Self-reported depressive symptoms, diagnosed clinical depression and cardiac morbidity and mortality after myocardial infarction

Background
Self-reported depressive symptoms and clinical depression after myocardial infarction (MI) are both associated with poor cardiac prognosis. It is important to distinguish between the two when assessing cardiac prognosis, but few studies have done so. The present article evaluates the independent prognostic impact of self-reported depressive symptoms and clinical depression on cardiac outcomes after MI.

Methods
2704 MI-patients were administered the Beck Depression Inventory (BDI) and underwent the Composite International Diagnostic Interview at 3 months post-MI. All-cause mortality, cardiac mortality and cardiovascular readmissions were evaluated up till 10 years post-MI (mean: 6 years), representing 16,783 persons-years of follow-up. Event-free survival was evaluated using Cox regression analysis.

Results
Analyses on mortality and cardiovascular readmissions included 2493 and 2434 patients respectively. Compared to patients scoring < 5 on the BDI, those scoring ≥ 19 had age- and sex-adjusted HR's (95% CI) of 3.20 (2.16-4.74, p < 0.001) for all-cause mortality, 3.97 (2.06-7.65, p < 0.001) for cardiac mortality, and 1.45 (1.08-1.95, p < 0.05) for cardiovascular readmissions. Cardiac disease severity and cardiac risk factors explained one third to half of the relationship. The presence of clinical depression was associated with all-cause (HR: 1.72 (1.29-2.30, p < 0.001)) and cardiac mortality (HR: 1.67 (1.01-2.77, p < 0.05)). However, adjusting for BDI-scores decreased these HR's with 53% and 72% respectively, rendering them non-significant. Dichotomized BDI-scores remained to predict cardiac prognosis independently from the presence of clinical depression.

Conclusions
After MI, self-reported depressive symptoms are a more accurate predictor of cardiac morbidity and mortality than clinical depression. This association is confounded largely by cardiac disease severity.