Objective: To evaluate the effect of echocardiographically determined left ventricular hypertrophy (LVH) on survival in comparison with number of stenosed vessels and left ventricular systolic dysfunction.
Design: Cohort study based on a consecutive sample from a hospital registry, with a mean follow-up of 5 years.
Setting: An inner-city public hospital in Chicago, Ill.
Patients: The study included 1089 consecutive black patients who underwent both coronary angiography and M-mode echocardiography as part of a diagnostic evaluation.
Results: Nonstenosed coronary arteries, single-vessel disease, and multivessel disease were found in 48%, 16%, and 36% of patients, respectively; LVH (left ventricular mass index > 131 g/m2 in men and > 100 g/m2 in women) was detected in 50% of patients. Hypertrophy without coexistent obstructive coronary disease was associated with a lower survival rate than that observed for single-vessel disease and was similar to multivessel disease. When LVH, number of diseases vessels, and left ventricular dysfunction were subjected to multivariate analysis, hypertrophy conferred a relative risk (RR) of 2.4 (95% confidence interval [CI], 1.7 to 3.2). By comparison, the presence of a single stenosed vessel did not increase the risk of death. Multivessel disease and ejection fraction less than 45% were associated with an RR of 1.6 (95% CI, 1.1 to 2.2) and 2.0 (95% CI, 1.4 to 2.7), respectively. Calculation of the attributable risk fraction demonstrated that for every 100 deaths in this cohort, LVH independently accounted for 37. The corresponding attributable risk fractions were 1%, 22%, and 9% for single-vessel disease, multivessel disease, and ventricular dysfunction, respectively.
Conclusions: Left ventricular hypertrophy was associated with a greater RR and attributable risk than the traditional measures of coronary disease severity. The high prevalence and powerful risk of LVH make an important contribution to the adverse survival rates among black patients with heart disease and may account for much of the black-white differential.