Alcohol consumption and subclinical and clinical coronary heart disease: a Mendelian randomization analysis
Takashi Hisamatsu, Katsuyuki Miura, Yasuharu Tabara, Yuichi Sawayama, Takashi Kadowaki, Aya Kadota, Sayuki Torii, Keiko Kondo, Yuichiro Yano, Akira Fujiyoshi, Takashi Yamamoto, Yoshihisa Nakagawa, Minoru Horie, Takeshi Kimura, Tomonori Okamura, Hirotsugu Ueshima
Abstract
AIMS: The potential effect of alcohol consumption on coronary heart disease (CHD) remains unclear. We used the variant rs671 in the aldehyde dehydrogenase 2 gene (ALDH2) as an instrument to investigate the causal role of alcohol intake in subclinical and clinical CHD. METHODS: We conducted two Mendelian randomization studies: a cross-sectional study of coronary artery calcification (CAC) on computed tomography of 1029 healthy men (mean age, 63.8 years) and a case-control study of 421 men with CHD [acute coronary syndrome (ACS) or stable angina pectoris] who underwent coronary revascularization and 842 age-matched male controls. RESULTS: In the CAC study, medians (25%tiles, 75%tiles) of alcohol consumption by ALDH2-rs671 *2 homozygotes [n = 86 (8.4%)], *1*2 heterozygotes [n = 397 (38.5%)], and *1 homozygotes [n = 546 (53.1%)] were 0.0 (0.0, 0.0), 28.0 (0.0, 129.0), and 224.0 (84.0, 350.0) g/week, respectively. In age-adjusted Poisson regression with robust error variance, compared with *2 homozygotes, relative risks for prevalent CAC score >0, ≥100, and ≥300 in *1 homozygotes were 1.29 (95% confidence interval, 1.06-1.57), 1.76 (1.05-2.96), and 1.81 (0.80-4.09), respectively. In age-adjusted ordinal logistic regression for CAC distributions, we observed higher odds among *1 homozygotes [odds ratio, 2.19 (1.39-3.46)] and even among *1*2 heterozygotes [1.77 (1.11-2.82)] compared with *2 homozygotes. In the case-control study, conditional logistic regression revealed lower prevalence of *1 homozygotes among men with CHD [odds ratio, 0.54 (0.35-0.82)], especially ACS [0.46 (0.27-0.77)], than controls. CONCLUSION: Our findings indicate a positive association of alcohol consumption with CAC burden but an inverse association with clinical CHD, especially ACS.