Scientific Name(s): Glycine max (L.) Merr.
Common Name(s): Soy, Soy isoflavones, Soya, Soybean
Medically reviewed by Drugs.com. Last updated on Nov 22, 2022.
Soy is commonly used as a source of fiber, protein, and minerals. Several meta-analyses evaluating soy use for various indications are available; however, evidence is lacking to support use in the treatment of asthma, menopausal symptoms, obesity, osteoporosis, diabetes, or heart disease. Limited benefit has been demonstrated for irritable bowel disease, polycystic ovary syndrome, and chronic kidney disease. Negative effects of soy consumption have also been noted.
The Italian Society of Diabetology (ISD) and the Italian Society for the Study of Arteriosclerosis (ISSA) state that soy supplementation for cholesterol-lowering purposes may be advised, though with some level of uncertainty and based on low-level evidence, in the general population and in patients with mild hypercholesterolemia who have low to moderate cardiovascular risk.
Dosages of various forms of soy in clinical studies evaluating various uses have included 22.7 to 300 mg/day of soy isoflavones, up to 40 g/day of isolated soy protein, 120 g/day of dietary soy foods, 50 to 150 g/day of unfermented soy foods, up to 450 mg/day of genistein, up to 300 mg/day of daidzein, or 70 g/day of whole soy nuts.
Contraindications have not been identified. Women with current or a history of estrogen-dependent tumors, including breast cancer, should consult their physician prior to consuming soy in amounts higher than those typically found in food. An increased risk of some cancers has been documented in men consuming soy.
Soy has "generally recognized as safe" (GRAS) status when used as food. Avoid dosages above those found in food; safety and efficacy have not been established. Soy isoflavones are passed into breast milk. Because of nutritional disadvantages of soy-based infant formulas, the European Society of Pediatric Gastroenterology, Hepatology, and Nutrition (ESPGHAN) and the American Academy of Pediatrics (AAP) recommend amino acid–based and extensively hydrolyzed infant formulas over infant soy formula in non-breastfed infants with cow's milk protein allergy; a soy formula may be considered in infants on a vegan diet.
None well documented; use caution when administering with estrogen derivatives.
Soybeans and their products are generally well tolerated. Minor GI disturbances have been reported. Cross-sensitivity between poly-gamma-glutamic acid (PGA) of fermented soybeans and PGA from jellyfish has been documented in several case reports. Soybeans contain natural goitrogens and can cause hypothyroidism in susceptible patients.
- Fabaceae (bean)
Soybean is an annual plant that grows 0.3 to 1.5 m in height. The bean pods, stems, and leaves are covered with short, fine hairs, and the pods contain up to 4 oval, yellow to brown seeds. Cotyledons ("seed leaf") account for most of the seed's weight and contain nearly all the oil and protein. Legumes such as soy are able to fix free nitrogen from the air into a useable form for growth via the bacterium Rhizobium japonicum, which is associated with the roots.Ensminger 1994, USDA 2019
In 2838 BC, Chinese emperor Shung Nang described soybeans as China's most important crop. The plant was introduced to Japan, Europe, and eventually to the United States by the early 1800s. The United States now produces 49% of the world's soybeans. Soy foods have become increasingly popular among health-conscious individuals since the early 1990s. In 2000, approximately 27% of US consumers reported using soy products at least once a week, nearly double the 1998 figure. As a food source, soy has been used in Asian cultures for thousands of years, with Asian populations consuming 60 to 90 g/day of soy, compared with Western diets that contain approximately one-tenth of that amount. Soybean products are numerous and include milk, flour, curd, sufu, tofu, tempeh (Indonesian ingredient), miso (fermented soybean paste), sprouts, soy sauce, soybean oil, textured soy proteins (in meat extenders), soy protein drinks, and livestock feeds. Because of its low cost, good nutritional value, and versatility, soy protein is used as part of food programs in less developed countries.Craig 1997, Ensminger 1994, Messina 2001, Polunin 1997
Soybeans are high in nutritional value and contain up to 35% oil, 24% carbohydrates, and 50% protein.Ensminger 1994 Isolation of proteins and analytical protein determination methods are often used to characterize soybean proteins.Garcia 1997 Fatty acids in beans include linoleic (55%), palmitic (9%), and stearic (6%) acids. Soybeans are rich in minerals and trace elements, including calcium, iron, potassium, amino acids, and vitamins, and are also a good fiber source.Ensminger 1994, Polunin 1997 Soybeans contain isoflavone compounds known as phytoestrogens that are structurally similar to human estrogen and progesterone. The plant's isoflavones include genistein and daidzein, which are the most abundant, as well as glycitein and equol,Messina 2001 with soy protein preparations varying widely depending on the processing technique.Garcia 1997 Isoflavones remain in soy preparations that are not extracted with alcohol. The dehulling, flaking, and defatting of soybeans produces a relatively pure preparation that is low in isoflavones. Isoflavone concentrations range from approximately 2 mg/g of protein in textured soy protein, soy flour, and soy granules to 0.6 to 1 mg/g of protein in isolated soy protein.Sacks 2006
Uses and Pharmacology
Isoflavones, the phytoestrogens in soybean, have weak functional effects similar to those of the female hormone estradiol, including hormonal and nonhormonal actions.(Messina 2001) Hydrolysis of isoflavone glycosides by intestinal glucosidases yields the active intermediates genistein, daidzein, and glycitein, which undergo further metabolism by the gut microbiome to equol and p-ethyl phenol. This metabolism is highly variable among individuals and may depend on factors such as effects of carbohydrate intake on intestinal fermentation, as well as race or geographic dietary influences; 35% of Westerners and up to 60% of Asian populations produce equol. Isoflavones are secreted into bile via enterohepatic circulation. Plasma half-life of genistein and daidzein is approximately 8 hours, with peak concentration achieved in 6 to 8 hours in adults. Elimination is renal, primarily as glucuronide conjugates.(Barnes 1998, Reverri 2015) Absorption and bioavailability of nutrients, including calcium, from soy-based formulas and beverages have been found to be much lower than with nonsoy comparators.(Koletzko 2012)
In a post hoc analysis of a prospective multicenter trial of adult patients with inadequately controlled asthma (N=300), patients submitted dietary questionnaires that were used to evaluate soy genistein intake. Patients underwent baseline pulmonary function evaluation and were followed for a 6-month period to observe the incidence of episodes of poor asthma control. Subgroups included minimal, moderate, or high intake of soy genistein. Patients with minimal or no genistein intake had a significantly lower baseline forced expiratory volume in the first second of expiration (FEV1) compared with the moderate- and high-intake groups (2.26 L vs 2.53 L and 2.47 L, respectively; P=0.01). The incidence of episodes of poor asthma control during follow-up was also significantly higher in the minimal soy intake group compared with the moderate- and high-intake groups (54% vs 35% vs 40%, respectively; P<0.001). Between-group comparisons were not significantly different for the moderate-intake group compared with the high-intake group for these outcomes.(Bime 2012) In contrast, in a randomized, placebo-controlled, double-blind, multicenter trial (N=386), the use of soy isoflavone (isoflavones 100 mg/day for 24 weeks) as add-on therapy (ie, to daily inhaled corticosteroids and/or a leukotriene modifier) did not improve lung function (FEV1) or clinical outcomes in patients older than 12 years (median, 36 years) with poorly controlled asthma.(Smith 2015)
Isoflavones are selective estrogen receptor modulators, but they also possess nonhormonal properties. The weak estrogenic action of soy isoflavones and other phytoestrogens suggests a potential ability to lessen the deleterious effects of more potent endogenous estrogens on breast and endometrial cancer.
