title
stringlengths
1
1.19k
keywords
stringlengths
0
668
concept
stringlengths
0
909
paragraph
stringlengths
0
61.8k
PMID
stringlengths
10
11
Main Outcomes and Measures
depressive symptoms
ADVERSE EFFECTS
The primary outcome was days until attainment of a therapeutic TCA plasma concentration. Secondary outcomes were severity of depressive symptoms (measured by HAMD-17 scores) and frequency and severity of adverse effects (measured by Frequency, Intensity, and Burden of Side Effects Rating scores).
PMC10167565
Results
Of 125 patients randomized, 111 (mean [SD] age, 41.7 [13.3] years; 69 [62.2%] female) were included in the analysis; of those, 56 were in the PIT group and 55 were in the control group. The PIT group reached therapeutic concentrations faster than the control group (mean [SD], 17.3 [11.2] vs 22.0 [10.2] days; Kaplan-Meier
PMC10167565
Conclusions and Relevance
MDD, depressive
ADVERSE EFFECTS
In this randomized clinical trial, PIT resulted in faster attainment of therapeutic TCA concentrations, with potentially fewer and less severe adverse effects. No effect on depressive symptoms was observed. These findings indicate that pharmacogenetics-informed dosing of TCAs can be safely applied and may be useful in personalizing treatment for patients with MDD.
PMC10167565
Trial Registration
ClinicalTrials.gov Identifier: This randomized clinical trial examines whether tricyclic antidepressant (TCA) dosing based on pharmacogenetics according to the Dutch Pharmacogenetics Working Group guidelines results in faster attainment of therapeutic TCA plasma concentrations compared with usual treatment.
PMC10167565
Introduction
ADVERSE EFFECTS
Pharmacogenetics has the potential to personalize antidepressant treatment, yet implementation in psychiatry is still very limited.Antidepressants are metabolized by various isoforms of cytochrome P450 (CYP450) enzymes, notably the CYP450 2D6 (CYP2D6) and CYP450 2C19 (CYP2C19) isozymes.Currently, several guidelines are available for optimizing pharmacotherapy by pharmacogenetic testing, among which the guidelines by the Dutch Pharmacogenetics Working Group (DPWG) and the Clinical Pharmacogenetics Implementation Consortium are internationally well recognized.In the present study, we aimed to examine whether dosing based on pharmacogenetics according to the DPWG guidelines resulted in faster attainment of therapeutic TCA plasma concentrations compared with usual treatment. Furthermore, we investigated whether faster attainment of therapeutic concentrations was associated with higher effectiveness and fewer and less severe adverse effects. We hypothesized that application of the DPWG guidelines would result in faster attainment of therapeutic plasma concentrations, higher effectiveness, and a lower rate of adverse effects.
PMC10167565
Methods
PMC10167565
Study Design
The Pharmacogenetics for Individualized Tricyclic Antidepressant (PITA) dosing study was a multicenter RCT in which patients were enrolled between June 1, 2018, and January 1, 2022. The trial protocol is provided in The participating institutions were hospitals and mental health care institutions in the Netherlands (eTable 1 in
PMC10167565
Flowchart of the PITA (Pharmacogenetics for Individualized Tricyclic Antidepressant [TCA]) Study
The reference group consisted of nonrandomized patients with a cytochrome P450 2D6 isozyme (CYP2D6) normal metabolizer phenotype receiving usual treatment. PIT indicates pharmacogenetics-informed treatment; TAU, treatment as usual.
PMC10167565
Participants
unipolar nonpsychotic MDD
Patients were enrolled by their treating psychiatrist. They had a primary diagnosis of unipolar nonpsychotic MDD according to
PMC10167565
Randomization
With the use of stratified block randomization, patients were randomized (1:1) to the PIT or control group by a staff member not involved in this study. The stratification was performed in the following order: (1) CYP2D6 metabolizer phenotype (and CYP2C19 metabolizer phenotype for those receiving imipramine), categorized as poor metabolizer (PM), intermediate metabolizer (IM), normal metabolizer (NM), and ultrarapid metabolizer (UM); (2) prescribed drug (nortriptyline, clomipramine, or imipramine); and (3) clinical setting (inpatient or outpatient), except for patients treated with imipramine who had a deviant metabolizer phenotype (PM, IM, or UM) for both CYP2D6 and CYP2C19 (eTable 3 in In our original design, only patients with aberrant metabolizer phenotypes (PM, IM, or UM) were randomized to the PIT or control group, and patients with a CYP2D6 NM phenotype were assigned to the reference group. This reference group received the same treatment as the control group. After acceptance of a protocol amendment on November 25, 2019, all patients were randomized to the PIT or control group.
