Alcohol and coronary artery calcification: an investigation using alcohol flushing as an instrumental variable

KE Yun, Y Chang, SC Yun… - International Journal …, 2017 - academic.oup.com
KE Yun, Y Chang, SC Yun, G Davey Smith, S Ryu, S Cho, EC Chung, H Shin, YH Khang
International Journal of Epidemiology, 2017academic.oup.com
Background: We examined whether alcohol flushing could be used as an instrumental
variable (IV) and investigated the effect of alcohol consumption on coronary calcification
using alcohol flushing status as an IV. Methods: We analysed cross-sectional data from 24
681 Korean adults (20 696 men and 3985 women) who had been administered a
questionnaire assessing alcohol consumption and alcohol flushing, as well as a coronary
artery calcium (CAC) measurement. The associations of alcohol flushing status with …
Abstract
Background: We examined whether alcohol flushing could be used as an instrumental variable (IV) and investigated the effect of alcohol consumption on coronary calcification using alcohol flushing status as an IV.
Methods: We analysed cross-sectional data from 24 681 Korean adults (20 696 men and 3985 women) who had been administered a questionnaire assessing alcohol consumption and alcohol flushing, as well as a coronary artery calcium (CAC) measurement. The associations of alcohol flushing status with potential confounders and alcohol consumption were examined. We employed two-stage predictor substitution methodology for the IV analysis.
Results: The prevalence of alcohol flushing did not differ depending on gender, education, household income, cigarette smoking or physical activity. Balanced levels of confounders were observed between alcohol flushers and non-flushers. Alcohol flushing was closely related to alcohol consumption and levels of liver enzymes. In men, a doubling in alcohol consumption was associated with increased odds of coronary calcification in both the IV analysis [odds ratio (OR) of CAC scores of 1 or over = 1.11; 95% confidence interval (CI) = 1.03-1.20) and the multivariable regression analysis (OR = 1.04; 95% CI = 1.01-1.07). For cardiovascular risk factors, the IV analysis showed a positive association between alcohol consumption and blood pressure and high-density lipoprotein-cholesterol.
Conclusions: Alcohol flushing can be used as an IV in studies evaluating the health impact of alcohol consumption, especially in East Asian countries. Through such an analysis, we found that increased alcohol consumption was associated with an increased risk of subclinical coronary atherosclerosis.
Oxford University Press