Clinical predictors of short-term outcome in electroconvulsive therapy.

Abstract

Electroconvulsive therapy (ECT) remains one of the most effective biological treatments for major depression. However, there is little information on the clinical use of ECT, and most studies were conducted before the introduction of newer antidepressants and before improvements in ECT delivery. This study examined ECT use in a university teaching hospital to determine predictors of short-term ECT outcome. This was a retrospective chart review of ECT over the period 1994-1996. Data extracted from the chart included demographic information, clinical features of depression, and documented antidepressant trials. Outcome measures, based on the chart notes, included Clinical Global Impression (CGI) and cognitive side effects of ECT at 1 week post-ECT or at discharge if sooner. Of 174 patients who received ECT, 130 had a diagnosis of unipolar major depressive disorder. Of these 130 patients, 92% were refractory to at least 1 antidepressant medication. After a clinical course of ECT, 87% were rated as "much" or "very much" improved on the CGI. Moderate side effects were noted in 16% of patients, while only 7% had marked side effects. Medication resistance was not related to ECT response. No significant clinical predictors (symptoms, chronicity, number of antidepressant trials) of ECT outcome were found on a stepwise multiple-regression analysis. These results support previous studies showing that ECT results in very good short-term response in major depressive disorder and that the cognitive side effects of ECT are reasonable. Despite the limitations of this study (retrospective, chart review, global measures), the results will inform clinicians who are recommending ECT for their patients.