Objectives:Clinically useful predictors of response to electroconvulsive therapy (ECT) are warranted, especially in the case of bipolar depression. The aim of this study was to explore the associations between response and its known and putative correlates. Methods:Six hundred seventy bipolar depressive patients treated with ECT were included in the study. The association between response (CGI-I <= 2) and mean seizure duration, number of treatments, age, sex, bipolar subtype, episode duration, HAM-D and YMRS scores, psychomotor disturbances and psychotic symptoms assessed through BPRS-EV were evaluated by means of univariate and multivariate logistic regression models, including quadratic and/or linear effects of continuous variables. Results:Four hundred eighty three patients (72%) were responders. Among known correlates of response, significant quadratic effects were found for seizure duration and number of treatments, while a linear association was confirmed for episode duration. Among putative correlates, severe motor retardation, tension or agitation, hyperactivity and delusions of guilt were significantly associated with response (p<.01) and a significant quadratic effect was found for YMRS score (p<.01). Conclusion:Bipolar depressive patients with severe psychomotor disturbances, mood-congruent delusions and severe mixed features are highly responsive to ECT. A significant improvement in response prediction is expected when considering those clinical characteristics.
