Background Osteoprotegerin is a cytokine involved in bone metabolism as well as vascular calcification and atherogenesis. Although circulating osteoprotegerin levels are robustly associated with incident cardiovascular disease ( CVD ) in the general population, its relevance as a biomarker among populations at high CVD risk is less clear. Methods and Results Three independent reviewers systematically searched PubMed, EMBASE , and Web of Science to identify prospective studies that had recruited participants on the basis of having conditions related to high CVD risk. A total of 19 studies were eligible for inclusion, reporting on 27 450 patients with diabetes mellitus (2 studies), kidney disease (7 studies), preexisting heart disease (5 studies), or recent acute coronary syndromes (5 studies) at baseline. Over a mean follow-up of 4.2 years, 4066 CVD events were recorded. In a random-effects meta-analysis, the pooled risk ratio for CVD events comparing people in the top versus the bottom tertile of osteoprotegerin concentration was 1.30 (95% confidence interval, 1.12-1.50; P<0.001; I=68.3%). There was evidence for presence of publication bias ( P value from Egger's test=0.013). Correction for publication bias using the trim-and-fill method reduced the risk ratio to 1.21 (95% confidence interval, 1.03-1.42; P<0.001). The risk ratios did not vary significantly by population type, geographical region, statistical adjustment, sample or assay type, age, sex, or length of follow-up. Conclusions In populations at high CVD risk, elevated circulating osteoprotegerin levels are associated with a higher risk for future CVD events. The magnitude of association appears weaker than in the general population.