Page 141 - Vitamin D and Cancer

128 R. Scragg

incident hypertension had a weak inverse association with vitamin D but not
vitamin D supplements [148]. A possible explanation for the failure of most of
these studies to find an association is that dietary sources of vitamin D contribute
only a small proportion of the total vitamin D entering the body each day, which is
mainly derived from sun exposure [34]. Interestingly, when two of these cohort
studies were reanalyzed using plasma 25OHD, which measures vitamin D from all
sources, both measured 25OHD and estimated 25OHD were inversely associated
with risk of incident hypertension in both men and women [149]. For example,
participants in the lowest baseline quartile of plasma 25OHD (<15 ng/mL) had a
3.18 (95% CI: 1.39, 7.29) increased risk of developing hypertension over 4 years
than those in the highest 25OHD quartile (³30 ng/mL). This finding is supported
by a recent publication from the cross-sectional NHANES III study (n = 12,644)
which found that serum 25OHD was inversely associated with both systolic blood
pressure and pulse pressure [53]. However, another cross-sectional study from the
Netherlands (n = 1,205) did not observe any association between serum 25OHD and
blood pressure, possibly because the elderly sample had relatively high vitamin D
levels, although there was a significant positive association between serum parathy-
roid hormone and blood pressure [150].
Further intervention studies, both from Germany, have also been carried out.
A randomized clinical trial in elderly women found that 800 IU of vitamin D3/day
(with 1,200 mg of calcium) after 8 weeks significantly decreased systolic blood
pressure by 5 mmHg, but not diastolic, compared with placebo [151]. Another
randomized trial of patients with hypertension found that exposure to UV-B radia-
tion over 6 weeks, which increases vitamin D, lowered blood pressure by 6 mmHg
compared with the UV-A control group (p < 0.05) [152].

6.4.3 Studies of Vitamin D Supplementation

The Women’s Health Initiative trial is the only randomized trial to date which has
examined the effect of vitamin D on CV disease in the general population [139].
Postmenopausal women aged 50–79 years (n = 36,282) at 40 clinical sites in the
USA were randomized to take calcium carbonate 500 mg with vitamin D 200 IU
twice daily or placebo. Both fatal and nonfatal disease events were recorded. After
7 years of follow-up, the adjusted hazard ratios in the treated group versus control
were 1.04 (95% CI: 0.92, 1.18) for coronary heart disease and 0.95 (95% CI: 0.82,
1.10) for stroke. Thus, this study did not detect any effect of vitamin D and calcium
supplementation on CV disease.
However, this study has some major design limitations which prevent it from
being a proper test of the hypothesis that vitamin D protects against CV disease
[139, 153, 154]. Firstly, the dose of vitamin D was only 400 IU/day, which would
have raised serum 25OHD levels only by about 10 nmol/L [155], way below the
daily dose of 1,700 IU required to raise 25OHD levels above 80 nmol/L that is
currently considered optimum [156]. The actual vitamin D ingested would have