 of 30 patients (80%) 6 months after the introduction of TAC. Among patients achieving a CR at 12 months, mucosal healing was maintained in 19 of 27 patients, and in 12 of 16 patients at 24 months (Figure 3).The average eGFR and serum creatinine at baseline were 108.4 mL/min/1.73 m2 and 0.61 mg/dL, respectively. Renal function was reduced in 35.9% (14/39) of patients showing a 30% decrease in eGFR relative to the baseline during the observation period. The average oral dose after 3 months of administration was 5 mg/day. The mean plasma trough TAC level was controlled at 5–10 ng/mL and remained flat over time (Figure 4). Although nephropathy is usually dependent on the dosage of TAC, eGFR gradually decreased in cases treated with long-term TAC administration, even though plasma trough TAC levels remained flat (Figure 4 and Figure 5a). Of patients with a 30% decrease in eGFR, three patients had a drop in eGFR of more than 60% (Figure 5b). Changes in renal function were analyzed in 23 cases treated with TAC for more than 6 months and whose course could be followed for more than 1 year after the discontinuation of TAC treatment. Renal function tended to improve after discontinuation of TAC in 19 cases. However, eGFR remained reduced more than 30% compared to values before the administration of TAC even after TAC had been discontinued for 1 year in 4 cases (Figure 6a). Moreover, although renal function was within normal range, eGFR remained reduced after the discontinuation of TAC in cases with eGFR decreases greater than 30% (p < 0.05) (Figure 6b).The goals of UC management are to maintain clinical remission and mucosal healing while avoiding hospitalization and surgery. TAC is used for the management of refractory UC and has excellent short-term efficacy [8,12,13,14]. Although maintenance therapy with thiopurine is usually recommended after induction of remission with TAC, several studies evaluated the long-term efficacy of TAC for maintaining remission [8,15,16,17,18,19].Yamamoto et al. reported the results of retrospective analysis of UC patients who maintained remission with either TAC or thiopurine. There was no significant difference in the therapeutic effect between the two agents. However, the relapse-free survival rate was significantly lower in the thiopurine-refractory TAC group than in the thiopurine group [18]. This report is interesting because it shows that TAC may be an option for the maintenance of remission in patients who are intolerant of immunomodulators. Additionally, some other retrospective studies reported the usefulness of TAC as a maintenance therapy [8,17,20,31,32]. Endo et al. reported the long-term therapeutic efficacy of TAC, and compared therapeutic efficacies between patients treated with infliximab (IFX) and TAC. Though IFX and TAC showed similar short-term efficacy, the IFX-based strategy had better outcomes with regard to long-term efficacy for the treatment of steroid-refractory UC [31]. According to studies reporting long-term prognosis with TAC treatment, eventfree survival rates were 57–82%, 37–56%, and 38–50% at 6, 12, and 24 months, respectively [17,20,31,32]. In our study, relapse-free survival among patients treated with TAC at 6, 12, and 24 months was 82%, 69%, and 41%, respectively. Although patients’ backgrounds differed between our studies and those of the others, our treatment outcomes were consistent with those in previous reports. Furthermore, in this study, patients who maintained clinical remission had a high rate of mucosal healing. These data suggest that TAC may have potential as a maintenance therapy.Although TAC is useful for intractable UC, its prolonged use appeared to increase the risk of nephrotoxicity. However, there are few reports of TAC-induced renal damage in UC. A cohort study of patients who had undergone transplantation of organs other than kidneys revealed that 16.5% of chronic renal dysfunction occurred during a 36-month observation period. Surprisingly, 28.9% of these patients were reported to have progressed to end-stage renal failure [33]. According to the survey of specific uses of TAC for UC, the incidence of renal dysfunction was seen in 8.4% of the patients. Among these cases, 16.1% were reported to be severe. TAC can cause acute or chronic nephrotoxicity, particularly when used in high doses or over a long period. Acute nephrotoxicity due to calcineurin inhibitors is mainly observed early in administration and caused by the vasoconstitution of imported arterioles. Vasoconstitution is thought to be caused by an imbalance of factors such as endothelin, thromboxane, activation of the renin–angiotensin system, or vasodilators such as prostaglandin E2, prostacyclin, and nitric oxide. These changes depend on the dose of the calcineurin inhibitor and are reversible [34]. Therefore, reducing dose and lowering trough concentration are usually needed when nephrotoxicity is observed during TAC administration. In our study, mean plasma trough TAC level was controlled at 5–10 ng/mL, and the TAC dose or trough level remained flat over time (Figure 4). However, eGFR gradually decreased during long-term TAC administration (Figure 5a). These data showed that TAC troughs and doses do not always reflect renal function during prolonged TAC use. Chronic nephropathy is characterized by renal tubular interstitial fibrosis and can lead to irreversible nephrotoxicity, although the mechanism is not as clear as that for acute nephropathy [35]. eGFR is known to be more sensitive than serum creatinine levels to observed changes in renal function during prolonged TAC dosing. However, caution is needed, as eGFR could be impaired, but could still remain within a normal range in some cases. In our study, renal function was reduced in 35.9% (14/39) of patients with more than 30% decrease in eGFR relative to the baseline. However, eGFR was still observed to be within the normal range (Figure 5a). Remarkably, eGFR remained reduced even after the discontinuation of TAC in cases when eGFR was decreased by more than 30%, while eGFR improved in cases where the decrease in eGFR remained within 30%. Our study demonstrated that the discontinuation of TAC treatment should be considered before eGFR is decreased by more than 30% during TAC administration. While TAC has the potential for maintenance therapy, we must not forget the risk of nephrotoxicity. Since TAC troughs and doses do not always reflect renal function, eGFR is an important predictor of kidney reserve, and it is necessary to regularly evaluate eGFR. Detection of nephrotoxicity may enable us to appropriately adjust treatment. Limitations of our study included the sample size, retrospective design, single-center experience, and the heterogeneity of previous and current treatments among patients. Urine tests were also not examined in this study. Further studies are needed to elucidate the cause of irreversible renal damage due to the long-term administration of TAC.In conclusion, long-term administration of TAC appeared to prevent the relapse of UC. There are still no clinical trials that prospectively examined the effectiveness of the long-term administration of TAC, and it is necessary to accumulate evidence on the usefulness and safety of TAC dosing for the maintenance of remission. However, this study demonstrated the potential use of TAC as an effective option in the long-term medical management of patients with UC. On the other hand, prolonged use of TAC tended to increase the risk of nephrotoxicity. In most cases, renal function improved upon discontinuation or reduction of the dose, but in some patients, there was a significant decrease. There is a need for careful monitoring of renal function during TAC dosing. Conceptualization and study design, K.H. and T.S.; methodology, T.S., D.I., and T.O.; Data curation, K.H, T.S., K.O., O.N., K.N., D.I., N.S., and T.O.; formal analysis, K.H., K.N., D.I., and K.O.; visualization, K.H. and O.N.; project administration, T.S.; writing—original draft, K.H.; review and editing, K.H., T.S., and A.N.; supervision, A.N. All authors performed a critical check of the final draft for important intellectual content, and approved the final version prior to submission and publication. All authors have read and agreed to the published version of the manuscript. This research received no external funding.The authors have no potential conflicts (financial, professional, or personal) that are relevant to the manuscript. Flowchart of treatment outcomes. Clinical outcomes were assessed at 3, 6, 12, 24, and 36 months after initiating tacrolimus (TAC).Kaplan–Meier plots for relapse-free survival of patients treated with tacrolimus for more than 3 months (n = 39).Endoscopic evaluation of patients treated with tacrolimus. Endoscopic severity was assessed by endoscopic-activity index before tacrolimus therapy, and 6, 12, and 24 months after administration.Transition of tacrolimus (TAC) trough level and dose. (a) Target through level; (b) TAC does. Whole-blood levels of TAC were maintained at a range of 5–10 ng/mL after trough level of 10–15 ng/mL for first 2 weeks to induce remission.Changes in renal function. (a) Chronological changes in estimated glomerular filtration rate (eGFR) after administration of tacrolimus. (b) Decrease in eGFR during observation period (n = 39).Changes in estimated glomerular filtration rate (eGFR) after tacrolimus (TAC) discontinuation. (a) Changes in eGFR in overall patient population. Changes in renal function were analyzed in 23 patients treated with TAC for more than 6 months and whose course could be followed for more than 1 year after discontinuation of TAC treatment. * p < 0.05 (analyzed using Mann–Whitney test). (b) Changes in eGFR in patients with eGFR reductions less than 30% (n = 12) and in patients with eGFR reductions more than 30% (n = 11) after administration. * p < 0.05 (analyzed using Mann–Whitney’s test).Patient characteristics.1 eGFR, estimated glomerular filtration rate; 2 TAC, tacrolimus; 3 TNF, tumor necrosis factor.

Med-MDPI/jcm_2/jcm-09-06-01772.txt

@@ -0,0 +1 @@
+The nucleoside trifluridine/tipiracil (TAS-102) and the multikinase inhibitor regorafenib significantly improved survival in metastatic colorectal cancer patients (mCRC). Both treatments are characterized by different treatment-related adverse events but detailed analyses of predictive side effects are rare. In this retrospective, observational, real-life study, clinical data on mCRC patients treated with trifluridine/tipiracil or regorafenib at the Medical University of Vienna, Austria and the University Hospital Zurich, Switzerland were collected. The correlation between adverse events and response or survival rates were calculated performing Fisher’s exact test and log-rank test, respectively. Common adverse events of any grade included fatigue (52%), nausea/vertigo (34%), anemia (26%), and leukopenia (22%) in trifluridine/tipiracil patients and fatigue (42%), hand-foot-skin syndrome (36%) and hoarseness (34%) in patients upon regorafenib treatment. In trifluridine/tipiracil patients the prevalence of leukopenia (p = 0.044) and weight loss (p = 0.044) was prognostic, whereas leukopenia (p = 0.044) and neutropenia (p = 0.043) predicted PFS. The disease control rate was not significantly affected. In regorafenib-treated patients, the prevalence of nausea (p = 0.001) was prognostic, while oral mucositis predicted PFS (p = 0.032) as well as the DCR (p = 0.039). In conclusion, we underline the efficacy of trifluridine/tipiracil and regorafenib in the real-life setting. We describe predictive adverse events like neutropenia/leukopenia, which might be used as surrogate marker in anticancer therapy beyond second line treatment. Colorectal cancer (CRC) is a challenging global health problem. An estimated 1.1 million new cases of CRC were diagnosed worldwide in 2018, making it the third most common type of malignancy [1]. The 5-year survival rate of CRC patients highly depends on tumor stage at the time of diagnosis, as it shows a decrease from 90% in non-metastatic patients to merely 13% in patients with metastatic colorectal cancer (mCRC) [2,3].In recent years, new treatment options for patients with metastatic CRC have prolonged overall survival (OS) in these patients to approximately 32 months [4,5,6]. The current standard of care therapy for mCRC comprises conventional chemotherapeutical compounds as well as targeted therapy with antibodies against anti-epidermal growth factor receptor (EGFR) and vascular endothelium growth factor (VEGF) [7,8]. The implementation of these agents in standardized first-and second line treatment regimens resulted in a significant improvement in prognosis and outcome in mCRC patients. Recent advances in treatment could further improve the prognosis of mCRC patients’ options after the failure of second line treatment: Based on the results of the CORRECT trial, the multikinase inhibitor regorafenib has been approved for the third line treatment in patients with refractory mCRC [9]. In this particular study, regorafenib, compared to best-supportive care alone, led to a significant increase in median OS (6.4 months vs. 5.0 months, hazard ratio [HR] 0.77, p = 0.005) and a significant prolongation of median progression-free survival (PFS) (1.9 vs. 1.7 months, HR 0.49, p < 0.001). These findings were corroborated by the positive results demonstrated in the CONCUR study in an Asian population [10]. Moreover, recent studies have investigated the effects of regorafenib in other indications, leading to approval of regorafenib for the treatment of GIST and hepatocellular carcinoma [11,12,13,14].Trifluridine/tipiracil is a recently developed oral nucleoside compound [15,16]. The efficacy of trifluridine/tipiracil in mCRC patients was investigated in the international RECOURSE trial, a phase III study comparing trifluridine/tipiracil against placebo in refractory mCRC patients [16]. The study showed positive results: patients treated with trifluridine/tipiracil showed a significant improvement in median OS (7.1 vs. 5.3 months, HR 0.58, p < 0.001) and median PFS (2.0 vs. 1.7 months, HR 0.48, p < 0.001) compared to placebo.Considering these positive phase III trials, regorafenib and trifluridine/tipiracil offer new options in the third-line treatment of refractory mCRC. However, the most effective and safest treatment sequence in this setting remains unclear. Each agent presented a distinct toxicity profile in clinical studies. Previous studies have shown that the most common grade 3 side effects under regorafenib therapy include fatigue, hand-foot skin reactions, rash and elevation of liver enzymes [17,18]. Studies investigating the toxicity of trifluridine/tipiracil have found that hematological side effects of grade 3 or higher are common in trifluridine/tipiracil patients, followed by less common grade 3 side effects such as nausea and loss of appetite [17,19]. In summary, both drugs have shown similar effects on OS and PFS in mCRC patients, while their toxicological profile is highly different. A clinical head-to-head trial comparing regorafenib and trifluridine/tipiracil in mCRC patients is not available and analyses of the two compounds’ efficacy and side effects are scarce [20]. Treatment adherence and improved quality of life with reduced side effects was already described when regorafenib was gradually escalated in cycle 1-starting with 80 mg/day-compared to the standard dose of 160 mg/day (ReDOS) [21]. Additionally, flexible dosing showed numeric improvement on several parameters that increased tolerance, such as fatigue, hypertension, or hand-foot syndrome as shown in the REARRANGE trial [22]. Therefore, this retrospective real-life observational study aimed to investigate the efficacy and side-effects of treatment with regorafenib or trifluridine/tipiracil in mCRC patients. Moreover, we tried to elucidate the question of whether any of the reported side effects bear predictive quality for survival or disease control.The retrospective, observational, real-life study was approved by the institutional ethics committee of the Medical University of Vienna and Zurich (EC Nr.: 2189/2017) and carried out in accordance with the requirements of the International Conference on Harmonization E6 for Good Clinical Practice as laid down in the Helsinki Declaration.The patients were selected from respective institutional registries, either at the Medical University of Vienna, Austria or the University Hospital Zürich, Switzerland from January 2013 to December 2017. All patients fulfilled following criteria: histologically proven adenocarcinoma of the colon or rectum with metastasis (stage IV) and an Eastern Cooperative Oncology Group (ECOG) performance status ranging from 0 to 3; pretreatment with fluoropyramidines, oxaliplatin, irinotecan, bevacizumab and, in case of RAS w.t., cetuximab or panitumumab was required, according to the label. Patients treated with trifluridine/tipiracil or regorafenib were included for statistical analysis if at least one follow up scan was performed. A total of 143 patients informed consented to one of the two treatment options, whereby 31 patients did not start treatment for different reasons (alternative treatments, deterioration of performance status, lost in follow up, disease-related events). From 112 patients who started the respective treatment, 85 patients had at least one follow-up CT scan and were considered for this retrospective analysis. Scans were performed according to respective institutional recommendations. Patients’ characteristics are displayed in Table 1. Therapy was provided upon informed consent. Further therapy algorithms, prior treatment regimens, resection and radiation, as well as tumor characteristics, were registered for patient characterization.Trifluridine/tipiracil or regorafenib were prescribed for patients with mCRC as salvage therapy. The treatment was discontinued if the disease progressed, severe adverse events occurred or at the patient’s request. Median duration for treatment with regorafenib was 3.4 cycles, while median duration of trifluridine/tipiracil therapy was 3.2 cycles. The occurrence of adverse events was surveyed during the first cycle of application. Adverse events were described referring to CTCAE. Due to the complex analysis, the different degrees of adverse events were not considered in the evaluation.The distribution of categorical variables (e.g., localization or adverse event “yes” or “no”) was described by counts and percentages. In order to evaluate the association between different adverse events during the first cycle and the effectiveness of treatment, we performed univariate and multivariate analysis. The Fisher exact test was applied for tumor response and the log-rank test for PFS and OS. Hazard ratios or odds ratios to quantify the potential impact of an adverse event were calculated using Cox proportional-hazards regression models. Additionally, a multivariate analysis considering predefined patient background factors (age, sex, RAS mutation and localization) was performed. p-value < 0.05 was considered statistically significant and all conducted tests were two-sided. All calculations were carried out with IBM SPSS Statistics (version 24).In this multicenter analysis, 85 patients were identified as suitable candidates between January 2013 and December 2017, as shown in Table 1. All patients had colorectal cancer with liver-, lung-, lymph node-, bone- or other metastases and have been previously treated with current standard first- and second-line therapies.A total of 35 patients from our database were treated with regorafenib as last line therapy, while 16 patients received only trifluridine/tipiracil. Additionally, a total of 32 patients received regorafenib before trifluridine/tipiracil, whereas only two patients were treated with trifluridine/tipiracil before the initiation of regorafenib therapy.The regorafenib and trifluridine/tipiracil cohorts comprised 69 and 50 patients, respectively. The regorafenib cohort included 49 (71%) men and 20 (29%) women, as compared to 34 (68%) men and 16 (32%) women in the trifluridine/tipiracil cohort. The median age at initial diagnosis was 60 (33–81) years in the regorafenib group and 59 (33–80) years in the trifluridine/tipiracil group. The majority of tumors, 78% of each treatment cohort, were located on the left side of the colon, including the rectum. Ras mutation status was evenly distributed between both treatment groups, whilst most of the patients of both treatment arms were BRAF wildtyp (91% and 88%). The starting dose of regorafenib was 160 mg/d in 46 (67%) patients, while 23 (33%) patients started with 120 mg or less (Table 1).Among the 69 patients treated with regorafenib, the most frequently reported adverse event was fatigue, with 29 cases (42.0%). The second most frequent side effect, with 25 cases (36.2%), was hand-foot-skin-reaction, followed by 24 patients (34.8%) who reported hoarseness. A further 21 cases (30.4%) of weight loss or anorexia, 14 cases (20.3%) each of nausea or vertigo and hypertension, 13 cases (18.8%) of diarrhea, 11 cases (15.9%) of oral-mucositis and stomatitis, 8 cases (11.6%) each of abdominal pain and loss of appetite and 6 cases (8.7%) of paresthesia and neuropathy were reported, as shown in Table 1.In the adverse event single observation for regorafenib, a log rank test for OS showed a significant difference (p < 0.001) with respect to the occurrence of nausea and vertigo vs. nausea and vertigo absent, and a log rank test for PFS showed a significant difference (p = 0.032) with respect to the occurrence of oral mucositis and stomatitis vs. the absence of of oral mucositis and stomatitis. With regard to disease control rate (DCR), there was a significant difference in the incidence of oral mucositis and stomatitis (exact Fischer-test: p = 0.039).These differences remained even after consideration of the co-variables age, sex, Ras status and localization in a multivariate model, as shown in Table 2. We found that the occurrence of nausea or vertigo is significantly associated with a shorter overall survival (HR = 3.621; 95% CI: 1.519–8.630; p = 0.004) under regorafenib therapy. The multivariate analysis also showed that the occurrence of oral mucositis and stomatitis is a statistically significant parameter for shortened PFS (HR = 3.258; CI: 1.381–7.687; p = 0.007) in regorafenib therapy. Oral mucositis and stomatitis appears to be a statistically significant protective parameter in regard to the DCR (OR = 0.071; CI: 0.007–0.718; p = 0.0025). In addition, our multivariate analysis demonstrated that Ras-gene wildtype conditions were significantly associated with shorter PFS in patients that experienced oral mucositis or stomatitis (HR = 2.385; CI: 1.113–5.113; p = 0.025), as well as nausea or vertigo (HR = 2.197; 95% CI: 1.033–4.671; p = 0.041).In the trifluridine/tipiracil treatment arm, fatigue also occurred in 26 patients (52.0%) as the most common adverse event. A total of 17 patients (34.0%) suffered from nausea and vertigo, while, in 13 cases (26.0%), anemia was recorded as the third most frequent adverse event. Another 12 cases (24.0%) of diarrhea, 11 cases (22.0%) of leukopenia, 10 cases (20.0%) of neutropenia, 9 cases (18.0%) of vomiting, 8 cases (16.0%) of weight loss or anorexia and 7 cases (14.0%) of abdominal pain were reported, as shown in Table 1.A log rank test for PFS showed significant differences with respect to the occurrence of neutropenia (p = 0.043) and leukopenia (p = 0.010) under trifluridine/tipiracil treatment, as shown in the Kaplan–Meier curves in Figure 1A,B, respectively. Regarding OS, the occurrence of leukopenia (Figure 1C), and weight loss or anorexia, showed a significant difference in the log rank test (p = 0.044 for each). In respect of the a priori-defined co-variables, the differences in the PFS remained. However, the difference in OS persisted only in terms of the occurrence of weight loss or anorexia, as shown in Table 3.Our multivariate analysis showed that the occurrence of weight loss or anorexia during trifluridine/tipiracil treatment is a statistically significant parameter (p = 0.022) for shortened overall survival (HR = 5.595; 95% CI: 1.286–24.344). Notably, the analysis also showed that neutropenia (HR = 0.345; 95% CI: 0.134–0.893; p = 0.028) and leukopenia (HR = 0.194; 95% CI: 0.066–0.575; p = 0.003) were significantly associated with prolonged PFS under trifluridine/tipiracil treatment. Note that febrile neutropenia was rare—only one patient (n = 1)—and therefore not considered in further analysis. To assess which neutrophil count bears an impact on PFS in more detail, the neutrophil counts were classified into 3 groups (< = 500, 500–1000, >1000). The results regarding PFS showed a p-value of (log-rank test) = 0.408. Additionally, we have analysed a possible impact of neutropenia/leucopenia on PFS/OS on the subgroups “colon” and “rectum”. The results did not show any significant differences. Next to those findings, the disease control rate did not correlate with common side effects in trifluridine/tipiracil-treated patients (Figure 1). In this study, we characterize the efficacy and distinctive toxicity profile of regorafenib and trifluridine/tipiracil in the late-line treatment setting of mCRC patients. The safety profiles of regorafenib or trifluridine/tipiracil treatment resemble the reported adverse events of the respective randomized trials [9,16]. The most frequently reported adverse event of any grade, upon regorafenib treatment, was fatigue (42%), followed by hand-foot-skin reactions (36%), hoarseness (35%), weight loss (30%), nausea or vertigo (20%) and hypertension (20%). Notably, the frequency of weight loss and nausea or vertigo in the regorafenib cohort of any grade in our study was higher than in the CORRECT trial, while the frequency of hand-foot-skin reactions, diarrhea, oral mucositis, hypertension was lower [9]. In the trifluridine/tipiracil cohort, the most frequent adverse events of any grade were fatigue (52%), nausea or vertigo (34%), anemia (26%), diarrhea (24%), leukopenia (22%), neutropenia (20%), vomiting (18%), weight loss or anorexia (16%) and abdominal pain (14%). When compared to the RECOURSE trial, the frequency of fatigue and anemia was higher in our study, whereas the frequency of nausea, neutropenia, diarrhea, vomiting and abdominal pain was lower [16]. The frequency of weight loss or anorexia was not reported in the RECOURSE trial, however, the incidence we found in this study conforms to other previously published data [17,19].When compared to each other, both study compounds show distinguishing toxicity profiles. While trifluridine/tipiracil commonly caused side effects of a hematological nature, regorafenib caused a higher number of adverse events. This finding was corroborated by a recently published meta-analysis [20].Finally, we analyzed the adverse events reported under therapy with regorafenib or trifluridine/tipiracil for predictive value. We found that in the regorafenib cohort, the occurrence of nausea or vertigo was predictive for shorter overall survival (HR = 3.621, 95% CI: 1.519–8.630, p = 0.004). Furthermore, we were able to show in our model that the presence of oral mucositis predicted a shortened PFS (HR = 3.258, 95% CI: 1.381–7.687, p = 0.007), although it positively influenced the DCR (HR = 0.071, CI: 0.007–0.718, p = 0.025). A hypothesis for the rationale behind this notable finding remains to be further explored in future trials, but it might be speculated that mucositis more often led to treatment interruptions.In regard to trifluridine/tipiracil, we found that weight loss or anorexia predicted a shortened OS (HR = 5.595; 95% CI: 1.286–24.344; p = 0.022). On the other hand, longer PFS was predicted by the occurrence of leukopenia (HR = 0.194, CI: 0.066–0.575, p = 0.003) and neutropenia (HR = 0.345; CI: 0.134–0.893; p = 0.028) in our patients and those side effects seemed to have positive effects on OS, however, this trend did not reach statistical significance in our analysis. We also investigated which neutrophil count bears an impact on PFS in more detail. Furthermore, we have analysed a possible impact of neutropenia/leucopenia on PFS/OS on the subgroups “colon” and “rectum”. However, no significant impact was determined, possibly due to the small sample size of the study. While the predictive quality of neutropenia for PFS in trifluridine/tipiracil therapy has already been reported in a prior trial by Hamauchi et al., our study is the first to report predictive value of leukopenia for PFS [23]. The reason for this effect still has to be elucidated. One possible explanation might be the fact that those patients who did not show better survival parameters might suffer from an increased tumor burden, leading to an increased baseline neutrophil count resulting in less neutropenia during treatment. Another explanation might be that those patients experiencing neutropenia might have been treated with the optimal dosage, while the others might be under-dosed.Our findings underline that some of the common adverse events seen in patients treated with either one of the study drugs actually have prognostic value. These insights might have impact on the clinical routine, as they allow clinicians to evaluate certain adverse events such as neutropenia or leukopenia in patients treated with trifluridine/tipiracil in a more differentiated manner. Furthermore, our characterization of the toxic effects of the investigated drugs might facilitate the process of finding the adequate treatment for each individual mCRC patient based on their medical history, tolerances and performance status.Our study has also limitations: Baseline pathological assessment of patients in a late line clinical trial was sometimes several years earlier, thus, no MMR status of the patients was available, which might have been interesting for its impact. Additionally, the low patient number with BRAF V600E tumors (n = 2) did not allow further statistical analysis on its role. The analysis was of a retrospective nature and the sample size was limited, although the two cohorts were well matched and recruited in two different centers, ensuring the high intern validity and comparability of our findings. A prospective validation of our observations is recommended. In conclusion, we demonstrated that both regorafenib and trifluridine/tipiracil are effective therapeutic options in the late line setting of mCRC therapy and show a distinct toxicological profile. Some of the adverse events caused by the compounds have predictive value that might support the clinician in the process of developing an individualized, patient-centered therapy in order to ensure their patient’s best possible outcome. Both regorafenib and trifluridine/tipiracil have proven their effectiveness in large, placebo-controlled trials, but in order to further investigate their toxicological profile and their optimal use in the salvage therapy of mCRC, we advocate for prospective head-to-head trials in large patient cohorts. Ultimately, the results of such trials might lead to the establishment of novel clinical practice guidelines for the salvage therapy of mCRC.Conceptualization, T.W. and G.W.P.; Data curation, M.U., S.F., M.J., M.D., A.S., D.B., M.K., H.P. and C.M.; Formal analysis, M.U. and D.B.; Methodology, M.K., C.M. and G.W.P.; Visualization, H.P.; Writing—original draft, M.U., M.J., A.S. and T.W.; Writing—review & editing, G.W.P. All authors have read and agreed to the published version of the manuscript.This research received no external funding.The analysis was supported by an independent research grant from Bayer, Austria and Servier, Switzerland.The authors declare no conflict of interest.A + B: Kaplan–Meier curves showing progression-free survival under trifluridine/tipiracil. (A): KMC adjusted to neutropenia (yes or no) (log rank test: p = 0.043). (B): KMC adjusted to leukopenia (yes or no) (log rank test: p = 0.010). (C): Kaplan–Meier curves showing overall survival under trifluridine/tipiracil. KMC adjusted to leukopenia (yes or no) (log rank test: p = 0.044).Baseline patient characteristics and adverse events (CTCAE any grade).Association between progression-free survival (PFS)/overall survival (OS)/disease control rate (DCR) and adverse events with consideration of a-priori covariates in regorafenib.OS: overall survival, PFS: progression free survival, DCR: disease control rate, HR: hazard ratio, OR: odds ratio.Association between progression-free survival (PFS)/overall survival (OS)/disease control rate (DCR) and adverse events with consideration of a-priori covariates in trifluridine/tipiracil.OS overall survival, PFS progression free survival, DCR disease control rate, HR hazard ratio, OR odds ratio.

