European Journal of Clinical Nutrition

https://doi.org/10.1038/s41430-019-0434-9

REVIEW ARTICLE

Effect of probiotics and synbiotics on selected anthropometric and

biochemical measures in women with polycystic ovary syndrome: a
systematic review and meta-analysis
1
Amir Hadi ●
Sajjad Moradi2,3 Abed Ghavami4 Saman Khalesi
● ●
5 ●
Marzieh Kafeshani6

Received: 1 November 2018 / Revised: 24 March 2019 / Accepted: 12 April 2019

© Springer Nature Limited 2019

Abstract
This study aimed to systematically review randomized clinical trials (RCTs) to clarify the effects of pro-/synbiotic
supplementation on anthropometric and biochemical measurements in women with polycystic ovary syndrome (PCOS).
PubMed, Scopus, ISI Web of Science, Cochrane Library, and Google Scholar were searched through September 2018. Eight
RCTs (nine treatment arms) were included. Pro-/synbiotic supplementation significantly reduced fasting blood sugar
1234567890();,:

(−2.52 mg/dl, 95% confidence interval (CI): −4.10 to −0.95), insulin (−2.27 µIU/mL, 95% CI: −3.40 to −1.14),
1234567890();,:

homeostasis model assessment for insulin resistance index (−0.69, 95% CI: −0.98 to −0.40), C-reactive protein (−1.69
Hedges’, 95% CI: −3.00 to −0.38), and total testosterone (−0.12 ng/mL, 95% CI: −0.17 to −0.08) in women with PCOS.
However, changes in the mean difference of weight and body mass index did not reach a statistically significant level. The
findings suggest that pro-/synbiotic supplementation may improve glucose homeostasis parameters, hormonal, and
inflammatory indices in women with PCOS.

Introduction

Polycystic ovary syndrome (PCOS) is a common gyne-

cological endocrine disorder with an estimated prevalence
Supplementary information The online version of this article (https:// of 6–26% worldwide, affecting women reproduction and
doi.org/10.1038/s41430-019-0434-9) contains supplementary
material, which is available to authorized users. metabolism and leading to other complications such as
diabetes, cardiovascular events, and breast cancers [1].
* Saman Khalesi Therefore, effective treatment and management of PCOS
S.khalesi@cqu.edu.au have great clinical and public health importance. Various
* Marzieh Kafeshani therapeutic methods such as lifestyle modification (diet-
Marzikafeshani@hlth.mui.ac.ir
ary modification and exercise) and pharmacological
1
therapies have shown benefits [2]. However, the high
Halal Research Center of IRI, FDA, Tehran, Iran
incidence of PCOS, and the possible side-effects of
2
Department of Community Nutrition, School of Nutrition and pharmacological therapies [3] call for safer and more
Food Science, Food Security Research Center, Isfahan University
effective interventions.
of Medical Sciences, Isfahan, Iran
3
The intestinal microflora and its relationship with health
Nutritional Sciences Department, School of Nutritional Sciences
have been studied increasingly in the last decades. The
and Food Technology, Kermanshah University of Medical
Sciences, Kermanshah, Iran DOGMA theory by Tremellen and Pearce [3] was the first
4 theory linking microflora to PCOS. Insulin resistance and
Department of Nutrition, School of Nutrition, Tabriz University of
Medical Sciences, Tabriz, Iran high testosterone production increase the risk of PCOS.
5 When the balance of gut flora is disrupted (e.g., poor diet
Physical Activity Research Group, Appleton Institute & School of
Health Medical and Applied Sciences, Central Queensland and lifestyle, or obesity) dysbiosis occurs [3]. As a result,
University, Brisbane, Australia the intestinal permeability increases, leading to ‘leaky gut’
6
School of Nutrition and Food Science, Food Security Research allowing bacteria, toxins, and antigens from the lumen to
Center, Isfahan University of Medical Sciences, Isfahan, Iran
 A. Hadi et al.

