Influence of vitamin D supplementation on plasma lipid profiles: A meta-analysis of randomized controlled trials

Cardiovascular disease (CVD) remains the leading cause of death and disability in the world [1]. Although cardiovascular mortality rates have declined in some high-income countries, more than 17 million people died from CVD in 2008 and it is estimated that by 2030, almost 23.6 million people will die from CVD. Thus, extraordinary effort has been devoted to determining the modifying risk factors to prevent atherosclerosis, the main cause of CVD. There is now increasing evidence that vitamin D, beyond its well-known effects on bone metabolism, also plays an important role in the development of CVD [2]. Epidemiologic studies have shown that vitamin D deficiency was closely associated with increased risk of major adverse CVD events [3,4]. Furthermore, randomized intervention trials showed a tendency towards a reduction in CVD risk with vitamin D supplementation, though the tendency is statistically nonsignificant [5]. Taking into account that vitamin D deficiency is highly prevalent across the world while vitamin D supplementation is simple, safe, and inexpensive [6], the deficiency of vitamin D may be a common and easily treatable risk factor for CVD prevention [7].There are several possible mechanisms contributing to the association between vitamin D and CVD, such as insulin sensitivity, parathyroid hormone elevation and inflammation [8]. It is reasonable that dyslipidemia should also be considered as a potential link because dyslipidemia is a well-described independent risk factor for CVD. Observational studies have indicated that high 25-hydroxyvitamin D [25(OH)D] levels were associated with a favorable serum lipid profile [9]. However, a solid rationale for such association is difficult to determine unless there is an effect of vitamin D supplementation on serum lipids in placebo-controlled randomized trials. Unfortunately, the intervention studies gave divergent results that some showing a positive and some a negative effect [9]. Therefore, we conducted a meta-ana