Med-MDPI/jcm_2/jcm-09-06-01773.txt

@@ -0,0 +1,2 @@
+The supine sleep position in late pregnancy is a major risk factor for stillbirth, with a population attributable risk of 5.8% and one in four pregnant women reportedly sleeping in a supine position. Although the mechanisms linking the supine sleep position and late stillbirth remain unclear, there is evidence that it exacerbates pre-existing maternal sleep disordered breathing, which is another known risk factor for adverse perinatal outcomes. Given that maternal sleep position is a potentially modifiable risk factor, the aim of this study was to characterize and correlate uteroplacental and fetal hemodynamics, including cardiac function, in a cohort of women with apparently uncomplicated pregnancies with their nocturnal sleep position. This was a prospective observational cohort study at an Australian tertiary obstetric hospital. Women were asked to complete a series of questions related to their sleep position in late pregnancy after 35 weeks of completed gestation. They also underwent an ultrasound assessment where Doppler indices of various fetoplacental vessels and fetal cardiac function were measured. Regional cerebral perfusion was also assessed. Pregnancy outcome data was extracted from the electronic hospital database for analysis. A total of 274 women were included in the final analysis. Of these, 78.1% (214/274) reported no supine sleep, and 21.9% (60/274) reported going to sleep in a supine position. The middle cerebral artery, anterior cerebral artery, and vertebral artery pulsatility indices were all significantly lower in the supine sleep cohort, as was the cerebroplacental ratio. There were no significant differences in the mode or indication for delivery or in serious neonatal outcomes, including 5-min Apgar score < 7, acidosis, and neonatal intensive care unit admission between cohorts. Women in the supine cohort were more likely to have an infant with a BW > 90th centile (p = 0.04). This data demonstrates fetal brain sparing in association with the maternal supine sleep position in a low-risk population. This data contributes to the growing body of literature attempting to elucidate the etiological pathways responsible for the association of late stillbirth with the maternal supine sleep position.Epidemiological data suggests that up to one in three women in early pregnancy and one in five women in late pregnancy sleep in the supine position [1], and overall, almost 27% of pregnant women spend at least some time sleeping on their backs during the night [2]. Sleeping in the supine position is potentially causally implicated in late stillbirth [3,4,5,6,7,8], with almost 56% of women experiencing this complication believing that the time of fetal demise was at night [9]. Currently, it is estimated that the supine sleep position is associated with a 5.8% population attributable risk for late stillbirth [8], although a recent meta-analysis suggested the risk may be substantially higher at 37% [10]. This highlights the importance of maternal sleep position as a possible predisposing etiological cause, and importantly, as a potentially modifiable risk factor to reduce the risk of late stillbirth. Although the mechanisms linking the supine sleep position and late stillbirth remain unclear, there is evidence that sleeping in the supine position exacerbates pre-existing maternal sleep disordered breathing (SDB) [8,11], which is a known risk factor for adverse perinatal outcomes [12,13].In the supine position, the inferior vena cava and aorta are compressed against the maternal spine by the heavy gravid uterus, which causes not only significant reduction in venous return to the maternal heart, but also diminution of aortic blood flow by almost 30% [14,15]. The subsequent reduction in maternal cardiac output results in a drop in uteroplacental perfusion, causing impaired fetal oxygenation and changes in fetal Doppler indices [16,17,18,19,20].Given this background, the aim of this study was to prospectively characterize and correlate uteroplacental and fetal hemodynamics, including cardiac function, in a cohort of women with apparently uncomplicated pregnancies with their nocturnal sleep position.This was a prospective observational study of women aged between 18 and 50 years, with non-anomalous fetuses and uncomplicated singleton pregnancies. To be eligible for inclusion, women were required to have a full medical history available and to have had their pregnancy dated within the first trimester. The study was undertaken at the Mater Mothers’ Hospital in Brisbane, Australia between July 2017 and June 2019. Relevant ethics, governance, and privacy approvals were obtained from the Mater Human Research Ethics Committee and Governance office, respectively (HREC/17/MHS/34). After informed written consent was obtained, all women underwent a single ultrasound assessment in late pregnancy (from 35 weeks of gestation) and completed a sleep questionnaire (Supplementary Table S1), which included specific questions relating to sleep position. The questionnaire asked women to describe their sleep position both going to sleep and waking from sleep before they were pregnant, over the last one month of pregnancy, over the last week of pregnancy, and the night prior to completing the questionnaire. Women were able to simply answer “yes” or “no” to each question and were also asked to quantify the amount of time they spent in each position as rarely (<10%), sometimes (10–50%), or mostly (>50%).For the purposes of analysis, if women answered “yes,” “sometimes,” or “mostly” to question of going to sleep in the supine position at any time over the preceding month, week, or night of pregnancy, they were categorized into the supine cohort, whereas women who said “no” or “rarely” were classified as controls.All ultrasound examinations were carried out using an Affinity 70G (Philips, USA) or a Voluson E8 (GE, Zipf, Austria) ultrasound machine by a single experienced sonographer (NR). The estimated fetal weight (EFW) was calculated using Hadlock’s formula [21]. Doppler parameters of various fetal vessels were measured, namely the middle cerebral artery pulsatility index (MCA PI), anterior cerebral artery pulsatility index (ACA PI), posterior cerebral artery pulsatility index (PCA PI), vertebral artery pulsatility index (VertA PI), umbilical artery pulsatility index (UA PI), and umbilical vein time averaged flow velocity (UV TAV). In addition, fetal cardiac output (CO) and maternal uterine artery pulsatility index (UtA PI) were measured. All measurements were recorded in triplicate over multiple cardiac cycles in the absence of maternal or fetal breathing movements with the average values reported.The MCA, ACA, and PCA were identified in a transverse section of the fetal head using a light transducer pressure and color Doppler. The insonation angle was kept as close to zero degrees as was achievable. The MCA PI was recorded just distal to the circle of Willis, the ACA PI was recorded in the first segment distal to the junction with the internal carotid artery, and the PCA PI was recorded in the second segment distal to the junction with the posterior communicating artery. The VertA was identified in the nuchal region using the color Doppler and was recorded at its location between the first cervical vertebra and the occipital bone. The UA PI was measured from a free loop of cord with the insonation angle < 30°. The UV diameter was recorded in a transverse section of a magnified image with calipers placed at the inner edges of the vessel wall. The UtA PI was recorded at the level of the maternal iliac vessels, and the mean of the left and right uterine arteries was used.The cerebroplacental ratio (CPR) was calculated as the ratio of the MCA PI to the UA PI. Ratios were also generated for the other cerebral vessels to the UA PI, namely the ACA PI/UA PI, PCA PI/UA PI, and VertA PI/UA PI. UV flow in milliliters per minute was calculated using the following formula: Time Averaged Velocity (centimeters per second) × 0.3 × cross-sectional area of the vessel (square millimeters) corrected for estimated fetal weight in kilograms [22].Fetal CO (Left (LCO), Right (RCO), and Combined (CCO)) was calculated using a formula [23] incorporating the stroke volume (SV); time-velocity integral (TVI), obtained by manually tracing the pulse wave waveform, from the left and right outflow tracts, respectively [24]; pulmonary or aortic valve radius (r), respectively; and fetal heart rate (FHR) and corrected for EFW (mL/min/kg):SV (ml) = π × r2 × TVI = π × (valve diameter/2)2 × TVI(1)
+      CO (mL/min) = SV × FHR(2)All obstetric caregivers and participating women were blinded to the ultrasound findings unless a malpresentation or small for gestational age (EFW < 5th centile) or large for gestational age (EFW > 95th centile) fetus was detected. Maternal demographic data and intrapartum and neonatal outcomes were extracted from the electronic hospital database. Indications for operative birth were recorded as those made contemporaneously by the treating obstetric team. A diagnosis of intrapartum fetal compromise (IFC) was made either on the basis of a pathological FHR pattern or an abnormal fetal scalp lactate (>4.2 mmol/L). Neonatal acidosis was defined as a cord artery pH < 7 or base excess of <−12mmol/L. EFW centiles were based on an Australian population and corrected for gender and gestational age [25]. The CPR centiles were also based on an Australian population and corrected for gestational age [26].Sample size calculation was performed by conservatively estimating that the mean MCA PI for gestation would decrease by approximately 10% (from 1.71 to 1.54) at a gestation of 36 weeks in women who slept in a supine position [26]. Using a standard deviation of 0.25, an alpha of 0.05, and a power of 0.9, a sample size of 46 in each group was required.The statistical software package Stata, Release 13, for Windows (StataCorp LP, College Station, TX, USA) was used to perform the statistical analysis.Demographic characteristics were summarized using mean and standard deviation for normally distributed continuous variable, median and interquartile range for non-normally distributed continuous variables, and frequency and percent for categorical variables. A logistic regression model was used to examine the association between supine sleep and ultrasound parameters, as well as perinatal outcomes after adjusting for relevant demographic factors.A total of 411 women met the inclusion criteria and were approached to participate in this study. Although 302 women were eventually recruited, only 274 women answered the complete questionnaire and underwent the ultrasound scan and were included in the final analysis (Figure 1). Overall, 78.1% (214/274) of participants were in the control group and 21.9% (60/274) were in the supine cohort. There were no significant demographic differences between the two groups (Table 1).Table 2 shows the differences for the various ultrasound parameters between the two groups. There were three cerebral parameters that were significantly lower in the supine cohort: The mean MCA PI (1.62 (0.25) vs. 1.74 (0.29), p = 0.001), ACA PI (1.52 (0.27) vs. 1.59 (0.30). p = 0.046), and VertA PI (1.57 (0.27) vs. 1.68 (0.35), p = 0.019). The CPR was also lower in the supine sleep cohort (1.98 (0.37) vs. 2.13 (0.42); p = 0.008) (Figure 2 and Supplementary Table S2). There were however no differences in the UA PI or the ACA/UA, PCA/UA, or VertA/UA ratios. There were also no differences in UtA PI, UV flow, EFW, or EFW centile.Additionally, there were no differences in the fetal cardiac output parameters between the cohorts. The LCO, RCO, and CCO were all lower in the supine sleep cohort (even when corrected for fetal weight) but did not reach statistical significance: Mean LCO (465 mL/min (118.6) vs. 492.5 mL/min (147.0), p = 0.30), RCO (758.93 mL/min (178.37) vs. 778.67 mL/min (189.22), p = 0.68), CCO (1223.9 mL/min (266.5) vs. 1271.1 mL/min (297.9), p = 0.44).Table 3 details the obstetric and neonatal outcomes between the two cohorts. There were no differences in mode of birth, with similar rates of spontaneous vaginal delivery (56.3% (120/213) vs. 66.7% (40/60), p = 0.20]) elective caesarean section (8.9% (19/213) vs. 5.0% (3/60), p = 0.35), instrumental delivery (18.3% (39/213) vs. 15.0% (9/60), p = 0.64), and emergency caesarean section (16.4% (35/213) vs. 13.3% (8/60), p = 0.59) between the control and supine sleep groups, respectively. There was also no difference observed in the indication for operative delivery or serious neonatal outcomes between the two groups. Women in the supine cohort were more likely to have an infant with a BW > 90th centile (p = 0.04).The results of this study demonstrate significant differences in fetal cerebral Doppler indices in nonobese women who slept in a supine position in the third trimester of pregnancy. Specifically, we found that the MCA PI, CPR, ACA PI, and VertA PI were all lower in the supine sleep position cohort consistent with cerebral redistribution.Although the fetal MCA, due to the ease of its imaging, is the cerebral vessel most commonly reported in Doppler studies of fetal wellbeing, there is evidence that cerebral redistribution occurs in a regional, stepwise temporal fashion with early changes occurring in the frontal lobe. The PCA supplies the occipital brain, and changes in this vessel tend to occur later in the redistributive process [27]. Changes in the ACA [27,28] and VertA/UA ratio [29,30] have been shown to be associated with adverse perinatal outcomes or suboptimal fetal growth, although there is a lack of evidence demonstrating its superiority over other parameters of fetal wellbeing [30,31]. The differences in cerebral perfusion are interesting given the lack of fetal or birth weight discordance between our cohorts, suggesting that the maternal supine sleep position may cause subtle circulatory changes independent of perturbations in growth that may increase the vulnerability of these fetuses to adverse outcomes. The lack of difference in UA PI suggests no overt degree of placental dysfunction, however the altered cerebral vascular indices suggest that there is some degree of compensation occurring within the fetus. The lack of evidence of systemic alterations in perfusion could also be due to the low-risk nature of the cohort, which was reflected in the fact there were no differences in perinatal outcomes between cohorts.Interestingly, women in the supine sleep cohort were more likely to have an infant with a BW > 90th centile. Although the literature regarding altered fetal growth and maternal sleep is conflicting, increased birth weight has been previously reported in women with SDB [32,33,34,35]. Nevertheless, given the relatively limited numbers of women in this study, it is possible that this particular finding may have arisen by chance.As the fetal brain sparing effect is mediated by alterations in cardiac output [24,36], we sought to ascertain if we could detect any changes in cardiac output and correlate this with regional cerebral perfusion. Fetal cardiac function has been shown to be impaired in growth restricted cohorts [37,38], as well as appropriately grown fetuses that go on to develop IFC [39]. We previously showed that term fetuses that develop IFC have lower LCO and higher RCO. More recent data have suggested that the CPR, MCA PI, and UV flow are also associated with alterations in fetal cardiac parameters, including the LCO [40]. However, the same study demonstrated that only around 17% of the CPR could be explained by alterations in cardiac indices [40]. The mechanisms underpinning cardiac adaptations to a hostile intrauterine environment are complex. Given that the fetuses in our study were not small, it is possible that the cerebral blood flow changes we identified were mediated more by vascular homeostatic mechanisms (changes in resistance, etc.), rather than more overt cardiac mechanisms. Given that overall our cohort was normally grown and low risk, it is possible that the alterations were too subtle to detect, or that our sample size was not large enough given that it was powered based on MCA Doppler changes.The rate of supine sleep in our cohort was 21.9%, which is consistent with rates reported by other investigators [1,2]. Warland et al. proposed that altered uterine perfusion as a result of supine positioning may be relevant in a scenario where the fetus is already ‘vulnerable’ to hypoxic stress as a result of other factors being present. These may include maternal demographic risk factors, co-morbid conditions, or the presence of pre-existing impaired placental function [41]. Other work has demonstrated that maternal position affects fetal behavioral states and heart rate variability [42] and is a modifiable risk factor that could be targeted to decrease the incidence of stillbirth. Indeed, a recent publication indicated that intervention aimed at reducing time spent in a supine sleep position was feasible, improved maternal and fetal parameters, and did not negatively impact maternal sleep quality [43]. Another recent publication suggested that the supine position was not associated with stillbirth [44], but there were various methodological differences that could have accounted for this alternate finding [45], the most significant of which was the gestation at assessment being earlier when the gravid uterus had less of an impact on maternal circulation.One of the constraints of this study was the use of the self-reported sleep position [46]. However, this is an accepted method of determining sleep position in pregnancy and has been used in other large studies notwithstanding its limitations [3,4,5,7,47]. The going-to-sleep position using self-reported questionnaires has good concordance with video surveillance of the maternal sleep position [48]. Our rationale for using the going-to-sleep position was also relevant, as it is a potentially modifiable risk factor as opposed to the maternal waking-from-sleep position. Reporting women who sleep supine less than 10% of the night with women who sleep supine ≥ 10% of the night may also be a limitation of this study if the fetal response is considered to be dose-dependent. However, in our view, the physiological changes associated with pregnancy in the third trimester would be more magnified when the gravid uterus is at its heaviest, so any time spent in a supine position would have an effect, hence our rationale for including all women who reported sleeping supine ≥ 10% of the time. The use of data from a single time point is also a possible limitation, however the difficulty of pinpointing the exact gestation at which supine sleep began makes a pre-supine sleep assessment difficult to obtain, particularly if a woman sleeps in a supine position throughout pregnancy. The use of a robust sample size, prospective study design, and comprehensive fetal Doppler and cardiac assessment were strengths of this study. As our study was exploratory and hypotheses-generating, it was not powered to demonstrate differences in obstetric or perinatal outcomes. Interestingly, Dunietz et al. [49] objectively assessed women in the third trimester for sleep position, maternal respiratory events, and perinatal outcomes. This study also found no association between maternal supine sleep position and perinatal outcomes. However, it did find an association between maternal supine sleep and more maternal respiratory and oxygen desaturation events, which may account for the fetal redistribution pattern observed in this study.The data presented in this manuscript provides information regarding some of the fetal circulatory changes in women who sleep in the supine position during late pregnancy and may provide some insight into the mechanisms associated with late stillbirth and adverse obstetric and perinatal outcomes in these women given the association between a low CPR and poor outcomes [50,51,52].Our results are pertinent because it is now well known that the majority of late stillbirths, particularly at term, are not small or growth-restricted [53]. Indeed, there is now considerable evidence that fetal cerebral redistribution is associated with adverse outcomes, including perinatal death, even in appropriately grown fetuses [29,54,55,56]. In animal models, it has been proposed that the regulation of cerebral blood flow is area-specific, with some areas more prone to cerebral injury in the event of a hypoxic event [57,58]. It has also been proposed that fetal brain injury may actually be initiated as a consequence of cerebral redistribution, with increased flow as a particular risk factor [59,60]. Animal models exhibiting the same brain sparing effect as human fetuses [61] have demonstrated that pregnant mice experiencing an acute hypoxic event are particularly vulnerable to any further hypoxic injury leading to fetal death [62]. Extrapolating animal and other human data to the context of supine sleep position suggest that similar mechanisms could also be responsible for the increased risk of stillbirth seen in women who sleep on their backs. Further studies are clearly required to elucidate fetal and maternal physiological changes potentially linking supine sleep position to fetal compromise and death.Our findings and those of others [20,63] suggest that some degree of cerebral redistribution occurs in the maternal supine sleep position. Furthermore, there is strong epidemiological evidence that the supine sleep position may be causative for some adverse outcomes, including stillbirth. Given that cerebral redistribution is a physiological response to fetal compromise [64], it is tempting to postulate that supine sleep causes placental dysfunction from the reduction in placental perfusion and results in the subsequent change in fetal hemodynamics. However, in our view, the evidence for such extrapolation is limited and the association should not be considered causative at this stage. Public health advice in many jurisdictions already cautions women against sleeping on their back during pregnancy. Our findings provide further evidence of some of the potential physiological mechanisms underpinning the increased perinatal risks associated with the supine sleep position. However, we are unable to provide any advice regarding the frequency of ultrasound surveillance or timing of birth for women who spend the majority of time sleeping in the supine position, and further research is required.The following are available online at https://www.mdpi.com/2077-0383/9/6/1773/s1, Table S1. Sleep questionnaire, Supplementary Table S2: Maternal demographics by cerebral redistributionN.R. and S.K. conceived the study. N.R. recruited all participants and together with S.O. performed the statistical analysis. S.K. and N.R. drafted the manuscript. All authors have read and agreed to the published version of the manuscript.This research was supported by a Fellowship from the Mater Foundation Limited.The authors declare no conflicts of interest.Recruitment flowchart.Maternal supine sleep position and cerebroplacental ratio (CPR) centile.Maternal demographics by maternal sleep position.†: Student t-test; ‡: Mann-Whitney U-test; §: Chi-squared test; ∆: Fisher’s exact test. BMI: Body mass index; ATSI: Aboriginal and Torres Strait Islander; SD: Standard deviation; IQR: Interquartile range.Fetal Doppler and cardiac parameters by maternal sleep position.* Adjusted for gestational age at ultrasound assessment; # Adjusted for EFW; MCA PI: Middle cerebral artery pulsatility index; UA PI: Umbilical artery pulsatility index; CPR: Cerebroplacental ratio; ACA PI: Anterior cerebral artery pulsatility index; PCA PI: Posterior cerebral artery pulsatility index; VertA PI: Vertebral artery pulsatility index; LCO: Left cardiac output; RCO: Right cardiac output; CCO: Combined cardiac output; UtA PI: Uterine artery pulsatility index; UV: Umbilical vein; mL/min/kg: Milliliters per minute per kilogram; EFW: Estimated fetal weight.Intrapartum and perinatal outcomes by maternal sleep position.* Adjusted for gestational age at ultrasound assessment; IQR: Interquartile range; SVD: Spontaneous vaginal delivery; CS: Cesarean section; IFC: Intrapartum fetal compromise; Em CS: Emergency cesarean section; BW: Birth weight; BE: Base excess; NICU: Neonatal intensive care unit.

Med-MDPI/jcm_2/jcm-09-06-01774.txt

@@ -0,0 +1 @@
+Background. The pandemic of Novel Coronavirus Disease 2019 (COVID-19) is challenging, given the large number of hospitalized patients. Cardiovascular co-morbidities are linked to a higher mortality risk. Thus, patients with Congenital Heart Disease (CHD) might represent a high-risk population. Nevertheless, no data about them are available, yet. Hence, we conducted a nationwide survey to assess clinical characteristics and outcomes in patients with congenital heart disease affected by COVID-19. Methods and Results. This is a multi-centre, observational, nationwide survey, involving high-volume Italian CHD centres. COVID-19 diagnosis was defined as either “clinically suspected” or “confirmed”, where a severe acute respiratory syndrome coronavirus 2 (SARS-CoV2) test had been performed and was positive. Cardiovascular comorbidities were observed among adult patients—atrial fibrillation (seven; 9%), hypertension (five; 7%), obesity (seven; 9%) and diabetes (one; 1%)—but were absent among children. Cardiovascular complications were mainly observed in the “confirmed” COVID-19+ group, consisting of heart failure (9%), palpitations/arrhythmias (3%), stroke/TIA (3%) and pulmonary hypertension (3%). Cardiovascular symptoms such as chest pain (1%), myocardial injury (1%) and pericardial effusion (1%) were also recorded. On the contrary, CHD patients from the clinically suspected COVID-19 group presented no severe symptoms or complications. Conclusions. Despite previous reports pointing to a higher case-fatality rate among patients with cardiovascular co-morbidities, we observed a mild COVID-19 clinical course in our cohort of CHD patients. Although these results should be confirmed in larger cohorts to investigate the underlying mechanisms, the findings of low cardiovascular complications rates and no deaths are reassuring for CHD patients.Facing Novel Coronavirus Disease 2019 (COVID-19) is complex, given the high transmission rate (R0 = 1.4–3.8) and the large number of hospitalized patients, often needing intensive care and ventilation support [1,2,3].According to initial reports from Wuhan and the Hubei region, patients with cardiovascular co-morbidities are at higher risk of morbidity and mortality [4,5,6]. A meta-analysis of observational studies [6] showed the incidences of hypertension, cerebrovascular diseases and diabetes were higher in Intensive Care Unit (ICU)/severe cases (28.8%, 16.7% and 11.7%, respectively) than in their non-ICU/severe counterparts (14.1%, 6.2% and 4.0%, respectively) [6]. Additionally, a single-centre study by Guo et al. [5], on hospitalized COVID-19 patients, stated that the case fatality rate was higher (13.3%) for patients with underlying cardiovascular comorbidities compared to those without (7.6%), with a dramatic increase (69.4%) in patients with underlying cardiovascular comorbidities experiencing myocardial injury also during hospitalization.For this reason, subjects that might be considered at particularly high risk should comprise patients with Congenital Heart Disease (CHD) [7,8], both children and adults.CHD is the most common and global inborn defect. With an increasing life expectancy amongst them, CHD patients are susceptible to acquired cardiovascular and other diseases and to environmental threats, including infectious diseases [9].This risk might be further multiplied if CHD is associated with additional comorbidities—lung disease, pulmonary hypertension and heart failure—or in presence of complex congenital heart disease [7,8,9,10,11].Nevertheless, several knowledge gaps remain around this association, and to date, there are no available published studies on COVID-19 patients, both children and adults, with congenital heart disease.For these reasons, we collected data from a multi-centre, observational, nationwide survey in order to evaluate clinical characteristics and the prevalence of adverse outcomes in patients with congenital heart disease affected by COVID-19.We conducted a multi-centre, cross-sectional, observational, nationwide survey aimed at evaluating consecutive patients with congenital heart disease admitted to Italian Congenital Heart Disease Units affiliated and associated with the CHD working group of the Italian Society of Cardiology, during a six-week period of the initial COVID-19 outbreak in Italy: 21 February–4 April. All patients admitted with CHD, who were diagnosed with COVID-19 and either treated and discharged or who died during hospitalization in the 6-week window, were included independently of their age. COVID-19 was diagnosed according to the guidelines of the World Health Organization [12]. Nasopharyngeal swabs were acquired during hospitalization. Real-time polymerase chain reaction tests were applied to diagnose COVID-19, according to recommended protocols. Infection was established as whether at least two positive test results were observed [13].Data were reported for two groups of patients: (i) confirmed diagnosis of COVID-19 (CHD-Covid-19+), if they presented positive nasopharyngeal swabs with suggestive clinical symptoms; and (ii) suspected diagnosis of COVID-19 (suspCHD-Covid-19), if they were exposed to contagion and presented a suggestive history and clinical symptoms, in the absence of confirmation from nasopharyngeal swabs, where they were not performed or not available.The Congenital Heart Disease-Working Group of the Italian Society of Cardiology (Società Italiana di Cardiologia, SIC) invited all affiliated and associated Italian hospitals to get involved in the survey, including academic and non-academic hospitals with Congenital Heart Disease Units receiving CHD patients (Bergamo Hospital; Ancona Hospital; Policlinico San Donato of Milan; University of Padua; Bambino Gesù Children’s Hospital of Rome; Bambino Gesù Pediatric Hospital of Taormina; University of Messina; Monaldi Hospital of Naples, Italy).Clinical data on patients admitted for the above-reported diagnosis were recorded, including epidemiological, demographical and clinical symptoms and complications, management and outcomes.At each site, a coordinating investigator was in charge of screening consecutive patients admitted to the hospital, data collection and quality. Data were collected using report sheets at single centres. After collection, the participating centers submitted filled-in report sheets to the coordinating unit at Padua University that was in charge of transferring all the data into Excel-based electronic worksheets. The data were finally checked for missing or contradictory entries at the coordinating centre.Cardiovascular complications, as adjudicated through medical records, were: (a) palpitations/arrhythmias (defined as rapid ventricular tachycardia lasting more than 30 s), (b) chest pain, (c) myocardial injury (if serum levels of troponin T (TnT) were above the 99th percentile upper reference limit), (d) heart failure (HF) (defined by a physician diagnosis of HF, which included documentation of one of the following: impaired LV systolic or diastolic dysfunction, pulmonary edema/congestion, dilated ventricles or reduced right ventricular function), (e) stroke/TIA (defined according to the AHA Definition [14]), (f) pulmonary hypertension (increased >20 mmHg compared to baseline values), (g) pericardial effusion, and (h) respiratory failure (defined according to the Berlin Definition [15]).Categorical variables are presented as frequency rates and percentages. Continuous variables are presented as mean and standard deviation (SD).A total of eight high-volume CHD centres (Bergamo Hospital; Ancona Hospital; Policlinico San Donato of Milan; University of Padua; Bambino Gesù Children’s Hospital of Rome; Bambino Gesù Pediatric Hospital of Taormina; University of Messina; Monaldi Hospital of Naples, Italy) participated in data collection. Overall, 76 COVID-19-infected CHD cases were registered from 21 February to 4 April 2020 and were reported in this study, including four children (less 18 years old) and 72 adults. The distributions of the CHDs reported are listed in Figure 1.The children group comprised two girls and two boys, with a mean age of 0.9 years (range: 2 months to 2 years). The adult group included 38 men and 34 women, with a mean age of 36.6 years (range: 21 to 76 years).Looking at single subgroups, nine patients presented a confirmed diagnosis of COVID-19 and were included in the CHD-Covid-19+, while the remaining 67 had no microbiological confirmation (no test available during the outbreak) and were classified according to the WHO guidelines [12] (suspCHD-Covid-19). Underlying cardiovascular comorbidities and risk factors are summarised in Table 1. They were observed in adult patients—especially atrial fibrillation (seven; 9%), hypertension (five; 7%), obesity (seven; 9%) and diabetes (one; 1%)—but were virtually absent in the paediatric population.Cardiovascular complications were mainly observed in the CHD-Covid-19+ group (Figure 2). The most represented cardiovascular complication, among the whole population collected for the survey, was heart failure (9%), followed by palpitations/arrhythmias (3%), stroke/TIA (3%) and pulmonary hypertension (3%). Chest pain (1%), myocardial injury (1%) and pericardial effusion (1%) were also recorded (Figure 2).When CHD-Covid-19+ patients were considered alone (Table 2), heart failure was observed in 55% of the total cases, while palpitations/arrhythmias, stroke/TIA and pulmonary hypertension were observed in 22% of the cases. In one out of nine CHD-Covid-19+ patients were observed chest pain, myocardial injury or pericardial effusion.Finally, one patient from the CHD-Covid-19+ group presented severe respiratory failure and was treated with C-PAP in ICU.All the patients from the suspCHD-Covid-19 group presented with mild infection, without any severe symptoms or complications (Figure 3). They recovered within 1 to 2 weeks with prompt symptomatic treatment. They underwent no specific cardiac treatment.Among the confirmed COVID-19 positive cases, two patients developed shortness of breath and received oxygen therapy, two were managed in ICU because of cardiac complications and were treated with inotropic drugs (one received ECMO implantation), and three had pneumonia for which one was treated with CPAP therapy (Figure 3). No death was reported in either group (Figure 3), and the rate of discharge from the hospital was 100%.Based on the sources of data used in this study, patients—both children and adults—with CHD were infected with COVID-19 and were admitted to Italian CHD Units between 21 February–4 April.To the best of our knowledge, this is the first study on COVID-19 including patients with congenital heart disease.We reported detailed cardiovascular information on seventy-six COVID-19-infected patients with CHD, with a particular focus on the number of cardiovascular complications, underlying cardiovascular comorbidities/risk factors, and fatal outcomes.In the current study, among the 76 patients with COVID-19, 9% exhibited heart failure, which was the most common cardiovascular complication, followed by palpitations/arrhythmias (3%), stroke/TIA (3%), pulmonary hypertension (3%), and myocardial injury (1%).Furthermore, when CHD-Covid-19+ patients were considered alone (Table 2), heart failure was observed in 55% of the total, while palpitations/arrhythmias, stroke/TIA and pulmonary hypertension were observed in 22% of cases.In agreement with study from Guo et al. [5], demonstrating that patients with underlying CVD and other comorbid conditions are more prone to experience myocardial injury during the course of COVID-19, we also observed a high amount of myocardial involvement in CHD COVID-19-infected patients by summing-up those with myocardial injury and those with overt heart failure.Moreover, similar to other previous reports from the general population [4,16,17], when analysing cardiovascular comorbidities and/or risk factors from our CHD patients, cardiovascular comorbidities other than CHD were relatively common in adults and were absent in the paediatric population. In fact, atrial fibrillation and obesity accounted for 9%, hypertension for 7%, and diabetes for 1% of the total.The child group comprised two girls and two boys, with a mean age of 0.9 years. On the contrary, the adult group, with a mean age of 36.6 years, included 38 men and 34 women, in line with previous studies showing higher percentages of infection in men than in women [4,17].Finally, against what has been reported to date on patients with cardiovascular disease [4,5,6], no deaths were observed in the whole cohort of patients with CHD; most of them presented with mild infection, without any severe symptoms or complication. The implications of the results in relation to general or immunological COVID-19 therapies [18] will be considered in an extended study.Given the tremendous number of infections reported in Italy [19], the number of COVID-19-infected patients with CHD identified was relatively limited, especially if we consider the high CHD case load per year (about 300 per year for each) admitted to Italian CHD Units participating in our survey.Some potential explanations might be speculated. Firstly, this phenomenon may be partially due to reduced chances of exposure of those patients, or incomplete recognition due to mild or asymptomatic disease, rather than resistance to infection. In our experience, families tend to be very protective of children or young adults with a history of CHD, and this might also play a role in reducing the exposure of CHD patients. Additionally, the Italian Government imposed a strict lockdown during the accelerating phase of the pandemic [20]. Consequently, patients with CHD together with those with comorbidities were advised to protect themselves from potential exposure to the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) even more than the general population.While it is true that congenital heart disease affects a relatively younger population, therefore less at risk for COVID-19 adverse outcomes, on the other hand, we did not observe, in our cohorts of congenital heart disease patients, the same critical respiratory outcomes and the higher mortality risk that have been previously described for cardiovascular diseases [4,5,6] other than CHD. These findings may suggest that that the occurrence of cardiovascular risk factors, like older age, obesity, hypertension and diabetes, which is much lower in CHD than that reported in studies on the general population [21], may play a major role in determining COVID-19 mortality, rather than cardiac disease per se. However, further evidence is awaited, as are randomised control trials with larger populations to confirm our observations.The main limitation of the study was the possible lack of inclusion of asymptomatic patients. Even though our survey was conducted systematically through eight high-volume Italian CHD Units, covering different geographical areas, the epidemic has been spreading rapidly, and the incomplete identification of infected patients is possible.Despite several reports suggesting that SARS-CoV-2 may have a more severe clinical presentation and higher fatality rate, in patients with cardiovascular co-morbidities, most of the patients included in our observational cross-sectional study presented an overall mild clinical course, and there were virtually no deaths for the congenital heart disease patients. Although the underlying reasons are not yet understood, these results are reassuring in the current literature panorama reporting an association between cardiovascular risk factors and the fatality rate in COVID-19 patients.Conceptualization, G.D.S. and J.S.; Methodology, G.D.S.; Software, M.A., V.B. and F.B.; Validation, G.D.S., C.I. and M.C.; Formal Analysis, G.D.S. and J.S.; Investigation, P.F., M.C., F.B., P.C., A.S., L.O., M.A., V.B., G.C., M.P.C., M.G.R., P.P.B., P.G. and C.I.; Resources, NONE; Data Curation, G.D.S., C.I. and J.S.; Writing—Original Draft Preparation, J.S.; Writing—Review and Editing, G.D.S. and C.I.; Visualization, P.F., M.C., F.B., P.C., A.S., L.O., M.A., V.B., G.C., M.P.C., M.G.R., P.P.B. and P.G.; Supervision, G.D.S. and C.I.; Project Administration, G.D.S. and C.I.; Funding Acquisition, NONE. All authors have read and agreed to the published version of the manuscript.This research received no external funding.This study was conducted by the Working Group on Congenital Heart Disease of the Italian Society of Cardiology (SIC).The authors declare no conflict of interest.Congenital Heart Disease distribution on admission to high-volume Italian CHD centres with “clinically suspected” or “confirmed” COVID-19 diagnoses. CHD = Congenital Heart Disease; ASD = atrial septal defect; VSD = ventricular septal defect; PA = pulmonary stenosis; AS/BAV = aortic stenosis/bicuspid aortic valve; COA = coarctation of the aorta; DORV = double outlet right ventricle; TOF = tetralogy of Fallot; TGA = transposition of the great arteries; ccTGA = congenitally corrected transposition of the great arteries; AVSD = atrioventricular septal defect; TCPC = total cavopulmonary connection; CHD-Covid-19+ = confirmed diagnosis of COVID-19; suspCHD-Covid-19 = clinically suspected COVID-19 (no confirmation test available during the outbreak).Cardiovascular complications of patients with congenital heart disease and CHD-Covid-19+ (confirmed diagnosis of COVID-19) or suspCHD-Covid-19 (clinically suspected COVID-19). N = number of cases.Clinical manifestations and outcome of patients with Congenital Heart Disease and “clinically suspected” or “confirmed” COVID-19 diagnosis.Clinical characteristics of patients with Congenital Heart Disease and COVID-19 infection.Values are mean or n (%).Characteristics of 9 patients with Congenital Heart Disease infected with COVID-19.CHD = congenital heart disease; CV = cardiovascular; TCPC = total cavopulmonary connection; TGA = transposition of the great arteries; LV = left ventricle; PA = pulmonary artery; PAx = pulmunary atresia; VSD = ventricular septal defect; pAVSD = partial atrioventricular septal defect; PAB = pulmonary artery banding; DM = diabetes mellitus; AF = atrial fibrillation; PH = pulmonary hypertension; PLE = pleural effusion; HF = heart failure; MyoInj = myocardial injury; PEE = pericardial effusion; COA = coarctation of the aorta.

