Skip to main content
NCBI home page
Journal of Diabetes and Metabolic Disorders logoLink to Journal of Diabetes and Metabolic Disorders
. 2024 May 21;23(2):1573–1587. doi: 10.1007/s40200-024-01441-3

The effects of ketogenic diet on metabolic and hormonal parameters in patients with polycystic ovary syndrome: a systematic review and meta-analysis of clinical trials

Niloofarsadaat Eshaghhosseiny 1, Mohammad Ahmadi 2, Bahareh Izadi 3, Mohebat Vali 4, Maryam Akbari 3, Isaac Azari 5,✉, Hamed Akbari 6,✉
PMCID: PMC11599545  PMID: 39610476

Abstract

Purpose

In recent years, using the ketogenic diet (KD) as a potential intervention for polycystic ovary syndrome (PCOS) has gained attention. Therefore, this study aimed to conduct a meta-analysis to determine the effects of KD on sexual hormones, glycemic and lipid parameters in women diagnosed with PCOS.

Methods

A comprehensive literature search was performed using online databases such as Medline/PubMed, Scopus, Web of Science (ISI), Embase, and the Cochrane Library and clinical trials were selected based on the inclusion criteria. Data extraction and quality assessment were conducted independently by two investigators using appropriate tools. The effects of a KD on metabolic biomarkers and hormonal parameters were pooled using a random-effects model and were considered as the weighted mean difference (WMD) with corresponding 95% confidence intervals (CIs). Heterogeneity across studies was assessed using Cochran's Q test and the I-square test.

Results

Ten studies including 408 women were analyzed in this analysis. Findings showed that KD significantly decreased triglycerides levels (WMD = -44.03 mg/dL; 95% CI, -56.29, -31.76), total cholesterol (-18.95 mg/dL; -29.06, -8.83), and low-density lipoprotein cholesterol (LDL) (-18.11 mg/dL; -29.56, -6.67) compared to the control groups. KD also led to a notable reduction in fasting glucose (-10.30 mg/dL; -14.10, -6.50) and HOMA-IR (-1.93; -3.66, -0.19). Also, this diet led to a significant decrease in luteinizing hormone (LH) levels (-3.75 mIU/mL; -3.84, -3.65) and total testosterone levels (-7.71 ng/dL); -12.08, -3.35), while follicle-stimulating hormone (FSH) increased (0.43 mIU/mL; 0.29, 0.57).

Conclusion

The KD demonstrated promising outcomes in improving metabolic and hormonal parameters in women diagnosed with PCOS.

Keywords: Hormonal parameters, Ketogenic diet, Metabolic biomarkers, Polycystic ovary syndrome, Systematic review

Introduction

Polycystic ovary syndrome (PCOS) is known as one of the most common endocrine disorders among women of reproductive age, with a reported prevalence of 6%-15% based on different diagnostic criteria. [1]. A previous study in Iran showed that according to different criteria, the prevalence of PCOS in adolescent girls aged 14–19 is between 0.7% and 4.2% [2]. Another study showed that the prevalence of PCOS in Iran is comparable to the global average, and estimates indicate that approximately 7–15% of women of reproductive age may be affected by this disease [3]. In addition, PCOS is considered a complex condition [4].

Weight loss can help to improve PCOS. Even a moderate weight loss of 5%–10% can improve ovulation function and increase pregnancy rates by lowering insulin and free testosterone levels. Lifestyle modifications, such as calorie restriction and exercise, are considered the primary treatment for PCOS [5, 6]. Although dietary macronutrient composition is essential, caloric restriction appears to be the most important factor in achieving weight loss and improving metabolic function in women with PCOS [5, 6]. However, there is limited research on macronutrient modification as a therapeutic approach [7–9]. The effect of diet composition on reproductive and metabolic outcomes in women with PCOS remains a topic of debate. The KD is a nutritional protocol in which carbohydrate intake is restricted to less than 30 g per day or 5% of the total energy intake, with a relative increase in the ratio of protein and fat [10–12]. Some early studies [13–15] have suggested that KD may improve insulin resistance, weight loss, and other metabolic parameters in women with PCOS. It has also been shown in a study that after 45 days of intervention with a ketogenic diet (KD) among women with PCOS, a significant improvement in the level of reproductive hormones was observed [16]. In addition, a study in PCOS rats showed that KD treatment improved ovarian function, especially in rats treated with KD for three weeks [17].

Nevertheless, a complete understanding of the effects of KD on PCOS outcomes underscores the need for comprehensive studies. Using food as medicine has significant social and economic implications for healthcare costs and overall quality of life. We hypothesized that KD may improve sexual hormones, lipid, and glycemic profiles. This study can help healthcare providers and patients understand the potential benefits and risks of KD as a dietary intervention for PCOS. Therefore, the present study aimed to conduct a systematic review and meta-analysis to determine the effects of KD on some sexual hormones and lipid and glycemic parameters in women with PCOS.

Methods and materials

The Preferred Reporting Items for Systematic Reviews and Meta-analyses (PRISMA) checklist was utilized to guide the conduct and reporting of this research.

Data sources and search strategy

A comprehensive systematic literature search was performed to identify studies investigating the effects of a KD on metabolic and hormonal parameters in patients with PCOS. The search was conducted in Medline/PubMed, Scopus, Web of Science (ISI), Embase, and the Cochrane Library. The search was conducted up to March 28th, 2024. To capture all relevant randomized controlled trials (RCTs), a combination of the following keywords and MESH terms was used: "diet, ketogenic" OR "ketogenic diets" OR "ketogenic*" OR "ketogenic diet" AND "polycystic ovary syndrome" OR "ovary syndrome, polycystic" OR "PCOS" AND "metabolic parameters" OR "glycemic control" OR "blood sugar" OR "lipid profiles" OR "hyperlipidemia" OR "lipid metabolism" OR "triglycerides (TG)" OR "triglycerides" OR "high-density lipoproteins (HDL)" OR "low-density lipoprotein (LDL)" OR "total cholesterol (TC)" OR "hormonal parameters" OR "luteinizing hormone (LH)" OR "follicle-stimulating hormone (FSH)" OR "testosterone" OR "sex hormone binding globulin (SHBG)". Additionally, the reference lists of previous reviews and selected articles were manually screened to identify further relevant publications. This process was conducted independently by two investigators using EndNote software.

