Functional connectivity signatures of major depressive disorder: machine learning analysis of two multicenter neuroimaging studies.

The promise of machine learning has fueled the hope for developing diagnostic tools for psychiatry. Initial studies showed high accuracy for the identification of major depressive disorder (MDD) with resting-state connectivity, but progress has been hampered by the absence of large datasets. Here we used regular machine learning and advanced deep learning algorithms to differentiate patients with MDD from healthy controls and identify neurophysiological signatures of depression in two of the largest resting-state datasets for MDD. We obtained resting-state functional magnetic resonance imaging data from the REST-meta-MDD (N = 2338) and PsyMRI (N = 1039) consortia. Classification of functional connectivity matrices was done using support vector machines (SVM) and graph convolutional neural networks (GCN), and performance was evaluated using 5-fold cross-validation. Features were visualized using GCN-Explainer, an ablation study and univariate t-testing. The results showed a mean classification accuracy of 61% for MDD versus controls. Mean accuracy for classifying (non-)medicated subgroups was 62%. Sex classification accuracy was substantially better across datasets (73-81%). Visualization of the results showed that classifications were driven by stronger thalamic connections in both datasets, while nearly all other connections were weaker with small univariate effect sizes. These results suggest that whole brain resting-state connectivity is a reliable though poor biomarker for MDD, presumably due to disease heterogeneity as further supported by the higher accuracy for sex classification using the same methods. Deep learning revealed thalamic hyperconnectivity as a prominent neurophysiological signature of depression in both multicenter studies, which may guide the development of biomarkers in future studies.

Outcomes of electroconvulsive therapy in patients with depressive symptoms with versus without comorbid personality disorders/traits: A systematic review and meta-analysis.

AIMS: To assess electroconvulsive therapy (ECT) outcomes in patients affected by depressive symptoms with versus without additional comorbid personality disorders/traits. METHODS: We identified observational studies investigating ECT clinical outcomes in patients affected by depressive symptoms with versus without comorbid personality disorders/traits in Embase/Medline in 11/2022. Our protocol was registered with PROSPERO (CRD42023390833). Study quality was evaluated using the Newcastle-Ottawa-Scale. Our primary outcomes were ECT response and remission rates. Meta-regression analyses included effects of in/outpatient percentages, age, number of ECT sessions, and electrode placement; subgroup analyses included the assessment methods for personality disorders/traits. We performed sensitivity analyses after excluding poor-quality studies. RESULTS: A total of 20 studies (n = 11,390) were included in our analysis. Patients with comorbid personality disorders/traits had lower remission rates (OR = 0.42, 95% CI = 0.31, 0.58, p < 0.001) with substantial heterogeneity (I2 = 93.0%) as well as lower response rates (OR = 0.35, 95% CI = 0.24, 0.51, n = 5129, p < 0.001) with substantial heterogeneity (I2 = 93.0%) compared with patients without comorbid personality disorders/traits. Relapse rates were higher in patients with versus without comorbid personality disorders/traits (OR = 3.23, 95% CI = 1.40, 7.45, k = 4, n = 239, p = 0.006) with moderate heterogeneity (I2 = 75.0%) and post-ECT memory impairment was more frequent in patients with versus without comorbid personality disorders/traits (OR = 1.41, 95% CI = 1.36, 1.46, k = 4, n = 471, p < 0.001) with minimal heterogeneity (I2 = 0.0%). Dropout rates were higher in patients with versus without comorbid personality disorders/traits (OR = 1.58, 95% CI = 1.13, 2.21, k = 3, n = 6145, p = 0.008). CONCLUSIONS: Patients with comorbid personality disorders/traits treated with ECT are reported to have lower response and remission rates and higher rates of side effects and relapse rates compared with patients without personality disorders/traits.

The boon and bane of nitrous oxide.