In 1990 and 2005, following reports of decreased chemically induced rat mammary cancer after the addition of soy protein to a typical diet, the National Cancer Institute held workshops and recommended that the impact of isoflavones on breast tissue be evaluated at the cellular level in high-risk women.(Messina 2006)
Available literature includes reviews of cohort and case-control studies evaluating effects of soy intake on the risk of breast cancer incidence, as well as meta-analyses of prospective studies evaluating effects of soy on the risk of breast cancer and its recurrence.(Dong 2011, Enderlin 2009, Magee 2012, Wu 2016) Overall, the data are not persuasive that adult consumption of soy affects the risk of developing breast cancer or that soy consumption affects the survival of breast cancer patients. According to one meta-analysis of 14 prospective studies, summary relative risk (RR) for the association of soy isoflavone consumption and incidence of breast cancer was 0.89 (95% CI, 0.79 to 0.99); however, when the data were evaluated based on ethnicity, a protective effect was found only for Asian populations.(Dong 2011) When dietary sources of protein were investigated in a dose-response meta-analysis, consumption of soy food was linearly associated with an overall decreased risk of breast cancer (by 9%), with a summary RR of 0.91 (95% CI, 0.84 to 1); summary RR for the highest versus lowest consumption was 0.92 (95% CI, 0.84 to 1; I2=0%). However, an appropriate method for calculating the amount of protein in soy food was not available.(Wu 2016) Data for risk of breast cancer recurrence from 4 studies yielded similar results (summary RR, 0.84; 95% CI, 0.7 to 0.99). Another modifier may be menopausal status because no association was evident in premenopausal women. No dose-response relationship was revealed.(Dong 2011) In another meta-analysis, soy did not appear to interfere with tamoxifen and anastrozole therapy in breast cancer patients.(Magee 2012)
Data regarding an increased breast cancer risk with use of soy isoflavones exist; it is possible that isoflavones in soy stimulate breast tumor growth through their estrogenic activity.(Bolca 2010, de Lemos 2001, Enderlin 2009) In a meta-analysis of 8 clinical trials of isoflavone supplementation, there was a modest increase in breast tissue density among premenopausal women but not among postmenopausal women. The clinical importance of this finding is unclear.(Hooper 2010) In a 12-month, randomized, controlled study, no difference in breast density was observed in postmenopausal women who received a daily supplement containing 60 mg of soy isoflavones plus Lactobacillus sporogenes compared with the control group.(Colacurci 2013) Similar results were reported in another 12-month, randomized, placebo-controlled trial in which 85 female breast cancer patients received soy isoflavones 50 mg/day. No significant differences were found between groups in breast density, as measured by either digital mammograms or magnetic resonance imaging. History of breast cancer, menopausal status, age, race, body mass index (BMI), baseline breast density, and 12-month blood isoflavone levels did not affect the results. However, differences in breast density based on BMI were statistically significant (P<0.0001).(Wu 2015)
In a 2014 meta-analysis of epidemiological studies reporting associations between soy and breast cancer, subanalyses of data were conducted based on menopausal status (pre- or postmenopausal), region of study (Asian or Western), soy intake (soy isoflavone, soy protein, soy food), and study design (cohort, nested case control, case control). Statistically significant inverse associations between soy consumption and risk of breast cancer were identified in premenopausal Asian women but not premenopausal women from Western nations, with soy bean/soy products showing a stronger association, and in postmenopausal women with the highest soy isoflavone intakes, with a stronger association in Asian women than in Western women.(Chen 2014)
The Society for Integrative Oncology's updated guideline on the evidence-based use of integrative therapies during and after breast cancer treatment (2017) recommends against the use of soy for hot flashes in patients with breast cancer due to lack of effect (grade D).(Greenlee 2017)
Soy isoflavones have estrogenic, antiandrogenic, and other activities that could theoretically prevent prostate cancer or slow its progression.(Messina 2003) Prostate cancer incidence appears to decrease with increased isoflavone intake.(Sirtori 2001)
Animal and in vitro data
Rats fed soy protein diets showed a reduced incidence of prostate tumors compared with rats fed casein. Tumor latency was increased only in the rats fed a diet containing isoflavone-rich, isolated soy protein.(Sirtori 2001) In prostate cancer cells, genistein reduced the synthesis of prostate-specific antigen (PSA), a marker of prostate cancer development and progression.(Sarkar 2003) In vitro, genistein inhibited the growth of androgen-dependent and androgen-independent prostate cancer cells in a dose-dependent manner.(Yan 2009)
Meta-analyses of randomized controlled trials and observational studies of soy consumption and risk of prostate cancer have been published.(Hwang 2009, van Die 2014, Yan 2009) In a meta-analysis of 8 randomized controlled trials evaluating supplemental or dietary soy/soy isoflavone consumption (preparations and doses varied) in men with prostate cancer or men with a clinically identified risk of developing prostate cancer, a statistically significant reduction in prostate cancer diagnosis (risk ratio, 0.49; 95% CI, 0.26 to 0.95) was observed in men with risk of developing prostate cancer; however, no significant differences between groups were observed for PSA level or sex steroid end points (sex hormone–binding globulin, testosterone, free testosterone, dihydrotestosterone, or estradiol). Additionally, no significant association was found between equol production and PSA levels.(van Die 2014) Other studies have also reported an inverse association between soy consumption and risk of prostate cancer (combined RR/odds ratio [OR], 0.7; 95% CI, 0.63 to 0.89).(Yan 2009) Data suggest that benefits may be limited to nonfermented soy products and Asian populations.(Hwang 2009, Yan 2009) These earlier data are supported by a 2018 systematic review and broad-scope meta-analysis (30 studies and 266,699 participants) that identified statistically significant inverse associations between risk of prostate cancer and dietary total soy intake (RR, 0.71; 95% CI, 0.58 to 0.85; P<0.001), unfermented soy food intake (RR, 0.65; 95% CI, 0.56 to 0.83; P<0.001), genistein intake (RR, 0.9; 95% CI, 0.84 to 0.97; P=0.008), daidzein intake (RR, 0.84; 95% CI, 0.73 to 0.97; P=0.018), and tofu intake (RR, 0.73; 95% CI, 0.57 to 0.94; P=0.013). Heterogeneity among studies evaluated for associations between prostate cancer risk and total soy intake or unfermented soy food intake was high. Fermented soy food intake, total isoflavone intake, and circulating isoflavones were not associated with prostate cancer risk. Neither soy food intake nor circulating isoflavones were associated with advanced prostate cancer risk. Covariates of study design, geographical location of studies, and type of soy intake impacted significance in subgroup analyses. For example, risk was reduced with total soy intake in North American and European studies but not in Asian studies, and with unfermented soy food intake in case-control studies in North America and Asia but not in cohort studies.(Applegate 2018)
In a pilot study enrolling men with prostate cancer, soy isoflavones reduced adverse urinary, intestinal, and sexual effects of radiation.(Ahmad 2010) In a 12-week study in 33 men undergoing androgen deprivation therapy for prostate cancer, 20 g/day of soy protein supplementation (isoflavone 160 mg) produced no effect on any outcomes measured (cognition, sleep quality, vasomotor symptoms, or quality of life).(Sharma 2009) In another trial, genistein 450 mg and daidzein 300 mg for 6 months showed no effect on PSA levels despite an increase in serum isoflavone levels.(deVere White 2010) Data from a preliminary study in 86 men with localized prostate cancer suggest that gene expression in prostate tissue may be altered by intake of soy isoflavones (80 mg/day; 51 mg/day of aglucon units), but serum biomarkers were not affected.(Hamilton-Reeves 2013) Gene expression and genome-wide DNA methylation profiling of tissue from men with localized prostate cancer suggest that genistein may have some demethylation effects, particularly in the genes NOTCH3 and JAG1.(Bilir 2017) A randomized, placebo-controlled trial evaluating soy protein isolate supplementation (20 g/day) for the prevention of prostate cancer recurrence in men who underwent radical prostatectomy was stopped early when interim analysis showed no benefit (hazard ratio [HR], 0.96; 95% CI, 0.53 to 1.72; log-rank P=0.89).(Bosland 2013)