Med-MDPI/jcm_2/jcm-09-06-01775.txt

@@ -0,0 +1,384 @@
+Purpose: This feasibility study investigated whether a year-long combined fitness wristband-based and personalized exercise prescription intervention improves Chinese breast cancer survivors’ (BCSs) health outcomes. Methods: Ninety-five BCSs (Xage = 44.81 ± 7.94; XBMI = 22.18 ± 3.48) were recruited from Southern region of China and were delivered the exercise intervention across 12 months, using a single group pretest–posttest design. Participants’ lipid profile (e.g., total cholesterol, high-density lipoprotein, low-density lipoprotein, and triglycerides), blood glucose, breast cancer biomarkers (e.g., carcinoembryonic antigen and cancer antigen 15-3), and functional fitness (e.g., strength in arms and legs, endurance, balance, agility, and flexibility) were assessed at baseline and 12-month post-intervention. Results: Thirty-three BCSs successfully completed the intervention. A significant change in blood glucose (mean difference (MD): −0.22, 95% confidence interval (CI): −0.41–−0.03, t = −2.25, p = 0.028) was observed, with participants demonstrating lower levels of blood glucose at the 12-month post-intervention versus the baseline assessment. Notable changes in functional fitness were also discerned, including agility and balance (MD: −0.47, 95% CI: −0.68–−0.26, t = −4.336, p < 0.001), aerobic endurance (MD: 89.25, 95% CI: 73.82–104.68, t = 11.336, p < 0.001), lower-body flexibility (left) (MD: 4.58, 95% CI: −4.4–13.56, t = 4.653, p < 0.001), and lower-body flexibility (right) (MD: 4.84, 95% CI: −4.65–14.33, t = 4.092, p < 0.001). Conclusion: The observations suggested that our behavioral change program might promote certain health outcomes in Chinese BCSs, yet we are unable to recommend such a program given existing limitations. Future randomized control trials with diverse samples are warranted to confirm our findings.The most recent global estimates suggested that about 1.7 million new cases of breast cancer are diagnosed every year worldwide and that the vast majority of whom are females [1]. Cancer brings patients with physical, physiological, and psychological side effects including muscular atrophy, weight changes, lowered aerobic capacity, decreased strength and flexibility, fatigue, depression, and an overall decline in the quality of life [2]. Due to treatment advancements, luckily, survival following breast cancer diagnosis has greatly increased, with the average five-year net survival for women having invasive breast cancer remaining highest in developed countries such as U.S. (90%) [3] and Canada (87%) [4]. In contrast, however, approximately 60% of deaths from breast cancer occurred in developing countries [1], representing a huge variation among countries. In China, the five-year survival rate for breast cancer patients has reached 83.2%, which illustrated a 7.3% increase over the past decade [5]. However, the five-year survival rate is still lower than that of developed countries. Clinical practice guidelines recommend that breast cancer survivors (BCSs) should receive health promotion counselling in relation to obesity, physical activity (PA), nutrition, and smoking cessation [6,7,8]. Supported by scientific evidence, major health organizations such as the U.S. Department of Health and Human Services, the American Cancer Society, and the American College of Sports Medicine (ACSM) commonly identify the importance of incorporating PA into cancer care and advise patients to be physically active [9,10]. Review evidence also demonstrated the beneficial role of PA in mitigating several adverse effects of breast cancer and its treatment, including fatigue [11], upper-limb dysfunction [12], PA level, aerobic fitness, muscular strength, anxiety, self-esteem, and quality of life [13]. A recent systematic review indicated that patients who exercised following a diagnosis of cancer tended to have a lower relative risk of cancer mortality and recurrence and to experience fewer/less severe adverse effects [14], echoing the view that PA has been an important holistic therapy for the disease management in BCSs. BCSs’ PA levels decrease remarkably after breast cancer diagnosis and increase only slowly after the treatment period, causing a decline in functional fitness and overall well-being [15,16]. Despite successfully promoting people’s knowledge and favorable intentions to adopt healthy behaviors, interventions to change health behaviors have largely failed to establish enduring healthy lifestyle habits due to insufficient intervention lengths [17], highlighting that long-term behavior change is the key to developing healthy habits. This corresponds with previous research that the short-term PA programs would likely limit the ability to detect specific physiological responses with interventions in cancer patients [18], and thus, larger trials that examine the long-term benefits of PA are needed for this patient group [19].While evidence for the benefits of PA in BCSs continues to grow, the application of well-established exercise-training principles in the field of PA research and practice are not commonly being reported in the oncology literature [20]. Incomplete reporting of the exercise prescription (i.e., components of frequency, intensity, time, and type) and adherence to the prescription would restrict the reproducibility of PA intervention and the ability to determine the PA dose received by BCSs. This mirrors most review evidence that no PA studies in women diagnosed with breast cancer (1) attend to all key principles of exercise training, (2) describe all components of the exercise prescription in the methods, and (3) report adherence to the prescribed intervention in the results [21], all of which underline the need for reliable clinical trials that focus on delivering evidence-based, elaborate, and personalized exercise prescriptions. Today, the application of emerging technology in health promotion has generated substantial public interest [22], with wearable technology such as smartwatches and fitness wristbands being the most exciting and technologically advanced. This is because of its ability to track health metrics like step counts, heart rate, energy expenditure, and sleep [23]. Applying emerging technology to encourage PA among various populations, therefore, has generated substantial public interest [22]. It is worth noting that wearable activity trackers are perceived as useful and acceptable tools by BCSs for self-regulation of PA and sedentary behaviors and thus are recommended for health promotion [24]. Previous studies have also observed wearable devices such as Polar smartwatch to be effective in promoting PA, quality of life, and healthy weight status in BCSs [25,26], yet little to no research has been conducted on the long-term effectiveness of wearable activity trackers in the promotion of this population’s physical and clinical outcomes. It is also suggested that interventions may be more effective if focused on culturally appealing personalized exercise behaviors [27]. Therefore, we conducted the current study to investigate the feasibility of using a commercially available fitness tracker, employed in conjunction with a personalized exercise prescription to promote BCSs’ health outcomes.BCSs were recruited via posted flyers and word of mouth in Guangdong Provincial People’s Hospital. Interested BCSs were thereafter contacted by the researcher staff for screening against the following inclusion criteria: (1) previously diagnosed with stages 0–III breast cancer; (2) completed primary cancer treatment between 1–5 years earlier with no new cancer diagnosis or recurrence; and (3) did not participate in other health promotion programs. Exclusion criteria included (1) currently undergoing breast cancer treatment and (2) having any contraindications that might interfere with PA engagement, such as medical conditions and pacemaker implant. The current study was conducted between April 2016 and July 2017 in the city of Guangzhou, a tier-one metropolitan area in Southern China, using a single group pretest-posttest design. Specifically, BCSs were recruited from early April to mid-June 2016, with those who were interested in participating contacted by the research staff and screened against inclusion criteria. During this period, the baseline assessment was conducted with the eligible BCSs, including demographic/anthropometric information, clinical characteristics, and health outcomes. Notably, ethnicity and socioeconomic status were not used as enrollment screening criteria and thus were not collected in the current study. The BCSs were then provided with a Xiaomi wristband (Mi Band 1S Heart Rate Wristband, Xiaomi Corporation, Beijing, China) to monitor activity behaviors for one month at the end of June 2016. Like other fitness trackers (e.g., Fitbit Charge 3 and Polar A370), the Mi band can track health metrics, including steps taken, heart rate, calories burned, etc. In comparison to other commercially available fitness trackers, however, the Mi band has the advantage of lower price, which only costs between $20 to $25 USD. At the end of July 2016, all BCSs were instructed via phone to upload their health metrics data on the Mi band to a shared online server provided by the research team with the assistance of a high-tech company.It has been recommended that sedentary women should take 10,000 steps per day to receive health benefits [28,29]. Therefore, 10,000 steps per day was set as the median cutoff points for the 36 step categories while clustering BCSs with specific exercise prescriptions (see Table 1). Based upon daily steps taken for the past month, BCSs were first categorized into groups: (1) ≤7999 steps/day, (2) 8000–12,799 steps/day, and (3) ≥12,800 steps/day. In each category, BCSs were further classified as 12 different levels of exercise prescription. For example, if one was categorized as group 1 (e.g., ≤7999 steps/day) and her actual daily steps taken were less than 2499, she was then provided the level 1.1 exercise prescription. If one’s daily steps were 2500–2999, then that patient was given the level 1.2 exercise prescription (see Appendix A). The specific classification is presented in Table 1. To initiate the intervention, a personalized exercise prescription for next month (4 weeks) was delivered to each BCS by the research staff at the end of July 2016. For instance, BCSs given the level 1.1 exercise prescription were instructed to perform 10-min strengthening (e.g., squat and bicycle crunch) and 30-min moderate aerobic exercise (e.g., brisk walk) on day 1 and 10-min strengthening on different muscle groups and 30-min moderate aerobic exercise on day 2. Some sample specific exercise prescriptions for several levels (i.e., 1.1, 1.2, 2.1, and 2.2) are described in Appendix A, which were guided by the ACSM Exercise Guidelines for Cancer Survivors [9].To ensure that our health behavior change program worked while avoiding other confounders, BCSs were instructed not to participate in other health promotion programs for the duration of the study. During the 12-month intervention, BCSs were asked not to change their usual diet and were required to implement their own exercise prescriptions weekly and to wear the Mi band to track their steps taken daily. The activity data (i.e., average daily steps taken) collected by the Mi band for the past month were manually uploaded to an online server and received by the research staff at the end of each month. Afterwards, the personalized exercise prescriptions for the next month based on steps were delivered to BCSs. The intervention lasted 12 months until July 2017. Post-intervention assessments including all outcomes outlined previously were conducted at the end of July 2017. All BCSs received the Mi band as the nonmonetary incentive, regardless of whether they successfully completed all study procedures or not. All procedures performed with BCSs in the current study were in accordance with the ethical standards of the institutional and/or national research committee and with the 1964 Helsinki declaration and its later amendments or comparable ethical standards [30]. University approval and informed consent were documented prior to any data collection.Demographic and Anthropometric Information: BCSs self-reported date of birth, breast cancer diagnosis stage, menstrual status, and types of breast cancer. Height without shoes was measured to the nearest 0.1 cm using a Seca stadiometer (Seca, Hamburg, Germany) while weight and body fat percentage were assessed via the Tanita BC-558 IRONMAN® Segmental Body Composition Monitor (Tanita, Tokyo, Japan) after voiding or wearing light clothing and no shoes or socks. Body mass index (BMI) was defined as weight/height2 and was expressed in kilograms per square meter. Waist circumference was measured to the nearest 0.1 cm using a retractable medical tape in the standing position at the level of the umbilicus, and hip circumference was measured to the nearest 0.1 cm at the largest posterior extension of the buttocks. Health Outcomes. All testing was conducted at Guangdong Provincial People’s Hospital, including lipid profile (i.e., total cholesterol, high-density lipoprotein, low-density lipoprotein, and triglycerides), blood glucose, breast cancer biomarkers (i.e., carcinoembryonic antigen and cancer antigen 15-3), and functional fitness (i.e., strength in arms and legs, endurance, balance, agility, and flexibility) were assessed at baseline and 12-month post-intervention. Specifically, fasting venous blood samples of the participants were drawn and analyzed by the ADVIA 2400 Chemistry System (Siemens Healthcare Diagnostics) for the measurements of total cholesterol, high- and low-density lipoprotein, triglycerides, and blood glucose. In addition, serum levels of carcinoembryonic antigen and cancer antigen 15-3 were determined via electrochemiluminescence immunoassay and chemiluminescence enzyme immunoassay, using a kit provided by Roche Diagnostics Ltd. (Rotkreuz, Switzerland) and Fujirebio Inc. (Tokyo, Japan), respectively. To ensure the precision and accuracy of measurement, daily internal quality control was performed with commercial quality control materials (BioRad Labs). Patients’ functional fitness was assessed via the Senior Fitness Test (SFT), which consists of six functional measures of strength in arms and legs, endurance, balance, agility, and flexibility [31]. The detailed description of the SFT is presented in Table 2. The SFT battery has been shown high reliability and validity (intraclass correlation coefficient (ICC) ranged between 0.79 and 0.97) in assessing physical functioning [32] and has been widely used among various clinical populations, including BCSs [33]. The SFT battery requires approximately 30–40 min per patient to administer. All analyses were conducted in the open-access software RStudio version 3.5.0 (The R Foundation, Vienna, Austria). Data were first screened for the outliers and normality of distributions, after which the outliers were adjusted to lessen the impact of extreme scores and the skewed data were log transformed to improve shape characteristics. Second, descriptive statistics were calculated and presented as mean ± standard deviation and as frequency, unless stated otherwise. Lastly, to evaluate significant changes on BCSs’ health outcomes, linear mixed models were performed, with the participant being treated as the random effect. This analytic approach reduces concerns regarding missing data, as it allows to incorporate all available data and takes into account the fact that participants were lost during the study period due to withdrawal or loss of contact [34]. All models were fitted using the lme4 package [35], and the t-tests used in the interpretation of model effects were computed using the ImerTest package [36]. Statistical significance was set at p < 0.05. Notably, sensitivity analysis was also performed using only the complete-case data and showed similar results. A total of 175 BCSs were screened for eligibility, with 80 who failed to meet inclusion criteria were excluded. As a result, 95 BCSs (Xage = 44.81 ± 7.94; XBMI = 22.18 ± 3.48) initially enrolled and started the intervention. Notably, 62 BCSs were considered incomplete cases due to (1) inadequate adherence (defined as less than 50% of prescribed sessions attended) to the prescribed intervention; (2) inability to finish the post-assessment; (3) changes in health status unrelated to the study and thus cessation of the intervention; (4) withdrawal; and (5) loss of contact. Finally, a total of 33 BCSs successfully finished all the study protocols and had complete pre- and post-assessment data, causing a retention rate of 35%. Complete baseline characteristics of the samples, including initial, nonadherence, and adherence are presented in Table 3. Despite low retention, a good to excellent adherence to the exercise prescription was self-reported by these participants, with adherence rate ranging from 75% to 90%, and the mean adherence rate was 83%. Of note, we also examined the baseline differences in participants characteristics between nonadherence and adherence samples and found no significant difference between these two subgroups (data are not shown).Descriptive characteristics for the outcomes of interest at baseline and 12-month as well as inferential statistics for baseline and posttest on health outcomes are shown in Table 4. Specifically, there was a significant change in blood glucose over time (mean difference (MD): −0.22, 95% confidence interval (CI): −0.41–−0.03, t = −2.25, p = 0.028), with BCSs demonstrating lower levels of blood glucose at the 12-month versus baseline assessment. Significant changes on other health outcomes were not observed, including cancer antigen 15-3 (MD: 0.006, 95% CI: −1.394–1.406, t = 0.008, p = 0.993); carcinoembryonic antigen (MD: 0.13, 95% CI: −0.12–0.38, t = 0.301, p = 0.759); high-density lipoprotein (MD: −0.05, 95% CI: −0.14–0.04, t = −0.929, p = 0.362); low-density lipoprotein (MD: −0.03, 95% CI: −0.10–0.04, t = −0.230, p = 0.820); total cholesterol (MD: −0.14, 95% CI: −0.42–0.14, t = −0.754, p = 0.456); and triglyceride (MD: 0.18, 95% CI: −0.18–0.54, t = 0.916, p = 0.364). The results also suggested that our behavior change program was effective at improving BCSs’ functional fitness, including agility and balance (MD: −0.47, 95% CI: −0.68–−0.26, t = −4.336, p < 0.001), aerobic endurance (MD: 89.25, 95% CI: 73.82–104.68, t = 11.336, p < 0.001), lower-body flexibility (left) (MD: 4.58, 95% CI: −4.4–13.56, t = 4.653, p < 0.001), and lower-body flexibility (right) (MD: 4.84, 95% CI: −4.65–14.33, t = 4.092, p < 0.001). Of note, BCSs’ aerobic endurance demonstrated the greatest improvements among all outcomes after the 12-month behavior change program. No significant changes were detected in lower-body strength (MD: −0.47, 95% CI: −1.39–0.45, t = −0.486, p = 0.629), upper-body flexibility (left) (MD: 1.40, 95% CI: −1.34–4.41, t = 1.632, p = 0.110), upper-body flexibility (right) (MD: 0.43, 95% CI: −0.41–1.27, t = 0.782, p = 0.439), upper-body strength (left) (MD: 2.31, 95% CI: −2.21–6.83, t = 1.983, p = 0.053), and upper-body strength (right) (MD: 1.70, 95% CI: −1.64–5.04, t = 1.456, p = 0.151). Visual comparisons between the baseline and 12 months across all outcomes of interest are exhibited in Figure 1. BCSs are at risk of greater physical inactivity and sedentary behavior [37]. PA engagement after a breast cancer diagnosis may reduce the risk of death from this disease, with patients who follow PA guidelines are more likely to improve their survival [38]. Previous research suggested that the personalized interventions are effective in promoting healthful lifestyle changes among cancer survivors [39]. Additionally, BCSs have found acceptable and expressed interest in the use of wearable activity trackers to self-regulate their activity behaviors [24]. Therefore, this study examined the feasibility of a combined wristband-based and personalized exercise prescription intervention in the promotion of improved health outcomes in Chinese BCSs. Our observations demonstrated that this type of intervention might improve patients’ blood glucose levels, agility and balance, aerobic endurance, and lower-body flexibility. Physiological and biological indicators play an important role in the management of BCSs [40] and provide a guide for the treatment of breast cancer. Review evidence from randomized controlled trials (RCTs) that included clinical endpoints demonstrated that exercise may trigger beneficial changes in circulating insulin level, insulin-related pathways, inflammation, serum level of high-mobility group box 1, serum interleukin 6 and tumor necrosis factor-α, leptin, total cholesterol, cutaneous T cell-attracting chemokine, and possibly immunity [41,42,43]. Notably, while the positive changes after PA intervention are promising, these changes were not consistently and statistically significant across all included studies, suggesting that the current evidence concerning the effects of exercise on biomarkers in BCSs is controversial. Our observations were congruent with this assertion; only detected blood glucose was significantly improved following a 12-month combined aerobic and resistance training intervention, among other health outcomes. Interestingly, a recent study (Breast Cancer and Exercise Trial in Alberta) [44] by Friedenreich et al. also examined the effects of a year-long PA intervention on BCSs’ biomarkers, including C-reactive protein, insulin, glucose, homeostatic model assessment of insulin resistance, estrone, sex hormone-binding globulin, total estradiol, and free estradiol. In their study, the 400 BCSs were randomly assigned to either moderate-vigorous aerobic exercise, 5 days/week (3 days/week supervised) for 30 min/session (moderate), or 60 min/session (high) conditions. Corresponding with our findings, this study observed that some biomarkers improved for both groups after one year but that neither group observed a significantly greater improvement over the other. This signifies that performing either moderate-duration or high-duration exercises can reduce the risk of breast cancer. It is noteworthy, however, in their study that some biomarkers such as estrogen and insulin stayed the same over time because they are already low in postmenopausal women.It is suggested that PA may be more pronounced on certain biomarkers (e.g., insulin pathway) in obese or physically inactive BCSs [41,45]. Notably, BCSs in our study were relatively fit and active; this could partially explain why other biomarkers were not found to be improved. Moreover, there was also a suggestion that PA may be more effective at modifying certain biomarkers such as insulin-like growth factor 1 in BCSs who were not taking tamoxifen [41]. However, in the present study, we did not investigate whether and how many BCSs were taking tamoxifen during the intervention period, which, to some extent, might be a confounding factor that impacts our experimental results. In closing, while our study observed that a year-long PA program may improve BCSs’ blood glucose levels, future research is warranted to confirm the effects of this innovative intervention on multiple biomarkers.Breast cancer and oncologic treatment may generate significant negative effects on physical functioning [46]. Women who have been diagnosed with breast cancer often experience a rapid decline in functional fitness [47]. The American Cancer Society recommends that BCSs should engage in PA regularly to improve their functional fitness and to reduce the risk of developing new cancers [48]. Evidence consistently demonstrated that PA led to significant improvements in aerobic fitness, increased flexibility, and strength [9]. For example, Milne et al. [49] conducted an RCT of combined aerobic and resistance exercise in BCSs over 24 weeks and observed significant improvements in functional fitness, including aerobic fitness (i.e., the Aerobic Power Index cycle test) and muscular strength (i.e., bicep curls, leg press, and chest extension). Likewise, another larger RCT by Winters-Stone et al. [50] found that BCSs in the resistance training group had better testing scores on maximal leg and bench press strength compared to the stretching group over the course of a year-long intervention, despite no significant improvements noted in grip strength, chair stand, best 4-m usual walk, and one-leg standing balance tests. Moreover, Foley and Hasson [51] recently conducted a 12-week (30 min each, totaling 90 min twice weekly) multimodal exercise intervention (a combination of aerobic conditioning, resistance, balance, and flexibility training) using one single-group pretest–posttest design, to promote BCSs’ functional fitness. The study observed significant improvements in mobility (Timed Up and Go, 6-min walk test), muscular strength (leg press strength and chest press strength), upper-extremity flexibility (back scratch test), and balance (functional reach test and single-leg stance time). Interestingly, our study only observed notable changes in agility and balance, aerobic endurance, and lower-body flexibility but not in lower-body strength, upper-body flexibility, and upper-body strength. It is noteworthy, however, that our sample was recruited from China, whereas Foley and Hasson’s sample was from North America. Previous research appealed that multilevel analysis of individuals and populations within specific contexts would strengthen clinical practice for cancer management [52,53]. Therefore, future studies with racially and ethnically diverse groups are encouraged to ascertain such disparities. The widening disparities in cancer outcomes between Asian- and Euro-Americans challenges the current research and practice paradigms for cancer control. A Cultural Systems Approach would strengthen future studies. This paradigm requires multi-level analyses of individuals and populations within specific contexts in order to identify culturally based strategies to improve practice along the cancer care continuum.The strengths of this study include (1) the application of a combined wristband-based and personalized exercise prescription intervention in BCSs, making it one of its first kind in the literature while providing the preliminary evidence for the acceptability and feasibility of this type of innovative intervention; (2) all components of the customized exercise prescription are driven by clinical practice guidelines, warranting the tolerability and safety of the intervention; and (3) detailed reporting of both exercise prescription and adherence to exercise prescription, which well responds to the call to action [21] and would allow full replication of positive findings in clinical settings. Several limitations within this study, however, should also be noted. First, this study was conducted in one geographic location in a sample of Chinese BCSs, which limits generalizability of observations and somewhat hinders the ability to identify racial differences. Second, we only studied a single group of BCSs and did not include a control group, impeding any ability to draw initial intervention effectiveness conclusions. Third, the study implementation was limited by a high attrition rate that may increase the risk of bias. Indeed, it must be admitted that the retention strategies in this study may not be efficient, which resulted in a high attrition rate while increasing the risk of bias. Nevertheless, clinically promising trends in certain biomarkers and functional fitness have been noted. Fourth, as a feasibility study, we failed to collect patients’ past treatment exposures, which can be confounders and may potentially influence our observations. Future studies are encouraged to expand on this effort. Fifth, due to our recruitment method, differences in physical fitness may preexist among BCSs and thus may potentially impact all outcome measures. Finally, and relatedly, due to low retention rate, qualitative approach was not undertaken, which limits the ability to optimize our design. This study provides the preliminary evidence that a year-long combined fitness wristband-based and individually tailored aerobic and resistance training might be able to promote improved blood glucose levels, agility and balance, aerobic endurance, and lower-body flexibility in Chinese BCSs. However, our observations do not suggest that the programming confers additional beneficial effects on other health outcomes. Given that the findings of this feasibility study were related to the preceding limitations, we are unable to recommend implementing such behavior change program in BCSs yet. Future studies addressing the above limitations with more rigorous study design (e.g., RCTs with mixed methods), racially and ethnically diverse samples, as well as multiple health assessments are warranted to confirm and generalize our findings. While conducting this study, N.Z. played a role in developing measurement strategies, in sorting/analyzing data, and in writing the manuscript. N.L. played a role in coordinating intervention, in data collection, and in editing the manuscript. C.H. played a role in data collection and in editing the manuscript. W.L. played a role in editing the manuscript. Z.G. played a role in developing the idea, in overseeing data collection, and in editing the manuscript. All authors agree with the order of presentation of the authors. All authors have read and agreed to the published version of the manuscript.This research received no external funding.The authors would like to thank Guangdong Provincial People’s Hospital and the research staff who assisted with the study. We also want to thank Zachary, C. Pope from the University of Minnesota for designing the exercise prescriptions for the patients. The results of this study are presented clearly; honestly; and without fabrication, falsification, or inappropriate data manipulation.The authors declare no conflict of interest.(Selected exercise prescriptions)
+          Level 1. Exercise Made Easy: Beginners Programming
+        Below is a 12-week outline (48 workouts) of exercise programming for individuals with little experience in the gym or those just coming back from a long time away from the gym. As such, this workout program is programmed for exercise 4 times a week for roughly an hour, with simple exercises placed within. Notably, under each of the main exercises for each day are suggested alternative exercises which may better appeal to breast cancer survivors recovering from treatment.Finally, for each cardio workout listed below, you will use the Borg Rating of Perceived Exertion scale to gauge your intensity level. This scale is listed below, allowing for comprehension of the intensity level; each cardio workout should be completed at. 
+          Month 1: Endurance Phase
+        For each strength exercise, the endurance phase of this program will include higher numbers of repetitions (10–12 repetitions for each set) for each exercise with lower weight. You should rest for 60 s to 1.5 min between each strength exercise. For each cardio workout, feel free to select your favorite cardio exercise. Some cardio exercise options include but are not limited to walking, jogging, running, biking, swimming, stairclimbing, rowing, elliptical exercise, etc.Day 1:
+		  Strength Workout
+		  ○Dumbbell Lunges (count every other leg): 3 sets × 10–12 repetitions
+		  ■Alternative: Walking Lunges without Dumbbells (still counts every other leg)○Hamstring Curls: 3 sets × 10–12 repetitions
+		  ■Alternative: Glute Kickbacks○Calf Raises: 3 sets × 50 repetitions
+		  ■Alternative: Seated Heel Lifts○Bicycle Crunches: 2 sets × 50 revolutions■Alternative: Standard CrunchesCardio Workout○30-min selected cardio exercise at moderate pace: rating of perceived exertion (RPE) Scale: 4–6Strength Workout
+		  ○Dumbbell Lunges (count every other leg): 3 sets × 10–12 repetitions
+		  ■Alternative: Walking Lunges without Dumbbells (still counts every other leg)○Hamstring Curls: 3 sets × 10–12 repetitions
+		  ■Alternative: Glute Kickbacks○Calf Raises: 3 sets × 50 repetitions
+		  ■Alternative: Seated Heel Lifts○Bicycle Crunches: 2 sets × 50 revolutions■Alternative: Standard CrunchesDumbbell Lunges (count every other leg): 3 sets × 10–12 repetitions
+		  ■Alternative: Walking Lunges without Dumbbells (still counts every other leg)Alternative: Walking Lunges without Dumbbells (still counts every other leg)Hamstring Curls: 3 sets × 10–12 repetitions
+		  ■Alternative: Glute KickbacksAlternative: Glute KickbacksCalf Raises: 3 sets × 50 repetitions
+		  ■Alternative: Seated Heel LiftsAlternative: Seated Heel LiftsBicycle Crunches: 2 sets × 50 revolutionsAlternative: Standard CrunchesCardio Workout30-min selected cardio exercise at moderate pace: rating of perceived exertion (RPE) Scale: 4–6Day 2:Strength Workout○Dumbbell Bench Press: 3 sets × 10–12 repetitions
+			■Alternative: Push-Ups on Knees (only go down as far as your mobility will allow)
+			○Tricep Kickbacks: 3 sets × 10–12 repetitions
+			■Alternative: Tricep Dips on a Chair (only go down as far as your mobility will allow)
+			○Dumbbell Chest Flies (go light and pretend to “hug a tree”): 3 sets × 10–12 repetitions
+			■Alternative: Wide-Grip Wall Push-Ups
+			○Bicycle Crunches: 2 sets × 50 revolutions
+			■Alternative: Standard Crunches
+			
+			Cardio Workout
+			○30-min selected cardio exercise at high-intensity intervals: These intervals will be comprised of 1 min easy (RPE Scale: 2–3) and 1 min hard (RPE Scale: 7–8) and will be repeated until time is up.
+			
+		  Strength Workout○Dumbbell Bench Press: 3 sets × 10–12 repetitions
+			■Alternative: Push-Ups on Knees (only go down as far as your mobility will allow)
+			○Tricep Kickbacks: 3 sets × 10–12 repetitions
+			■Alternative: Tricep Dips on a Chair (only go down as far as your mobility will allow)
+			○Dumbbell Chest Flies (go light and pretend to “hug a tree”): 3 sets × 10–12 repetitions
+			■Alternative: Wide-Grip Wall Push-Ups
+			○Bicycle Crunches: 2 sets × 50 revolutions
+			■Alternative: Standard Crunches
+			
+			Dumbbell Bench Press: 3 sets × 10–12 repetitions
+			■Alternative: Push-Ups on Knees (only go down as far as your mobility will allow)
+			Alternative: Push-Ups on Knees (only go down as far as your mobility will allow)Tricep Kickbacks: 3 sets × 10–12 repetitions
+			■Alternative: Tricep Dips on a Chair (only go down as far as your mobility will allow)
+			Alternative: Tricep Dips on a Chair (only go down as far as your mobility will allow)Dumbbell Chest Flies (go light and pretend to “hug a tree”): 3 sets × 10–12 repetitions
+			■Alternative: Wide-Grip Wall Push-Ups
+			Alternative: Wide-Grip Wall Push-UpsBicycle Crunches: 2 sets × 50 revolutions
+			■Alternative: Standard Crunches
+			Alternative: Standard CrunchesCardio Workout
+			○30-min selected cardio exercise at high-intensity intervals: These intervals will be comprised of 1 min easy (RPE Scale: 2–3) and 1 min hard (RPE Scale: 7–8) and will be repeated until time is up.
+			30-min selected cardio exercise at high-intensity intervals: These intervals will be comprised of 1 min easy (RPE Scale: 2–3) and 1 min hard (RPE Scale: 7–8) and will be repeated until time is up.Day 3:
+		  Strength Workout○Upright Rows: 3 sets × 10–12 repetitions
+			■Alternative: Forward Arm Raises
+			○One-Arm Rows (pretend you are “starting a lawnmower”): 3 sets × 10–12 repetitions
+			■Alternative: Single Arm Circles
+			  ○Lat Pulldowns: 3 sets × 10–12 repetitions
+			■Alternative: Seated Reaching Exercise (count every other side)
+			  ○Dumbbell Shrugs: 2 sets × 10–12 repetitions
+			■Alternative: Shoulders to Ears
+			  ○Sit-ups: 2 sets × 25 repetitions
+			■Alternative: Standard Crunches
+			  
+			Cardio Workout
+			○30-min selected cardio exercise at moderate pace: RPE Scale: 4–6
+			
+		  Strength Workout○Upright Rows: 3 sets × 10–12 repetitions
+			■Alternative: Forward Arm Raises
+			○One-Arm Rows (pretend you are “starting a lawnmower”): 3 sets × 10–12 repetitions
+			■Alternative: Single Arm Circles
+			  ○Lat Pulldowns: 3 sets × 10–12 repetitions
+			■Alternative: Seated Reaching Exercise (count every other side)
+			  ○Dumbbell Shrugs: 2 sets × 10–12 repetitions
+			■Alternative: Shoulders to Ears
+			  ○Sit-ups: 2 sets × 25 repetitions
+			■Alternative: Standard Crunches
+			  
+			Upright Rows: 3 sets × 10–12 repetitions
+			■Alternative: Forward Arm Raises
+			Alternative: Forward Arm RaisesOne-Arm Rows (pretend you are “starting a lawnmower”): 3 sets × 10–12 repetitions
+			■Alternative: Single Arm Circles
+			  Alternative: Single Arm CirclesLat Pulldowns: 3 sets × 10–12 repetitions
+			■Alternative: Seated Reaching Exercise (count every other side)
+			  Alternative: Seated Reaching Exercise (count every other side)Dumbbell Shrugs: 2 sets × 10–12 repetitions
+			■Alternative: Shoulders to Ears
+			  Alternative: Shoulders to EarsSit-ups: 2 sets × 25 repetitions
+			■Alternative: Standard Crunches
+			  Alternative: Standard CrunchesCardio Workout
+			○30-min selected cardio exercise at moderate pace: RPE Scale: 4–6
+			30-min selected cardio exercise at moderate pace: RPE Scale: 4–6Day 4:
+		  Cardio Workout Only○60-min selected cardio at light/moderate pace: RPE Scale: 3–6
+			
+		  Cardio Workout Only○60-min selected cardio at light/moderate pace: RPE Scale: 3–6
+			60-min selected cardio at light/moderate pace: RPE Scale: 3–6Day 5:
+		  Strength Workout
+			○Goblet Squats with Dumbbell: 3 sets × 10–12 repetitions
+			■Alternative: Air Squats
+			○Hamstring Curls: 3 sets × 10–12 repetitions
+			■Alternative: Glute Kickbacks
+			○Calf Raises: 3 sets × 50 repetitions
+			■Alternative: Seated Heel Lifts
+			○Leg Raises: 2 sets × 25 raises
+			■Alternative: Seated Leg Lifts
+			
+			Cardio Workout
+			○30-min selected cardio exercise at moderate pace: RPE Scale: 4–6
+			  Strength Workout
+			○Goblet Squats with Dumbbell: 3 sets × 10–12 repetitions
+			■Alternative: Air Squats
+			○Hamstring Curls: 3 sets × 10–12 repetitions
+			■Alternative: Glute Kickbacks
+			○Calf Raises: 3 sets × 50 repetitions
+			■Alternative: Seated Heel Lifts
+			○Leg Raises: 2 sets × 25 raises
+			■Alternative: Seated Leg Lifts
+			
+			Goblet Squats with Dumbbell: 3 sets × 10–12 repetitions
+			■Alternative: Air Squats
+			Alternative: Air SquatsHamstring Curls: 3 sets × 10–12 repetitions
+			■Alternative: Glute Kickbacks
+			Alternative: Glute KickbacksCalf Raises: 3 sets × 50 repetitions
+			■Alternative: Seated Heel Lifts
+			Alternative: Seated Heel LiftsLeg Raises: 2 sets × 25 raises
+			■Alternative: Seated Leg Lifts
+			Alternative: Seated Leg LiftsCardio Workout
+			○30-min selected cardio exercise at moderate pace: RPE Scale: 4–630-min selected cardio exercise at moderate pace: RPE Scale: 4–6Day 6:
+		  Strength Workout
+			○Incline Dumbbell Bench Press: 3 sets × 10–12 repetitions■Alternative: Push-Ups on Knees (only go down as far as your mobility will allow)
+		  ○Diamond Push-Ups: 3 sets × 10–12 repetitions■Alternative: Close-Grip Wall Push-Ups
+		  ○Dumbbell Chest Flies (go light and pretend to “hug a tree”): 3 sets × 10–12 repetitions■Alternative: Wide-Grip Wall Push-Ups
+		  ○Jack Knives: 2 sets × 20 repetitions■Alternative: Standard Crunches
+		  
+		  Cardio Workout
+			○30-min selected cardio exercise at high-intensity intervals: These intervals will be comprised of 1 min easy (RPE Scale: 2–3) and 1 min hard (RPE Scale: 7–8) and will be repeated until time is up.
+          
+			  Strength Workout
+			○Incline Dumbbell Bench Press: 3 sets × 10–12 repetitions■Alternative: Push-Ups on Knees (only go down as far as your mobility will allow)
+		  ○Diamond Push-Ups: 3 sets × 10–12 repetitions■Alternative: Close-Grip Wall Push-Ups
+		  ○Dumbbell Chest Flies (go light and pretend to “hug a tree”): 3 sets × 10–12 repetitions■Alternative: Wide-Grip Wall Push-Ups
+		  ○Jack Knives: 2 sets × 20 repetitions■Alternative: Standard Crunches
+		  
+		  Incline Dumbbell Bench Press: 3 sets × 10–12 repetitions■Alternative: Push-Ups on Knees (only go down as far as your mobility will allow)
+		  Alternative: Push-Ups on Knees (only go down as far as your mobility will allow)Diamond Push-Ups: 3 sets × 10–12 repetitions■Alternative: Close-Grip Wall Push-Ups
+		  Alternative: Close-Grip Wall Push-UpsDumbbell Chest Flies (go light and pretend to “hug a tree”): 3 sets × 10–12 repetitions■Alternative: Wide-Grip Wall Push-Ups
+		  Alternative: Wide-Grip Wall Push-UpsJack Knives: 2 sets × 20 repetitions■Alternative: Standard Crunches
+		  Alternative: Standard CrunchesCardio Workout
+			○30-min selected cardio exercise at high-intensity intervals: These intervals will be comprised of 1 min easy (RPE Scale: 2–3) and 1 min hard (RPE Scale: 7–8) and will be repeated until time is up.
+          30-min selected cardio exercise at high-intensity intervals: These intervals will be comprised of 1 min easy (RPE Scale: 2–3) and 1 min hard (RPE Scale: 7–8) and will be repeated until time is up.Day 7:
+		  Strength Workout○Upright Rows: 3 sets × 10–12 repetitions■Alternative: Forward Arm Raises
+			○One-Arm Rows (pretend you are “starting a lawnmower”): 3 sets × 10–12 repetitions
+			■Alternative: Single Arm Circles
+			○Lat Pulldowns: 3 sets × 10–12 repetitions■Alternative: Seated Reaching Exercise (count every other side)
+			○Seated Cable Row: 2 sets × 10–12 repetitions■Alternative: Single Arm Circles
+			○Sit-ups: 2 sets × 25 repetitions■Alternative: Standard Crunches
+			
+          Cardio Workout○30-min selected cardio exercise at moderate pace: RPE Scale: 4–6
+			
+			
+		  Strength Workout○Upright Rows: 3 sets × 10–12 repetitions■Alternative: Forward Arm Raises
+			○One-Arm Rows (pretend you are “starting a lawnmower”): 3 sets × 10–12 repetitions
+			■Alternative: Single Arm Circles
+			○Lat Pulldowns: 3 sets × 10–12 repetitions■Alternative: Seated Reaching Exercise (count every other side)
+			○Seated Cable Row: 2 sets × 10–12 repetitions■Alternative: Single Arm Circles
+			○Sit-ups: 2 sets × 25 repetitions■Alternative: Standard Crunches
+			
+          Upright Rows: 3 sets × 10–12 repetitions■Alternative: Forward Arm Raises
+			Alternative: Forward Arm RaisesOne-Arm Rows (pretend you are “starting a lawnmower”): 3 sets × 10–12 repetitions
+			■Alternative: Single Arm Circles
+			Alternative: Single Arm CirclesLat Pulldowns: 3 sets × 10–12 repetitions■Alternative: Seated Reaching Exercise (count every other side)
+			Alternative: Seated Reaching Exercise (count every other side)Seated Cable Row: 2 sets × 10–12 repetitions■Alternative: Single Arm Circles
+			Alternative: Single Arm CirclesSit-ups: 2 sets × 25 repetitions■Alternative: Standard Crunches
+			Alternative: Standard CrunchesCardio Workout○30-min selected cardio exercise at moderate pace: RPE Scale: 4–6
+			
+			30-min selected cardio exercise at moderate pace: RPE Scale: 4–6
+			Day 8:Cardio Workout ONLY○60-min selected cardio at light/moderate pace: RPE Scale: 3–6
+			
+		  Cardio Workout ONLY○60-min selected cardio at light/moderate pace: RPE Scale: 3–6
+			60-min selected cardio at light/moderate pace: RPE Scale: 3–6
+			
+            Level 2. Exercise Made Easy: Intermediate Programming
+          
+            Month 1: Endurance Phase
+          For each strength exercise, the endurance phase of this program will include higher numbers of repetitions (10–12 repetitions for each set) for each exercise with lower weight. You should rest for 60 s to 1.5 min between each strength exercise. For each cardio workout, feel free to select your favorite cardio exercise. Some cardio exercise options include but are not limited to walking, jogging, running, biking, swimming, stairclimbing, rowing, elliptical exercise, etc.Day 1:
+		  Strength Workout
+			○Dumbbell Lunges (count every other leg): 3 sets × 10–12 repetitions■Alternative: Walking Lunges without Dumbbells (still count every other leg)
+			○Hamstring Curls: 3 sets × 10–12 repetitions■Alternative: Glute Kickbacks
+			○Calf Raises: 3 sets × 50 repetitions■Alternative: Seated Heel Lifts
+			○Bicycle Crunches: 2 sets × 50 revolutions■Alternative: Standard Crunches
+			
+		  Cardio Workout○30-min selected cardio exercise at moderate pace: RPE Scale: 4–6
+			
+		  Strength Workout
+			○Dumbbell Lunges (count every other leg): 3 sets × 10–12 repetitions■Alternative: Walking Lunges without Dumbbells (still count every other leg)
+			○Hamstring Curls: 3 sets × 10–12 repetitions■Alternative: Glute Kickbacks
+			○Calf Raises: 3 sets × 50 repetitions■Alternative: Seated Heel Lifts
+			○Bicycle Crunches: 2 sets × 50 revolutions■Alternative: Standard Crunches
+			
+		  Dumbbell Lunges (count every other leg): 3 sets × 10–12 repetitions■Alternative: Walking Lunges without Dumbbells (still count every other leg)
+			Alternative: Walking Lunges without Dumbbells (still count every other leg)Hamstring Curls: 3 sets × 10–12 repetitions■Alternative: Glute Kickbacks
+			Alternative: Glute KickbacksCalf Raises: 3 sets × 50 repetitions■Alternative: Seated Heel Lifts
+			Alternative: Seated Heel LiftsBicycle Crunches: 2 sets × 50 revolutions■Alternative: Standard Crunches
+			Alternative: Standard CrunchesCardio Workout○30-min selected cardio exercise at moderate pace: RPE Scale: 4–6
+			30-min selected cardio exercise at moderate pace: RPE Scale: 4–6Day 2:
+		  Cardio Workout Only
+			○30-min selected cardio at light/moderate pace: RPE Scale: 3–6
+			
+		  Cardio Workout Only
+			○30-min selected cardio at light/moderate pace: RPE Scale: 3–6
+			30-min selected cardio at light/moderate pace: RPE Scale: 3–6Day 3:
+		  Strength Workout
+			○Dumbbell Bench Press: 3 sets × 10–12 repetitions
+			■Alternative: Push-Ups on Knees or on Toes (only go as far down as your mobility will allow)
+			○Tricep Kickbacks: 3 sets × 10–12 repetitions
+			■Alternative: Tricep Dips on Chair (only go as far down as your mobility will allow)
+			○Dumbbell Chest Flies (go light and pretend to “hug a tree”): 3 sets × 10–12 repetitions
+			■Alternative: Wide-Grip Wall Push-Ups
+			○Bicycle Crunches: 2 sets × 50 revolutions
+			■Alternative: Sit-Ups
+			
+			Cardio Workout
+			○30-min selected cardio exercise at high-intensity intervals: These intervals will be comprised of 1 min easy (RPE Scale: 2–3) and 1 min hard (RPE Scale: 7–8) and will be repeated until time is up.
+			
+		  Strength Workout
+			○Dumbbell Bench Press: 3 sets × 10–12 repetitions
+			■Alternative: Push-Ups on Knees or on Toes (only go as far down as your mobility will allow)
+			○Tricep Kickbacks: 3 sets × 10–12 repetitions
+			■Alternative: Tricep Dips on Chair (only go as far down as your mobility will allow)
+			○Dumbbell Chest Flies (go light and pretend to “hug a tree”): 3 sets × 10–12 repetitions
+			■Alternative: Wide-Grip Wall Push-Ups
+			○Bicycle Crunches: 2 sets × 50 revolutions
+			■Alternative: Sit-Ups
+			
+			Dumbbell Bench Press: 3 sets × 10–12 repetitions
+			■Alternative: Push-Ups on Knees or on Toes (only go as far down as your mobility will allow)
+			Alternative: Push-Ups on Knees or on Toes (only go as far down as your mobility will allow)Tricep Kickbacks: 3 sets × 10–12 repetitions
+			■Alternative: Tricep Dips on Chair (only go as far down as your mobility will allow)
+			Alternative: Tricep Dips on Chair (only go as far down as your mobility will allow)Dumbbell Chest Flies (go light and pretend to “hug a tree”): 3 sets × 10–12 repetitions
+			■Alternative: Wide-Grip Wall Push-Ups
+			Alternative: Wide-Grip Wall Push-UpsBicycle Crunches: 2 sets × 50 revolutions
+			■Alternative: Sit-Ups
+			Alternative: Sit-UpsCardio Workout
+			○30-min selected cardio exercise at high-intensity intervals: These intervals will be comprised of 1 min easy (RPE Scale: 2–3) and 1 min hard (RPE Scale: 7–8) and will be repeated until time is up.
+			30-min selected cardio exercise at high-intensity intervals: These intervals will be comprised of 1 min easy (RPE Scale: 2–3) and 1 min hard (RPE Scale: 7–8) and will be repeated until time is up.Day 4:Cardio Workout Only
+			○30-min selected cardio at light/moderate pace: RPE Scale: 3–6
+			
+		  Cardio Workout Only
+			○30-min selected cardio at light/moderate pace: RPE Scale: 3–6
+			30-min selected cardio at light/moderate pace: RPE Scale: 3–6Day 5:
+		  Strength Workout
+			○Upright Rows: 3 sets × 10–12 repetitions
+			■Alternative: Forward Arm Raises
+			○One-Arm Rows (pretend you are “starting a lawnmower”): 3 sets × 10–12 repetitions
+			■Alternative: Single Arm Circles
+			  ○Lat Pulldowns: 3 sets × 10–12 repetitions
+			■Alternative: Seated Reaching Exercise (count every other side)
+			  ○Dumbbell Shrugs: 2 sets × 10–12 repetitions
+			■Alternative: Shoulders to Ears
+			  ○Sit-ups: 2 sets × 25 repetitions
+			■Alternative: Bicycle Crunches or Standard Crunches
+			  
+			Cardio Workout
+			○30-min selected cardio exercise at moderate pace: RPE Scale: 4–6
+			 
+		  Strength Workout
+			○Upright Rows: 3 sets × 10–12 repetitions
+			■Alternative: Forward Arm Raises
+			○One-Arm Rows (pretend you are “starting a lawnmower”): 3 sets × 10–12 repetitions
+			■Alternative: Single Arm Circles
+			  ○Lat Pulldowns: 3 sets × 10–12 repetitions
+			■Alternative: Seated Reaching Exercise (count every other side)
+			  ○Dumbbell Shrugs: 2 sets × 10–12 repetitions
+			■Alternative: Shoulders to Ears
+			  ○Sit-ups: 2 sets × 25 repetitions
+			■Alternative: Bicycle Crunches or Standard Crunches
+			  
+			Upright Rows: 3 sets × 10–12 repetitions
+			■Alternative: Forward Arm Raises
+			Alternative: Forward Arm RaisesOne-Arm Rows (pretend you are “starting a lawnmower”): 3 sets × 10–12 repetitions
+			■Alternative: Single Arm Circles
+			  Alternative: Single Arm CirclesLat Pulldowns: 3 sets × 10–12 repetitions
+			■Alternative: Seated Reaching Exercise (count every other side)
+			  Alternative: Seated Reaching Exercise (count every other side)Dumbbell Shrugs: 2 sets × 10–12 repetitions
+			■Alternative: Shoulders to Ears
+			  Alternative: Shoulders to EarsSit-ups: 2 sets × 25 repetitions
+			■Alternative: Bicycle Crunches or Standard Crunches
+			  Alternative: Bicycle Crunches or Standard CrunchesCardio Workout
+			○30-min selected cardio exercise at moderate pace: RPE Scale: 4–6
+			30-min selected cardio exercise at moderate pace: RPE Scale: 4–6Day 6:
+		  Strength Workout
+			○Goblet Squats with Dumbbell: 3 sets × 10–12 repetitions
+			■Alternative: Air Squats
+			○Hamstring Curls: 3 sets × 10–12 repetitions
+			■Alternative: Glute Kickbacks
+			  ○Split Jumps: 3 sets × 16 repetitions
+			■Alternative: Walking Lunges without Dumbbells (count every other leg)
+			  ○Leg Raises: 2 sets × 25 raises
+			■Alternative: Standing Leg Raises (count every other leg)
+			  
+			Cardio Workout
+			○30-min selected cardio exercise at moderate pace: RPE Scale: 4–6
+			
+		  Strength Workout
+			○Goblet Squats with Dumbbell: 3 sets × 10–12 repetitions
+			■Alternative: Air Squats
+			○Hamstring Curls: 3 sets × 10–12 repetitions
+			■Alternative: Glute Kickbacks
+			  ○Split Jumps: 3 sets × 16 repetitions
+			■Alternative: Walking Lunges without Dumbbells (count every other leg)
+			  ○Leg Raises: 2 sets × 25 raises
+			■Alternative: Standing Leg Raises (count every other leg)
+			  
+			Goblet Squats with Dumbbell: 3 sets × 10–12 repetitions
+			■Alternative: Air Squats
+			Alternative: Air SquatsHamstring Curls: 3 sets × 10–12 repetitions
+			■Alternative: Glute Kickbacks
+			  Alternative: Glute KickbacksSplit Jumps: 3 sets × 16 repetitions
+			■Alternative: Walking Lunges without Dumbbells (count every other leg)
+			  Alternative: Walking Lunges without Dumbbells (count every other leg)Leg Raises: 2 sets × 25 raises
+			■Alternative: Standing Leg Raises (count every other leg)
+			  Alternative: Standing Leg Raises (count every other leg)Cardio Workout
+			○30-min selected cardio exercise at moderate pace: RPE Scale: 4–6
+			30-min selected cardio exercise at moderate pace: RPE Scale: 4–6Day 7:Cardio Workout Only○30-min selected cardio at light/moderate pace: RPE Scale: 3–6
+			
+		  Cardio Workout Only○30-min selected cardio at light/moderate pace: RPE Scale: 3–6
+			30-min selected cardio at light/moderate pace: RPE Scale: 3–6Day 8:Strength Workout
+			○Incline Dumbbell Bench Press: 3 sets × 10–12 repetitions
+			■Alternative: Push-Ups on Knees or Toes (only go as far down as your mobility will allow)
+			○Diamond Push-Ups: 3 sets × 10–12 repetitions
+			■Alternative: Close-Grip Wall Push-Ups
+			  ○Dumbbell Chest Flies (go light and pretend to “hug a tree”): 3 sets × 10–12 repetitions
+			■Alternative: Wide-Grip Wall Push-Ups
+			  ○Jack Knives: 2 sets × 20 repetitions
+			■Alternative: Sit-Ups or Standard Crunches
+			  
+			Cardio Workout○30-min selected cardio exercise at high-intensity intervals: These intervals will be comprised of 1 min easy (RPE Scale: 2–3) and 1 min hard (RPE Scale: 7–8) and will be repeated until time is up.
+			
+		  Strength Workout
+			○Incline Dumbbell Bench Press: 3 sets × 10–12 repetitions
+			■Alternative: Push-Ups on Knees or Toes (only go as far down as your mobility will allow)
+			○Diamond Push-Ups: 3 sets × 10–12 repetitions
+			■Alternative: Close-Grip Wall Push-Ups
+			  ○Dumbbell Chest Flies (go light and pretend to “hug a tree”): 3 sets × 10–12 repetitions
+			■Alternative: Wide-Grip Wall Push-Ups
+			  ○Jack Knives: 2 sets × 20 repetitions
+			■Alternative: Sit-Ups or Standard Crunches
+			  
+			Incline Dumbbell Bench Press: 3 sets × 10–12 repetitions
+			■Alternative: Push-Ups on Knees or Toes (only go as far down as your mobility will allow)
+			Alternative: Push-Ups on Knees or Toes (only go as far down as your mobility will allow)Diamond Push-Ups: 3 sets × 10–12 repetitions
+			■Alternative: Close-Grip Wall Push-Ups
+			  Alternative: Close-Grip Wall Push-UpsDumbbell Chest Flies (go light and pretend to “hug a tree”): 3 sets × 10–12 repetitions
+			■Alternative: Wide-Grip Wall Push-Ups
+			  Alternative: Wide-Grip Wall Push-UpsJack Knives: 2 sets × 20 repetitions
+			■Alternative: Sit-Ups or Standard Crunches
+			  Alternative: Sit-Ups or Standard CrunchesCardio Workout○30-min selected cardio exercise at high-intensity intervals: These intervals will be comprised of 1 min easy (RPE Scale: 2–3) and 1 min hard (RPE Scale: 7–8) and will be repeated until time is up.
+			30-min selected cardio exercise at high-intensity intervals: These intervals will be comprised of 1 min easy (RPE Scale: 2–3) and 1 min hard (RPE Scale: 7–8) and will be repeated until time is up.Day 9:
+		  Cardio Workout Only○30-min selected cardio at light/moderate pace: RPE Scale: 3–6
+			
+		  Cardio Workout Only○30-min selected cardio at light/moderate pace: RPE Scale: 3–6
+			30-min selected cardio at light/moderate pace: RPE Scale: 3–6Day 10:
+		  Strength Workout
+			○Upright Rows: 3 sets × 10–12 repetitions
+			■Alternative: Forward Arm Raises
+			○Back Extension: 3 sets × 10–12 repetitions■Alternative: Good Morning Exercise (keep back straight through range of motion)
+			  ○Lat Pulldowns: 3 sets × 10–12 repetitions■Alternative: Seated Reaching Exercise
+			  ○Seated Cable Row: 2 sets × 10–12 repetitions■Alternative: Single Arm Circles
+			  ○Sit-ups: 2 sets × 25 repetitions■Alternative: Bicycle Crunches or Standard Crunches
+			  
+			Cardio Workout
+			○30-min selected cardio exercise at moderate pace: RPE Scale: 4–6
+			
+		  Strength Workout
+			○Upright Rows: 3 sets × 10–12 repetitions
+			■Alternative: Forward Arm Raises
+			○Back Extension: 3 sets × 10–12 repetitions■Alternative: Good Morning Exercise (keep back straight through range of motion)
+			  ○Lat Pulldowns: 3 sets × 10–12 repetitions■Alternative: Seated Reaching Exercise
+			  ○Seated Cable Row: 2 sets × 10–12 repetitions■Alternative: Single Arm Circles
+			  ○Sit-ups: 2 sets × 25 repetitions■Alternative: Bicycle Crunches or Standard Crunches
+			  
+			Upright Rows: 3 sets × 10–12 repetitions
+			■Alternative: Forward Arm Raises
+			Alternative: Forward Arm RaisesBack Extension: 3 sets × 10–12 repetitions■Alternative: Good Morning Exercise (keep back straight through range of motion)
+			  Alternative: Good Morning Exercise (keep back straight through range of motion)Lat Pulldowns: 3 sets × 10–12 repetitions■Alternative: Seated Reaching Exercise
+			  Alternative: Seated Reaching ExerciseSeated Cable Row: 2 sets × 10–12 repetitions■Alternative: Single Arm Circles
+			  Alternative: Single Arm CirclesSit-ups: 2 sets × 25 repetitions■Alternative: Bicycle Crunches or Standard Crunches
+			  Alternative: Bicycle Crunches or Standard CrunchesCardio Workout
+			○30-min selected cardio exercise at moderate pace: RPE Scale: 4–6
+			30-min selected cardio exercise at moderate pace: RPE Scale: 4–6Comparisons between baseline and 12 months across all outcomes of interest. Note: AB = agility and balance; AE = aerobic endurance; BG = blood glucose; CA15-3 = cancer antigen 15-3; CEA = carcinoembryonic antigen; HDL = high-density lipoprotein; LBFL = lower-body flexibility left; LBFR = lower-body flexibility right; LDL = low-density lipoprotein; LBS = lower-body strength; UBFL = upper-body flexibility left; UBFR = upper-body flexibility right; TC = total cholesterol; TG = triglyceride; UBSL = upper-body strength left; UBSR = upper-body strength right.Categories of exercise prescription.Description of senior fitness test.Participant baseline characteristics.Note: a Data are presented as mean ± standard deviation; b data are presented as frequency; no statistically significant difference in baseline characteristics between nonadherence and adherence.Inferential statistics for baseline and posttest on health outcomes.Note: All values are presented as mean ± standard deviation; estimates were determined from unadjusted mean differences in group change from baseline to 12 months; 95% CI = 95% confidence interval; * statistical significance within-group changes between baseline and 12 months.

