Stroke Risk in New-Onset Atrial Fib Goes Up With Greater Alcohol Intake

June 17, 2021

There's abundant evidence linking higher alcohol intake levels to greater stroke risk and, separately, increasing risk for new-onset atrial fibrillation (AF). Less settled is whether moderate to heavy drinking worsens the risk for stroke in patients already in AF, and whether giving up alcohol can attenuate that risk. A new observational study suggests the answer to both questions is yes.

The risk for ischemic stroke was only around 1% over about 5 years in a Korean nationwide cohort of almost 98,000 patients with new-onset AF. About half the patients followed were nondrinkers, as they had been before the study, 13% became abstinent soon after their AF diagnosis, and 36% were currently drinkers.

But stroke risk went up about 30% with "moderate" current alcohol intake, compared with no intake, and by more than 40% for current drinkers reporting "heavy" alcohol intake, researchers found in an adjusted analysis.

However, abstainers who had mild to moderate alcohol-intake levels before their AF diagnosis "had a similar risk of ischemic stroke as nondrinkers," write the authors, led by So-Ryoung Lee, MD, PhD, and colleagues, Seoul National University Hospital, Republic of Korea, in their report published June 7 in the European Heart Journal. In a secondary analysis, binge drinking was also independently associated with risk for ischemic stroke.

The findings suggest "alcohol abstinence after the diagnosis of AF could reduce the risk of ischemic stroke," they conclude. "Lifestyle interventions, including attention to alcohol consumption, should be encouraged as part of a comprehensive approach in the management of patients with a new diagnosis of AF" for lowering the risk for stroke and other clinical outcomes.

"These results are pretty comparable to those obtained in the more general population," David Conen, MD, MPH, not connected to the analysis, told | Medscape Cardiology.

In the study's population with new-onset AF, there is an alcohol-dependent risk for stroke "that goes up with increasing alcohol intake, which is more or less similar to that found without atrial fibrillation in previous studies," said Conen, from the Population Health Research Institute, McMaster University, Hamilton, Ontario, Canada.

The study, "which overall I think is very well done," he said, is noteworthy for suggesting also that binge drinking, which was scrutinized in a secondary analysis, appeared independently to worsen the risk for stroke in its AF population.

Conen said the observed 1% overall risk for stroke was very similar to the rate he and his colleagues saw in a recent combined analysis of two European cohorts with AF that was usually longer standing; the median was 3 years. That analysis, in contrast, showed no significant association between increasing levels of alcohol intake and risk for stroke or systemic embolism.

However, "our confidence limits did not exclude the possibility of a small to moderate association," he said. Given that, and the current study from Korea, there might indeed be "a weak association between alcohol consumption and stroke" in patients with AF.

"Their results are just more precise because of the larger sample size. That's why they were able to show those associations," said Conen, who was senior author on the earlier report, which covered a pooled analysis of 3852 patients with AF in the BEAT-AF and SWISS-AF cohort studies. It was published January 25 in CMAJ, with lead author Philipp Reddiess, MD, Cardiovascular Research Institute Basel, Switzerland.

The two published studies contrast in other ways that are worth noting, and together suggest the stroke rate might have been 1% in both by chance, Conen said. "The populations were pretty different."

In the earlier study, for example, the overwhelmingly European patients had more comorbidities and had been in AF for much longer; their mean age was 71 years, and 84% were on oral anticoagulation (OAC).

In contrast, the Korean cohort averaged 61 years in age and only about 24% were taking oral anticoagulants. Given their distribution of CHA2DS2-VASc scores and mean score of 2.3, more than twice as many should have been on OAC, Conen speculated. "Even if you take into account that some patients may have contraindications, this is clearly an underanticoagulated population."

The European cohort might have been "a little bit more representative because atrial fibrillation is a disease of the elderly," Conen said. But "the Korean paper has the advantage of being a population-based study."

It involved 97,869 patients from a Korean national data base who were newly diagnosed with AF from 2010 to 2016. Of the total, 49,781 (51%) were continuously nondrinkers before and after their diagnosis; 12,789 (13%) abstained from alcohol only after their AF diagnosis; and 35,299 (36%) were drinkers during the follow-up, either because they continued to drink or newly started after their diagnosis.

Of the cohort, 3120 were diagnosed with new ischemic stroke over a follow-up of 310,926 person-years, for a rate of 1 per 100 person-years.

The adjusted hazard ratio (HR) for ischemic stroke over a 5-year follow-up, compared with nondrinkers, was:

  • 1.127 (95% CI, 1.003 - 1.266) among abstainers

  • 1.280 (95% CI, 1.166 - 1.405) for current drinkers

The corresponding HR, compared with current drinkers, was:

  • 0.781 (95% CI, 0.712 - 0.858) for nondrinkers

  • 0.880 (95% CI, 0.782 - 0.990) among abstainers

No significant interactions with ischemic stroke risk were observed in groups by sex, age, CHA2DS2-VASc score, or smoking status. The risk rose consistently with current drinking levels.

Hazard Ratio* for Ischemic Stroke After AF Diagnosis by Current Level of Alcohol Intake, Compared With Sustained Nondrinking
Intake Level by Weekly Consumption Hazard Ratio (95% CI)
Abstinence (0 g) 1.130 (1.006–1.270)
Mild (>0 to <105 g) 1.231 (1.109–1.366)
Moderate (105 to <210 g) 1.315 (1.134–1.525)
Heavy (≥210 g) 1.440 (1.231–1.684)
*Adjusted for age, sex, baseline body mass index, alcohol intake level, regular exercise, low income, comorbidities (including hypertension, diabetes, dyslipidemia, heart failure, previous myocardial infarction, peripheral artery disease, chronic kidney disease, chronic obstructive pulmonary disease, and cancer), baseline CHA2DS2-VASc score, and medications such as oral anticoagulants, antiplatelets, and statins.

The overall stroke rate of 1% per year is "very low," and "the absolute differences are small, even though there is a clear significant trend from nondrinking to drinking," Conen said.

However, "the difference becomes more sizable when you compare heavy drinking to abstinence."

Lee reports no conflicts of interest; disclosures for the other authors are in their report. Conen reports receiving speaker fees from Servier Canada; disclosures for the other authors are in their report.

Eur Heart J. 2021;ehab315. Full text

CMAJ. 2021;193:E117-E123. Full text

Follow Steve Stiles on Twitter: @SteveStiles2. For more from | Medscape Cardiology, 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.