Animal and in vitro data
Genistein has demonstrated inhibition of early cancer markers in human epithelial cells.(Katdare 1998) In one report, genistein slowed growth of implanted tumors in mice and in vitro.(Record 1997) These anticancer effects of genistein may be related to its ability to reduce expression of stress response–related genes. Induction of stress proteins in tumor cells protects them against cell death; therefore, inhibition of this stress response by isoflavone is beneficial.(Zhou 1998)
Meta-analyses of data regarding effects of soy intake on endometrial, ovarian, gastric, and colorectal cancer risk have been published. A protective effect was reported for high soy intake over low intake in meta-analyses of case-control and cohort studies of endometrial and ovarian cancer (OR, 0.61; 95% CI, 0.53 to 0.72). Subgroup analyses revealed preventative effects against endometrial cancer for case-control studies, soy food intake, use of soy-derived isoflavones, unfermented soy food intake, and postmenopausal status; benefit remained significant for both Asian and non-Asian populations.(Myung 2009, Sirtori 2001, Zhang 2015) Reduced risk of ovarian cancer was also demonstrated in a case-control study conducted in Chinese women (N=1,000) who consumed at least 120 g/day of soy foods compared with 61 g/day. A similarly significant inverse relationship was noted between isoflavone intake and ovarian cancer risk (P<0.001).(Lee 2014) However, no association between the risk of ovarian cancer and soy phytoestrogen consumption was observed in the Women's Lifestyle and Health Cohort study.(Hedelin 2011)
Among Japanese and Korean populations, one meta-analysis showed a significant increase in risk of gastric cancer with high intake of fermented soy products (OR,1.22; 95% CI, 1.02 to 1.44) and a significant decrease in risk of gastric cancer with high intake of nonfermented soy products (OR, 0.64; 95% CI, 0.54 to 0.77).(Kim 2011) Similarly, in a meta-analysis of 13 prospective studies (N=517,106 and 58 cases), pooled data showed an inverse association between high intake of total soy food and risk of gastric cancer (highest vs lowest category: RR, 0.78; 95% CI, 0.62 to 0.98). Dose-response meta-analysis showed that high intake of nonfermented soy food (150 g/day) was inversely related to gastric cancer risk; in groups receiving fermented soy food or lower dosages of nonfermented soy products (50 g/day and 100 g/day), no significant association with gastric cancer risk was observed. In men, a high intake of miso soup, particularly 1 to 5 cups/day, was associated with a significant increase in risk of gastric cancer (highest vs lowest category; RR, 1.17; 95% CI, 1.02 to 1.36).(Weng 2017) Results from meta-analyses of studies evaluating the protective effects of soy against colorectal cancer are equivocal.(Yan 2010, Yu 2016) One study established no association, except when a subgroup analysis conducted by gender revealed a decreased risk in women,(Yan 2010) while a systematic review and meta-analysis found an overall risk reduction of 0.77 (95% CI, 0.72 to 0.82; P=0.024) among 17 trials that reported inconsistent results. Statistically significant benefit was found with soy foods/products in Asian populations, as well as with case-control study designs.(Yu 2016) In a meta-analysis of 34 cohort and case-control studies, subgroup analyses showed a strong, statistically significant inverse association between overall risk of GI cancers and dietary intake of soy isoflavones (risk estimate, 0.73; 95% CI, 0.59 to 0.92), but only a slight reduction in risk was related to soy food intake. Modest statistically significant reductions were noted in both male and female subgroups. No significant association between fermented versus nonfermented soy products and GI cancer risk was found. When analyzing according to GI sites, colon cancer and colorectal cancer showed weak but statistically significant reductions in risk; gastric and rectal cancers were associated with small but nonsignificant reductions in risk.(Tse 2016)
A group of investigators evaluated the relationship between soy intake and risk of lung cancer in a cohort (N=74,942) of 40- to 70-year-old women during a mean follow-up period of 9.1 years. Of the 370 lung cancer cases, 340 were in women who had never smoked. Soy consumption was documented during patient interviews, and the highest quintile of soy intake had a significantly lower incidence of lung cancer compared with the lowest quintile of soy intake (HR, 0.63; 95% CI, 0.44 to 0.9). The investigators then conducted a meta-analysis of 11 studies (7 case-control and 4 cohort studies including the current one; N=231,494); the summary RR for the pooled results was 0.83 (95% CI, 0.72 to 0.96).(Yang 2012)
Cardiovascular disease risk factors
Soy isoflavones exhibit strong biological properties in animal studies, causing arterial vasodilation, lowering of serum cholesterol, and inhibition of atherosclerosis in postmenopausal monkeys.(Anthony 1998, Clarkson 2001) However, beneficial effects observed in animal models have not translated well to studies in humans. The widespread availability of clinical trial data makes data from animal studies largely irrelevant.(Sacks 2006)
Soy protein has gained considerable attention for its potential role in improving risk factors for cardiovascular disease.(Sacks 2006) Neither the efficacy of isoflavone content in soy preparations nor the relevance of baseline lipid profiles has been well delineated, nor has a dose-response relationship been confirmed.(Sacks 2006, Vogel 2005, Zhang 2013) Clinical trials, as well as several meta-analyses of clinical trials conducted in adults and children, evaluating the effects of soy, soy proteins, soy flour, and/or soy isoflavones on lipid parameters have been published, with equivocal results. Influences on total cholesterol, triglycerides, low-density lipoprotein (LDL) cholesterol, and high-density lipoprotein (HDL) cholesterol, as well as on lipoprotein(a) and blood pressure, remain inconsistent.(Beavers 2010, Campbell 2010, Carmignani 2014, Gonciulea 2017, Harland 2008, Hooper 2008, Maki 2010, Malhotra 2014, Mohammadifard 2015, Padhi 2015, Qin 2013, Reynolds 2006, Ruscica 2018, Sacks 2006, Santo 2010, Taku 2007, Taku 2010, Zhang 2016)
Risk of stroke and coronary heart disease (CHD) with soy consumption was assessed in a 2016 meta-analysis of observational studies (4,954 stroke cases and 7,616 CHD cases). No association was found between dietary soy intake and stroke risk overall or in subgroup analyses of gender or Asian populations; however, an inverse association was found in case-control studies (high heterogeneity). Overall results were similar for risk of CHD, including in subgroup analysis of Asian populations. In contrast, soy intake was associated with an increased risk of CHD in males (summary RR, 1.36; 95% CI, 1 to 1.85). Analysis of case-control studies showed an inverse relationship between soy intake and CHD risk (no heterogeneity); a significant inverse association was also observed in subgroup analyses of low-quality studies, female subjects, and studies reporting incidental CHD as the outcome. No association was found between soy isoflavone intake and risk of either disease. The majority of studies were of high quality.(Lou 2016)
