Antidepressants in Coronary Heart Disease

Abstract

IN THIS ISSUE OF JAMA, LESPERANCE ET AL 1 REPORT THE results of the Canadian Cardiac Randomized Evaluation of Antidepressant and Psychotherapy Efficacy (CREATE) trial, a blinded, randomized controlled clinical trial testing the efficacy of the selective serotonin reuptake inhibitor (SSRI) citalopram and interpersonal psychotherapy in 284 patients with major depressive disorder. The efficacy of citalopram and interpersonal psychotherapy in the treatment of major depression in patients primarily free of coronary heart disease (CHD) is well documented. The clinical importance of CREATE is that it involved patients with comorbid CHD. Depression is a painful, functionally impairing, and frequently recurrent condition that in patients with either coronary or cerebral vascular disease also significantly increases the risk of cardiovascular morbidity and mortality. A recent report found 17 studies indicating that depression following a coronary event was associated with a 3-fold increase in cardiac mortality. Despite the ominous implications of depression occurring in patients with CHD, there is limited information to guide its treatment and questions have even been raised about whether treatment is needed. Some physicians consider major depressive disorder that is detected after an acute coronary syndrome (ACS) episode as an understandable reaction to stress that will remit when the coronary condition stabilizes. However, recent studies show that most major depressive disorders observed after ACS began long before the coronary event, which therefore could not have provoked the depression. In 1993, Frasure-Smith et al first raised the question of whether treating depression after ACS would reduce cardiac morbidity and mortality. At that time, neither safety nor efficacy data were available for any antidepressant during the post-ACS period. The Sertraline Antidepressant Heart Attack Randomized Trial (SADHART) tested the safety and antidepressant efficacy of sertraline in patients with ACS as a step to test mortality reduction. SADHART found no adverse cardiovascular effects associated with sertraline treatment and reported that depressive symptoms improved in patients whose depression was severe or began before the coronary event or those who had a history of prior episodes. Among patients whose depressive episode was mild or first began after their coronary event, placebo response rates were high with no significant drug-placebo difference. SADHART, with 369 patients, was not powered to determine whether sertraline reduced death or recurrent myocardial infarction (MI). The Enhancing Recovery in Coronary Heart Disease (ENRICHD) trial, which randomized 1834 depressed, post-MI patients, was adequately powered to test whether cognitive behavioral therapy would reduce death compared with usual care. Treatment with cognitive behavioral therapy reduced depression modestly but not death or recurrent MI. For ethical reasons, patients enrolling with more severe depression or failing to respond to cognitive behavioral therapy were offered an antidepressant drug. Approximately 20% received an SSRI, usually sertraline. Patients treated with an SSRI, whether assigned to receive cognitive behavioral therapy or usual care, had a 42% reduction in death or recurrent MI compared with the 1388 depressed patients not receiving an antidepressant. Although not based on randomized treatment, this difference was highly statistically significant. The CREATE trial documents the antidepressant efficacy of a second SSRI, citalopram, in a comorbid CHD population. Although CREATE, ENRICHD, and SADHART all enrolled patients meeting criteria for major depressive disorder, half the participants in ENRICHD and SADHART had less severe depression than that required for enrollment in CREATE. The majority of patients participating in SADHART and ENRICHD began treatment in the first month after experiencing ACS when risk of cardiovascular events is high. In contrast, patients in the CREATE trial began treatment on average more than 18 months after their index event. Despite these differences, SADHART and CREATE found similar drug-response characteristics. CREATE limited enrollment to patients with moderate to severe depression and found citalopram had a response rate of 31% greater than