Study selection and selection criteria

Articles were selected based on the following PICOS criteria: participants (obese/overweight PCOS patients), intervention (KD), comparison (any comparisons), outcomes (glycemic control, lipid profile, hormonal parameters), and study design (clinical trials).

Inclusion Criteria: Clinical trials reporting sufficient data for direct extraction or calculation of mean (SD) changes in the effects of KDs on outcomes, including glycemic control (glucose, HOMA-IR), lipid profile (TG, TC, LDL, HDL), and hormonal parameters (LH, FSH, testosterone, SHBG), among obese or overweight patients with PCOS. Additionally, articles that published in English language were included.

Exclusion Criteria: Animal studies, observational studies, letters, commentaries, editorials, review articles, duplicate studies, and studies lacking sufficient data were excluded from the meta-analysis.

Data extraction and quality assessment

Two independent investigators (BI and MV) extracted data from the included articles using an Excel spreadsheet. Discrepancies between the two investigators were resolved by consulting the census or through discussion with a third author (HA). The extracted information includes the name of the author, year of publication, trial design, study location, basic characteristics of the study populations such as gender, mean age, and number of patients. It also includes details about the type and duration of intervention used in the study, as well as the mean (SD) change observed in the outcomes studied. Using the Cochrane Collaboration Risk of Bias Tool, the quality of the included RCTs was assessed across specific domains, including random sequence generation, allocation concealment, blinding, incomplete outcome data, selective reporting, and other potential sources of bias [18]. RCTs with high-risk items or more than two unclear domains were considered low quality. Regarding before-after clinical trials, the methodological risk of bias was evaluated using the National Institutes of Health (NIH) tool [19]. The scoring of the before-after trials was categorized as poor (< 50%), fair (50–75%), and good (≥ 75%) quality. The methodological quality assessment findings of the included trials are presented in Table 1.

Table 1.

The results of methodological quality assessment for before–after trials and RCT; ketogenic diet as a potential intervention for polycystic ovary syndrome

A) The National Institutes of Health (NIH) quality assessment tool for Pre/Post included trials
Criteria

Cincione et al. (2021)

[20]

Avolio et al. (2020)

[21]

Metropoulos et al. (2005)

[22]

Jian et al. (2022)

[23]

Magagnini et al. (2022)

[24]

Paoli et al. (2020)

[25]

 Khalid et al. (2024)[26]
1. Was the study question or objective clearly stated? Yes Yes Yes Yes Yes Yes Yes
2. Were eligibility/selection criteria for the study population prespecified and clearly described? Yes Yes Yes Yes Yes Yes Yes
3. Were the participants in the study representative of those who would be eligible for the test/service/intervention in the general or clinical population of interest? Yes Yes NO Yes Yes Yes Yes
4. Were all eligible participants who met the prespecified entry criteria enrolled? Yes Yes Yes CD CD No CD
5. Was the sample size sufficiently large to provide confidence in the findings? Yes Yes NO Yes CD Yes Yes
6. Was the test/service/intervention clearly described and delivered consistently across the study population? Yes Yes Yes CD Yes Yes Yes
7. Were the outcome measures prespecified, clearly defined, valid, reliable, and assessed consistently across all study participants? Yes Yes Yes Yes Yes No Yes
8. Were the people assessing the outcomes blinded to the participants' exposures/interventions? No CD CD No CD No CD
9. Was the loss to follow-up after baseline 20% or less? Were those lost to follow-up accounted for in the analysis? Yes Yes NO Yes NA Yes Yes
10. Did the statistical methods examine changes in outcome measures from before to after the intervention? Were statistical tests done that provided p values for the pre-to-post changes? Yes Yes Yes Yes CD Yes CD
11. Were outcome measures of interest taken multiple times before and multiple times after the intervention (i.e., did they use an interrupted time-series design)? Yes No No Yes No No No
12. If the intervention was conducted at a group level (e.g., a whole hospital, a community, etc.), did the statistical analysis take into account the use of individual-level data to determine effects at the group level? NA NA NA NA NA NA NA
Final quality rating Good Good Fair Fair Poor Fair Fair
Note. CD = cannot determine; NA = not applicable; NIH = National Institute of Health
B) the Cochrane Collaboration Risk of Bias tool for quality assessment tool for included RCTs
First Author Random sequence generation (selection bias) Allocation concealment (selection bias) Blinding of participants and personnel (performance bias) Blinding of outcome assessment (detection bias) Incomplete outcome data addressed (attrition bias) Selective reporting (reporting bias) other sources of bias (e.g., bias of study design, trial stopped early, extreme baseline imbalance, and fraudulent trial)
Cincione et al. (2023)[13] Low risk Low risk Unclear Unclear Unclear Low risk Unclear
Li et al. (2021) [14] Low risk Low risk High risk Unclear Unclear Low risk Unclear
Pandurevic et al. (2023)[27] Low risk Low risk High risk Unclear Low risk Low risk Low risk
Open in a new tab

Data analysis and statistical methods

The effects of KD were examined by pooling the mean (SD) change from the trials using a random-effects model. The effect sizes were reported as weighted mean difference (WMD) with corresponding 95% confidence intervals (CIs). For trials that did not provide the mean changes and their corresponding (SDs), the following formula was used to calculate them: [meanAfter—meanBefore] for the mean change, and [√ ([SDAfter2 + SDBefore2]—[2 × R × SDAfter × SDBefore])] for the corresponding SD. In this calculation, a correlation coefficient of 0.8 was assumed for the value of "R". Inter-study heterogeneity was assessed using Cochran's Q test and the I-square test. Significance for heterogeneity was set at P < 0.1, with an I2-value > 50% indicating significant heterogeneity across trials. Subgroup analyses were conducted to investigate potential sources of heterogeneity based on moderator variables. Additionally, sensitivity analyses were performed to evaluate the robustness of the pooled findings after removing each trial. To assess the presence of publication bias among the included trials, Egger's linear regression was applied. All statistical analyses were performed using STATA (Stata Corp., College Station, Texas, USA).