A 2013 meta-analysis evaluating effects of soy isoflavone supplementation on body weight (9 studies [n=528]) and glucose metabolism (11 studies for fasting glucose [n=1,182]; 11 studies for fasting insulin [n=1,142]) pooled results of high-quality data. A statistically significant overall improvement in body weight was observed with soy isoflavone dosages of 40 to 160 mg/day given over 8 weeks to 1 year compared to placebo (P=0.008). Greater reductions in body weight were observed in the lower-dose subgroup (less than 100 mg/day), in those receiving supplementation for shorter durations (less than 6 months), and in women who were not obese at baseline (BMI less than 30 kg/m2) (P=0.011, P=0.009, and P=0.013, respectively). Fasting insulin was also significantly reduced overall (P=0.021) and in nonobese subjects (P=0.028), whereas a significant difference in blood glucose was observed only with a longer duration of intervention (at least 6 months; P=0.001).(Zhang 2013) In a study comparing plant- and animal-based sources of dietary protein over 6 weeks in 173 postmenopausal women (mostly non-Hispanic white),(Gonciulea 2017, Ruscica 2018) no significant changes in total cholesterol to HDL ratios, triglycerides, fasting glucose, fasting insulin, or insulin resistance were observed. No significant differences were observed with nonsoy plant–based diets compared with animal- and dairy-based diets, with no differences observed between soy and nonsoy plant–based diets or between the 2 animal-based diets (nondairy animal and dairy diets).(Gonciulea 2017) Similarly, in 56 men and postmenopausal women with metabolic syndrome who followed a 12-week Mediterranean-based diet in which protein was sourced from either soy or animals, no differences were found between groups in abdominal adiposity and related adipokines, visceral adipose and related biomarkers, fasting glucose, insulin, insulin resistance, or markers of inflammation.(Ruscica 2018) In a small study (N=17), a surrogate marker for arterial stiffness/endothelial function was statistically significantly improved in men and postmenopausal women with metabolic syndrome who consumed whole soy nuts (70 g/day) for 4 weeks compared with when they crossed over to the control snack. However, little to no clinically relevant changes were seen in inflammatory or oxidative stress biomarkers, hemoglobin A1c, insulin, insulin sensitivity, or fasting glucose. None of the effects were contingent on participants being equol or O-desmethylangolensin producers.(Reverri 2015)
Meta-analyses examining the effect of soy isoflavones(Liu 2012, Zhang 2016) and soy nuts(Mohammadifard 2015) on blood pressure report equivocal results. A meta-analysis of 11 randomized controlled trials with sample sizes ranging from 18 to 302 assessed blood pressure effects of soy protein containing isoflavones (65 to 153 mg/day) ingested for 1 to 12 months. Pooling the equivocal data yielded no significant overall effects on systolic blood pressure (SBP) or diastolic blood pressure (DBP) measures; heterogeneity was high. Statistically significant reductions in both SBP (−5.94 mm Hg; P=0.01) and DBP (−3.35 mm Hg; P=0.04) were observed in patients with hypertension at baseline compared to normotensive populations. Diabetes and hypercholesterolemia significantly interfered with the positive effects of soy isoflavone, although these findings were of limited statistical power.(Liu 2012) Similarly, in a double-blind, randomized, controlled trial conducted in 213 patients with elevated LDL cholesterol, no changes in SBP or DBP were observed with consumption of 25 g/day of soy protein or whey protein (in the form of muffins made from soy flour) for 6 weeks.(Padhi 2015) In contrast, in a double-blind, randomized, controlled trial in early menopausal women (N=200), those consuming soy protein plus isoflavones showed a 3.2 mm Hg reduction in SBP compared to those consuming only soy protein (−0.8 mm Hg); treatment difference was −2.5 mm Hg, which was statistically significant (P<0.01). Additionally, significant improvement was seen in glucose metabolism but not in lipid parameters, and a detrimental effect on thyroid function was observed, specifically in the first 3 months of supplementation.(Sathyapalan 2017)
A 2005 American Heart Association advisory statement(Sacks 2006) and a 2006 expert panel report from the American College of Cardiology(Vogel 2005) found that the evidence regarding clinical benefit of soy in reducing the risk of cardiovascular disease is uncertain and that soy cannot be routinely recommended.(Sacks 2006, Vogel 2005) The 2017 joint position statement of the ISD and ISSA on nutraceuticals for the treatment of hypercholesterolemia encourages dietary intake of soy products as a useful substitute for animal source foods, but notes that evidence regarding cholesterol-lowering effects of soy supplements is contradictory; the ISD/ISSA state that supplementation for cholesterol-lowering purposes may be advised, though with some level of uncertainty and based on low-level evidence, in the general population and in patients with mild hypercholesterolemia who have low to moderate cardiovascular risk (Level II, Strength C).(Pirro 2017)
Chronic kidney disease
The impact of dietary soy intake on patients with chronic kidney disease was assessed in a meta-analysis of 12 randomized trials (N=280); contents of isolated soy protein consumption ranged from 14 to 30.7 g/day, and isoflavones ranged from 35.5 to 61 mg/day. Subgroup analyses were conducted based on lipid levels and dialysis status. Some variations in methodologies and study groups among the trials included diets adopted during the trial (ie, low-protein, standard diabetes, and usual diets), duration of follow-up (7 weeks to 4 years), type of soy used (ie, isolated soy protein with isoflavones, soy diets, soy products), and comorbidities (ie, diabetes, hypertension). Overall, pooled results showed significant reductions in the soy group compared with control for 24-hour urine protein excretion (−0.13 mg/day; P<0.00001), serum urea nitrogen (BUN) (−3.07 mg/dL; P=0.04), and C-reactive protein (−0.98 mg/L; P<0.00001), with one study contributing 99.8% of the weight for the latter effect. In nondialysis patients, soy intake was associated with statistically significant reductions in serum creatinine (−0.05 mg/dL; P=0.04) and serum phosphorus (−0.13 mg/dL; P=0.04). No significant effects on creatinine clearance, glomerular filtration rate, serum albumin, body weight, or BMI were observed.(Jing 2016)
In a 2.5-year, randomized, placebo-controlled trial in 313 postmenopausal women (Women's Isoflavone Soy Health [WISH] trial), soy isoflavone supplementation, at a dose selected to mimic soy intake in a typical Asian diet, was evaluated for effects on cognition. The long-term soy isoflavone supplementation improved visual memory but showed no benefit on cognitive function.(Henderson 2012) Post hoc analysis of data from the WISH clinical trial in postmenopausal women (mean age, 61 years) determined that change in overnight urinary excretion of isoflavonoids was inversely associated with changes in a factor score representing general intelligence, but was not associated with change in a composite score of global cognition; the inverse association between change in isoflavonoids and change in general intelligence factor score was of greater magnitude in women who had undergone surgically induced menopause.(St John 2014) Similarly, no effect on overall cognitive performance was observed with 6 months of soy isoflavone supplementation in a study of men and postmenopausal women with Alzheimer disease. However, limited data supported improvements in 2 of the 4 cognitive domains: speeded dexterity and verbal fluency. A positive correlation was found between the latter and total plasma equol level; only 25% of participants demonstrated measurable increases in equol production by study end. Neither gender nor APOE e4 genotype influenced the response to supplementation.(Gleason 2015)
Vitamin B12 deficiency has been linked to an increased risk of dementia, possibly due to hyperhomocysteinemia that results from this deficiency. The Academy of Nutrition and Dietetics' 2016 updated position statement on vegetarian diets states that neither fermented soy products nor unfortified plant foods are a reliable source of active vitamin B12.(Melina 2016)
Diabetes and glucose metabolism
Encouraging data from rats fed a high–soy isoflavone diet revealed improved insulin secretion and better glycemic control, and led to studies in humans.(Liu 2011, Ricci 2010) The widespread availability of clinical trial data makes data from animal studies largely irrelevant.
