Prediction of estimated risk for bipolar disorder using machine learning and structural MRI features.

BACKGROUND: Individuals with bipolar disorder are commonly correctly diagnosed a decade after symptom onset. Machine learning techniques may aid in early recognition and reduce the disease burden. As both individuals at risk and those with a manifest disease display structural brain markers, structural magnetic resonance imaging may provide relevant classification features. METHODS: Following a pre-registered protocol, we trained linear support vector machine (SVM) to classify individuals according to their estimated risk for bipolar disorder using regional cortical thickness of help-seeking individuals from seven study sites (N = 276). We estimated the risk using three state-of-the-art assessment instruments (BPSS-P, BARS, EPIbipolar). RESULTS: For BPSS-P, SVM achieved a fair performance of Cohen's κ of 0.235 (95% CI 0.11-0.361) and a balanced accuracy of 63.1% (95% CI 55.9-70.3) in the 10-fold cross-validation. In the leave-one-site-out cross-validation, the model performed with a Cohen's κ of 0.128 (95% CI -0.069 to 0.325) and a balanced accuracy of 56.2% (95% CI 44.6-67.8). BARS and EPIbipolar could not be predicted. In post hoc analyses, regional surface area, subcortical volumes as well as hyperparameter optimization did not improve the performance. CONCLUSIONS: Individuals at risk for bipolar disorder, as assessed by BPSS-P, display brain structural alterations that can be detected using machine learning. The achieved performance is comparable to previous studies which attempted to classify patients with manifest disease and healthy controls. Unlike previous studies of bipolar risk, our multicenter design permitted a leave-one-site-out cross-validation. Whole-brain cortical thickness seems to be superior to other structural brain features.

Machine Learning Prediction of Estimated Risk for Bipolar Disorders Using Hippocampal Subfield and Amygdala Nuclei Volumes.

The pathophysiology of bipolar disorder (BD) remains mostly unclear. Yet, a valid biomarker is necessary to improve upon the early detection of this serious disorder. Patients with manifest BD display reduced volumes of the hippocampal subfields and amygdala nuclei. In this pre-registered analysis, we used structural MRI (n = 271, 7 sites) to compare volumes of hippocampus, amygdala and their subfields/nuclei between help-seeking subjects divided into risk groups for BD as estimated by BPSS-P, BARS and EPIbipolar. We performed between-group comparisons using linear mixed effects models for all three risk assessment tools. Additionally, we aimed to differentiate the risk groups using a linear support vector machine. We found no significant volume differences between the risk groups for all limbic structures during the main analysis. However, the SVM could still classify subjects at risk according to BPSS-P criteria with a balanced accuracy of 66.90% (95% CI 59.2-74.6) for 10-fold cross-validation and 61.9% (95% CI 52.0-71.9) for leave-one-site-out. Structural alterations of the hippocampus and amygdala may not be as pronounced in young people at risk; nonetheless, machine learning can predict the estimated risk for BD above chance. This suggests that neural changes may not merely be a consequence of BD and may have prognostic clinical value.

Daring to Feel: Emotion-Focused Psychotherapy Increases Amygdala Activation and Connectivity in Euthymic Bipolar Disorder-A Randomized Controlled Trial.

BACKGROUND: In bipolar disorder (BD), the alternation of extreme mood states indicates deficits in emotion processing, accompanied by aberrant neural function of the emotion network. The present study investigated the effects of an emotion-centered psychotherapeutic intervention on amygdala responsivity and connectivity during emotional face processing in BD. METHODS: In a randomized controlled trial within the multicentric BipoLife project, euthymic patients with BD received one of two interventions over 6 months: an unstructured, emotion-focused intervention (FEST), where patients were guided to adequately perceive and label their emotions (n = 28), or a specific, structured, cognitive behavioral intervention (SEKT) (n = 31). Before and after interventions, functional magnetic resonance imaging was conducted while patients completed an emotional face-matching paradigm (final functional magnetic resonance imaging sample of patients completing both measurements: SEKT, n = 17; FEST, n = 17). Healthy control subjects (n = 32) were scanned twice after the same interval without receiving any intervention. Given the focus of FEST on emotion processing, we expected FEST to strengthen amygdala activation and connectivity. RESULTS: Clinically, both interventions stabilized patients' euthymic states in terms of affective symptoms. At the neural level, FEST versus SEKT increased amygdala activation and amygdala-insula connectivity at postintervention relative to preintervention time point. In FEST, the increase in amygdala activation was associated with fewer depressive symptoms (r = 0.72) 6 months after intervention. CONCLUSIONS: Enhanced activation and functional connectivity of the amygdala after FEST versus SEKT may represent a neural marker of improved emotion processing, supporting the FEST intervention as an effective tool in relapse prevention in patients with BD.