Results

Study selection and study characteristics

Figure 1 illustrates the step-by-step process of searching for and selecting relevant trials. The initial literature search yielded 381 reports, of which 257 were duplicates and irrelevant studies were excluded. Following the screening of titles and abstracts, the full texts of 21 trials were assessed for eligibility according to our inclusion criteria. After careful examination of the full literature, a total of 10 trials [13, 14, 20–27], involving 408 obese or overweight women with PCOS were identified.

Fig. 1.

Fig. 1

Open in a new tab

Study selection of PRISMA flowchart

Among these trials, 3 had RCT design [13, 14, 27], while the others utilized a before-after design [20–26]. The duration of interventions ranged from 6 to 24 weeks. Six trials were conducted in Italy, two in China, and one in the USA and Pakistan. Further details of the eligible trials are presented in Table 2.

Table 2.

Main characteristic of included studies; ketogenic diet as a potential intervention for polycystic ovary syndrome

Authors Country Patients Total sample size Study design Interventions Duration Outcomes
Cincione et al. (2023) [13] Italy PCOS patients with overweight or obesity 144 RCT The very-low-calorie ketogenic diet in this study included the maximum intake of carbohydrates was set at 30 g per day and the total caloric intake per day was established at around 600 kcal 6 weeks Glucose, HOMA-IR, LH, FSH, total testosterone, SHBG,
Li et al. (2021) [14] China PCOS and liver dysfunction who were obese 18 RCT The ketogenic diet was composed of approximately 5%–10% of energy derived from carbohydrates (≤ 50 g/day), 18%–27% from protein, and 70%–75% from fat 12 weeks Glucose, LH, FSH, total testosterone, TG, TC, LDL, HDL
Pandurevic et al. (2023) [27] Italy obese women with PCOS 30 RCT The low carbohydrate ketogenic diet in this study included consumption of less than 50g of carbohydrates per day, primarily from vegetables, along with a daily intake of 10g of olive oil 16 weeks Glucose, HOMA-IR, total testosterone, SHBG, TG, TC, LDL, HDL
Cincione et al. (2021) [20] Italy obese women with PCOS 17 Before-after trial The ketogenic diet followed with daily intake of carbohydrates was limited to 30g 6 weeks Glucose, HOMA-IR, LH, FSH, total testosterone, SHBG, TG, TC, LDL, HDL
Avolio et al. (2020) [21] Italy PCOS women, ages 18–38 years 57 Before-after trial The low carbohydrate ketogenic diet involved limiting the daily carbohydrate intake to less than 20g 24 weeks HOMA-IR
Mavropoulos et al. (2005) [22] USA Overweight and obese women with PCOS 5 Before-after trial The low carbohydrate ketogenic diet involved consuming fewer than 20 g of carbohydrates per day 24 weeks Glucose, total testosterone, TG, TC, LDL, HDL
Jian et al. (2022) [23] China Overweight and obese women with PCOS 51 Before-after trial The ketogenic diet involved a daily intake of 50 g or less of carbohydrates 4 weeks Glucose, TG, TC, LDL, HDL
Magagnini et al. (2022) [24] Italy Obese women with PCOS 25 Before-after trial The ketogenic diet consists of approximately 18 g of protein, 4 g of carbohydrates, and 3 to 4 g of fat 12 weeks HOMA-IR, SHBG
Paoli et al. (2020) [25] Italy Overweight women with PCOS 14 Before-after trial The ketogenic diet is characterized by high protein content, with each portion containing 19 g of protein, and very low carbohydrate content, with each portion containing 3.5 g of carbohydrates 12 weeks Glucose, HOMA-IR, LH, FSH, total testosterone, SHBG, TG, TC, LDL, HDL
Khalid et al. (2024) [26] Pakistan Overweight women with PCOS 47 Before-after trial Ketogenic mediterranean with phytoextracts diet which is very low caloric ketogenic diet (1000 kcal) + phytoextracts for first 3 weeks, low caloric KD (1200 kcal) + phytoextracts for second 3 weeks and Mediterranean diet (1600 kcal) + phytoextracts for last 6 weeks 12 weeks Glucose, LH, FSH, TG, TC, LDL, HDL
Open in a new tab

RCT: Randomized controlled trial; PCOS: polycystic ovary syndrome; KD: ketogenic diet; HOMA-IR: Homeostatic Model Assessment; LH: Luteinizing hormone; FSH: Follicle-stimulating hormone

Effects of a KD on metabolic biomarkers

Six of the ten included trials reported the effects of KD on lipid profiles. The pooled findings demonstrated that a KD significantly decreased levels of TG [WMD = -44.03 mg/dL; 95% CI, -56.29, -31.76; p < 0.01; I2 = 82.73% (with seven trials)], TC [WMD = -18.95 mg/dL; 95% CI, -29.06, -8.83; p < 0.01; I2 = 97.33% (with seven trials)], and LDL [WMD = -18.11 mg/dL; 95% CI, -29.56, -6.67; p < 0.01; I2 = 92.15% (with seven trials)] among obese or overweight women with PCOS comparing to control groups. However, the levels of HDL did not show a significant increase [WMD = 3.51 mg/dL; 95% CI, -3.74, 10.76; p = 0.34; I2 = 97.65% (with seven trials)] (Fig. 2A-D).