Several meta-analyses of trials conducted in peri- and postmenopausal non-Asian and Asian populations have revealed equivocal data regarding the impact of soy isoflavone on body weight, fasting blood glucose, insulin levels, and insulin resistance.(Fang 2016, Gonciulea 2017, Li 2018, Liu 2011, Ricci 2010, Ruscica 2018, Zhang 2013, Zhang 2016) Subgroup analyses suggest that benefit might lie preferentially with nonobese patients, intervention durations of longer than 6 months,(Zhang 2013) whole soy foods,(Liu 2011) and/or genistein dosages of less than 70 mg/day.(Fang 2016) Risk of type 2 diabetes was inversely associated with consumption of soy protein and/or isoflavones in 2 meta-analyses; however, heterogeneity among the 8 trials in one meta-analysis was determined to be unacceptably high to yield meaningful results.(Li 2018, Tian 2017) Further conflicting results were made evident in subgroup meta-analyses of patients with type 2 diabetes and metabolic syndrome in which blood glucose, LDL, and C-reactive protein improved with longer durations of administration (at least 6 months) of soy isoflavones, whereas insulin levels and insulin resistance improved with shorter durations of intervention (less than 6 months).(Zhang 2016)
A small, randomized, crossover trial in patients with type 2 diabetes and diabetic nephropathy (N=25) evaluated 4-week courses of soy milk versus cow's milk for impact on indicators of inflammation, coagulation, and oxidative stress. The only significant difference seen with soy milk was a reduction in D-dimer (−3.77% vs +16.13%; P<0.05). After controlling for confounders, soy milk consumption had no significant effects on tumor necrosis factor alpha, interleukin 6, high-sensitivity C-reactive protein, or malondialdehyde levels compared to cow's milk consumption.(Miraghajani 2012) In a study comparing protein derived from either soy or animal sources in men and postmenopausal women (N=56) with metabolic syndrome who followed a 12-week Mediterranean-based diet, no differences were observed between groups in fasting glucose, insulin, or insulin resistance.(Ruscica 2018) In a study evaluating the effects of soy isoflavones on metabolic status in women with polycystic ovary syndrome, soy isoflavone 50 mg/day for 12 weeks resulted in significant improvements from baseline in measures of insulin resistance, beta-cell function, insulin sensitivity, triglycerides, very low–density lipoprotein (VLDL), and malondialdehyde. The differences experienced in the soy group were significantly greater than those with placebo for all of these measures (P<0.001 to P=0.02), as well as for nitric oxide (P=0.02).(Jamilian 2016) Improved glycemic control was reported in a study of pregnant women with gestational diabetes who consumed a soy-based diet for 6 weeks compared with those on a control diet. However, these results must be interpreted with caution due to inconsistent application of turmeric and tomato paste to the daily preparation of soy textured protein but not to the control diet. Both turmeric and tomato paste contain known potent anti-inflammatory and/or antioxidant constituents (curcumin and concentrated lycopene, respectively); therefore, extrapolating benefit to be related solely to soy is difficult.(Jamilian 2015)
In a randomized, double-blind, crossover, postprandial study in healthy males (N=12), the addition of soy-soluble polysaccharides to glucose solutions or liquid and gelled dairy products was associated with minimal effects on postprandial blood glucose or insulin levels. Although glucose area under the curve and glycemic index values were significantly lower with flaxseed gum (2.5 g) and soy-soluble polysaccharide–fortified (2.25 g) dietary fiber products than with the glucose reference, the significant inverse correlation to product viscosity (which was greatest with flaxseed gum) suggested postprandial glycemia is more dependent on product viscosity than fiber concentration. Fortification of food products, especially dairy, with 1% low-viscosity soluble fibers does not appear to be a useful means of attenuating postprandial blood glucose or insulin response.(Au 2013)
The American Diabetes Association standards of care (2014) recommend an increase in foods containing alpha-linolenic acid, an omega-3 fatty acid found in soy, as a component of medical nutrition therapy for patients with type 1 or type 2 diabetes, based on observed beneficial effects on lipoprotein profiles, heart disease prevention, and overall positive health in observational studies of patients with diabetes (moderate-quality evidence).(ADA 2014)
Food allergy/intolerance in infants
Allergy to cow's milk protein affects approximately 2.5% of infants, with peak presentation usually occurring during the first year of life. Clinically, the situation can be difficult to discern because nonspecific symptoms may be easily confused with functional GI disorders or lactose intolerance. In these cases, alternative protein sources from vegetables, particularly soy, are substituted for cow's milk.(Koletzko 2012) Soy-based infant formulas are also indicated for lactose malabsorption or sensitivity, galactosemia, acute diarrhea, general GI discomfort, and as a source of nutrition for infants in vegetarian families.(Lasekan 2015)
Although the majority of infants tolerate soy-based formulas, approximately 10% to 14% react to soy proteins. Soy allergy is one of the most common food allergies globally and is in the top 8 of products on the World Health Organization's allergic foods list.(Asero 2016) However, the cumulative prevalence of true IgE-mediated soy allergy or sensitization is likely overestimated. A systematic review found the rate in children younger than 19 years to be less than 0.3%.(Katz 2014) Soy allergy can also be present in a nonspecific manner and complicate the management of GI disorders such as celiac disease.(Syrigou 2014) Once tolerance of soy protein has been established, soy protein–based formula is an option for infants older than 6 months. However, soy-based formulas have nutritional disadvantages over breast milk. In addition to lower bioavailability and/or absorption of nutrients, soy isoflavones have weak estrogenic effects that can lead to high serum concentrations in infants.(Koletzko 2012, Osborn 2006)
In a 35-day feeding trial in healthy term infants (N=120), addition of mixed carotenoids, with and without sucrose, to soy-based infant formula did not negatively affect tolerance or hydration status to a considerable degree and thus may be a means of improving the nutrient profile.(Lasekan 2015) Data on the use of alternative parenteral lipid emulsions (eg, medium-chain/long-chain triglycerides, medium-chain triglycerides/olive/fish/soy) in preterm infants are of low to very low quality and are insufficient to provide recommendations for or against their use over conventional parenteral soy-based lipid emulsion formulas.(Kapoor 2015) However, a study of infants receiving prolonged parenteral nutrition suggests that replacement of standard soy-based parenteral lipid nutrition (10% Intralipid) with a fish oil–based lipid preparation (10% Omegaven) could halt or even reverse parenteral nutrition–associated cholestasis in infants dependent on parenteral nutrition.(Lam 2014)
The 2016 Academy of Nutrition and Dietetics' updated position statement on vegetarian diets advises that breastfeeding be provided for the first 6 months in infants. However, if breastfeeding is not possible, then commercial, infant formula should be the primary beverage for the first year and full-fat, fortified soy milk was noted as a suitable primary beverage for children starting at 1 year of age who are growing normally and eating a variety of foods.(Melina 2016) Although many infants with food intolerance become tolerant over time, evidence is insufficient to suggest that substitution with soy milk can prevent the development of atopy (hereditary hypersensitivity) or food intolerance.(Osborn 2006) Because of the nutritional disadvantages of soy-based formula, both the ESPGHAN and the AAP consider extensively hydrolyzed infant formula and amino acid–based formula to be preferable to infant soy formula in non-breastfed infants with cow's milk protein allergy; a soy formula may be considered for infants on a vegan diet. ESPGHAN and AAP recommend against the use of soy milk in infants because it is not suitable to meet infant nutritional needs.(Koletzko 2012)
In one report, the use of fiber-supplemented soy formula reduced the duration of diarrhea in 44 infants.(Vanderhoof 1997) Soy has also been investigated for the treatment of infantile colic(Lucassen 1998) and recurrent abdominal pain in childhood(Huertas-Ceballos 2002); however, there is a lack of evidence suggesting that soy has any beneficial effect in these conditions. In a 2×2 factorial, randomized, placebo-controlled study conducted in 100 adult women with irritable bowel syndrome (IBS), significant improvements in symptom severity scores were observed in those who received soy isoflavones for 6 weeks compared to those who did not (P=0.001). The soy supplements were taken twice daily and contained 20 mg of soy isoflavones (daidzein 10 mg, genistein 8.5 mg, and glycerin 1.5 mg); vitamin D was taken once biweekly and contained 50,000 units of cholecalciferol. Severity of abdominal pain (P=0.018), duration of abdominal pain (P=0.001), satisfaction of bowel habits (P=0.011), and life disruption (P<0.001) were all significantly improved with soy compared to placebo, with some data suggesting a synergistic effect with coadministration of vitamin D, based on total IBS scores and satisfaction of bowel habits.(Jalili 2016)
Because of their weak estrogenic activity, soy isoflavones have been hypothesized to improve several estrogen-dependent conditions, including perimenopausal vasomotor symptoms (eg, hot flashes) and postmenopausal bone loss. Interest in the use of soy and its derivatives for the treatment of menopausal symptoms has been encouraged by observations of a lower prevalence of menopausal complaints, especially hot flashes, among women in Asian countries where soy is an important component of the traditional diet.(Lethaby 2007)
The widespread availability of clinical trial data makes data from animal studies largely irrelevant.
