Critique 075: Moderate drinking associated with lower risk of stroke in women — 15 March 2012

Jimenez M, Chiuve SE, Glynn RJ, Stampfer MJ, Camargo Jr CA, Willett WC, Manson JE, Rexrode KM.  Alcohol consumption and risk of stroke in women.  Stroke 2012, pre-publication. http://stroke.ahajournals.org. DOI: 10.1161/STROKEAHA.111.639435

Authors’ Abstract

Background and Purpose—Light-to-moderate alcohol consumption has been consistently associated with lower risk of heart disease, but data for stroke are less certain. A lower risk of stroke with light-to-moderate alcohol intake has been suggested, but the dose response among women remains uncertain and the data in this subgroup have been sparse.

Methods—A total of 83,578 female participants of the Nurses’ Health Study who were free of diagnosed cardiovascular disease and cancer at baseline were followed-up from 1980 to 2006. Data on self-reported alcohol consumption were assessed at baseline and updated approximately every 4 years, whereas stroke and potential confounder data were updated at baseline and biennially. Strokes were classified according to the National Survey of Stroke criteria.

Results—We observed 2,171 incident strokes over 1,695,324 person-years. In multivariable adjusted analyses, compared to abstainers, the relative risks of stroke were 0.83 (95% CI, 0.75– 0.92) for < 5 g/d, 0.79 (95% CI, 0.70–0.90) for 5 to 14.9 g/d, 0.87 (0.72–1.05) for 15 to 29.9 g/d, and 1.06 (95% CI, 0.86 –1.30) for 30 to 45 g/d. Results were similar for ischemic and hemorrhagic stroke.

Conclusions—Light-to-moderate alcohol consumption was associated with a lower risk of total stroke. In this population of women with modest alcohol consumption, an elevated risk of total stroke related to alcohol was not observed.

Forum Comments

While most previous observational studies have suggested that light-to-moderate drinking is associated with a slightly lower risk of ischemic stroke (the most common type of stroke in western countries), data are not as clear for hemorrhagic stroke.  The current paper is based on the very large Nurses’ Health Study with repeated assessments of alcohol intake.  Of the total of 2,171 incident strokes occurring among the cohort, 1,206 were confirmed as ischemic strokes and 363 as hemorrhagic strokes.

For total stroke, the main findings reported by the authors were that there was a lower risk of stroke for drinkers of > 0 – 4.9 g/day and 5.0 – 14.9 g/day of alcohol.  The estimated risk of stroke in the highest category of reported alcohol intake (30-45 grams of alcohol, or up to more than 3 typical drinks) per day was 1.06, but the difference was not statistically different.  There were very few heavy drinkers in the cohort.

The authors also state in the text that former drinkers did not exhibit an elevated risk of stroke compared with lifetime abstainers, and that analyses by alcohol type provided similar results as results from total alcohol intake, but results are not shown for either comparison. In this study, the results between alcohol and stroke were similar for ischemic and hemorrhagic types of stroke.

In sensitivity analyses, the only significant differences by sub-categories were for the presence of atrial fibrillation: there was no demonstrable effect on stroke for subjects with atrial fibrillation, but the number of subjects with atrial fibrillation was very small. For the 365 strokes occurring among subjects without atrial fibrillation, risk was significantly lower for all categories of alcohol intake in comparison with non-drinkers.

Specific comments on the present study: Forum reviewers were unanimous is stating that this paper was based on a very well-done analysis with excellent ascertainment of alcohol exposure. The estimates of alcohol intake were based on the most recent measurement, although the investigators also evaluated the association of stroke with an index of cumulative alcohol consumption throughout follow up, and state that the results were similar. The measurement of all potentially confounding variables was done appropriately.

The frequency of use of hormones, multivitamins, physical activity, high cholesterol, and family history of myocardial infarction were very similar across categories of alcohol intake, suggesting that the women who consumed varying amounts of alcohol were generally similar in these characteristics to non-drinkers. The findings of lower BMI, increased smoking, greater HDL-cholesterol, and lower risk of diabetes with increasing alcohol consumption were expected, and further suggest that the assessments of alcohol were reasonably accurate.

As one Forum reviewer noted: “The Nurses’ Health Study is a strong study: a large sample size with 26 years of follow-up including admirable updated information on alcohol intake and confounders.  However, few participants were in the higher alcohol intake categories: 7% consuming 15.0 – 29.9 g/day and 4% consuming 30-45 g/day, with wide confidence limits as the obvious consequence.  These analyses may be underpowered to estimate the inflection point of greater risk of stroke at higher levels of alcohol consumption.

