Giving Up Alcoholic Drinks in Atrial Fib May Cut AF Burden, Recurrence Risk

March 26, 2019

NEW ORLEANS — Adults who are regular drinkers and have had paroxysmal or persistent atrial fibrillation (AF) may cut their chances of AF recurrence by almost half if they stop drinking, or at least substantially lower their alcohol intake, suggests a small randomized trial.

It also suggested such alcohol "abstainers" might show a decline in AF burden and symptom severity, and at the same time lower their blood pressure and maybe lose some weight.

The results from the Alcohol-AF study, which included mostly men, are in line with abundant observational data on alcohol intake as a promoter of AF that had apparently never been tested in a randomized controlled trial.

The new study comes with the virtues not only of an abstinence group compared with a continued-drinking control group, 70 patients in each, but also careful tracking of compliance using urine checks for ethanol metabolites and monitored documentation of AF recurrences.

Although zero alcohol consumption was prespecified for patients assigned to the abstinence group, only 43 of them, about 61%, actually gave up alcohol entirely for the study's 6 months.

The remaining abstainers ranged in their actual alcohol consumption, even up to their intake levels before the trial.

Still, overall, "86% of them were able to drastically reduce their alcohol intake by more than 70% of what they were having before the study," Aleksandr Voskoboinik, MBBS, Alfred Hospital, Melbourne, Australia, told | Medscape Cardiology.

For example, the abstainers cut their average initial intake of 17 drinks per week by about 88%, to about two per week.

"We still saw benefits in those who significantly cut back," observed Voskoboinik, who had presented Alcohol-AF here at the American College of Cardiology 68th Annual Scientific Session (ACC.19).

So, the trial's formal conclusion wasn't that abstinence should be recommended for all moderate drinkers with AF; rather, it was that all such patients should at least "significantly reduce" their intake.

He and his colleagues wanted an evidence-based but also practical message that would be palatable for patients with AF, and abstinence from alcohol is a hard sell for people who are at least moderate drinkers, Voskoboinik said.

"I don't think we have the evidence base to be quite that dogmatic, and if you have more realistic goals, I think they'll also be more achievable."

Indeed, "in some ways, one of the most remarkable findings is that you had an abstinence rate of 61%," observed P. Barton Duell, MD, Oregon Health and Science University, Portland, as a panelist after Voskoboinik's presentation of the trial. "So the question is, how did you do that?"

Nearly 700 patients were considered for the trial, all of whom met the inclusion criteria and "would have been perfect for the study," Voskoboinik replied. Of the 521 patients excluded for various reasons, 491 refused to participate because they wouldn't agree to be potentially assigned to the abstinence group.

"So we were left with a selected, motivated group of patients."

Whether the motivation was enough for abstainers to continuing keeping their alcohol intake down after the study's conclusion is not objectively known, Voskoboinik said when interviewed. But anecdotally, "a large proportion of the patients were drinking less than they were when they entered the study."

They tended to be those whose symptoms had improved during the trial, whereas those who didn't feel a difference were more likely to return to their prestudy drinking levels, he said.

"These are much needed data. We have been recommending this to our patients without evidence-based trials, so I'm so pleased that you finally did it," commented Annabelle S. Volgman, MD, Rush University, Chicago, from the panel after Voskoboinik's presentation.

But she queried whether the trial's consideration of alcohol intake levels and its definition of "moderate" drinking were different for women and for men. "Women don't have as much leeway with alcohol intake," she observed.

Voskoboinik acknowledged that there was no such adjustment by sex, a limitation of the trial. "Unfortunately we don't have sufficient data to extrapolate this to women. More than 80% of our population was male."

The study entered 140 patients (mean age, 62 years) at six centers with paroxysmal (63%) or persistent (37%) AF — in sinus rhythm, after cardioversion if needed — who had been consuming at least 10 drinks (a drink defined as containing 12 g of ethanol) per week. Their mean weekly intake was 17.

Seventy were randomly assigned to an abstinence group, counseled not to consume alcohol at all, and saw study clinicians in person every month. An equal number were counseled to continue their usual alcohol consumption as the control group.

The abstinence group underwent regular urine testing for ethyl glucuronide (EtG), a product ethanol metabolism.

For most patients, cardiac rhythm was monitored with the Kardia Mobile (AliveCor) smartphone-based electrocardiographic (ECG) device, which they were instructed to use twice daily or on feeling symptoms suspected to be from AF. Fifty of the trial's patients instead were monitored continuously by implantable loop recorder or their previously implanted pacemakers.

The AliveCor device was actually a plus for study participants, in that it provided regular feedback. "The patients really liked it, compliance was actually pretty good," Voskoboinik said when interviewed.

On the other hand, "this is a great use for wearables," proposed Volgman from the panel, on the heels of the Apple Heart Study presentation at ACC.19. Monitors like the Apple Watch in the study might be especially suited for people at risk who consume a lot of alcohol, she said.

At 6 months, the coprimary end point of recurrent AF was seen in 37 patients in the abstinence group and 51 in the control group, for an adjusted hazard ratio (HR) of 0.52 (95% CI, 0.30 - 0.89; P = .014) by intention to treat. Recurrent AF was defined as any atrial tachyarrhythmia lasting at least 30 seconds after the study's initial 2-week blanking period.

The other coprimary end point, the percentage of time spent in AF, also was significantly lower in the abstinence group.

Both end points were blindly adjudicated.

Six-Month Outcomes in Abstinence and Control Groups by Intention to Treat
End Point Abstinence, %
(n = 70)
Control, %
(n = 70)
P Value
Recurrent AF* 53 73 .014
Mean AF burden* 5.6 8.2 .016
No AF at 6 months 47 25 .01
AF symptom severity, mild or no symptoms (EHRA criteria) 90 68 <.05
AF-related hospitalizations 9 20 .053
*coprimary end point.
EHRA=European Heart Rhythm Association.

Weight and body-mass index decreased significantly in the abstinence group (P < .001 for both); they rose significantly in the control group (P = .04 and P = .03, respectively). Blood pressure also fell significantly in the abstinence group.

Six-Month Changes in Weight and Systolic Blood Pressure: Abstinence vs Control Groups
End Point Abstinence
(n = 70)
(n = 70)
P Value
Change in weight, kg –2.7 +0.8 <.01
Change in systolic BP, mm Hg –12.4 –1.0 .02

However, AF recurrence was not predicted by changes in weight or systolic blood pressure, nor by age, sex, type of AF, duration of previous AF, or whether the patient had a previous AF ablation.

The current study strengthens the evidence base at the disposal of clinicians when talking to patients with AF who drink, Voskoboinik told | Medscape Cardiology.

"Just as with patients with coronary artery disease — whom we ask about other risk factors, like smoking, hypertension, and dyslipidemia — I think now AF is probably on a similar sort of footing," he said.

"In the past, we'd just deal with symptoms and anticoagulation, and this sort of reinforces the message to talk to patients about alcohol intake, weight loss, sleep apnea, and all those other things. This gives you another potential area to intervene on."

Keeping AF at bay with risk factor modification, "I think, will be an ongoing challenge, just like maintaining weight loss," Voskoboinik said. "But at the end of the day, it's giving the patient something that's potentially empowering them to make a positive change in their outcomes."

Voskoboinik had no relevant disclosures.

American College of Cardiology (ACC) 68th Annual Scientific Session: Abstract 413-08. Presented March 18, 2019.

Follow Steve Stiles on Twitter: @SteveStiles2. For more from, follow us on Twitter and Facebook.


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.