Med-MDPI/jcm_2/jcm-09-06-01776.txt

@@ -0,0 +1 @@
+Background and Study Aim: In terms of therapeutic management, gastrointestinal stromal tumors (GISTs) seem to be the most difficult group of subepithelial gastrointestinal lesions (SELs). Despite various treatment option, choice of optimal management remains a dilemma in daily practice. Our aim was to evaluate a new hybrid resection technique of gastric GISTs type III as a modality of endoscopic full-thickness resection. Methods: Three males and one female (mean age of 68) were qualified for the procedure. Endoscopic full-thickness resections consisted of the endoscopic resection combined with suturing by Apollo OverStitch System. The main inclusion criterium was a complete diagnosis of GISTs (computed tomography (CT), endoscopic ultrasound (EUS), fine-needle biopsy (FNB)) with the evaluation of the tumor features, especially, the location in the gastric wall. All of the tumors were type III with a diameter between 20–40 mm. The lesions were located in the corpus (1), antrum (1) and between gastric body and fundus (2). All procedures were performed in 2019. Results: The technical and therapeutic success rate was 100% and the mean resection time 107.5 min. Neither intra- nor postprocedural complications were observed. In all four cases, R0 resection was achieved. Histopathologic assessment confirmed GIST with <5mitose/50HPF in all of the tumors, with very low risk. Conclusion: Based on our outcomes, endoscopic resection combined with the sewing by Apollo OverStitch of gastric GISTs type III, with the diameter between 20–40 mm, seems to be an effective therapeutic option with a good safety profile, however further studies with a larger treatment group are needed.In terms of therapeutic management, gastrointestinal stromal tumors (GISTs) seem to be the most challenging group of subepithelial gastrointestinal lesions (SELs). Despite various treatment strategies considering general clinical assessment and the type of tumor according to Miettinen classification, chose of optimal management in daily practice remains a dilemma for some group of tumors. After the confirmation of malignancy potential, regular follow-up for most GISTs with the diameter below 10 mm is widely accepted. Nevertheless, in case of GISTs with the diameter > 10 mm, R0 resection should be considered [1,2,3,4,5,6,7].Endoscopic ultrasound (EUS) with immunohistochemical and pathologic assessment is a crucial method characterized by high sensitivity and specificity for the differential diagnosis of GISTs [8,9,10]. However, recent studies show that in 40.6% performed biopsies (19 and 22 G needles) obtained material was sufficient for evaluation the mitotic count on 50 HPF [11,12,13,14]. Considering the mitotic index as a malignancy predictor and its clinical implication, a complete resection of the tumor allows for the estimation of malignant potential and healing.Presently, the first-line treatment for local GISTs is surgical resection [15,16,17]. In recent years, along with the development of advanced endoscopic techniques, new therapeutic pathways have been introduced for tumors with a diameter of < 40 mm. Recent studies have demonstrated feasibility and safety of endoscopic resection for tumors with the diameter below 50 mm [18]. Endoscopic management may be considered after complete diagnosis claimed by imaging examinations, pathologic result and after exclusion of high-risk factors [19]. Among the endoscopic resection modalities of GISTs located in the stomach were found endoscopic band ligation (EBL), endoscopic submucosal excavation (ESE), endoscopic submucosal dissection (ESD) and various combination of endoscopic full-thickness resection (EFTR) [20]. Submucosal tunneling endoscopic resection (STER) is the treatment option, though reserved for gastric GISTs localized mainly in the cardia. The selection of endoscopic techniques should consider the connection with muscularis propria (MP), size of the tumor and relation to individual layers of the stomach wall. The classification presented by Kim et al. [7] concerning the endoscopic/surgical approach depends on the connection with muscularis propria, may be useful in a choice of resection technique (Figure 1).According to Kim et al. [7] classification, type I has a narrow connection with the MP layer, therefore the best therapeutic option for gastric GISTs are: EBL, ESE and ESD. Type II has a wider connection with muscularis propria and may be resected the same technique as type I. In the case of type III and IV, the achievement of complete resection by endoscopic methods is nearly impossible. Therefore, as well as EFTR techniques and surgical treatment should be considered for both of these types [7].In current study, we present a results of combined endoscopic resection technique for type III of gastric GISTs.The selection of patients was based on the evaluation of particular clinical features and optimal care of patients after multidisciplinary committee qualification. All procedures were performed in 2019 at two Endoscopic Units (Hospital of the Ministry of Interior and Administration Szczecin, Poland and Specialist Hospital of Alfred Sokolowski, Wałbrzych, Poland). Every patient signed a written consent and was thoroughly informed about the course of the treatment. All the patients included into the research fulfilled the following criteria:(1)disqualification from surgical treatment due to concomitant diseases or lack of consent for surgical treatment,(2)gastric GIST with a diameter of >20 and <40 mm, evaluated with endoscopic ultrasound and confirmed by EUS-guided biopsy.(3)confirmed the local disease, without metastases or infiltration of local tissues (CECT, EUS).disqualification from surgical treatment due to concomitant diseases or lack of consent for surgical treatment,gastric GIST with a diameter of >20 and <40 mm, evaluated with endoscopic ultrasound and confirmed by EUS-guided biopsy.confirmed the local disease, without metastases or infiltration of local tissues (CECT, EUS).The procedures were performed in a total of four patients (3 males and 1 female) at the mean age of 68 (62–77). In summary, three of the patients did not agree with surgical management. One patient was disqualified due to thrombocytopenia related to myelodysplastic syndrome. All patients underwent endosonographic evaluation of the tumor and surrounding tissues. Among the significant features of the were assessed: size and morphology of the tumor, location in the gastrointestinal wall and connection with particular layers (Figure 2). All EUS examination were performed by A.W-K. with the Pentax linear echoendoscope (processor Hi Vision Preirus, Hitachi Aloka Medical, Wallingford, CT). The tumors were located in the middle of the gastric body (1), in the antrum (1) and between the body and fundus of the stomach (2), with a mean size of 28.75 mm (20.0–40.0 mm). The deep connection with the muscularis propria was confirmed and the tumors were located in the middle of the gastric wall (type 3 according to Kim et al. classification). EUS-guided fine-needle biopsy (FNB; type of the needle—Expect™ 19 Flex; Boston Scientific, Natick, MA, USA) was performed for the pathologic conformation and assessment of immunohistochemistry. Selected markers like: SMA (-), DOG-1 (+), CD 117 (+), CD 34 (+) and Vimentin (+) were determined. Relevant data of the patients and lesions were described in Table 1.All procedures were performed under general anesthesia in a supine position. For the standard endoscopic resection as a first step of the procedure, endoscope GIF-HQ 190 was used. In turn, for the suturing, endoscope GIF 2TH180 (Olympus America, Center Valley, Penn) was applied. All resections were performed by A.R. The procedure began with the injection of indigo carmine solution underneath the mucosal layer and circular incision of the mucosa around the tumor using the Dual Knife (Olympus, Tokyo, Japan) (Figure 3A–B). The next step was the dissection of submucosa. Then, in order to prevent perforation at the cutting site, Apollo OverStitch FM (Austin TX) was used to duplicate the gastric wall below the tumor (Figure 3C–D; Figure 5A–B). Duplication consists of doubling the whole gastric wall just underneath the tumor, which is possible by making them close together, through the sutures (2.0 continuous suture). This leads to the tumor elevation on duplicated folds. The duplication using Apollo OverStitch is starting at the site of previously cut submucosa. After the first duplication of MP, further cutting the muscularis propria above the sutured doubled gastric wall up was performed (Figure 4A–D). When the site below the tumor was completely duplicated, the tumor with muscularis propria and serosa was resected (Figure 5C–E). The whole tumor was removed from the stomach using a standard endoscope.All procedures were elective, and the main steps of the resection were alike in all cases. Mean resection time was 107.5 min, however in case of the tumor with the diameter approximately 40 mm, the procedure was extended by about 30 min. There were no adverse events during or after the procedure (adverse events rate 0%). The length of hospital stay was 3 days. The post-procedural observation period was uneventful. Prophylactic antibiotics were administered before all of procedures. In addition, PPI were used for 6 weeks after the procedure. Technical success was 100% and complete resection was confirmed in post-resection material trough pathologic assessment. Resection margin was R0 with a therapeutic success of 100%. All tumors were confirmed to be GIST with <5 mitoses/50 HPF. Based on GEIS guidelines, risk of progression was very-low, therefore adjuvant therapy and follow-up were not required [21]. (Table 2, Figure 6)The safety and effectiveness of a hybrid endoscopic technique for gastric GISTs type III, with a diameter between 20 and 40 mm, were evaluated in this study. The optimal treatment of GISTs depends on various factors like the tumor size, presence of metastases and stage of the disease. The additional factor determining further management is the malignancy potential. Despite the malignancy risk estimation supported by pathologic assessment, the resection of tumors significantly increases the chance for cure. Therefore, resection of GISTs is recommended as diagnostic and therapeutic approach [22]. European Society for Medical Oncology (ESMO) and Japanese Society of Clinical Oncology recommend surgical resection GISTs smaller than 20 mm [15,16]. Based on ESMO guidelines, the standard approach to GISTs ≧ 20 mm should consider surgical treatment as a first option. In terms of qualification for complementary management, the tumor location should be taken into account. The presence of the tumor with gastric location deprived metastases is related to a very low risk of progression without indication for adjuvant therapy and follow-up [15].In our group, all patients were qualified for endoscopic treatment after multidisciplinary consensus. As mentioned before, endoscopic resection methods of gastric GISTs include EBL, ESE, ESD and EFTR. However, considering features of the tumor, techniques included EBL, ESE and ESD were not recommended due to possible incomplete resection. Therefore, EFTR as a resection technique of the tumor with a safe margin, was chosen. This method was firstly presented by Suzuki et al. [23] and further studies have confirmed the effectiveness of this resection modality for tumors infiltrating the MP. In two studies, a total of 59 gastric GISTs were endoscopically removed using EFTR without laparoscopic assistance [24,25]. Technically, for the procedure, an endoscopic snare, clips and an endoloop were used. Moreover, the key to the method was performing a controlled perforation [26]. Despite the high effectiveness with the technical success of 100% and a low risk of other complications, this technique seems advanced and time-consuming. In turn, gFTRD (gastric FTRD; Ovesco Endoscopy, Tübingen, Germany) may be other option in terms of cost and availability. Meier et al. [27] resected six gastric GISTs with a diameter of under 20 mm. Complete R0 resection was achieved in three tumors, with a total technical success 89.7% (26/29 of submucosal lesions). This method seems to be less technically demanding than previously described. However, gFTRD is dedicated to diagnostic or therapeutic full-thickness-resection in the case of GISTs < 20 mm. Huang et al. performed endoscopic full-thickness resection of 13 gastric GISTs with the closure of perforations with metal clips followed by an abdominal paracentesis to decrease the abdominal cavity pressure [20]. For more extensive perforation, the retina mending method is recommended [26]. Moreover, postoperative gastrointestinal decompression, proton pump inhibitors and antibiotics should be administered to prevent postoperative infection [26]. Currently presented variants of gEFTR may not allow for curable resection. In particular, when the significant MP involvement, extraluminal protrusion or infiltration of the whole gastric wall is expected.Presented method in this manuscript, it consisted of two main steps. The performance of the incision around the tumor, cutting submucosa and closing neighboring gastric walls in the near of the tumor leading to lift up the lesion and provide the full-thickness wall resection with the pathologic lesion. The application of Apollo OverStitch TM set was expanded beyond bariatric indications, such as suturing after endoscopic resection [28]. However, from available tools, Apollo OverStitch TM set allows for bringing neighboring gastric walls together and resect tumors without the necessity of the perforation performance.There are couple of limitations of this method. Firstly, current procedure may be related with higher costs and the required technical skills of the endoscopic operator. However, the technical success rate was 100% with no complications during or after the procedure. The limitation of complications occurrence may be related to the fact that in this method the perforation is not performed during the resection due to using a suturing system before the knife cutting. Moreover, hospital stay duration is shorter and general costs are lower than after surgical treatment. It is important to note that therapeutic success rates are comparable to standard surgical treatment, especially in the case of low-risk tumors, when lymph nodes resection is not indicated. In terms of procedure duration, our hybrid approach could be comparable with ESD. However, ESD can only be used for GISTs originating from the superficial MP layer. In addition, major complications such as perforation (8.2%) and bleeding (15.6%) and less common such aspiration pneumonia, venous thromboembolism and air embolism should be taken into account [19]. In turn, the duration time of the presented hybrid technique was lower than surgical procedures, both laparoscopic and classical (107.5 min vs. 147.8 ± 59.3 vs. 139.2 ± 62.1, respectively) [29].Additional, for standardization of this method an increment of case volume is necessary. However, it seems to be an alternative method for patients with gastric GISTs type III and size between 20–40 mm, disqualified from surgical treatment.All confirmed GISTs require removal either by standard surgery or alternative endoscopic procedures. There is currently no optimal method to remove gastric GISTs with tumor size of 20 to 40 mm, type III. In the presented hybrid approach patient benefits outweigh its limitations. Due to a small patient population, further studies are necessary to compare treatment modality with other techniques.Conceptualization—K.M.P., A.W.-K., A.R.; methodology—K.M.P., A.R., A.W.-K.; validation—K.M.P. Formal Analysis—K.M.P., A.W.-K., A.R; investigation—A.W.-K., A.R.; data curation—A.R., K.M.P/Writing—original draft preparation, All authors; writing—review & editing—All authors; visualization, K.M.P, A.W.-K., A.R. Supervision—A.R., A.W.-K.; project administration—K.M.P; All authors have read and agreed to the published version of the manuscript.This research received no external funding.The authors declare no conflict of interest.Adapted from Kim et al. Classification of GISTs (gastrointestinal stromal tumors) [7]. Type I—GIST with a very narrow connection with the proper muscle layer and protrudes into the luminal side, like a polyp; Type II—with a wider connection with the proper muscle layer and protrudes into the luminal side at an obtuse angle; Type III—is located in the middle of the gastric wall; Type IV—protrudes mainly into the serosal side of the gastric wall.Endoscopic ultrasound evaluation of gastric GISTs. Typical appearance of GIST in endoscopic ultrasound. All tumors of a mixed echogenicity were located in the middle of the gastric wall and have a deep connection with muscularis propria. The average size was 28.75 mm.First step of the procedure (case 4). (A,B) Marking the tumor and circular incision around the tumor with submucosal dissection; (C,D) then adjacent folds of the stomach wall with Apollo OverStitch were bringing together. Tumor was partly lifted up, which provided protection against the perforation during further resection.Step two. (A–D) Subsequent layers below the tumor were cut with the incision of muscularis propria.Step three. Step three. (A,B) After revealing the connection of the tumor with the deep layer of the muscularis propria, the adjacent gastric folds have been again brought together in another site (C–F) to lift up the resection site, allowed for safe and transmural cut off the tumor; (A,B) successive sutures were inserted (C–F) to enucleate it completely from the MP layer, thus the full-wall resection was performed.Immunohistochemical markers of the tumors. Membranous and cytoplasmic staining for (A) CD 117 (+), CD 34 (+), DOG-1 (+). (Case 1–4); (B) CD34 (+) strong positive (4×); (C) CD117 (+) mild positive (×4); (D) strong and diffuse expression of DOG1 (+), (D) membranous and cytoplasmic (2×), (E) 10×; (F) 40× H&E—mixed epithelioid/spindle cell morphology with subnuclear vacuoles, eosinophilic cytoplasm and abundant myxoid stroma with thin-walled blood vessels.Patient characteristics.Results of the treatment, complications, follow-up.Continuous variables were reported using mean ± standard deviation (SD). Categorical variables were reported using proportion (%).