enter the bloodstream and trigger inflammation, leading to eight RCTs included in this study (a total of 540 women
insulin resistance and overproduction of testosterone [3]. with PCOS), four used probiotics [2, 6, 7, 10] and the
Probiotics (live microorganisms with potential benefits) remaining used synbiotics [1, 5, 8, 9]. Seven studies
and synbiotics (probiotics combined with prebiotics which [1, 2, 5, 7–10] compared the dietary intake pre- and post-
are food for probiotics) are proposed to improve the health intervention and reported no significant differences. A high
of gut microflora, reduce gut leakiness, inflammation, and level of compliance was reported in five trials [2, 5, 7–9].
insulin sensitivity [4]. Therefore, they may propose benefits According to Jadad scores, all studies were classified as
to prevent or manage PCOS. However, only a few clinical high-quality studies (score ≥ 3) (Table 1).
studies have investigated the effect of pro-/synbiotic on The overall effect of pro-/synbiotics on body weight,
PCOS management. The present study aimed to system- BMI, FBS, insulin, HOMA-IR index, CRP Hedges’, and
atically review the effect of pro-/synbiotic supplementation total testosterone compared to the placebo is presented in
on anthropometric indices, blood glucose level, hormonal, Table 2. A significant effect was observed for FBS
and inflammatory factors in women with PCOS. (−2.52 mg/dl, 95% CI: −4.10 to −0.95, I2 = 0.0%, n = 6),
insulin concentration (−2.27 µIU/mL; 95% CI: −3.40 to
−1.14, n = 6, I2 = 42.9%), HOMA-IR index (−0.69; 95%
Methods CI: −0.98 to −0.40, n = 6, I2 = 40.2%), CRP level (−1.69
Hedges’; 95% CI: −3.00 to −0.38, n = 5, I2 = 96.5%), and
PubMed, Scopus, Cochrane Library, ISI Web of Science, testosterone level (−0.12 ng/mL; 95% CI: −0.17 to −0.08,
and Google Scholar were searched from inception to n = 4, I2 = 43.4%). The meta-analysis effect for weight and
September, 2018. Studies were eligible if they were rando- BMI did not reach a statistically significant level (Table 1).
mized controlled trials; included women ≥18 years; and The subgroup analyses of age and supplement type are
supplemented live bacteria as probiotics or synbiotics (not in presented in Table 2. The subgroup of probiotics supple-
combination with other drugs or supplements). Information ments resulted in a significant reduction in body weight and
on studies characteristics and main outcomes including FBS. Synbiotics subgroup had a significant reduction in
weight, body mass index (BMI), fasting blood sugar (FBS), FBS, insulin, HOMA-IR index, and CRP level. Subgroups
serum insulin, homeostasis model assessment-estimated of age <30 years resulted in a significant reduction in FBS,
insulin resistance (HOMA-IR) index, C-reactive protein but a significant reduction in HOMA-IR index and total
(CRP), and total testosterone were extracted from each eli- testosterone subgroup was observed in the subgroup of age
gible trial. The methodology quality of included studies was ≥30 years (Table 2).
assessed using the Jadad checklist. Further details on the
methodology are presented in the Supplementary file.
Statistical analyses were carried out using the STATA, Discussion
version 11.2 (Stata Corp, College Station, TX). The overall
effect was calculated using the mean and standard devia- The findings from this study suggest that pro-/synbiotic
tion (SD) of changes and presented as mean difference and supplementation may be beneficial in reducing FBS,
95% confidence interval (CI). As CRP reported in different insulin, HOMA-IR, CRP, and total testosterone levels in
units across the studies, Hedges’ adjusted g was used to women with PCOS. There are several possible explana-
calculate the effect size of this variable. A random-effects tions for the observed effect. The dysbiosis of gut
model was used. The between-study heterogeneity was microbiota leads to an increase in the lipopolysaccharides
assessed using I2 index. Subgroups of intervention types (LPS) producing bacteria that can induce obesity, chronic
(pro-/synbiotic) and participants’ age (30< or ≥30 years) inflammation, and insulin resistance [1–3]. Probiotics may
were analyzed to explore heterogeneity. The one-out improve the dysbiosis, lowering LPS producing bacteria,
method was also used to assess the sensitivity of meta- and increasing beneficial bacteria. Probiotics may regulate
analysis to individual studies. A P < 0.05 was considered the secretion of some mediators of the brain–gut axis
statistically significant. including serotonin, ghrelin, and peptide YY hormones
[1, 2, 6, 8], reducing appetite, improving systemic energy
homeostasis, and luteinizing hormone secretion [5, 6].
Results Also, the production of SCFAs by probiotic bacteria can
enhance intestinal mucosal integrity, reducing the trans-
Overall, eight records [1, 2, 5–10] met the eligibility criteria location of microbial endotoxins, inflammation, and
(Supplementary Fig. 1). Basic characteristics of the inclu- insulin resistance [5]. Since a high level of circulating
ded trials and participants are summarized in Table 1. Of the insulin stimulates testosterone secretion, (which increases
Table 1 Characteristics of included studies
First author Country Total Mean Mean BMI Definition of RCT design Duration Intervention of experimental group Intervention of Outcomes Jadeda
(publication year) sample size age (year) (kg/m2) PCOS (blinding) (weeks) control group score

