De Schutter, Alban; Lavie, Carl J.; Milani, Richard V.
Background: Previous research has demonstrated a high prevalence of psychological risk factors in patients with coronary heart disease (CHD), as well as the benefits of cardiac rehabilitation (CR) on psychological distress (PD) in showing its potential to improve mortality. We examined the impact of CR on mortality by anxiety and/or hostility symptoms in a large population of CHD patients with symptoms of depression following CR. Patients and Methods: We studied 538 patients with CHD who had completed formal CR. Using a validated questionnaire, symptoms of PD were obtained in 3 domains: anxiety, hostility, and depression. Subjects were divided into 3 groups: nondepressed (n = 502), depression alone (n = 14), and depression with anxiety and/or hostility (n = 22). A multivariate analysis for mortality was performed using a composite PD score (PD = sum of scores for anxiety, depression, and hostility). Subjects were analyzed by total mortality over 3-year follow-up by the National Death Index. Results: Mortality was significantly higher in the group with depressive symptoms compared with those without depressive symptoms (19% vs 3%; P < 0.0001). The comorbid depressed group had a slight trend toward higher mortality (22.7% [5 of 22 patients] vs 14% [2 of 14 patients]; P = 0.52). After adjusting for age, body mass index (BMI), ejection fraction, exercise tolerance, and sex using Cox proportional regression, the number of psychiatric comorbidities (ie, depression, anxiety, and hostility), as well as the sum of their raw scores, were significantly associated with increased mortality. However, this effect disappeared after adjusting for depression score (comorbidity hazard ratio, 1.7; 95% confidence interval, 1.0-2.8 vs comorbidity hazard ratio, 1.2; 95% confidence interval, 0.4-3.7 after adjusting). Conclusion: Psychological distress is an independent predictor of mortality in stable CHD patients following CR; although anxiety and hostility may also modulate this effect, the overall impact seems to be mostly mediated through depression. Patients with persistent depression following CR may need further intervention.