Serum 25-Hydroxyvitamin D and Incidence of Fatal and Nonfatal Cardiovascular Events: A Prospective Study With Repeated Measurements

Abstract

Several studies suggested that low serum concentrations of 25-hydroxyvitamin D (25(OH)D) are associated with an increased risk of cardiovascular disease (CVD). However, the evidence is still inconclusive, mostly based on CVD mortality and studies with single 25(OH)D measurements. We aimed to assess the association of 25(OH)D with fatal and nonfatal CVD in the same study population, using repeated 25(OH)D measurements and competing risks analysis. This was a population-based cohort study (ESTHER study, baseline 2000-2002). Follow-up data, including survival status, were collected after 2, 5, and 8 years. The response rate for survival was 99.9%. Participants were recruited during a health screening examination by their general practitioners. 25(OH)D was measured in blood samples collected at baseline and the 5-year follow-up visit. A total of 9949 men and women, aged 50 to 74 years at baseline, with sufficient knowledge of the German language and resident in the German state of Saarland were included in the study. Outcomes included CVD, coronary heart disease (CHD), and stroke, in total and differentiated into fatal and nonfatal events. Overall, 854 study participants had a nonfatal and 176 a fatal CVD event during 8 years of follow-up. Comparing subjects with 25(OH)D levels below 30 nmol/L and above 50 nmol/L resulted in a hazard ratio of 1.27 (95% confidence interval = 1.05-1.54) for total CVD and 1.62 (95% confidence interval = 1.07-2.48) for fatal CVD in a model adjusted for important potential confounders. No significant association for nonfatal CVD was observed. In dose-response analysis, we observed an increased cardiovascular risk at 25(OH)D levels below 75 nmol/L. Results for CHD and stroke were comparable to the results obtained for the composite outcome CVD. Our results support evidence that low 25(OH)D levels are associated with moderately increased risk of CVD and indicate that the observed association is much stronger for fatal than for nonfatal events.