Boosting the Theory of Mind Network: Specific Psychotherapy Increases Neural Correlates of Affective Theory of Mind in Euthymic Bipolar Disorder.

BACKGROUND: In bipolar disorder, impaired affective theory of mind (aToM) performance and aberrant neural activation in the ToM brain network partly explain social functioning impairments. However, it is not yet known whether psychotherapy of bipolar disorder influences neuroimaging markers of aToM. METHODS: In this study, conducted within the multicentric randomized controlled trial of the BipoLife consortium, patients with euthymic bipolar disorder underwent 2 group interventions over 6 months (mean = 28.45 weeks): 1) a specific, cognitive behavioral intervention (specific psychotherapeutic intervention [SEKT]) (n = 31) targeting impulse regulation, ToM, and social skills and 2) an emotion-focused intervention (FEST) (n = 28). To compare the effect of SEKT and FEST on neural correlates of aToM, patients performed an aToM task during functional magnetic resonance imaging before and after interventions (final functional magnetic resonance imaging sample of pre- and postcompleters, SEKT: n = 16; FEST: n = 17). Healthy control subjects (n = 32) were scanned twice with the same time interval. Because ToM was trained in SEKT, we expected an increased ToM network activation in SEKT relative to FEST postintervention. RESULTS: Both treatments effectively stabilized patients' euthymic state in terms of affective symptoms, life satisfaction, and global functioning. Confirming our expectations, SEKT patients showed increased neural activation within regions of the ToM network, bilateral temporoparietal junction, posterior cingulate cortex, and precuneus, whereas FEST patients did not. CONCLUSIONS: The stabilizing effect of SEKT on clinical outcomes went along with increased neural activation of the ToM network, while FEST possibly exerted its positive effect by other, yet unexplored routes.

The big five model in bipolar disorder: a latent profile analysis and its impact on longterm illness severity.

BACKGROUND: Using a personality typing approach, we investigated the relationship between personality profiles and the prediction of longterm illness severity in patients with bipolar disorder (BD). While previous research suggests associations between BD and traits from the NEO-FFI profiles, the current study firstly aimed to identify latent classes of NEO-FFI profiles, and, secondly, to examine their impact on the longterm prognosis of BD. METHODS: Based on the NEO-FFI profiles of 134 euthymic patients diagnosed with BD (64.2% female, mean age = 44.3 years), successive latent profile analyses were conducted. Subsequently, a subsample (n = 80) was examined prospectively by performing multiple regression analysis of the latent classes to evaluate the longitudinal course of the disease (mean: 54.7 weeks) measured using a modified Morbidity Index. RESULTS: The latent profile analyses suggested a 3-class model typifying in a resilient (n = 68, 51%), vulnerable (n = 55, 41%) and highly vulnerable (n = 11, 8%) class. In the regression analysis, higher vulnerability predicted a higher longterm Morbidity Index (R2 = 0.28). CONCLUSIONS: Subgroups of patients with BD share a number of discrete personality features and their illness is characterized by a similar clinical course. This knowledge is valuable in a variety of clinical contexts including early detection, intervention planning and treatment process.

Individuals at increased risk for development of bipolar disorder display structural alterations similar to people with manifest disease.

