Discriminating between bipolar and major depressive disorder using a machine learning approach and resting-state EEG data.

OBJECTIVE: Distinguishing major depressive disorder (MDD) from bipolar disorder (BD) is a crucial clinical challenge as effective treatment is quite different for each condition. In this study electroencephalography (EEG) was explored as an objective biomarker for distinguishing MDD from BD using an efficient machine learning algorithm (MLA) trained by a relatively large and balanced dataset. METHODS: A 3 step MLA was applied: (1) a multi-step preprocessing method was used to improve the quality of the EEG signal, (2) symbolic transfer entropy (STE), an effective connectivity measure, was applied to the resultant EEG and (3) the MLA used the extracted STE features to distinguish MDD (N = 71) from BD (N = 71) subjects. RESULTS: 14 connectivity features were selected by the proposed algorithm. Most of the selected features were related to the frontal, parietal, and temporal lobe electrodes. The major involved regions were the Broca region in the frontal lobe and the somatosensory association cortex in the parietal lobe. These regions are near electrodes FC5 and CPz and are involved in processing language and sensory information, respectively. The resulting classifier delivered an evaluation accuracy of 88.5% and a test accuracy of 89.3%, using 80% of the data for training and evaluation and the remaining 20% for testing, respectively. CONCLUSIONS: The high evaluation and test accuracies of our algorithm, derived from a large balanced training sample suggests that this method may hold significant promise as a clinical tool. SIGNIFICANCE: The proposed MLA may provide an inexpensive and readily available tool that clinicians may use to enhance diagnostic accuracy and shorten time to effective treatment.

A Machine Learning Algorithm to Discriminating Between Bipolar and Major Depressive Disorders Based on Resting EEG Data.

Distinguishing major depressive disorder (MDD) from bipolar disorder (BD) is a crucial clinical challenge due to the lack of known biomarkers. Conventional methods of diagnosis rest exclusively on symptomatic presentation, and personal and family history. As a result, BD-depressed episode (BD-DE) is often misdiagnosed as MDD, and inappropriate therapy is given. Electroencephalography (EEG) has been widely studied as a potential source of biomarkers to differentiate these disorders. Previous attempts using machine learning (ML) methods have delivered insufficient sensitivity and specificity for clinical use, likely as a consequence of the small training set size, and inadequate ML methodology. We hope to overcome these limitations by employing a training dataset of resting-state EEG from 71 MDD and 71 BD patients. We introduce a robust 3 steps ML technique: 1) a multi-step preprocessing method is used to improve the quality of the EEG signal 2) symbolic transfer entropy (STE), which is an effective connectivity measure, is applied to the resultant EEG signals 3) the ML algorithm uses the extracted STE features to distinguish MDD from BD patients. Clinical Relevance--- The accuracy of our algorithm, derived from a large sample of patients, suggests that this method may hold significant promise as a clinical tool. The proposed method delivered total accuracy, sensitivity, and specificity of 84.9%, 83.4%, and 87.1%, respectively.

Evolved mechanisms in depression: the role and interaction of attachment and social rank in depression.

Evolved mechanisms underpinning attachment and social rank behavior may be the basis for some forms of major depression, especially those associated with chronic stress. We note the heterogeneity of depression, but suggest that some of its core symptoms, such as behavioral withdrawal, low self-esteem and anhedonia, may have evolved in order to regulate behavior and mood and convey sensitivity to threats and safety. Focusing on the evolved mental mechanisms for attachment and social rank helps to make sense of (1) depression's common early vulnerability factors (e.g., attachment disruptions, neglect and abuse), (2) the triggering events (e.g., loss of close relationships, being defeated and/or trapped in low socially rewarding or hostile environments), and (3) the psychological preoccupations of depressed people (e.g., sense of unlovableness, self as inferior and a failure). This focus offers clues as to how these two systems interact and on how to intervene.

Transcranial magnetic stimulation in the treatment of mood disorder: a review and comparison with electroconvulsive therapy.

