Introduction 1 5 2 3 Methods Searching strategy 4 Selection of eligible trials We included both single and double blind randomized controlled trials with baseline and after treatment values for synthesizing risk (mean) differences. The outcome measures were differences of serum total cholesterol (TC), low density lipoprotein cholesterol (LDL-C), high density lipoprotein cholesterol (HDL-C) and triglycerides (TG) between post-randomization baselines and after treatments. Eligible interventions were isoflavones tablets of single isolated component or any mixtures of genisteins, daidzein, formononectin and biochanin A. Isoflavones interventions in forms of soy protein supplements or diets were also included as long as they compared isoflavones containing treatments with isoflavones depleting controls. Treatments with soy polysaccharides, fiber, and phytosterols as their primary interventions were excluded. Validity assessment and data abstraction Two independent investigators reviewed the articles obtained without masking. There were no scorings to the included trials. Data were entered twice to reduce input errors. Inter-rater reliability was not performed. Data were abstracted with a designed form before analysis. Duplicated trials or studies with the same population were counted once to reduce the duplicated publication bias. Data disagreement between the two reviewers was resolved by discussion. Study characteristics The characteristics of the obtained studies were tabulated with subtypes of isoflavones interventions, subjects' serum lipids status, dosages of isoflavones and lengths of treatment. Sub-group analysis was performed with different forms of isoflavones interventions, such as isolated genistein or mixtures of isoflavones tablets versus placebo, isoflavones containing versus depleting soy protein diets. Sensitivity analysis was also made across different population characteristics and lengths of treatment. Funnel plots were used to detect possible publication bias or treatment heterogeneity across sample sizes. Quantitative data synthesis We obtained the risk differences (RD) from the post-randomization baselines and after-treatment values in each trial and calculated the pooled standard deviation of the RD as:  1  1  2  2  6 31 Results Trial flow 1 7 23 24 27 28 29 Figure 1 Trial flow chart of including isoflavones studies.  7 23  Study characteristics 1 Table 1 Characteristics of the 17 included studies. There were 9 crossover and 8 parallel studies. The isoflavones tablets introduced were mainly in aglycone form. The abbreviations were listed below table. Study Treatment (Rx) † Isoflavones content in Rx (mg/d) ‡ Control Isoflavones content in control (mg/d) No. of subjects Lipid status § Subject gender || Length of treatment (weeks) 7 soy biscuits NA soy biscuits 0 10 H male 4 8 T 57 T      8 T 85.5  28.5 46 B female 26 9 T 150 G = 40; D = 50; glycosides = 60 placebo 0 36 H female 24 10 T 80 G = 40; D = 40 placebo 0 29 NS post 2 11 soy protein capsule 100 G = 70; D = 18; glycitein = 12 soy protein capsule 0 78 N peri 16 12 T 55 G = 30; D = 1; B = 16; F = 8 placebo 0 59 N male & female 8 13 T 40 G = 1; D = 0.5; B = 26; F = 16 placebo 0 75 NS post 5 14 ISP 65 ISP < 4 49 H post 12 15 ISP 113–144 ISP      15 ISP 55–74  9–11 13 N female 3 16 T 40–80 G= 22–43; D = 17–33; glycitein= 1–3 placebo 0 21 B female 5 17 T 80 G = 8; D = 7; B = 49; F = 16 placebo      17 T 40 G = 4; D = 3.5; B = 24.5; F = 8  0 13 B female 5 18 T 86 G = 8.6; D = 7.4; B = 51.4; F = 18.6 placebo 0 14 N pre 16 19 soy burger 56 G = 34.8; D = 21.2 soy burger 2 22 N male & female 2 20 T 80 placebo 0 20 N post 8 21 T 54 G= 54 placebo 0 60 N female 24 22 T 70 G = 42; D = 27 ISP 3 28 N elderly men 6 23 ISP 110–154 ISP      23 ISP 54–76  6–8.2 18 NS post 4 * crossover design † 'T' = isoflavone tablet; 'ISP' = isolated soy protein ‡ 'G' = Genistein; 'D' = Daidzein; 'B' = Biochanin A; 'F' = Formononectin §'H' = hyperlipidemic subjects, 'N' = normolipidemic subjects; 'B' = both hyper- and normolipidemic subjects; 'NS' = not specified || 'pre' = premenopausal women; 'peri' = perimenopausal women; 'post' = postmenopausal women Quantitative data synthesis 2 2 11 9 10 16 19 22 22 3 Figure 2 Funnel plot of risk difference of isoflavones tablets versus placebo on serum total cholesterol.   