Meta-analysis Of The Effects Of Isoflavones On The Lipid Profiles In Peri-and Postmenopausal Women

Backgrounds A large-scale epidemiological study showed that there were gender differences in the incidence rate of cardiovascular diseases.The incidence of coronary heart disease(CHD)in premenopausal women was significantly lower than that in man.However,the incidence rates of the peri-and postmenopausal women increase rapidly,because of the changes of sex hormones' level in postmenopausal women,especially the decrease in endogenous estrogen levels.As the main active ingredient in the soy foods and mimic the chemical structure of endogenous estrogen,isoflavones can bind to estrogen receptor.Studies have showed isoflavones may reduce the risk of breast cancer.And so far,there have been no report that referred to the increasing of the probability of serious side effect.Objective This meta-analysis was primarily conducted to evaluate the main effects ofisoflavones on lipid profiles of peri-and postmenopausal women.Isoflavone may serve as a natural SERM for postmenopausal hormone replacement therapy to improve women's life quality and life expectancy.Methods PUBMED,CNKI and Cochrane Library were searched for randomized controlled trials published until May,2016 that described the effects of soy protein intake in humans.Nineteen studies were selected for the meta-analysis.Data were obtained from the nineteen studies which included the study characteristics(first author's name;year of publication;study design;number of participants in each group;daily amount of isoflavones or soy protein in active arm;and duration)and were tabulated for analysis.Quality of the studies was evaluated using the randomized controlled trial(RCT)quality evaluation standard which was recommended by Cochrane Handbook,which highlights the significance of random sequence generation,allocation sequence concealment,double-blinded experiment design,the completeness of results and bias minimization.The principal effect we used in this investigation was the mean difference between the change in lipid concentrations among the subjects consuming the isoflavones or soy protein(final value minus initial value)and that among the subjects on the control diet.We performed this analysis with the application of REVMAN for WINDOWS softtware(version 5.3;update on June 13,2014;Cochrane Collaboration,Oxford,United Kingdom).Estimates of the weighted mean differences(WMDs)and 95%confidence interval(CI)were obtained using a random-effect model or a fixed-effect model.Heterogeneity among studies was assessed by the Cochran's Q test and 12 statistic to evaluate the magnitude of heterogeneity.In addition,we applied Egger's test to detect potential publication bias.The trim and fill method is also used in an attempt to eliminate bias.Results Isoflavones were associated with significant decreases in serum total cholesterol,-0.1(95%CI-0.12,-0.07)mmol/L,triacylglycerols,-0.08(95%CI-0.1,-0.05)mmol/L,and LDL cholesterol,-0.14(95%CI-0.16,-0.11)mmol/L,and significant increases in serum HDL cholesterol,0.04(95%CI 0.03,0.04)mmol/L.There was no significant difference in incidence rate of adverse events between the isoflavone group and the control group.When the ingestion of isoflavones is greater than 80mg/d,TC,TG and LDL cholesterol decreased significantly and HDL cholesterol increased significantly,with TC-0.10 mmol/L(95%CI-0.13,-0.07),TG-0.09 mmol/L(95%CI-0.11,-0.06),LDL-C-0.15 mmol/L(95%CI-0.18,-0.13),HDL 0.04 mmol/L(95%CI 0.03,0.05).When the intervention duration was longer than 3 month,TC,TG and LDL cholesterol decreased significantly and HDL cholesterol increased significantly,with TC-0.10mmol/L(95%CI-0.12;-0.07),TG-0.09 mmol/L(95%CI-0.11,-0.06),LDL-C-0.14 mmol/L(95%CI-0.16,-0.12),HDL 0.04 mmol/L(95%CI 0.03,0.05).The side effects results showed that there was no significant difference in incidence of adverse events between the isoflavone group and the control group.The result of TC,TG and HDL cholesterol did not show publication bias,suggesting that combining these studies for a meta-analysis was valid.Trim and fill method was used to correct the result of LDL cholesterol.The result of LDL cholesterol was valid.Conclusions The meta-analysis showed that the intake of isoflavones was associated with a significant decrease in TC,LDL-C,and TG concentrations as well as a significant increase in serum HDL-C.As the heterogeneity was not significant,the results were stable.Further analysis showed that the lipid level decreased significantly when the duration was more than three months and the dose was more than 80mg/d.The analysis of adverse reactions showed that the adverse reactions of isoflavones were less and mild,mainly manifested as nausea and abdominal distension,and there was no significant difference in the incidence of adverse events during treatment with the control group.As natural SERMs,treatment of isoflavones with optimal dosage and duration might be a safe alternative for hormone replacement therapy in peri-and postmenopausal women.