The role of serum vitamin D in cardiovascular diseases risk : a retrospective study

Abstract

Cardiovascular disease (CVD) is one of the leading causes of mortality in developed regions including Hong Kong. A cost effective primary preventive measure of CVD is indispensable. Studies have been conducted to investigate novel strategies on the prevention and management of CVD. Vitamin D deficiency is common globally, whereas accumulating evidence suggest that vitamin D deficiency is associated with cardiovascular morbidity and mortality. However, the findings of vitamin D on the cardiovascular system are controversial and the relationships remain unclear. In this thesis, the role of serum vitamin D in CVD was evaluated using a retrospective cohort study. 1,159 participants [mean age at baseline 49.7 years (yrs)] from the Hong Kong Osteoporosis Study were included in this retrospective study. 53 of the participants had CVD at baseline. During a median follow-up of 10 yrs, 110 participants developed CVD. Using receiver operating characteristic (ROC) analysis of 1,159 participants, serum vitamin D level had a poor ability to discriminate individuals with or without CVDs [area under curve=0.58 [95% confidence interval (CI) = 0.53-0.63, p= 0.001]. In the fully adjusted logistic regression analysis of 1,159 participants, log serum vitamin D levels were not significantly associated with CVD [odd ratio (OR) =1.67, 95% CI = 0.85-3.28, p=0.135]. Participants were divided into subgroups based on their gender, vitamin D status and vitamin D quartile. All associations were statistically insignificant except in the group of participants with serum vitamin D levels > 75 nmol/L, log serum vitamin D was associated with CVD (OR: 59.33, 95% CI: 1.38-2559.26, p=0.034). In this group, 31% of the CVD cases were diagnosed before the measurements of baseline data. The lifestyle changes of these patients, evidenced by their healthier lifestyle at baseline, after being diagnosed with CVD were believed to be the account for the unexpected results obtained in this group. 1106 participants who were CVD free at baseline were further studied in Cox regression analysis. The findings using Cox regression analysis were similar to those from logistic regression analysis, except no significant association was observed between log serum vitamin D and CVD in the in the group of participants with serum vitamin D levels > 75 nmol/L. In conclusion, serum vitamin D levels were not associated with the risk of CVD.