Dr. Wood and colleagues from the University of Aberdeen recently reported that vitamin D has no effect on conventional cardiovascular risk factors. They published results from their randomized controlled trial in the Journal of Clinical Endocrinology and Metabolism.
Adrian D. Wood et al. Vitamin D3 Supplementation Has No Effect on Conventional Cardiovascular Risk Factors. A Parallel-Group, Double-Blind, Placebo-Controlled RCT. J Clin Endocrin Metab. First published ahead of print August 3, 2012 as doi:10.1210/jc.2012-2126
Let’s look at the who, how, and what:
267 healthy 60-70 year-old Caucasian women who lived at home participated and completed the study. Subjects were seen every 2 months. “Healthy” means that folks with known diseases or problems were screened out.
There were 3 groups in the study, all doses administered for 12 months:
They measured surrogate heart disease markers:
They also assessed: bone density, weight, waist circumference, activity levels, diet, and UVB exposure (with badges).
Mean baseline vitamin D level was 13-14 ng/ml in all three groups, so almost all were deficient or insufficient. After 12 months of supplementation, mean serum vitamin D levels were:
The authors reported that risk factor levels varied by season in all groups independently of serum D levels; this raises issues about seasonality data in observational studies. As for their heart disease markers, they found no difference among any group. They concluded: “Improving vitamin D status through dietary supplementation is unlikely to reduce CVD risk factors.”
The authors’ conclusion is stated in their title: “Vitamin D3 supplementation has no effect on conventional cardiovascular risk factors.” This title and conclusion overstates the actual finding. It would more accurate to say that 1000 IU/day or 400 IU/day for a year has no effect on cardiovascular risk factors, not that vitamin D has no effect.
The authors agreed that they may have had low D serum effects: “Additionally, vitamin D3 doses of 400 or 1000 IU may have been too low to have beneficial effects on surrogate markers of CVD risk.” Dr. Robert Heaney, for example, has suggested that a vitamin D level of 50 ng/ml should be considered the minimum threshold, a level that the 1000 IU group did not come close to.
The study was too small and for too short of time to look at actual cardiovascular disease (CVD) onset or events. Obviously CVD is a long latency disease and takes many years to develop.
Correlating low serum D levels with markers is basically correlating with a correlation, which is weaker than correlating with actual events. Excluding sick folks also distorts the sample.
I was disappointed that there was no discussion of their secondary data (activity, diet, bone density, UVB, etc.).
In summary, here is what I would highlight about this study:
Why do researchers study low doses? All human use research has to be approved by committees who are typically very risk averse, though we are starting to see more and more trials using better and higher doses of vitamin D.