Fig. 2.

Fig. 2

Fig. 2

Open in a new tab

A-F. The effects of a ketogenic diet on A)TG, B) TC, C) LDL, D) HDL, E) FG, and F) HOMA-IR levels in patients with PCOS

The pooled results, using a random-effects model, indicated that the KD appeared to be more effective in decreasing levels of fasting glucose [WMD = -10.30 mg/dL; 95% CI, -14.10, -6.50; p < 0.01; I2 = 91.97% (with eight trials)] and HOMA-IR [WMD = -1.93; 95% CI, -3.66, -0.19; p = 0.030; I2 = 99.61% (with six trials)] (Fig. 2E–F).

According to Table 3, due to significant heterogeneity observed across trials, subgroup analyses were conducted based on several variables including study design, country, and intervention duration. The results of the subgroup analysis indicated that in studies with an RCT design, there were no significant effects of a KD on metabolic biomarkers among patients with PCOS. Additionally, for TC, LDL, HDL and HOMA-IR levels in studies with an intervention duration of less than 10 weeks, no significant changes were observed.

Table 3.

Subgroup analyses results; ketogenic diet as a potential intervention for polycystic ovary syndrome

Outcomes Subgroups Non studies WMD 95% CI I2
(%)
TG (mg/dL) Study design RCT 2 -47.07 -113.62, 19.47 39.40
Before-after trial 5 -44.62 -57.64, -31.60 87.90
Country China 2 -54.95 -70.07, -39.82 0.00
Italy 3 -48.25 -74.10, -22.39 86.07
Pakistan 1 -34.75 -38.96, -30.54 -
USA 1 -28.60 -40.53, -16.66 -
Intervention duration  ≤ 10 2 -62.58 -78.19, -46.97 59.59
 > 10 5 -34.61 -38.33, -30.88 0.00
TC (mg/dL) Study design RCT 2 -13.19 -26.54, 0.16 0.00
Before-after trial 5 -20.51 -32.00, -9.02 98.19
Country China 2 -7.54 -14.18, -0.89 0.00
Italy 3 -27.70 -41.37, -14.03 95.47
Pakistan 1 -21.45 -23.04, -19.86 -
USA 1 4.40 -20.74, 29.54 -
Intervention duration  ≤ 10 2 -23.65 -56.02, 8.72 98.73
 > 10 5 -21.05 -25.46, -16.64 47.31
LDL (mg/dL) Study design RCT 2 -10.47 -21.71, 0.76 0.00
Before-after trial 5 -20.38 -33.76, -7.01 94.06
Country China 2 -6.55 -12.80, -0.29 0.000
Italy 3 -25.65 -39.23, -12.00 81.23
Pakistan 1 -32.17 -35.15, -29.19 -
USA 1 11.88 -9.95, 33.55 -
Intervention duration  ≤ 10 2 -20.29 -48.91, 8.31 96.65
 > 10 5 -17.46 -30.43, -4.48 85.70
HDL (mg/dL) Study design RCT 2 -1.22 -4.64, 2.20 0.00
Before-after trial 5 5.26 -3.90, 14.43 98.39
Country China 2 -3.94 -8.98, 1.09 63.99
Italy 3 7.48 -3.16, 18.12 95.26
Pakistan 1 9.45 7.78, 11.12 -
USA 1 -1.60 -8.67, 5.47 -
Intervention duration  ≤ 10 2 4.56 -15.81, 24.94 99.36
 > 10 5 3.24 -2.56, 9.05 89.09
Fasting glucouse (mg/dL) Study design RCT 3 -5.47 -15.55, 4.61 76.03
Before-after trial 5 -12.56 -16.96, -8.16 94.44
Country China 2 -69.46 -183.28, 45.35 98.35
Italy 4 -6.65 -10.63, -2.66 79.64
Pakistan 1 -9.08 -9.63, -8.53 -
USA 1 -18.20 -31.11, -5.28 -
Intervention duration  ≤ 10 3 -45.89 -88.35, -3.42 97.01
 > 10 5 -7.85 -10.35, -5.35 77.71
HOMA-IR Study design RCT 2 -0.14 -1.07, 0.79 0.00
Before-after trial 4 -2.30 -4.29, -0.31 99.76
Country China - - - -
Italy 6 -1.92 -3.65, -0.19 99.61
Pakistan - - - -
USA - - - -
Intervention duration  ≤ 10 2 -1.54 -3.47, 0.38 0.00
 > 10 4 -3.46 -5.47, -1.46 99.76
LH (mIU/mL) Study design RCT 2 -1.16 -6.06, 3.73 0.00
Before-after trial 3 -3.74 -3.84, -3.65 0.00
Country China 1 -0.1 -6.74, 6.54 -
Italy 3 -3.89 -4.35, -3.44 0.00
Pakistan 1 -3.74 -3.83, -3.64 -
USA - - - -
Intervention duration  ≤ 10 2 -4.82 -6.59, -3.06 0.00
 > 10 3 -3.74 -3.84, -3.65 0.00
FSH (mIU/mL) Study design RCT 2 0.14 -1.59, 1.88 42.49
Before-after trial 3 0.43 0.29, 0.56 40.49
Country China 1 -0.52 -1.91, 0.87 -
Italy 3 0.60 0.21, 1.01 27.54
Pakistan 1 0.39 0.35, 0.43 -
USA - - - -
Intervention duration  ≤ 10 2 1.02 0.36, 1.68 0.00
 > 10 3 0.39 0.35, 0.43 0.00
Total testosterone (ng/dL) Study design RCT 3 -8.94 -18.11, 0.21 0.00
Before-after trial 3 -6.73 -8.57, -4.88 91.83
Country China 1 -1.73 -5.74, 2.28 -
Italy 4 -9.73 -14.38, -5.07 84.59
Pakistan - - - -
USA 1 -3.80 -18.16, 10.56 -
Intervention duration  ≤ 10 2 -8.52 -13.78, -3.26 0.00
 > 10 4 -7.09 -12.61, -1.57 90.54
SHBG (nmol/L) Study design RCT 2 10.25 7.63, 12.87 0.00
Before-after trial 3 20.41 -5.52, 46.35 99.43
Country China - - - -
Italy 5 16.34 -1.15, 33.85 99.19
Pakistan - - - -
USA - - - -
Intervention duration  ≤ 10 2 11.25 0.17, 22.33 0.00
 > 10 3 19.47 -2.70, 41.65 99.59
Open in a new tab

