Improved estimation of the risk of manic relapse by combining clinical and brain scan data

INTRODUCTION

Estimating the risk of manic relapse could help the psychiatrist individually adjust the treatment to the risk. Some authors have attempted to estimate this risk from baseline clinical data. Still, no studies have assessed whether the estimation could improve by adding structural magnetic resonance imaging (MRI) data. We aimed to evaluate it.

MATERIAL AND METHODS

We followed a cohort of 78 patients with a manic episode without mixed symptoms (bipolar type I or schizoaffective disorder) at 2-4-6-9-12-15-18 months and up to 10 years. Within a cross-validation scheme, we created and evaluated a Cox lasso model to estimate the risk of manic relapse using both clinical and MRI data.

RESULTS

The model successfully estimated the risk of manic relapse (Cox regression of the time to relapse as a function of the estimated risk: hazard ratio (HR)=2.35, p=0.027; area under the curve (AUC)=0.65, expected calibration error (ECE)<0.2). The most relevant variables included in the model were the diagnosis of schizoaffective disorder, poor impulse control, unusual thought content, and cerebellum volume decrease. The estimations were poorer when we used clinical or MRI data separately.

CONCLUSION

Combining clinical and MRI data may improve the risk of manic relapse estimation after a manic episode. We provide a website that estimates the risk according to the model to facilitate replication by independent groups before translation to clinical settings.

Differential failure to deactivate the default mode network in unipolar and bipolar depression

OBJECTIVES

Neuroimaging studies have revealed evidence of brain functional abnormalities in bipolar depressive disorder (BDD) and major depressive disorder (MDD). However, few studies to date have compared these two mood disorders directly.

METHODS

Matched groups of 26 BDD type I patients, 26 MDD patients and 26 healthy controls underwent functional magnetic resonance imaging (fMRI) while performing the n-back working memory task. A whole-brain ANOVA was used to compare the three groups and clusters of significant difference were examined further using region-of-interest (ROI) analysis.

RESULTS

The whole-brain ANOVA revealed a single cluster of significant difference in the medial frontal cortex. The BDD and MDD patients both showed failure to deactivate in this area compared to the controls. The BDD patients showed significantly greater failure of deactivation than the MDD patients, which was not accounted for by differences in severity or chronicity of illness between them.

CONCLUSIONS

Failure of deactivation, considered to reflect default mode network dysfunction, is present to a greater extent in bipolar than unipolar depression. The study of this network may be useful in the search for brain markers that distinguish the two disorders.

Evaluation of machine learning algorithms and structural features for optimal MRI-based diagnostic prediction in psychosis

A relatively large number of studies have investigated the power of structural magnetic resonance imaging (sMRI) data to discriminate patients with schizophrenia from healthy controls. However, very few of them have also included patients with bipolar disorder, allowing the clinically relevant discrimination between both psychotic diagnostics. To assess the efficacy of sMRI data for diagnostic prediction in psychosis we objectively evaluated the discriminative power of a wide range of commonly used machine learning algorithms (ridge, lasso, elastic net and L0 norm regularized logistic regressions, a support vector classifier, regularized discriminant analysis, random forests and a Gaussian process classifier) on main sMRI features including grey and white matter voxel-based morphometry (VBM), vertex-based cortical thickness and volume, region of interest volumetric measures and wavelet-based morphometry (WBM) maps. All possible combinations of algorithms and data features were considered in pairwise classifications of matched samples of healthy controls (N = 127), patients with schizophrenia (N = 128) and patients with bipolar disorder (N = 128). Results show that the selection of feature type is important, with grey matter VBM (without data reduction) delivering the best diagnostic prediction rates (averaging over classifiers: schizophrenia vs. healthy 75%, bipolar disorder vs. healthy 63% and schizophrenia vs. bipolar disorder 62%) whereas algorithms usually yielded very similar results. Indeed, those grey matter VBM accuracy rates were not even improved by combining all feature types in a single prediction model. Further multi-class classifications considering the three groups simultaneously made evident a lack of predictive power for the bipolar group, probably due to its intermediate anatomical features, located between those observed in healthy controls and those found in patients with schizophrenia. Finally, we provide MRIPredict (https://www.nitrc.org/projects/mripredict/), a free tool for SPM, FSL and R, to easily carry out voxelwise predictions based on VBM images.

Prevalence of cavum vergae in psychosis and mood spectrum disorders

BACKGROUND

Midline brain abnormalities might increase susceptibility to both first-episode and chronic mental disorder. Evidence of cavum vergae (CV) abnormality in mental disorders is scarce.

METHODS

The presence of CV was assessed by a researcher blind to clinical information in a cross-disorder sample of 639 patients with mood and psychotic disorders and in 223 healthy controls. Homogeneous magnetic resonance imaging methods of acquisition and assessment were applied.

RESULTS

Seven out of 639 patients with mood or psychotic disorders were detected with CV which corresponds to a prevalence of 1.1%. There were no concurrent cases of CV in the healthy control group. Identified cases which are briefly described were diagnosed from bipolar I disorder (n=2), delusional disorder (n=1), brief psychotic disorder (n=1) and schizoaffective disorder (n=3). Patients with CV had descriptively lower current IQ, executive functioning and memory scores in relation to patients without CV but this was not statistically significant.

