Light to Moderate Drinking: Likely Cardioprotective, But Recommended?

February 28, 2011

February 28, 2011 (Calgary, Alberta) — It's time to acknowledge the pile of evidence that light to moderate alcohol consumption is not only good for cardiovascular health, it could potentially be recommended for CV risk reduction, according to authors of two meta-analyses published online February 22, 2011 in BMJ [1,2].

An analysis of prospective cohort studies showing alcohol effects on cardiovascular end points, with first author Dr Paul E Ronksley (University of Calgary, AB), suggested that most any level of alcohol intake is protective against CV mortality, incident CHD, and CHD mortality, while intake of up to one drink per day is protective against incident stroke and stroke mortality. The findings are consistent with a vast evidence base linking light to moderate alcohol intake with reduced CV risk.

The other report focused on clinical intervention studies of alcohol effects on biomarkers associated with CV disease and concluded that moderate intake raises levels of HDL cholesterol, apolipoprotein A1, and adiponectin and reduces fibrinogen levels. In their analysis, Dr Susan E Brien (University of Calgary) et al defined moderate alcohol intake as up to a drink per day for women and two drinks per day for men.

Although observational studies can't establish cause and effect, the group observes, together the two reports make a compelling case for alcohol as an actual cause of the reduced CV risk long associated with light to moderate drinking.

But What Do You Do With the Information?

"Then the question becomes, what do you do with that [information]? There are two levels at which it needs to be considered. One is: what does a doctor say to a patient in a clinical setting? And the second is: what do public-health officials tell the public?" Dr William A Ghali (University of Calgary), senior author of both reports, said for heartwire .

"With respect to public-health messages," he and his coauthors write, "there may now be an impetus to better communicate to the public that alcohol, in moderation, may have overall health benefits that outweigh the risks in selected subsets of patients."

In clinical practice, they contend, the evidence base supporting CV benefits from alcohol intake could be the basis for "counseling for selected patients to incorporate moderate amounts of alcohol into their diets to improve their coronary heart disease risk."

But before that could happen, the strategy would have to be evaluated "in pragmatic clinical trials that assess the questions of optimal patient selection, compliance, risks, and benefits."

I just don't think we should close our eyes and ears to this evidence.

Such trials wouldn't focus as much on links between alcohol intake and clinical outcomes, but more on "the receptivity of both physicians and patients to the recommended consumption of alcohol for therapeutic purposes and the extent to which it can be successfully and safely implemented."

Ghali recognizes that he and his colleagues are extending the public debate about alcohol and CV health beyond its comfort zone, in which people who already partake are assured their moderate drinking may improve cardiovascular health. Teetotalers are not currently advised that if they start drinking, it will be good for their hearts.

On the other hand, public-health messages advocating "consumption of alcohol for therapeutic purposes" would likely be offensive to "people who view, and people who have good reason to view, alcohol as a very dangerous substance," Ghali acknowledged.

"I think we should undertake a process of dialog and debate around what should be said and then evaluate whatever we come up with in terms of public-health messages to see what impact that has on behaviors," he said. "I just don't think we should close our eyes and ears to this evidence."

Outcomes and Biomarker Studies: Consistent Message

The group's outcomes meta-analysis included 84 prospective cohort studies of adults without preexisting CV disease, both drinkers and nondrinkers. Follow-up ranged from 2.5 to 35 years (mean 11 years); 85% of the studies followed participants for more than five years.

Adjusted Relative Risk (RR) for Outcomes, Drinkers Compared With Nondrinkers

End point Number of studies RR (95% CI)
CV mortality 21 0.75 (0.70–0.80)
Incident CHD 29 0.71 (0.66–0.77)
CHD mortality 31 0.75 (0.68–0.81)
Incident stroke 17 0.98 (0.91–1.06)
Stroke mortality 10 1.06 (0.91–1.23)

The adjusted RR for death from any cause was 0.87 (95% CI 0.83–0.92) for drinkers compared with nondrinkers, an analysis that accounted for the opposing effects of potentially fatal alcohol-related risk (as from car accidents or liver cirrhosis) and any protective effect on CV health, Ghali observed.

A look at end points by level of alcohol intake showed that the CV-event risk was lowest at one to two drinks per day (figuring that one drink contains 2.5 g to <15 g alcohol). Reductions in stroke risk or stroke mortality weren't seen at greater than one drink per day.

Adjusted Relative Risk (RR) for Outcomes by Daily Alcohol Intake, Drinkers vs Nondrinkers

End point <2.5 g/d, RR (95% CI) 2.5–14.9 g/d* RR (95% CI) 15–29.9 g/d RR (95% CI)
CV mortality 0.71 (0.57–0.89) 0.77 (0.71–0.83) 0.75 (0.70–0.80)
Incident CHD 0.96 (0.86–1.06) 0.75 (0.65–0.88) 0.66 (0.59–0.75)
CHD mortality 0.92 (0.80–1.06) 0.79 (0.73–0.86) 0.79 (0.71–0.88)
Incident stroke 0.81 (0.74–0.89) 0.80 (0.74–0.87) 0.92 (0.82–1.04)
Stroke mortality 1.00 (0.75–1.34) 0.86 (0.75–0.99) 1.15 (0.86–1.54)

*About one drink per day

The adjusted risk of incident stroke was significantly increased at a daily intake exceeding 60 g: RR 1.62 (95% CI 1.32–1.98).

The group's "companion review" included 44 studies that "evaluated the circulating blood levels of [prespecified] biomarkers during a period of intentional, prescribed alcohol feeding vs a period of no alcohol use."

No significant effect of alcohol use was seen on total cholesterol levels or levels of LDL cholesterol, triglycerides, or Lp(a). However, "significant changes in levels of high-density lipoprotein cholesterol, fibrinogen, and adiponectin after alcohol consumption were well within a pharmacologically relevant magnitude."

The effects were independent of whether alcohol was consumed in beer, wine, or liquor.

Pooled Mean Differences in Biomarker Levels During Periods of Alcohol Use vs Periods of Nonuse

Biomarker Number of studies Mean difference (95% CI)
HDL cholesterol (mmol/L) 33 0.094 (0.064–0.123)
Apolipoprotein A1 (g/L) 16 0.101 (0.073–0.129)
Fibrinogen (g/L) 7 −0.20 (−0.29 to −0.11)
Adiponectin (mg/L) 4 0.56 (0.39–0.72)

All changes alcohol use vs nonuse p<0.01

The effect on HDL cholesterol went up with alcohol-intake level, increasing by 0.072 mmol/L at one to two drinks per day up to 0.14 mmol/L at more than four drinks per day (p=0.013 for the trend), the group reports.

"This degree of increase is greater than any currently available single pharmacological therapy, including fibrates (approved by the Food and Drug Administration for people with low levels of high-density lipoprotein cholesterol)."

"I don't want to come across as a proponent of widespread consumption of alcohol," Ghali said; concern that promoting light to moderate alcohol consumption could cause harm is appropriate. "But there isn't proof that recommending it would lead to harm, and that's worth just keeping in the back of our minds."

Ghali said the authors have no conflicts of interest. The analyses were supported by the Robert Wood Johnson Foundation, Substance Abuse and Mental Health Services, and Administration Center for Substance Abuse Treatment. Individual coauthors were supported by the Alberta Heritage Foundation for Medical Research and the Canadian Institutes of Health Research.


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.