Nitrous oxide (N2O) has been known since the end of the eighteenth century. Today, N2O plays a huge role as a greenhouse gas and an ozone-depleting stratospheric molecule. The main sources of anthropogenic N2O emissions are agriculture, fuel combustion, wastewater treatment, and various industrial processes. By contrast, the contribution of medical N2O to the greenhouse effect appears to be small. The recreational and medical uses of N2O gradually diverged over time. N2O has analgesic and anesthetic effects, making it widely used in modern dentistry and surgery. New research has also begun studying N2O's antidepressant actions. N-methyl-D-aspartate (NMDA) antagonism and opioid effects are believed to be the main underlying biochemical mechanisms. At this point, numerous questions remain open and, in particular, the conduct of larger clinical trials will be essential to confirm N2O's use as a rapid-acting antidepressant. The N2O concentration delivered, the duration of a single inhalation, as well as the number of inhalations ultimately required, deserve to be better understood. Finally, the non-medical use of N2O has gained significant attention in recent years. Sudden deaths directly attributed to N2O are primarily due to asphyxia. Heavy, chronic N2O use may result in vitamin B12 deficiency, which, among other things, may cause megaloblastic anemia, venous thrombosis, myeloneuropathy, and skin pigmentation. Helpful biochemical tests include homocysteine and methylmalonic acid. The centerpiece of treatment is complete cessation of N2O use together with parenteral administration of vitamin B12.

Postpartum hemorrhage and postpartum depression: A systematic review and meta-analysis of observational studies.

OBJECTIVE: To assess the postpartum depression (PPD) risk in women with postpartum hemorrhage (PPH) and moderators. METHODS: We identified observational studies of PPD rates in women with versus without PPH in Embase/Medline/PsychInfo/Cinhail in 09/2022. Study quality was evaluated using the Newcastle-Ottawa-Scale. Our primary outcome was the odds ratio (OR, 95% confidence intervals [95%CI]) of PPD in women with versus without PPH. Meta-regression analyses included the effects of age, body mass index, marital status, education, history of depression/anxiety, preeclampsia, antenatal anemia and C-section; subgroup analyses were based on PPH and PPD assessment methods, samples with versus without history of depression/anxiety, from low-/middle- versus high-income countries. We performed sensitivity analyses after excluding poor-quality studies, cross-sectional studies and sequentially each study. RESULTS: One, five and three studies were rated as good-, fair- and poor-quality respectively. In nine studies (k = 10 cohorts, n = 934,432), women with PPH were at increased PPD risk compared to women without PPH (OR = 1.28, 95% CI = 1.13 to 1.44, p < 0.001), with substantial heterogeneity (I2 = 98.9%). Higher PPH-related PPD ORs were estimated in samples with versus without history of depression/anxiety or antidepressant exposure (OR = 1.37, 95%CI = 1.18 to 1.60, k = 6, n = 55,212, versus 1.06, 95%CI = 1.04 to 1.09, k = 3, n = 879,220, p < 0.001) and in cohorts from low-/middle- versus high-income countries (OR = 1.49, 95%CI = 1.37 to 1.61, k = 4, n = 9197, versus 1.13, 95%CI = 1.04 to 1.23, k = 6, n = 925,235, p < 0.001). After excluding low-quality studies the PPD OR dropped (1.14, 95%CI = 1.02 to 1.29, k = 6, n = 929,671, p = 0.02). CONCLUSIONS: Women with PPH had increased PPD risk amplified by history of depression/anxiety, whereas more data from low-/middle-income countries are required.

Deep Learning in the Identification of Electroencephalogram Sources Associated with Sexual Orientation.

INTRODUCTION: It is unclear if sexual orientation is a biological trait that has neurofunctional footprints. With deep learning, the power to classify biological datasets without an a priori selection of features has increased by magnitudes. The aim of this study was to correctly classify resting-state electroencephalogram (EEG) data from males with different sexual orientation using deep learning and to explore techniques to identify the learned distinguishing features. METHODS: Three cohorts (homosexual men, heterosexual men, and a mixed sex cohort), one pretrained network on sex classification, and one newly trained network for sexual orientation classification were used to classify sex. Further, Grad-CAM methodology and source localization were used to identify the spatiotemporal patterns that were used for differentiation by the networks. RESULTS: Using a pretrained network for classification of males and females, no differences existed between classification of homosexual and heterosexual males. The newly trained network was able, however, to correctly classify the cohorts with a total accuracy of 83%. The retrograde activation using Grad-CAM technology yielded distinctive functional EEG patterns in the Brodmann area 40 and 1 when combined with Fourier analysis and a source localization. DISCUSSION: This study shows that electrophysiological trait markers of male sexual orientation can be identified using deep learning. These patterns are different from the differentiating signatures of males and females in a resting-state EEG.

