Objective: Middle-aged and elderly men with the Lewis blood group phenotype Le(a-b-), have a two-fold higher prevalence of obesity than others. We investigated if the association could be ascribed to differences in lifestyle, or if obesity determinants had a different impact in this group.
Method: This was a cross-sectional study of 3,290 men aged 53-74 years with a mean of 63 years drawn from the Copenhagen Male Study. The main outcome measure was prevalence of obesity [body mass index (BMI) ≥30 (kg/m(2))].
Results: Among men with the Le(a-b-) phenotype (n = 315), the prevalence of obesity was 15.6%; it was 8.1% among men with other phenotypes (n = 2,975, p < 0.001). Obesity-related life style habits did not differ between Le(a-b-) and other men. In multiple logistic regression analyses, the role of obesity covariates was conspicuously different between men with the Le(a-b-) phenotype and men with other Lewis phenotypes. However, among Le(a-b-) men, a much stronger inverse association was found between high coffee consumption (>7 cups/day vs. less) and obesity, and between lacking fat avoidance and obesity; the odds ratio [95% confidence interval (CI)] for high coffee consumption was 0.2(0.1-0.7) and for avoidance of fatty foods 2.0(1.04-3.7). The association of obesity with leisure time physical activity, smoking, alcohol use, use of sugar in hot beverages, and tea consumption, did not differ between Le(a-b-) men and others.
Conclusions: Our findings suggests a gene-environment interaction between lifestyle and obesity. Because some ethnic groups have up to 40% with the Le(a-b-) phenotype, the above observations may be of significant public health interest.