WMD: weighted mean difference; RCT: randomized controlled trial

Sensitivity analyses demonstrated that the pooled results remained unchanged for metabolic biomarkers, including TG, TC, LDL, and fasting glucose levels after the removal of individual studies. However, the pooled results for HDL levels changed when excluding each trial conducted by Jian et al. [23] (WMD = 5.38, 95% CI: 0.11, 10.66). Regarding HOMA-IR levels, the sensitivity analyses showed that the pooled results changed after excluding each article conducted by Cincione et al. [20] (WMD = -1.63, 95% CI: -3.52, 0.25) and Avolio et al. [21] (WMD = -2.11, 95% CI: -4.51, 0.28) (Supplementary Fig. 1s A-F).

Effects of a KD on hormonal parameters

The effects of the KD on hormonal parameters were also examined. As demonstrated in Fig. 3A-B, the implementation of the KD resulted in a significant decrease in LH levels [WMD = -3.75 mIU/mL; 95% CI, -3.84, -3.65; p < 0.01; I2 = 00.0% (with five trials)] and total testosterone levels [WMD = -7.71 ng/dL; 95% CI, -12.08, -3.35; p < 0.01; I2 = 84.58% (with six trials)] compared to the control groups. Conversely, there was an increase in FSH levels [WMD = 0.43 mIU/mL; 95% CI, 0.29, 0.57; p < 0.01; I2 = 28.66% (with five trials)] following the consumption of a KD (Fig. 3C). However, the level of SHBG [WMD = 16.35 nmol/L; 95% CI, -1.16, 33.85; p = 0.07; I2 = 99.19% (with five trials)] did not significantly decrease as a result of the KD (Fig. 3D).

Fig. 3.

Fig. 3

Open in a new tab

A-D. The effects of a ketogenic diet on A) LH, B) total testosterone, C) FSH, and D) SHBG levels in patients with PCOS

Subgroup analysis revealed that among studies with an RCT design, KD did not have significant effects on hormonal parameters in patients with PCOS, except for SHBG levels. However, for SHBG levels in studies with an intervention duration of more than 10 weeks, significant changes were observed (Table 3).

Sensitivity analyses were conducted to assess the robustness of the pooled results. The results showed that the pooled results for LH, FSH, and total testosterone remained unchanged after removing each study. In terms of SHBG levels, the sensitivity analyses demonstrated that the pooled results changed after excluding the article conducted by Magagnini et al. [28] (WMD = 9.16, 95% CI: 7.27, 11.05) (Supplementary Fig. 2s A-D).

Publication bias

No evidence of publication bias was found for the following biomarkers and parameters, based on the findings of Egger linear regression analyses: TG (Egger Test's P = 0.32), TC (P = 0.61), LDL (P = 0.16), HDL (P = 0.81), fasting glucose (P = 0.45), HOMA-IR (P = 0.34), LH (P = 0.95), FSH (P = 0.54), total testosterone (P = 0.75), and SHBG levels (P = 0.61).

Discussion

This systematic review and meta-analysis aimed to investigate the effects of the KD on metabolic biomarkers and hormonal parameters in patients with PCOS. Overall, this study confirmed the researchers' initial hypothesis and showed that KD could reduce lipid profiles. The results of this study showed that this diet can significantly reduce metabolic biomarkers including TG, TC, and LDL, but it could not significantly increase HDL, but reducing the level of HOMA-IR seems to be more effective. There was significant heterogeneity in the studies, which examined the sources of heterogeneity, including study design, had no significant effect on metabolic biomarkers, and no significant changes were observed in TC, LDL, HDL, and HOMA-IR levels in studies with an intervention period of less than 10 weeks. In addition, this diet was able to significantly decrease hormonal parameters, including LH and total testosterone levels, and increase FSH, but it caused an insignificant decrease in SHBG levels. In the subgroup analysis, significant changes were observed in SHBG levels in studies with RCTs design.

PCOS is associated with obesity, weight changes, cardiovascular diseases, and changes in carbohydrate metabolism, such as IR and insulin secretion [29], which disrupt the production of sex hormones by hyperinsulinemia [30]. There is an urgent need for an anti-inflammatory dietary intervention to treat PCOS due to the important role of chronic inflammation in the pathogenesis of many chronic diseases and complications [31, 32].

