Plasma vitamins E and A inversely correlated to mortality from ischemic heart disease in cross-cultural epidemiology.

In eight study populations in which the medians of total plasma cholesterol did not differ significantly (mean 5.8 mM, p greater than 0.05) and therefore did not correlate with the IHD mortality (r2 = 0.05), the median of absolute plasma level of vitamin E (alpha-tocopherol) was inversely related to the IHD mortality (r2 = 0.55; p = 0.003). Vitamin A behaved similarly (r2 = 0.51; p = 0.046). The absolute levels of vitamins E and A together with cholesterol predicted (by multiple regression analysis) the IHD mortality of these eight populations fairly well (r2 = 0.81; p = 0.06). Considering all 12 study populations analyzed thus far, total plasma cholesterol correlated with the IHD mortality directly as expected (r2 = 0.51; p less than 0.01), but the median of the plasma alpha-tocopherol individually standardized for cholesterol and triglycerides (220 mg/dL + 110 mg/dL, respectively) maintained a strong inverse association with the IHD mortality (r2 = 0.49; p = 0.01). In the partial regression analysis, lipid-standardized vitamin E exhibited an even stronger inverse correlation with IHD mortality (r2 = 0.69; p less than 0.001). Again, vitamin A behaved similarly to vitamin E, that is, after lipid-standardization of individuals (r2 = 0.33; p = 0.07), as well as in the cholesterol-independent partial regression analysis (r2 = 0.74; p less than 0.001). Both vitamins may act singularly, for after lipid-standardization they vary de facto independently (rs = 0.012) in individuals. The combination of vitamins E and A as obtained by multiple partial regression predicted the actual IHD mortality to a large extent (r2 = 0.89; p less than 0.001), whereas the three-variable prediction model, with the median of total cholesterol and of individually lipid-standardized vitamins E and A, fit the actual IHD mortality of these 12 populations almost completely (r2 = 0.94; p less than 0.001). In conclusion, the plasma status of vitamins E and A are important, hitherto underrated risk factors of IHD, which may act independently, but can, if combined, predict at least 53% of the cross-cultural differences of IHD mortality. After inclusion of total cholesterol into a multivariate model, up to 94% of the IHD mortality can be predicted. The present epidemiological data are in agreement with the hypothesis that these vitamins have physiological functions in the protection of lipoproteins against peroxidation and atherogenic apo-B modifications, respectively, but that does not exclude additional beneficial effects of vitamin E and A in the arterial wall.