Effect of isoflavones on serum homocysteine and c-reactive protein levels in postmenopausal women

Abstract

Cardiovascular diseases (CVDs) are the leading cause of mortality and morbidity among the postmenopausal women in developed as well as developing countries. The incidence of CVD is lower in premenopausal women than in men; however, CVD risk in postmenopausal women is 3.4 times than that in premenopausal women. These differences in risk are partially related to increases in homocysteine (Hcy) and C-reactive protein (CRP). Elevated Hcy is an independent, modifiable risk factor for atherosclerotic CVD. Elevated CRP, a marker of acute inflammation, is a reliable predictor of CVD. Soy milk is rich in phytoestrogen, it has been claimed that phytoestrogens, particularly soy isoflavones, have beneficial effects on cardiovascular diseases (CVDs) and diabetes; however, the underlying mechanism for these effects still need to be established. In the present study the effect of soy milk on homocysteine (Hcy, an important risk for CVDs), C-reactive protein (CRP, one of the marker of inflammatory status) and insulin resistance (a major defect in diabetes & CVDs) were explored in response to short-term intake of soy milk. Soy beans, cultivated in Bangladesh, were collected directly from the producers, and milk was prepared by following a standardized manual procedure. It was an un-blinded open ended trial where 36 women, under a randomized cross-over design, consumed 350ml of the milk 2 times/day (calculated with a target 30 mg isoflavone daily) for 21 days with a wash out period of 10 days. The subjects were advised to follow the usual lifestyle during the period. The Control Group did not receive any other placebo product. Fasting (10-12 h) serum samples were analyzed for glucose & lipids (enzymatic-colorimetric method), insulin and CRP (ELISA method) and Hcy (Fluorescence Polarization Immunoassay). Pooled data for the two groups were used to calculate the baseline and final values which were then compared by paired t-test. Insulin sensitivity (HOMA%S) was calculated by homeostatic model assessment. Among the lipidemic profile serum HDL (mg/dl) was significantly higher (p=0.005) at the time point of 21st days compared to 0 day in Soy milk group. Serum triglycerides (mg/dl) and NEFA (mmol/l) were significantly lower (p=0.088 and p=0.010 respectively) in Soy milk group at 21st days compared to 0 day. However the insulinemic profile (fasting insulin, HOMA%B, HOMA%S and HOMA-IR) and inflammatory status (CRP and Hcy) did not show any significant difference between the two time points. When considering the time point of 31st and 52nd days (after 10 days wash out period) serum glucose (mmol/l) was significantly lower (p=0.005) in Soy milk group at 52nd days compared to 31st days. Serum triglyceride was significantly lower (p=0.049) in both groups at 52nd days compared to 31st days. Among inflammatory status and insulinemic profile there were no significant difference between the two time points (p=ns). On the other hand, at the time point of baseline and endpoint serum blood sugar, triglycerides and NEFA were significantly lower (p=0.009; p=0.008 and p=0.013) in endpoint compared to the baseline values. On Pearson’s correlation analysis CRP showed significant positive correlation with BMI (r=0.416, p=0.012), WHR (r=0.330, p=0.049) and fasting insulin (r=0.286/p=0.015) at baseline, while at the endpoint CRP showed significant positive correlation with BMI (r=0.538, p=0.001) and WHR (r=0.343, p=0.041). On the other hand, Hcy showed significant positive correlation with age (r=0.451, p=0.006) and NEFA (r=0.407/p=<0.001) at baseline while at endpoint it showed correlation only with age (r=0.508, p=0.002). Logistic regression analysis of CRP was done with age, BMI, WHR, lipid profile and HOMA%B confounding independent variables. A negative significant association was found with HOMA%B (=-0.352 95% CI: -0.021-<0.002, p=0.043) at baseline. At end point CRP showed significant negative association with HOMA%B (=-0.360 95% CI: -0.017-<0.001, p=0.041) after adjusting the effects of major confounders (age, BMI, W/P ratio and lipid profile). End point group Hcys showed significant negative association with HOMA%S (=-0.521 95% CI: -0.157-(-0.019), p=0.041) and HOMA IR (=-0.382 95% CI: -1.111-(-0.001), p=0.050) after adjusting the effects of major confounders (age, BMI, W/P ratio), these associations disappeared after adjusting the same confounders at baseline. From the above results it may concluded that Soy Isoflavones seem to have beneficial effects on hyperglycemia and dyslipidemia among postmenopausal women and the effects seems to be associated through lowering of insulin resistance and chronic subclinical inflammation.