PMC10167565
Intervention
Patients randomized to the PIT group received an initial TCA dosage according to the DPWG guidelines (eTables 2 and 3 in
PMC10167565
Outcome Measures
depression, depressive symptoms
ADVERSE EFFECTS
The primary outcome was time (in days) to therapeutic TCA plasma concentrations. Secondary outcomes were severity of depressive symptoms and frequency and severity of adverse effects. Severity of depressive symptoms was measured weekly through the HAMD-17 (score range, 0-52, with higher scores indicating greater depression severity)
PMC10167565
Sample Size Calculation
SECONDARY
To our knowledge, no previous studies have assessed time to attainment of a therapeutic TCA plasma concentration by pharmacogenetics-informed dosing according to DPWG guidelines. We assumed that 50% of the control group would reach a therapeutic plasma concentration within 4 weeks and that 50% of the PIT group would reach a therapeutic concentration within 2 weeks. Taking α = .05 and a power of 80% (2-sided log-rank test), a sample size of 44 patients per treatment group was required. For the secondary end points, we needed 63 patients per group (independent
PMC10167565
Statistical Analysis
depression
ADVERSE EFFECTS
All analyses were performed according to a modified intention-to-treat principle, meaning that patients were included if at least 1 TCA dose was administered. The primary analysis was presented in a Kaplan-Meier survival curve and conducted using a 2-sided log-rank test. In case no therapeutic plasma concentration was reached during the study, censoring was applied. We performed subgroup analyses per antidepressant (nortriptyline, clomipramine, and imipramine). Secondary outcome measures were analyzed through linear mixed-model analyses using a linear time trend (weeks of treatment) and the interaction term between study group (PIT or control) and time as an independent variable. We examined the interaction between treatment group and time for both depression severity (HAMD-17 score) and frequency and severity of adverse effects. Independent
PMC10167565
Results
PMC10167565
Baseline Characteristics
depression, depressive symptoms
Among 171 included patients, 125 patients (73.1%) were randomized to the PIT (n = 63) or control (n = 62) group. A total of 14 randomized patients (11.2%) were excluded from the analysis because TCA treatment was not initiated due to early improvement of depressive symptoms or violation of the study protocol (At baseline, there were no differences between treatment groups with regard to sex, age, depression characteristics, CYP2D6 and CYP2C19 phenotype distribution, and type of TCA (
PMC10167565
Baseline Characteristics of Patients
depression, Depression
Abbreviations: CYP2C19, cytochrome P450 2C19 isozyme; CYP2D6, cytochrome P450 2D6 isozyme; HAMD-17, 17-item Hamilton Rating Scale for Depression; IM, intermediate metabolizer; NM, normal metabolizer; PIT, pharmacogenetics-informed treatment; PM, poor metabolizer; TCA, tricyclic antidepressant; UM, ultrarapid metabolizer.Participants in the PIT and usual treatment groups were compared using χScore range, 0-52, with higher scores indicating greater depression severity.
PMC10167565
Attainment of Therapeutic Plasma Concentrations
A total of 47 patients (83.9%) in the PIT group and 45 patients (81.8%) in the control group attained a therapeutic concentration. The PIT group reached therapeutic concentrations significantly faster than the control group (mean [SD], 17.3 [11.2] days vs 22.0 [10.2] days; Kaplan-Meier χ
PMC10167565
Survival Curves for Time to Therapeutic Plasma Concentrations of Tricyclic Antidepressants Overall
Vertical bars on survival curves represent times of censoring. PIT indicates pharmacogenetics-informed treatment; TAU, treatment as usual.Post hoc analyses demonstrated that the effect was mainly found for nortriptyline (χ
PMC10167565
Survival Curves for Time to Therapeutic Plasma Concentrations of Nortriptyline and Clomipramine
Vertical bars on survival curves represent times of censoring. PIT indicates pharmacogenetics-informed treatment; TAU, treatment as usual.
PMC10167565
Effects on Depressive Symptoms
depression
At baseline, depression severity (measured by HAMD-17 score) was similar between the PIT and control groups (
PMC10167565
Treatment Effectiveness and Severity of Adverse Effects
Depression, HAMD-17
ADVERSE EFFECTS
Treatment effectiveness was measured by the 17-item Hamilton Rating Scale for Depression (HAMD-17), and severity of adverse effects was measured by item 2 of the Frequency, Intensity, and Burden of Side Effects Rating (FIBSER). Whiskers represent SEs. PIT indicates pharmacogenetics-informed treatment; TAU, treatment as usual.
PMC10167565
Adverse Effects
ADVERSE EFFECTS, INTERACTIONS
The interaction between severity of adverse effects and time was significantly different between the PIT and control groups (Interactions between the frequency of adverse effects (FIBSER item 1) and time (
PMC10167565
Reference Group
In the reference group consisting of 46 nonrandomized patients with a CYP2D6 NM phenotype receiving usual treatment, 9 patients did not initiate TCAs; therefore, analyses were conducted for 37 patients. Baseline characteristics are shown in eTable 5 in
PMC10167565
Discussion
PMC10167565
Main Findings
depressive symptoms
ADVERSE EFFECTS, SECONDARY
In this first RCT to date comparing PIT with standard treatment, we found that PIT resulted in faster attainment of therapeutic TCA plasma concentrations without exposing patients to more severe adverse effects. Our analyses indicated a mean reduction of 5 days in time to attainment of therapeutic concentrations compared with usual treatment. This effect was primarily due to faster attainment of therapeutic nortriptyline concentrations, which showed a median reduction of 6 days compared with usual treatment. Based on these findings, we conclude that the DPWG guidelines can be effectively used to safely accelerate attainment of therapeutic TCA plasma concentrations.The difference between PIT and usual treatment can be explained by personalized initial dosages and bypassing of the buildup phase in the PIT group. The apparent difference between secondary (nortriptyline) and tertiary (clomipramine and imipramine) TCAs might be explained by metabolization pathways,To date, only 1 studyRegarding the secondary outcome measures, we found that faster attainment of therapeutic plasma concentrations did not translate into a significantly greater reduction in depressive symptoms or adverse effects. However, for adverse effects, we observed a different pattern in severity over time, suggesting that patients in the PIT group experienced gradually fewer and less severe adverse effects compared with those in the control group. In addition, it is clear that clinical outcome is influenced by many other biological and nonbiological factors in addition to antidepressant plasma concentrations.Most previous studies
PMC10167565
Limitations
This study has several limitations. First, dosage adjustments based on therapeutic drug monitoring were performed weekly in both study groups; therefore, usual treatment was of higher quality than that found in standard clinical practice in which it takes several weeks until plasma concentrations are measured.
