Objective: We examined the prospective associations of sleep disturbances and sleep duration with the long-term incidence of major cardiovascular disease (CVD) events, in a large cohort of Italian adult men. Methods: A total of 2277 men aged 35-74 years of age and CVD free at baseline from the MONICA-Brianza and PAMELA population-based cohorts were followed up for a median of 17 years, for first coronary heart disease (CHD) or ischemic stroke events (fatal or nonfatal; n =293). Sleep disturbances, based on the Jenkins Sleep Questionnaire, were categorized as none/some, moderate, or severe. Sleep durations were ≤6 hours (short), seven to eight hours, and ≥9 hours (long) per night. Results: At baseline, 855 men (38%) either reported sleep disturbances or were short or long sleepers. The presence of severe sleep disturbances increased the risk of first CVD (hazard ratio [HR]=1.80, 95% confidence interval [CI]=1.07-3.03) and CHD events (HR=1.97, 95% CI=1.09-3.56), in particular from the age of 48 years onward. In comparison to men sleeping seven to eight hours, long sleepers experienced a higher CVD risk (HR=1.56, 95% CI=1.10-2.22), due mainly to ischemic strokes, and starting at older ages (≥60 years). A joint effect between disturbed sleep and short sleep duration on CVD and CHD events was also observed. Adjustments for physical activity and depression did not substantially modify these associations. Conclusion: Severe sleep disturbances and long sleep duration were associated with specific CVD endpoints and age at onset, potentially suggesting distinct underlying mechanisms. A short questionnaire discriminating different levels of sleep disturbances and sleep duration should be routinely adopted in CVD prevention programs to identify men at increased risk for early-onset events.