The KD includes low carbohydrate, moderate protein, and high fat contents [33]. Carbohydrates are the main sources of energy in the body. When the body's intake of carbohydrates decreases by 50 g, insulin secretion in the body decreases, and the body enters a catabolic state. This is because the reduced reserves force the body to make certain metabolic changes [33, 34]. Gluconeogenesis is one of the metabolic changes caused by this diet. During this process, when the availability of glucose is further reduced, the endogenous production of glucose cannot meet the body's needs, and ketogenesis begins to provide an alternative source of energy in the form of ketone bodies. Ketone bodies replace glucose as their primary energy source. During ketogenesis, owing to low blood glucose feedback, the stimulus for insulin secretion is also low, which greatly reduces the stimulus for fat and glucose storage [33]. This may be due to the use of this diet, a decrease in fat factors, except HDL, and a decrease in fasting glucose levels. Other hormonal changes may contribute to the increased breakdown of fats, leading to fatty acid production [33]. In addition, the type of the KD has led to concerns regarding the effect of a low-carbohydrate diet on lipids, especially LDL. A recent systematic review study [35], contrary to our study, showed a small to neutral increase in LDL; however, similar to our study, a decrease in TG levels and an increase in HDL have been observed, especially in low-carbohydrate diet interventions. In a systematic review study [36], total cholesterol, LDL, and HDL decreased and TG increased, showing that the results of our study were slightly different. The difference in various studies regarding the effect of the KD on lipid factors can be due to the difference in the studied population and it seems that comorbidities and the condition of patients are highly affected by metabolic and hormonal factors. A previous study [37], it was also stated that overweight and obese populations with healthy metabolism have a more specific reaction to diet fat than people with unhealthy metabolism. Therefore, the condition of the examined patients and their metabolism will be effective in changing the lipid profile when consuming the KD.

Our study also showed the effects of diet on hormonal parameters. In fact, as mentioned, this diet can significantly reduce LH and total testosterone levels, as well as significantly decrease SHBG levels and increase FSH levels. A previous study [20], demonstrated the effect of the KD on the improvement of biochemical parameters such as LH, FSH, SHBG, and HOMA-IR. Also, another study [24] dealt with the effect of this diet on markers that predict metabolic disorder and ovulation, and in this study, it showed a significant decrease in the serum levels of anti-Müllerian hormone, but unlike our study, it showed a significant increase in progesterone and SHBG. However, this diet can be beneficial for the ovarian reserve and luteal function [24]. Various studies [9, 38, 39] have confirmed the effect of the KD on metabolic and endocrine effects by improving body weight, free testosterone percentage, LH/FSH ratio and fasting insulin levels. This leads to a decrease in androgen secretion and an increase in sex hormone-binding globulin, improving insulin sensitivity and normalizing endocrine function.

In previous studies [33, 40, 41], the short-term effects (up to 2 years) of the KD have been well reported and proven. We also observed that more changes in metabolic and hormonal biomarkers in studies with an intervention period of less than 10 weeks. These changes may not be significant for some biomarkers, but they show more changes than longer interventions. However, the long-term health consequences are not well-known owing to the limited number of studies.

Therefore, it seems that the KD can be effective in improving various clinical manifestations of PCOS and improve them. However, due to the side effects and obstacles, this diet should be used under the supervision of a specialist doctor and specific patients. Moreover, more solid evidence is needed before a specific nutritional approach to PCOS management is recommended. Before nutritional intervention (ideally individualized) is implemented for the clinical treatment of PCOS, a careful nutritional assessment is critical to determine the best strategy.

This study had both strengths and limitations that should not be overlooked. One of the limitations was the high heterogeneity, which we attempted to address by conducting subgroup analyses based on study design, country, and intervention duration variables. Additionally, we performed a sensitivity analysis to identify any studies that may have influenced the results because of high heterogeneity. In addition, the results of our study may have been biased by the inclusion of only English articles. Future studies should include articles published in other languages to reduce the risk of publication bias. The results of our study may not be generalizable to other countries or populations because most of the studies were related to Italy, followed by America and China, and only one study was related to Pakistan. Therefore, future studies should include data from a wider range of countries to improve the generalizability of our findings.

In conclusion, the KD demonstrated promising findings in improving metabolic and hormonal markers in women with PCOS. However, further research is needed to fully understand its long-term effects and to determine the best approach for individual patients. It is important to consult a specialist and incorporate proper exercise when considering this diet as a treatment method. Future studies should focus on the manner in which KD affects metabolic and hormonal biomarkers. Additionally, it is suggested that researchers explore how this diet can be used in combination with other methods, including the Mediterranean diet and exercise. Furthermore, investigating the optimal duration of intervention is recommended as an area for future studies.

Acknowledgements

The authors have no acknowledgments to declare.

Author contributions

NE, MA, BI, MV, MA, IA, and HA contributed equally to this systematic review. NE, IA, MA, HA conceived the study and designed the research. BI and MV conducted the literature search and data extraction. NE, MA, and MA analyzed the data. All authors participated in the interpretation of the results. NE, IA, MA, and HA drafted the manuscript, and all authors critically reviewed and approved the final version.

Funding

None.

Data availability

The data that support the findings of this study are available on request from the corresponding author.

Declarations

Conflict of interest

All authors declare that they have no conflicts of interest.

Sources of support

The authors did not receive support from any organization for the submitted work.

Footnotes

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Contributor Information

Isaac Azari, Email: Isaacazari@gmail.com.

Hamed Akbari, Email: hamedakbari6989@gmail.com, Email: hakbari2@ualberta.ca.