PMC10167565
Conclusions
MDD, depressive symptoms
This RCT found that application of the DPWG guidelines in TCA treatment of MDD could be safely applied and resulted in faster attainment of therapeutic plasma concentrations. No effect on depressive symptoms was found. The results of this study imply that the benefits of preemptive pharmacogenetic testing may vary between antidepressants. Therefore, further research that takes into account specific gene-antidepressant interactions with clinical outcomes is necessary.
PMC10167565
Introduction
CORTEX
Neurofeedback (NF) is “a non-invasive brain stimulation technique equipped with a closed-loop control mechanism, whereby information on the dynamics, usually non-observable, is made observable to subjects, who can then use it to retroact on it, and push it towards functionally desirable goal states. NF involves defining: i) the general goal; ii) a neural target as feature; iii) an appropriate stimulation schedule” [Many NF paradigms target the modulation of neural functions related to executive control: the higher-order brain processes that support goal formation, planning, monitoring, and controlling complex, goal-directed thoughts and behaviors. Central to this system are the anterior cingulate cortex (ACC), which monitors for conflict and the demand for control, and the prefrontal cortex (PFC), which exerts goal-directed modulatory influences on sensory, perceptual, and memory processes represented in subcortical and posterior-cortical regions [
PMC10035884
Theta: A promising neural target for training executive control
Extensive research has shown that EEG-theta synchronization is central to executive control [In a meta-analytic review of the NF literature, Rogala et al. [As cited by Rogala et al. [More recently, Brandmeyer & Delorme [Wang & Hsieh [We are not aware of NF research designed to promote Overall, research on Fmθ NF training has revealed that young adult participants instructed to sustain elevated levels of Fmθ during training is associated with phasic theta power increases and improvement on some, but not all executive tasks relative to those who receive sham NF [
PMC10035884
Unresolved issues and gaps
One issue of previous NF training studies which has not been fully addressed concerns the sham control. Sham control conditions have typically been implemented as various forms of false feedback (e.g., receiving feedback from another participant’s previous recordings) or random feedback from one’s own recordings (e.g., non-contingent reward). Considering that accurate contingent feedback linking response and reward is required for optimal learning to occur [Another key limitation of previous NF studies, including those targeting Fmθ, is that they modulate neural activity only in one direction (either solely up-modulation or solely down-modulation). However, to support adaptive behavior in real-world situations, neural activity must exhibit dynamic variability such that activation and inhibition must be coordinated in response to varied internal/external demands. There are a number of theoretical concepts rooted in nonlinear systems that underscore the importance of such dynamic variability such as Representing another key issue, analogous to the above on sham control issues, the Fmθ activation targeted by NF is often confounded with the neural functions required to modulate brain responses in NF protocols. Specifically, NF is an attention-demanding task requiring the activation of frontal executive control networks during training irrespective of the frequency or region targeted by feedback [
PMC10035884
Current study
From a dynamical systems perspective, it is conceivable that designing NF to more flexibly train variable multistable states (e.g., transitioning between activation and inhibition) by rapidly alternating increasing and decreasing Fmθ feedback may lead to improved executive task performance relative to feedback designed to more rigidly train monostable states (e.g., consistently high or low activation). We are not aware of any NF studies which have asked this question, however, Papo [In both NF training groups, accurate contingent feedback was provided and short discrete trials (30-s) followed by a brief rest (10-s) were implemented rather than long sustained trial periods (several min). We implemented such short trials in an effort to promote dynamic variability in brain state that purportedly supports the dynamic executive control functions targeted by the NF training (i.e., dynamically alternating between goal states of activation and inhibition in the Go-NoGo task). We were also interested in whether Fmθ up-modulation training was truly driven by effective NF (self-regulation based on accurate contingent reward), as inferred in previous research, or by differences in non-specific executive control processes deployed during NF training (mental effort, attentional focus). This question was addressed with our aforementioned design where we compared Fmθ responses between alternating up/down-modulation to up-modulation alone. We tested two competing hypotheses regarding the effects of NF training on concomitant levels of Fmθ power. If Fmθ changes are due to self-regulation afforded by accurate contingent feedback, then the INC group should exhibit greater increases within and across training sessions. If, on the other hand Fmθ increases more in the ALT group, or no differences are observed between groups, then self-regulation based on accurate contingent reward mechanisms underlying reinforcement learning would be challenged. These hypotheses assume that self-regulation processes based on operant conditioning are differentiable from non-specific executive control processes. Further, they assume an adequate stimulation schedule to enable sufficient reinforcement learning. Assuming that Fmθ is effectively shaped by self-regulation in both groups, we hypothesized that the INC group would exhibit greater increases in Fmθ within and across NF training sessions. However, in contrast with previous research, we hypothesized that the ALT group would improve at a greater rate on the Go-NoGo shooting task over sessions, implying that training to dynamically modulate increases and decreases in Fmθ provides greater training specificity of dynamics-to-function mapping than does training to systematically increase Fmθ. Furthermore, we hypothesized higher Fmθ power in the ALT group during performance of the Go-NoGo task over sessions, suggesting more refined task specificity and context-dependent adaptations, or more specific dynamics-to-function mapping, in the ALT group.