“The similar results for ischemic and hemorrhagic stroke are surprising and not easily explained in the context of the acute and delayed antithrombotic effects of alcohol in humans.”¹ On the other hand, in the large Kaiser Permanente study, Klatsky et al,² reported that an increase in risk of hemorrhagic stroke was seen only among much heavier consumers of alcohol (those averaging 6 or more drinks/day).  In the present study, there were not enough heavy drinkers to adequately test whether there is a significant increase in stroke risk with heavier drinking.”No data are presented on the pattern of drinking. As one Forum reviewer noted, “Their first category of alcohol drinkers reported an average of < 5 gm of alcohol per day. This is the equivalent of about one third of a typical glass of an alcoholic beverage; does this mean a very small amount every day, 1 glass every 3 days, or 2 glasses on a weekend day?” From the data available in this study, it is not possible to judge whether or not “binge drinking” affects risk of stroke.

Another Forum reviewer commented: “The results are in harmony with previous studies from this big population. While they did not show an increase in risk from alcohol for hemorrhagic stroke, the authors state that the documentation according to type of stroke was not always available. Further, only 11% of the participants had an alcohol consumption >15 g/day.” Said another reviewer: “This paper appears to have no major methodologic flaw. The similarity of results between ischemic and hemorrhagic stroke is of interest, in view of some previous data suggesting a linear tendency to increased hemorrhagic stroke with increasing amounts of alcohol.”

Interesting comments were received from a European member of the Forum: “The study on alcohol and stroke is straight forward and in line with the previous publications of this group; it is a high caliber study. The only problem I have is with the very likely underreporting of alcohol in this group. The group of ‘drinkers’ reporting below 5 g/day of alcohol seems pretty ridiculous for a central or southern European population, where regular drinking (and not just on week-ends) may be more common.”

The reviewer adds: “The authors mention that they have analysed changes in alcohol consumption over time but they do not give details on this topic; multiple observations reduce the risk of misclassification of alcohol intake. In light of the likely underreporting of alcohol consumption, it is impressive that the risk curve for total stroke exceeds that of non-drinkers only at 38 g of alcohol per day. This is a lot of alcohol for an American lady and I would like to warn the authors that they might be accused of telling the female population of the US that they may drink more than two glasses of wine or nearly a liter of beer a day without running the risk of developing a stroke, be it ischemic or hemorrhagic. In Europe we would be less hesitant with this interpretation, but we would certainly take other risks such as breast cancer into account.”

References from Forum Review

1. Lacoste L, Hung J, Lam JYT. Acute and delayed antithrombotic effects of alcohol in humans. Am J Cardiol 2001:87: 82-85

2. Klatsky AL, Armstrong MA, Friedman GD, Sidney S. Alcohol drinking and risk of hemorrhagic stroke. Neuroepidemiology 2002;21:115-122.

Forum Summary

A well-done analysis from the Nurses’ Health Study shows that the risk of total stroke is slightly lower among light-to-moderate consumers of alcohol than among subjects reporting no alcohol intake. In comparison with non-drinkers, the estimated risk is 17-21% lower for women averaging up to 15 grams of alcohol per day (a little over one drink/day by US definitions of approximately 12 grams of alcohol for a typical “drink.”). For consumers of larger amounts of alcohol, the risk of stroke appears to be slightly increased, but not statistically significantly so.

Data on the pattern of drinking (regularly, binge, etc.) were not reported. Among these predominantly light drinkers, there were no differences between effects on the risk of the most common type of stroke, ischemic stroke (due to atherosclerotic obstruction of a artery or an embolic clot) or the less-common hemorrhagic stroke from bleeding into the brain).

The results, with full adjustment for other factors that may affect risk, suggest a “J-shaped” curve for total stroke, with reductions in risk for light-to-moderate drinking and possibly an increase with greater amounts. In this study, the point at which the risk of drinkers exceeds that of non-drinkers was about 38 grams of alcohol per day (the equivalent of about 3 typical “drinks”), with a 95% confidence interval of 28 to 57 grams/day.

The study supports many previous reports from observational epidemiologic studies that have shown a slight reduction in risk of the ischemic type of stroke from moderate alcohol intake. Some, but not all previous studies, show an increase in hemorrhagic stroke for any amounts of alcohol, but that was not seen in this study, possibly because there were few heavy drinkers in this group of nurses.

*                                     *                                      *                                  *                                   *

Comments on this paper were provided by the following members of the International Scientific Forum on Alcohol Research:

Harvey Finkel, MD, Hematology/Oncology, Boston University Medical Center, Boston, MA, USA

Erik Skovenborg, MD, Scandinavian Medical Alcohol Board, Practitioner, Aarhus, Denmark

Tedd Goldfinger, DO, FACC, Desert Cardiology of Tucson Heart Center, Dept. of Cardiology, University of Arizona School of Medicine, Tucson, Arizona, USA

Gordon Troup, MSc, DSc, School of Physics, Monash University, Victoria, Australia

Giovanni de Gaetano, MD, PhD, Research Laboratories, Catholic University, Campobasso, Italy

R. Curtis Ellison, MD, Section of Preventive Medicine & Epidemiology, Boston University School of Medicine, Boston, MA, USA

Arne Svilaas, MD, PhD, general practice and lipidology, Oslo University Hospital, Oslo, Norway

Ulrich Keil, MD, PhD, Institute of Epidemiology and Social Medicine,  University of Münster, Münster, Germany