Background: The influence of testosterone on risk for cardiovascular events in men is uncertain. Previous observational studies of sex hormones and incident cardiovascular disease in men have reported inconsistent findings, limited by cohort sizes and different selection criteria.

Objective: To analyze associations of serum total testosterone and sex hormone-binding globulin (SHBG) with incident cardiovascular events in men.

Design: Cohort study.

Participants: Community-dwelling men aged 40 to 69 years.

Measurements: Testosterone and SHBG were assayed, and free testosterone was calculated. Cox proportional hazards regression was done, with outcomes of incident myocardial infarction (MI), hemorrhagic stroke (HS), ischemic stroke (IS), heart failure (HF), and major adverse cardiovascular events (MACE), adjusted for sociodemographic, lifestyle, and medical factors.

Results: Of 210 700 men followed for 9 years, 8790 (4.2%) had an incident cardiovascular event. After adjustment for key variables, lower total testosterone concentrations were not associated with incident MI, HS, IS, HF, or MACE. Men with lower calculated free testosterone values had a lower incidence of MACE. Lower SHBG concentrations were associated with higher incidence of MI and lower incidence of IS and HF (HR, 0.69, but not with HS or MACE.

Conclusion: Men with lower total testosterone concentrations were not at increased risk for MI, stroke, HF, or MACE. Calculated free testosterone may be associated with risk for MACE. Men with lower SHBG concentrations have higher risk for MI but lower risk for IS and HF, with causality to be determined.