Comparison of associated features and drug treatment between co-occurring unipolar and bipolar disorders in depressed eating disorder patients

BACKGROUND: To examine the differences of associated characteristics and prescription drug use between co-occurring unipolar and bipolar disorders in patients with eating disorders (EDs). METHODS: Patients with EDs and major depressive episode (MDE) were recruited from psychiatric outpatient clinics. They were interviewed and completed self-administered measures assessing eating and general psychopathology. The prescribed drugs at the index outpatient visit were recorded. Clinical characteristics and prescription drugs of groups with major depressive disorder (ED-MDD), MDE with lifetime mania (ED-BP I), and MDE with lifetime hypomania (ED-BP II) were compared. Continuous variables between groups were compared using generalized linear regression with adjustments of age, gender, and ED subtype for pair-wise comparisons. Multivariate logistic regression with adjustments of age, gender, and ED subtype was employed to estimate adjusted odds ratios with 95% confidence intervals between groups. RESULTS: Two hundred and twenty-seven patients with EDs had a current MDE. Among them, 17.2% and 24.2% experienced associated manic and hypomanic episodes, respectively. Bipolar I and II patients displayed significantly poorer weight regulation, more severe impulsivity and emotional lability, and higher rates of co-occurring alcohol use disorders than ED-MDD patients. ED-BP I patients were found to have the lowest IQ, poorest working memory, and the most severe depression, suicidality and functional impairment among all patients. Patients with ED-BP II shared affect and behavioral dysregulations with ED-BP I, but had less severe degrees of cognitive and functional impairments than ED-BP I. Patients with ED-BP I were significantly less likely than those in the ED-MDD and ED-BP II groups to be on antidepressant monotherapy, but a great rate (27%) of ED-BP I individuals taking antidepressant monotherapy had potential risk of mood switch during the course of treatment. CONCLUSIONS: Our study identified discriminative features of bipolar I and II disorders from MDD among a group of depressed ED patients. We suggest that the associated mania, hypomania, and mood lability are predictors of clinical severity and should be identified from ED patients presented with depressive features. Accurate diagnosis of bipolar disorders may have implications for pharmacotherapy in patients with EDs. ELECTRONIC SUPPLEMENTARY MATERIAL: The online version of this article (doi:10.1186/s12888-017-1243-0) contains supplementary material, which is available to authorized users.