Mediterranean diet adherence and metabolic parameters in women with polycystic ovary syndrome: a retrospective cross-sectional study

Abstract

Objective

Polycystic ovary syndrome (PCOS) is a common endocrine disorder frequently accompanied by metabolic disturbances and low-grade inflammation. The Mediterranean diet (MD) has been associated with improved cardiometabolic health; however, evidence in women with PCOS remains limited and inconsistent.

Materials and methods

This retrospective cross-sectional study aimed to assess adherence to the MD and examine its relationship with biochemical and anthropometric parameters in women with PCOS. Data from 420 women aged 18–50 years who received medical nutrition therapy between December 2023 and June 2024 were analyzed. Adherence to the MD was evaluated using the 14-item Mediterranean Diet Adherence Screener (MEDAS). Anthropometric measurements and biochemical markers, including lipid profile, oral glucose tolerance test (OGTT) values, HbA1c, insulin and C-reactive protein (CRP), were obtained from medical records.

Results

The mean MEDAS score was 6.8 ± 2.3, indicating generally low adherence. MEDAS scores showed weak negative correlations with body mass index (BMI), body fat percentage, CRP, and 1-hour OGTT, and weak positive correlations with HDL-C, HbA1c, fasting glucose, 2-hour OGTT, and insulin (p < 0.05). CRP levels were moderately positively correlated with BMI and body fat percentage (r = 0.501 and r = 0.522, respectively; p < 0.001).

Conclusions

Low adherence to the Mediterranean diet was common among women with PCOS. Although several statistically significant associations were observed between MEDAS scores and metabolic parameters, these relationships were generally weak. Larger prospective studies are needed to clarify the potential role of the Mediterranean diet in modulating inflammation and metabolic health in PCOS.

Introduction

Polycystic ovary syndrome (PCOS) is one of the most common endocrine and metabolic disturbances, affecting approximately 5–26% of women of reproductive age worldwide, with prevalence varying according to diagnostic criteria [1, 2]. Its high frequency and multisystemic effects make it a significant public health concern, as it impacts reproductive, metabolic, and psychological health. Women with PCOS may experience profound reproductive impairments, disrupted metabolic functions (such as insulin resistance and type 2 diabetes), together with psychological disorders (notably depression and anxiety) and additional health issues spanning the period from adolescence to menopause [3–5].

PCOS is considered a heterogeneous and multifactorial disorder. Although the exact mechanisms responsible for its development have not been fully clarified, current evidence suggests that genetic predisposition interacts with environmental exposures, dietary patterns, and lifestyle behaviors in the pathogenesis of the syndrome [6]. These factors may influence hormonal regulation and metabolic pathways, contributing to the diverse clinical manifestations observed in women with PCOS. Consequently, women with PCOS may present with a wide range of symptoms, including menstrual irregularities, infertility, hirsutism, acne, and metabolic disturbances. The variability in clinical presentation highlights the complex nature of the disorder and the need for comprehensive approaches in its management [7].

In addition to its reproductive consequences, PCOS is strongly associated with several metabolic risk factors [8]. Insulin resistance is among the most frequently reported metabolic abnormalities in women with PCOS and plays a central role in the progression of the syndrome. Hyperinsulinemia resulting from insulin resistance may stimulate ovarian androgen production and further disrupt hormonal balance. Moreover, particularly abdominal obesity is commonly observed in women with PCOS and may cause metabolic dysfunction [9]. Dyslipidemia, impaired glucose tolerance, and metabolic syndrome are also frequently reported in this population. Therefore, PCOS is increasingly recognized not only as a reproductive disorder but also as a metabolic condition that may have long-term implications for overall health [10]. These metabolic disturbances may increase the risk of developing type 2 diabetes and cardiovascular disease later in life.

Lifestyle modification is considered an important component of PCOS management, and dietary habits play a key role in this process. Nutritional strategies that support weight management, improve insulin sensitivity, and reduce inflammation may contribute to the control of metabolic complications associated with the syndrome [11].

The Mediterranean diet is characterized by a high intake of plant-based foods such as vegetables, fruits, legumes, whole grains, and nuts, with olive oil serving as the primary source of dietary fat. Moderate consumption of fish and seafood, limited intake of red and processed meat, and moderate dairy consumption are also typical components of this dietary pattern. This dietary model provides high amounts of monounsaturated fatty acids, dietary fiber, and antioxidant compounds, which may contribute to improved metabolic health. Previous studies have shown that adherence to the Mediterranean diet may increase insulin sensitivity, reduce inflammatory processes, and improve cardiovascular risk factors and anthropometric indicators [12–14]. These potential benefits make the Mediterranean diet a promising nutritional approach for individuals with PCOS.

Despite the recognized metabolic benefits of the Mediterranean diet, evidence regarding its specific relationship with PCOS remains limited and inconsistent. Moreover, most existing studies have focused on general metabolic outcomes, while fewer studies have specifically examined dietary adherence patterns in women diagnosed with PCOS [15, 16]. Considering that traditional Mediterranean dietary patterns coexist with increasingly westernized eating habits in Türkiye, understanding how closely women with PCOS adhere to this dietary model and how this adherence relates to metabolic indicators represents an important gap in the literature.