Reviews and meta-analyses of clinical trials evaluating the efficacy of soy products and phytoestrogens in managing the symptoms of menopause are available, including a Cochrane meta-analysis.(Krebs 2004, Kronenberg 2002, Lethaby 2007, Li 2015) Heterogeneity of included study populations, treatment regimens, outcome measures, and trial methodology exists. A placebo effect was noted in most trials, making adequate blinding and randomization vital to the results in these studies.(Lethaby 2007, Li 2015)
Data from the included studies are conflicting. In one meta-analysis, overall combined results and results by subgroup (soy concentrates [genistein or daidzein], soy extracts, or dietary soy supplementation) showed a significant tendency in favor of soy compared to placebo for the management of menopausal vasomotor symptoms, but establishing conclusive results was difficult due to high heterogeneity among the studies.(Bolaños 2010) A Cochrane meta-analysis found no evidence of effectiveness in alleviating menopausal symptoms with use of phytoestrogen treatments, including soy.(Lethaby 2007) In a model-based meta-analysis, soy isoflavones showed a quantitatively slight and slow benefit in attenuating hot flushes compared to estradiol (48 weeks vs 12 weeks to achieve 80% of the maximum effect, respectively), with no difference in effect observed between soy and placebo.(Li 2015) Another analysis evaluated the effect of soy protein and isoflavones on circulating hormones in pre- and postmenopausal women. No effect on hormones (estradiol, estrone, sex hormone–binding protein, follicle-stimulating hormone [FSH], and luteinizing hormone [LH]) was reported in postmenopausal populations, and only a modest effect was found on FSH and LH in premenopausal women.(Hooper 2009) Two small, randomized clinical trials published subsequent to earlier meta-analyses evaluated the effect of dietary soy, hormone replacement therapy, and placebo on menopausal symptoms and revealed a reduction in the severity of hot flashes, bone/joint pain, and/or vaginal dryness for soy and hormone therapy.(Carmignani 2010, Carmignani 2014) As in other studies, no change in FSH or estradiol was observed with soy supplementation.(Carmignani 2014) A double-blind, randomized, placebo-controlled trial conducted in 180 postmenopausal (natural) women investigated the effect of omega-3 fatty acid or soybean supplementation on menopausal symptoms. Daily supplementation with omega-3 fatty acids 1,000 mg (EPA 180 mg, DHA 120 mg), soybean extract 500 mg, or placebo for 3 months resulted in significant improvements with both active interventions compared to placebo for the overall Menopause Rating Scale score (P<0.001) with no difference between the omega-3 and soybean groups. Likewise, significant differences were observed for both active interventions compared to placebo in subscale scores related to hot flushes/sweating, sleep problems, somato-vegetative symptoms, depression, and psychological scores, with no significant differences between omega-3 fatty acid and the soybean extract groups. All treatments were well tolerated; however, 2 participants in the soybean extract group reported allergic reactions.(Purzand 2020)
In a 2-year trial evaluating soy isoflavones as an estrogen replacement for prevention of menopausal bone loss and menopausal symptoms (Soy Phytoestrogens As Replacement Estrogen [SPARE] trial), no significant differences regarding changes in bone mineral density (BMD) of the spine, total hip, or femoral neck were observed between participants receiving soy and those receiving placebo. A significantly larger proportion of participants in the soy group experienced hot flashes and constipation compared with the control group.(Levis 2011)
Preliminary data suggest that soy isoflavones, but not soy protein, have a beneficial effect on bone turnover in postmenopausal women.(Pawlowski 2015, Sathyapalan 2017) In a randomized crossover trial, various soy isoflavone interventions were evaluated for effects on bone calcium retention. Bone calcium retention, indicative of bone strength, was most effectively increased in 24 postmenopausal women (by 7.6%) by a lower dose of naturally occurring mixed soy isoflavones (total isoflavones 105 mg/day) compared with high-dose mixed soy isoflavones (by 5.5%), mixed soy/genistein (by 5.8%), and low-dose genistein (by 3.4%); though significant, these effects were less than that observed with risedronate (15.3% increase in bone calcium retention). No significant increase was observed in the high-dose genistein group. At the lower genistein dose, addition of daidzein and glycitein appeared to enhance the antiresorptive effect. Equol production status of the women did not affect the results.(Pawlowski 2015) In a study of 200 women within 2 years of menopause onset, daily supplementation with soy protein (15 g) plus isoflavone (66 mg) for 6 months resulted in a statistically significant reduction in both bone resorption and bone formation markers (P<0.01 for both) compared with soy protein alone. These effects were more pronounced between months 3 and 6.(Sathyapalan 2017)
The American Association of Clinical Endocrinologists guidelines for clinical practice for the diagnosis and treatment of menopause (2011) note that although soy may have some estrogen agonist properties, long-term safety issues, particularly in patients with breast cancer, remain of concern for high-dose therapy. The updated position statement (2017) advises against the use of phytoestrogens for treatment of hot flushes in breast cancer survivors.(Cobin 2017, Goodman 2011) The Society of Obstetricians and Gynaecologists of Canada's revised clinical practice guidelines on managing menopausal vasomotor symptoms (2021) state that a standard dose of fermented soybean extract may be effective for the relief of vasomotor symptoms. Additionally, despite data suggesting that s-equol may be beneficial for vasomotor symptoms, evidence is inconclusive regarding the use of soy isoflavonoids.(Yuksel 2021) The Endocrine Society's clinical practice guidelines for the treatment of symptoms of menopause (2015) recommend counseling patients on the lack of consistent evidence for benefit of complementary therapies, including soy isoflavones, as an alternative nonhormonal therapy for vasomotor symptoms (weak recommendation; low-quality evidence).(Stuenkel 2015) A North American Menopause Society (NAMS) position statement for nonhormonal management of menopause-associated vasomotor symptoms (2015) recommends the use of soy with caution, specifically the S-equol derivative of isoflavones and extract, in postmenopausal women with mild to moderate hot flushes and without a history of soy intolerance or allergy. Alternative treatment should be sought if there is no response within 12 weeks (level II evidence).(NAMS 2015)
Obesity and nutrient intake
Improvement in body weight and BMI,(Liu 2010) as well as in abdominal and/or overall fat,(Cha 2014, Christie 2010) has been documented with soy in individual trials and a meta-analysis(Zhang 2013); however, data are inconsistent.(Akhlaghi 2017, Neascu 2014, Ruscica 2018, van Nielen 2014) In clinical trials, the source of dietary protein (soy based vs animal based) did not differentially affect weight loss, satiety, gut hormones, or motivation to eat in 20 obese men(Neascu 2014); body composition or glucose metabolism in 15 obese postmenopausal women(van Nielen 2014); or weight gain in 86 teenaged college females.(Berger 2014) Preliminary data suggest that individuals with peroxisome proliferator-activated receptor (PPAR)-gamma mutant TT allele polymorphism may be more susceptible to visceral fat loss than those with CC or CT variants following fermented soybean supplementation for 12 weeks. Of note, subjects with the TT allele variant consumed significantly fewer carbohydrates and less sodium during the study period compared to subjects in other groups.(Cha 2014)
A meta-analysis of randomized controlled trials published through August 2016 assessed the effect of soy (N=1,634) and soy isoflavones (N=1,113) on the obesity-related factors of weight, BMI, waist circumference, fat mass, and abdominal fat in otherwise healthy adults. Studies examining diseases other than hyperlipidemia and nonalcoholic fatty liver disease were excluded. Additionally, interventions that employed other soy varieties or specific extracts (eg, black soy, fermented soy products, daidzein, genistein) were excluded. Dosages ranged from 7.5 to 116 g/day of soy protein and from 33.3 to 300 mg/day of soy isoflavones for up to 2 years. Overall, no significant pooled effect was found for soy protein or soy isoflavones on weight or BMI, respectively, or for either intervention on waist circumference or fat mass. Subgroup analysis showed that soy produced a significant increase in weight compared to controls in subjects with baseline obesity (P=0.02), in those receiving doses of 40 g/day or greater (P=0.03), and those receiving shorter durations of soy consumption (1 to 3 months) (P=0.03). When compared with meat and whey protein, but not casein/whole milk or usual diet/cereals, soy protein led to a significant decrease in weight (P=0.03 and P=0.04, respectively). Subgroup effects on other outcomes for soy protein and soy isoflavones were either not significant or were less clear due to high heterogeneity and/or small sample sizes.(Akhlaghi 2017) Similarly, in a study of men and postmenopausal women (N=56) with metabolic syndrome who followed a Mediterranean-based diet for 12 weeks in which protein was sourced from either soy or animals, no differences were observed between groups in abdominal adiposity and related adipokines, visceral adipose and related biomarkers, fasting glucose, insulin, or insulin resistance.(Ruscica 2018) In contrast, a meta-analysis of randomized controlled trials conducted in non-Asian postmenopausal women consuming dietary soy isoflavones and not receiving hormone replacement therapy revealed a statistically significant overall improvement in weight. Pooled results of high-quality data from 528 subjects revealed a statistically significant overall improvement in body weight with soy isoflavone dosages of 40 to 160 mg/day (treatment duration range, 8 weeks to 1 year) compared to placebo (P=0.008); heterogeneity was not significant. Greater reductions in body weight were observed in the lower-dose subgroup (less than 100 mg/day), in those receiving supplementation for shorter durations (less than 6 months), and in women who were not obese at baseline (BMI less than 30 kg/m2) (P=0.011, P=0.009, and P=0.013, respectively).(Zhang 2013)
The effects of soy protein with and without isoflavones have been studied in animal models, with conflicting results.(Breitman 2005, Gallo 2005, Nakai 2005, Register 2003) The widespread availability of clinical trial data makes data from animal studies largely irrelevant.
