STUDY OBJECTIVE: It has been suggested that low levels of dehydroepiandrosterone sulphate (DHEAS) are predictive for cardiovascular diseases in men. We aimed to review the available evidence from prospective cohort studies and retrospective case-control studies. METHODS: We extracted summary statistics from 4 case-control studies and 8 cohort studies, and calculated the pooled relative risk associated with a 2 micromol/l increase in DHEAS. MAIN RESULTS: The number of subjects included in each of the individual studies ranged from 94 to 2134, mean age from 48 to 83 years and mean DHEAS levels from 1.2 to 7.3 pmol/l. In men, coronary mortality was available as outcome in 3 cohort studies and 1 case-control study. Combining data from these 4 studies showed a 15% (95% CI: 4%-28%, p = 0.008) increase in fatal coronary heart disease associated with a 2 micromol/l decrease in DHEAS. However, statistical significance was lost when the retrospective study causing significant heterogeneity (p = 0.02) was excluded. Fatal and non-fatal coronary events were reported in 1 cohort study and 3 case-control studies. The average increase in fatal plus non-fatal coronary heart disease associated with a 2 micromol/l decrease in DHEAS amounted to 13% (2%-26%, p = 0.02). The available data did not allow drawing any conclusions on the prognostic value of DHEAS in women, nor on the relationship between DHEAS and total or cardiovascular mortality or stroke in men. CONCLUSIONS: The present findings suggest that, in men, low serum levels of DHEAS may be associated with coronary heart disease. However, whether DHEA supplementation has any cardiovascular benefit is not clear. Data from prospective randomised trials are needed.