PMC10035884
Materials and methods
PMC10035884
Participants
SESSION
Thirty (N = 30; 13 female) young healthy adults (ages 18–40 yr; mean 24.99 ±3.21) participated in seven separate sessions within a three-week interval: one Orientation Session, five combined NF training + Go-NoGo shooting (SH) task testing sessions, and one post-training SH task testing session. Electroencephalography (EEG) was not recorded in the Orientation Session but was continuously recorded during NF training and SH task periods throughout each of the following six sessions.
PMC10035884
Study design and timeline
The participants were randomly assigned to either the INC or ALT group. Those in the INC group (n = 12; 5 female) were instructed to increase their Fmθ power during each of six trials in each of six blocks in each of five sessions. Those in the ALT group (n = 18; 8 female) were instructed to either increase or decrease their Fmθ during each of six trials within each block, with three blocks of increase and three blocks of decrease alternated by random assignment across the six blocks in each of five sessions. In the first session, the SH task was conducted before NF training as a pre-test performance assessment. In sessions 2–5, NF training was conducted before the SH task, and in the 6
PMC10035884
Task design.
Experiment design illustrating the Intake/Orientation session, five sessions of NF training and six sessions of Go-NoGo SH task testing. Insert above shows six blocks (6 trials/block) of NF training for the INC (all increase Fmθ) and ALT (3 increase and 3 decrease Fmθ randomly assigned) groups with eyes-open rest periods before and after NF training (EO1-EO2). Insert below shows four blocks (90 trials/block) in both Low and High time stress conditions (counterbalanced order by subjects and sessions), with eyes-open rest periods before and after each SH task condition (EO1-EO3).
PMC10035884
Equipment and apparati
PMC10035884
Neurofeedback training and Go-NoGo shooting task
EVENT
An HTC Vive virtual reality system (The SH task paradigm was programmed to simulate a target range from a first-person (egocentric) perspective and transmitted the following event markers online in real-time from the scenario generation computer via parallel cable connection to the EEG recording system: (1) onset of target exposure, (2), location of target (left, center, right), (3) distance of target (near, mid, far), (4) identity of target (enemy or friendly), (5) onset of trigger pull (if shot fired), (6) result of shot (hit or miss target), and (7) offset of target exposure (time of target down if not fired upon or fired upon and missed).
PMC10035884
Electroencephalography (EEG) recording
EEG data were acquired at 2048 Hz and referenced online to the Common Mode Sense (CMS) and Direct Right Leg (DRL) electrodes using a 64 (+8 external) channel BioSemi system (Amsterdam, The Netherlands;
PMC10035884
Data integration and synchronization
All data were integrated and synchronized using Lab Streaming Layer (LSL [
PMC10035884
Procedures
PMC10035884
Intake/Orientation session
Anxiety
BLIND
Interested participants were first asked to come to the Virtual Reality Laboratory at the Technology Research Center (TRC) at UMBC for an Intake/Orientation meeting. During this session, the participants were provided a more thorough description of the study and further explanation of procedures and demands, shown brief introductions and demonstrations of the NF and SH task simulations, physiological recording preparation and procedures, and encouraged to ask questions or express any concerns they might have before consenting to participate. Volunteers who agreed to participate were asked to read and sign an Informed Consent Agreement (approved by the Human Use Committee at ARL and the Institutional Review Board at UMBC, in accordance with the Declaration of Helsinki and the U.S. Code of Federal Regulations). At this time, the participant was randomly assigned to either the control or experimental group. To avoid potential biases or preconceptions of the participants in a manner that might confound the results of this study, we informed all participants that "The purpose of this study is to investigate how humans learn to control technological systems with their brain signals through short-term neurofeedback training". Thus, participants were blind with respect to the existence of control and experimental groups or the real purpose of the study. We then administered a basic demographics information form and the trait version of the State-Trait Anxiety Inventory (STAI [
PMC10035884
Familiarization and training procedure (FTP)
SESSION
To minimize novelty and learning effects during experimental testing, we first presented the participants a Familiarization and Training Procedure (FTP) during the Orientation Session. The FTP was conducted without physiological recordings or instrumentation beyond the Vive headset and hand controller, which was used in the implementation of the SH task. If the participant exhibited stable performance in the FTP (asymptotic hit rates on enemy targets), we proceeded to the individualized threshold testing procedure (described below), else we repeated the FTP until stable performance was exhibited. In the FTP, a block consisting of 100 trials with all-enemy targets (presented with parameters: inter-target-intervals (ITIs) = 2000 ± 1000 ms spread over a Gaussian-distributed range of ITIs in 100 ms step increments presented in a randomized sequence and target exposure times (TETs) ranging from 400–1500 ms in equally distributed 100 ms bin step increments and presented in a randomized sequence across nine target positions. At this stage of training in the experiment, we sought to emphasize reflexive but accurate shooting responses to enemy targets to reinforce the buildup of a pre-potent response bias, which later in the experiment by design, would induce a greater likelihood of higher friendly-fire error rates during the experiment.