References

  • 1.Trikudanathan S. Polycystic ovarian syndrome. Medical Clinics. 2015;99(1):221–35. 10.1016/j.mcna.2014.09.003. [DOI] [PubMed] [Google Scholar]
  • 2.Pourhoseini SA, Babazadeh R, Mazlom SR. Prevalence of Polycystic Ovary Syndrome in Iranian Adolescent Girls Based on Adults and Adolescents’ Diagnostic Criteria in Mashhad City. J Reprod Infertil. 2022;23(4):288–95. 10.18502/jri.v23i4.10815. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Mehrabian F, Khani B, Kelishadi R, Ghanbari E. The prevalence of polycystic ovary syndrome in Iranian women based on different diagnostic criteria. Endokrynol Pol. 2011;62(3):238–42. [PubMed] [Google Scholar]
  • 4.Papadakis G, Kandaraki EA, Garidou A, Koutsaki M, Papalou O, Diamanti-Kandarakis E, et al. Tailoring treatment for PCOS phenotypes. Expert Rev Endocrinol Metab. 2021;16(1):9–18. 10.1080/17446651.2021.1865152. [DOI] [PubMed] [Google Scholar]
  • 5.Norman RJ, Dewailly D, Legro RS, Hickey TE. Polycystic ovary syndrome. The Lancet. 2007;370(9588):685–97. 10.1016/S0140-6736(07)61345-2. [DOI] [PubMed] [Google Scholar]
  • 6.Hoeger KM, Oberfield SE. Do women with PCOS have a unique predisposition to obesity? Fertil Steril. 2012;97(1):13–7. 10.1016/j.fertnstert.2011.11.026. [DOI] [PubMed] [Google Scholar]
  • 7.Stamets K, Taylor DS, Kunselman A, Demers LM, Pelkman CL, Legro RS. A randomized trial of the effects of two types of short-term hypocaloric diets on weight loss in women with polycystic ovary syndrome. Fertil Steril. 2004;81(3):630–7. 10.1016/j.fertnstert.2003.08.023. [DOI] [PubMed] [Google Scholar]
  • 8.Moran LJ, Noakes M, Clifton PM, Wittert GA, Williams G, Norman RJ. Short-term meal replacements followed by dietary macronutrient restriction enhance weight loss in polycystic ovary syndrome. Am J Clin Nutr. 2006;84(1):77–87. 10.1093/ajcn/84.1.77. [DOI] [PubMed] [Google Scholar]
  • 9.Tsagareli V, Noakes M, Norman RJ. Effect of a very-low-calorie diet on in vitro fertilization outcomes. Fertil Steril. 2006;86(1):227–9. 10.1016/j.fertnstert.2005.12.041. [DOI] [PubMed] [Google Scholar]
  • 10.Paoli A. Ketogenic diet for obesity: friend or foe? Int J Environ Res Public Health. 2014;11(2):2092–107. 10.3390/ijerph110202092. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 11.Paoli A, Bianco A, Grimaldi KA. The ketogenic diet and sport: a possible marriage? Exerc Sport Sci Rev. 2015;43(3):153–62. 10.1249/JES.0000000000000050. [DOI] [PubMed] [Google Scholar]
  • 12.Muscogiuri G, Barrea L, Laudisio D, Pugliese G, Salzano C, Savastano S, et al. The management of very low-calorie ketogenic diet in obesity outpatient clinic: a practical guide. J Transl Med. 2019;17:1–9. 10.1186/s12967-019-2104-z. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 13.Cincione IR, Graziadio C, Marino F, Vetrani C, Losavio F, Savastano S, et al. Short-time effects of ketogenic diet or modestly hypocaloric Mediterranean diet on overweight and obese women with polycystic ovary syndrome. J Endocrinol Invest. 2023;46(4):769–77. 10.1007/s40618-022-01943-y. [DOI] [PubMed] [Google Scholar]
  • 14.Li J, Bai WP, Jiang B, Bai LR, Gu B, Yan SX, et al. Ketogenic diet in women with polycystic ovary syndrome and liver dysfunction who are obese: a randomized, open-label, parallel-group, controlled pilot trial. Journal of Obstetrics and Gynaecology Research. 2021;47(3):1145–52. 10.1111/jog.14650. [DOI] [PubMed] [Google Scholar]
  • 15.Pandurevic S, Dionese P, Mancini I, Mitselman D, Magagnoli M, Teglia R, et al. Efficacy of very low calorie ketogenic diet in obese PCOS: a randomized controlled study. Endocr Abstr. 2022. 10.1530/endoabs.81.P193. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 16.Khalid K, Apparow S, Mushaddik IL, Anuar A, Rizvi SAA, Habib A. Effects of ketogenic diet on reproductive hormones in women with polycystic ovary syndrome. J Endocr Soc. 2023;7(10):bvad112. 10.1210/jendso/bvad112. [DOI] [PMC free article] [PubMed]
  • 17.Liu S, Yao Q, Li X, Wu H, Sun C, Bai W, et al. Effects of a ketogenic diet on reproductive and metabolic phenotypes in mice with polycystic ovary syndrome†. Biol Reprod. 2023;108(4):597–610. 10.1093/biolre/ioad004. [DOI] [PubMed] [Google Scholar]
  • 18.Higgins JPT, Altman DG, Gøtzsche PC, Jüni P, Moher D, Oxman AD, et al. The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. Bmj. 2011;343. 10.1136/bmj.d5928 [DOI] [PMC free article] [PubMed]
  • 19.Ma LL, Wang YY, Yang ZH, Huang D, Weng H, Zeng XT. Methodological quality (risk of bias) assessment tools for primary and secondary medical studies: what are they and which is better? Mil Med Res. 2020;7(1):7. 10.1186/s40779-020-00238-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 20.Cincione RI, Losavio F, Ciolli F, Valenzano A, Cibelli G, Messina G, et al. Effects of mixed of a ketogenic diet in overweight and obese women with polycystic ovary syndrome. Int J Environ Res Public Health. 2021;18(23): 12490. 10.3390/ijerph182312490. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 21.Avolio E, Ferraro S, Mocciaro R, De Pergola G, Buzzanca F, Venturella R, et al. The effect of ketogenic and low carb diet (Cyclic Alternation) on the ovarian morphology and insulin metabolism in polycystic ovary syndrome patients. Clin Obstet Gynecol. 2020;6:1–6. 10.15761/COGRM.1000307.