Med-MDPI/jcm_2/jcm-09-06-01777.txt

@@ -0,0 +1 @@
+Denotes equal contribution.Mesenteric fibrosis (MF) constitutes an underrecognized sequela in patients with small intestinal neuroendocrine neoplasms (SI-NENs), often complicating the disease clinical course. The aim of the present systematic review, carried out by Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) methodology, is to provide an update in evolving aspects of MF pathogenesis and its clinical management in SI-NENs. Complex and dynamic interactions are present in the microenvironment of tumor deposits in the mesentery. Serotonin, as well as the signaling pathways of certain growth factors play a pivotal, yet not fully elucidated role in the pathogenesis of MF. Clinically, MF often results in significant morbidity by causing either acute complications, such as intestinal obstruction and/or acute ischemia or more chronic conditions involving abdominal pain, venous stasis, malabsorption and malnutrition. Surgical resection in patients with locoregional disease only or symptomatic distant stage disease, as well as palliative minimally invasive interventions in advanced inoperable cases seem clinically meaningful, whereas currently available systemic and/or targeted treatments do not unequivocally affect the development of MF in SI-NENs. Increased awareness and improved understanding of the molecular pathogenesis of MF in SI-NENs may provide better diagnostic and predictive tools for its timely recognition and intervention and also facilitates the development of agents targeting MF.Small intestinal neuroendocrine neoplasms (SI-NENs) comprise a rare tumor entity with rising incidence, but overall favorable survival outcomes despite frequent presentation at distant stage disease [1]. They originate in enterochromafin (EC) cells, usually in the terminal ileum and less commonly in the jejunum. The primary tumor is typically submucosal and < 1cm in size, whereas in up to 50% of patients multifocal tumors can be found arising independently in the small intestine [2,3]. Mesenteric lymph node (LN) metastases are commonly present at diagnosis and tend to grow conspicuously large and often induce mesenteric fibrosis (MF) in the surrounding tissue of the mesentery.The release of various amines, peptides and growth factors (GFs) may induce hormonal symptoms that constitute the carcinoid syndrome (CS), but also fibrosis in cellular systems leading to MF and carcinoid heart disease (CHD) [4,5]. In addition, neuroendocrine neoplasm (NEN) cells overexpress a plethora of proangiogenic molecules that promote both recruitment and proliferation of endothelial cell precursors becoming highly vascularized neoplasms [6].Locoregional LN metastases together with the accompanying MF may occur in up to 50% of SI-NENs with encasement of the superior mesenteric vessels at different levels in the mesentery [7,8]. Locoregional resective surgery when feasible constitutes the mainstay of treatment in patients with MF-related symptoms. Although advances in systemic and liver-targeted treatment of SI-NENs have been made, alleviating hormonal symptoms and prolonging progression-free (PFS) and overall survival (OS), effective treatment options targeting MF are not fully elucidated. As improved insight in the complex pathogenesis of MF is key to the development of new therapies, we have reviewed evolving aspects of tumor microenvironment (TME), mediators and signaling pathways involved in the molecular pathogenesis of MF in SI-NENs, in order to refine the management of patients suffering from fibrotic complications.We performed a systematic review of the literature available on PubMed, Cochrane Library, Embase, Web of Science and SCOPUS databases about MF in SI-NENs until March 15th, 2020, as specified in the “Search Strategy Supplement” (Table S1). Articles were independently evaluated by two of the authors (K.D. and K.I.A.) for the relevance to the planned scope of the review. Reference lists of key publications were also reviewed for eligibility. We followed Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines for reporting [9]. The literature search and the selection of included studies are presented in the PRISMA flow diagram (Figure 1).In SI-NENs, MF represents an abnormal repair process, mimicking the mechanism operating in certain types of connective tissue disorder, associated with serious complications during the course of the disease [10,11]. Serotonin and vasoactive substances were the first effectors to be implicated in the pathogenesis of CS and into its more severe complications, the MF or CHD, representing the fibrotic-related sequelae of CS. There is currently evidence that CHD pathophysiology is mostly based on circulating effectors, specifically on high serotonin levels as depicted by the high urine level of 5-hydroxyindoleacetic acid (5-HIAA), whereas MF seems to be caused by a complex network of autocrine and paracrine mediators produced by neoplastic cells and TME elements [11].TME represents an important functional unit comprising a cluster of endothelial and inflammatory cells and mesenchymal stroma elements with fibroblasts. This network of cells is interconnected and interacting, leading to tumor growth and development of fibrosis, possibly through epigenetic changes of tumor cells [6,12,13]. TME has been recently depicted as a dynamic milieu where an injury with fragmentation of extracellular matrix elements results in chemo-attractant molecules recruiting immune cells that promote a tumorigenic environment, followed by pro-angiogenic elements during tissue repair.Proteomic analysis of SI-NENs and associated MF primarily revealed differences in mesenteric stroma showing that certain proteins are involved in the formation and regulation of the extracellular matrix and the complement cascade [14]. In addition, a multitude of immune cells including B and T cells, natural killer (NK) cells, mast cells, dendritic cells and macrophages have been reported to infiltrate NENs, resulting in an immunosuppressed TME, which promotes tumor growth [6]. NEN patients display immune recognition of their tumors, as CD8+ T-cells target specific SI-NEN-associated antigens including chromogranin A and tryptophan hydroxylase [15]. Recently, several studies have addressed the expression of programmed death-ligand 1 (PD-L1) as a potential mechanism of NEN immune evasion, with the upregulation of PD-L1 exhibited in a subset of SI-NENs [16,17,18].Importantly, in vitro studies in currently available pancreatic NEN cell lines cannot serve as an ideal model for assessment of MF. However, the development of three-dimensional organoids/spheroids may be a promising model to study the SI-NEN stroma complexity and its influence in epithelial tumorigenesis in the context of MF. Recently, SI-NEN cells have been cultured from surgically removed tumors as spheroids in ECM, forming a three-dimensional matrix that encapsulates SI-NEN cells and mimics TEM [19]. Although, organoid/spheroid systems lack some important components present in vivo, such as fibroblasts, endothelial cells and immune cells, coculture of organoids with other cell types has been attempted in order to generate a more “physiological” TEM, and to study cell–cell interactions with potential future implications in the study of MF [20].The effort to understand TME elements sheds light on the paracrine role that GFs may exert, affecting the molecular pathways implicated in fibroblast growth, proliferation and action [12,21]. A number of mediators have been shown to play an essential role in MF pathogenesis (Table 1 and Figure 2).Serotonin (5-hydroxytryptamine; 5-HT) has been classically considered as the main mediator of MF in SI-NENs. However, no consistent association has been documented between MF and elevated serotonin levels [10,56]. Serotonin exhibits both mitogenic and fibrogenic effects in fibroblasts, smooth muscle cells and endothelial cells.The strongest evidence that implicates serotonin in the interplay of fibrosis is the finding of higher levels of 5-HIAA in patients with CHD compared to patients without cardiac involvement [57,58,59]. Serotonin administration induces CHD through a transforming growth factor beta (TGFβ)-mediated mechanism [60], and the presence of serotonin likely modulates connective tissue growth factor (CTGF) function. Moreover, serotonin acts in association with other molecules, potentiating the mitogenic activity of the β-fibroblast growth factor (FGF) and the combined action of epidermal growth factor (EGF) and insulin [26].Transcript levels and the secretion of TGβ1, CTGF and FGF2 were significantly reduced in SI-NEN treated with a specific 5-HT2B receptor antagonist in vitro [12]. Thus, 5-HT2B receptors and serotonin have an important role in the regulation of peri-tumoral fibrosis and angiogenesis, sustaining TME [12]. In addition, the MEK/extracellular signal-regulated kinase (ERK) pathway is potentially playing a role in this process, since serotonin acts via the G-protein coupled 5-HT receptors, which are activating mitogenic pathways [12].Another mechanism for the pathogenesis of MF may be the decreased expression of serotonin-degrading enzymes in the stromal compartment of mesenteric metastases [61]. Recently, the tryptophan metabolism pathway was investigated by label-free proteomics in MF. Quantification of h-5-HT and h-5-HIAA in blood and tissues was simple, highly specific and predictive of steady state changes in 5-HT, opening the possibility for rapid screening of TPH1 inhibitor dose response, such as that of telotristat etiprate [62].GFs act locally by a combination of paracrine and autocrine actions to stimulate cell proliferation and differentiation. Platelet-derived growth factor (PDGF), insulin growth factor 1 (IGF-1) and -2, EGF, TGF-α and TGF-β have shown mitogenic properties on fibroblasts [47].TGFβ can stimulate both the production and deposition of the extracellular matrix via increased expression of collagen, fibronectin, proteoglycan and integrin expression in parallel to decreased expression of proteases and increased expression of protease inhibitors [63]. TGFβ is also chemotactic for fibroblasts and macrophages and may induce a rapid local fibrotic response [28]. TGFβ1 and the receptor subtype-2(TGFβrII) have been identified in SI-NEN samples, confirming the interaction of tumor cell with TME to exert its pro-fibrotic role in NENs [31].CTGF is involved in the coordination of complex biologic processes such as differentiation, tissue repair and angiogenesis. The low-density lipoprotein (LDL) receptor-related protein/α-2-macroglobulin receptor (LRP2) has been demonstrated to be the receptor for CTGF in fibroblasts [64]. By signaling via integrins, a mechanistic interpretation is provided for the chemotactic and mitogenic properties of CTGF, as well as for its functions in ECM remodeling during angiogenesis and tissue repair [65]. CTGF is transcriptionally activated principally through TGFβ1 [40]. However, PDGF, EGF and FGF also activate CTGF gene expression at the transcriptional level [41,66]. Processes such as TGFβ-induced fibroblast proliferation, collagen synthesis and myofibroblast differentiation are regulated by CTGF-dependent pathways. Consequently, CTGF represents a downstream mediator of the profibrotic activities of TGFβ1 acting on fibroblasts, being a mediator of MF in SI-NENs [10,37,38].SI-NEN tissues express high CTGF levels [67]. Proteomic and tissue microarray analysis demonstrated that SI-NENs synthesized and secreted CTGF. Serum levels of both CTGF and TGFβ1 were significantly increased in SI-NENs, suggesting that CTGF could be a marker for MF in SI-NENs [67,68]. Overall, TGF-β1 has an autocrine role in SI-NENs besides its paracrine role in MF by inducing CTGF [31]. The CTGF expression pattern is also correlated with the severity of MF.All these data suggest that an MF network operates, where the initial event is secretion of TGFβ1, which in turn induces CTGF to act in conjunction, thus leading to the overproduction of collagen after bypassing tissue repair suppression. This network seems to be even more complex since other molecules are possibly participating (Figure 2). In vitro experiments have also demonstrated that the N-terminal region of CTGF stimulates myofibroblast differentiation and collagen production, while the C-terminal stimulates fibroblast cell proliferation [37]. CTGF can be found in SI-NEN patient serum and it is potentially involved in MF [69].SI-NENs express PDGF chains and PDGF receptor types [42]. An autocrine growth may be justified by α-receptors, whereas, tumor cells express B- and A-chains, which may also be synthesized by stromal cells stimulating tumor growth in a paracrine manner [42,43]. PDGF was found to be expressed in tumor cells and stroma in SI-NENs. PDGF α-receptors were found on tumor cells and adjacent stroma, whereas PDGF β -receptors were found only in stroma. Hence, PDGF may be involved in the autocrine stimulation of tumor cells and stimulation of stromal cell growth and matrix deposition.Cryosections from SI-NENs showed that the PDGF β-receptor is expressed in 66%, as opposed to 10% of non-NENs tissues respectively. The immunoreaction was present in connective tissue cells adjacent to tumor cell clusters, but not in the tumor cells per se and were more intense in the proximity to the tumor compared to more distant foci [13]. This finding implies that SI-NEN cells may directly or indirectly induce expression of the PDGF-β receptor on adjacent stromal cells in the tumor tissue, which may possibly contribute to the stimulation of connective tissue cell proliferation in SI-NENs [10,13]. However, in one study no detection of PDGFR expression in SI-NENs was reported [46]. In parallel with this, a phase-II study of imatinib, specifically inhibiting the tyrosine kinase domain in PDGFR, showed no substantial regression of SI-NEN, but a significant number of patients with progressive disease achieved disease stabilization [44].Insulin growth factor 1 plays an important role in the physiological regulation of cell growth and differentiation. IGF-1 receptor is present in SI-NEN, suggesting that it may possibly act as an autocrine stimulator of tumor growth [36,37,46]. The recent marker IGF-IEc, an isoform produced by the transcripts of the alternative splicing of the IGF-1 gene has been found to be expressed in SI-NENs [45]. However, since it is in an in vivo SI-NEN model, the serotonin-IGF-1 axis was found to act differentially depending on serotonin levels; hence, more focused studies on MF have yet to be performed [48].EGF in addition to the transforming growth factor alpha (TGF-α) and -β families of GFs, can stimulate or inhibit proliferation and differentiation in multiple tissues [70]. TGF-α was found to be abundant in tumor cells mediating its effects by binding to the EGF receptor (EGF-R) [47]. TGF-α and EGF receptors are expressed in SI-NENs [46,47], supporting the role of TGF-α in the autocrine regulation of tumor growth [49]. The presence of EGF-R in the stromal component of SI-NENs potentially implies the enhanced role of EGF in the process of MF.FGF 2 is a well-known stimulator of vascular endothelial cells and β-FGF is a potent endothelial cell mitogen, playing a crucial role in tumor angiogenesis [12].Her2/neu and vascular endothelial growth factor (VEGF)-receptors (VEGF-Rs) were overexpressed in SI-NENs [46]. More recently, it was demonstrated that interleukin-6 (IL-6), VEGF and monocyte chemo-attractant protein 1 (MCP-1) are actively secreted by cancer-associated fibroblasts, to induce tumor cell proliferation [6]. VEGF expression in vitro was modulated by oxygen levels through a pathway regulated by hypoxia-inducible factor 1 (HIF-1α) [50]. In addition, human fibroblasts have been shown to secrete nerve growth factors (NGFs) that have regulatory effects on angiogenesis [36]. In another study, the fibrotic marker vascular adhesion protein-1 (VAP-1) was highly expressed in SI-NEN stroma in close association with the extracellular matrix [51].In summary, TME represents an important functional unit comprising a cluster of endothelial and inflammatory cells and mesenchymal stroma elements with fibroblasts interconnected and interacting. Fragmented extracellular matrix elements and chemo-attractant molecules recruit immune cells that act as a pro-tumorigenic environment, interacting with pro-angiogenic elements. In specific circumstances, 5-HT via its receptor may be mitogenic in stroma cells acting through a TGFβ-mediated mechanism or potentiating the effects of PDGF, β-FGF or EGF and insulin. CTGF is transcriptionally activated principally through TGFβ1, acting as a downstream mediator of its profibrotic activities on fibroblasts, whereas PDGF, EGF and FGF are also activating CTGF gene expression at the transcriptional level. Of note, infiltration of immune cells is not high in SI-NENS, but tumor-associated macrophages may display a tumor-promoting role whilst they suppress the adaptive immune system and stimulate fibrosis by secretion of profibrotic factors such as TGFβ (Figure 2).Patients with SI-NENs are rarely diagnosed before locoregional LN metastases have developed, and often present with advanced symptomatic disease. In this setting, the disease clinical manifestations are most commonly due to MF around locoregional tumor deposits. MF can cause contraction and tethering of the adjacent bowel loop leading to acute symptoms, such as intestinal obstruction, intussusception and/or intestinal ischemia by impaired blood supply to the intestines [71,72]. The presence of advanced MF associated with intestinal ischemia has been recently identified as a poor prognostic factor for OS [5]. Venous ischemia due to superior mesenteric vein involvement is more often encountered as compared to its arterial counterpart, resulting in venous stasis, abdominal pain, aggravated diarrhea, ascites, malabsorption and malnutrition [73]. In advanced cases, MF surrounding LN metastases in the root of the mesentery may lead to shrinkage and fixation of it to the retroperitoneum that can cause in turn obstruction of the small intestine, duodenum and/or transverse colon. Chyloabdomen may also occur in patients with advanced MF owing to lymphatic obstruction. Occasionally, CS may accompany MF with retroperitoneal extension, when tumor secretory products exceed the detoxifying capacity of the liver, or bypass it, draining directly into the systemic circulation through retroperitoneal lymphatic spread [4,5,71,74]. Extensive retroperitoneal fibrosis may be complicated with obstructive uropathy, hydronephrosis and renal failure late in the disease clinical course [5]. Moreover, involvement of the vasculature in the root of the mesentery may result in collaterals that could predispose for lower gastrointestinal hemorrhages.Apart from acute symptoms, MF can cause partial bowel obstruction and chronic postprandial abdominal pain, influencing patients’ food intake and nutritional status in the long term. In addition, profound diarrhea due to refractory CS, but also due to venous stasis and malabsorption in cases with extensive MF may induce nutritional deficiencies and quality of life impairment. As part of malnutrition in this setting, there are often encountered deficiencies in fat-soluble vitamins, mainly vitamin D, whereas severe disturbances in water and electrolyte homeostasis may also occur [75]. Moreover, in patients with CS, up to 60% of tryptophan is used for serotonin production leading to tryptophan deficiency and subsequently lower niacin levels and sometimes pellagra in a subset of patients [75,76,77]. However, although malnutrition in cancer patients can result in low levels of trace elements like cobalt, copper, fluorine, iodine, selenium and zinc, little is known about trace elements’ deficiencies owing to aggravated diarrhea related to MF in patients with SI-NENs [75].The presence of MF is radiologically defined as enhancing spiculated soft-tissue mass with fibrotic bands radiating outward in the mesenteric fat in a stellate pattern around an LN metastasis on cross-sectional imaging [72,78] (Figure 3). The radiological severity of MF is based on the following classification: (a) no radiological evidence of MF (absence of radiating strands), (b) mild MF (≤ 10 thin radiating strands), (c) moderate MF (> 10 thin strands or < 10 thick strands) and (d) severe MF (≥ 10 thick strands) [78]. Of note, the radiological severity described by Pantograg—Brown is not correlated with any kind of clinical aspects such as symptoms or prognosis [78]. The histological assessment of MF is based on the histological section of the surgical specimen with the maximum amount of fibrous tissue. The mesenteric LN metastasis and surrounding tissue are examined for MF using Sirius Red staining and the width of the thickest fibrous band surrounding the tumor is measured.Importantly, the extension of mesenteric LN metastases and associated MF below or above the horizontal part of the duodenum is a crucial factor for surgical planning [79]. An arbitrary staging system of the mesenteric LN metastases and accompanying MF has been developed by the Uppsala group (Stage I: LN metastases close to the intestine; Stage II: LN metastases higher in the mesentery; Stage III: LN metastases along, but not encasing the mesenteric vessels at the level of the horizontal duodenum; Stage IV: LN metastases extending above the horizontal part of the duodenum, encasing the superior mesenteric vessels) [8]. However, the validity of this system and its effect on surgical management of SI-NEN patients has not been established to date.Notably, when utilizing histopathological, surgical and radiological correlations, computed tomography (CT) imaging has limitations and may underestimate the presence of MF [80,81]. This needs to be considered when making clinical decisions regarding the management of these patients in view of the considerable morbidity associated with MF. Importantly, as most SI-NENs express somatostatin receptors on the cell surfaces, functional imaging with somatostatin receptor-based Gallium-68 positron emission tomography–computed tomography (68Ga–PET–CT) has been shown to detect more primary SI-NENs and also metastatic tumor deposits, including peritoneal carcinomatosis lesions frequently encountered in the proximity of advanced MF, as compared to conventional contrast-enhanced CT [5,82]. Therefore, functional imaging with 68Ga-PET–CT may have a role in the accurate detection and staging of lymph node metastases surrounded by MF with important implications in surgical planning. On the other hand, fluorine-18 ((18)F)-fluorodeoxyglucose ((18)F-FDG) PET may be of limited value in low-proliferative well-differentiated SI-NENs. However, it could be particularly helpful for visualizing a subset of more aggressive SI-NENs, such as well-differentiated tumors with Ki67 values > 10%. According to limited data, (18)F-FDG-avid tumor lesions, even in slow-growing NENs, may indicate a more aggressive biological behavior [83]. Nevertheless, as available pathology data from SI-NEN patients with MF fibrosis have confirmed fibrotic reactions in close proximity to mesenteric LN metastases [5], molecular imaging studies along with conventional cross-sectional imaging can be used to delineate the management of complex cases with MF.When considering the imaging modalities used in other fibrotic conditions, such as liver fibrosis, current cross-sectional and functional methods of evaluating its extent often provide an incomplete picture, whereas certain methods such as ultrasonography and elastography are not applicable in the setting of MF. A novel PET tracer has recently been proposed in a preclinical study demonstrating that imaging hepatic integrin αvβ3 with PET and (18F)-Alfatide may offer a non-invasive method for monitoring the progression of liver fibrosis [84].With respect to diagnostic biomarkers for MF, only a few studies have assessed the utility of non-invasive monoanalyte biomarkers (serum CTGF and urinary 5-HIAA) in this setting [69,85,86]. However, these markers perform modestly and have not been specifically applied in the clinical practice for the assessment of MF as yet. Recently, a set of five profibrotic circulating transcripts with known roles in fibrosis (CTGF, CD59, amyloid precursor-like protein 2 (APLP2), frizzled homologue (FZD) 7 and BCL2/adenovirus E1B 19 kDa protein-interacting protein 3-like (BNIP3L)), taking the name of “fibrosome” has been assessed [81]. The combination of these five circulating transcripts (from the entire 51-gene molecular signature of the NETest) has demonstrated an accuracy of 100% for the detection of microscopic MF and a higher accuracy, as compared to radiological and surgical assessments. Given the ability of the multianalyte NETest to act as a liquid biopsy that can capture the multidimensionality of NEN, the “fibrosome” set appears to be a promising biomarker for MF detection [81]. However, further validation of these results is warranted in prospective studies.Importantly, the mainstay of treatment for MF in SI-NEN patients is locoregional resective surgery either with a curative intent or to provide symptomatic relief, depending on the stage of the disease (Figure 4) [87]. However, compromise of the mesenteric vasculature in the root of the mesentery may render surgical resection technically challenging and endanger the circulation to substantial parts of the intestine, leading to devastating complications such as short-bowel syndrome. Nevertheless, radical resection is not always possible due to the location of large metastatic LN conglomerates and the encasement of the superior mesenteric vessels in the root of the mesentery [5,79]. The National Comprehensive Cancer Network Guidelines on SI-NENs advocate against the resection of a small, asymptomatic, relatively stable primary tumor in the presence of unresectable metastatic disease. A recent study has confirmed no OS benefit in asymptomatic patients with distant-stage disease who underwent upfront locoregional resective surgery, however radiological assessment of MF was not included [88]. These results are also consistent with the findings of two recent cohort studies of SI-NEN patients with MF that confirmed no survival benefit of locoregional resective surgery in the presence of MF in asymptomatic patients with stage IV disease [85,89]. However, a further study has revealed that the presence of MF is linked with a more aggressive biological behavior when PFS was used as a surrogate endpoint, despite similar Ki-67 labelling indices [90]. In addition, size and morphologic features of mesenteric tumor deposits appear to have no prognostic impact; instead, deposit multifocality may be associated with patient outcomes, whereas the extent of bowel resection seems to be associated with the extent of MF [91,92]. A significant association of MF with metastatic patterns and with criteria of functionality has been confirmed, suggesting a common pathophysiological mechanism [86]. It should be noticed though that while asymptomatic patients with MF and disseminated, SI-NENs may not benefit from up-front prophylactic surgery; some patient populations might, such as older patients and patients with progressive loco-regional disease [93]. Importantly, in order to be able to identify patients who might benefit from a prophylactic surgical approach, more insight is needed into the development of MF in SI-NENs [94]. Of note, multivisceral transplantation including the small intestine is a seldom-utilized approach worldwide for highly selected patients with extensive metastatic burden, i.e., potentially some SI-NEN patients with no liver metastases but extensive metastatic disease in the root of the mesentery and associated MF threatening a vascular catastrophe of the gut by encasement of mesenteric vessels [95].SI-NEN patients with an advanced degree of arterial and venous involvement in the root of the mesentery and in the presence of extensive MF are unlikely to receive radical locoregional surgery and should therefore be spared unnecessary laparotomies if asymptomatic [91]. Interestingly, prolonged OS has been demonstrated after minimal invasive interventions in patients with inoperable locoregional disease due to MF, as compared to those with no intervention and a similar stage IV tumor burden [5]. Transhepatic endoluminal stenting of the superior mesenteric vein appears to be a relatively safe procedure that may be undertaken in cases with advanced inoperable disease and severe ischemic symptoms resistant to conservative medical control [5,97]. Stent failure is most likely attributed to thrombosis and the role of anticoagulation in this setting remains to be defined in further studies with longer follow-up [96]. Other minimal invasive measures with palliative purposes in order to preserve renal function in advanced stages may include intra-ureteral J-stent and percutaneous nephrostomy for the management of obstructive uropathy due to advanced SI-NEN-related retroperitoneal fibrosis [5].The CLARINET and PROMID trials have shown that long-term treatment with long-acting somatostatin analogues (SSAs), lanreotide autogel and octreotide slow release, may exert an anti-tumor activity on SI-NENs [98,99]. Treatment with SSAs designed to reduce circulating tumor metabolites (including 5-HT), has been shown to achieve symptomatic and antitumoral response and may confer the additional advantage of retarding or preventing the progression of CHD [100]. In addition, assessment of the level of serum fibrosis markers CTGF and TGFβ in patients with CS treated with prolonged-release SSAs revealed a reduction of CTGF and TGFβ levels, which could in turn prevent the formation of MF [30]. This study confirmed other preclinical ones exhibiting a potential inhibitory effect of SSAs in local fibrosis formation by reducing the secretion of profibrotic growth factors such as TGFβ [101,102]. Despite the control of CS, including normalization of serotonin secretion and reduction of bioactive peptides, the effect of SSAs in MF remains to be determined prospectively in further studies.As the FGF pathway is highly activated in SI-NENs with CS owing to elevated serotonin, it has been implied that therapies that inhibit serotonin synthesis may prevent CHD and MF [71,103]. Efficacy of telotristat ethyl, a novel agent inhibiting peripheral serotonin synthesis was assessed in two phase III clinical trials: TELESTAR and TELECAST [104,105]. This prodrug is converted into its active metabolite: telotristat etiprate that inhibits tryptophan function, resulting in a reduction of the serotonin bloodstream levels [106]. The TELESTAR trial reported a reduction in diarrhea in those patients treated with telotristat ethyl together with a significant reduction of urinary 5-HIAA levels [104]. The TELECAST study provided further confirmation about the efficacy and the safety of telotristat ethyl [105]. With respect to the impact of telotristat ethyl on fibrosis, two of the patients recruited into the TELESTAR trial with known CHD had stable valvular function on follow-up [106,107]. However, no evidence of antitumor effect or impact of telotristat ethyl on MF is available as yet.Apart from inhibition of 5-HTsynthesis, 5-HT receptor antagonists may target serotonin signaling and potentially reduce MF in SI-NENs. The 5HT2 antagonists ketanserin and cyproheptadine have been found to inhibit attacks of flushing, diarrhea and dyspnea in patients with CS [108,109]. Their therapeutic benefit would appear to be a peripheral effect, as 5-HIAA excretion in these patients was not reduced [109]. Another 5-HT receptor antagonist, terguride, reduces the pro-fibrotic potential in scleroderma and suppresses pathways implicated in the regulation of pro-fibrotic genes, suggesting that 5-HT inhibitors might reduce MF via suppression of TGF-β1-mediated non-canonical signaling pathways [110]. Therefore, serotonin synthesis antagonists and 5-HT receptor inhibitors may have an important role not only in the management of severe CS, but also in the prevention of MF, pending confirmation of prospective clinical studies in SI-NENs.The phase III RADIANT-4 trial reported that everolimus significantly improved PFS in patients with progressive SI-NENs [111]. However, the response to treatment is not durable, most likely due to incomplete and unsustained inhibition of mTORC1 signaling by this compound, and/or activation of mTOR complex 2 (mTORC2), underscoring the need for alternative therapies and drug combinations. In a recent preclinical study, in vitro mTOR kinase inhibition effectively stabilized progressive NENs and delayed cardiac impairment [112]. mTOR inhibition may result in a decrease of the expression of 5-HT receptors, as well as that of other players that may be implicated in fibrosis development: however, the exact mechanisms underlying mTOR inhibition’s impact on MF are currently unknown [112,113].SI-NEN proliferation is responsive to a number of GFs, such as EGF, PDGF, TGFα, TGFβ and CTGF, and therefore may be susceptible to tyrosine kinase inhibitors (TKIs) targeting these receptors or associated signaling pathways [23,114,115]. In particular, TGFα stimulation can be inhibited at several points of the MAPK pathway, but success is limited to NEN models and is not evident in the clinical setting. In addition, although some NENs are inhibited by TGFβ1, paradoxical growth has been seen in experimental models of SI-NENs [114]. CTGF expression is associated with more malignant clinical phenotypes, as it promotes growth in SI-NEN models, and is implicated as a mediator of local and distant fibrosis. However, the anti-proliferative effect of CTGF inhibition has not been tested in NENs [114]. Notably, although TKIs have not gained approval and their efficacy seems limited in SI-NENs; their impact on MF and CHD have not been specifically addressed [116,117]. Contemporary research on fibrotic diseases focuses on promising TKIs, such as imatinib-targeting c-alb kinases and PDGF receptors [118,119]. The multiple and overlapping signaling pathways that characterize SI-NENs suggest targeting these tumors at a number of levels may be required to provide efficacy [114]. Although the studies assessing the efficacy of TKIs in SI-NENs did not focus on MF, the molecular pathways involved in its development are similar between SI-NENs and other fibrotic conditions that TKIs can target. Thus, the use of TKIs in SI-NENs could be extended beyond their anti-tumor effects and also be evaluated in the context of prevention or reversal of MF.The efficacy of PRRT in the management of SI-NENs has been confirmed in the NETTER-1 trial [120]. The delivery of targeted PRRT locoregionally in the mesenteric LN conglomerate could be associated with a risk of inducing a local reaction, hence increasing the risk of MF-related complications. In a recent study investigating this, the severity of MF did not seem to be affected after PRRT [121].Angiotensin-converting-enzyme (ACE) inhibitors through renin–angiotensin system inhibition may reduce tumor progression and potentially affect SI-NEN-related MF [122]. Ex vivo angiotensin-converting enzyme (ACE) inhibition has been shown to reduce BON1 cell proliferation, but further in vivo and human studies are needed to address ACE inhibitors’ clinical efficacy in patients with SI-NENs [28]. In addition, tamoxifen has been used in SI-NENs in the 1980s for control of CS and also for that of tumoral growth [33,34,35,123]. More recently, it has been used in other fibrotic diseases, however its clinical value in MF, probably mediated by TGFβ inhibition, has not been determined as yet [124,125].SI-NENs are often associated with the development of locoregional fibrotic reactions in the mesentery that may lead to severe symptomatology with significant associated morbidity. Locoregional resective surgery remains the mainstay of treatment in symptomatic cases of all stages, however its survival benefit in asymptomatic patients with disseminated disease has been recently challenged. Moreover, in inoperable cases, prompt recognition of MF-related complications and minimally invasive interventions as clinically indicated per patient seem effective in disease palliation. As the impact of modern multimodality systemic and/or targeted treatments in SI-NEN-related MF has not been adequately investigated, elucidation of the molecular basis of MF in this setting may guide our ability to develop predictive biomarkers as well as targeted drug therapies for the prevention or reversal of MF. Three-dimensional organoids/spheroids and coculture of these with other cell types may be a promising model for several translational applications that may facilitate the study of NEN stroma complexity and its influence in epithelial tumorigenesis in the context of MF. In addition, the investigation of epigenetic changes in SI-NEN as a driver of MF and the clinical utility of profibrotic circulating transcripts as a novel biomarker for MF stratification should be assessed in further studies. Finally, composite endpoints in SI-NEN future clinical studies, such as health-related quality of life, symptom-specific survival or development of symptoms specifically owing to growing mesenteric LN metastases and/or increasing MF may be more valid than OS and PFS as endpoints when assessing the effects of SINEN-related MF. Although the antitumor efficacy of TKIs and 5-HT synthesis inhibitors and receptor antagonists in SI-NENs may be limited or may not have been fully determined respectively, further studies on these drugs and other novel agents targeting MF are warranted to guide the development of medical treatments that would reduce MF and alleviate associated symptoms.The following are available online at https://www.mdpi.com/2077-0383/9/6/1777/s1, Table S1: Systematic Literature Search Strategy.Conceptualization, G.K. and K.D.; methodology, A.K., K.I.A. and K.D.; data curation, K.I.A., G.W. and K.D.; writing—original draft preparation, A.K., K.I.A. and K.D.; writing—review and editing, all authors.; visualization, A.K., G.W. and K.I.A.; supervision, K.D. All authors have read and agreed to the published version of the manuscript.K.D. was supported by the Royal Swedish Academy of Sciences. The other authors of the study have not received any funding.The authors would like to thank Linda Bejerstrand at the Medical Library of Örebro University for her assistance with systematic literature retrieval. The authors declare that there is no conflict of interest. The funder had no role in the design of the study; in the collection, analyses, or interpretation of data; in the writing of the manuscript, or in the decision to publish the results.Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) flow diagram of study selection. Abbreviations: ENETS; European Neuroendocrine Tumor Society; NENs: Neuroendocrine Neoplasms.Tumor microenvironment represents an important functional unit comprising a cluster of endothelial and inflammatory cells, and mesenchymal stroma elements with fibroblasts interconnected and interacting. An initial event, an injury, induces fragmentation of extracellular matrix elements, chemo-attractant molecules with recruitment of immune cells acting as a pro-tumorigenic environment, followed by pro-angiogenic elements. In specific circumstances, serotonin (5-HT) via a receptor (5-HT Rc) may be mitogenic in stroma cells acting through a TGFβ-mediated mechanism or potentiating the effects of PDGF, β-FGF, or EGF and insulin. TGF-β represents a pro-fibrotic mediator; TGFβ1 and the receptor subtype-2 (TGFβr2) have been identified in SI-NEN. CTGF is transcriptionally activated principally through TGFβ1, acting as a downstream mediator of its profibrotic activities on fibroblasts; PDGF, EGF and FGF are also activating CTGF gene expression at the transcriptional level. PDGF is a mitogen for the fibroblasts and the smooth muscle cells. Infiltration of immune cells is not high in SI-NENS, but tumor-associated macrophages suppress the adaptive immune system and stimulate fibrosis by secretion of profibrotic factors such as TGFβ. Abbreviations: 5-HT: 5-hydroxytryptamine; 5-HT Rc: 5-hydroxytryptamine receptor; TGFβ: transforming growth factor-β; PDGF: platelet-derived growth factor; β-FGF: β-fibroblast growth factor; EGF: Epidermal growth factor; TGFβr2: transforming growth factor-β receptor subtype-2; CTGF: connective tissue growth factor; CAF: cancer-associated fibroblast; GFs: growth factors; IGF-1: insulin growth factor type 1; VEGF: vascular endothelial growth factor; VEGF R: vascular endothelial growth factor receptor; CD8: cluster of differentiation 8; CD4: cluster of differentiation 4; FoxP3: forkhead box P3; Th1: T helper type 1; IFN-α: Interferon alpha.Presence of mesenteric fibrosis in a patient with a small intestinal neuroendocrine tumor as an enhancing soft-tissue mass with fibrotic bands radiating outward in the mesenteric fat in a stellate pattern around a lymph node metastasis in coronary and transverse planes.Management of Small Intestinal NENs with Mesenteric Fibrosis-Flow diagram. † Systemic treatment recommendations adapted from ESMO 2020 Clinical Practice Guidelines for diagnosis, treatment and follow-up of Gastroenteropancreatic Neuroendocrine Neoplasms [96]. Abbreviations. BSC: best supportive care; CAP: capecitabin; CS: carcinoid syndrome; FOLFOX: Folinic acid, Fluorouracil (5-FU) and Oxaliplatin; IF-a: interferon alpha; MF: mesenteric fibrosis; MWA: microwave ablation; NP: nephropyelostomy; PD: progressive disease; PRRT: Peptide Receptor Radionuclide Therapy; RE: radio-embolization; RF: Radiofrequency ablation; TACE: trans-arterial chemo-embolization; SMV: superior mesenteric vein; SSA: somatostatin analogue; SSTR: somatostatin receptor; TAE: trans-arterial embolization; TE: telotristat etiprate; TEM: temozolomide; WD-SINENs: well-differentiated small intestinal neuroendocrine neoplasms.Most studied profibrotic mediators secreted by small intestinal neuroendocrine neoplasms (SI-NENs) and their implications as potential therapeutic targets related to mesenteric fibrosis (MF).Abbreviations: 5-HT: serotonin; ACE: angiotensin-converting enzyme; BMP: bone morphogenic protein; CAFs: cancer-associated fibroblasts; CTGF, connective tissue growth factor; EGF: epidermal growth factor; EGFR: epidermal growth factor receptor; ERK: extracellular-signal-regulated kinase; FGF2, fibroblast growth factor 2; β-FGF: fibroblast growth factor beta; hVAP-1: human vascular adhesion protein-1; IGF-1: Insulin-like growth factor 1; K02288: bone morphogenetic protein (BMP) type I receptor inhibitor; MAPK: Mitogen-activated protein kinase; MEK: Mitogen-activated protein; MF: mesenteric fibrosis; PDGF: platelet-derived growth factor; PKC: Protein kinase C; SI-NEN, small intestinal neuroendocrine neoplasm; SMAD: an acronym from the fusion of Caenorhabditis elegans Sma genes and the Drosophila Mad, Mothers against decapentaplegic; SSAs: somatostatin analogues; TGFα: transforming growth factor alpha; TGFβ, transforming growth factor beta.

