10.mar.07
The Gazette (Montreal)
Joe Schwarcz, director of McGill University's Office for Science and Society (www.OSS.McGill.ca) who can be heard every Sunday from 3-4 p.m. on CJAD, writes that antioxidant supplement manufacturers are circling the wagons and taking aim at the messenger. In this case the target is Goran Bjelakovic, leader of a research group that has published a study on the effectiveness of antioxidant supplements in disease prevention.
The surprising conclusion? Rather than preventing disease, beta carotene, vitamin A and vitamin E actually increased the risk of death! This startling analysis is not from the pages of some tabloid, it was in the Journal of the American Medical Association, one of the prime peer-reviewed scientific publications in the world.
Understandably, the dietary supplement industry is upset. For years we have heard frightening tales of treacherous free radicals roaming through our body, wreaking havoc with molecules essential to our health. Heart disease, cancer, macular degeneration and even Alzheimer's disease have been linked with free radical damage. Since antioxidants are the natural enemies of free radicals, exploring potential benefits of dietary supplements containing antioxidants is reasonable. After all, fruits and vegetables loaded with antioxidants are clearly linked with good health, and at least in the test tube, antioxidants can wipe out free radicals.
Preventing disease with antioxidants is an alluring theory. But a theory has to be put to the test before it can be accepted as doctrine. The path of scientific progress is littered with ideas that crumbled in the face of experimental evidence. At one time the Earth being the centre of the universe made sense as did the idea that atoms were minuscule hard spheres. It made sense to put heart attack victims on long-term bed rest and to associate ulcers with stress and spicy foods. But of these notions fell by the wayside as experimental evidence against them mounted. Now, the theory that antioxidant supplements are helpful in preventing disease is being challenged.
Bjelakovic and his group receive no funding from any commercial enterprise and appear to have no reason to either knock or support dietary supplements. Their motivation is purely scientific. With people popping antioxidant pills like candy, why not try to determine if they are getting the hoped-for benefits? Manufacturers maintain that they are, and point to piles of studies in scientific literature that buttress their arguments.
But the literature in this area is vast, and is replete with studies that do not support the use of supplements. In general these do not attract journalists' attention and are not commonly reported. Bjelakovic decided to scour the literature for all relevant studies and pool the results, hoping for a legitimate conclusion about the value of antioxidants. Such a "meta analysis" is a common way to try to shed light on a confusing topic. Individual studies may show trends that are not very significant, but pooled results can be very meaningful.
There is certainly no dearth of research in this area. Bjelakovic's team tracked down hundreds of published trials and whittled these down to 68 that met the criteria for proper blinded, randomized, placebo-controlled studies that investigated the effects of beta-carotene, vitamin A, vitamin C, vitamin E and selenium on health. All are established antioxidants and excellent candidates for efficacy in preventing disease. Some trials used low doses of supplements, some high; some trials lasted months, others many years. Some used single antioxidants, others used combinations. But the strength of a meta analysis lies in pooling results from many studies, evening out variables and allowing an overall conclusion. In this case, the surprising result was the increased mortality among subjects who had been taking antioxidants.
Perhaps this shouldn't have been that surprising, given that in 2004 Bjelakovic had published another meta analysis in the prime British medical journal, The Lancet, in which he had found no benefit from antioxidants in the prevention of gastrointestinal cancers and had also noted increased mortality associated with antioxidants. In the face of much criticism from supplement manufacturers, Bjelakovic has now expanded his analysis from the effect of antioxidants on gastrointestinal cancers to their effect on all diseases. The data appear to be robust. More than 230,000 participants were involved in the 68 trials, 21 one of which focused on healthy subjects who were taking antioxidants to prevent disease. The results are the same as in the previous analysis: no apparent benefit and possibly increased risk.
The criticism has been furious. Many relevant trials were excluded, critics claim. Causes of death were not determined and may have had nothing to do with supplements. Subjects took other supplements and prescription drugs that could have clouded the issue. There is no biological mechanism that can explain potential harm by antioxidants.
Well, that is not exactly true. Certain white blood cells, for example, attack toxins by generating free radicals and possibly antioxidants can interfere with this activity. No doubt, some of the criticisms aimed at the Bjelakovic analysis are valid, but given that so many subjects and so many studies were involved, any significant benefit from the antioxidants examined would have become apparent.
I don't think antioxidant supplements will kill you, but they have been hyped beyond what scientific evidence can support. The assumption has been that the benefits of fruit and vegetable consumption can be ascribed to their antioxidant content. Actually, these foods contain thousands of compounds that can interfere with disease processes in various ways and it is becoming clear that isolating a few and putting them into a pill just doesn't deliver. Concentrate instead on eating your berries, apples, lentils, onions, garlic, cabbage and broccoli. If you insist on popping a pill, do it with orange juice. At least then you're guaranteed some benefit. And if you want to take a shot at this messenger, well, throw tomatoes.