Shoaei [10] Iran 55 26 26 Rotterdam Parallel (Yes) 8 Lactobacillus casei 7 × 109 Placebo (MDX) FBS,insulin, HOMA- 4
CFU/g, Lactobacillus acidophilus IR, CRP
2 × 109 CFU/g,
Lactobacillus rhamnosus
1.5 × 109 CFU/g, Lactobacillus
bulgaricus 2 × 108 CFU/g,
Bifidobacterium breve
2 × 1010 CFU/g, Bifidobacterium longum
7 × 109
CFU/g and Streptococcus thermophiles
1.5 × 109 CFU/g
Ahmadi [2] Iran 60 25 26 Rotterdam Parallel (Yes) 12 Lactobacillus acidophilus Placebo (starch) Weight, BMI, FBS, 5
2 × 109 insulin, HOMA-IR
CFU/g, Lactobacillus casei 2 × 109 CFU/
g and
Bifidobacterium bifidum 2 × 109 CFU/g
Karamali [7] Iran 60 27 23 Rotterdam Parallel (Yes) 12 2 × 109 CFU/g of three species bacteria Placebo (starch) Weight, BMI, CRP, total 5
(Lactobacillus acidophilus, Lactobacillus testosterone
casei and Bifidobacterium)
Esmaeilinezhad [5] Iran 44 30 26 Rotterdam Parallel (Yes) 8 2 liter of pomegranate juice + 20 g of inulin 2 liter of Weight, BMI, FBS, 5
+ 2 × 108 CFU/g of five species bacteria pomegranate Juice insulin, HOMA-IR, total
(Lactobacillus casei, Lactobacillus testosterone
ramnosous, Lactobacillus plantroum, Bacillus
koagolans and Bacillus indicus)
Esmaeilinezhad [5] Iran 42 30 27 Rotterdam Parallel (Yes) 8 2 liter of water + 20 g of 2 liter of water Weight, BMI, FBS, 5
inulin + 2 × 108 CFU/g of five species insulin, HOMA-IR, total
bacteria (Lactobacillus casei, Lactobacillus testosterone
ramnosous, Lactobacillus plantroum, Bacillus
koagolans and Bacillus indicus)
Karimi [1] Iran 99 28 32 Rotterdam Parallel (Yes) 12 20 g of inulin + Lactobacillus acidophilus Placebo (MDX and FBS, insulin, HOMA- 5
3 × 1010 CFU/g, Lactobacillus casei starch) IR, CRP
3 × 109 CFU/g, Lactobacillus bulgaricus
5 × 108 CFU/g, Lactobacillus
rhamnosus 7 × 109 CFU/g, Bifidobacterium
longum
1 × 109 CFU/g,
Bifidobacterium breve 2 × 1010 CFU/g and
Streptococcus
thermophilus 3 × 108 CFU/g
Ghanei [6] Iran 60 29 26 Rotterdam Parallel (Yes) 12 1 × 109 CFU/g of four species bacteria Placebo (MDX) Weight, BMI, CRP 5
(Acidophilus, Lactobacillus plantarum,
Lactobacillus fermentum, and Lactobacillus
Effect of probiotics and synbiotics on selected anthropometric and biochemical measures in women with. . .

