Nine RCTs (n = 432) were included in the review. The studies were generally considered to be good quality with Jadad scores of between 3 and 5 (out of 5 points). Sample sizes ranged from 23 to 78 subjects.
Isoflavones significantly decreased urinary Dpyr (WMD -2.08 nmol/nmol, 95% CI: -3.82, -0.34, nine RCTs). Results of subgroup analyses indicated that the effect was only significant in studies using isoflavone tablets and not in those using isolated soy protein. In subgroup analyses significant reductions were found for studies evaluating isoflavone doses of 90 mg/day or lower (seven studies), but not for doses higher than 90 mg/day (two studies), for studies lasting less than 12 weeks (six studies), but not for longer term studies (three studies), for Asian women (three studies) but not for western women (seven studies) and for postmenopausal women (seven studies), but not for pre-menopausal women (two studies).
Isoflavones significantly increased serum BAP (WMD 1.48 microgram/l, 95% CI: 0.22, 2.75).
There was no strong evidence of publication bias and the failsafe N was calculated to exceed 30.