PMC10035884
Individual performance thresholding procedure (IPTP)
To induce a task-relevant time-stress effect, while accounting for individual differences in the ability to perform the task, we determined each participant’s individual reaction time thresholds (i.e., individualized target exposure durations) corresponding to the 50th (High time stress) and 90th (Low time stress) percentile hit-rates for enemy targets using psychophysical methods (method of limits [
PMC10035884
Compensation and scheduling
SESSION
We provided monetary compensation to attract willing participants, reduce attrition, and incentivize completion of all training and testing sessions. We paid $10/hr as base pay for participation in each session and an additional $100 bonus for completion of the entire study. We scheduled dates and times of the six remaining sessions after the Orientation Session within a three-week interval (two sessions/week). In each training + testing session, the participants were instrumented for physiological recordings (~60 min) and asked to conduct NF training (~30 min) and perform the SH task (~45 min), except for the last session, in which they were asked only to perform the SH task.
PMC10035884
Statistical analysis
Multilevel linear modeling (MLM) was applied to examine effects of NF training on Fmθ, and to examine training-related changes in Fmθ and behavioral performance measures during the Go-NoGo shooting task (% errors of commission, accuracy, RT). Models were constructed in four steps and participants were used as the level 2 grouping variable. Several models were run for each dependent measure: intercepts only models, inclusion of level 1 predictors, inclusion of the level 2 predictors of session number and interactions with session number, and the full model including the level 2 predictors group and group interactions, see equations 1 and 2. Data were analyzed using R (version 4.1.0) in conjunction with several packages: lme4, jtools, ggplot2, and interactions.EQ1. Equations for full NF modelsLevel 1: ŷLevel 2: π        π        πEQ2. Equations for full SH task modelsLevel 1: ŷLevel 2: π        πBefore conducting the modeling, several assumptions related to performing an MLM analysis were examined and outliers were removed. MLM analyses are better able to handle assumptions of independence compared to traditional general linear models but are still subject to many of the same assumptions including normality, linearity, multicollinearity, homogeneity of variances, issues related to residuals, and outliers [Fixed effects are reported as unstandardized beta weights with standard errors. Additionally, intraclass correlation coefficients (ICC) were obtained from each model, large ICCs and design effects of greater than 2 provide evidence for multilevel structure to the data. The estimation method used was restricted maximum likelihood (REML). REML tends to lead to more accurate results than full maximum likelihood (FML), when the sample size of the level two identifier is small [
PMC10035884
Responder-only analyses
REGRESSION
Responder-only analyses were also conducted, which are reported in Supporting information. For the analyses, participants were considered responders if either session or moderation could predict their Fmθ power during NF training using univariate regression analyses. Using these criteria 18 participants were considered responders and 12 were non-responders.
PMC10035884
Results
PMC10035884
Neurofeedback training
An intercepts only model was conducted, Fmθ (
PMC10035884
Fmθ changes within and across sessions for each group during NF.
SE
Changes in Fmθ over Blocks and Sessions of NF training for each Group (error bars are SE).
PMC10035884
FCz full spectrum and topographic maps of theta and alpha power for the first and last training sessions for each group during NF.
SESSION
Topographic maps of mean theta (4–7 Hz; top row) and alpha (8–13 Hz; bottom row) averaged over Blocks for Session 1 (left map) and Session 5 (right map) of NF training and corresponding mean power spectra at FCz from 0–30 Hz for INC group (left spectra), ALT group, down-modulation (middle spectra), and ALT group, up-modulation blocks (right spectra).
PMC10035884
Full MLM of Fmθ during NF training.
PMC10035884
Go-NoGo shooting task
PMC10035884
Frontal theta
Fmθ from shooting task trials (
PMC10035884
Fmθ changes across sessions in low and high time-stress conditions for each group during SH.
SE
Changes in Fmθ over Sessions during Go-NoGo shooting task in Low (left) and High (right) time-stress conditions for each Group (error bars are SE).
PMC10035884
FCz full spectrum and topographic maps of theta and alpha power for the first and last testing sessions for each group during SH.
SESSION
Topographic maps of mean theta (4–7 Hz; top row) and alpha (8–13 Hz; bottom row) power averaged over Blocks for Session 1 (left map) and Session 6 (right map) during SH task and corresponding mean power spectra at FCz from 0–30 Hz for INC group (column 1) and ALT group (column 2) in Low (row 1) and High (row 2) time stress conditions.
PMC10035884
Full MLMs of Fmθ during Go-NoGo shooting task.
PMC10035884
Behavioral performance
PMC10035884
% Errors of commission
Examining commission errors (
PMC10035884
Changes in errors of commission across sessions in low and high time-stress conditions for each group.
