Cutting Booze to Cut AF Is Difficult for Patients to Swallow

The Alcohol-AF Study

Interviewer: John M. Mandrola, MD; Interviewee: Aleksandr Voskoboinik, MBBS


April 18, 2019

This transcript has been edited for clarity.

John M. Mandrola, MD: Hi, everyone. This is John Mandrola from | Medscape Cardiology. I'm here at the American College of Cardiology meeting in New Orleans and am pleased to be with Dr Alex Voskoboinik from Melbourne, Australia. He presented a very interesting study on alcohol and atrial fibrillation (AF). Alex, welcome.

Aleksandr Voskoboinik, MBBS: Thank you for having me, John.

Alcohol-AF Study in a Nutshell

Mandrola: Tell us the topline results of your study first, and then we'll get into the details.

Voskoboinik: Essentially, in moderate drinkers (classified as 10 or more standard drinks per week) with AF, abstinence from alcohol resulted in a significant reduction in AF burden and reduced recurrences. Secondary benefits of abstinence included improved blood pressure control and some weight loss.[1] Those were our key findings.

Mandrola: This is very important. Tell us about the background of this study.

Voskoboinik: Lifestyle intervention has really come to the forefront in the past 5-10 years. Obviously, a lot of money has been spent on AF ablation, and we have toxic drugs. But we've always had an interest in primary prevention. If we can give our patients some additional recommendations that may circumvent some of those things, I think that would be really important.

There is really a gap in the literature, in the sense that we've always known that there is a relationship.[2] A number of observational studies have looked at alcohol and AF with respect to substrate and the atrium, and also as a trigger, but today there have been no randomized studies looking specifically at that question.

Many AF patients are middle-aged, and alcohol is ubiquitous in society. It was important to look particularly in those with more moderate levels and heavier levels of alcohol use to see if significant reduction or even abstinence would make a difference.

Cardiac Function and Alcohol Consumption

Mandrola: You have published a very elegant paper showing that reduction of alcohol can actually change atrial function.

Voskoboinik: Yes. We looked at 75 patients using detailed high-density mapping, and we found in multivariate analysis that the heavier drinkers (average 14 or more standard drinks per week) seemed to have indication of more atrial substrate for AF with conduction slowing in the atrium and also lower main voltages.[3] The relationship of more habitual consumption with AF is part of the theme we've been looking at.

Mandrola: Maybe there is a plausible link for reduction of alcohol.

Voskoboinik: Yes, AF is a very heterogeneous condition, so there is probably a combination of mechanisms. There is the effect on trigger with acute autonomic effects and acute atrial electrical effects, and then there is the more long-term remodeling of the atrium itself. Obviously there are links between alcohol and other factors, such as obstructive sleep apnea and hypertension. Alcoholic cardiomyopathy has been well described as well. In some patients [alcohol] may be a relevant factor.

Patient Population

Mandrola: You did an observational study and prestudy work, but now you have to show that reduction of alcohol actually makes a difference. Who are these patients who were drinking but then stopped drinking?

Voskoboinik: This was a real challenge. The study was originally planned to run for 12 months but we shortened it to 6 months because of that challenge. Alcohol is ubiquitous in our society. In the end it was quite a selected population. We had just over 500 screened and eligible patients who did not wish to be randomized.

The patients in our study were motivated. At the end of the day, we wanted to have a study where we had a degree of compliance because that was going to be key. We know that stopping drinking in our culture is so difficult, and it will remain a challenge even with this study's findings.

Having the motivated patient in the run-in period was key, and even with that we shortened the study to 6 months to make it a feasible period of time for patients to undertake the study.

There were 70 patients in the abstinence arm and 70 patients in the control arm, so 140 patients in total with 1:1 randomization. Patients were all-comers, so paroxysmal or persistent AF. During the run-in phase they were to be in sinus rhythm at randomization and stable sinus rhythm as well. We didn't want a situation where patients would be changing antiarrhythmics during the course of the 6 months. We wanted a degree of stability. In the end, about two thirds of the patients had paroxysmal AF and one third had a history of persistent AF but were in sinus rhythm at randomization.

Mandrola: How do you know whether somebody is not drinking alcohol?

Voskoboinik: There was monthly contact with investigators. Patients kept a diary. They were provided encouragement and positive reinforcement during the contact, and we also tested urine for an alcohol metabolite during the follow-up.

Self-reporting is a limitation of the study, but we were confident that we had, at least to some degree, a motivated cohort. Even with that, complete abstinence for the entire 6 months was achieved by 61% of the patients, and over 85% reduced intake by more than 70% of their baseline.

Study Results

Mandrola: If anything, that would bias toward the null. You said in the beginning that there was a reduction in AF burden. How big was the reduction?

Voskoboinik: Because of the heterogeneous nature, it was not normally distributed, so the median burden was 0.5% in the abstinence group and 1.2% in the control group, but with the mean burden it comes out to about a 5.5%-8% difference. If you look at the actual spread of burdens, there were significantly more patients (about double) in the abstinence group who had no AF and, similarly, more patients in the control group who had more than 5% burden. That fit with the Kaplan-Meier curve as well, with a 20% lower recurrence rate overall. The findings were statistically significant, and anecdotally some patients certainly benefited more than others.

Mandrola: Were there any key secondary endpoints?

Voskoboinik: The abstinence group had a modest reduction in weight of 2.5 kg and a reduction in systolic blood pressure of just over 10 mm Hg. There was a small reduction in the control arm, but these were the two key significant secondary endpoints.

Mandrola: This is provocative because reducing alcohol intake or stopping drinking is significant. What do you think are the clinical implications and how will you translate this to your patients?

Clinical Implications

Voskoboinik: That is a really important question. I think the key message is to take an alcohol history in the AF population. Similar to coronary artery disease, which is a heterogeneous condition and has a lot of risk factors, treating the patient with AF in the clinic is not just about prescribing the antiarrhythmic. It's also about looking at weight, sleep apnea, and alcohol intake. A trial of abstinence would be something we would potentially recommend, particularly in those with higher levels of alcohol intake.

In the broader population there are some unanswered questions. There is obviously extensive epidemiologic data with the U-shape curve. We didn't look at whether mild amounts of alcohol have cardioprotective benefits specifically because the focus of this study was on AF, AF burden, and AF symptoms. But that remains an unanswered question. At the end of the day, it's part of the discussion to have with the patient when looking at their individual risk profile.

We now have a bit of data to be able to guide clinicians about additional lifestyle advice with respect to alcohol intake. We've always had that anecdotal experience and observational studies, and now we have some randomized data.

Mandrola: I also want to point out to the listeners that this was a randomized controlled trial and it was not funded.

Voskoboinik: It was an investigator-initiated study, so it didn't have any external industry support. It was run out of six hospitals in Melbourne. We are very grateful to our patients and to our collaborators who were able to complete this study. It was certainly a tough study to run, on various levels, but we got there in the end. We're delighted to have our numbers and to have results we can present.

Mandrola: Congratulations and thank you for contributing to knowledge on AF.

Voskoboinik: Thank you very much for the interview, John.

Follow Medscape on Facebook, Twitter, Instagram, and YouTube


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.