Med-MDPI/jcm_2/jcm-09-06-01778.txt

@@ -0,0 +1 @@
+These authors contributed equally to this work.Danggui Liuhuang (DLH) decoctions are traditional herbal medicines that are widely used for menopausal symptoms. The objective of this systematic review was to investigate evidence of the efficacy of DLH decoctions for menopausal symptoms. Fifteen databases were searched from inception until 29 May, 2020. We included randomized controlled trials (RCTs) testing any type of DLH decoction. All RCTs investigating DLH decoctions or modified DLH decoctions were included. The methodological quality of the RCTs was evaluated using Cochrane’s risk of bias assessment tool. We measured the certainty of evidence (CoE) according to the GRADE approach. A total of five RCTs met all of the inclusion criteria and were analyzed. The included RCTs had unclear risks of bias in most domains. Based on studies using the Kupperman index for the measurement of menopausal symptoms, DLH decoctions may have ameliorative effects on menopausal symptoms equivalent to those of conventional drug therapies, but we are very uncertain (MD 3.03, 95% CI −3.17 to 9.25, two studies, very low CoE). Compared with conventional drug therapies, DLH decoctions may reduce hot flashes (MD 0.17, 95% CI 0.28 to 0.06, three studies, low CoE). The difference in the response rate between treatments is very uncertain (risk ratio 1.1, 95% CI 1.07 to 1.25, three studies, very low CoE); the results may indicate that compared with drug therapy, DLH decoction therapy elicits responses in 84 more cases per 1000 cases. In conclusion, there is limited evidence that DLH decoctions improve menopausal symptoms equivalently compared with conventional drug therapies. However, the studies had unclear risks of bias, and the CoEs were very low in general. Additional large and rigorous studies are needed.Menopausal women suffer from a wide variety of symptoms, such as hot flashes, night sweats, depression, anxiety, poor sleep, joint pains, and dry skin, that cause poor quality of life [1,2]. Menopausal women are also vulnerable to cardiovascular disease (CVD) because estrogen withdrawal has a detrimental effect on cardiovascular function and metabolism [3,4]. Hormone replacement therapy (HRT) is the most effective treatment for menopausal syndrome, but current evidence shows that the adverse effects of HRT include increasing the risk of breast cancer. A recent individual participant meta-analysis showed that HRT increases the incidence of breast cancer according to starting age and duration [5]. However, because of the side effects [6,7], many people seek complementary alternative medicines (CAMs) for the management of menopausal symptoms [8,9]. In particular, herbal medicine is the most popular CAM, and a high percentage of menopausal women use it for the management of their symptoms [8,9].There have been five systematic reviews on the use of various herbal medicines for the management of menopausal symptoms [10,11,12,13,14]. Two reviews concluded that there is insufficient evidence for the effectiveness of herbal medicines in improving menopausal symptoms [12,14]. The third review also reported limited evidence for the ameliorative effects of one herbal product on menopausal symptoms [13]. The fourth review suggested that further study is necessary to confirm the effects of herbal medicine on menopausal hot flashes [10]. The fifth study reported the beneficial effects of adjunctive herbal medicines for menopausal depression [11]. However, these studies assessed the efficacy of various types of herbal medicines together, had intrinsic heterogeneities, and are outdated.One herbal medicine, Danggui Liuhuang (DLH) decoction, is widely used to treat menopausal symptoms in clinical practice, and its mechanism has been reported in several studies [15,16,17]. However, none of the previous systematic reviews included studies on the use of DLH decoctions for the management of menopausal symptoms. Therefore, the aim of this review was to summarize the results of the randomized controlled trials (RCTs) with the purpose of analyzing the efficacy and safety of DLH decoctions for the treatment of menopausal symptoms.The protocol of this review was registered with PROSPERO under the number CRD42017079189. We published the protocol in advance and followed the methods to perform this review [18].Fifteen databases were searched from their inception to 29 May, 2020: PubMed, EMBASE, the Cochrane Central Register of Controlled Trials (CENTRAL), AMED, CINAHL, seven Korean medical databases (OASIS, the Korean Traditional Knowledge Portal, the Korean Studies Information Service System, Korea Med, the Korean Medical Database, the Research Information Sharing Service, and DBPia), and three Chinese databases (CNKI, Wanfang, and VIP). We also searched the Chinese clinical trial registry for ongoing trials (http://www.chictr.org.cn/index.aspx). The searches were conducted in Korean, English, and Chinese. The search string used was as follows: “danggui liuhuang decoction” OR “danggui liuhuang tang” OR “danggui liuhuang” and “climacteric” OR ���menopause” OR “menopausal” OR “perimenopause” OR “peri-menopausal” OR “perimenopause period” OR “menopausal syndrome” OR “climacteric syndrome” OR “female climacteric syndrome”. The details of the search strategies are in the in the Supplementary Material.All RCTs and quasi-RCTs comparing DLH decoctions with Western medicine were included. Case studies, qualitative studies, uncontrolled trials, and reviews were excluded.The participants were menopausal women. We excluded women in menopause secondary to surgery, chemotherapy, and/or radiotherapy.Studies using all types of DLH decoctions or modified DLH decoctions were included. The DLH decoctions included the following seven formulas: Angelicae Gigantis Radix, Astragali Radix, Coptidis Rhizoma, Rehmanniae Radix Crudus, Rehmanniae Radix Preparata, Phellodendri Cortex, and Scutellariae Radix. Modified DLH decoctions were defined as single decoctions with one or more supplemental herbs. We excluded DLH decoctions combined with other types of traditional medicine therapies, such as acupuncture, moxibustion, and cupping. The control groups included Western medicine, placebo, and no-treatment groups.Menopausal symptoms (overall, hot flashes, and insomnia) were assessed using response rates and the Kupperman index.The secondary outcomes included adverse events (AEs).All trials from the electronic database searches were reviewed by two authors (J.H.J. and H.Y.L.). They selected the relevant trials through a review of the titles and abstracts. They extracted data according to the pre-defined criteria. Details such as the participants, interventions, outcomes, and results were obtained from each report.Any disagreements were resolved by discussion among the two authors (J.H.Z. and H.Y.L.) and an arbiter (M.S.L.). The authors of the included trials were contacted for clarification if necessary. Data were collected from the included trials by two authors (J.H.J. and H.Y.L.).The risk of bias was assessed using the risk of bias assessment tool from the Cochrane Handbook for the following seven domains: (1) random sequence generation, (2) allocation concealment, (3) blinding of participants and personnel, (4) blinding of outcome assessment, (5) incomplete outcome data, (6) selective reporting, and (7) other bias [19]. We used “L”, “H” and “U” as keys for assessment of the risk of bias, with “L” indicating a low risk of bias, “H” indicating a high risk of bias, and “U” indicating an unclear risk of bias. M.S.L. made the final decision as an arbiter for any unresolved disagreements.We used the GRADE Pro GDT software (https://gradepro.org/) to create a Summary of Findings table. The certainty of evidence (CoE) was assessed for seven categories: number of studies, study design, risk of bias, inconsistency, indirectness, imprecision, and other considerations [20].All of the statistical analyses were performed using Cochrane Collaboration’s software program, Review Manager Software (Version 5.3). For dichotomous data, we present the treatment effects as the relative risk (RR) values with 95% confidence intervals (CIs). Mean differences (MDs) with 95% confidence intervals are used to present the treatment effects for continuous data. We have converted other forms of data into MDs. The Chi-squared test and Higgins I2 test were used to assess heterogeneity.Our search identified 803 potentially relevant trials, of which five studies were included (Figure 1). The data from all of the included trials are summarized in Table 1 [21,22,23,24,25]. All of the included RCTs were performed in China and were published between 2014 and 2019. Four RCTs measured menopausal symptoms using the Kupperman index [21,22,23,24], and four RCTs assessed response rates [21,23,24,25]. There were no ongoing clinical trials for this topic.The risk of bias was unclear in most of the domains for the included RCTs (Figure 2). All of the RCTs reported the use of random sequence generation [21,22,23,24,25], and none of the RCTs mentioned allocation concealment or blinding of outcome assessment. The included trials were comparative studies on DLH decoctions versus conventional drug therapies, and blinding of participants or personnel could not be performed. One study did not report the reasons for drop-out and withdrawal of participants [22]. None of the included studies published or registered their protocols, and the reporting bias was unclear.Four RCTs tested the effects of DLH decoctions compared with drug therapies on menopausal symptoms with the Kupperman index [21,22,23,24]. Two studies reported the total menopausal scores [22,23]. The results of the meta-analysis failed to show superior effects of DLH decoctions compared with drug therapies on total menopausal symptom scores (MD 3.30, 95% CI −3.19 to 9.25, two trials, n = 174, p = 0.34, I2 = 92%, Figure 3A).Three studies reported hot flash scores [21,22,24]. Two studies showed superior effects of DLH decoctions compared with drug therapies in reducing hot flashes [21,24], while the other one failed to do so [22]. The results of the meta-analysis showed favorable effects of DLH decoctions compared with drug therapies in reducing hot flashes (MD −0.17, 95% CI −0.28 to −0.06, three trials, n = 264, p = 0.003, I2 = 82%, Figure 3B).Three studies assessed response rates based on total menopausal symptoms [21,23,24]. Two RCTs reported superior effects of DLH decoctions compared with drug therapies [21,24], while the other one showed equivalent effects between the two groups [23]. The meta-analysis failed to show superior effects of DLH therapy compared to drug therapy on response rates (RR 1.10, 95% CI 0.97 to 1.25, three trials, n = 270, p = 0.15, I2 = 60%, Figure 3C).Two RCTs reported the response rates for DLH decoctions compared with drug therapies and showed equivalent effects between the two groups [23,25]. The meta-analysis also failed to show significant differences between the two groups (RR 1.07, 95% CI 0.76 to 1.51, two trials, n = 136, p = 0.71, I2 = 83%, Figure 3D).Three RCTs assessed the AEs [22,24,25], while the other two studies did not measure the AEs. One RCT reported several AEs in both groups, including breast pain, gastrointestinal reaction, stomach ache, and vaginal bleeding [22]. The other two studies reported AEs for only drug therapies, including breast pain, vaginal bleeding, daytime sleepiness, dizziness, and fatigue [23,25].Few rigorous studies have investigated the effects of DLH decoctions on menopausal symptoms. Evidence from the included studies showed equivalent ameliorative effects between DLH therapy and conventional drug therapies on menopausal symptoms, including hot flashes and insomnia. However, the small number of studies, small sample sizes, and unclear risks of bias prevent firm conclusions from being drawn.DLH decoctions may be better than single-agent therapies at improving menopausal symptoms and may reduce both the required doses of conventional drugs and the associated AEs. Despite such promising results, caution should be practiced before generalization of these results, given the diversity of symptoms and the different states of herbal medicines in various countries.The CoE was consistently very low or low for all outcomes (Table 2). The most common reasons for the low values were a lack of allocation concealment and a lack of blinding, which are known to result in the overestimation of effect sizes. The CoEs were low due to indirectness and imprecision. High or unclear risks of bias in the included studies reduced their reliability. Although a blinded study was unlikely to be possible due to the use of a decoction, it was unfortunate that no studies provided a placebo decoction to the control group.This review has several limitations. Firstly, although considerable efforts were made to retrieve all RCTs on the subject, we cannot be absolutely certain that we were successful. Moreover, selective publishing and reporting are major causes of bias that must be considered. It is conceivable that several RCTs with negative results were not published, which could have distorted the overall picture. Secondly, all of the included studies were conducted in China, and a degree of uncertainty regarding the applications of the findings remains. Finally, the reviewed studies had high risks of bias, and there was a paucity of published studies; therefore, the conclusions of this review might be overstated.There have been no previous reviews on this type of herbal decoction. However, the evidence level is similar to that of the systematic reviews on other herbal medicines [10,13,14].The short-term clinical efficacy of the DLH decoctions was good, and their safety was comparable to that of conventional drug therapies. Several approaches were suggested for the management of menopause including lifestyle changes; phytoestrogen; HRT; and sole, combination, or conjugated therapies with estrogen [26]. However, several adverse events associated with HRT and estrogen have been reported, including increased rates of breast cancer. Thus, DLH decoctions appear to be an option for clinical use. However, due to the lack of standardized authoritative assessment indicators of efficacy, the objectivity of the efficacy assessment was compromised. Therefore, we relied mainly on self-reported measures to determine the effects on the main outcomes.Two of the main limitations of the included studies were the lack of detailed reporting and the lack of transparency regarding the research process. These limitations resulted in low CoEs and reduced confidence in the pooled results. Future studies should follow the recommended reporting guidelines, including the CONSORT guidelines, to better enable other researchers to understand the study designs and the experimental, analysis, and interpretation methods [27]. They should also utilize adequate allocation concealment, optimal treatment dosages, and sample sizes based on recognized sample size calculations. In addition, important procedures, including the use of validated primary outcome measures and adequate statistical tests for intention-to-treat and missing data, should be undertaken in future research. We excluded the studies with patients with menopause induced by cancer treatment in this study. It may be worthwhile to focus on these patients, who may particularly benefit from nonhormonal treatment, in future study.In conclusion, the existing studies show equivalent effects of DLH decoctions and conventional drug therapies on menopausal symptoms. However, due to the small number of studies and the high risks of bias, the evidence is limited. Further rigorous RCTs are needed to overcome the many limitations of the current evidence.The following are available online at https://www.mdpi.com/2077-0383/9/6/1778/s1, supplementary material: Search strategies.H.W.L., J.H.J. and M.S.L. conceived the study, developed the criteria, searched the literature and wrote the review. J.Z., F.Y. advised on the review design and revised the manuscript. All authors have read and approved the final manuscript. All authors have read and agreed to the published version of the manuscript.This study was supported by the Korea Institute of Oriental Medicine (KSN2013210 and KSN2013240), Korea.The authors declare no conflict of interest.Flow chart of the trial selection process. CCT: clinical controlled trial; RCT: randomized controlled trial.Risks of bias. (A) Risks of bias of the included studies. The authors reviewed each item’s risk of bias for each included study. (B) Risks of bias of individual studies. +: low risk of bias; −: high risk of bias; ?: unclear.Forest plots of (A) Kupperman menopausal index scores, (B) hot flash scores, (C) response rates (total), and (D) response rates (insomnia).Summary of randomized clinical studies of Danggui Liuhuang (DLH) decoction for menopausal symptoms.DLH decoction: Danggui Liuhuang decoction; KI: Kupperman index; n.r.: not reported; NS: not significant. * We excluded two additional groups: TCM patent prescription group (n = 42) and DLH decoction plus TCM patent prescription group (n = 42). † We excluded these from the analysis because they were not in the included criteria.Summary of findings.* The risk in the intervention group (and its 95% confidence interval) is based on the assumed risk in the comparison group and the relative effect of the intervention (and its 95% CI). CI: Confidence interval; DLH: Danggui Liuhuang; MD: Mean difference; RR: Risk ratio. GRADE Working Group grades of evidence: High certainty (⨁⨁⨁⨁): We are very confident that the true effect lies close to that of the estimate of the effect. Moderate certainty (⨁⨁⨁◯): We are moderately confident in the effect estimate: The true effect is likely to be close to the estimate of the effect, but there is a possibility that it is substantially different. Low certainty (⨁⨁◯◯): Our confidence in the effect estimate is limited: The true effect may be substantially different from the estimate of the effect. Very low certainty (⨁◯◯◯): We have very little confidence in the effect estimate: The true effect is likely to be substantially different from the estimate of effect. Explanations: a Downgraded by one level for study limitations: unclear or high risk of bias in half of the domains in included studies. b Downgraded by one level for substantial heterogeneity in included studies. c Downgraded by two levels for imprecision: wide confidence interval crossing assumed threshold of minimal clinically important difference. d Downgraded by one level for imprecision: small sample size.

