Background: One half to one third of healthy middle-age and older adults in the United States have vitamin D deficiency, probably from inadequate exposure to sunlight and insufficient dietary intake. According to some data, in addition to its known musculoskeletal effects, vitamin D deficiency is associated with elevated blood pressure and cardiovascular disease (CVD). Ecologic studies report that hypertension and coronary heart disease rates increase with further distance from the equator, which may be related to decreasing exposure to sunlight and subsequent vitamin D deficiency. However, these associations cannot definitively attribute cardiovascular risk to vitamin D deficiency. To evaluate the possible correlation, Wang and colleagues prospectively studied vitamin D levels in a large group of ambulatory, community-based persons without CVD at baseline.

The Study: The study included the Framingham Offspring cohort. Between 1996 and 2001, vitamin D (25-hydroxyvitamin D) levels were measured in 1,972 consecutive patients at their regular medical examinations. Of these patients, 233 were excluded for prevalent CVD or renal disease. The remaining 1,739 patients were followed for up to 7.6 years (mean follow-up was 5.4 years) to assess the risk of CVD.

Vitamin D, lipid, and C-reactive protein levels were drawn from fasting morning samples. Detailed medical histories were recorded for each participant and included age, sex, diagnoses of diabetes and hypertension, smoking status, physical activity, and total dietary intake of vitamin D. Medical records from hospitalizations and office visits were reviewed to determine the diagnosis of a cardiovascular event, defined as myocardial infarction, coronary insufficiency, angina, stroke, transient ischemic attacks, peripheral claudication, or heart failure.

Results: The mean level of 25-hydroxyvitamin D was 19.7 ng per mL (50 nmol per L). Levels less than 15 ng per mL (38 nmol per L) were considered deficient, and levels less than 10 ng per mL (24 nmol per L) were considered severely deficient. The overall prevalence of vitamin D deficiency was 28 percent, with 9 percent in the severely deficient range. There were 120 fatal or nonfatal cardiovascular events during the study. The age- and sex-adjusted five-year rate of CVD was approximately twice as high in those with vitamin D deficiency, with the highest rate occurring in those who also had hypertension. Adjusting for C-reactive protein levels, physical activity, vitamin D supplementation, patient education level, and time of year did not decrease this association. The risk of CVD increased as 25-hydroxyvitamin D levels decreased from 15 or higher, less than 15, and less than 10 ng per mL.

Conclusion: This observational study suggests a link between vitamin D deficiency and the development of CVD, which is compounded in persons with hypertension. The authors caution that additional clinical studies are needed to determine whether correcting vitamin D deficiency has a role in preventing CVD.