In psychiatry, there has been a growing focus on identifying at-risk populations. For schizophrenia, these efforts have led to the development of early recognition and intervention measures. Despite a similar disease burden, the populations at risk of bipolar disorder have not been sufficiently characterized. Within the BipoLife consortium, we used magnetic resonance imaging (MRI) data from a multicenter study to assess structural gray matter alterations in N = 263 help-seeking individuals from seven study sites. We defined the risk using the EPIbipolar assessment tool as no-risk, low-risk, and high-risk and used a region-of-interest approach (ROI) based on the results of two large-scale multicenter studies of bipolar disorder by the ENIGMA working group. We detected significant differences in the thickness of the left pars opercularis (Cohen's d = 0.47, p = 0.024) between groups. The cortex was significantly thinner in high-risk individuals compared to those in the no-risk group (p = 0.011). We detected no differences in the hippocampal volume. Exploratory analyses revealed no significant differences in other cortical or subcortical regions. The thinner cortex in help-seeking individuals at risk of bipolar disorder is in line with previous findings in patients with the established disorder and corresponds to the region of the highest effect size in the ENIGMA study of cortical alterations. Structural alterations in prefrontal cortex might be a trait marker of bipolar risk. This is the largest structural MRI study of help-seeking individuals at increased risk of bipolar disorder.

Impulsivity predicts illness severity in long-term course of bipolar disorder: A prospective approach.

BACKGROUND: Bipolar disorder is a common, severe and chronic mental illness. Despite this, predictors of illness severity remain poorly understood. Impulsivity is reported to be associated with bipolar disorder and aggravating comorbidities. This study therefore sought to examine the predictive value of impulsivity for determining illness severity in euthymic bipolar disorder patients. METHODS: Baseline trait impulsivity of 120 bipolar euthymic patients (81 bipolar disorder I [68%], 80 female [67%]) and 51 healthy controls was assessed using Barratt Impulsiveness Scale 11. The impact of impulsivity on illness severity (measured with morbidity index) was prospectively tested in 97 patients with sufficient follow-up data (average observation time: 54.4 weeks), using linear regression analysis. RESULTS: Barratt Impulsiveness Scale 11 total (ß = 0.01; p < 0.01) and in particular Barratt Impulsiveness Scale 11 attentional subscale scores (ß = 0.04; p < 0.001) predicted illness severity in bipolar disorder, while controlling for other clinical variables. Only age at onset persisted as an additional, but less influential predictor. Barratt Impulsiveness Scale 11 total scores and Barratt Impulsiveness Scale 11 attentional subscale scores were significantly higher in euthymic patients compared to controls. This was not observed for the motor or non-planning subscale scores. LIMITATIONS: The average year-long observation time might not be long enough to account for the chronic course of bipolar disorder. CONCLUSION: Trait impulsivity and particularly attentional impulsivity in euthymic bipolar patients can be strong predictors of illness severity in bipolar disorder. Future studies should explore impulsivity as a risk assessment for morbidity and as a therapeutic target in bipolar disorder patients.

Sexual Dysfunction in Unmedicated Patients with Schizophrenia and in Healthy Controls.

INTRODUCTION: Sexual dysfunction figures prominently in patients with schizophrenia; however, medication effects may play a role. The objective of this case control study was to assess differences in the presence of sexual dysfunction in unmedicated patients with schizophrenia versus healthy controls. METHODS: Sexual dysfunction was assessed using the Derogatis Inventory for Sexual Function self-rating in a cross-sectional design controlling for age and gender effects. A brief sexual anamnesis was applied to describe the psychosocial background of the mostly male sample further. RESULTS: Results show a significant difference with patients reporting more problems in most domains and with a significant correlation between severity of psychosis (Positive and Negative Syndrome Scale total scores) and the impairment of orgasm experience. The study revealed reduced sexual activities and less pleasure during sexual activities of patients. DISCUSSION: This study implies that schizophrenia has an impact on the presence of sexual dysfunction and that sexual dysfunction is partly independent of antipsychotic treatment. Since the sample consisted mostly of first-episode males, conclusions might only be valid for this subgroup.

Improving functional outcome in bipolar disorder: A pilot study on metacognitive training.

