Objective: To clarify the effect of soy isoflavones on prevention of osteoporosis, and the effective dosage of soy isoflavones and its duration.

Methods: Random control trials that investigated the association of soy isoflavones and osteoporosis were included in the meta-analysis by researching MEDLINE, EMBASE and the Chinese Biomedical Database up to October 2011. The Rev Man software was used for all of the statistical analysis.

Results: The present meta-analysis found that soy isoflavones significantly increased the bone mineral density by 54% and decreased the bone resorption marker urinary deoxypyridinoline (DPD) by 23% compared to baseline in women. Using random effects model, the effect of isoflavones on bone mineral density (BMD) regarding menopausal status and isoflavone dose revealed higher weighted mean difference changes were found in postmenopausal women and isoflavone dose above 75 mg/d. Subgroup analysis of trials with menopausal status, supplement type, isoflavone dose and intervention duration that used soy isoflavone extracts resulted in significant different overall effect of DPD using by random effects model. Sensitivity analysis indicated that the effect of soy isoflavones on BMD and DPD was robust.

Conclusions: The present meta-analysis reveals that soy isoflavone supplements significantly increase bone mineral density and decrease the bone resorption marker urinary DPD. It shows no significant effect on bone formation markers serum bone alkaline phosphatase. The significant effect of soy isoflavones on BMD and urinary DPD is relative to menopausal status, supplement type, isoflavone dose and intervention duration.