Episode length and mixed features as predictors of ECT nonresponse in patients with medication-resistant major depression.

This study aimed to ascertain predictors of nonresponse to electroconvulsive therapy (ECT) in a large sample of major depressive patients resistant to pharmacologic treatment. A total of 208 depressive patients (31 with major depression [UP], 101 with bipolar disorder II [BP II], and 76 with bipolar disorder I [BP I] according to DSM-IV criteria) were included in the study and treated with bilateral ECT on a twice-a-week schedule. The patients were assessed before (baseline) and a week after the ECT course (final score) using the Hamilton Rating Scale for Depression-17 items (HAM-D-17), the Young Mania Rating Scale (YMRS), the Brief Psychiatric Rating Scale (BPRS), and the Clinical Global Improvement (CGI). Responders were defined as those patients with a reduction of at least 50% in HAM-D-17 score and a rating of 2 ("much improved") or 1 ("very much improved") in the CGI-Improvement subscale. At the end of the ECT course, 152 patients (64%) were classified as responders and 56 patients (36%) were classified as nonresponders. On backward stepwise logistic regression, bipolar subtype (odds ratio [OR]=17.85; 95% confidence level [CL]=1.786-178.407), higher mean baseline YMRS scores (OR=1.094; 95% CL=1.025-1.166), lower mean baseline HAM-D-17 scores (OR=0.928; 95% CL=0.860-1.002), and length of current episode (OR=1.047; 95% CL=1.009-1.086) were identified as statistically significant predictors of nonresponse. ECT was an effective treatment for approximately two-thirds of the patients with medication-resistant depression who were included in this study. ECT nonresponse was associated with bipolar subtype, presence of manic symptoms during depression, slightly less severe depressive symptomatology, and protracted duration of the episode.