BACKGROUND: Effective group psychological interventions in bipolar disorder are rare. In this study, we present "metacognitive training (MCT) for bipolar disorder"-an adaption of a group intervention that has proven effective in other severe psychiatric disorders. MCT is a structured, interactive approach that addresses cognitive biases, social cognition, and self-esteem. In this pilot study, we investigated psychosocial functioning as primary outcome measure, as well as the feasibility of MCT and its acceptance among bipolar patients. METHODS: Thirty-four outpatients with bipolar disorder were recruited. Inclusion criteria were euthymia and psychosocial functioning with a score >11 assessed by the Functional Assessment Short Test. The subjects received eight weekly MCT sessions. Before and after the intervention, psychosocial functioning, quality of life (QoL), and patient views were assessed. RESULTS: Patients improved significantly in global psychosocial functioning, with a large effect size from baseline to post-treatment. Over the intervention period, patient QoL improved significantly in terms of their physical health, however not for other QoL subdomains. Treatment adherence was 80%, and patients' appraisal of the training was positive. LIMITATIONS: As this study lacks a control group, it is not possible to ascertain whether the positive treatment effects are attributable to MCT. Additionally, it is unclear whether gains in psychosocial functioning would have been maintained long term. CONCLUSIONS: This pilot trial conclusively shows that MCT is feasible and provides preliminary evidence for both the acceptance and efficacy of MCT. Further studies with larger samples and control condition will be necessary to build on these findings.

Autobiographical Memory in Bipolar Disorder and Its Link to Neuropsychological Functioning.

OBJECTIVE: The aim of this study was to examine the specificity of autobiographical memory (AM) in bipolar disorder (BD) and to investigate the association between AM and neuropsychological functions. METHOD: Twenty bipolar patients and 22 matched healthy controls (HCs) were included in this study. AM was assessed with an extended version of the Autobiographical Memory Test (AMT) including rejection cue words. A neuropsychological test battery was used to examine verbal memory, executive functions, and attention. RESULTS: Across both groups, the number of specific memories in the AMT was significantly smaller in response to rejection cue words and positive cue words than in response to negative cue words. Participants with BD and HCs did not differ significantly in neuropsychological measures. Across both groups, scores of verbal memory, executive functions, and attention were significantly correlated with specificity of retrieved memories. LIMITATIONS: Although our clinical sample consisted of clinically stable outpatients, 6 out of 20 patients were not rated as euthymic but as mildly depressed. All BD patients were medicated. CONCLUSION: Contrary to previous results, patients with BD did not differ in the number of specific memories compared to an HC group. Our findings suggest that neuropsychological functioning is associated with AMT specificity. Further research is required to gain a better understanding of the underlying mechanisms which may influence the ease of memory retrieval.

Behavioral measures and self-report of impulsivity in bipolar disorder: no association between Stroop test and Barratt Impulsiveness Scale.

BACKGROUND: Impulsivity as a tendency to act quickly without considering future consequences has been proposed as a dimensional factor in bipolar disorder. It can be measured using behavioral tasks and self-report questionnaires. Previous findings revealed patients to show worse performance on at least one behavioral measure of impulsivity. Additionally, self-reported impulsivity seems to be higher among bipolar patients, both parameters being possibly associated with a more severe course of illness. In this study, our primary aim was to investigate the relationship between these two constructs of impulsivity among bipolar patients. METHODS: A total of 40 euthymic patients with bipolar disorder (21 female, 22 Bipolar I) and 30 healthy controls were recruited for comprehensive neuropsychological assessment. To assess inhibition control as a behavioral measure of impulsivity, the Stroop Color and Word Test (Stroop) was used. Additionally, both groups completed the Barratt Impulsiveness Scale (BIS) as a self-report of impulsivity. To compare the groups' performance on the Stroop and ratings on the BIS, the non-parametric Mann-Whitney U test was used. Within the bipolar group, we additionally examined the possibility of an association between Stroop performance and BIS total scores using Pearson's Correlation r. RESULTS: Patients and controls differed significantly on the Stroop and BIS, with patients performing worse on the Stroop and scoring higher on the BIS. However, there was no association between the Stroop and BIS within the bipolar group. As an exploratory analysis, a positive correlation between Stroop performance and number of episodes was found. Further, we detected a statistical trend in the direction of poorer Stroop performance among patients treated with polypharmacy. CONCLUSIONS: Both difficulties with behavioral inhibition and self-reported impulsivity were observed to be higher in bipolar patients than controls in the current study. However, within the patient group we did not observe an association between patients' behavioral performance and self-report. This indicates that the parameters likely constitute distinct, dimensional factors of bipolar disorder. In future research, studies with larger samples should investigate which of the two markers constitutes the better marker for the illness and is more suitable to differentiate the most severe patients.