Med-MDPI/jcm_2/jcm-09-06-01779.txt

@@ -0,0 +1,2 @@
+Diabetic foot infections (DFIs) are severe complications of long-standing diabetes, and they represent a diagnostic challenge, since the differentiation between osteomyelitis (OM), soft tissue infection (STI), and Charcot’s osteoarthropathy is very difficult to achieve. Nevertheless, such differential diagnosis is mandatory in order to plan the most appropriate treatment for the patient. The isolation of the pathogen from bone or soft tissues is still the gold standard for diagnosis; however, it would be desirable to have a non-invasive test that is able to detect, localize, and evaluate the extent of the infection with high accuracy. A multidisciplinary approach is the key for the correct management of diabetic patients dealing with infective complications, but at the moment, no definite diagnostic flow charts still exist. This review aims at providing an overview on multimodality imaging for the diagnosis of DFI and to address evidence-based answers to the clinicians when they appeal to radiologists or nuclear medicine (NM) physicians for studying their patients.Diabetic foot infection (DFI) is a common complication of longstanding diabetes, and it is associated with considerable morbidity, the increased risk of lower extremity amputation, and a high mortality rate [1]. The development of DFI derives from a complex interplay among peripheral neuropathy, peripheral arterial disease (PAD), and the immune system.Neuropathy is the most prominent risk factor for diabetic foot ulcerations (DFU). Motor neurons damage results in foot deformities that contribute to the injury of foot tissues and bones. Sensory neurons damage leads to a loss of protective sensation. Therefore, neuropathic patients could develop skin ulcers that might remain unrecognized for a long time, thus exposing the adjacent soft tissues to the colonization of pathogens and causing a soft tissue infection (STI). If not promptly identified and treated, the infection could spread to the underlying bone and cause osteomyelitis (OM).PAD further facilitates micro-organisms invasion and rapid progression to infection, since insufficient tissues oxygenation might impair the healing of ulcers, creating an optimal substratum for the colonization of pathogen. In addition to this, PAD reduces granulocytes migration and antibiotic penetration into infected site, thus contributing to the spread of infection and complicating its therapeutic management. Moreover, patients with severe PAD are prone to sudden ischemia resulting from arterial thrombosis with consequent critical limb ischemia and an increased risk of amputation [2,3]. Indeed, both ischemia and infection are the most important factors in determining the prognosis of foot ulcerations, since patients with PAD and infection show more severe comorbidities and worst clinical outcomes compared with the classic “neuropathic foot patients” [4]. Uncontrolled hyperglycaemia represents another pivotal aspect in the pathogenesis of DFI being responsible of an impairment of both cell-mediated and humoral immune response mainly characterized by altered leukocyte’s functions, reduced chemotaxis, and phagocytosis proprieties [5,6].A prompt identification of foot ulcers, STI, and OM and an accurate evaluation of the extent of the infective process is crucial for prognostication of the patients and for planning the most appropriate treatment that usually requires a combination of metabolic control, medical treatment with specific antibiotic regimen, and surgical approach. The International Working Group (IWGDF) and the Infectious Diseases Society (IDSA) proposed a single scheme to assess the presence and the severity of infection [7,8], and this classification is currently applied for predicting the need for hospitalization, the likelihood of undergoing lower extremity amputation, and other adverse outcomes [9]. However, in the latest update of these guidelines, OM and STI have been addressed separately, since they are two distintict conditions, although they may coexist in the same patient, with different diagnostic, therapeutic, and prognostic implications [10].Clinical suspicion through a comprehensive history and physical exam are the starting points for the diagnosis of DFI, which is validated by a complete laboratory evaluation, microbiologic assessment, and imaging.Clinical diagnosis of superficial STI is based on the presence of at least two local signs of inflammation: rubor, calor, dolor, tumour, or purulent secretion. Other secondary features suggestive of infection may be present, such as friable or discolored granulation tissue, necrosis, and failure of the wound to heal [11]. Clinical manifestations of acute deep infection include abscess, necrotizing fasciitis, and gangrene. In those cases, the infection process may involve one or more foot compartments and may require a first urgent surgical treatment and eventually distal revascularization to reduce the amputation level [12].The development of an OM is one of the most serious and disabling complications of diabetes, being associated with prolonged antibiotic therapy and hospitalization, as well as higher re-infection rates and risk of amputations compared with patients with STI, resulting in high social costs [13].Diagnosing OM is sometimes a challenge for the clinicians, since it may occur in the absence of local or systemic signs of infection and inflammation, especially in chronic infections. Several wound characteristics, in particular the width and depth of the lesion, may be helpful in predicting the presence of bone infection. A lesion’s surface greater than 2 cm2 has a sensitivity of 56% and a specificity of 92% for the diagnosis of OM. Similarly a deep ulcer over 3 mm is significantly associated with an underlying OM in comparison with a more superficial one (82% versus 33%) [14].Another diagnostic criterion is represented by the possibility to reach the bone with a blunt at the base of the lesion, the “probe-to-bone test”. Combining the results of the probe-to-bone test with those of plain radiography improves the overall diagnostic accuracy of OM [15,16].However, the gold standard for the definitive diagnosis of OM still remains the bone biopsy that provides histological and microbiological information and, at the same time, it is useful to determine the susceptibility to various antibiotics [7]. Although bone biopsy is the more accurate technique in identifying the pathogenic germs, it is an invasive procedure, and it is not always feasible. However, a culture of deep soft tissue that is in direct contact with the bone shows a good correlation with bone biopsy in identifying the responsible pathogen and, therefore, this approach may be useful in alternative to bone biopsy [17].Imaging offers the possibility to diagnose DFI by using a less invasive approach that is complementary to physical examination, laboratory, and microbiological evaluations. A wide panel of modalities may be very helpful for the clinicians to better understand whether the patient has a STI, OM, or sterile inflammation that is a hallmark of Charcot osteoarthropathy, for example (Table 1). To achieve an accurate differential diagnosis is mandatory in the optic of promptly starting an appropriate treatment reducing the need for hospitalization and the risk of major amputations, but univocal consensus on diagnostic criteria for imaging modalities still does not exist.This review aims at providing an overview of radiologic and nuclear medicine (NM) modalities able to achieve an accurate differential diagnosis between the different kinds of DFI and to guide therapeutic strategies.Surgical management for diabetic foot (DF) deformities and complications is a critical aspect in dealing with these patients. Understanding of the DF ‘syndrome’ has improved the approach to diabetic patients affected by a complicated foot. In the last decades, we observed an increasing interest in developing less invasive surgical procedures as alternatives to major lower extremity amputation. They are focused on local resections and the drainage of infected underlying soft tissue, toes, and metatarsal heads for neuropathic or neuroischemic complicated DF [18,19]. In this optic, imaging plays a crucial role in diagnosing the infection and defining its extent, aiming at selecting those cases candidate to more conservative approaches.Structural deformities and high plantar pressures are a predisposing risk factor to diabetic foot ulceration (DFU) [20,21,22,23]. Common deformities include hammertoes, prominent metatarsal heads, hallux limitus, Charcot foot, and previous toe or partial foot amputations [24]. Each leads to high pressures that contribute, in the case of an insensitive DF, to tissue inflammation and ulceration. Ameliorating these high pressures by structurally realigning or removing bony prominences is the rationale for foot surgery. In the presence of infection, phlegmon, and/or OM, surgery becomes a critical urgent component of care [25]. A proposed scheme for classifying the types of foot surgery in diabetic patients refers to the presence of open wounds and their acuity [26]:Prophylactic procedures are those performed in neuropathic patients to reduce the risk of ulceration or recurrent ulceration in the absence of open wounds;Curative surgery when cutaneous ulcers are present is often performed to provide a cure by joint resection, removing underlying bony prominences (surgical decompression), osteomyelitis, or by draining underlying abscesses or phlegmons;Urgent procedures are performed for severe deep or ascending infections (infectious gangrene, necrotizing fasciitis, etc.) to control the progression of infection. These procedures are performed emergently and usually consist in wide open drainages or minor amputations at the foot level.Prophylactic procedures are those performed in neuropathic patients to reduce the risk of ulceration or recurrent ulceration in the absence of open wounds;Curative surgery when cutaneous ulcers are present is often performed to provide a cure by joint resection, removing underlying bony prominences (surgical decompression), osteomyelitis, or by draining underlying abscesses or phlegmons;Urgent procedures are performed for severe deep or ascending infections (infectious gangrene, necrotizing fasciitis, etc.) to control the progression of infection. These procedures are performed emergently and usually consist in wide open drainages or minor amputations at the foot level.In daily clinical practice, curative and urgent procedures are most frequent, since usually patients arrive to a surgical referral with an active more or less complicated DFU.When dealing with deep infected cutaneous ulcers, the primary principle in treating surgical infection is source control. Most infected DFUs respond well to local debridement, the administration of culture-specific antibiotics, and offloading of the foot with specific footwear. Some develop a rapid spread of infection along the tissue planes and tendon sheaths and present with local tissue necrosis, spreading cellulitis and systemic inflammatory response [27].According to the T.I.M.E. (Tissue, Infection, Moisture and Edges) procedure, source control includes the resection and/or debridement of any dead/infected tissue/bone and avoid fluid stasis by draining any hidden infected site [28]. However, Time stands also for “do not waste Time” in referral the patient to specialists who can better deal with the patient’s need and also stands for “Timing”, indicating untimely or adequate choice of procedure (for example, limb revascularization) to treat the patient at its presentation. Since deep foot infection can potentially be limb threatening without timely intervention, delay will lead to further tissue loss. In this case, we can state that “Time is Tissue”.The endpoints of curative approach to deep foot ulcer and osteomyelitis are:Treat and cure the infection;Reduce pain (not always present because of neuropathy);Retain foot and allow best function (rehabilitation);Reduce recurrency.Treat and cure the infection;Reduce pain (not always present because of neuropathy);Retain foot and allow best function (rehabilitation);Reduce recurrency.Radical surgical resection, including healthy bone and soft tissue, is sometimes required and must follow an “oncologic approach” in the case of deep foot infections and OM [29,30], since they are difficult to treat and they could relapse.Our understanding of the pathophysiology has been greatly improved by the biofilm model, which explains the wide variety of symptoms, courses, and the complex therapeutic management. The pathogens first form the surface layer of colonies, which then multiply into a three-dimensional structure. This biofilm structure offers the bacteria protection from mechanical influences and makes it harder for antibiotics, the body’s own defensive cells, and antibodies to penetrate, functioning as a diffusion barrier. The pathogens pass from a planktonic, free-floating phase with a high metabolic rate and rapid multiplication into a sessile form with greatly reduced metabolism and slowed biological reactions. This phenotypic change makes them more resistant to antibiotics compared to planktonic counterparts, since cellular growth within biofilms produces a matrix that protects the pathogens from the immune system and antimicrobial drugs. In OM and prosthesis-related infections, it has been calculated that this particular type of growth can reduce their sensitivity to antibiotics by a factor of 103 [31]. The time required for a mature biofilm formation is about 24–48 h [32]. Mechanical forces of surgical debridement are effective in disrupting that matrix, exposing bacteria to the effects of antibiotics and body’s immune response. Therefore, with a surgical medication, we can realize a therapeutic window of 1–2 days where a sharp debridement should be repeated in order to remove all instable tissues and biofilm covering the wound bed.All foreign bodies including screws and stitches must be removed, since they might be biofilm carriers. Any infected tendons and bone should be cleaned and irrigated in order to remove necrotic and/or infected tissues. The remaining tissues must be viable and well-perfused. There are no objective criteria for defining bone resection limits; therefore, it remains an individual decision of the surgeon, but generally, it should be up to when a hard bone is touched with the surgical instrument [33]. In some cases, non-infected bones must be removed or reduced in order to relieve the pressure to the underlying ulcerated cutaneous plane. The size of the defect produced by the procedure is not a primary consideration; only the vascular supply should be evaluated and preserved. What happens next depends on how radical the débridement and resection has been. Thereafter, the most important aspect is the management of dead space, which, if not treated properly, may lead to the early recurrence of infection and inadequate rehabilitation, especially if it involves the foot plantar surface. Surgical drainage is mandatory for the prevention of any fluid or exudate stasis that might be responsible for persistent bacterial contamination, biofilm, or infection and wound-healing impairment and delay [34] (Figure 1).Concluding, the surgical management of DF complications is challenging and it requires an appropriate diagnosis in order to correctly identify the problem and to promptly start an adequate and a personalized treatment for the single patient. An interdisciplinary approach derived from close collaboration between clinicians, surgeons, radiologists, NM physicians, microbiologists, podiatrists, and nurses is mandatory.Although the reference standard for the diagnosis of diabetes-related OM still remains bone biopsy, the diagnosis is largely based on the presence of clinical and laboratory findings such as an erythrocyte sedimentation rate (ESR) >70 mm/h, and a positive result of a probe-to-bone test (palpation of bone in the depths of infected pedal ulcers) [10,11]. However, it should be kept in mind that (1) an ESR of more than 70 mm/h is highly specific for OM, but has a sensitivity of only 28% [35], and (2) the reliability of the probe-to-bone test may vary with the performing clinician’s experience and ulcer location [10,11,35]. In addition, the benefit of this test is substantially influenced by the pre-test probability of the patient having an OM or not. A positive probe-to-bone test suggests the diagnosis in a high-risk patient. A negative test indicates a low probability of OM in a low-risk patient [36,37]. Hence, the diagnosis of DFI may be difficult when based on clinical and laboratory findings alone. Advanced imaging of the foot has improved our ability to evaluate the possibility of OM, and it may be helpful for the diagnosis and definition of deep or soft-tissue purulent collections.Radiography and magnetic resonance imaging (MRI) are the most commonly used radiological modalities to evaluate the DF infective complications. Ultrasounds can be employed for guiding the aspiration of suspect fluid collections or removing foreign bodies; however, it is not currently recommended by the IWGDF [10] and the diabetic foot guidelines of the American College of Radiology [38]. Computed tomography (CT), despite its higher sensitivity compared with radiography and MRI in detecting cortical erosions, periosteal reaction, small sequestra, soft tissue gas, and calcifications within sites of chronic osteomyelitis, plays a limited role in the imaging of diabetic patients with suspected OM or STI of the foot [38]. The main disadvantages of CT are the low soft tissue contrast resolution and the inability to detect the bone marrow edema seen in the early stages of infection. If MRI is contraindicated or unavailable, post-contrast CT may be used to detect soft-tissue and osseous abscess formation. However, the risk of use of iodinated contrast in diabetic patients should be taken into account, as diabetic nephropathy progressing to end-stage renal disease is commonly a comorbidity in patients with diabetes [39].The sensitivity of radiography is rather low in this setting, since radiographic findings of DF infective complications can be undetected for up to four weeks after the onset of infection, and these changes can be caused by Charcot osteoarthropathy and other disorders such as gout [40,41]. However, radiography should be the first-line imaging modality in any patient with suspected infection. It is cheap, widely available, and when radiographic findings such as demineralization, bone resorption, cortical destruction, periosteal reaction, bowing, or the obliteration of fat stripes and fascial planes, arthropathic changes, and the presence of soft tissue gas and foreign bodies are interpreted by an experienced radiologist, they are highly suggestive of DF infective complications [10].After initial radiography, MRI with fluid-sensitive, fat-suppressed sequences (i.e., short-tau inversion recovery [STIR] or fat-saturated T2-weighted images) is the modality of choice for investigating OM and associated soft-tissue complications [42,43] with high sensitivity and high specificity (90% and 83%, respectively) in the diagnosis of OM [38,44]. Post-contrast images improve the evaluation of soft tissue pathology, as they help in detecting abscesses and sinus tracts more easily [43]. Moreover, its radiation-free assessment becomes particularly important in the young population and when repeated follow-up imaging is likely to be necessary. However, standard MRI is typically based only on morphologic sequences, which provide only structural information. In the last years, technical improvements have allowed the capability to add functional quantitative information to structural information. The application of Dixon sequences improves image quality and increases the detection of sinus tracts and intraosseus sequestrums [45]. Diffusion-weighted imaging and the apparent diffusion coefficient value can help in the differentiation of diabetic neuropathic osteoarthropathy from OM with excellent inter-observer agreement [45].Bone marrow (BM) with normal signal intensity excludes the diagnosis of OM in diabetic patients with STIs. Early OM is characterized by BM edema with low marrow signal intensity on T1-weighted images, high marrow signal intensity on fluid-sensitive fat-suppressed sequences, and post-contrast enhancement (Figure 2).However, there are several mimickers of diabetes-related OM that may present problems to making a correct MRI diagnosis by showing BM edema and post-contrast enhancement. Furthermore, these conditions, including biomechanical stress changes related to altered weight bearing, recent post-operative surgery, inflammatory arthritis, and primarily neuropathic osteoarthropathy, may coexist with OM, further complicating the ability to make an accurate diagnosis. Consequently, if marrow edema is used as the primary diagnostic criterion, MRI may not be very specific.Several secondary features such as subtending skin ulcer, sinus tract, abscess, tenosynovitis, or septic arthritis tend to be associated with OM. Their presence strongly suggests that osteomyelitis is present and can improve diagnostic accuracy [42,46].
+        Skin ulcer: Skin ulceration is typified by focal interruption of the cutaneous line, with raised margins (secondary to preexisting callus formation). Acute ulcer appears hyperintense on fluid-sensitive fat-suppressed images, with marked peripheral post-contrast enhancement, which is a finding that is indicative of granulation tissue at the base of the ulcer. Chronic ulcer may be associated with fibrous healing and thus appears as a mass with low signal intensity on T1-weighted images and low to intermediate signal intensity on fluid-sensitive fat-suppressed images [42,43,46].Sinus tract and abscess: Sinus tracts and abscesses are some of the major findings in osteomyelitis. Morrison et al. determined the usefulness of primary and secondary MRI signs of OM and found that the identification of a sinus tract showed high specificity (average, 85%) for the diagnosis of osteomyelitis in the adjacent bone [47]. Sinus tracts typically extend from skin ulcers to tendon sheaths, bones, or joints, and they represent a route for the subsequent spread of infection leading to abscesses, septic tenosynovitis, and/or osteomyelitis [47]. Sinus tracts appear as linear fluid signal intensity on fluid-sensitive fat-suppressed images and display a characteristic “tram-track” pattern of the enhancement on contrast-enhanced images. The latter are the most sensitive MRI feature for detecting sinus tracts (Figure 3). Abscess is seen as a focal fluid collection that is hypointense on T1-weighted images and hyperintense on fluid-sensitive fat-suppressed images, with a thick rim post-contrast enhancement, due to the presence of granulation tissue (Figure 3). The presence of rim enhancement is essential in distinguishing abscesses from cellulitis or phlegmons, which present diffuse post-contrast enhancement [42,43,46].Septic tenosynovitis: Septic tenosynovitis generally results from the contiguous spread of infection from an adjacent ulcer, abscess, or sinus tract. On MRI, it is characterized by an abnormal increase in fluid within the tendon sheath, and post-contrast images may show a thick rim enhancement around the tendon, due to inflamed synovium. The tendon loses its constant low signal intensity and becomes thickened and indistinct [42,43,46].Septic arthritis: Similar to OM and tenosynovitis, septic arthritis occurs also as a result of contiguous spread from an adjacent ulcer, abscess, or sinus tract. No single MRI feature can differentiate septic from nonseptic arthritis; increased joint fluid and synovial thickening with contrast enhancement may also be seen in non-infectious inflammatory arthropathies. However, in pedal infections, the diagnosis of septic arthritis may be more specific if an ulcer and adjacent soft-tissue infection directly abut the joint, or, a sinus tract extends into the joint. Septic arthritis may demonstrate edema with post-contrast enhancement in adjacent soft tissue and on both sides of the joint. Reactive BM oedema, secondary to septic arthritis, should be differentiated from a superimposed OM. A low signal intensity on T1-weighted images, and proximal extension of subchondral edema beyond the subchondral bone usually indicate OM [42,48].Skin ulcer: Skin ulceration is typified by focal interruption of the cutaneous line, with raised margins (secondary to preexisting callus formation). Acute ulcer appears hyperintense on fluid-sensitive fat-suppressed images, with marked peripheral post-contrast enhancement, which is a finding that is indicative of granulation tissue at the base of the ulcer. Chronic ulcer may be associated with fibrous healing and thus appears as a mass with low signal intensity on T1-weighted images and low to intermediate signal intensity on fluid-sensitive fat-suppressed images [42,43,46].Sinus tract and abscess: Sinus tracts and abscesses are some of the major findings in osteomyelitis. Morrison et al. determined the usefulness of primary and secondary MRI signs of OM and found that the identification of a sinus tract showed high specificity (average, 85%) for the diagnosis of osteomyelitis in the adjacent bone [47]. Sinus tracts typically extend from skin ulcers to tendon sheaths, bones, or joints, and they represent a route for the subsequent spread of infection leading to abscesses, septic tenosynovitis, and/or osteomyelitis [47]. Sinus tracts appear as linear fluid signal intensity on fluid-sensitive fat-suppressed images and display a characteristic “tram-track” pattern of the enhancement on contrast-enhanced images. The latter are the most sensitive MRI feature for detecting sinus tracts (Figure 3). Abscess is seen as a focal fluid collection that is hypointense on T1-weighted images and hyperintense on fluid-sensitive fat-suppressed images, with a thick rim post-contrast enhancement, due to the presence of granulation tissue (Figure 3). The presence of rim enhancement is essential in distinguishing abscesses from cellulitis or phlegmons, which present diffuse post-contrast enhancement [42,43,46].Septic tenosynovitis: Septic tenosynovitis generally results from the contiguous spread of infection from an adjacent ulcer, abscess, or sinus tract. On MRI, it is characterized by an abnormal increase in fluid within the tendon sheath, and post-contrast images may show a thick rim enhancement around the tendon, due to inflamed synovium. The tendon loses its constant low signal intensity and becomes thickened and indistinct [42,43,46].Septic arthritis: Similar to OM and tenosynovitis, septic arthritis occurs also as a result of contiguous spread from an adjacent ulcer, abscess, or sinus tract. No single MRI feature can differentiate septic from nonseptic arthritis; increased joint fluid and synovial thickening with contrast enhancement may also be seen in non-infectious inflammatory arthropathies. However, in pedal infections, the diagnosis of septic arthritis may be more specific if an ulcer and adjacent soft-tissue infection directly abut the joint, or, a sinus tract extends into the joint. Septic arthritis may demonstrate edema with post-contrast enhancement in adjacent soft tissue and on both sides of the joint. Reactive BM oedema, secondary to septic arthritis, should be differentiated from a superimposed OM. A low signal intensity on T1-weighted images, and proximal extension of subchondral edema beyond the subchondral bone usually indicate OM [42,48].Distinguishing OM from neuropathic osteoarthropathy, in the absence of secondary signs of infection, is a common and difficult clinical and radiological problem. An accurate differentiation is mandatory, because the early detection of OM is essential to initiate prompt medical and/or surgical treatment. The location and distribution of anatomical changes may be helpful. Indeed, neuropathic osteoarthropathy usually involves the tarsometatarsal and metatarsophalangeal joints, while OM mostly involves the calcaneum, malleoli, and forefoot [49]. The biggest diagnostic problem arises in the midfoot. In this region, MRI findings may be inconclusive, and secondary signs of infection are invaluable in determining the presence of OM. Furthermore, neuropathic osteoarthropathy is primarily an articular disease; thus, BM oedema is limited to juxta-articular locations, whereas OM, which almost invariably results from an ulcer or abscess in contiguous soft tissue, shows diffuse marrow changes (Figure 3) [38,50].The differentiation of infected from non-infected neuropathic osteoarthropathy remains extremely challenging, as the clinical and radiological findings may overlap. However, several MRI findings may be useful for distinguishing between these two conditions. Sinus tract formation, the replacement of soft tissue fat, fluid collections and diffuse marrow abnormality, diffuse joint fluid enhancement, and joint erosion support superimposed infection [43,51]. Thin rim enhancement of effusion, the presence of subchondral cysts, or intraarticular bodies indicate the absence of infection [19]. Bones that “disappear” on T1-weighted images and then “reappear” on contrast-enhanced or T2-weighted images (the “ghost sign”) is another MRI feature that indicates the presence of a superimposed infection. In uncomplicated neuropathic osteoarthropathy, the “ghost sign” is absent because there is bone destruction, but there is no infiltration of the marrow by inflammatory cells resulting in absence of the “ghost sign” [42,43,46].NM techniques offer the possibility to image a process from a functional point of view, and they allow the identification of pathophysiological changes even before they become clinically detectable. Several radiopharmaceuticals are available for imaging infection and inflammation for both single photon emission computed tomography (SPECT) and positron emission tomography (PET) modalities, and most of them are currently applied for the diagnosis and follow-up of DFI.Radiolabelled white blood cells (WBC) scintigraphy using both 111In or 99mTc represent the NM cornerstone for the diagnosis of infection, since it specifically targets activated granulocytes, thus representing a surrogate marker of bacterial infections [52]. Several guidelines have been published by the European Society of Nuclear Medicine (EANM) with the aim to standardize labeling procedures, acquisition protocols, and interpretation criteria in all the centers [53,54,55]. In particular, to provide an in vivo imaging of the physiologic dynamic process of migration of granulocytes into the infective site, it is recommended to acquire images with times corrected for isotope decay at three time points after the reinjection of autologous cells. Once correctly acquired and displayed, the correct interpretation derives from the comparison of uptake extent and intensity between late images, acquired 20 h (h) post injection (p.i.) and delayed images (3 h p.i.). By following these recommendations, we can easily differentiate between a bone infection from a sterile inflammation. Indeed, in the first situation, the uptake increases over time in terms of extent and/or intensity, whereas in inflammation, the uptake decreases or remains stable over time [55,56,57]. By using these recommendations, and when combined with SPECT/CT acquisitions for the evaluation of the extent of the process and for the precise localization of the uptake, this modality reaches a very high accuracy in diagnosing an infection [58] (Figure 4). In a recently published meta-analysis and systematic review comparing the diagnostic performance of WBC scan, Fluorine-18 Fluorodeoxyglucose positron emission tomography ([18F]FDG PET/CT) and MRI for the detection of DF osteomyelitis (DFO), the pooled sensitivity and specificity of radiolabelled WBC were respectively 91% and 92% for 99mTc hexamethylpropylene amine oxine (HMPAO) and 92% and 75% for 111In-oxine. In particular, 99mTc-HMPAO WBC scintigraphy, followed by [18F]FDG PET/CT, showed higher specificity than other imaging modalities in the diagnosis of DFO, whereas the sensitivities were similar (approximately 90% for all) [59].However, data regarding the use of radiolabelled WBC scintigraphy in DF are very discordant in the literature [60]. The sensitivity and specificity of this modality range from 75% [61] to 100% [62,63,64] and from 67% [64] to 100% [65] respectively, depending on deviation from the suggested labeling procedure, interpretation criteria adopted and, of course, different acquisitions protocols. In particular, several papers adopted only one-time point images, while others adopted outdated protocols of acquisition using fixed times or a fixed count, thus reflecting a wide heterogeneity of approach and results [64,66,67,68,69]. Hybrid imaging with SPECT/CT has also a great role in determining the accuracy of radiolabelled WBC scintigraphy, especially in discriminating superficial STIs from deeper infections. This differentiation is not easy achievable by using only planar images, but it is crucial for the correct management of the patient. Indeed, the primary goal for a correct therapeutic intervention derives from an accurate diagnosis of foot complications and, in particular, from the differentiation between sterile inflammation, STI, OM, and Charcot foot with or without a superimposed infection.In this optic, radiolabelled WBC scintigraphy is the most accurate NM imaging modality able to achieve this differential diagnosis, since it provides an in vivo demonstration of the pathophysiology that underlies inflammatory and infective processes. However, the accuracy of radiolabelled WBC in differentiating OM from STI also depends on the district of the foot [60]. Despite previous considerations may be applied for a correct discrimination between these two conditions in forefoot disorders, in mid- and hindfoot, the presence of Charcot osteoarthropaty may also be considered. In this situation, radiolabelled WBC uptake could also be related to physiological BM expansion secondary to chronic inflammation, thus resulting in a lower specificity of this modality [70,71,72]. Therefore, in order to overcome this limitation and to improve the accuracy of WBC scintigraphy, it is suggested to perform an additional bone marrow scintigraphy (BMS) using nanocolloids. Indeed, both radiopharmaceuticals accumulate in BM but only WBC accumulate in infective foci, so if the images of these two modalities are congruent (match), the diagnosis of Charcot is the most probable; conversely, in case of mismatch (positive at WBC scintigraphy and negative at colloids), the diagnosis of OM may be done.Despite radiolabelled WBC scintigraphy still representing the NM gold standard for the diagnosis of infections, some practical and technical issues, unfortunately, limit its use in all the centers. Indeed, this modality requires qualified personnel, adequate laboratories, and equipment. Moreover, it is a time-consuming procedure for both labeling and images acquisition, since acquisition at three time points is necessary. However, its accuracy has no peers in this field, and the availability of closed and single use kits has simplified the separation and labeling procedures, making all the steps safer for the operator [73].The use of monoclonal antibodies (MoAbs) or antibodies fragments (Fab’) direct against specific antigens expressed by activated granulocytes has been proposed as an alternative to radiolabelled WBC scintigraphy, but they also have several cons mainly related to the high molecular weight of the entire antibodies that constitutes a limiting factor for their diffusion into the infective focus, their long plasma half-life, and their non-specific accumulation into inflamed sites. Furthermore, MoAbs induce human murine antibodies (HAMA) in the host, thus limiting their use at only one time in the life. Moreover, the role of MoAbs or Fab’ fragments has not been extensively investigated in DF, and data in the literature are mainly based on small groups of patients [74,75,76]. Moreover, at the moment, there are no standardized protocols for the acquisition and interpretation, and the few data in literature are not sufficient to conclude that MoAbs or their fragments have to be preferred to radiolabelled WBC scintigraphy in the assessment of DF disorders.In the last decades, [18F]FDG PET/CT has gained an important role also for several indications in the field of infection and inflammation as specifically summarized in the guidelines published in 2013 by EANM and Society of Nuclear Medicine and molecular Imaging (SNMMI) [77].[18F]FDG offers several advantages over conventional scintigraphy. First of all, it avoids the manipulation of potentially infected blood; secondly, the acquisition time is considerably shorter than radiolabelled WBC, and thirdly, the images’ quality resolution is better than those obtained with planar scintigraphy. Moreover, in the presence of CT co-registration, it is possible to have a precise definition of the anatomical landmarks, therefore evaluating the extent of the infective process into soft tissues or bone. However, [18F]FDG accumulates in infections, inflammations, malignancies, reparative processes, and in all the other conditions in which the glucose is metabolized as a source of energy.In a meta-analysis published in 2013, the per-patients-based analysis showed a pooled sensitivity of 74% and a specificity of 91% [78]. Nevertheless, this meta-analysis was conducted only on 4 studies. Another more recent meta-analysis including 6 studies on 254 patients, in which the sensitivity and specificity of [18F]FDG PET/CT were 89% and 92%, respectively [59]. CT co-registration, of course, has a great influence on the accuracy of this imaging modality, but it also relies on correct interpretation criteria for a [18F]FDG PET/CT scan that, unfortunately, are not still well defined and standardized.In a large cohort of 110 diabetic patients with suspected pedal OM, Nawaz et al. compared [18F]FDG PET and MRI. In this series, the first modality was less sensitive (81% versus 91%) but more specific (93% versus 78%) and accurate (90% versus 81%) than the second [79]. In this study, the diagnosis of OM was based on visual assessment of [18F]FDG uptake on bony structures without any semi-quantitative analysis of maximum Standardized Uptake Value (SUVmax). Furthermore, no CT co-registration was performed in this study, which may be influencing the relative low sensitivity compared to MRI.Basu et al. explored the role of semi-quantitative analysis with SUVmax on 63 patients with DF disorders [80]. Patients with OM showed higher SUVmax values than patients with Charcot and uncomplicated DF, thus concluding that SUVmax could be a good parameter for differentiating these conditions. Although these findings were confirmed by other groups, some others did not find any correlation between SUVmax values and the different DF complications [81].So, concluding, at present, well-defined interpretation criteria for differentiating infection, inflammation, STI, OM, and Charcot do not exist yet for [18F]FDG, thus representing a great limiting factor for this specific clinical indication. CT co-registration, although useful for localizing the uptake into bone rather than in soft tissue, does not solve the problem of discriminating an infection from inflammation/degeneration [82] (Figure 5).Aiming to develop a more specific radiopharmaceutical for PET imaging, WBC have also been labeled with [18F]FDG, but published studies on DF still do not exist in literature.During the 3rd European Congress of Infection and Inflammation organized in Rome in December 2019, several specialists evaluating patients with DF complications gave their lectures on this topic from different points of view. Here, we summarize several statements that emerged from the following round table, aiming to provide evidence-based answers to the most frequent clinical questions.Radiography should be the first-line imaging modality when evaluating for bone involvement in the DF. This approach is cheap, widely available, and associated with minimal harm. It provides an anatomic overview of the area of interest and any preexisting conditions that could influence the selection and interpretation of subsequent imaging modalities. Although we are not aware of any studies of the role of serial radiographs to diagnose OM, useful information can be obtained by performing serial radiographs to detect progressive bony changes.From a radiological point of view, a negative radiographic examination is not enough to rule out OM, since it is not sensitive in the detection of early stages of acute OM [10]. Radiographs may remain unremarkable for up to four weeks after the onset of infection. Furthermore, when radiographic changes of OM such as demineralization, bone resorption, and periosteal reaction become detectable, they may be difficult to be correctly interpreted because similar abnormalities may occur with Charcot osteoarthropathy and other disorders such as gout [40]. Therefore, the appeal to advanced imaging is madatory in order to achieve an accurate diagnosis.MRI is not appropriate as first line imaging modality to diagnose OM; however, it is strongly recommended as an additional modality after initial radiography, when OM is suspected. MRI provides excellent spatial resolution and precise anatomical details; it allows preoperative mapping of the extent of infection, thus being helpful in minimizing the area of resection. Moreover, its radiation-free assessment becomes particularly important in the young population and when repeated follow-up imaging is necessary, and it is now widely available and less expensive than other imaging modalities [38].There is no relevant literature to support the use of MRI in the follow-up of DFO. However, this imaging modality can be very appropriate for determining whether the patients healed from the infection after treatment. Given that normal marrow signal reliably excludes OM [42], this condition should not be considered “cured” until there has been no evidence of recurrence for at least a year [83]. The radiation-free assessment as well as the high sensitivity and specificity for determining the presence or absence of pedal OM and STI [44] makes MRI imaging very suitable as a follow-up imaging modality, especially in young people.The main disadvantage of planar NM imaging techniques is the limited spatial resolution and the lack of anatomic landmarks, which is especially a problem in the foot, where all the bony structures are very small and close each other. Indeed, an uptake on the soft tissues at planar images may overlap the underlying bone and vice versa, leading to a wrong interpretation of the scan and consequently to a wrong treatment. Therefore, as previously mentioned, the appeal to hybrid images is mandatory in order to improve the diagnostic accuracy of planar images.Several authors explored the added value of SPECT/CT in the diagnosis [66,67,84,85,86] and therapy monitoring of DFO [87,88] and, despite the different protocols of acquisition adopted among the different studies, all authors concordantly agree that hybrid imaging is able to better localize the uptake into bone or soft tissues with an excellent definition of the extent of the infective process. Przybylski et al. reported a sensitivity, specificity, and diagnostic accuracy of 99mTc WBC scintigraphy with SPECT/CT were 87.5%, 71.4%, and 80% respectively [85]. Heiba et al. examined 272 patients by using a combined approach with 111In WBC scintigraphy and bone scan [66], concluding that dual isotope SPECT/CT was superior than bone scan or WBC scintigraphy with SPECT/CT alone in discriminating STI from OM and, in another paper, they concluded that this combined approach is associated with a reduced length of hospitalization [67]. In the series studied in 2009 by Filippi et al., the interpretation of planar images substantially changed with the addition of SPECT/CT in 52.6% of cases, being able to rule out the infection in 6 cases, to diagnose OM in 1 case and to better define the extent of the process in 3 cases [86].Therefore, concluding, data in the literature support the use of SPECT/CT in addition to planar images in the evaluation of DFI, in order to better localize the infection into bone or soft tissues and to accurately assess the extent of the process.The answer to this specific question is not easy, because it mainly depends on the local center’s equipment and facilities. As previously mentioned, the labeling procedures of leukocytes requires classified environments and laboratories and with isolators or class A wood with laminar flow, depending on local regulations. Personnel must be specifically trained to perform this procedure and must attend certified courses, thus impacting the department’s costs. Moreover, it is a time-consuming procedure that requires multiple times-point acquisition, and therefore, the patients need to come back to the hospital the day after in order to complete the examination. Aiming to overcome these limitations, several authors suggested the use of alternative approaches and [18F]FDG PET/CT, of course, represents the most attractive one due to higher quality images, shorter length of execution, and the easier and quicker length of handling of radioactive compounds, which does not require the manipulation of potentially infected blood. However, the well-known low specificity of [18F]FDG in differentiating an infection from a sterile inflammation and the lack of unanimous consensus on interpretation criteria make the diagnosis uncertain in most cases.One paper published in 2011 by Familiari et al. [81] perfectly fits this question. In a small cohort of 13 patients with suspected OM, they compared [18F]FDG PET/CT and planar images of 99mTc-HMPAO WBC scintigraphy, acquiring both modalities at three times point and using qualitative and semi-quantitative criteria of interpretation with SUVmax and Target/Background ratio (T/B) ratio at each time point. They identified a cutoff of more than 2.0 at late images as the best interpretation criterion for WBC scintigraphy: an increase of this cutoff between 3 and 20 h was suggestive for OM, while a stable or decreased uptake over time was suggestive for STI. Similarly, for [18F]FDG PET/CT, the best criterion for defining an OM was a SUVmax greater than 2.0 at 1 and 2 h p.i. and increasing with time. Whereas, if the uptake remaines stable or decreased over time, the scan was suggestive for STI. By using these criteria, the sensitivity, specificity, and accuracy of radiolabelled WBC scintigraphy were higher compared with [18F]FDG PET/CT. Therefore, they concluded that radiolabelled WBC scintigraphy should not be replaced by [18F]FDG PET/CT.In accordance with this view, we strongly recommend the use of radiolabelled WBC scintigraphy in suspected pedal OM, as it also emerged from a recently published retrospective multicenter study [89]. [18F]FDG PET/CT could represent a valid alternative if it is not possible to perform this imaging modality due to the limitations of the single center, but the interpretation of a PET scan, at the moment, really relies on personal experience, and therefore, the scans must be evaluated with caution.PET provides better resolution images and allows easier quantification methods than SPECT imaging. With radiolabelled WBC imaging, it is only possible to calculate T/B ratios between delayed and late images and to assess whether there is an increase or decrease of this value. With SUVmax evaluation on PET imaging, it is possible to quantify the uptake of by [18F]FDG at the infectious focus. Therefore, several authors tried to assess whether the use of this SUVmax could be beenficial in order to better differentiate between sterile inflammation, STI, OM, and Charcot.For example, Basu et al. found higher SUVmax values in patients with OM compared with patients with Charcot and uncomplicated DF (2.9–6.2 versus 0.7–2.4 versus 0.2–0.7), concluding that SUVmax could be a useful parameter for differentiating these conditions [80]. Kagna at al. found a statistically significant difference in SUVmax between OM and STI (6.7 ± 3.7 versus 4.4 ± 2.4) [90]. However, these results were not confirmed by other studies [79,81], which may be because several techinical and practical factors may influence the variation of SUVmax calculations among different centers. Moreover, universally recognised cutoff values to differentiate among Charcot’s neuropathy, OM, and STI have not been defined yet.Therefore, as it also stands for oncologic diseases, there is currently insufficient evidence to recommend that SUVmax could be a reliable tool for discriminating among different foot complications.The issue that ongoing antibiotic treatment could influence the sensitivity of radiolabelled WBC scintigraphy is still a matter of debate, and the opinions are very contrasting. From some papers, it emerges that the diagnostic accuracy of radiolabelled WBC is not significantly affected by the administration of antibiotics [87,88,91,92]. In 2013, Glaudemans et al. retrospectively studied a large population of patients with prosthetic joint infection, and they did not find significant differences in terms of diagnostic performance between patients under antibiotic treatment and patients that were not receiving therapy. Although this study was not focused on DFI, it was in support of the idea that this imaging modality retains a high sensitivity and specificity in detecting residual disease, independently by the administration of antibiotics. [93]. Indeed, as also indicated in recently published EANM guidelines [55], “patients receiving antibiotic treatment should not be excluded a priori since reports regarding their effect on WBC scintigraphy give various results”.However, not all NM physicians place very much trust in performing this exam during antibiotic treatment, because possible false negative scans may be observed. Therefore, despite a perfect timing to perform WBCs scintigraphy following antimicrobial therapy not being clearly indicated in the literature, it is often a common practice to delay the radiolabelled WBC scintigraphy until 2 weeks after therapy withdrawal or to repeat the scan, in case of doubts in patients receiving antibiotics 2 weeks later.Data in the literature on therapy monitoring in DF are mainly based on small series and do not allow drawing definite conclusions, but preliminary results seem to encourage the use of radiolabelled WBC scintigraphy, especially with SPECT/CT acquisitions, for the assessment of treatment response [87,88]. Similarly, [18F]FDG PET/CT could be used to follow signs of inflammation in the foot that may be still present, although the patient is considered clinically recovered [94], but definitive evidences are still lacking in the literature.Charcot osteoarthropathy is a condition that further complicates the challenging diagnosis of DFI. Radiolabelled WBC uptake in mid/hind-foot must be always interpreted with caution considering the possible physiologic accumulation into expanded BM that is typically present in a Charcot foot, independently by the presence of an infection or not. Therefore, a BMS is strongly suggested in order to have a scintigraphic map of BM and to compare with WBC images. Palestro described two criteria to diagnose OM in the presence of Charcot’s arthropathy: (1) the presence of labeled leukocyte uptake without corresponding activity on marrow images and (2) the spatially incongruent distribution of two radiopharmaceuticals [70,71].[18F]FDG also shows several limitations in the evaluation of Charcot, because the uptake in this condition is usually very intense and diffuse involving all the tarsal and metatarsal joints, reflecting the evident changes in bony architecture typical of this conditions. Therefore [18F], FDG is not able to discriminate whether Charcot is infected or not.An accurate identification and differentiation among different types of DFI still represent a challenge for the clinician. The appeal to multimodality imaging and a multidisciplinary approach are mandatory in order to plan the most appropriate therapeutic strategy for the single patient. Several radiological and NM approaches are available, being MRI, radiolabelled WBC scintigraphy, and [18F]FDG PET/CT the most appropriate, but larger multicenter studies are still needed in order to create standardized diagnostic flow charts that could be applied worldwide.Conceptualization, C.L.; methodology, C.L. and A.S.; resources, C.L., A.L., M.C., A.S., L.G., L.U.; writing—original draft preparation, C.L., A.L., M.C., A.S., L.G., L.U.; writing—review and editing, C.L.; supervision, A.S. All authors have read and agreed to the published version of the manuscript.This research received no external funding.The authors declare no conflict of interest.Extended plantar phlegmon. Left: With a probe, it is possible to follow the real spaces produced by the phlegmon spread along tissue plans. Where the end of the tract becomes superficial toward the skin, interposed tissues and the skin are pierced and incised in order to pass through the probe. Middle: A silastic tube is, thereafter, anchored to the probe in order to pass it backward along the fistula tract. Once this drainage is passed, the two ends are tied together with two silk stitches in order to construct the ulcer piercing ring (UP ring). Right: A diabetic foot ulceration (DFU) completely healed after 8 months in an out patient facility with daily medications and irrigations and with occasional antibiotic therapy and resulting with a small plantar scar. (Courtesy of Marco Cavallini).OM of the scaphoid and medial cuneiform. (From left to right) Sagittal T1-weighted, T2-fat-suppressed, and post-contrast T1-weighted fat-suppressed MRI clearly show that the marrow in the scaphoid and medial cuneiform has a low signal on a T1-weighted image (circle in left panel), increased signal on a fluid-sensitive fat-suppressed image (circle in middle panel), and post-gadolinium enhancement (circle in panel). These findings are indicative of OM.Forefoot ulcer, sinus track, and abscess associated with OM in a 57-year-old diabetic man with a 16-year history of insulin-dependent diabetes. Sagittal T2 fat-suppressed (left panel), and post-contrast T1-weighted fat-suppressed MRI (right panel) shows a dorsal thick rim-enhancing abscess adjacent to the first metatarsal head (arrows). Note a plantar ulcer appearing as a focal skin interruption and a sinus tract with rim-like enhancement (small arrow in right panel) extending near the first proximal interphalangeal joint. Given these findings, the hyperintensity (** in left panel), and post-contrast enhancement (** in right panel) in the first metatarsal, and proximal phalanx respectively, are indicative of OM.Example of 99mTc-HMPAO WBC scintigraphy in a patient with a skin ulcer in the medial right malleolus region, previous amputations of left leg, and of all metatarsal heads of the right foot. From left to right: Planar images acquired after 30 min, 3 h, and 20 h p.i. of radiolabelled autologous leukocytes show an increased amount of activity over time in terms of intensity and extent, being consistent with OM with an involvement of adjacent soft tissues. Further SPECT/CT acquisition (right panel) correctly localized the uptake in the right talus and accurately evaluated its extent into surrounding soft tissues. HMPAO: hexamethylpropylene amine oxine, WBC: white blood cell.Example of [18F]FDG PET/CT in a patient with Charcot osteoarthropathy. Fused images (upper panels) show moderate and diffuse uptake, which is interesting in particular bones and joints of the mid and hind-foot. Co-registered low-dose CT images (lower panels) show the evident destruction of bony architecture. These findings are consistent with the diagnosis of Charcot foot, but they do not allow discriminating a pan-inflammation from a possible superimposed infection.Common interpretation criteria of different imaging modalities in diabetic foot infections.BMS: bone marrow scintigraphy; CT: computer tomography; OM: osteomyelitis; STI: soft tissue infection; DFI: diabetic foot infection; WBC: white blood cells; BM: bone marrow; BMS: bone marrow scan; SPECT: single photon emission computed tomography; T1-w: T1-weighted; FN: false negative; FP: false positive.