Table 2 Subgroup analysis of the effects of isoflavones on serum total cholesterol. Isoflavones were shown to have significant benefits when given in the form of ISP or soy protein capsule. Regardless the forms of prescription, isoflavones decreased serum total cholesterol among normolipidemic but not hyperlipidemic subjects. Subgroup outcome No. of comparisons No. of subjects || Treatment effect on TC (mmol/L) ¶ Heterogeneity p-value § Form of intervention †:      a.) isoflavones tablets 8 10 12 13 16 18 20 21 482 0.01 (-0.17, 0.18) 1.00  b.) ISP(+) 14 15 22 23 229 -0.11 (-0.21, -0.01)* 0.34  c.) soy protein capsule 11 78 -0.69 (-1.19, -0.19)* NA  d.) soy biscuit 7 20 -0.05 (-0.47, 0.37) NA  e.) soy burger 19 44 0.09 (-0.38, 0.56) NA  f.) Overall 7 23 853 -0.09 (-0.18, -0.01)* 0.77 Isoflavones mixture ‡:      G&D&B&F 12 13 17 18 214 -0.06 (-0.34, 0.21) 0.96  G&D 9 11 16 19 22 285 -0.22 (-0.41, -0.03)* 0.06  G 21 60 0.10 (-0.35, 0.55) NA Isoflavones intake (mg per day):      < 50 8 13 17 132 -0.12 (-0.54, 0.29) 0.99  51–100 8 10 12 14 15 17 23 484 -0.10 (-0.21,0.01) 0.74  101 – 150 9 11 15 23 176 -0.10 (-0.24, 0.05) 0.09 Design :      parallel crossover 8 14 21 447 -0.08 (-0.27, 0.10) 0.48   a.) no washout 22 56 -0.43 (-0.73, -0.13)* NA   b.) washout mentioned 15 19 20 23 208 -0.07 (-0.17, 0.03) 0.91 Gender:      male 7 22 76 -0.30 (-0.54, -0.05)* 0.15  female 8 11 13 18 20 21 23 674 -0.07 (-0.17, 0.02) 0.89 Treatment length:      2–10 weeks 7 10 12 13 15 17 19 20 22 23 541 -0.09 (-0.18, 0.00) 0.84  11–20 weeks 11 14 18 155 -0.22 (-0.52, 0.07) 0.07  21–30 weeks 8 9 21 157 0.02 (-0.28, 0.31) 0.96 Menopausal status:      pre-menopausal 15 18 80 -0.06 (-0.17, 0.06) 0.58  peri-menopausal 11 78 -0.69 (-1.19, -0.19)* NA  post-menopausal 8 10 13 14 16 17 20 21 23 516 -0.05 (-0.20, 0.10) 1.00 Subjects with:      normolipidemia 11 12 15 18 19 21 22 377 -0.11 (-0.21, -0.01)* 0.08  hyperlipidemia 7 9 14 105 -0.05 (-0.36, 0.26) 1.00 * statistically significant † 'ISP (+)' = isoflavone containing isolated soy protein; 'ISP (-)' = isoflavone depleting isolated soy protein ‡ 'G' = Genistein; 'D' = Daidzein; 'B' = Biochanin A; 'F' = Formononectin § NA = not applicable || Since subjects acted as their own controls in a crossover trial, the 'calculated' total number was thus doubled in a single pair comparison. ¶ 95% confidence interval in parenthesis Table 3 Effects of isoflavones on serum LDL-cholesterol, HDL-cholesterol and triglycerides levels. Isoflavones in the forms of tablets or isolated soy protein (ISP) did not show significant benefits over serum LDL-cholesterol, HDL-cholesterol and triglycerides levels. The results were not heterogeneous to combine. Type of intervention* No. of trials No. of comparisons No. of subjects Treatment effect  Heterogeneity p-value 1.) Isoflavones tablets       LDL 9 8 10 12 13 16 18 20 21 446 0.00 (-0.14, 0.15) 0.94  HDL 10 8 10 12 13 16 18 20 21 482 0.01 (-0.05, 0.06) 0.98  TG 9 8 10 12 13 16 17 20 21 482 0.03 (-0.06, 0.12) 0.93 2.) ISP (+)       LDL 3 14 15 23 173 -0.06 (-0.16, 0.03) 0.84  HDL 3 14 15 23 173 -0.01 (-0.07, 0.05) 0.97  TG 3 14 15 23 173 0.02 (-0.05, 0.09) 1.00 * 'ISP (+)' = isoflavones containing isolated soy protein; 'LDL' = Low density lipoprotein cholesterol; 'HDL' = High density lipoprotein cholesterol; 'TG' = Triglycerides † 95% confidence interval in parenthesis Discussion Main findings 2 9 10 16 19 22 2 11 3 12 18 21 11 22 32 33 35 Validity and limitation 36 Biological plausibility 37 39 40 42 30 43 44 45 1 46 Suggestion to future trials Tablets of single isolated isoflavone with considerable dosages may be important in future randomized trials. Intervention with isoflavones mixtures may be helpful but interactions between components should be handled with extra cares. Introduction of phytoestrogens in forms of soy diets is not suggested because we found much heterogeneity inside the diets, ranging from soy biscuits to burger, which may be difficult to combine and analyze. Confounding factors such as fiber, fatty acids, amino acids and energy intake are especially hard to control. Complications associated with hyperlipidemia, such as coronary heart disease (CHD) or cardiovascular accident (CVA), may be selected as other endpoints in future trials. It may help constructing convincing funnel plots with less heterogeneity as trials accumulate. Conclusions Isoflavones tablets, up to 150 mg per day, had insignificant effects in lowering serum total cholesterol, LDL-cholesterol and triglyceride. There was also insignificant benefit over serum HDL-cholesterol. The results were consistent when tablets were introduced as isolated genistein, mixture of genistein and daidzein, or mixture of genistein, daidzein, formononectin and biochanin A. No significant effects were found among participants with normo- or hyperlipidemia and women with pre- or postmenopausal status. Isoflavones interventions in the forms of soy proteins, such as isolated soy protein (ISP), soy diets or soy protein capsule, were inconclusive due to inadequate sample size, heterogeneity and presence of potentially uncontrolled confounders. List of abbreviations 'ISP' = isolated soy protein; 'TC' = total cholesterol; 'LDL' = Low density lipoprotein cholesterol; 'HDL' = High density lipoprotein cholesterol; 'TG' = Triglycerides Competing interests None declared. Authors' contributions JY participated in the design of this manuscript. JY and YTF participated in abstracted the data and performed statistical analysis. All authors read and approved the final manuscript.