  • 22.Mavropoulos JC, Yancy WS, Hepburn J, Westman EC. The effects of a low-carbohydrate, ketogenic diet on the polycystic ovary syndrome: a pilot study. Nutr Metab. 2005;2(1):1–5. 10.1186/1743-7075-2-35. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 23.Jian LI, Wenpei BAI, Bo J, Feiran LIU, Yanrong C. Effects of ketogenic diet on vitamin D and glycolipid metabolism in overweight or obese patients with polycystic ovary syndrome. Journal of Clinical Medicine in Practice. 2022;26(4):14–7. 10.7619/jcmp.20220496. [Google Scholar]
  • 24.Magagnini MC, Condorelli RA, Cimino L, Cannarella R, Aversa A, Calogero AE, et al. Does the Ketogenic Diet Improve the Quality of Ovarian Function in Obese Women? Nutrients. 2022;14(19): 4147. 10.3390/nu14194147. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.Paoli A, Mancin L, Giacona MC, Bianco A, Caprio M. Effects of a ketogenic diet in overweight women with polycystic ovary syndrome. J Transl Med. 2020;18(1):1–11. 10.1186/s12967-020-02277-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 26.Khalid R, Jabeen S, Farooq S, Abid F. Influence of ketogenic mediterranean with phytoextracts diet on hyperandrogenism markers in married women with polycystic ovarian syndrome.
  • 27.Pandurevic S, Mancini I, Mitselman D, Magagnoli M, Teglia R, Fazzeri R, et al. Efficacy of very low-calorie ketogenic diet with the Pronokal® method in obese women with polycystic ovary syndrome: a 16-week randomized controlled trial. Endocri Connect. 2023;1(aop). 10.1530. [DOI] [PMC free article] [PubMed]
  • 28.Magagnini MC, Condorelli RA, Cimino L, Cannarella R, Aversa A, Calogero AE, et al. Does the Ketogenic Diet Improve the Quality of Ovarian Function in Obese Women? Nutrients. 2022;14(19). 10.3390/nu14194147 [DOI] [PMC free article] [PubMed]
  • 29.Muscogiuri G, Colao A, Orio F. Insulin-Mediated Diseases: Adrenal Mass and Polycystic Ovary Syndrome. Trends Endocrinol Metab. 2015;26(10):512–4. 10.1016/j.tem.2015.07.010. [DOI] [PubMed] [Google Scholar]
  • 30.Haffner SM, Katz MS, Stern MP, Dunn JF. The relationship of sex hormones to hyperinsulinemia and hyperglycemia. Metabolism. 1988;37(7):683–8. 10.1016/0026-0495(88)90091-1. [DOI] [PubMed] [Google Scholar]
  • 31.Barrea L, Frias-Toral E, Verde L, Ceriani F, Cucalón G, Garcia-Velasquez E, et al. PCOS and nutritional approaches: Differences between lean and obese phenotype. Metabol Open. 2021;12: 100123. 10.1016/j.metop.2021.100123. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Barrea L, Marzullo P, Muscogiuri G, Di Somma C, Scacchi M, Orio F, et al. Source and amount of carbohydrate in the diet and inflammation in women with polycystic ovary syndrome. Nutr Res Rev. 2018;31(2):291–301. 10.1017/S0954422418000136. [DOI] [PubMed] [Google Scholar]
  • 33.Masood W, Annamaraju P, Uppaluri KR. Ketogenic diet. StatPearls [Internet]: StatPearls Publishing; 2022. [PubMed]
  • 34.Mohorko N, Černelič-Bizjak M, Poklar-Vatovec T, Grom G, Kenig S, Petelin A, et al. Weight loss, improved physical performance, cognitive function, eating behavior, and metabolic profile in a 12-week ketogenic diet in obese adults. Nutr Res. 2019;62:64–77. 10.1016/j.nutres.2018.11.007. [DOI] [PubMed] [Google Scholar]
  • 35.Gjuladin-Hellon T, Davies IG, Penson P, Amiri BR. Effects of carbohydrate-restricted diets on low-density lipoprotein cholesterol levels in overweight and obese adults: a systematic review and meta-analysis. Nutr Rev. 2019;77(3):161–80. 10.1093/nutrit/nuy049. [DOI] [PubMed] [Google Scholar]
  • 36.Lu M, Wan Y, Yang B, Huggins CE, Li D. Effects of low-fat compared with high-fat diet on cardiometabolic indicators in people with overweight and obesity without overt metabolic disturbance: a systematic review and meta-analysis of randomised controlled trials. Br J Nutr. 2018;119(1):96–108. 10.1017/S0007114517002902. [DOI] [PubMed] [Google Scholar]
  • 37.Schwingshackl L, Hoffmann G. Comparison of effects of long-term low-fat vs high-fat diets on blood lipid levels in overweight or obese patients: a systematic review and meta-analysis. J Acad Nutr Diet. 2013;113(12):1640–61. 10.1016/j.jand.2013.07.010. [DOI] [PubMed] [Google Scholar]
  • 38.Valenzano A, Polito R, Trimigno V, Di Palma A, Moscatelli F, Corso G, et al. Effects of very low calorie ketogenic diet on the orexinergic system, visceral adipose tissue, and ROS production. Antioxidants. 2019;8(12): 643. 10.3390/antiox8120643. [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 39.Monda V, Polito R, Lovino A, Finaldi A, Valenzano A, Nigro E, et al. Short-Term Physiological Effects of a Very Low-Calorie Ketogenic Diet: Effects on Adiponectin Levels and Inflammatory States. Int J Mol Sci. 2020;21(9). 10.3390/ijms21093228. [DOI] [PMC free article] [PubMed]
  • 40.Ma S, Suzuki K. Keto-adaptation and endurance exercise capacity, fatigue recovery, and exercise-induced muscle and organ damage prevention: a narrative review. Sports (Basel). 2019;7(2). 10.3390/sports7020040. [DOI] [PMC free article] [PubMed]
  • 41.Oh R, Uppaluri KR. Low carbohydrate diet. StatPearls. Treasure Island: StatPearls Publishing; 2020. [PubMed]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Data Availability Statement

The data that support the findings of this study are available on request from the corresponding author.

Articles from Journal of Diabetes and Metabolic Disorders are provided here courtesy of Springer

RESOURCES