It has been a well accepted message in cardiovascular medicine, that alcohol harms the heart and blood vessels. Similar messages have been delivered by experts in liver disease as well as by specialists from several other disciplines. In most cases, such messages have been handed out in general terms, with no information regarding the role of the doses of alcohol nor the type of patients using alcohol. However, recent studies have clarified that such messages should be refined and adapted to newer well established scientific information. Recent meta-analyses have specified that the quantity of alcohol could play a major role in the relationship between alcohol consumption and cardiovascular events.
In a first meta-analysis (1), the effect of low or moderate alcohol consumption was examined on biological markers associated with the risk for coronary heart disease. A series of 13 biomarkers were included, using fixed or random effects models. Moderate alcohol use was associated to an increase in HDL-Cholesterol, an increase in apolipoprotein A1 and adinopectin; there also was a decrease in fibrinogen levels. All these effects are accepted as having a favorable role in the estimation of risk of coronary artery disease. One would expect that an increase in triglyceride levels would decrease the favorable impact of the other markers but this was not the case: there was no change in triglyceride levels. Thus, favorable changes were seen in this study on several CV biomarkers in subjects consuming low or moderate doses of alcohol.
Obviously, one should wonder how such findings translate into clinical reality. A second meta-analysis (2) investigated whether such findings could be accompanied by a decrease in cardiovascular events. Eighty-four studies were included in this second analysis. The pooled adjusted relative risks for alcohol drinkers compared to non drinkers was 0.75 for cardiovascular mortality and more for coronary mortality in particular; all of these being clearly significant (see table 1). Dose-response analysis revealed that the lowest risk for coronary heart disease occurred at 1-2 drinks per day; for stroke mortality it occurred at less than one drink per day. These data are in line with the J-shaped curves of the relationship between alcohol consumption and mortality in patients with cardiovascular disease published a few months earlier (3) pointing toward lower mortality in such patients at low doses of alcohol.
One could wonder whether these results are applicable to all subjects. This point was examined in another meta-analysis (4) performed on almost 200.000 women and 75.000 men. The study confirmed the above-mentioned results both in men and in women; they also came out as such in younger adults but the absolute gain was smaller in younger subjects than in middle-aged or older adults. Authors explained this by the expected fact that total risk in younger is obviously smaller than in older people. These findings may have far reaching consequences. The message regarding the relationship between alcohol and cardiovascular disease should be revised. There is no doubt that alcohol can do harm to different parts of the body, more particularly the heart and other important organs like the liver. However, at low doses (1-2 drinks a day), alcohol not only is neutral, but comes out quite convincingly in these studies, as associated with benefit as was proven in these recent meta-analyses carried out in large series of subjects and published in prestigious journals.
Nevertheless, all precautions should be added in our advice, in this respect, to the population. People would like to extrapolate such a message to alcohol use in general, especially at higher doses which cause harm to the heart. Also, the risk of developing alcohol addiction should be kept in mind in all cases. Moreover, all care should be taken for people with particular responsibility such as drivers and many other sensitive cases. Such precautions need to be strongly emphasized for patients, where the advice of their physician is of essential importance as it brings into balance the clinical cardiac and vascular condition on one hand and possible benefits or harms that alcohol might bring to them on the other.
Still, these newer data let us reflect again on the dose-response curve of risk factors to cardiovascular disease and events. They are another example of the superior finesse of the cardiovascular responses not acting as a uniform "all or none" but rather in individual small steps resulting in responses specific for this or that part of the curve and as such, better adapted to body needs and life conditions.
Table 1. Adjusted relative risk for drinkers versus non drinkers (data from ref. 2). Lowest risk for coronary mortality occurred at 1-2 drinks per day.
95% Confidence Intervals
|Incident Coronary Disease||0.71||0.66-0.77|
|All cause mortality||0.87||0.83-0.92|