Exposure and response prevention therapy augmented with naltrexone in kleptomania: a controlled case study using galvanic skin response for monitoring.

BACKGROUND: Kleptomania is a disease that shares features with obsessive compulsive spectrum disorders (OCD) and with substance abuse disorders (SAD). This is underlined by therapeutic approaches in kleptomania ranging from cognitive behavioural therapy and selective serotonin reuptake inhibitors that are effective in OCD, and opioid antagonists that are currently being used in SAD. However, almost no literature exists about exposure and response prevention (ERP) therapy in kleptomania. Furthermore, there is a clear lack of objective markers that would allow a therapeutic monitoring. AIM: To show the effectiveness of ERP therapy in kleptomania in a single case report. METHOD: An ERP therapy under real-world conditions and later augmentation with the opioid antagonist naltrexone is described. Continuous measurements of galvanic skin response (GSR) before, during and after therapy sessions are reported in association with changes of the Kleptomania Symptom Assessment Scale (KSAS) self-questionnaire. RESULTS: While KSAS scores showed a clear treatment response to ERP sessions, the GSR was significantly lower during ERP treatment in comparison with baseline measures. However, during augmentation with naltrexone, GSR measures increased again and clinical severity did not further improve. CONCLUSIONS: This case shows the possible usefulness of ERP-like approaches and therapy monitoring using electrophysiological markers of arousal for individualized treatment in kleptomania.

Length of Involuntary Hospitalization Related to the Referring Physician's Psychiatric Emergency Experience.

Although involuntary commitment (IC) is a serious intervention in psychiatry and must always be regarded as an emergency measure, the knowledge about influencing factors is limited. Aims were to test the hypothesis that duration of involuntary hospitalization and associated parameters differ for IC's mandated by physicians with or with less routine experience in psychiatric emergency situations. Duration of involuntary hospitalization and duration until day-passes of 508 patients with IC at the University Hospital of Psychiatry Zurich were analyzed using a generalized linear model. Durations of involuntary hospitalization and time until day-passes were significantly shorter in patients referred by physicians with less routine experience in psychiatric emergency situations than compared to experienced physicians. Shorter hospitalizations following IC by less-experienced physicians suggest that some IC's might be unnecessary. A specific training or restriction to physicians being capable of conducting IC could decrease the rate of IC.

Objective markers for sleep propensity: comparison between the Multiple Sleep Latency Test and the Vigilance Algorithm Leipzig.

The regulation of wakefulness is important for high-order organisms. Its dysregulation is involved in the pathomechanism of several psychiatric disorders. Thus, a tool for its objective but little time-consuming assessment would be of importance. The Vigilance Algorithm Leipzig allows the objective measurement of sleep propensity, based on a single resting state electroencephalogram. To compare the Vigilance Algorithm Leipzig with the standard for objective assessment of excessive daytime sleepiness, a four-trial Multiple Sleep Latency Test in 25 healthy subjects was conducted. Between the first two trials, a 15-min, 25-channel resting electroencephalogram was recorded, and Vigilance Algorithm Leipzig was used to classify the sleep propensity (i.e., type of vigilance regulation) of each subject. The results of both methods showed significant correlations with the Epworth Sleepiness Scale (ρ = -0.70; ρ = 0.45, respectively) and correlated with each other (ρ = -0.54). Subjects with a stable electroencephalogram-vigilance regulation yielded significant increased sleep latencies compared with an unstable regulation (multiple sleep latency 898.5 s versus 549.9 s; P = 0.03). Further, Vigilance Algorithm Leipzig classifications allowed the identification of subjects with average sleep latencies <6 min with a sensitivity of 100% and a specificity of 77%. Thus, Vigilance Algorithm Leipzig provides similar information on wakefulness regulation in comparison to the much more cost- and time-consuming Multiple Sleep Latency Test. Due to its high sensitivity and specificity for large sleep propensity, Vigilance Algorithm Leipzig could be an effective and reliable alternative to the Multiple Sleep Latency Test, for example for screening purposes in large cohorts, where objective information about wakefulness regulation is needed.

Personalized Medicine: Review and Perspectives of Promising Baseline EEG Biomarkers in Major Depressive Disorder and Attention Deficit Hyperactivity Disorder.