OBJECTIVE: To review repetitive transcranial magnetic stimulation (rTMS) as a mode of therapy for depression. METHOD: The following aspects of rTMS were reviewed and compared with electroconvulsive therapy (ECT): history, basic principles, technical considerations, possible mode of action, safety, adverse effects, and effects on mood in both healthy individuals and those suffering from bipolar disorder (BD) or depression. RESULTS: rTMS may selectively increase or decrease neuronal activity over discrete brain regions. As a result of this focused intervention with TMS, the potential for unwanted side effects is substantially reduced, compared with ECT. In open trials, rTMS and ECT are reported to be equally efficacious for patients having depression without psychosis, but the therapeutic benefits reported in double-blind sham-rTMS controlled trials are more modest. CONCLUSION: The antidepressant and antimanic effects of rTMS depend on technical considerations such as stimulus frequency, intensity, and magnetic coil placement, which may not yet be optimized. Biological heterogeneity among the patients treated with rTMS may also contribute to differing efficacy across clinical trials. rTMS may possess tremendous potential as a treatment for mood disorder, but this has not yet been realized. rTMS must still be regarded as an experimental intervention requiring further refinement.

Gabapentin as an adjunctive treatment in bipolar disorder.

OBJECTIVE: To evaluate the efficacy of gabapentin as an adjunctive treatment for bipolar disorder in both depressed and manic phases. METHOD: Thirty seven patients with bipolar type I or II with or without a rapid cycling course were openly treated with gabapentin added to current treatment for up to six months. Mood symptoms were rated weekly for 12 weeks then monthly for 3 months utilizing the HamD and YMS. RESULTS: Participants experienced a significant reduction in both depressive and manic symptoms. CONCLUSIONS: These findings are consistent with others in establishing the efficacy of gabapentin in both phases of bipolar disorder. LIMITATIONS: Small sample size and the use of an open uncontrolled design limit interpretation of results.

Comorbidity of obsessive compulsive disorder in bipolar disorder.

The comorbidity of OCD and bipolar disorder has not been systematically examined. Therefore, we determined the frequency of patients meeting DSM-III criteria for OCD syndrome in a sample of 149 inpatients with DSM-III major affective disorder who had received a clinically reviewed structured diagnostic interview. The frequency of OCD syndrome was not significantly different between subjects with major depression (35.2%, n = 105) and bipolar disorder (35.1%, n = 37). This suggests that OCD is equally common in bipolar as in unipolar patients.

The symptom structure of panic attacks in depressed and anxious patients.

This study examined the panic symptom profiles of three diagnostic groups: those with panic disorder and no history of major depression; those with panic disorder with a history of major depressive episode but no current depression; and those current major depression with panic disorder. Patients were compared on the frequency of specific panic attack symptoms based on structured interview responses. The symptom profiles of all three groups were significantly correlated. The patients with past and current depressive episodes had the most similar symptom structure.

The cholinergic-adrenergic hypothesis of depression reexamined using clonidine, metoprolol, and physostigmine in an animal model.

The role of central nervous system (CNS) cholinergic and noradrenergic mechanisms in the pathogenesis of depression and hypothalamic-pituitary-adrenal (HPA) axis hyperactivity is examined using the Behavioral Despair rat model of depression. Immobility (IM), the analog of depression in this model, and plasma corticosterone (C) were increased by physostigmine (PHYSO). Neostigmine (NEO), which does not cross the blood-brain barrier, produced the same peripheral cholinomimetic effects and motor inhibition as PHYSO, but did not change IM. PHYSO's effects on C and IM were blocked by metoprolol pretreatment and partially blocked by clonidine pretreatment. PHYSO increased acetylcholine in the striatum. In this animal model of depression, cholinergic and noradrenergic mechanisms are interactively involved in the regulation of behavioral depression and the HPA axis.

Effect of concurrent medical illness on dexamethasone suppression test results in depressed inpatients.

