When people discover that I’m a liver doctor and a wine writer, they invariably ask, “How much can I drink without developing liver disease?” They never ask, “How much should I drink to stay healthy?”
People know that alcohol – and wine – can cause liver disease. But since the headline-making “60 Minutes” segment in 1991 popularizing the seductively simple French Paradox (the French eat a high-fat diet but have less heart disease because they drink red wine), people have embraced the concept that drinking wine, especially red wine, prevents heart disease. Despite the absence of an ironclad linkage, it’s what everyone wants to believe: Alcohol, a forbidden fruit, is actually good for you.
As a physician, I am skeptical of the health claims made for wine – or any single food, for that matter. Such medical skepticism hasn’t stopped some in the wine industry from promoting wine as a health drink. Wineries boast about the amount of resveratrol – a modern-day fountain of youth – in their wines. And yes, red wines are filled with antioxidants.
But how strong is the evidence that wine prevents heart disease? At this point, not strong enough.
How we got here
The possibility that wine might reduce heart disease came from the same type of study that suggested alcohol could be responsible for liver disease. In separate studies, World Health Organization researchers and others in the United States and Europe in 1970s compared the amount of alcohol that countries consumed in total with the country’s rate of liver or heart disease. The countries with the highest alcohol consumption – France, Italy and Spain – had the highest rates of liver disease, but also had low rates of heart disease.
These kinds of studies, called observational studies, can never determine cause and effect. They only highlight associations – possibilities. But they’re important because they direct future research. Is alcohol causing liver disease – or preventing heart disease – or is something else the real cause?
Advertisers trumpet associations, hoping to dupe consumers into thinking that their product causes the desired effect. To promote sales of prepared dinner entrees on its Web site, Stouffer’s touts research that “teens who have frequent family dinners are likelier to say they get mostly A’s and B’s in school.” The clear implication is that eating together causes students to excel in school. A more plausible explanation is that families who eat together have higher incomes and support their children’s education. Eating together doesn’t cause better grades; it’s associated with behavior that results in good grades.
We liver doctors debated for years whether it was alcohol or poor nutrition common among alcoholics that caused liver disease. Dr. Charles Lieber at the VA Medical Center in the Bronx, N.Y., settled the debate in 1974 when he fed baboons alcohol and a nutritionally complete diet. The animals developed every stage of human alcoholic liver disease.
Without proof, how has it become conventional wisdom that that wine prevents heart disease?
Observational studies around the world have shown the same association: People who drink moderately – one to two drinks daily – have less heart disease than those who don’t drink at all or who drink heavily. Although they don’t show cause and effect, the results are compelling.
Researchers also proposed several ways alcohol might be beneficial. Consumption, up to two drinks a day, increases blood levels of HDL cholesterol, so-called “good” cholesterol, which unclogs arteries. Cardiologists believe that raising HDL levels, as occurs with exercise, reduces the chance of heart disease.
Like aspirin, alcohol inhibits blood clotting; it “thins the blood.” Aspirin decreases the chance of a heart attack in some people. Similarly, alcohol’s anti-clotting effect potentially could reduce the chance of a heart attack. Does it definitively? That’s unknown.
Wine isn’t necessarily like other drinks. In July at the 32nd World Congress of Vine and Wine in Zagreb, Croatia, Serge Hochar of Lebanon’s Chateau Musar, one of the wine world’s most compelling and controversial figures, rightly chided people for lumping wine with other alcohol because wine contains additional components that have important health impacts – at least in the test tube.
Even with the alcohol removed, wine raises blood levels of polyphenols and antioxidants such as resveratrol and has dramatic effects on blood vessels. Because the oxidation of cholesterol is thought to play a role in the development of atherosclerosis – the narrowing of arteries – and heart disease, perhaps it’s the nonalcohol component of red wine that’s beneficial. And there is other evidence for the benefits of white wine, which has a different chemical makeup, that further complicate the picture.
With the barrage of commercials, and even some wineries, raving about the benefit of antioxidants, you could be forgiven for believing that polyphenols are what make wine so healthy.
But we don’t know that. Other foods contain antioxidants, including resveratrol, but it is wine that captures the public’s attention because it satisfies two desires simultaneously. It’s a silver bullet, a simple and easy solution to the complex problem of heart disease. Hard-to-follow advice – exercise and eat reasonably – can be substituted with the more palatable – literally – recommendation to drink without guilt.
On a visceral level and as a wine lover, I hope these claims are true. On an intellectual level, I know there are no magic bullets. And drinking comes with inherent risks. A doctor might precede a recommendation to exercise with a stress test to make sure you’re in proper health. But there are no tests to ensure healthy drinking.
