Low dehydroepiandrosterone sulfate and heart disease in middle-aged men: Cross-sectional results from the Massachusetts male aging study

PURPOSE: Serum concentrations of the adrenal androgen dehydroepiandrosterone (DHEA) and its sulfate ester (DHEAS), both of which decline with age more markedly than other sex hormone levels, have been alternately credited and discredited as "protective" correlates of heart disease. Baseline data from the Massachusetts Male Aging Study (MMAS), collected in 1987-89, provided a large population-based random sample (n = 1709) in which to examine cross-sectionally the relation of DHEA and DHEAS levels to heart disease, while controlling for a comprehensive set of potential confounders including serum lipid and hormone levels as well as smoking, alcohol intake, obesity, hypertension, diabetes, diet, medication, physical activity, and psychological measures. METHODS: The subjects were men aged 40-70 years, randomly sampled from the Massachusetts state census listing, measured and interviewed at home. Nonfasting blood samples were assayed for hormones and lipids. RESULTS: In all strata of age, smoking, and alcohol intake, the age-adjusted odds ratio for self-reported heart disease was between 0.55 and 0.85 per standard deviation (SD) of log DHEA and DHEAS concentration. Multiple logistic regression analysis indicated a strong independent role for DHEAS as a predictor of self-reported heart disease, controlling for age and the potential confounders listed above. The multiply-adjusted odds ratio for heart disease was 0.64 per SD log DHEAS concentration, with 95% confidence interval (CI) 0.50-0.83 (P = 0.0002). The DHEAS effect was not diminished by controlling for use of cardiac, vasodilator, antihypertensive, or lipid-lowering medication. CONCLUSIONS: These findings suggest that serum DHEAS levels bear an inverse relationship to heart disease, independently of a large set of established cardiovascular risk factors. The cross-sectional nature of this study requires that the findings be interpreted with caution. (C) 1998 Elsevier Science Inc.