The Alcohol-AF study was a multicenter, prospective, open-label study conducted at six Australian hospitals. All subjects underwent comprehensive rhythm monitoring via an implantable loop recorder or an existing pacemaker. Twice daily they triggered the AliveCor mobile phone app together with Holter monitoring. Patients assigned to abstinence were counseled to abstain completely, got written and verbal advice to help with compliance, and had contact with the investigators on a monthly basis. They also underwent urine testing to monitor compliance. All study participants kept a weekly alcohol intake diary, reviewed monthly by investigators.
Participants averaged 17 standard drinks per week at enrollment. Two-thirds were primarily wine drinkers. At the 6-month mark, patients in the abstinence arm averaged two drinks per week, for an 88% reduction from baseline, and 43 of the 70, or 61%, had remained completely abstinent. The control arm averaged a 20% reduction in weekly consumption.
One of the two co-primary endpoints was recurrent AFib episodes lasting for at least 30 seconds. The rate was 53% in the abstinence group and 73% in controls. The time to first recurrence averaged 118 days in the abstinence group, compared to 86 days in controls, for a 37% prolongation through abstinence.
The other primary endpoint was total AFib burden. During the 6-month study, the abstinence group spent a mean of 5.6% of their time in AFib, compared to 8.2% in controls. The median times were 0.5% and 1.2%.
In a multivariate analysis, neither AFib type, duration, history of AFib ablation, age, nor gender predicted AFib recurrence. In fact, the only significant predictor was alcohol abstinence, which conferred a 48% reduction in risk.
Symptom severity, a secondary endpoint, showed an impressive between-group difference at follow-up. Ten percent of patients in the abstinence group had moderate or severe symptoms at follow-up as assessed via the European Heart Rhythm Association score of atrial fibrillation, compared to 32% of controls, for an absolute 22% difference. And 90% of the abstinence group had no or mild symptoms, as did 68% of controls. Nine percent of the abstinence group had an AFib-related hospitalization, as did 20% of controls.
Cardiac MRI, performed in all subjects, showed that the abstinence group experienced a significant reduction in left atrial area, from 29.5 cm2 at baseline to 27.1 cm2 at follow-up. They also experienced significant improvement in left atrial mechanical function, with their left atrial emptying fraction climbing from 42% to 50%.
The 3-kg weight loss from a baseline of 90 kg in the abstinent group represented a 0.7 kg/m2 reduction in body mass index.
In terms of the likely mechanism of benefit of alcohol abstinence in moderate drinkers, Dr. Voskoboinik noted that he and his colleagues recently demonstrated that regular moderate alcohol consumption, but not mild intake, is associated with potentially explanatory atrial electrical and structural changes, including conduction slowing, lower global bipolar atrial voltage, and an increase in complex atrial potentials (Heart Rhythm. 2019 Feb;16[2]:251-9).
Alcohol has been shown to be linked with AFib through multiple mechanisms. Alcohol has adverse effects on the atrial substrate, including promotion of left atrial remodeling, dilation, and fibrosis via oxidative stress, and alcohol has a contributory role in hypertension and obstructive sleep apnea. Alcohol can act as an acute trigger of AFib through binge drinking – the holiday heart syndrome – which causes autonomic changes, electrolyte abnormalities, and electrophysiologic effects, he continued.
Epidemiologic evidence in support of the notion that moderate drinking increases the risk of AFib includes a Swedish meta-analysis of seven prospective studies comprising 12,554 patients with AFib. The researchers concluded that the relative risk of AFib rose by about 8% with each drink per day, compared with the risk of a reference group composed of current drinkers at a rate of less than one drink per week (J Am Coll Cardiol. 2014 Jul 22;64[3]:281-9).
Although discussants were wowed by the 61% complete abstinence rate in the trial, Dr. Voskoboinik cautioned that the study participants were a highly motivated subset of the universe of moderate drinkers with AFib. And even so, “It was an incredibly challenging study to run,” he said.
“I think lifestyle change is a challenge, and with alcohol particularly so, because alcohol is so ubiquitous in our society. I think the important key message for us as physicians is to take an alcohol history and have a discussion with the patient. And we now have some data to show them. But at the end of the day, it’s up to the patient in conjunction with the physician,” he observed.
Discussant Annabelle S. Volgman, MD, noted that 85% of study participants were men, and because of important differences in AFib between men and women, she doesn’t consider the findings applicable to women.
Bruce Jancin/MDedge News
Dr. Annabelle S. Volgman
“You need to redo the study in women,” advised Dr. Volgman, professor of medicine at Rush University, Chicago, and director of the Rush Heart Center for Women.
She suggested that an interventional trial such as this one is a great opportunity to utilize wearable device technology, such as the Apple Watch, which can detect irregular pulse rhythms as demonstrated in the Apple Heart Study. This technology is especially popular with younger individuals.
“A lot of young men and women drink a lot of alcohol,” Dr. Volgman noted.
“That’s a great idea. It gets around a problem with AliveCor, which can miss episodes while you’re sleeping,” Dr. Voskoboinik replied.
He reported having no financial conflicts regarding the investigator-initiated and -funded Alcohol-AF trial.
SOURCE: Voskoboinik A. ACC 19, Session 413-08.