By Priscilla Abercrombie, RN, NP, PhD, AHN-BC

Women's Health & Healing, Healdsburg, CA

Dr. Abercrombie reports no financial relationships relevant to this field of study.

SYNOPSIS: This 12-week randomized, double-blind, placebo-controlled trial found that soy isoflavones affected some metabolic parameters but not others in women with polycystic ovary syndrome.

SOURCE: Jamilian M, Asemi Z. The effects of soy isoflavones on metabolic status of patients with polycystic ovary syndrome. J Clin Endocrinol Metab 2016;101:3386-3394.

SUMMARY POINTS

  • The investigators undertook a rigorously designed study to understand the impact of soy isoflavones on metabolic status in women with polycystic ovary syndrome (PCOS).
  • Seventy women with PCOS were randomized to either receive 50 mg/d of soy isoflavones containing 37.5 mg genistein, 10 mg daidzein, and 2.5 mg glycitein or placebo for 12 weeks.
  • Few conclusions can be drawn from this study because the results were mixed for each of the aspects of PCOS studied: metabolic, hormonal, and inflammatory.

The most common endocrine disorder among reproductive age women is polycystic ovary syndrome (PCOS).1 Prevalence is 6-10%, depending on the diagnostic criteria utilized.2 Consensus guidelines developed by the American Association of Clinical Endocrinologists and the Androgen Excess and PCOS Society1 state the diagnosis of PCOS is made by the existence of two of the following: chronic anovulation, hyperandrogenism (clinical and biological), and polycystic ovaries. These criteria are consistent with the Rotterdam Criteria developed in 2003.3 All women with PCOS are at increased risk for metabolic syndrome and should be evaluated for impaired glucose intolerance, hypertension, and lipid abnormalities. In addition, women with PCOS are at an increased risk for infertility, endometrial hyperplasia, and gestational diabetes. Goals of treatment vary according to symptomatology, metabolic risk factors, and desire for pregnancy. Typical treatments emphasize weight loss and employ the use of pharmaceuticals such as metformin, combined oral contraceptives, and spironolactone.

The role of adipose tissue and chronic inflammation in the pathophysiology of PCOS is a burgeoning area of research.4 Current investigations focus on the link between the development of metabolic, endocrine, and reproductive dysfunctions of PCOS and adipose tissue dysfunction and chronic low-grade inflammation.5 Women with PCOS frequently suffer from abdominal adiposity. Adipose tissue is involved in the release of inflammatory mediators such as cytokines, acute phase proteins, and adipokines that may contribute to insulin resistance and cardiovascular risk factors.

Soy typically contains three isoflavones: genistein, daidzein, and glycitein. Soy isoflavones are being studied for their role as anticancer, antioxidant, and anti-inflammatory agents in the context of metabolic disorders.6 A recent meta-analysis of randomized, clinical trials suggested that soy improves lipid profiles, especially among those who are hypercholesteremic, but it is more beneficial when eaten as a whole food vs. an isoflavone supplement.7 Another meta-analysis found that soy improved glucose metabolism in postmenopausal women.8 These meta-analyses suggest that soy may have a role in improving lipids and glucose metabolism in some populations.

The purpose of the study by Jamilian and Asemi was to determine if supplementation with soy isoflavones had an effect on metabolic status of women with PCOS. Seventy women with PCOS and between ages 18-40 years were recruited from a clinic in Iran during a three-month period ending in February 2016. Participants met the Rotterdam criteria for PCOS. Pregnant women and those with elevated prolactin, thyroid disorder, endocrine diseases including diabetes or impaired glucose tolerance, and gastrointestinal problems were excluded. Participants were matched for age, phenotypes of PCOS, and body mass index (BMI), and then were randomly assigned to receive either soy isoflavone supplements (n = 35) or placebo (n = 35) for 12 weeks. A power analysis determined an adequate sample size of 30 in each group, allowing for five dropouts in each group with testosterone as the primary outcome variable. Computer-generated random numbers were used to assign participants to groups. Both researchers and participants were blinded to randomization. Participants were advised not to change their physical activity or take any nutritional supplements during the trial. All participants completed three-day food records and physical activity records at baseline, three, six, and nine weeks and the end of the trial. Macronutrient and micronutrient intakes were analyzed with nutrition software. Physical activity was determined by metabolic equivalents (METs) in hours per day. Anthropometric measurements were taken and included BMI, waist circumference, and hip circumference. Clinical assessments included documenting hirsutism, acne, and alopecia using a scoring system.