Personalized medicine in psychiatry is in need of biomarkers that resemble central nervous system function at the level of neuronal activity. Electroencephalography (EEG) during sleep or resting-state conditions and event-related potentials (ERPs) have not only been used to discriminate patients from healthy subjects, but also for the prediction of treatment outcome in various psychiatric diseases, yielding information about tailored therapy approaches for an individual. This review focuses on baseline EEG markers for two psychiatric conditions, namely major depressive disorder and attention deficit hyperactivity disorder. It covers potential biomarkers from EEG sleep research and vigilance regulation, paroxysmal EEG patterns and epileptiform discharges, quantitative EEG features within the EEG main frequency bands, connectivity markers and ERP components that might help to identify favourable treatment outcome. Further, the various markers are discussed in the context of their potential clinical value and as research domain criteria, before giving an outline for future studies that are needed to pave the way to an electrophysiological biomarker-based personalized medicine.

EEG-vigilance regulation is associated with and predicts ketamine response in major depressive disorder.

Ketamine offers promising new therapeutic options for difficult-to-treat depression. The efficacy of treatment response, including ketamine, has been intricately linked to EEG measures of vigilance. This research investigated the interplay between intravenous ketamine and alterations in brain arousal, quantified through EEG vigilance assessments in two distinct cohorts of depressed patients (original dataset: n = 24; testing dataset: n = 24). Clinical response was defined as a decrease from baseline of >33% on the Montgomery-Åsberg Depression Rating Scale (MADRS) 24 h after infusion. EEG recordings were obtained pre-, start-, end- and 24 h post- infusion, and the resting EEG was automatically scored using the Vigilance Algorithm Leipzig (VIGALL). Relative to placebo (sodium chloride 0.9%), ketamine increased the amount of low-vigilance stage B1 at end-infusion. This increase in B1 was positively related to serum concentrations of ketamine, but not to norketamine, and was independent of clinical response. In contrast, treatment responders showed a distinct EEG pattern characterized by a decrease in high-vigilance stage A1 and an increase in low-vigilance B2/3, regardless of whether placebo or ketamine had been given. Furthermore, pretreatment EEG differed between responders and non-responders with responders showing a higher percentage of stage A1 (53% vs. 21%). The logistic regression fitted on the percent of A1 stages was able to predict treatment outcomes in the testing dataset with an area under the ROC curve of 0.7. Ketamine affects EEG vigilance in a distinct pattern observed only in responders. Consequently, the percentage of pretreatment stage A1 shows significant potential as a predictive biomarker of treatment response.Clinical Trials Registration: https://www.clinicaltrialsregister.eu/ctr-search/trial/2013-000952-17/CZ Registration number: EudraCT Number: 2013-000952-17.

Combined Effects of Nasal Ketamine and Trauma-Focused Psychotherapy in Treatment-Resistant Post-Traumatic Stress Disorder: A Pilot Case Series.

OBJECTIVE: This pilot case series investigated the feasibility and efficacy of an eight-week therapy program, combining nasally administered ketamine (0.5 mg/kg) with trauma-focused psychotherapy, for individuals with chronic, treatment-resistant post-traumatic stress disorder (PTSD). METHOD: Three patients with chronic, treatment-resistant PTSD underwent the eight-week therapy program. Clinical assessments included the Clinician-Administered PTSD Scale for DSM-5 (CAPS-5) and the Hamilton Depression Rating Scale (HAMD) at baseline, post-assessment, and follow-up assessment, along with additional measures assessing other relevant symptoms and side effects. RESULTS: The results demonstrated clinically relevant reductions in PTSD symptoms, indicated by a change in the CAPS-5 score at post-assessment (M = -18.00; SE = 6.48) and follow-up assessment (M = -25.33, SE = 5.58). Additionally, depressive symptoms showed notable improvement, with changes in HAMD scores at post-assessment (M = -8.33, SE = 3.07) and follow-up assessment (M = -9.00, SE = 3.77). Positive effects were also observed in anxiety reduction, decreased dissociations, and improvements in emotion regulation and disturbances of self-organization. CONCLUSIONS: Despite potential variations in clinical profiles among the patients, the therapy program demonstrated positive outcomes for all participants. Nasally administered ketamine was well tolerated and resulted in immediate symptom reduction in tension, anxiety, and common PTSD symptoms. However, to validate these findings and compare treatment efficacy, future randomized controlled trials are warranted, especially in comparison with trauma-focused therapy alone.

