Interactive effects of ApoE polymorphism, alcohol and smoking on age-related trends of blood pressure levels in elderly men: the Bambuì Cohort Study of Ageing (1997-2008).

Data from a population-based prospective study were used to examine longitudinal changes in blood pressure (BP) and seek interactions between apolipoprotein E (ApoE) genotypes, smoking and alcohol in a cohort of 557 elderly Brazilian men. Repeated BP measurements were obtained in four waves, and multi-level random-effects pattern-mixture models were used to evaluate age-related BP trajectories while accounting for non-ignorable dropouts/deaths and handling heterogeneities as random parameter variations. Alcohol was associated with high systolic BP in ε2 carriers and those with the ε3/3 genotype, but not in ε4 carriers. This was dependent on age and smoking habits: at the age of 60, expected systolic BP in alcohol drinking ε2 carriers was 16.5 mm Hg higher than in the reference group of non-smokers/non-drinkers if they were not smokers (P=0.049), and 28.6 mm Hg higher if they were also smokers (P=0.004). The youngest smoking/non-drinking ε2 carriers had lower systolic BP, but it increased rapidly and led to higher expected levels among older carriers. Alcohol consumption, alone or together with smoking, interacts with the effects of ApoE genotype on systolic BP, probably nullifying the more favourable lipid profile of ε2 carriers. The interactions of gene-modifiable risk factors have major public health implications.