The Inverse Association between Alcohol Consumption and Coronary Artery Disease in Japanese Men Undergoing Coronary Angiography

Several Japanese cohort studies reported that alcohol consumption was inversely associated with coronary artery disease (CAD). However, these results were inconsistent. We investigated alcohol consumption and CAD in 802 patients undergoing coronary angiography, in whom foods, beverages and alcohol intakes were assessed. Alcohol intake was classified into 4 categories: nondrinkers, light (<23 g/d), moderate (23-45 g/d), and heavy (≥46 g/d) drinkers. Among 583 men, CAD was found in 404, of whom 137 had myocardial infarction (MI). The prevalence of CAD and MI stepwisely decreased on alcohol intake: 80% and 30% in nondrinkers, 65% and 20% in light, 60% and 17% in moderate, and 54% and 16% in heavy drinkers (p<0.001). However, serum γ-GTP levels were higher in heavy drinkers than in nondrinkers. In multivariate analyses, odds ratios for <23 g/d, 23-45 g/d and ≥46 g/d alcohol intakes compared to nondrinkers was 0.53 (95%CI: 0.33-0.85), 0.44 (0.24-0.81), and 0.33 (0.16-0.68) for CAD, and 0.56 (0.34-0.90), 0.62 (0.32-1.20), and 0.43 (0.18-1.01) for MI. Among 219 women, CAD and MI were present in 107 and 36. Compared with nondrinkers, light and moderate to heavy drinkers less often had CAD (52%, 41%, and 29%) and MI (19%, 9%, and 14%), but these differences did not reach statistical significance. In women, no significant association was found between alcohol intake and CAD. Thus, in Japanese men undergoing coronary angiography, alcohol consumption was inversely associated with CAD, but heavy consumption was associated with abnormally high γ-GTP levels. In women, no significant association was found between alcohol consumption and CAD.