SE
Percentage errors of commission over Sessions in the Low (left) and High (right) time-stress conditions for each Group (error bars are SE).
PMC10035884
Full MLMs for behavioral measures from Go-NoGo shooting task.
PMC10035884
Accuracy
The intercepts only model for accuracy (
PMC10035884
Changes in shooting accuracy across sessions in low and high time-stress conditions for each group.
SE
Percentage enemy targets hit over Sessions in the Low (left) and High (right) time-stress conditions for each Group (error bars are SE).
PMC10035884
Reaction times
Reaction time (
PMC10035884
Changes in RTs across sessions in low and high time-stress conditions for each group.
SE
RTs to enemy targets over Sessions in the Low (left) and High (right) time-stress conditions for each Group (error bars are SE).
PMC10035884
Discussion
SAID
We investigated neurobehavioral effects of a novel NF training protocol that targeted both up- and down-modulation of Fmθ. This novel alternating up/down modulation condition (ALT) was compared to a more common NF training condition targeting solely up-modulation of Fmθ (INC). Accurate contingent feedback was provided in both groups; only the top-down goals of the training differed between groups, such that ALT group required switching of top-down training goals whereas the INC group required maintenance of a consistent top-down training goal. Differences between groups were investigated both in terms of (1) NF training-related changes in Fmθ and (2) changes in Fmθ and performance outcomes during a Go-NoGo shooting task. Overall, the present findings indicate greater Fmθ increases across training sessions in the ALT versus INC group. Accuracy and reaction time metrics improved at a steeper rate in the INC relative to the ALT group. Taken together, these results are consistent with the notion that alternating up- and down-modulation of Fmθ incurred switching costs that led to overall increases in theta (across sessions). It should be noted that these inferences are merely suggestive in light of insufficient number of sessions to explore alternative accounts of the findings (e.g., learning). Future studies should corroborate these findings using more practice sessions. Nevertheless, this study is an important step in unraveling the breadth of complex neurocognitive mechanisms at play in NF protocols.One question about using sham control groups, the gold standard in NF training research [In the present study, the difference in training between ALT and INC groups was the top-down goals of increasing, or alternating increasing and decreasing Fmθ. Considering that real contingent feedback was provided in both groups, the coupling of top-down and bottom-up error-related processing of visual stimuli should be comparable between groups (i.e., both groups should be able to self-regulate Fmθ based on accurate contingent feedback). In previous research where up-modulation of Fmθ groups were compared to sham control groups receiving false or non-contingent feedback, the opposite could be said. That is, top-down goals were the same between groups (up-modulation), but the coupling with bottom-up error-related processing demands would be artificially altered in sham control group subjects because the lack of correspondence between top-down goals and bottom-up feedback. On the other hand, because experimental and sham control groups receive the same visual stimuli (when pairs of subjects are matched), they experience identical bottom-up visual feedback (except where artifacts are produced). The difference would be the natural coupling of top-down and bottom-up attention signals in experimental group, but decoupling in the sham group. According to Corbetta and Shulman [
PMC10035884
Neurofeedback training
SESSION
In an attempt to disentangle confounding effects of training-related increases in Fmθ observed in previous research in real NF versus sham control groups, i.e., self-regulation based on contingent feedback, or other mechanisms, we tested two competing hypotheses while holding accurate contingent feedback constant between groups. First, if Fmθ increases are greater in the INC group, it would suggest that the effects of increased activation are due to reinforcement learning of self-regulation afforded by accurate contingent feedback, in support of previous research. Second, if Fmθ increases are greater in the ALT group, it would suggest that the effects of increased activation are not due to due to self-regulation, as ALT participants should exhibit lower Fmθ, especially during down-modulation blocks. Rather, greater increased activation in the ALT group may be associated with greater task demands imposed by having to switch between top-down goal states of up- and down-modulation (i.e., switch-costs). Support for the latter hypothesis would call into question previous research suggesting that Fmθ increases observed in NF training protocols designed to up-modulate Fmθ are unambiguously due to reinforcement learning of self-regulation. Based on the significant Group x Session x Block interaction observed in the present study, we found support for the hypothesis that the effects of increased activation in the ALT versus INC group within and across NF training sessions are not due to reinforcement learning of self-regulation afforded by accurate contingent feedback. According to Womelsdorf, Vinck, Leung, and Everling [We’re not aware of any studies on Fmθ NF which have directly compared up- vs. down-modulation training protocols to each other in the same study in young healthy participants. Our results suggest that Fmθ is involved in the attempted active modulation of brain states when accurate contingent feedback is provided when attempting to alternate increases and decreases within the same session, whether the top-down goal is to increase or decrease it. According to meta-analytic fMRI NF research, Emmert et al. [
PMC10035884
Go-NoGo shooting task performance