Primary outcomes were markers of insulin resistance and androgens. Secondary outcomes were lipid profiles and biomarkers of inflammation and oxidative stress. The following biochemical assessment was obtained from fasting blood samples at the beginning and end of the study: serum insulin, homeostasis model of assessment-insulin resistance (HOMA-IR), homeostasis model of assessment ß-cell (HOMA-B), quantitative insulin sensitivity check index (QUICKI), total testosterone (T), SHBG, free T, DHEA, free androgen index (FAI), fasting plasma glucose, serum triglycerides, very low-density lipoprotein (VLDL), total LDL, high-density lipoprotein (HDL), high sensitivity C-reactive protein (hs-CRP), nitric oxide (NO), total antioxidant capacity (TAC), total glutathione (GSH), and malondialdehyde (MDA). Detailed testing parameters and types of testing kits can be found in the article.

The intervention group received 50 mg/d of soy isoflavones containing 37.5 mg genistein, 10 mg daidzein, and 2.5 mg glycitein for 12 weeks based on a previous study with PCOS patients. This dosage is equivalent to consuming 500 mL of soy milk daily or two servings of soy. Placebo capsules were of similar shape and size. Supplements were distributed every four weeks and included an extra three-day supply to last until the next scheduled visit. Participants were instructed to return unused pills at each two-week visit. All participants received reminders on their cell phones daily to increase compliance.

All statistical methods employed to evaluate the study appear to be appropriate for the type of variables studied.

All 70 participants completed the trial with a > 90% compliance rate of capsules taken. No side effects were reported. There were no significant differences in anthropometric measurements, METs, or nutrition intake from baseline to the end of the trial between the two groups.

The results are displayed in Table 1. Alopecia significantly decreased in the soy supplement group (31.6 vs. 4.3%; P = 0.01) and the modified Ferriman-Gallwey (mFG) score for hirsutism significantly improved. Neither group had a change in acne. Serum levels of insulin, HOMA-IR, and HOMA-B decreased, and QUICKI increased. There were significant reductions in total T, FAI, mFG scores, triglycerides, and VLDL compared to the placebo group. In addition, there were significant increases in SHBG and plasma GSH and a decrease in MDA in the treatment group. There was no significant effect of the intervention on other lipids or inflammatory and oxidative stress markers. In summary, there were some changes in metabolic and hormonal biomarkers but no changes in most markers of inflammation or oxidative stress.

Table 1: Comparing Baseline and Intervention Groups After 12 Weeks in Women with PCOS

 

Placebo Group (n = 35)

Intervention Group (n = 35)

 

Lab Test

Change in mean

SD

P value*

Change in mean

SD

P value*

P value**

ANOVA

FPG, mg/dL

-0.02

±4.5

0.73

1.5

±6.7

0.18

0.19

Insulin, µIU/mL

2.8

±4.7

0.002

-1.2

±4.0

0.08

<0.001

HOMA-IR

0.6

±1.1

0.002

-0.3

±1.0

0.08

<0.001

HOMA-B

10.7

±18.2

0.001

-4.4

±15.0

0.09

<0.001

QUICKI

-0.01

±0.03

0.01

0.00009

±0.01

0.83

0.01

Total T, ng/mL

0.1

±0.3

0.03

-0.2

±0.4

0.19

0.01

SHBG, nmol/L

-1.3

±3.5

0.03

3.9

±6.2

0.001

<0.001

FAI

0.02

±0.03

0.004

-0.03

±0.04

0.001

<0.001

mFG scores

-0.2

±0.8

0.08

-1.1

±0.9

<0.001

<0.001

Free T, pg/mL

-0.4

±1.9

0.19

-0.7

±2.3

0.06

0.54

DHEAS, µg/mL

-0.1

±1.4

0.49

-0.7

±2.3

<0.001

0.96

Triglycerides, mg/dL

10.3

±24.5

0.01

-13.3

±62.2

0.21

0.04

VLDL, mg/dL

2.0

±4.9

0.01

-2.7

±12.4

0.21

0.04

Total cholesterol, mg/dL

-0.2

±14.8

0.93

3.4

±19.4

0.30

0.38

LDL, mg/dL

-2.4

±12.8

0.27

4.9

±24.5

0.24

0.12

HDL, mg/dL

0.2

±4.0

0.81

1.2

±5.3

0.21

0.38

Total/HDL ratio

-0.01

±0.3

0.80

0.005

±0.5

0.94

0.86

hs-CRP, mg/mL

0.2

±2.9

0.82

-0.2

±3.6

0.82

0.75

NO, µmol/L

2.7

±20.2

0.44

3.0

±5.1

0.001

0.93

TAC, mmol/L

33.4

±251.6

0.43

30.0

±68.0

0.01

0.93

GSH, µmol/L

22.7

±157.8

0.75

96.0

±102.2

<0.001

0.04

MDA, µmol/L

0.8

±2.3

0.05

-0.7

±0.8

<0.001

0.001

*Paired-samples t test; ** Time x group interaction (one way repeated measures ANOVA)

