John M. Mandrola, MD


March 21, 2019

People drink alcohol because it tastes good and sometimes causes nice sensations. Emphasis on sometimes.

Many studies[1] link alcohol intake with atrial fibrillation (AF), which can induce decidedly unpleasant sensations. Doctors rightly counsel patients with AF to reduce or eliminate alcohol intake.

One of the criticisms of lifestyle recommendations, such as alcohol abstinence, is that most of the evidence is observational and therefore confounded. Show me a randomized controlled trial, skeptics say.

Well, Alex Voskoboinik, MD, PhD, and colleagues from Melbourne, Australia, have completed one. The positive results of the Alcohol-AF trial, presented here at ACC.19, will do two things: Bolster common-sense advice with evidence and create difficult decisions for patients and doctors alike.

Recruiting for Alcohol AF

Researchers screened patients with AF at six hospitals for randomization to a trial of alcohol abstinence or continued usual intake. Obviously, to be enrolled, patients had to drink moderate amounts of alcohol.

The trial planned for 1-year follow-up but had to be shortened to 6 months because of "challenges" in adherence. The average alcohol intake of enrolled patients was 16 drinks per week. Perhaps you can see the challenge?

Voskoboinik said the researchers screened nearly 700 patients and ended up with 70 patients in each arm. He made clear that these were highly motivated patients—an important point for translating results of the trial.

Enrolled patients look like those we see in an AF clinic: average age 61, average CHADSVASC score of 1.5, about two thirds with paroxysmal AF and about one third having undergone AF ablation.


On average, alcohol intake markedly decreased in the abstinence arm, but only 43 of 70 (61%) patients achieved complete abstinence; most (86%) cut their intake by more than 70%.

Even so, the first primary endpoint, time to AF recurrence, was prolonged by 37% in the abstinence arm (P = .004). The second co-primary endpoint, mean AF burden, was also significantly reduced, and 46 patients in the abstinence group vs 25 in the control arm had 0% AF burden (P = .01).

Other good things happened: Body mass index and blood pressure were significantly reduced in the abstinence arm. The researchers used MRI to document statistically significant decreases in left atrial (LA) area and increases in LA emptying fraction in the abstinence arm.

They concluded that moderate alcohol consumption of more than 10 (standard) drinks per week is a potentially modifiable risk factor for AF. Abstinence (or a good attempt at it) was associated with reduction of AF burden, AF recurrence rates, reduction in AF symptoms, and improvement in weight loss and blood pressure.


What struck me most about Voskoboinik was his humility. In our interview and from the podium, he repeatedly and clearly noted the limitations of the study: It is not yet published, the patients were highly selected and motivated, not all patients had implantable loop recorders, and alcohol abstinence was confirmed mostly by self-report.

If more clinical scientists displayed this degree of humility about their work, science might have less issues with trust.

I look forward to the full paper, but plausibility and concordance with previous studies suggest these findings represent a true causal effect.

Observational data strongly associate alcohol intake—in a dose-dependent manner—with AF.[1] Alcohol exerts pro-fibrillatory autonomic,[2] electrical,[3] and structural[4] effects on the atrium. What's more, the observed effect in this study might have been larger had more participants been truly abstinent.

The findings of lower BP in the abstinence arm also aligns well with a recent systematic review showing that decreasing alcohol intake in people who drank more than two drinks per day was associated with significant blood pressure reduction.[5] And one hardly needs a reference to confirm that lower intake of carbohydrate-laden beverages would induce weight loss.

Voskoboinik concluded that reducing alcohol intake should be considered as part of the lifestyle intervention in moderate drinkers with AF.

I would go further. When this paper is published, it will create a bit of a moral challenge.

If the doctor and patient know that reduction of alcohol may eliminate AF, should that not be a mandatory first step before expensive and risky drugs or procedures are used? Given the vast inequities of access to healthcare, what would it say if we were ablating people so that they can continue drinking alcohol without experiencing AF?


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.