Large language models can support generation of standardized discharge summaries - A retrospective study utilizing ChatGPT-4 and electronic health records.

OBJECTIVE: To evaluate whether psychiatric discharge summaries (DS) generated with ChatGPT-4 from electronic health records (EHR) can match the quality of DS written by psychiatric residents. METHODS: At a psychiatric primary care hospital, we compared 20 inpatient DS, written by residents, to those written with ChatGPT-4 from pseudonymized residents' notes of the patients' EHRs and a standardized prompt. 8 blinded psychiatry specialists rated both versions on a custom Likert scale from 1 to 5 across 15 quality subcategories. The primary outcome was the overall rating difference between the two groups. The secondary outcomes were the rating differences at the level of individual question, case, and rater. RESULTS: Human-written DS were rated significantly higher than AI (mean ratings: human 3.78, AI 3.12, p < 0.05). They surpassed AI significantly in 12/15 questions and 16/20 cases and were favored significantly by 7/8 raters. For "low expected correction effort", human DS were rated as 67 % favorable, 19 % neutral, and 14 % unfavorable, whereas AI-DS were rated as 22 % favorable, 33 % neutral, and 45 % unfavorable. Hallucinations were present in 40 % of AI-DS, with 37.5 % deemed highly clinically relevant. Minor content mistakes were found in 30 % of AI and 10 % of human DS. Raters correctly identified AI-DS with 81 % sensitivity and 75 % specificity. DISCUSSION: Overall, AI-DS did not match the quality of resident-written DS but performed similarly in 20% of cases and were rated as favorable for "low expected correction effort" in 22% of cases. AI-DS lacked most in content specificity, ability to distill key case information, and coherence but performed adequately in conciseness, adherence to formalities, relevance of included content, and form. CONCLUSION: LLM-written DS show potential as templates for physicians to finalize, potentially saving time in the future.

Separation of low-voltage EEG-activity during mental activation from that during transition to drowsiness.

When dealing with electroencephalograms (EEGs) recorded under resting conditions, periods of low-voltage activity might indicate drowsiness, but mental activation as well. Thus, low-voltage EEG retrieves a notorious source of confusion. The simultaneous occurrence of drowsiness related slow horizontal eye movements (SEM) allow to assign low-voltage EEG-activity to a brain state of reduced vigilance instead of mental activation. The aim of this study was to investigate, whether under resting conditions with eyes closed low-voltage EEG with SEM (B1+) and without SEM (B1-) differ in spectral and spatial distribution of EEG-activity. EEGs of 35 healthy subjects where analyzed, each containing at least 10 s of low-voltage EEG recorded during a calculation task (calc, as control condition), as well as 10 s of each B1- and B1+, recorded during following about 20 min of rest. Using standardized, low resolution brain electromagnetic tomography, cortical current density was computed in four individually adapted frequency bands (delta, theta, alpha, beta) for calc, B1- and B1+. Paired test comparison of cortical current densities revealed significant differences for B1- compared to B1+. In detail, B1- exhibited lower delta- and theta band activity, especially in the cingulate- and adjacent medial portions of the frontal, parietal and occipital cortices, as well as higher beta band activity in temporal cortices. Similar results where found in calc versus B1+. These findings support the association of B1- to a higher level of vigilance compared to B1+, thus justifying the separation of low-voltage EEG-activity by means of SEM.

EEG biomarkers in major depressive disorder: discriminative power and prediction of treatment response.

Major depressive disorder (MDD) has high population prevalence and is associated with substantial impact on quality of life, not least due to an unsatisfactory time span of sometimes several weeks from initiation of treatment to clinical response. Therefore extensive research focused on the identification of cost-effective and widely available electroencephalogram (EEG)-based biomarkers that not only allow distinguishing between patients and healthy controls but also have predictive value for treatment response for a variety of treatments. In this comprehensive overview on EEG research on MDD, biomarkers that are either assessed at baseline or during the early course of treatment and are helpful in discriminating patients from healthy controls and assist in predicting treatment outcome are reviewed, covering recent decades up to now. Reviewed markers include quantitative EEG (QEEG) measures, connectivity measures, EEG vigilance-based measures, sleep-EEG-related measures and event-related potentials (ERPs). Further, the value and limitations of these different markers are discussed. Finally, the need for integrated models of brain function and the necessity for standardized procedures in EEG biomarker research are highlighted to enhance future research in this field.

