Association of alcohol consumption with circulating biomarkers of atrial fibrillation-related pathways in a population at high cardiometabolic risk

Abstract

Background: The effect of alcohol consumption on cardiovascular health, including atrial fibrillation risk, remains controversial. Evaluating the association of alcohol consumption with circulating atrial fibrillation-related biomarkers may help better understand the relevant mechanistic underpinnings.

Methods: We studied 523 participants from 3 sites for the PREDIMED-Plus study, a weight-loss randomized intervention trial in metabolically unhealthy adults. N-terminal pro-B-type natriuretic protein (NTproBNP), high sensitivity troponin-T (hsTnT), high-sensitivity C-reactive protein (hsCRP), 3-nitrotyrosine (3-NT), and procollagen type 1 carboxy-terminal propeptide (PICP) were measured in fasting serum samples at baseline and years 3 and 5 of follow-up. We calculated alcohol consumption in drinks/day (1 drink = 14 grams alcohol) with validated food frequency questionnaires at each visit. Using multiple linear regression and mixed models we estimated the association of alcohol consumption with log-transformed biomarkers at baseline and longitudinally adjusting for potential confounders.

Results: Among 523 participants (mean age: 65 years, 40% female), mean alcohol consumption was 1 drink/day. Cross-sectionally, alcohol consumption was not associated with cardiac biomarker concentrations. Longitudinally, compared to non-consumers, heavy drinkers (≥4 drinks/day) had smaller increases in hsTnT (β: -0.11, 95%CI: -0.20, -0.01)and PICP (β: -0.15, 95%CI: -0.30, 0.01) over the 5-year follow-up. In contrast, those who increased alcohol consumption over the 5-year period experienced greater increases in hsCRP (β: 0.42, 95%CI: 0.11, 0.73) compared to those whose drinking behavior stayed the same.

Conclusion: Alcohol consumption was associated with complex changes in circulating biomarkers, including comparatively lower fibrotic and myocardial damage, but higher levels of overall inflammation over time. These results underscore the need for further research to better understand the effects of alcohol on cardiovascular health.

Figure 1.

Flow chart of selection into the study.

Figure 2.

Associations of change in alcohol over 5 years modeled as restricted cubic splines with change in biomarkers, using the median change in alcohol consumption as the reference point. Top left to right: NT-proBNP, Hs-TnT, CRP. Bottom left to right: PICP, 3-NT. Multiple linear regression models adjusted for age, sex, education, and intervention group.