Several meta-analyses of clinical trials evaluating the efficacy of soy preparations in protecting against decreases in BMD and bone turnover markers (published up to 2017) have been conducted.(Darling 2009, Liu 2009, Ma 2008, Ma 2008, Pawlowski 2015, Sathyapalan 2017, Taku 2010, Wei 2012) Heterogeneity was present among the included trials, and influences of ethnicity and basal BMD have not been determined. Results data are conflicting, with some meta-analyses reporting small improvements in bone density and/or urine deoxypyridinoline,(Darling 2009, Ma 2008, Ma 2008, Wei 2012) some showing benefits only in subgroups (ie, postmenopausal women, with soy isoflavone dosages greater than 75 mg/day),(Wei 2012) and others reporting no effect.(Liu 2009, Taku 2010) Data from long-term clinical trials published subsequent to earlier meta-analyses reveal no treatment effect after 3 years of soy isoflavone supplementation (Soy Isoflavones for Reducing Bone Loss [SIRBL] study),(Alekel 2010) an increase in whole BMD after 2 years but no influence at common spine and hip fracture sites,(Wong 2009) and no effect after 3 years on biomarkers in general in healthy postmenopausal women.(Shedd-Wise 2011) Studies evaluating reductions in fracture rates in women with osteoporosis are lacking.
In a randomized crossover trial, various soy isoflavone interventions were evaluated for effects on bone calcium retention. Bone calcium retention was most effectively increased in 24 postmenopausal women (by 7.6%) by a lower dose of naturally occurring mixed soy isoflavones (total isoflavones 105 mg/day) compared with high-dose mixed soy isoflavones (by 5.5%), mixed soy/genistein (by 5.8%), and low-dose genistein (by 3.4%); though significant, these effects were less than that observed with risedronate (15.3% increase in bone calcium retention).(Pawlowski 2015)
In a study of 200 women within 2 years of menopause onset, daily supplementation with soy protein (15 g) plus isoflavone (66 mg) for 6 months resulted in a statistically significant reduction in both bone resorption and bone formation markers (P<0.01 for both) compared with soy protein alone. These effects were more pronounced between months 3 and 6.(Sathyapalan 2017)
Guidelines concerning the use of soy in osteoporosis have been published. The updated American Association of Clinical Endocrinologists and American College of Endocrinology clinical practice guidelines for the diagnosis and treatment of postmenopausal osteoporosis (2020) note a lack of conclusive evidence to support the use of phytoestrogens or natural estrogen-receptor agonists (isoflavones) for increasing bone density or decreasing risk of fracture.(Camacho 2020) The American College of Obstetricians and Gynecologists updated guidelines for the management of postmenopausal osteoporosis (2022) state that studies of soy isoflavone supplements have produced mixed results regarding effects on BMD and that there are no prospective studies on the effect of soy isoflavones on fracture rate. No recommendations could be made regarding the use of soy, and they advised that patients at risk should be counselled regarding effective pharmacotherapies.(ACOG 2022) The North American Menopause Society's updated position statement on the management of osteoporosis in postmenopausal women (2021) does not recommend the use of isoflavones, such as those from soy, as effective strategies to prevent or treat postmenopausal osteoporosis.(NAMS 2021)
In vivo and ex vivo data
In a study exploring application of a topical soy-based cream (soy oligopeptides) to skin that had been UVB irradiated to simulate sunburn, concentrations of 5 and 10 units/mL, but not 2.5 units/mL, significantly improved the erythema index and subcutaneous hydration compared with untreated UVB-irradiated skin (P<0.05 each). Transepidermal water loss and melanin index were not affected.(Zhou 2016)
Polycystic ovary syndrome
The effect of soy isoflavones on the metabolic status of women with polycystic ovary syndrome was investigated in a double-blind, randomized, placebo-controlled trial (N=70). Patients completed food and activity diaries at baseline and every 3 weeks. After 12 weeks, the soy isoflavone (50 mg/day, equivalent to 500 mL/day of soy milk) group exhibited significant improvements from baseline in measures of insulin resistance, beta-cell function, insulin sensitivity, total testosterone, sex hormone–binding globulin, free androgen index, triglycerides, VLDL, and malondialdehyde compared to placebo (undefined). Those in the soy group also had significant improvements in glutathione levels and hirsutism. After adjustment for baseline values of biochemical markers, age, and baseline BMI, differences observed in the soy group were significantly greater than those in the placebo group for all the described measures (P<0.001 to P=0.02), as well as for nitric oxide (P=0.02). Soy isoflavone supplementation was well tolerated, with no reports of adverse effects.(Jamilian 2016)
Supplementation with soy protein concentrate during a 16-week resistance training program in healthy young males provided no additional increase in muscle mass. Similar results were found with other protein supplements (ie, whey protein, leucine).(Mobley 2017) In a small double-blind, randomized controlled trial in postmenopausal women (N=32), addition of soy protein to milk during a 16-week resistance training program resulted in no difference from placebo for effects on muscle mass; however, muscle strength was increased.(Orsatti 2018)
The Academy of Nutrition and Dietetics' updated position paper on vegetarian diets (2016) states that adequate nutrition can be provided by a well-planned vegetarian diet that includes legumes. Therapeutic vegetarian diets are useful in maintaining a healthy weight and BMI and are associated with a reduction in cardiovascular disease risk and type 2 diabetes. Regular use of soy products helps to ensure adequate intake of protein as well as other essential nutrients, such as zinc. Although soybeans contain natural goitrogens, they have not been associated with thyroid insufficiency in healthy people with adequate iodine intake. In toddlers as young as 1 year of age who are growing normally and eating a variety of foods, fortified soy milk can be started if breastfeeding is not possible. In older adult vegetarians and vegans, it is important that legumes and soy foods are included as protein sources to meet higher protein requirements associated with aging.(Melina 2016)
One gram of soy protein in traditional soy foods contains approximately 3.5 mg of isoflavones (expressed as aglycone weight).Hamilton-Reeves 2010
Data regarding the benefit of soy for most uses are equivocal, and subgroup analyses suggest that benefit may preferentially lie with specific populations and/or forms of soy (ie, unfermented) in certain circumstances. For example, improvements have been observed for some uses only when soy was taken short-term (less than 6 months), and for others only when it was taken for more than 6 months. Negative effects of soy have also been reported: For example, the risk of gastric cancer was increased in men who consumed 1 to 5 cups/day of miso soup. Dosages of various forms of soy in clinical studies evaluating various uses have included 22.7 to 300 mg/day of soy isoflavones, up to 40 g/day of isolated soy protein, 120 g/day of dietary soy foods, 50 to 150 g/day of unfermented soy foods, up to 450 mg/day of genistein, up to 300 mg/day of daidzein, or 70 g/day of whole soy nuts.Akhlaghi 2017, deVere White 2010, Fang 2016, Gonciulea 2017, Jing 2016, Lee 2014, Li 2018, Liu 2011, Reverri 2015, Ricci 2010, Ruscica 2018, Sharma 2009, Wei 2012, Weng 2017, Zhang 2013, Zhang 2016
Dosages ranging from 40 to 160 mg/day (treatment duration range, 8 weeks to 1 year) were evaluated for effects on body weight and glucose metabolism in a meta-analysis of postmenopausal women; greater reductions in body weight were observed with doses less than 100 mg/day for durations of less than 6 months in women who were not obese (BMI less than 30 kg/m2).Zhang 2013
Cardiovascular disease risk factors
Consumption of 70 g/day of whole soy nuts for 4 weeks, regardless of equol production, was used in a study evaluating effects on arterial stiffness/endothelial function in men and postmenopausal women with metabolic syndrome.Reverri 2015
Chronic kidney disease
Dosages ranging from 14 to 30.7 g/day of isolated soy protein or 35.5 to 61 mg/day of soy isoflavones (treatment duration range, 7 weeks to 4 years) were reported in a meta-analysis of studies evaluating effects on various renal parameters (ie, urine protein excretion, BUN, serum creatinine, serum phosphorus) in patients with chronic kidney disease.Jing 2016