FPG: fasting plasma glucose; HOMA-IR: homeostasis model of assessment-insulin resistance; HOMA-B: homeostasis model of assessment ß-cell; QUICKI: quantitative insulin sensitivity check index; Total T; total testosterone; SHBG: sex hormone-binding globulin; FAI: free androgen index; mFG: modified Ferriman-Gallwey score; DHEAS: dehydroepiandrosterone sulfate; VLDL: very low-density lipoprotein; HDL: high-density lipoprotein; hs-CRP: high sensitivity C-reactive protein; NO: nitric oxide; TAC: total antioxidant capacity; GSH: total glutathione; and MDA: malondialdehyde.

COMMENTARY

These researchers attempted to investigate the effect of a soy isoflavone supplement on multiple aspects (metabolic, hormonal, inflammatory) of a complex syndrome called PCOS. This small study had a rigorous double-blind, randomized, placebo-controlled design but there are a few methodological issues that could have affected the study results. Although the participants were advised not to change their physical activity or take nutritional supplements, soy intake was not assessed or monitored in the study.

Conditions such as impaired glucose tolerance and diabetes were in the exclusion criteria but hyperlipidemia was not. Medication use also was not discussed, but oral contraceptives frequently used to treat PCOS as well as other medications could affect biomarkers under study. In addition, the use of complementary approaches to health, such as acupuncture or Ayurveda, could affect study outcomes. All of these flaws in the study design could have affected the study results.

Previous studies investigating the effect of soy isoflavones on the metabolic and hormonal health of women with PCOS have shown mixed results. A pilot study with 12 obese hyperinsulinemic and dyslipidemic women with PCOS found that giving 36 mg/d of genistein for six months resulted in significant changes in total cholesterol, LDL, and LD-HDL ratios, but there were no significant changes in other lipids, anthropometric features, hormonal milieu, or menstrual regularity.9 A quasi-randomized trial (n = 146) found that 18 mg of genistein given to women with PCOS twice a day for three months improved luteinizing hormone (LH), triglycerides, LDL, DHEAS, and testosterone but not HDL and FSH.10 The results from the Jamilian and Asemi study also were mixed. For instance, when evaluating the effect of the soy supplement on lipids, only triglycerides and VLDL showed improvement while other lipids did not. This also was the case for biomarkers related to insulin resistance and hormonal status. The flaws in study design or the small sample size may account for these mixed results. Soy supplementation may have more of an effect on women with hyperlipidemia and insulin resistance. In addition, whole soy foods could have more beneficial health effects than soy isoflavone supplements, as other studies suggest.

It may be helpful to look at the results in the context of clinical care and their usefulness in caring for women with PCOS. The hormonal milieu of PCOS usually is monitored using free T, ideally performed with an equilibrium dialysis technique, progesterone, serum 17-hydroxyprogesterone, anti-Müllerian hormone; in some cases DHEAS is assessed.1 In this study, free T levels did not show improvement, but DHEAS improved significantly and the other laboratory tests were not obtained. The other hormonal tests performed in this study were not recommended in the 2015 clinical guidelines. The oral glucose tolerance test or fasting glucose to insulin ratio commonly is used to screen for impaired glucose tolerance in the clinical setting.11 The HOMA-IR has had questionable accuracy in studies, so it probably is not a useful marker. In this study, fasting glucose did not improve but insulin and HOMA-IR did improve. In a double-blind, placebo-controlled trial (n = 165), soy isoflavone supplements failed to improve insulin sensitivity among Chinese women with impaired glucose regulation.12 The results from this study and others make it difficult to determine if supplementation with soy isoflavones improves insulin resistance or the hormonal milieu of women with PCOS.

The underlying role of inflammation in the pathophysiology of PCOS is not completely understood, and consequently which biomarkers should be studied has not been determined. In a study that explored the relationship between chronic low-grade inflammation and PCOS, the inflammatory mediators investigated were CRP, IL6, other IL, TNFα, MCP1, sE-selectin, and sICAM.13 Many of the same biomarkers for inflammation also were used in large studies, such as Framingham, to determine risk of diabetes.14 In the Jamilian and Asemi study, there was no significant improvement in CRP, a marker of inflammation, but there was improvement in the antioxidant GSH and a marker of oxidative stress, MDA. Many of the markers of inflammation used in other studies were not included in this study, and these biomarkers are probably most useful within the context of exploratory research.