Regarding training-related performance outcomes, we hypothesized that the ALT group would improve at a greater rate on the Go-NoGo shooting task over sessions, as the dynamics-to-function mapping of activation and inhibition attempted in training would be more specific to the demands of the Go-NoGo task to which the desired training outcome should map onto [Previous research has shown that, although Fmθ increases in NF relative to sham control groups, performance on inhibitory control or conflict detection tasks (Stroop, stop-signal, local-global task) did not improve pre-post training for either NF training or sham control groups [In the present study, we implemented NF training within the same session before Go-NoGo task testing, except for the first (pre-) and last (post-) training sessions, whereas previous research conducted NF training in separate sessions between pre- and post-task performance evaluation sessions [It is interesting to note in Figs Finally, few studies have examined brain network adaptations associated with EEG NF training but novel methods and approaches are rapidly advancing to investigate the nature of dynamic functional interactions among different brain regions and adaptations associated with NF training, particularly in the fMRI literature [
PMC10035884
Limitations and future directions
fatigue
A major limitation of this study was that we did not include a third sham control group, and/or a down-regulation only NF training group. This was mainly because of practical concerns of time (for both experimenters and participants) and cost. Ideally, we would have included a sham control group and an exclusive down-modulation group. Additionally, the duration of training sessions, the structure of NF trials, and the number of sessions may also have been insufficient to elicit robust training effects. However, reducing NF trial durations from continuous 5 min or more in many studies to 30 s trials was to minimize the likelihood of boredom and fatigue and maximize operant conditioning principles of learning [
PMC10035884
Summary and conclusions
In the present study, the difference between ALT and INC training groups was task-switching of top-down goals of increasing or alternating increasing and decreasing Fmθ activity. Considering that accurate contingent feedback was provided in both groups, coupling between top-down goals and bottom-up error-related processing of visual feedback stimuli was comparable between groups. Our results are difficult to form any firm conclusions given its limitations. However, we hope that we have introduced a proof-of-concept based on new ideas to better stimulate future research based on the novel concepts offered. In particular, protocols designed to test greater outcome specificity, in terms of more specific dynamics-to-function mapping [
PMC10035884
Supporting information
PMC10035884
Responder-only changes in Fmθ over sessions during NF training and Go-NoGo shooting task and changes in errors of commission, accuracy, and RT over sessions during Go-NoGo shooting task.
(DOCX)Click here for additional data file.
PMC10035884
Full MLM for responder-only of Fmθ during NF training and Go-NoGo shooting task and errors of commission, accuracy, and RT over sessions during Go-NoGo shooting task.
(DOCX)Click here for additional data file.
PMC10035884
Results of responder-only MLM analyses of Fmθ during NF training and Go-NoGo shooting task and errors of commission, accuracy, and RT over sessions during Go-NoGo shooting task.
(DOCX)Click here for additional data file.We gratefully acknowledge Michael Hammond, Zheng Li, and Vikramaditya Battina for engineering system design and testing, and Theo Feng for scheduling, data collection, and study administration. We also thank the peer reviewers for their valuable contribution to this manuscript.
PMC10035884
Background
critically ill obese, non-obese, ARDS
OBESE, ARDS
Respiratory mechanics is a key element to monitor mechanically ventilated patients and guide ventilator settings. Besides the usual basic assessments, some more complex explorations may allow to better characterize patients’ respiratory mechanics and individualize ventilation strategies. These advanced respiratory mechanics assessments including esophageal pressure measurements and complete airway closure detection may be particularly relevant in critically ill obese patients. This study aimed to comprehensively assess respiratory mechanics in obese and non-obese ICU patients with or without ARDS and evaluate the contribution of advanced respiratory mechanics assessments compared to basic assessments in these patients.
PMC10476380
Methods
All intubated patients admitted in two ICUs for any cause were prospectively included. Gas exchange and respiratory mechanics including esophageal pressure and end-expiratory lung volume (EELV) measurements and low-flow insufflation to detect complete airway closure were assessed in standardized conditions (tidal volume of 6 mL kg
PMC10476380
Results
non-obese, ARDS
OBESE, ARDS
Among the 149 analyzed patients, 52 (34.9%) were obese and 90 (60.4%) had ARDS (65.4% and 57.8% of obese and non-obese patients, respectively,
PMC10476380
Supplementary Information
The online version contains supplementary material available at 10.1186/s13054-023-04623-2.
PMC10476380
Keywords
PMC10476380
Introduction
Respiratory mechanics is a key element in clinical practice to monitor mechanically ventilated patients and guide ventilator settings [
PMC10476380
Patients and methods
PMC10476380
Patients’ selection
cardiac arrest
CARDIAC ARREST, PNEUMOTHORAX
All patients admitted from March 2018 to January 2020 in two academic hospital ICUs (Angers and Strasbourg, France) intubated and mechanically ventilated for any cause were prospectively included in the study within 24 h after intubation. Exclusion criteria were age < 18 years, pneumothorax, contraindication to esophageal pressure measurement, and use of extracorporeal membrane oxygenation (ECMO) at the time of inclusion. Patients admitted after a cardiac arrest were excluded from the analysis and reported in another publication [
PMC10476380
Study protocol
PMC10476380
Settings
HEMODYNAMIC INSTABILITY, RECRUITMENT
All patients received deep sedation and neuromuscular blockers at the time of measurements and were ventilated using an EngströmEsophageal pressure was measured with a specific nasogastric feeding tube equipped with an esophageal balloon (NutriventTo normalize volume history, a recruitment maneuver was performed in volume-controlled ventilation in absence of hemodynamic instability by increasing PEEP level up to 20 cmHFlow, airway pressure, and esophageal pressure–time curves were recorded using a dedicated computer connected to the ventilator with a 40 ms sampling time for offline analysis.
