Antioxidants: Bad for the heart?
A new study, questioning conventional wisdom, suggests caution in using antioxidant supplements
Researchers from the University of Washington have reported that patients taking antioxidant vitamins in addition to statin and niacin therapy failed to increase their HDL cholesterol (the "good" cholesterol) as much as patients not taking antioxidants. These results, reported in the August 9, 2001 issue of Arteriosclerosis, Thorombosis, and Vascular Biology, are but the latest in a series of disappointing results in trials examining the ability of antioxidants to prevent heart disease.
Why antioxidants sound so attractive
It is known that the oxidation of LDL cholesterol contributes to the blood vessel damage that causes coronary artery disease and heart attacks. By helping to prevent the oxidation of LDL cholesterol (or, in scientific parlance, by reducing "oxidative stress,") it has long been believed that antioxidant supplements like vitamins E and C might help slow or even prevent the formation of plaques within coronary arteries. Indeed, several epidemiological studies have been published that seem to confirm that antioxidant vitamins can reduce the incidence of cardiac events such as heart attacks or the need for bypass surgery. So: the theory behind antioxidants has seemed sound, and several studies seem to confirm their effectiveness. As a result, millions of individuals (including many doctors) now routinely take vitamin E and C supplements.
Randomized trials have not been kind to antioxidants
Unfortunately, epidemiological studies are considered suboptimal for deciding whether a therapy works. In epidemiological studies, a population of patients is followed for a period of time, keeping track of which patients happen to be using the therapy in question and which do not. Outcomes are then compared between patients using and not using the therapy. Because there may be inherent (but unidentified) differences between those who use a therapy and those who do not, and because patients who choose to use a therapy might experience improvement due to a "placebo effect," such studies are regarded as offering suggestive evidence, but not proof, of a treatment's effectiveness. (With a placebo effect, a "treatment" that actually has no specific physiological benefit produces positive results because patients expect it to. The placebo effect is now recognized as sometimes being quite powerful in alleviating symptoms, and has confounded the evaluation of countless therapies in medicine.)
To "really" measure a treatment's effectiveness, randomized trials are usually considered necessary. In randomized trials, groups of "identical" patients (that is, as close to identical as possible) are randomly chosen to receive the treatment in question - those who are not randomized to receive the treatment are given placebo. Ideally, neither the patients nor their doctors are aware of whether they are actually receiving the treatment or the placebo. Randomized trials, then, at least in theory, minimize any hidden, systematic differences between patients taking or not taking the treatment being studied, and eliminate any falsely positive outcomes that may result from a "placebo effect" (since patients taking either the actual treatment or the identical-appearing placebo ought to have the same degree of placebo effect.)
During the last two years, a number of randomized trials using antioxidant vitamin supplements have finally been reported, and the results have generally been disappointing. Because of the failure of randomized trials to demonstrate a benefit from taking antioxidants, both the American Heart Association and the Institute of Medicine have released recent statements saying that, while a diet rich in antioxidant vitamins seems prudent, there is insufficient evidence to recommend using supplements of of vitamin C, vitamin E, beta-carotine, selenium, or other antioxidants to prevent heart disease.
The latest study suggests the possibility of harm
The study from the University of Washington, reported last week, brings up the possibility that antioxidant therapy may do more than merely fail to halt the progression of coronary artery disease. This new study suggests the possibility of harm.
In this trial, patients with coronary artery disease who also had low levels of HDL cholesterol were randomly assigned to one of 4 groups: 1) low-dose statin and niacin therapy; 2) low-dose statin and niacin therapy plus a cocktail of antioxidants; 3) a cocktail of antioxidants without statin and niacin, and 4) placebo. The cocktail of antioxidants consisted of vitamin E, vitamin C, beta-carotene, and selenium.
The study results showed that the increase in HDL levels seen in patients receiving statin-niacin therapy was eliminated when they also received the antioxidants. That is, in these patients the antioxidants were potentially harmful.
As it turns out, this "negative" finding is less than statistically perfect. (The finding that antioxidants would reduce HDL levels was not postulated ahead of time by the investigators, but instead was discovered in a routine review of their data, bringing up the possibility that the reported finding was a random data error instead of an actual physiologic phenomenon.) But the primary endpoints of the study also suggest that antioxidants blunt the benefits seen with statin-niacin therapy. (While patients receiving statin-niacin had a 4% reduction in coronary artery blockage, those who received antioxidants in addition to statin-niacin had a 7% increase in blockage. In contrast, those receiving antioxidants alone had a 15% increase in blockage, and those receiving placebo had a 34% increase in blockage.)
The bottom line: this new study seems to validate the unpopular stance on antioxidant supplements taken by both the American Heart Association and the Institute of Medicine.
What to do about antioxidants
DrRich is sorry to say that he must agree with the policy statements of the AHA and the IOM. So far, there is insufficient data from well-designed randomized trials to issue a general recommendation for people to take supplements of the antioxidant vitamins in order to prevent heart disease. Eating lots of fruits and vegetables - in other words, a good diet - seems the most prudent course at the moment.
But are antioxidants actively harmful? Based on data from randomized trials completed to date, it is likely that they are usually not harmful. But it does seem prudent to recommend that - especially if statins and niacin are being used in the attempt to increase HDL levels - either antioxidant supplements should be avoided, or at the very least, HDL levels should be remeasured both with and without antioxidant supplementation, in order to document whether those supplements are blunting the benefits of statin-niacin therapy.