Med-MDPI/jcm_2/jcm-09-06-01780.txt

@@ -0,0 +1 @@
+This study aims to evaluate clinical and ulcer characteristics as well the outcomes of patients with diabetic foot ulcers (DFUs). The study group was composed of DFUs patients managed by a limb salvage protocol according to guidance. Clinical and ulcers findings were described, and 1-year outcomes defined as limb salvage, healing, healing time, major amputation and death were compared between neuropathic and ischemic DFUs. One thousand, one hundred and ninety-eight subjects were included; 386 (32.2%) neuropathic and 812 (67.8%) ischemic DFUs. Neuropathic patients were younger (69.5 ± 11.5 vs. 74.5 ± 11.5, p < 0.0001) and reported less cases of nephropathy (22.8 vs. 39.6%, p < 0.0001), ischemic heart disease (22.8 vs. 36.9, p = 0.0004), cerebrovascular disease (8.3 vs. 17.2%, p = 0.002), heart failure (10.1 vs. 24.7%, p = 0.0002) and end-stage-renal-disease (ESRD) (5.4 vs. 27%, p = 0.0001) than ischemic patients; they also showed less cases of large (>5 cm2) (10.3 vs. 22.9%, p = 0.0007), infected (40.4 vs. 55.7%, p = 0.0005) and deep to the bone (22.3 vs. 39.2, p = 0.0002) ulcers, as well less multiple ulcerations (21.8 vs. 32.8%, p = 0.006) than patients with ischemic DFUs. The outcomes for neuropathic and ischemic DFUs were limb salvage (98.4 vs. 82.3%, p < 0.0001), healing (97.3 vs. 79.6%, p < 0.0001), healing time (34.9 vs. 35.6 weeks, p = 0.8), major amputation (0.5 vs. 6.6%, p = 0.0001), death (1.1 vs. 11%, p < 0.0001) respectively. Revascularization failure and ESRD were independent predictors of major amputation, while heart failure and number of co-morbidities (≥5) were independent predictors of death. Ischemic DFUs patients showed more severe clinical and ulcers features as well worse outcomes than neuropathic DFUs patients.Diabetic foot disease (DFD) is the most severe consequence of two diabetes related long-term complications: peripheral neuropathy (PN) and peripheral arterial disease (PAD). Foot ulceration is usually the main clinical expression of DFD [1]. Diabetic foot ulcers (DFUs) affect up to 15% of the diabetic population at some time in their life and they represent the first cause of hospitalization, minor and major amputation among diabetic subjects [2,3].Patients with DFD are often very fragile and foot ulceration may be just a part of an extremely complex clinical condition in which specific long-term complications (PN and PAD) and concomitant co-diseases affect the general health of patients. A 5-year mortality rate was reported following new-onset of DFU, which is between 25% and 60% [4,5] higher than several types of cancers. Moreover, cardiovascular and renal disease are the main causes of death. Diabetic foot (DF) patients have reported a complex interplay of several inflammatory markers which can affect cardio-vascular system and DF influence a faster progression of cardio-vascular damage and morbidity [6,7]. Therefore, not only does DFD require early management of foot ulceration but also the assessment of all comorbidities that may influence the outcomes.Within this framework, it is necessary to consider two patterns of DFUs in patients with or without peripheral arterial disease (PAD), termed respectively as neuro-ischemic/ischemic ulcers and neuropathic ulcers. It has been reported that until now, in developed countries, the rate of PAD in patients with foot ulceration is approximately 50% [8,9], while neuropathic ulcers are more prevalent in low income countries [10,11].PAD increases the risk of non-healing and major amputation [12,13,14], and it is associated with an increased risk of concomitant cardiovascular disease, as well as ischemic heart disease, cerebrovascular disease, and chronic kidney disease, and a high risk of mortality [15,16]. Therefore, it is essential that PAD in patients with DFUs is recognized early and managed accordingly.Based on their daily experience, the authors retain that a deep understanding of the patients DFU history, management and outcomes could lead to an improvement in our strategies.This study aims to evaluate the pattern of diabetes-related complications and co-morbidities in patients with DFUs, comparing neuropathic and ischemic patients. Furthermore, the characteristics of neuropathic and ischemic/neuro-ischemic DFUs will be reported and compared, as well as the long-term outcomes.Consecutive patients who were referred to our diabetic foot unit for a new diabetic foot problem between January 2010 and December 2018 were considered for this study. Patients included were those attending the clinic for a new foot ulceration, including both neuropathic and ischemic/neuro-ischemic DFUs. Subjects referred with an unsalvageable foot condition requiring major amputation, those with a life expectancy of less than 6 months and those who lost to follow-up during the first 12 months were excluded.All patients included were managed through a pre-set limb salvage protocol including revascularization in the case of ischemic/neuro-ischemic ulcers, antibiotic therapy and surgical treatment for infected wounds, dedicated off-loading, appropriate wound care, and management of diabetes and comorbidities according to guidance [17,18].Data were collected in a local database and retrospectively analyzed. Baseline demographic, clinical and ulcer findings were recorded.The study has been done and approved according to local ethics committee policy. At admission, patients provided their verbal consent.Diabetic retinopathy, neuropathy, and nephropathy were conditions reported. Retinopathy was considered in the case of proliferative or not proliferative retinopathy; peripheral neuropathy was considered in the case of loss of peripheral sensitivity detected through vibration perception (128 Hz tuning fork) or Semmes-Weinstein 10-g monofilament [17,18]; nephropathy was considered in the case of albuminuria, both micro (30–300 mcg/mg creatinine) and macroalbuminuria (>300 mcg/mg creatinine) [19].Ischemic heart disease (IHD) was considered in the case of previous acute coronary syndrome or coronary revascularization, evidence of angina, significant changes on electrocardiography (above or under-leveling ST, q wave, inversion of T wave, new left bundle branch block). Cerebrovascular disease was considered in the case of previous cerebrovascular ischemia, previous carotid revascularization, or significant carotid artery disease (occlusion >70%).Hypertension was considered in the case of blood pressure >130/80 mmHg persistently or current antihypertensive therapy [19]; hypercholesterolemia was defined as low density lipoproteins (LDL) > 70 mg/dL or needing statin therapy [19]; heart failure (HF) was considered in the case of typical symptoms and signs of HF reduced left ventricular ejection fraction (LVEF) (<40%) or normal or only mildly reduced LVEF and elevated levels of brain natriuretic peptides (BNP > 35 pg/mL and/or NT-proBNP > 125 pg/mL) without dilated left ventricle (LV), associated with relevant structural heart disease (LV hypertrophy/left atrial enlargement) and/or diastolic dysfunction [20]. End-stage-renal-disease (ESRD) requiring dialysis was considered in the case of chronic renal replacement therapy. Anemia was considered according to hemoglobin values at the first assessment. Patients were only considered smokers if they had a smoking habit at the time of treatment.Baseline ulcer characteristics (location, size, depth, infection) reported at the first assessment were recorded. Infection was considered in the case of clinical signs according to International Working Group on the Diabetic Foot (IWGDF) [17,18]. Ulcer was considered deep to the bone in the case of bone exposure or positive probe-to-bone test. The site of ulcer location was characterized as forefoot, midfoot and rearfoot; in the presence of more than one ulcer, it was considered multiple location.Neuropathic ulcers were considered in the case of patients with PN without PAD; ischemic ulcers were considered in the case of patients with PAD, regardless of the presence or not of PN. PAD was considered in the case of the absent pulses and ankle-brachial index of <0.9 or TcPO2 < 50 mmHg, in addition to evident stenosis and/or an obstruction at duplex ultra-sound [18,19,20].Micro and macrovascular diabetes-related complications, co-morbidities, ulcers characteristics and outcomes in patients with neuropathic and ischemic DFUs were reported and compared.Limb salvage, healing, healing time, amputation, and mortality after 1-year of follow-up were the primary outcomes considered. Limb salvage was considered in the case of healing or incomplete healing in patients with preserved limb function; healing was considered in the case of complete epithelialization of previous ulceration during the follow-up; healing time was considered as the time reported in weeks which occurred from the first assessment and the complete epithelialization; amputation was considered as any amputation above-the-ankle and included below and above-the-knee.As secondary endpoint, the association between the number of comorbidities and limb salvage in the whole population of neuropathic and ischemic subjects was reported.All potential predictors of major amputation and death where evaluated.Statistical analysis was performed by SAS (JMP12; SAS Institute, Cary, NC, USA) for personal computer. Data are expressed as means ± SD. Comparison between groups were reported by the Student’s t test (frequency data) or ANOVA (continuous data). Univariable logistic regression analysis was performed with all potential predictor variable with the outcome of interest (major amputation and death). Then, all positive predictors were entered simultaneously in a multivariate logistic regression model. These models yielded a set of variables that best predict the outcome of interest. p < 0.5 was considered statistically significant.One thousand, one hundred and ninety-eight patients were included. The mean age was 73 ± 12 years; 758/1198 (63.3%) were males, 1130/1198 (94.3%) had type 2 diabetes and the mean HbA1c was 62 ± 24 mmol/mol (Table 1).Three hundred eighty-seven (387) (33.2%) patients had neuropathic DFUs while 812 (67.8%) patients had ischemic/neuro-ischemic DFUs. Ischemic DF patients were older, had a longer duration of diabetes and higher baseline HbA1c values when compared to neuropathic DF subjects (Table 1).They were characterized by the high presence of long-term diabetes-related complications, mainly peripheral neuropathy (92%), retinopathy (51%), nephropathy (34%), PAD (68%) and ischemic heart disease (32%). Regarding concomitant co-morbidities, they frequently reported hypertension (79%), dyslipidemia (40%), and often ESRD requiring dialysis (22%) and heart failure (20%) (Table 2).Ischemic DF patients had more diabetic nephropathy and less peripheral reduced sensitivity than neuropathic DF subjects (Table 2).Ischemic DF patients showed a higher rate of ischemic heart disease and cerebrovascular disease than neuropathic subjects (Table 2).Ischemic DF patients showed more cases of HF, ESRD and anemia than neuropathic persons. No difference was recorded in terms of hypertension, dyslipidemia and smoking between the two groups (Table 2).Overall, 16.5% of patients reported 4 co-morbidities concurring with diabetes and 10.8% had 5 or more. More patients with ischemic DF presented with 4 or more concomitant co-morbidities than patients with neuropathic DF (Table 3).Ischemic DF patients showed more cases of multiple lesions in comparison to neuropathic DF patients, while neuropathic DF subjects showed more cases of forefoot localization than ischemic DF patients. Ischemic DFUs were larger, more infected and deep to the bone in more cases than neuropathic DFUs (Table 4).One thousand and fifty (1050) (87.7%) patients had limb salvage after 1-year of follow-up, 1022 (85.3%) patients healed in an average time of 35.4 weeks (range 31.9–39.1), 55 (4.6%) were amputated (major amputation), and 93 (7.7%) died (Figure 1).The outcomes for neuropathic and ischemic patients were, respectively: limb salvage (98.4 vs. 82.4%, p < 0.0001), healing (97.3 vs. 79.6%, p < 0.0001), average healing time (34.9 vs. 35.6 weeks, p = 0.8), amputation (0.5 vs. 6.6%, p = 0.0001), and death (1.1%vs. 11%, p < 0.0001) (Figure 1).In the whole population, the rate of limb salvage gradually decreased as the number of concomitant co-morbidities increased; furthermore, the rate of limb salvage in persons with ischemia was less than those with neuropathy, despite having a similar number of coexisting diseases (Table 5).At the multivariate analysis of all predictors found at univariate analysis, revascularization failure and ESRD were independent predictors of major amputation, while heart failure and the presence of five or more concomitant co-diseases were independent predictors of mortality (Table 6).This study offers a complete overview on ulceration findings, clinical characteristics, and long-term outcomes in a large cohort of diabetic foot patients including those with both neuropathic and neuro-ischemic/ischemic DFUs.Overall, 87.7% had limb salvage and 85.3% of patients healed in an average time of 35.4 weeks, while 4.6% had major amputation and 7.7% died.On the one hand, the reported data are very encouraging because through a specialized multi-disciplinary team approach, we achieved a great rate of limb salvage and healing, mainly in neuropathic subjects. On the other hand, we observed a significant increase in ischemic DFUs, which were approximately two times higher than neuropathic DFUs and, among those reported, showed a higher rate of amputation (6.6 vs. 0.5%) and death (11 vs. 1.1%), and a lower rate of healing (79.6 vs. 97.3%) in comparison to neuropathic subjects.These data are partially similar to those reported by the Eurodiale study [12] which showed that ischemic patients had a higher rate of non-healing in comparison to neuropathic patients; the same Eurodiale study described ischemic DFUs as larger, deeper and more infected than neuropathic DFUs, as we found in this current cohort. In addition, we reported that in most cases of ischemic DF, there was evidence of heel ulceration and multiple ulcers, while in neuropathic patients, there was higher involvement of the forefoot than in ischemic patients.The ulcer size and the presence of infection could increase the risk of non-healing [21]. Additionally, infection is a predictor of amputation [22] and mortality in frail patients with PAD [23]; furthermore, heel ulcers could influence both non-healing and amputation [24,25].Patients included in our study group were elderly (mean age >70 years), with a long diabetes duration (approximately 24 years) and poor glycemic control. They were characterized by the high presence of microvascular and macrovascular complication, and several concomitant co-morbidities, such as hypertension, dyslipidemia, ESRD and heart failure.Patients with ischemic ulcers were older than those with neuropathic ulcers and showed higher rates of microvascular (40% with nephropathy) and macrovascular (37% with ischemic heart disease and 17% with carotid artery disease) complications, and concomitant co-morbidities (82% with hypertension, 27% on dialysis, 25% with heart failure and 24% with anemia) in comparison to patients with neuropathic ulcers.In the whole population, approximately 16% of patients reported at least four concomitant co-morbidities, and 10.8% had five or more. Patients with ischemic DF were more likely to present with four or more concomitant co-morbidities than patients with neuropathic DF (32 vs. 16%). It is also very interesting to highlight the rate of limb salvage decreases with the increase in the number of co-morbidities. However, the limb salvage rate was 100% and 90%, respectively, in subjects without and with three concomitant co-diseases or fewer. Nonetheless, the rate of limb salvage is significantly lower in those with ischemic DF than neuropathic DF despite the similar number of co-morbidities. These data are already evident in the presence of one co-disease (100% of limb salvage in neuropathic patients and 88% in ischemic patients) and much more evident in the presence of three or more co-diseases, suggesting that co-morbidities in patients with PAD reduce the possibility of limb salvage, which is probably due to the impact of PAD perse.The role of co-morbidities is reinforced by the multivariate analysis which revealed that ESRD was an independent predictor of major amputation and heart failure, and, additionally, the number of concomitant co-diseases (≥5) were independent predictor of mortality.The role of co-morbidities in DFD is often underestimated, even though it is well known that they can significantly influence outcomes for patients with DFUs. It is maintained that ulcer-related outcomes may underestimate the morbidity and mortality associated with DFD [26], and recently clinicians are more often focused on the patient’s comorbidities in the management of DFD.Our data highlight that DF should be considered as a marker of multi-organ disease in ischemic subjects, with a significant impact on outcomes. Many papers have reported that co-morbidities, mainly dialysis and heart failure, increase the risk of major amputation and outcomes [23,27].Patients with DFUs, mainly ischemic, are extremely difficult to treat, and foot injury is often just a part of a very complex clinical condition. Ischemic patients often require hospitalization to be treated for revascularization and due to the presence of severe co-morbidities, in some cases, a fast limb salvage protocol may not only address limb salvage, but may also save the patient’s life.Similarly, the Eurodiale study showed that heart failure and ERSD had a greater incidence in patients with PAD, and in addition, ESRD was an independent predictor of non-healing [12].Gershater et al., in a prospective study on 1148 hundred and eighty patients, reported that the absence of uremia and heart disease were clinical factors related to primary healing in the whole population and in survivors. Conversely, diabetic nephropathy and uremia were predictors of non-healing in survivor patients with ischemic/neuro-ischemic ulcers, and uremia was related to major amputation in ischemic ulcers. Furthermore, deceased patients showed more ischemic ulcers and more co-morbidities than the other groups [28].Alpeqvist et al. showed that Creatinine values <130 μmol/L and the absence of congestive heart failure were independent predictors of primary healing in a population composed of 1115 patients with ischemic DFUs [29].Faglia et al. showed that ischemic heart disease was the leading cause of death; dialysis and a history of cardiac disease were independent predictors of death; dialysis was an independent predictor of major amputation in 554 patients with DFUs and CLI when treated using a limb salvage protocol, which includes revascularization [23].In our previous experience, we observed that dialyzed patients had higher risk of non-healing and major amputation than subjects with preserved renal function [30], and that patients with ischemic DFUs affected by heart failure and dialysis had a very high risk of one-year mortality (56%) [23].Therefore, co-morbidities, such as heart disease, including both coronary artery disease and heart failure, and ESRD not only reduce the chance of healing, but they are also independent predictors of mortality.In the current study, we confirm that ischemic heart disease is quite typical in diabetic foot patients with PAD. It has already been reported that approximately 50% of diabetic patients with PAD have a concomitant ischemic heart disease [31]. Heart impairment is common in patients with DFUs, even in asymptomatic persons. Londhal et al. reported that patients with chronic DFUs, in 69% of cases, presented with myocardial infarction and/or hypertension and/or heart failure; in 78% left ventricular dysfunction and/or hypertrophy and/or diastolic dysfunction, and in 76% echocardiographic signs of heart dysfunction without any previous history of cardiovascular disease [32].Therefore, patients with DFUs, mainly ischemic DFUs, should be considered as subjects with a high risk of heart disease—mainly heart failure secondary to ischemic heart disease.It is noteworthy that micro/macrovascular diabetes-related complications were more frequent in patients with ischemic DF than neuropathic DF, although a similar diabetes duration was reported. Therefore, it may be evident that age, concomitant co-morbidities (e.g., hypertension and renal impairment), poor glycemic control and individual susceptibility could increase the risk of developing PAD and the abovementioned complications.It is also necessary to highlight that revascularization failure was an independent predictor of major amputation. Revascularization failure was considered as technical recanalization failure of occluded vessels (defined as the impossibility to overcome the obstruction) and/or absence of arterial flow to the foot. On the one hand this data suggests that the severity of PAD could negatively influence the revascularization procedure and outcome, but on the other hand also confirms that failed revascularization is a predictor of major amputation as already reported by Faglia et al. [27] and our research group in a recent study [33].Therefore, DFD is a complex clinical condition characterized by foot injury, severe patterns of PAD, and several concomitant co-morbidities which could influence management and outcomes. The presence of a foot ulceration in diabetics should always be considered a strong risk factor for early and long-term mortality, mainly in ischemic subjects which have shown 5-year-mortality, of approximately 60% [34]. It is evident that persons with DFUs are often fragile and the onset of foot ulceration could worsen their frail health status.This study gives a complete overview on clinical and ulcers features, and the long-term outcomes in a very large cohort of patients with DFUs, which proves to identify specific characteristics in the current pattern of neuropathic and ischemic subjects. To the best of our knowledge it is the first study after the Eurodiale study to evaluate and compare the prevalence, characteristics and outcomes of neuropathic and ischemic DF patients in a very large cohort of patients; furthermore, it is the first to highlight that the number of comorbidities could influence the possibility of limb salvage, in addition to the fact that concomitant co-diseases may have a major influence on ischemic rather than neuropathic DFUs.The current study is a retrospective study and data were collected from one single diabetic foot center and, accordingly, outcomes are related to our comprehensive limb salvage protocol, performed by an expert multidisciplinary diabetic foot team. Future research may be useful to identify if outcomes are influenced only by the number of comorbidities or by a comorbidity score, whereby each concomitant disease has a specific burden.Our data illustrate that among current patients affected by DFD, there is a prevalence of ischemic DFUs in comparison to neuropathic DFUs, and there also seems to have been an increase in ischemic subjects over the last years. Patients with DFUs are complex subjects, who in addition to foot injury, are often older in age and display several diabetes-related complications and comorbidities—mainly cardiovascular. This study reinforces the concept that patients with ischemic DFUs report more severe wound and clinical features than those with neuropathic DFUs, and that there is a lower rate of limb salvage in spite of the similar number of concomitant co-diseases. Furthermore, co-morbidities appear to play a key role in the outcomes of patients with DFD.While aiming to reduce amputation and mortality, DFD should be considered a multi-organ disease, which means that patients need an intensive global and multidisciplinary treatment plan with close control of cardiovascular risk factors.Conceptualization, M.M. and L.U.; methodology, M.M. and L.U.; software, M.M. and L.U.; validation, M.M., V.I., L.G., J.L.L.-M. and L.U.; formal analysis, M.M.; investigation, M.M. and V.I.; resources, M.M. and V.I.; data curation, M.M.; writing—original draft preparation, M.M.; writing—review and editing, L.U. and J.L.L.-M.; visualization, M.M., V.I., L.G., J.L.L.-M.; supervision, J.L.L.-M.; project administration, M.M. and L.U. All authors have read and agreed to the published version of the manuscript.This research received no external funding.The authors declare no conflict of interest.Outcome of all subjects, neuropathic and ischemic/neuro-ischemic patients.Demographic and diabetes-related characteristic of all subjects, neuropathic and ischemic foot patients.Microvascular and Macrovascular related complications, and concomitant comorbidities.Microvasc.: microvascular; macrovasc.: macrovascular; ESRD: end-stage-renal-disease.Rate of co-diseases in all subjects, neuropathic and ischemic DF groups.Ulcer baseline characteristics in all subjects, neuropathic and ischemic DF patients.Rate of limb salvage according to the number of co-disease in all subjects, neuropathic and ischemic DF groups.Multivariate analysis of independent predictors of outcome (major amputation and death) found at univariate analysis.ESRD: end-stage-renal-disease; IHD: ischemic heart disease.