Wine intake, ABO phenotype, and risk of ischemic heart disease and all-cause mortality: the Copenhagen Male Study--a 16-year follow-up

Abstract

The association of alcohol intake with ischemic heart disease (IHD) and all-cause mortality may depend on ABO phenotype. We tested this hypothesis in a 16-year follow-up of 3,022 Caucasian men aged 53-74 years without overt cardiovascular disease. Potential risk factors and confounders included were ABO phenotypes, alcohol intake (wine, beer, and spirits), tobacco smoking history, leisure-time physical activity, social class, and age. During 16 years, 1985-1986 to end of 2001, 197 subjects (6.5%) died due to IHD, and 1,204 (39.8%) from all causes. Among non-O phenotypes (A, B, and AB) significantly fewer men who died due to IHD were wine consumers, 43.9% versus 55.7%, P<.01; with respect to all-cause mortality corresponding figures were 47.0% versus 60.1%, P<.001. No difference was found among men with phenotype O. Among men with phenotype A, compared to alcohol abstainers, in Cox analysis, the hazard ratio (HR) (95% confidence limit) for men drinking up to 8 beverages/wk was 0.5 (0.3-1.02), and among men consuming >8 beverages/wk (the highest quintile) the HR was 0.3 (0.2-0.8), P<.01. Among men with phenotype O, the association of wine intake with IHD mortality was slightly and not significantly U-shaped. The difference in the predictive role of wine intake between phenotype O and phenotype A men was supported in a statistical test for interaction. A similar association was found for all-cause mortality. The results suggest that the effect of wine intake on IHD and all-cause mortality among middle-aged and elderly men may depend on ABO phenotypes.