https://orcid.org/0000-0002-2527-8936
This editorial refers to ‘Alcohol consumption, cardiac biomarkers and risk of atrial fibrillation and adverse outcomes’, by D. Csengeri et al. †, on page .
Alcohol consumption and risk of cardiovascular associations per one standard drink.
Atrial fibrillation (AF) is associated with an increased risk of death and cardiovascular complications, and has been described as one of the cardiovascular epidemics of the 21st century.1 The pathogenesis of AF is complex, but several AF risk factors have been previously described, explaining ∼50% of the population attributable risk.1–4 Enhancing our understanding of the risk factors predisposing to AF is crucial, as it may lead to the development of new preventative strategies. Indeed, thus far, only oral anticoagulation has been unequivocally shown to improve hard outcomes, once a patient has developed AF.5
An important modifiable risk factor predisposing to AF is excessive alcohol consumption.6 , 7 Both acute and increased habitual drinking have been linked to the development of AF. Acute alcohol consumption has a direct effect on the heart leading to atrial tachyarrhythmias as is manifest in the Holiday Heart Syndrome.8 , 9 Chronic alcohol consumption is also associated with an increased risk of AF.6 , 7 In the Women’s Health Study, women consuming >2 alcoholic drinks per day had a 60% higher risk of incident AF after multivariable adjustment, whereas lower amounts did not increase risk, suggesting a threshold effect.6 More recently, a meta-analysis of seven prospective studies involving 12 554 AF cases reported a linear effect between alcohol and incident AF with a risk increase of 8% per alcoholic drink consumed.7 However, while there is robust data to support that elevated alcohol consumption is associated with new-onset AF, uncertainty surrounded the evidence whether smaller quantities of alcohol are significantly associated with AF risk.
In this issue of the Journal, Csengeri et al.10 addressed the relationship between alcohol intake and incident AF in a very large dataset. Csengeri et al. 10 combined five prospective community-based cohorts totalling 107 845 individuals free of AF at baseline, of whom 5854 developed new-onset AF over a median follow-up of 13.9 years. Alcohol consumption was determined at baseline from patient self-report, while incident AF was determined from hospitalization data via ICD codes or comorbidities listed on death certificates. Median age was 47.8 years, 48.3% were men and median alcohol consumption was 3 g/day. In Cox regression models stratified by sex and cohort, consumption of 1 alcoholic beverage (defined as 12 g of alcohol) per day, was associated with a 16% increased risk of incident AF [hazard ratio 1.16, 95% confidence interval (CI) 1.11–1.22, P < 0.001]. Importantly, there was a significant relationship between alcohol and increased AF risk at very low levels of alcohol intake. Even an average consumption of 3 g/day was robustly associated with AF risk (HR 1.04, 95% CI 1.02–1.05). This relationship was not significantly attenuated by adjustment for other AF risk factors. Although not reported in the manuscript, it is important to note that the absolute risk of AF was likely low in participants consuming low-to-moderate amounts of alcohol.
The authors found a J-shaped relationship between alcohol consumption and incident heart failure, with the lowest risk observed at levels of up to 20 g of alcohol per day. When examining the intermediary role of incident heart failure in the development of AF from alcohol intake, the interaction between alcohol- and time-dependent heart failure (HF) was not significant. Finally, N-terminal pro B-type natriuretic peptide (NT-proBNP) and high-sensitivity troponin-I (hsTnI) were weakly correlated with alcohol consumption and adjustment for these biomarkers did not attenuate the relationship between alcohol and AF.
The current study10 has several strengths. Harmonization of the outcome and covariates across five cohorts allowed the authors to perform standardized analyses with a very large sample size and great power to detect significant relationships between alcohol intake and AF, including the lower spectrum of alcohol consumption where the risk of incident AF is expected to be small. Accordingly, this study provides solid evidence that even very low levels of alcohol (i.e. <1 drink per day) remain significantly associated with an increased risk of new-onset AF.
The study of Csengeri et al.10 has limitations. First, AF episodes were not adjudicated and relied on International Classification of Diseases (ICD codes) from hospitalization data, which may have led to misclassification bias. Furthermore, AF cases that did not lead to hospital presentation or were mentioned on death certificates would not have been detected. Second, this study was not able to examine the relationship between binge drinking and incident AF. Third, the study did not report the absolute risks of AF associated with low levels of alcohol consumption. This important issue has to be taken into account when also considering the potentially beneficial associations of modest alcohol intake with other cardiovascular outcomes.11–14 Finally, the study was not designed to shed further light as to the mechanisms linking habitual alcohol consumption and the development of AF. Although the authors examined the potential role of NT-proBNP and hsTnI in this relationship, biomarker data were only available for a minority of patients and only at baseline. The effects of alcohol on atrial electrophysiology likely depend on multiple factors including changes in atrial repolarization, vagal tone, and direct myocardial injury and fibrosis.9 More studies examining how alcohol affects atrial electrophysiology via these potential pathways are needed.
In conclusion, the current article10 makes an important contribution to our understanding of the relationship between alcohol intake and incident AF, in particular at the lower spectrum of alcohol consumption. A significant relationship between alcohol and AF was identified, and even small quantities of alcohol were associated with an increased, albeit small, risk of incident AF. Together with a recent randomized trial showing that a reduction in alcohol intake led to a reduction in AF recurrence,15 these data suggest that lowering alcohol consumption may be important for both prevention and management of AF. Importantly, any reduction in low-to-moderate alcohol consumption to potentially prevent AF needs to be balanced with the potentially beneficial association low amounts alcohol may have with respect to other cardiovascular outcomes (Graphical abstract).11–14 The net clinical benefit of consuming low amounts of alcohol requires further study, ideally in adequately powered randomized trials. Until then, each individual has to make its own best educated decision as to whether consuming up to 1 alcoholic drink per day is worthwhile and safe.
Conflict of interest: D.C. received speakers fees from Servier Canada, outside of the current work. He holds a McMaster University Department of Medicine Mid-Career Research Award.
The opinions expressed in this article are not necessarily those of the Editors of the European Heart Journal or of the European Society of Cardiology.
Footnotes
† doi:10.1093/eurheartj/ehaa953.
References
