JAMA Meta-Analysis: Antioxidants Increase Mortality

COPENHAGEN, DenmarkA meta-analysis out of Copenhagen University Hospital found the use of antioxidant dietary supplementsspecifically beta-carotene, vitamin A and vitamin Emay increase mortality, and that the roles of vitamin C and selenium need further study. The surprising findings were released in the Journal of the American Medical Association (297, 8:842-57, 2007).

Goran Bjelakovic, M.D., and colleagues included 68 randomized trials with 232,606 participants that compared beta-carotene, vitamins A, C and E, and selenium, either singly or in combination, vs. placebo or no intervention. Effect of antioxidant supplements on all-cause mortality was analyzed with random effects meta-analysis, and meta-regression was used to assess the effect of covariates across the trials. Trials were further stratified by risk as low-bias (high quality) or high-bias (low quality).

When all trial outcomes were pooled, no significant associations were seen; however, in low-bias trials involving 180,938 participants, antioxidant supplements were associated with a 5 percent increased risk of mortality. Among low-bias trials, use of beta-carotene, vitamin A and vitamin E was associated with 7 percent, 16 percent and 4 percent, respectively, increased risk of mortality.

In their conclusion, the researchers noted a number of limitations of the analysis including inability to determine cause of increased mortality, inclusion of both secondary prevention and primary prevention trials, and wide variability in the nutrients, doses, durations and populations included. They also offered possible explanations for the link between antioxidants and increased mortality: Although oxidative stress has a hypothesized role in the pathogenesis of many chronic diseases, it may be the consequence of pathological conditions. By eliminating free radicals from our organism, we interfere with some essential defense mechanisms. Antioxidant supplements are synthetic and not subjected to the same rigorous toxicity studies as other pharmaceutical agents. Better understanding of the mechanisms and actions of antioxidants in relation to a potential disease is needed.

Industry members and associations sought to put the unexpected findings in context. While meta-analyses can be useful when the included studies are very similar in design and study population, this meta-analysis combined studies that differ vastly from each other in a number of important ways that compromise the results, said Andrew Shao, Ph.D., vice president of scientific and regulatory affairs at the Council for Responsible Nutrition (CRN). He added the combination of secondary and primary prevention trials is an unsound approach, and only by dividing the trials into low- or high-quality trials, using their own criteria, was there a statistically significant effect on mortality. This meta-analysis appears to be a predetermined conclusion in search of a method to support it, Shao said.

Daniel Fabricant, Ph.D., vice president of scientific affairs for the Natural Products Association, agreed with Shaos assessment. In some cases, meta-analysis works and works well, but not in the biased way it was used here, he said. Its the scientific equivalent of using a screw driver to hammer a nail. Looking at people in clinical trials to draw mortality conclusions is a fundamental mistake because most of them are not healthy to begin with, and thus already at greater risk of mortality. To call into question vitamins that have a well-documented history of providing health benefits is irresponsible and may cause those individuals most in need of supplementation to avoid it.