Does Comorbid Alcohol and Substance Abuse Affect Electroconvulsive Therapy Outcome in the Treatment of Mood Disorders?

Abstract

Antidepressant medications remain the principal agents used to treat patients with mood disorders, although 30% to 40% of these patients do not improve. One of the factors associated with poor medication response is alcohol and substance abuse. Persons with mood disorders are at the greatest risk for suicide, and alcoholism is a significant additional risk factor. Electroconvulsive therapy (ECT) is shown to be the most effective treatment for major depression especially when associated with psychosis, catatonia, and suicide intent. However, similar to most antidepressant trials, patients with depression and comorbid alcohol and substance abuse are excluded from ECT efficacy studies. Through a retrospective chart review, we compared response to ECT in patients with mood disorder and comorbid alcohol and drug abuse to those with mood disorder only. From 2004 to 2010, 80 patients with mood disorder received ECT. Fifty of these had comorbid alcohol or drug abuse. Using a 10-item psychopathology scale, we compared pre- and post-ECT symptom severity between the 2 groups. Outcome was determined by measuring a decrease in the pre-ECT and post-ECT score using Wilcoxon rank tests, with statistical significance at P = 0.05. There was no difference between the 2 groups in most demographics, ECT medication, or seizure quality. There was no difference in ECT outcome between those with comorbid alcohol abuse and those without based on percent decrease in pre- and post-ECT symptom scores (abuse: mean [SD], 0.89 [0.2] vs nonabuse: mean [SD], 0.93 [0.16]; Wilcoxon, 1332; P = 0.086). When we compared those who met the criteria for alcohol or drug dependence (19 patients) with those with no abuse, there was a trend for the dependence group to not do as well (dependence: mean [SD], 0.83 [0.25] vs nonabuse: mean [SD], 0.93 [0.16]; Wilcoxon, 405; P = 0.053). Those with combined drug and alcohol abuse (18 patients) did have a significantly worse outcome (combined: mean [SD], 0.82 [0.25] vs nonabuse: mean [SD], 0.93 [0.16]; Wilcoxon, 372; P = 0.033). Our results indicate that comorbid alcohol and drug abuse may influence the response to ECT in the treatment of mood disorders. Based on our results, patients with comorbid dependence and combined drug and alcohol abuse showed symptom improvement but did not do as well as those with nonabuse.