A retrospective evaluation of the clinical records of 138 depressed patients, who received the dexamethasone suppression test (DST) as part of a standardized physical and psychiatric assessment protocol, revealed that 60 had acute, chronic (mild or severe), stable or remitted medical conditions. The proportion of DST nonsuppressors did not differ between depressed subjects with medical conditions (45% nonsuppressors) and those without (34.6% nonsuppressors; p greater than 0.2). However, all of the six subjects with acute or chronic-severe medical conditions were found to be nonsuppressors (p = 0.003). These results may help clarify the medical exclusion criteria for the clinical application of the DST.

Plasma corticosterone is increased and correlated with brain acetylcholine in physostigmine- but not in neostigmine-treated rats.

Rats were given intraperitoneal injections of physostigmine (PHYSO), neostigmine (NEO) or saline (SAL). Either 15 or 30 min later the number and intensity of observable cholinomimetic effects (OCE) was determined, plasma was collected for corticosterone (Cst) measurement, and the cerebral cortex, striatum, hippocampus and hypothalamus were removed after microwave treatment for the measurement of tissue acetylcholine (ACh) and choline (Ch) concentrations. Plasma Cst correlated with the number of OCEs at both 15 and 30 min in both NEO- and PHYSO-treated animals. Although the number and intensity of OCE were the same in NEO- and PHYSO-treated animals 15 min after injection, plasma Cst was significantly higher in the PHYSO-treated group. ACh levels in the cortex were also increased in PHYSO- compared with NEO-treated animals 15 min after injection. Ch levels remained unchanged. Plasma Cst correlated positively with ACh levels in the cortex and striatum in PHYSO-treated rats both 15 and 30 min after injection. These data support the involvement of central cholinergic mechanisms in the regulation of the HPA axis.

Epstein-Barr virus as a cause of autoimmune disease and other medical morbidity in patients with affective disorders.

We hypothesize that psychiatric patients suffering from the major affective disorders (depression and manic-depressive illness) may commonly also suffer from a chronic active infection with the Epstein-Barr virus. This infection would be a consequence of the immune dysfunction known to be associated with these disorders of mood. According to this hypothesis, the increased medical morbidity and mortality reported in these psychiatric patients would be attributable in part to diseases in which Epstein-Barr virus is implicated or suspected as a cause.

Neurotransmitter metabolites and endocrine responses in depression.

Urinary 3-methoxy-4-hydroxyphenylethyleneglycol (MHPG), 3-4-dihydroxyphenylethyleneglycol (DHPG), 5-hydroxyindoleacetic acid (5-HIAA), plasma thyroid stimulating hormone (TSH), prolactin (PRL) and growth hormone (GH) were measured before and after the injection of thyrotropin releasing hormone (TRH) in healthy subjects and depressed patients with primary affective disorder. The TSH response to TRH did not differ in depressed compared with control subjects. A trend (.05 less than p less than .10) toward a lower PRL response appeared in male depressed compared with male control subjects. GH levels did not consistently change after TRH. In all subjects the TSH response correlated positively with pre- and post-TRH urinary MHPG. The PRL response correlated negatively with pre-TRH urinary 5-HIAA. Pre-TRH daytime urinary 5-HIAA levels were elevated in depressed subjects.

Elevated 3,4-dihydroxyphenylethyleneglycol (DHPG) excretion in dexamethasone resistant depressed patients.

Urinary and plasma DHPG and MHPG were estimated in patients with MADD and showing DST non-suppression as compared to those with normal suppression. Day and night 12h urine collections and morning plasma samples were analyzed by GC-MS for total MHPG and DHPG and free MHPG levels. Urinary DHPG excretion was significantly elevated in DST non-suppressors compared to suppressors, but no differences were found in urinary MHPG or plasma glycol levels. Elevated DHPG excretion in DST non-suppressors suggests that increased peripheral sympathetic NE activity occurs in association with dexamethasone resistance in MADD.