Mental health treatment seeking among patients with OCD: impact of age of onset.

BACKGROUND: Several studies have described the deficits in the health care provided to persons with obsessive-compulsive disorder (OCD), however, without making any distinction between psychiatric-psychotherapeutic professionals and general practitioners or other professionals. Also, the relation between subjectively defined early signs of the disorder, diagnosis and utilization of professional help has not yet been investigated systematically. The present study addresses these questions, using a self-rating questionnaire for patients with OCD (questionnaire on the utilization of professional help by patients with OCD). METHODS: Eighty-eight patients with OCD, who came to our University Hospital Outpatient Clinic for obsessive-compulsive disorders in Leipzig, participated in the study. The questionnaire, which had been developed specially for this survey, asked study participants to remember when they had first perceived signs of their disorder and their first-time utilization of professional help. RESULTS: Patients with OCD and early onset of disorder sought professional help later than did patients with later onset of disorder. When professional help was utilized, it took 2 years on average before the diagnosis was made. Patients with OCD first consulted a psychiatrist or psychotherapist and not the general practitioner as their first professional contact person. CONCLUSION: First signs in the early stages of OCD, particularly with an onset in childhood and adolescence have to be diagnosed at an earlier stage for appropriate treatment. Psychiatrists and psychotherapists, not primarily general practitioners, have a particularly high demand for further education about early diagnosis and treatment of OCD.

Autism spectrum disorders in adults and the autonomic nervous system: Heart rate variability markers in the diagnostic procedure.

The diagnostic assessment of autism spectrum disorders (ASD) in adults is a challenging and time-consuming procedure. In order to address the lack of specialised health-care professionals and improve the waiting time, we aimed to identify specific electrocardiogram (ECG) derived Heart Rate Variability (HRV) parameters that could be used for diagnostic purposes. 152 patients were diagnosed based on a standardised clinical procedure and assigned to one of three groups: ASD (n = 56), any other psychiatric disorder (OD) (n = 72), and patients with no diagnosis (ND) (n = 24). Groups were compared using ANOVA. Discriminative power of biological parameters and the clinical assessment were compared using receiver operating characteristic curves (ROCs). Patients with ASD showed reduced parasympathetic and increased sympathetic activity compared to ND. The accuracy determined by the area under the curve (AUC) of the biological parameters for discrimination between ASD vs. pooled OD/ND was 0.736 (95% CI = 0.652-0.820), compared to .856 (95% CI = 0.795-0.917) for the extensive clinical assessment. Our results confirmed the dysregulation of the autonomic nervous system in ASD with reduced parasympathetic and increased sympathetic activity as compared to ND. The discriminative power of biological markers including HRV was considerable and could supplement less sophisticated clinical assessments.

EEG wakefulness regulation in transdiagnostic patients after a recent suicide attempt.

OBJECTIVE: Decades of research have not yet produced statistically reliable predictors of preparatory behavior eventually leading to suicide attempts or deaths by suicide. As the nature of suicidal behavior is complex, it is best investigated in a transdiagnostic approach, while assessing objective markers, as proposed by the Research Domain Criteria (Cuthbert, 2013). METHODS: A 15-min resting-state EEG was recorded in 45 healthy controls, and 49 transdiagnostic in-patients with a recent (<6 months) suicide attempt. Brain arousal regulation in eyes-closed condition was assessed with the Vigilance Algorithm Leipzig (VIGALL) (Sander et al., 2015). RESULTS: A significant incline of median vigilance and vigilance slope was observed in patients within the first 3-min of the EEG recording. Additionally, a significant positive correlation of self-reported suicidal ideation with the vigilance slope over 15-min recording time, as well as a significant negative correlation with EEG vigilance stage A1 during the first 3-min was found. CONCLUSIONS: Transdiagnostic patients with a recent suicide attempt show a distinct vigilance regulation pattern. Further studies including a control group consisting of patients without life-time suicide attempts are needed to increase the clinical utility of the findings. SIGNIFICANCE: These findings might serve as potential objective markers of suicidal behavior.