gasseri)
Nasri [8] Iran 60 25 27 Rotterdam Parallel (Yes) 12 0.8 g inulin + 2 × 109 CFU/g of three species Placebo (MDX) Weight, BMI, CRP, total 4
bacteria (Lactobacillus acidophilus, testosterone
Lactobacillus casei and Bifidobacterium)
Samimi [9] Iran 60 27 27 Rotterdam Parallel (Yes) 12 0.8 g inulin + 2 × 109 CFU/g of three species Placebo (starch) Weight, BMI, FBS, 5
bacteria (Lactobacillus acidophilus, insulin, HOMA-IR
Lactobacillus casei and Bifidobacterium)

RCT randomized controlled trial, BMI body mass index, PCOS polycystic ovarian syndrome, CFU colony-forming unit, MDX maltodextrin, FBS Fasting blood sugar, HOMA-IR Homeostasis
model of assessment for insulin resistance, CRP C-reactive protein
a
Based on Jadad checklist for methodology quality of interventional studies
 A. Hadi et al.

Table 2 Overall and subgroup Subgrouped by No. Effect size 95% CI P valuea I2 (%)
analysis results for the effect of of trials
pro-/synbiotic supplementation
on glycemic parameters, Weight (kg)
anthropometric indices, and Overall 7 −0.67 −1.43, 0.10 0.08 90.8
hormonal and inflammatory Age
markers ≥30 2 −1.50 −3.10, 0.10 0.06 0.0
<30 5 −0.53 −1.38, 0.32 0.22 93.9
Type of intervention
Probiotic 3 −1.3 −1.93, −0.13 0.02 90.6
Synbiotic 4 −0.08 −0.58, 0.39 0.73 17.9
BMI (kg/m2)
Overall 7 −0.23 −0.55, 0.08 0.14 92.6
Age
≥30 2 −0.55 −1.11, 0.01 0.05 0.0
<30 5 −0.17 −0.52, 0.19 0.35 95.1
Type of intervention
Probiotic 3 −0.36 −0.74, 0.02 0.06 91.3
Synbiotic 4 −0.03 −0.25, 0.19 0.77 43.2
FBS (mg/dl)
Overall 6 −2.52 −4.10, −0.95 0.002 0.0
Age
≥30 2 −1.87 −4.28, 0.55 0.13 0.0
<30 4 −3.01 −5.08, −0.93 <0.001 0.0
Type of intervention
Probiotic 2 −4.70 −8.43, −0.96 0.01 0.0
Synbiotic 4 −2.05 −3.79, −0.31 0.02 0.0
Insulin (µIU/mL)
Overall 6 −2.27 −3.40, −1.14 <0.001 42.9
Age
≥30 2 −2.29 −3.47, −1.12 <0.001 27.5
<30 4 −2.27 −4.38, −0.17 0.03 59.3
Type of intervention
Probiotic 2 −2.07 −4.76, 0.63 0.13 59.8
Synbiotic 4 −2.40 −3.81, −0.99 <0.001 49.7
HOMA-IR index
Overall 6 −0.69 −0.98, −0.40 <0.001 40.2
Age
≥30 2 −0.86 −1.16, −0.56 <0.001 0.0
<30 4 −0.56 −1.02, −0.11 0.01 52.1
Type of intervention
Probiotic 2 −0.47 −1.05, 0.12 0.11 52.9
Synbiotic 4 −0.82 −1.10, −0.53 <0.001 14.1
CRP Hedges’
Overall 5 −1.69 −3.00, −0.38 0.01 96.5
Type of intervention
Probiotic 3 −2.80 −5.75, 0.15 0.06 98.2
Synbiotic 2 −0.51 −0.83, −0.20 <0.001 0.0
Total testosterone (ng/mL)
Overall 4 −0.12 −0.17, −0.08 <0.001 43.4
Age
≥30 2 −0.13 −0.18, −0.08 <0.001 33.1
<30 2 −0.20 −0.48, 0.08 0.15 67.8

FBS Fasting blood sugar, HOMA-IR Homeostasis model of assessment for insulin resistance, CRP C-reactive
protein
a
Calculated using random-effects model

PCOS risk), the reduction in insulin secretion observed The current meta-analysis has limitations. The small
following pro-/synbiotic supplementation may lower numbers of included studies (especially at subgroup ana-
PCOS risk [5, 7, 8]. lysis), the high heterogeneity observed (although explored),
Effect of probiotics and synbiotics on selected anthropometric and biochemical measures in women with. . .