LIMITATIONS

Effects of medication and lack of statistical power of the CV patient group.

CONCLUSIONS

Midline brain abnormalities, such as CV, might represent an unspecific risk factor for the development of severe mental disorders.

Brain functional changes across the different phases of bipolar disorder

BACKGROUND

Little is known about how functional imaging changes in bipolar disorder relate to different phases of the illness.

AIMS

To compare cognitive task activation in participants with bipolar disorder examined in different phases of illness.

METHOD

Participants with bipolar disorder in mania (n = 38), depression (n = 38) and euthymia (n = 38), as well as healthy controls (n = 38), underwent functional magnetic resonance imaging during performance of the n-back working memory task. Activations and de-activations were compared between the bipolar subgroups and the controls, and among the bipolar subgroups. All participants were also entered into a linear mixed-effects model.

RESULTS

Compared with the controls, the mania and depression subgroups, but not the euthymia subgroup, showed reduced activation in the dorsolateral prefrontal cortex, the parietal cortex and other areas. Compared with the euthymia subgroup, the mania and depression subgroups showed hypoactivation in the parietal cortex. All three bipolar subgroups showed failure of de-activation in the ventromedial frontal cortex. Linear mixed-effects modelling revealed a further cluster of reduced activation in the left dorsolateral prefrontal cortex in the patients; this was significantly more marked in the mania than in the euthymia subgroup.

CONCLUSIONS

Bipolar disorder is characterised by mood state-dependent hypoactivation in the parietal cortex. Reduced dorsolateral prefrontal activation is a further feature of mania and depression, which may improve partially in euthymia. Failure of de-activation in the medial frontal cortex shows trait-like characteristics.

Transcultural adaption and validation of the Spanish version of the Bipolar Depression Rating Scale (BDRS-S)

BACKGROUND

The Bipolar Depression Rating Scale (BDRS) arguably better captures symptoms in bipolar depression especially depressive mixed states than traditional unipolar depression rating scales. The psychometric properties of the Spanish adapted version, BDRS-S, are reported.

METHODS

The BDRS was translated into Spanish by two independent psychiatrists fluent in English and Spanish. After its back-translation into English, the BDRS-S was administered to 69 DSMI-IV bipolar I and II patients who were recruited from two Spanish psychiatric hospitals. The Hamilton Depression Rating Scale (HDRS), the Montgomery-Asberg Depression Rating Scale (MADRS) and the Young Mania Rating Scale (YMRS) were concurrently administered. 42 patients were reviewed via video by four psychiatrists blind to the psychopathological status of those patients. In order to assess the BDRS-S intra-rater or test-retest validity, 22 subjects were assessed by the same investigator performing two evaluations within five days.

RESULTS

The BDRS-S had a good internal consistency (Cronbach׳s α=0.870). We observed strong correlations between the BDRS-S and the HDRS (r=0.874) and MADRS (r=0.854) and also between the mixed symptom cluster score of the BDRS-S and the YMRS (r=0.803). Exploratory factor analysis revealed a three factor solution: psychological depressive symptoms cluster, somatic depressive symptoms cluster and mixed symptoms cluster.

LIMITATIONS

A relatively small sample size for a 20-item scale.

CONCLUSIONS

The BDRS-S provides solid psychometric performance and in particular captures depressive or mixed symptoms in Spanish bipolar patients.

Bipolar depressed patients show both failure to activate and failure to de-activate during performance of a working memory task

BACKGROUND

Bipolar depression has been found to be associated with changes in prefrontal cortex activity during performance of cognitive tasks. However, the role of task-related de-activations has been little investigated.

METHOD

Forty-one bipolar depressed patients and 41 matched normal controls underwent fMRI scanning while performing baseline, 1-back and 2-back versions of the n-back task. Linear models were used to obtain maps of within-group activations and areas of differential activation between the groups.

RESULTS

The bipolar depressed patients showed reduced activation in the dorsolateral prefrontal cortex (DLPFC) bilaterally and several other regions. After controlling for differences in task performance only differences in the DLPFC and cerebellum remained. Left DLPFC activation was inversely correlated with Hamilton and MADRS scores. The patients showed failure to de-activate in the medial prefrontal cortex, an area corresponding to the anterior medial node of the default mode network.

LIMITATIONS

To confirm default mode network dysfunction demonstration of resting-state connectivity abnormalities would also be required. The study was carried out on treated patients, and did not assess for presence of depressive symptoms in the healthy controls.

CONCLUSIONS

Both prefrontal cortical and default mode network dysfunction appear to characterise bipolar depression. The former, but not the latter, is associated with symptom severity.

ECT in a patient with Parkinson's disease and schizophrenia, with dopamine transporter visualisation using 123I-ioflupane SPET

A 58-year old woman with chronic schizophrenia developed worsening parkinsonian symptoms over the previous 6 years, and was eventually diagnosed as having Parkinson's disease. Antipsychotics were stopped because they worsened these symptoms. Antiparkinsonian treatment led to a significant increase in delusions and behavioural disorganisation. The patient underwent electroconvulsive-therapy which improved both her psychiatric and motor symptoms. After treatment, 123I-Ioflupane uptake was mildly increased in the left caudate nucleus, but uptake in right caudate nucleus was lower than in a pretreatment scan.