STUDY OBJECTIVES: The mechanisms linking short sleep duration to cardiovascular disease (CVD) are poorly understood. Emerging evidence suggests that endothelial dysregulation may lie along the causal pathway linking sleep duration to cardiovascular risk, although current evidence in humans is based on cross-sectional studies. Our objective was to evaluate the prospective association between objectively assessed sleep duration and clinical indices of endothelial health. METHODS: A total of 141 medically healthy adults underwent an overnight laboratory sleep study when they were between the ages of 21 and 60 years. Total sleep time was objectively assessed by polysomnography at study entry. Endothelial health, including brachial artery diameter (BAD) and flow-mediated dilation (FMD), was measured 18.9 ± 4.6 years later. Medical health and psychiatric status were assessed at both time points. Approximately half of the sample had a lifetime history of major depressive disorder. RESULTS: In univariate analyses, shorter sleep duration was associated with increased BAD (β = -0.24, p = .004) and decreased FMD (β = 0.17, p = .042). BAD, but not FMD, remained significantly associated with sleep duration after adjusting for sex, age, body mass index (BMI), smoking, diabetes, hypertension, and lifetime history of major depressive disorder (MDD) at T2. The association between sleep duration and BAD was stronger than the association between BAD and an aggregate measure of CVD risk including three or more of the following risk factors: male sex, age ≥ 65 years, smoker, BMI ≥ 30, diabetes, hypertension, and MDD. CONCLUSIONS: Objectively assessed short sleep duration was prospectively associated with increased BAD over a 12- to 30-year period.