In summary, this study, which examined the effect of soy isoflavones on the metabolic status of women with PCOS, had mixed results. At this time, supplementation with soy isoflavones does not appear to have a role in the clinical care of women with PCOS. Larger longitudinal studies that include women with PCOS who have metabolic disorders, that examine soy in the diet, and that employ labs recommended for clinical monitoring could shed more light on the effects of soy. Further study of the effect of soy isoflavones on the inflammatory process in PCOS should include replicating the use of biomarkers consistently used in large studies investigating metabolic disorders. This was an ambitious, rigorously designed study that was conducted in a low-resource setting with limited access to costly laboratory tests and research funding. Future studies should build on the efforts of these investigators who have sought to improve our understanding of the effect of soy isoflavones on the health of women with PCOS.

REFERENCES

  1. Goodman NF, Cobin RH, Futterweit W, et al. American Association of Clinical Endocrinologists, American College of Endocrinology, and Androgen Excess and PCOS Society Disease State Clinical Review: Guide to the Best Practices in the Evaluation and Treatment of Polycystic Ovary Syndrome--Part 1. Endocr Pract 2015;21:1291-1300.
  2. Bozdag G, Mumusoglu S, Zengin D, et al. The prevalence and phenotypic features of polycystic ovary syndrome: A systematic review and meta-analysis. Hum Reprod 2016; Sep 22 [Epub ahead of print].
  3. Rotterdam ESHRE/ASRM-Sponsored PCOS consensus workshop group. Revised 2003 consensus on diagnostic criteria and long-term health risks related to polycystic ovary syndrome (PCOS). Hum Reprod 2004;19:41-47.
  4. Spritzer PM, Lecke SB, Satler F, Morsch DM. Adipose tissue dysfunction, adipokines, and low-grade chronic inflammation in polycystic ovary syndrome. Reproduction 2015;149:R219-R227.
  5. Ojeda-Ojeda M, Murri M, Insenser M, Escobar-Morreale HF. Mediators of low-grade chronic inflammation in polycystic ovary syndrome (PCOS). Curr Pharm Des 2013;19:5775-5791.
  6. Behloul N, Wu G. Genistein: A promising therapeutic agent for obesity and diabetes treatment. Eur J Pharmacol 2013;698:31-38.
  7. Tokede OA, Onabanjo TA, Yansane A, et al. Soya products and serum lipids: A meta-analysis of randomised controlled trials. Br J Nutr 2015;114:831-843.
  8. Fang K, Dong H, Wang D, et al. Soy isoflavones and glucose metabolism in menopausal women: A systematic review and meta-analysis of randomized controlled trials. Mol Nutr Food Res 2016;60:1602-1614.
  9. Romualdi D, Costantini B, Campagna G, et al. Is there a role for soy isoflavones in the therapeutic approach to polycystic ovary syndrome? Results from a pilot study. Fertil Steril 2008;90:1826-1833.
  10. Khani B, Mehrabian F, Khalesi E, Eshraghi A. Effect of soy phytoestrogen on metabolic and hormonal disturbance of women with polycystic ovary syndrome. J Res Med Sci 2011;16:297-302.
  11. Goodman NF, Cobin RH, Futterweit W, et al. American Association of Clinical Endocrinologists, American College of Endocrinology, and Androgen Excess and PCOS Society Disease State Clinical Review: Guide to the Best Practices in the Evaluation and Treatment of Polycystic Ovary Syndrome - Part 2. Endocr Pract 2015;21:1415-1426.
  12. Ye YB, Chen AL, Lu W, et al. Daidzein and genistein fail to improve glycemic control and insulin sensitivity in Chinese women with impaired glucose regulation: A double-blind, randomized, placebo-controlled trial. Mol Nutr Food Res 2015;59:240-249.
  13. Spritzer PM, Lecke SB, Satler F, Morsch DM. Adipose tissue dysfunction, adipokines, and low-grade chronic inflammation in polycystic ovary syndrome. Reproduction 2015;149:R219-R227.
  14. Dallmeier D, Larson MG, Vasan RS, et al. Metabolic syndrome and inflammatory biomarkers: A community-based cross-sectional study at the Framingham Heart Study. Diabetol Metab Syndr 2012;4:28.