Antioxidants may protect tissues against damage, reduce systemic inflammation, and modulate antigen presentation by inflammatory cells. However, no large-scale investigation on whether dietary and/or supplement intake of antioxidants is protective against the development of autoimmune rheumatic disease. Here, Costenbader et al (American Journal of Epidemiology 2010; 172(2): 205) explore antioxidant intake and the risk of developing rheumatoid arthritis (RA) and systemic lupus erythematosus (SLE) in women enrolled in the Nurse’s Health Study.
Participants in the Nurses’ Health Study, a prospective cohort study of U.S. female nurses enrolling between 1976 – 2004, had food frequency questionnaires administered every four years that included quantifiable data on multivitamin intake and antioxidant intake from dietary sources. A total antioxidant intake composite score [called the “ferric-reducing ability of plasma” score (FRAP)] was calculated for each participant. The association with antioxidant exposure with incident RA and SLE was examined.
A total of 184,643 women without RA or SLE at baseline were studied. The majority of participants (95%) were Caucasian with an average age of approximately 40 years. Higher intakes of antioxidants were associated with a lower frequency of smoking, higher levels of physical activity, lower intakes of alcohol and caffeine, and higher intakes of calcium and protein. There were 787 incident cases of RA and 192 incident cases of SLE over a mean follow-up time of 11 – 21 years (depending on the enrollment cohort). There was no difference in the incidence of RA or SLE according to level of intake of any individual antioxidants (vitamins A, C, E, α-carotenoid, β-carotene, β-cryptoxanthine, or leutein) either from dietary sources, supplements, or combined, in crude analyses or those adjusted for age at menarche, menopausal status, use of oral contraceptives and hormone replacement, smoking, physical activity, body mass index or race. The incidence of RA or SLE was also not different according to low vs. high levels of the composite FRAP score in crude or adjusted analyses.
Female U.S. nurses with low intakes or antioxidants were not more susceptible to developing RA or SLE than their counterparts with higher intake levels.
This large, carefully analyzed study is very clear in its message of no effect of antioxidant intake on RA and SLE risk. It remains possible that misclassification of the exposures or the outcome could have nullified any association. However, it is unlikely that an interventional study in which antioxidants intake is regulated in patients without RA or SLE who are followed for development of disease will be performed. This is not the first instance of compelling preclinical information about the beneficial effect of antioxidants on immune mediated disease not being validated in clinical investigation, with examples in cancer and heart disease not demonstrating strong protective effects.
Antioxidant supplementation remains popular among the general population solely based on purported health benefits. In this instance, this study is convincing enough to discourage high antioxidant intakes solely for the purpose of reducing the risk of RA or SLE. In particular, high levels of supplementation of some antioxidant vitamins (e.g. E and β-carotene) have even been associated with increased risk of adverse outcomes (e.g. mortality) in some populations.