No Protective Effect of Moderate Drinking on Stroke Risk

Damian McNamara

April 10, 2019

Blood pressure and stroke risk steadily rise as alcohol intake increases, results from a large population-based study show.

In findings that contradict some previous research that suggests moderate drinking may protect against stroke, an innovative study that compared conventional epidemiologic analyses and genetic analyses shows this is not the case.

The study included more than 500,000 individuals in China and showed that alcohol was responsible for approximately 8% of ischemic stroke and 16% of intracerebral hemorrhage in men.

"There are no protective effects of moderate alcohol intake against stroke. Even moderate alcohol consumption increases the chances of having a stroke. The findings for heart attack were less clear-cut, so we plan to collect more evidence," study coauthor Zhengming Chen, DPhil, Nuffield Department of Public Health, University of Oxford, UK, said in a release.

The study was published online April 4 in The Lancet.

Mixed Findings

"It's always been uncertain whether the apparent protection from moderate alcohol intake influences stroke and heart attack," study coauthor Richard Peto, FRS, professor of Medical Statistics and Epidemiology at Oxford, told Medscape Medical News.

In other words, although results of previous epidemiologic studies were reassuring for moderate drinkers, such study designs could not completely rule out other factors lowering the risk of stroke or cardiovascular disease.

One observational study published in Stroke, and reported by Medscape Medical News at the time, suggested moderate drinking was associated with lower stroke risk among women in the Nurses' Health Study.

In contrast, researchers conducting another study looked at 600,000 individuals and found even moderate drinking may increase mortality risk.

"Moderate alcohol intake has been associated with reduced cardiovascular risk in many studies, in comparison with abstinence or with heavier drinking," the current researchers note.

Prospective epidemiologic studies typically yield a U-shaped curve of association between alcohol intake and stroke risk — where moderate drinking of about one to two drinks per day is linked to the lowest risk, Peto said. However, these findings on moderate drinking may be clouded by other, confounding factors.

To get a clearer picture, the investigators prospectively studied 512,715 men and women from the China Kadoorie Biobank and followed them for about 10 years.  

Participants were recruited between June 2004 and July 2008 from 10 diverse rural and urban areas of China. At baseline, interviewers recorded socioeconomic status, medical history, smoking, drinking, diet, and physical activity. 

Investigators also monitored cardiovascular disease — including ischemic stroke, intracerebral hemorrhage, and myocardial infarction — by linking morbidity and mortality registries and electronic hospital records. The cohort's mean age was 52.

Genetic Alcohol Intolerance

Based on self-reported current and past drinking patterns, the researchers classified participants as non-drinkers, occasional drinkers, current drinkers, or ex-drinkers.

The investigators performed both traditional epidemiologic and genetic epidemiologic analyses, which used Mendelian randomization to detect differences.

A total of 161,498 participants were genotyped for the two common genetic variants, rs671 for the ALDH2 gene and rs1229984 for the ADH1B gene, which together substantially reduce alcohol intake.

East Asians who feature genotypes associated with these variants can experience Alcohol Flushing Syndrome. The build-up of the alcohol metabolite acetaldehyde in this syndrome can cause uncomfortable facial flushing, nausea, and tachycardia.

Among men, these genetic variants caused a 50-fold difference in average alcohol intake, from near zero to about four drinks per day.

"Those affected cannot drink much or don't drink at all," Peto said. In a research sense, "it's a randomly allocated inability to drink."

Men with these genetic variants drink 10-fold less than unaffected men, Peto estimated. In addition, women in China drink about 20 times less alcohol than men, making them an ideal control group to rule out other factors that may reduce, or increase, stroke risk.

At baseline, 33% (69,897) of men and 2% (6245) of women reported drinking some alcohol in most weeks, mainly spirits. But there was wide variation in the prevalence of drinking across geographic areas.

Consistent with previous studies, the conventional analyses in men revealed U-shaped associations between self-reported alcohol intake, stroke, and myocardial infarction, with moderate alcohol use associated with the lowest risk.

No U-Curve

However, the genetic analyses showed that among men who drank an average of four drinks per day, the relative risk for ischemic stroke was 1.27 (95% confidence interval [CI], 1.13 – 1.43; P = .0001), 1.58 for intracerebral hemorrhage (95% CI, 1.36 – 1.84; P < .0001) and 1.38 for total stroke (95% CI, 1.26 – 1.51; P < .0001).

Consuming about four drinks daily also was also associated with a mean systolic blood pressure increase of 4.3 mm Hg. 

The mean intake of alcohol in all men was about 100 g per week. The relative risk for ischemic stroke at this consumption level was 1.09 (95% confidence interval, 1.04 – 1.14). The relative risk for intracerebral hemorrhage was 1.18 (95% CI, 1.12 – 1.24), and for total stroke it was 1.12 (95% CI, 1·09 – 1·16). Adjusting for potential confounders did not significantly alter the results.

Among men, both conventional and genetic analyses showed increasing alcohol intake was associated with elevated blood pressure, concentrations of HDL cholesterol and γ-glutamyl transferase, a serum marker for alcohol-related liver disease (all P < .0001).

The genetic findings suggest the "apparently protective effects of moderate alcohol intake against stroke are not mainly caused by alcohol itself, and are largely artifacts of reverse causation and confounding," the researchers note.

Since participants reported drinking mainly spirits, the effects of other drinks, such as red wine, could not be evaluated.

In terms of future research, the investigators would like to conduct a study with a larger cohort and longer follow-up that combines findings from their group and others, said Peto.

"What we would really like to do is answer the question about heart attacks. It's not as easy, however, because they experience more stroke in China, Korea, and Japan than heart attacks," he added.

More Research Warranted

In an accompanying editorial, Shiu Lun Au Yeung, PhD, and Tai Hing Lam, MD, praised the study for its prospective design.

The authors "showed that, at least for stroke, the apparently protective effects of moderate alcohol consumption were largely non-causal using Mendelian randomization, which is the optimal non-experimental design to minimize confounding,"

Future research comparing alcohol consumption and abstinence should take the time-varying nature of alcohol use into account and incorporate longer-term follow-up, added Yeung and Lam, who are both affiliated with School of Public Health, Li Ka Shing Faculty of Medicine, The University of Hong Kong, Hong Kong Special Administrative Region, China.

More information on the potential benefits of quitting alcohol for people at different consumption levels would be "ethical, informative, and warranted," they add.

In terms of addressing alcohol-attributable diseases on a global scale, the editorialists call for a global initiative to address alcohol use similar to the successful World Health Organization Framework Convention on Tobacco Control.

"The lag between the peak of alcohol prevalence and alcohol-induced diseases, which could be several decades, might confuse policy makers, and be used by the alcohol industry and related interests to argue against more stringent control measures by claiming that moderate drinkers should not be deprived of the supposed health benefits. We need to learn from tobacco control and unite to advocate for a Framework Convention for Alcohol Control," they write.

Peto and some coauthors report grants from the Hong Kong Kadoorie Charitable Foundation, Wellcome Trust, UK Medical Research Council, GlaxoSmithKline, the British Heart Foundation, and Cancer Research UK during the conduct of the study. Yeung and Lam have disclosed no relevant financial relationships.

The Lancet. Published online April 4, 2019. Full text, Editorial

Follow Damian McNamara on Twitter: @MedReporter. For more Medscape Neurology news, join us on Facebook and Twitter.


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.