We congratulate Sanchez-Guerrero and colleagues [1] for investigating postmenopausal estrogens and the risk for systemic lupus erythematosus [1]. Given the preponderance of the disease in women, adverse effects of the female gender and sex hormones in murine lupus, and numerous retrospective reports (often anecdotal and uncontrolled) describing a temporal association between estrogen exposure and the development or exacerbation of the disease, it is easy to accept that estrogens and systemic lupus erythematosus simply do not mix. Accordingly, the finding of an increased relative risk for the disease in a "naive" cohort exposed to postmenopausal estrogens is intuitive. Because Sanchez-Guerrero and colleagues' study was an omnibus prospective observational one, not a randomized trial designed to test the association of postmenopausal estrogens and the risk for systemic lupus erythematosus, confounders and biases are possible. The disease may have been overascertained in women receiving estrogens: These women had more than the "at least one per year" physician visit that characterized the unexposed cohort; in addition, when these women sought medical attention, their symptoms and estrogen use may have led to a blood test for antinuclear antibodies. Alternatively, those who developed unexplained mood swings, energy loss, low-grade fevers, facial flushing, and musculoskeletal symptoms may have received a diagnosis of the "postmenopausal syndrome" and been given estrogen rather than a diagnosis of late-onset systemic lupus erythematosus. Why did the disease not become manifest during pregnancy (presuming no patients were nulliparous), when estrogen levels are 100-fold greater than those achieved with postmenopausal estrogens?

Several salutary effects of postmenopausal estrogens assume importance in systemic lupus erythematosus [2]. The risks for osteoporosis, exaggerated by menopause (natural or cyclophosphamide-induced) and glucocorticoids, are major concerns. Moreover, hormonal replacement is associated with a 40% reduction in the risk for coronary artery disease [2], higher high-density lipoprotein cholesterol levels, and decreased low-density lipoprotein cholesterol levels [3]. These benefits are relevant in systemic lupus erythematosus, given the bimodal distribution of mortality with late deaths caused by atherosclerotic disorders [4]. Although Sanchez-Guerrero and colleagues did not evaluate women with systemic lupus erythematosus, their findings may add to the assumption that postmenopausal estrogens increase flare rates. However, in a controlled retrospective study of postmenopausal estrogens in systemic lupus erythematosus, disease exacerbations did not increase in the exposed cohort compared with the unexposed cohort [5].

Given the health needs of women with established systemic lupus erythematosus, large prospective studies on the safety of postmenopausal estrogens are critical.