The role of high density lipoprotein for alcohol intake and myocardial infarction: results from two prospective cohorts
Open Access
- Author:
- Huang, Shue
- Graduate Program:
- Nutritional Sciences
- Degree:
- Doctor of Philosophy
- Document Type:
- Dissertation
- Date of Defense:
- September 10, 2019
- Committee Members:
- Xiang Gao, Dissertation Advisor/Co-Advisor
Xiang Gao, Committee Chair/Co-Chair
Penny Margaret Kris-Etherton, Committee Member
Sy-Miin Chow, Outside Member
Gregory C Shearer, Dissertation Advisor/Co-Advisor
Gregory C Shearer, Committee Chair/Co-Chair
A Catharine Ross, Program Head/Chair - Keywords:
- alcohol intake
HDL
cardiovascular disease
prospective cohort
mediation analysis
longitudinal analysis - Abstract:
- Myocardial infarction (MI) is the most common form of coronary heart disease, which is by far the most common cardiovascular condition. Despite the improvement and increased use in evidence-based therapies and lifestyle interventions, it remains the number one cause of morbidity and mortality worldwide. Alcohol intake has both a beneficial and a harmful influence on atherosclerosis, which is the main pathological change in MI. Moderate alcohol intake has been found to be associated with a lower risk for MI in numerous epidemiological studies, and it is hypothesized that this protective effect is mainly mediated through high-density lipoprotein cholesterol (HDL-C) concentration. However, controversial results have been generated from epidemiological studies on this topic and the role of HDL-C concentration being a sufficient predictor for cardiovascular disease has been challenged. The objective of the first study in this dissertation was to determine the association between total alcohol intake and type of alcohol containing beverage, and 6-year (2006-2012) longitudinal change in HDL-C concentrations in a community-based cohort. In this study, we included 71,379 Chinese adults (mean age 50±11 yr.) who were free of cardiovascular diseases and cancer, and did not use cholesterol-lowering agents during follow-up. Alcohol intake was assessed by a questionnaire in 2006 (baseline) and participants were categorized as never, past, light (women: 0-0.4 servings/d, men: 0-0.9 servings/d), moderate (women: 0.5-1.0 servings/d, men: 1-2 servings/d), and heavy drinkers (women: >1.0 servings/d, men: >2 servings/d). HDL-C concentrations were measured in 2006, 2008, 2010, and 2012. We used generalized estimating equation models to examine the associations between baseline alcohol intake and change in HDL-C concentrations, adjusting for age, sex, smoking, physical activity, obesity, hypertension, diabetes, liver function and C-reactive protein concentrations. We observed an umbrella-shaped association between total alcohol consumption and changes in HDL-C concentration. Compared with never drinkers, past, light, moderate and heavy drinkers, respectively, experienced a 0.012 (95% CI: 0.008, 0.016), 0.013 (95% CI: 0.010, 0.016), 0.017 (95% CI: 0.009, 0.025), and 0.008 (95% CI: 0.005, 0.011) mmol/L per year slower decrease in HDL-C (P <0.0001 for all) after adjusting for potential confounders. Moderate alcohol consumption was associated with the slowest increase in total cholesterol/HDL-C and triglyceride/HDL-C ratios. We observed a similar association between hard liquor consumption and HDL-C change. In contrast, greater beer consumption was associated with slower HDL-C decreases, in a dose-response manner. The objective of the second study was to test the hypothesis that the lower risk of myocardial infarction (MI) associated with alcohol intake is mediated by raising HDL-C concentration. This study included 81,253 Chinese men and women (mean age: 51±12 yr.) from the Kailuan Study who were free of cardiovascular disease in 2006 (at baseline) and were followed up to Dec. 2016. At baseline, alcohol consumption was assessed via a questionnaire and the concentration of HDL cholesterol was measured. Incident MI at follow up was a first MI event, confirmed by medical record review. Multivariable Cox regression was used to model the association between habitual alcohol intake and risks of MI, adjusting for potential covariates including age, sex, education, monthly income, occupation, smoking status, physical activity, body mass index, waist circumferences, hypertension, diabetes and total cholesterol. Mediated effect through HDL cholesterol was assessed using a causal mediating analysis (SAS macro). During an average of 9.6 years of follow-up, we documented 1088 incident cases. The adjusted hazard ratio (HR) for MI was 0.74 (95% confidence interval (CI): 0.60, 0.91), 0.80 (95%CI: 0.56, 1.16), 0.56 (95%CI: 0.45, 0.70) for light, moderate, and heavy alcohol drinkers compared with non-drinkers. The ratio changed very slightly after further adjustment of HDL cholesterol concentration. Mediation analysis showed that HDL cholesterol concentrations mediated a small, non-significant proportion (Proportion mediated ~2%) of the association between alcohol and MI. The objective of the third study was to investigate whether and to what extent the association between alcohol intake and incident myocardial infarction (MI) is mediated through HDL-C concentration, HDL particles (HDL-P) concentration, and apoA-I concentrations. A total of 6,683 participants from the Multi-Ethnic Study of Atherosclerosis (MESA) were included in the analysis. Alcohol consumption was assessed via a questionnaire at baseline (exam 1). HDL-C concentrations and HDL-P were both measured at exam 1. Cox regression was used to model the association of habitual alcohol intake and risk for MI before and after adjusting HDL-C and HDL-P, in addition to adjustment of all potential covariates. Mediated effects through HDL-C, and HDL-P were estimated using the causal mediation analysis. After a median of 8 years follow-up, 171 incident MI cases were documented. Higher alcohol intake was associated with a lower risk for incident MI (P for trend =0.039). The relation between alcohol intake and MI was slightly attenuated by adjustment of HDL-C and moderately by HDL-P. Mediation analysis showed no mediating effect of HDL-C (HRNIE: 0.98; 95%CI: 0.94, 1.02; P-value=0.26), whereas HDL-P slightly mediated the association of habitual alcohol intake and MI (HRNIE: 0.95, 95%CI: 0.90, 0.99; P-value=0.027). The proportion of the total effect of alcohol on MI mediated by HDL-P was 15.8%. In conclusion, we found that moderate alcohol consumption was associated with slower HDL-C decreases; however, the type of alcoholic beverage was differently associated with the change in HDL-C concentrations. Alcohol consumption was associated with a lower risk for MI incidence in both Chinese and MESA cohorts. Our results suggest that the benefits on MI associated with moderate alcohol consumption are not related to the effects of alcohol on HDL cholesterol. The lower risk of MI related to alcohol intake appears to partially work through increasing HDL-P. The mechanism for much of alcohol’s effect to reduce MI risk remains unexplained. This suggests that HDL-P can be a target for MI prevention, however the mediating effect of HDL-P is very moderate. Future studies arewarranted to confirm our finding regarding HDL-P and further evaluate if HDL-P concentration may be a surrogate biomarker reflecting the anti-atherogenic HDL function to decide if HDL-P can be used as a new risk predictor and intervention target for CVD.