Page 134 - Vitamin D and Cancer

6 Vitamin D: Cardiovascular Function and Disease 121

The descriptive epidemiology of CV disease shows that rates are highest in
winter in both the northern and southern hemispheres, increase with increasing
latitude, and decrease with increasing altitude. Drawing this evidence together, the
author published a hypothesis in 1981 that sunlight and vitamin D may protect
against CV disease [48]. This hypothesis was also consistent with the increased CV
disease rates in population groups with lower vitamin D levels due to decreased
skin synthesis, such as older people and those with increased skin pigmentation
(e.g., African-Americans) [49, 50]. A more detailed review of the evidence in
support of the hypothesis was subsequently published [51].
Recent ecological studies of CV disease have continued to provide support for
the hypothesis. For example, an inverse association between UV insolation and
coronary heart disease mortality in men has recently been reported for the countries
of Western Europe [52]. Seasonal variations in vitamin D status, with low 25OHD
levels in winter, have been shown in both the northern and southern hemispheres
[53, 54]. Winter excesses in mortality and incidence have been reported for the full
spectrum of CV disease, including coronary heart disease [55–57], stroke [57–59],
heart failure [60, 61], ventricular arrhythmias [62], endocarditis [63], and pulmo-
nary embolism [64].
Importantly, the winter excess in CV disease has been observed in warm cli-
mates, such as Los Angeles [65], and in Hawaii despite a small seasonal variation
in temperature between 22.8°C and 27.8°C [66]. The winter excess in cardiovascu-
lar disease is attributed frequently to the cold temperatures of winter [67], but it
does not seem plausible that the mild temperatures experienced by people in the
above two locations during the winter months is a major factor in their raised CV
disease rates at that time of year.
The hypothesis that vitamin D protected against CV disease was tested in a
population-based case–control study of myocardial infarction carried out in New
Zealand by the author and colleagues in the 1980s and published in 1990 [68]. The
sample was restricted to incident cases from a register which provided blood
samples within 12 h of onset of symptoms since a pilot study showed that plasma
25OHD was unchanged during this period [69]. The unit of measurement for
25OHD in this study was actually nanograms per milliliter (ng/mL), rather than
nanomoles per liter (nmol/L) as reported. Mean plasma 25OHD was significantly
lower in cases (n = 179) than controls selected from the electoral roll who were
individually matched by age, sex, and date of blood collection (32.0 vs 35.0 ng/
mL; p = 0.017). An inverse association between plasma 25OHD and risk of myo-
cardial infarction, with the odds ratio for those in the highest 25OHD quartile
being 0.30 (95% confidence interval [CI]: 0.15, 0.61) compared with the lowest
quartile [68].

6.3.2 Animal Studies

Independently of the above epidemiological studies, research from animal models
in the 1980s was beginning to better define the effect of vitamin D on CV function