Clin Endocrinol (Oxf). 2003 Jun; 58(6): 704-9.

The effects of soy protein containing isoflavones on lipids and indices of bone resorption in postmenopausal women.

Dalais FS, Ebeling PR, Kotsopoulos D, McGrath BP, Teede HJ.

Department of Epidemiology and Preventive Medicine, Monash University, The Alfred Hospital, Prahran, Australia.

OBJECTIVE: To assess the effect of a dietary soy protein supplement containing isoflavones on lipids and indices of bone resorption in postmenopausal women. DESIGN: Placebo-controlled, double-blind, randomized study. PATIENTS: One hundred and six postmenopausal women were randomized to dietary soy supplementation (n = 51) or placebo (n = 55) for 3 months, of which 78 were included in the final analysis. MEASUREMENTS: Lipid profiles including total, low-density lipoprotein (LDL) and HDL cholesterol as well as triacylglycerol were measured. Pyridinoline and deoxypyridinoline were used as markers of bone resorption. Urinary isoflavone excretion was measured to assess compliance. RESULTS: There was a significantly greater increase in urinary isoflavone excretion detected in the soy group compared to placebo. Lipid profiles improved with significant decreases in LDL cholesterol (-0.60 +/- 0.10 vs.-0.29 +/- 0.09 mmol/l, P < 0.05), triacylglycerol (-0.22 +/- 0.07 vs. +0.01 +/- 0.05 mmol/l, P < 0.005) and the LDL : HDL ratio (-0.32 +/- 0.10 vs. +0.20 +/- 0.10, P < 0.005) in the soy group compared to placebo. There were no significant differences between the soy and placebo groups for urinary excretion of pyridinoline (-3.8 +/- 3.1 vs.-0.8 +/- 3.1 nmol/mmolCr, P = 0.4) or deoxypyridinoline (-0.8 +/- 0.9 vs.-0.3 +/- 0.7 nmol/mmolCr, P = 0.4). CONCLUSIONS: In postmenopausal women, dietary supplementation with soy protein containing isoflavones does not appear to have oestrogenic effects on markers of bone resorption. Soy protein favourably affected lipids; however, these effects (fall in triacylglycerol and no change in HDL) differ from those observed with oral oestrogen. These findings suggest that soy may not have biologically significant oestrogenic effects on bone resorption and we hypothesize that the lipid effects may be mediated, at least in part, through nonoestrogenic mechanisms.