Dosages of up to 150 g/day of nonfermented soy foods were used to evaluate effects on gastric cancer risk; it should be noted that high intake of miso soup (1 to 5 cups/day) was significantly associated with an increased risk of gastric cancer in men.Weng 2017
Irritable bowel syndrome
20 mg of soy isoflavones (each capsule containing daidzein 10 mg, genistein 8.5 mg, and glycerin 1.5 mg) twice daily (in addition to biweekly administration of vitamin D) was evaluated in a study of women with IBS.Jalili 2016
Dosages studied for prevention of osteoporosis ranged from 22.7 to 114 mg/day of soy isoflavones (treatment duration range, 1 month to 2 years) in a systematic review. Subgroup analysis showed benefit in postmenopausal women for dosages greater than 75 mg/day.Wei 2012 In a trial in menopausal women, soy protein plus isoflavone supplementation (15 g plus 66 mg daily for 6 months) was used to improve bone biomarkers.Sathyapalan 2017 In postmenopausal women, 105 mg/day of mixed soy isoflavones has been used for prevention of bone loss.Pawlowski 2015
Polycystic ovary syndrome
50 mg/day of soy isoflavones for 12 weeks was used in a small study evaluating effects on several metabolic parameters in women with polycystic ovary syndrome.Jamilian 2016
Pregnancy / Lactation
Information regarding safety and efficacy in pregnancy and lactation is limited. Soy is GRAS when used as food. Avoid dosages above those found in food. Consumption of soy products increases soy isoflavone content in breast milk; 36.5 g/day (250 mL) of a soy protein–based beverage over 2 to 4 days increased isoflavone content in breast milk by a factor of 10 to 15 compared with baseline.(Jochum 2017)
The 2016 Academy of Nutrition and Dietetics' updated position statement on vegetarian diets advises that breastfeeding be provided for the first 6 months in infants. However, if breastfeeding is not possible, then commercial infant formula should be the primary beverage for the first year and full-fat, fortified soy milk was noted as a suitable primary beverage for children starting at 1 year of age who are growing normally and eating a variety of foods.(Melina 2016) Because of nutritional disadvantages of soy-based infant formulas (lower absorption of minerals and trace elements; appreciable amounts of isoflavones with weak estrogenic action, leading to high serum concentrations in infants), both ESPGHAN and AAP consider extensively hydrolyzed infant formula and amino acid–based formula to be preferable to infant soy formula in non-breastfed infants with cow's milk protein allergy; soy formula may be considered in infants on a vegan diet. ESPGHAN and AAP recommend against the use of soy milk because it is not suitable to meet the nutritional needs of infants.(Koletzko 2012)
Estrogen Derivatives: Soy may enhance the adverse/toxic effect of estrogen derivatives. Monitor therapy.Goodman 2011, Zava 1998
Soybeans and their products are generally well tolerated. A 2-year trial of daily supplementation with 80 or 120 mg of soy isoflavones reported no effect on all measured laboratory indices, except for a minimal increase in BUN at the 2-year mark.(Steinberg 2011)
Effects of phytoestrogens in soy-based infant formulas and commercial soy preparations have been of concern.(Bluck 1997, Huggett 1997)
A meta-analysis of 15 clinical studies showed no effect on testosterone or sex hormone–binding globulin levels,(Hamilton-Reeves 2010) and in another clinical trial, semen quality in healthy men was unaffected by 2 months of high-dose isoflavones.(Beaton 2010)
A randomized clinical trial evaluated the effect of soy phytoestrogens on subclinical hypothyroidism over 8 weeks. Six participants in the study who received higher-dose phytoestrogen (16 mg daily) developed overt hypothyroidism, while secondary outcomes for the study populations were positive for decreased blood pressure and insulin resistance.(Sathyapalan 2011) Although case reports are lacking, soy formula–fed infants may be at risk of thyroid dysfunction; however, the US Department of Health and Human Services National Toxicology Program has concluded that there is minimal concern that soy infant formula adversely affects development parameters, including with respect to thyroid function.(Fitzpatrick 2000, NTP 2010) Soybeans, like cruciferous vegetables, are known to contain natural goitrogens and have been shown to lower thyroid function and cause severe hypothyroidism, particularly in susceptible patients, but have not been associated with thyroid insufficiency in healthy people.(Melina 2016, Nakamura 2017)
Allergy (eg, asthma, anaphylaxis) has mostly been reported in children but can also occur in adulthood.(Asero 2016) Although soybeans and peanuts, as well as other beans, are phylogenetically and antigenetically similar, data are insufficient to recommend soy avoidance in peanut-allergic patients.(Codina 1997, Inomata 2005, Sicherer 2000) However, anaphylaxis from suspected cross-sensitivity between PGA of fermented soybeans (natto) and PGA from jellyfish has been noted in several case studies of surfers or divers regularly stung by jellyfish who also had natto allergies.(Inomata 2014) Anaphylaxis to PGA in natto is often a late-onset reaction, with symptoms developing several hours after ingestion. Cases have been reported in both children and adults, many of whom also exhibited other immune-based conditions, including celiac disease or asthma.(Honda 2014, Inomata 2014, Mori 2017, Syrigou 2014) A 30-year-old woman with atopic dermatitis and severe hand eczema experienced an immediate anaphylactic reaction following percutaneous sensitization by soy-based ingredients in cosmetic lotions.(Yagami 2015) Other less obvious sources of soy exposure should be noted, including various pharmaceuticals that use soy lecithin as an emulsifier in their formulation; these products are administered topically, orally, intravenously, or by inhalation. In a case report of an 9-year-old with soy allergy, an injection of benzathine benzylpenicillin containing soy lecithin as a constituent of both the prefilled syringe and formulation powder and solvent for reconstitution resulted in a delayed papular rash at the injection site.(Barni 2015) Similarly, bronchospasm, dermatitis, and anaphylaxis have been induced by propofol formulated with 10% soybean oil in a few patients with food allergies to soy.(Richard 2016)
Evidence exists from animal studies regarding the adverse effects of genistein on the developing female reproductive tract, including decreased age at vaginal opening; abnormal estrous cyclicity; decreased fertility, implants, and litter size; and histopathology of the female reproductive tract.NTP 2010 Similarly, studies of male rats administered soy milk for 35 days have demonstrated endocrine disruption during prepubertal development, leading to decreased testosterone; this effect was magnified with concomitant consumption of glyphosate.Nardi 2017
This information relates to an herbal, vitamin, mineral or other dietary supplement. This product has not been reviewed by the FDA to determine whether it is safe or effective and is not subject to the quality standards and safety information collection standards that are applicable to most prescription drugs. This information should not be used to decide whether or not to take this product. This information does not endorse this product as safe, effective, or approved for treating any patient or health condition. This is only a brief summary of general information about this product. It does NOT include all information about the possible uses, directions, warnings, precautions, interactions, adverse effects, or risks that may apply to this product. This information is not specific medical advice and does not replace information you receive from your health care provider. You should talk with your health care provider for complete information about the risks and benefits of using this product.
This product may adversely interact with certain health and medical conditions, other prescription and over-the-counter drugs, foods, or other dietary supplements. This product may be unsafe when used before surgery or other medical procedures. It is important to fully inform your doctor about the herbal, vitamins, mineral or any other supplements you are taking before any kind of surgery or medical procedure. With the exception of certain products that are generally recognized as safe in normal quantities, including use of folic acid and prenatal vitamins during pregnancy, this product has not been sufficiently studied to determine whether it is safe to use during pregnancy or nursing or by persons younger than 2 years of age.
Frequently asked questions
More about soy
Related treatment guides
Always consult your healthcare provider to ensure the information displayed on this page applies to your personal circumstances.
Copyright © 2023 Wolters Kluwer Health