and the inclusion of all studies from one population, limit 4. Khalesi S, Johnson DW, Campbell K, Williams S, Fenning A,
the interpretation and generalization of the findings. Future Saluja S, et al. Effect of probiotics and synbiotics consumption on
serum concentrations of liver function test enzymes: a systematic
studies are required to confirm the beneficial effect of pro/
review and meta-analysis. Eur J Nutr. 2018;57:2037–53.
synbiotics in the treatment and management of PCOS. 5. Esmaeilinezhad Z, Babajafari S, Sohrabi Z, Eskandari M-H,
Amooee S, Barati-Boldaji R. Effect of synbiotic pomegranate
Funding source This research did not receive any specific grant from juice on glycemic, sex hormone profile and anthropometric indi-
funding agencies in the public, commercial, or not-for-profit sectors. ces in PCOS: a randomized, triple blind, controlled trial. Nutr
Metab Cardiovasc Dis. 2018;201–8.
6. Ghanei N, Rezaei N, Amiri GA, Zayeri F, Makki G, Nasseri E.
Compliance with ethical standards The probiotic supplementation reduced inflammation in polycystic
ovary syndrome: a randomized, double-blind, placebo-controlled
Conflict of interest The authors declare that they have no conflict of trial. J Funct Foods. 2018;42:306–11.
interest. 7. Karamali M, Eghbalpour S, Rajabi S, Jamilian M, Bahmani F,
Tajabadi-Ebrahimi M, et al. Effects of probiotic supplementation
Publisher’s note: Springer Nature remains neutral with regard to on hormonal profiles, biomarkers of inflammation and oxidative
jurisdictional claims in published maps and institutional affiliations. stress in women with polycystic ovary syndrome: a randomized,
double-blind, placebo-controlled trial. Arch Iran Med.
2018;21:1–7.
References 8. Nasri K, Jamilian M, Rahmani E, Bahmani F, Tajabadi-Ebrahimi
M, Asemi Z. The effects of synbiotic supplementation on hor-
1. Karimi E, Moini A, Yaseri M, Shirzad N, Sepidarkish M, monal status, biomarkers of inflammation and oxidative stress in
Hossein-Boroujerdi M, et al. Effects of synbiotic supplementation subjects with polycystic ovary syndrome: a randomized, double-
on metabolic parameters and apelin in women with polycystic blind, placebo-controlled trial. BMC Endocr Disord. 2018;18:21.
ovary syndrome: a randomised double-blind placebo-controlled 9. Samimi M, Dadkhah A, Kashani HH, Tajabadi-Ebrahimi M,
trial. Br J Nutr. 2018;119:398–406. https://doi.org/10.1017/ Hosseini ES, Asemi Z. The effects of synbiotic supplementation
s0007114517003920. e-pub ahead of print 2018/03/03 on metabolic status in women with polycystic ovary syndrome:
2. Ahmadi S, Jamilian M, Karamali M, Tajabadi-Ebrahimi M, Jafari a randomized double-blind clinical trial. Probiotics Antimicrob
P, Taghizadeh M, et al. Probiotic supplementation and the effects Proteins. 2018: 1–7.
on weight loss, glycaemia and lipid profiles in women with 10. Shoaei T, Heidari-Beni M, Tehrani HG, Feizi A, Esmaillzadeh
polycystic ovary syndrome: a randomized, double-blind, placebo- A, Askari G. Effects of probiotic supplementation on pan-
controlled trial. Hum Fertil. 2017;20:254–61. creatic beta-cell function and c-reactive protein in women with
3. Tremellen K, Pearce K. Dysbiosis of Gut Microbiota (DOGMA)-- polycystic ovary syndrome: a randomized double-blind pla-
a novel theory for the development of Polycystic Ovarian Syn- cebo-controlled clinical trial. Int J Prev Med. 2015;6:27.
drome. Med Hypotheses. 2012;79:104–12. https://doi.org/10. https://doi.org/10.4103/2008-7802.153866. e-pub ahead of
1016/j.mehy.2012.04.016. e-pub ahead of print 2012/05/01 print 2015/05/08