PMC10476380
Measurements
expiratory occlusion
BLIND
All esophageal pressure signal recordings were independently inspected by two investigators blind to the other clinical data, and those considered non-valid were excluded from the analyses including esophageal pressure data.Inspiratory and expiratory occlusion maneuvers were performed to measure Abdominal pressure (An arterial blood gas was performed after 15 min free of any occlusion maneuver, and EELV was measured at PEEP 5 cmHA low-flow inflation (5 L minComplete airway closure and corresponding AOP were identified by the inspection of the pressure–volume curves as previously described [
PMC10476380
Calculated variables
DPThe difference between P1 and Inspiratory (The lung driving pressure (DPThe elastance ratio (As DPThe ratios Dead space was assessed using ventilatory ratio, which was computed as minute ventilation (mL/min) × PaCO
PMC10476380
Other collected data
ARDS
OPACITIES, ARDS, BLIND, CHRONIC RESPIRATORY DISEASE
Age, height, weight, past medical history of chronic respiratory disease, immunodepression, Sequential Organ Failure Assessment (SOFA) score [The lung opacities were independently assessed on chest X-rays by two experienced investigators blind to clinical data.The diagnosis of ARDS was performed using the criteria of the Berlin definition by an adjudication committee blind to respiratory mechanics data [Survival was assessed at day 60 after inclusion.The number of ventilator-free days at day 28 was defined as the number of days between day 1 and day 28 on which patients breathed without assistance. A value of 0 ventilator-free day was assigned for patients who died before day 28.
PMC10476380
Statistical analysis
obesity
OBESITY
Results are presented as median [interquartile range] and number (percentage). Normality of the variables was assessed using the D’Agostino & Pearson test. The study population was divided into four groups according to the presence of obesity or not (BMI < or ≥ 30 kg m
PMC10476380
Results
PMC10476380
Main patients’ characteristics
non-obese, ARDS
ACUTE RESPIRATORY DISTRESS SYNDROME, OBESE, ARDS
One hundred and sixty-four patients were included in the study. One hundred and forty-nine of them were included in the analysis (15 patients were excluded because of lack of data due to technical issues in recordings or major deviations in study protocol). Valid esophageal pressure measurements were analyzed in 124 patients. Respiratory mechanics and gas exchange were assessed 10 [3.5–22] hours after intubation.Fifty-two patients (34.9%) were obese. Ninety (60.4%) patients fulfilled ARDS criteria (65.4% of obese and 57.8% of non-obese patients, Main characteristics of the patientsData are presented as median [interquartile range] or number (percentage)ARDS, Acute Respiratory Distress Syndrome; BMI, Body Mass Index; SOFA, Sequential Organ Failure Assessment; SAPS II, Simplified Acute Physiology Score II*
PMC10476380
Gas exchange
obesity, ARDS
OBESITY, ARDS
Gas exchange in the patients categorized according to the presence or not of obesity and/or ARDS is presented in Table Gas exchange and respiratory mechanicsData are presented as median [interquartile range] or number (percentage)RR, Respiratory Rate; VE, Minute ventilation; FiO*
PMC10476380
Airway closure and driving pressure
A complete airway closure assessed with a PEEP of 5 cmHDPΔP1-
PMC10476380
Lung driving pressure, plateau pressure of the lung and lung stress
DP
PMC10476380
Conclusion
obese, critically ill, non-obese ARDS, ARDS
OBESE, CRITICALLY ILL, ARDS
Basic respiratory mechanics and gas exchange features of obese patients are similar to those observed in non-obese ARDS patients. But an advanced assessment of respiratory mechanics allows to show that end-expiratory esophageal pressure, although largely distributed, is higher in obese patients. Chest wall compliance is not altered in obese or ARDS patients and is not easily predictable by patients’ general characteristics. A complete airway closure can be found in around 25% of critically ill patients ventilated with a PEEP of 5 cmH
PMC10476380
Acknowledgements
The authors would like to greatly acknowledge all the medical and non‐medical teams of the Angers and Strasbourg Medical ICUs.
PMC10476380
Author contributions
FM
FB, JCR, and AM designed the study. FB, HM, PYO, BP, CD, AS, EY, DC, AC, MC, HR, and FM conducted the study on enrolled patients. FB, JCR, PYO, CD, BP, EY, LP, AL, MC, and AM analyzed the data. FB and JCR interpreted the data and wrote the first draft of the manuscript. All authors contributed to drafting of the work and approved the final version of the manuscript.
PMC10476380
Funding
Some equipment used in this work was graciously provided by GE Healthcare. EY, CD, and DC received a 1‐year research fellowship grant from the University Hospital of Angers, France. BP received a 1‐year research fellowship grant from the University Hospital of Réunion, France.
PMC10476380
Availability of data and materials
The datasets analyzed during the current study are available from the corresponding author on reasonable request.
PMC10476380
Declarations
PMC10476380
Ethics approval and consent to participate
The study was performed in accordance with the ethical standards of the Declaration of Helsinki. It was approved by the appropriate legal and ethical authorities (ethics committee Sud-Est I, # 2017-A02842-51). Oral consent was obtained from all patients’ relatives after oral and written information.
PMC10476380