Effect of long-term intervention of soy isoflavones on bone mineral density in women: a meta-analysis of randomized controlled trials.

INTRODUCTION: A number of RCTs have examined the role of soy isoflavones on bone mineral density (BMD) and yielded inconsistent results. This meta-analyses aims to assess the overall effect of soy isoflavones on BMD.

METHODS: We searched for all articles published in English from January 1990 to March, 2008. We included RCTs of soy isoflavone supplementation in women of at least one year duration. The main outcomes were BMD changes from baseline at the lumbar spine, total hip and femoral neck.

RESULTS: We identified 10 eligible RCTs containing 896 women. A mean dose of 87 mg soy isoflavones for at least one year did not significantly affect BMD changes. The mean (95%CI) differences in BMD changes (in mg/cm(2)/year) were 4.1 (-1.6, 9.8) (0.4%) at the lumbar spine, -1.5 (-7.2, 4.3) (-0.3%) at the femoral neck under random-effects model, and 2.5 (-0.5, 5.4) (0.2%) at the total hip by fix-effects model, respectively. Similar results were obtained in subgroup analyses by isoflavone sources (soy protein vs. isoflavone extract), ethnic differences (Asian vs. Western). Larger dose (>or=80 mg/d), but not lower dose (<80 mg/d), of isoflavone intervention tended to have a weak beneficial effect on spine BMD (p=0.08 vs. p=0.94).

CONCLUSIONS: Soy isoflavone supplementation is unlikely to have significant favorable on BMD at the lumbar spine and hip in women.