So what’s the problem?
Scientists do not know the exact relevance of antioxidants in preventing heart disease. If they play only a minor role, increasing them even a thousand-fold would be of no clinical importance. Same with any wine-related factor.
That’s why I – like other physicians – am reluctant to recommend drinking for health reasons, despite the plausible health impacts and overwhelming observational evidence.
I am reminded of the certainty with which physicians recommended that women take estrogen supplements to reduce the risk of heart disease after menopause. Ample observational studies suggested a link: Women had less heart disease than men and this discrepancy disappeared after menopause. As with wine, there were plausible scientific explanations as to why.
In the 1990s the National Institutes of Health sponsored the Women’s Health Initiative, a cause-and-effect study of 16,000 post-menopausal women; half received supplemental hormones and half did not. Surprise. The women who received hormones had more heart disease, not less.
Why not commission a comparable study of alcohol and heart disease? The ethical considerations of chronically giving alcohol to people is one good reason.
Animal studies, similar to Lieber’s, might be possible. We need results that show cause and effect before recommending wine as a way to reduce heart disease. Try finding a government agency to fund that.
Without such studies, we are left with nagging possibilities. Maybe wine drinkers have less heart disease because they are more affluent, eat better, control their blood pressure better, exercise and do other things to take care of themselves. Perhaps we are not smart enough to know what these “other things” are.
To me, wine is not a health beverage. It’s to be enjoyed because it tastes good, and makes a meal and life more enjoyable. If moderate consumption turns out to be good for us, so much the better.
If it doesn’t, I’ll still have some with dinner.
The myths and facts, step by step.
How we got here
— In the 1970s and 1980s, observational studies by Dr. Arthur Klatsky in Oakland, Dr. Eric Rimm in Boston and many others showed that people who drank moderate amounts of alcohol had less heart disease. They raised the question “Is moderate alcohol beneficial?”
— In the 1990s, mechanisms for alcohol’s protective effects emerged. Dr. Michael Graziano in Boston showed that alcohol raised the levels of HDL cholesterol, the “good” cholesterol. Other scientists showed that alcohol inhibited clotting.
— In 2009, Dr. Kenneth Mukamal in Boston started a six-month experiment giving some people a small amount of alcohol – equivalent to one drink – daily and withholding all alcohol for others. He plans to compare blood lipids and plaque buildup in arteries of the two groups to determine if alcohol slows the cholesterol-clogging arterial disease atherosclerosis.
Does alcohol have magical powers?
Resveratrol, a polyphenol found in wine, has almost magical powers in animals, extending the lifespan of fruit flies and worms and reducing inflammation and protecting against the adverse effects of obesity in mice. These potentially ground-breaking discoveries need to be tempered by the knowledge that very high doses – the equivalent of 100-plus bottles of red wine daily – were given to mice. Importantly, after resveratrol is absorbed into the blood, it is broken down quickly and may have limited effect in humans.
The headline-creating clinical studies purporting to show that moderate wine consumption prevents Alzheimer’s disease do not show cause and effect, but are observational studies. A plausible explanation is that the healthy elderly can drink moderately without becoming impaired, but those with early or mild Alzheimer’s will be pushed over by even a small amount of alcohol and hence, don’t drink. Drinking wine, like driving a car safely, doesn’t prevent Alzheimer’s, it’s just more evidence that you don’t have it.
— Several large observational studies show an increased risk of breast cancer in women who drank moderate amounts of wine. As with the observational studies showing a reduction of heart disease, cause and effect remains unknown, but the results are worrisome.
— Women are more susceptible to all effects of alcohol, both potentially beneficial and harmful. They have a higher blood-alcohol level compared with men after drinking the same amount of wine or other form of alcohol because they have less of an enzyme in their stomach that starts to break down alcohol. Keeping the alcohol in the stomach, which occurs with eating, allows more of it to be broken down. Less is absorbed, so the blood alcohol level is lower.
— The calories in dry wine come entirely from alcohol. A 5-ounce glass of dry red or white wine (at 13 percent alcohol) contains about 110 calories.
Please note: Nothing in these articles should be construed as individual medical advice. For specific recommendations regarding alcohol consumption, consult your physician.
Wine writer Michael Apstein is a gastroenterologist at Beth Israel Deaconess Medical Center in Boston and an assistant professor of medicine at Harvard Medical School.
This article appeared on page K – 1 of the San Francisco Chronicle on Friday, August 23, 2009
Read more: http://www.sfgate.com/cgi-bin/article.cgi?f=/c/a/2009/08/23/FDF2196S5E.DTL&ao=3#ixzz1mkn8qCUT