RESUMO
BACKGROUND: Facial flushing response to drinking is observed in East Asians with deficient activity of the variant aldehyde dehydrogenase 2 genotype. This study examined the role of flushing response in the relationship between alcohol consumption and dyslipidemia. METHODS: This cross-sectional study included 1,443 Korean men, including 261 non-drinkers, 470 flushers and 712 non-flushers. Based on a questionnaire, weekly alcohol consumption was assessed and categorized into 4 groups. Dyslipidemia was defined by the National Cholesterol Education Program-Adult Treatment Panel III criteria. After adjusting for confounding factors, we evaluated the relationship between weekly alcoholic drinks and dyslipidemia in flushers and non-flushers by comparing with non-drinkers, using a multi-variable logistic regression analysis. RESULTS: Non-flushers had a significant relationship between hypertriglyceridemia and alcohol consumption in two groups (4-8 drinks: adjusted odds ratio [aOR] 1.937, 95% confidence interval [CI] 1.029-3.644; ≥16 drinks: aOR 2.118, 95% CI 1.272-3.527) in contrast to flushers, who showed no significant relationship between hypertriglyceridemia and alcohol consumption. Non-flushers had a significant relationship to low serum high density lipoprotein (HDL) cholesterol levels in four groups (<4 drinks: aOR 0.428, 95% CI 0.277-0.662; 4-8 drinks: aOR 0.409, 95% CI 0.216-0.774; 8-16 drinks: aOR 0.285, 95% CI 0.152-0.536; ≥16 drinks: aOR 0.343, 95% CI 0.207-0.568), and flushers had a significant relationship in two groups (8-16 drinks: aOR 0.234, 95% CI 0.102-0.536; ≥16 drinks: aOR 0.342, 95% CI 0.166-0.705). CONCLUSIONS: Our results suggest that drinking alcohol increases the risk of hypertriglyceridemia in non-flushers and the risk of low HDL cholesterol flushers.