Article By: Harvey Finkel, MD
People picture science differently. For some, it's a Gothic laboratory and Victor Frankenstein shrieking, "It's alive!" For others, it's a quiet and understated, sterile, well-lighted, modern and nondescript workplace. For still others, science is epidemiology, the statistical study of what affects the health of populations.
What science really is amounts to attitude and discipline: objective collection and analysis of data, while avoiding confounding factors. It must be reproducible, correctable, and accepting of progress. It is ever changing.
So it should be no surprise that we are often presented, simultaneously or sequentially, with varying answers to the same question.
A recent illumination casting shadows was the instantly and widely publicized long-term study of 38,O77 male health professionals reported by Kenneth J. Mukamal, et al. in The New England Journal of Medicine of January 2OO3. Frequency of drinking turned out to be the key determinant of cardiac health in this study. Those who consumed alcohol three or four days per week had 32 percent fewer heart attacks than those who drank less than once weekly. The men drinking five to seven days weekly sustained 37 percent fewer heart attacks than the non- or almost non-drinkers.
It didn't matter whether the amount consumed on a drinking day was less than one or more than three drinks. It didn't matter what kind of beverage was consumed. Drinking with meals appeared to make no difference. Ingrained beliefs got no respect.
An accompanying editorial, "To Drink or Not To Drink?", predictably advised caution vis-a-vis alcohol, and wondered helpfully whether it was time for a direct clinical trial of alcohol in cardiovascular disease, perhaps starting with those already afflicted.
Two additional research projects published this year may shed more light. Tolstrup, et al. (Addiction, March 2OO4) followed the relation of drinking pattern to mortality in 56,535 middle-aged men and women in Denmark during a ten-year period. The risk of death for both sexes was significantly lower among those who drank frequently than for those who, while consuming the same total quantity, concentrated their consumption. The highest risk occurred in women and men who did all their drinking on one day a week or more rarely. Like almost all other studies this one showed a correlation between moderate drinking and reduced mortality.
The same issue of Addiction contains a study of Trevisan, et al. from Buffalo and Berkeley, which addresses the effects of drinking patterns on the risks of non-fatal heart attacks of 427 white men, with 9O5 controls. The results agreed with Mukamal's and Tolstrup's that frequent (or regular) drinking is healthy, in moderate quantity, but determined that food with drink is favorable. Specifically, Trevisan's findings can be summarized that men who scatter their drinking throughout the week have only 45 percent of the risk of heart attack of abstainers; men who drink on weekends only have a higher risk than all-week drinkers and no less risk than abstainers; only those who drink mostly (at least 75 percent of the time) with food, even snacks, have significantly lower risk than abstainers.
During the last twenty years, a mountain of data, mostly epidemiological, has been accumulated indicating that alcohol consumption enhances health and reduces deaths, chiefly by reducing the risk of our biggest killer: heart attack. Heart attacks kill more people - male, female, whatever race - in the developed world than anything else. The "French paradox" dramatically encapsulated this relationship: French wine drinkers have fewer heart attacks despite indolence, fatty diets and smoking. Moderate drinkers live longer than both abstainers and heavy drinkers.
For a long time the notion that drinking could be healthy had to disencumber itself from attacks of critics who could not bring themselves to accept that alcohol in any form, in any quantity, could be beneficial to health. Eventually the weight of evidence muffled their protests, but they still abound, squirming in discomfort at any admission by a reluctant government or the medical establishment that drinking might be helpful. They are certainly not scientists.
A number of respectable hypotheses, still under study, on the salutary effects of alcohol continue to contend for attention. I feel confident that we shall clarify the relationships of cardiovascular health to drinking eventually, but science is difficult to hurry. The various views are not necessarily mutually exclusive.
The simplest position is that it is not the alcohol or wine at all that helps our hearts, except figuratively, but that the lifestyle of those who partake moderately of wine happens to be more heart-healthy than that of others. This view, however, ignores direct experimental as well as epidemiologic evidence that alcohol in moderation and the antioxidant components of wine possess attributes that preserve the health of the arterial blood vessels nourishing the heart and other organs.
Some investigators propose that it is only the alcohol that counts, that it doesn't matter what one drinks, but this supposition ignores the continually accumulating evidence that compounds of wine in addition to alcohol play a role. It is the general belief among physicians familiar with the data that alcohol itself contributes about half the benefits. Alcohol (ethyl alcohol or ethanol) stimulates the liver to increase production of high-density lipoproteins (HDL) - "good cholesterol" - which leach out damaging low-density lipoprotein (LDL) - "bad cholesterol" - from blood-vessel walls, then carry it back to the liver for excretion in the bile. Alcohol also modulates some pathways of blood clotting, and by this means may further reduce heart attack risk, for a blood clot on a diseased artery wall is the coup de grace of a heart attack.
Most of us believe that a group of polyphenolic antioxidants, derived from grape skins - where they act as antifungal agents and in other, unknown, ways, and extracted into wine during alcoholic fermentation - are, in concert with alcohol, major health benefactors, perhaps well beyond heart attack risk reduction. (Inhibition of oxidation may play roles in protection from cancer, aging, degenerative diseases, dementia, and various internal toxins.) Compounds of this sort enhance the color, flavor and preservation of wine. Among the most familiar are resveratrol, quercetin and the catechins. As antioxidants, they would act favorably on LDLs, which become most toxic only after oxidation. Polyphenols also modulate blood clotting and perhaps inflammation, and may be of importance in maintaining the health and vigor of the vital inner layers of the arterial walls by activating the production of the key nitric oxide. It is too early to assess the significance of their inhibition of the bacteria, Helicobacter pylori and Chlamydia pneumoniae, suspected provokers of atherosclerosis, or of endothelin-1, a small protein accused of being a prime factor in the genesis of coronary disease. Details of the influences of moderate drinking have been discussed frequently in these commentaries.
Now we have the new studies, which suggest that frequent dollops of alcohol in any form, no matter with or without meals, according to Mukamal, substantially reduce the risk of heart attack. This deserves attention. They are large and carefully done projects by experienced hands over many years. But they are heir to the uncertainties of epidemiologic observation, notoriously tricky, and must shoulder the burden of squaring their own results with those of the multitude alluded to in the preceding paragraph. They involve a minority of women. They cannot be taken as definitive.
As for me, I would continue to drink wine moderately for enjoyment, regularly and mostly with meals, and take continued comfort in the well-supported belief that both abstention and excess are distinct risks to health and longevity.