Therefore, this study retrospectively aimed to evaluate Mediterranean diet adherence in women with PCOS and to examine its association with selected biochemical and anthropometric parameters. The findings of this study are expected to contribute to a better understanding of the potential role of dietary patterns in PCOS management and to support the development of more individualized nutritional strategies.

Methods

Study design and sample

This retrospective cross-sectional study was conducted at the PCOS outpatient clinic of Etlik Zübeyde Hanım Women’s Health Training and Research Hospital (Ankara, Türkiye). The study population consisted of women registered at the PCOS outpatient clinic between December 2023 and June 2024. A total of 420 women aged 18–50 years who were diagnosed with PCOS according to the Rotterdam criteria and met the inclusion criteria were included, representing all eligible patients who received medical nutrition therapy during the study period. Participants with missing data for any of the variables included in the analyses were excluded from the study, and analyses were conducted using complete-case data.

Because the sample size was determined by feasibility and availability inherent to the retrospective study design, an a priori sample size calculation was not performed. Instead, a sensitivity power analysis was conducted using G*Power 3.1 for the primary correlation analyses. With a total sample size of 420, a two-sided significance level of α = 0.05, and 80% statistical power, the study was able to detect a minimum correlation coefficient of approximately |r| = 0.14.

Data were collected from patients who received medical nutrition therapy at the nutrition and diet outpatient clinic between December 2023 and June 2024. Information on medication and dietary supplement use was obtained, and participants using hormonal therapy, insulin sensitizers, or any dietary supplements within the previous three months were excluded. This study was conducted in accordance with the Declaration of Helsinki. Ethical approval for this retrospective study titled “Determination of Mediterranean Diet Adherence in PCOS Patients and Its Relationship with Biochemical and Anthropometric Parameters” was obtained from the Ethics Committee of Etlik Zübeyde Hanım Women’s Health Training and Research Hospital (approval number: 12–37, dated 18 November 2024). Due to the retrospective design of the study and the use of anonymized data, the requirement for informed consent was waived by the ethics committee.

PCOS diagnostic criteria

The diagnosis of PCOS was established according to the Rotterdam criteria (2003), requiring at least two of the following: (i) oligo- or anovulation; (ii) clinical and/or biochemical signs of hyperandrogenism; (iii) polycystic ovarian morphology on ultrasound [17].

Dietary assessment and mediterranean diet adherence

Adherence to the Mediterranean diet was assessed using the 14-item Mediterranean Diet Adherence Screener (MEDAS) questionnaire, which is routinely administered in our clinic prior to initiating Mediterranean diet therapy. The scale evaluates the frequency of consumption of key Mediterranean diet components, including healthy fats (olive oil), vegetables, fruits, whole grains, legumes, and fish. A total MEDAS score below 7 indicates low adherence, a score of 7–8 indicates acceptable adherence, and a score ≥ 9 indicates strict adherence. The Turkish version of the 14-item MEDAS questionnaire, which has demonstrated acceptable validity and reliability for use in Turkish adults, was employed in the study. Each item is scored as 0 or 1, yielding a total score of 0–14, with higher scores indicating greater adherence [18].

Anthropometric measurements

Anthropometric measurements were obtained in the morning following an overnight fast and according to the standardized clinical procedures of the outpatient clinic. Anthropometric measurements, including weight, height, body fat percentage, and muscle mass, were routinely recorded using a calibrated body composition analyzer (Tanita BC418 MA (Tanita Corporation, Tokyo, Japan). Body mass index (BMI; kg/m²) was calculated as weight (kg) divided by height squared (m²) [19].

Biomarker assessment

After an overnight fast of 8–12 h, venous blood samples were drawn between 08:00 and 10:00 A.M [20]., during the early follicular phase (between the 2nd and 5th days of menstruation) to minimize hormonal variability. During this period, no major changes in biochemical markers are expected. Serum glucose, insulin, lipid profiles (total cholesterol, triglycerides (TG), High-density lipoprotein cholesterol (HDL-C), Low-density lipoprotein cholesterol (LDL-C), Glycated haemoglobin A1c (HbA1c), and C-reactive protein (CRP) levels were analyzed using standardized enzymatic and immunoassay methods and an automated analyzer (e.g., Roche Cobas). Oral glucose tolerance test (OGTT) fasting and both 1st hour (1-h) and 2nd hour (2-h) plasma glucose levels were also assessed. The homeostasis model assessment of insulin resistance (HOMA-IR) was calculated using the following formula: fasting glucose (mg/dL) × fasting insulin (µU/mL)/405 [21].

Statistical analysis

Statistical analyses were performed using IBM SPSS Statistics version 22.0 (IBM Corp., Armonk, NY, USA) [22]. Continuous variables were assessed for normality using the Kolmogorov–Smirnov and Shapiro–Wilk tests, together with visual inspection of histograms and evaluation of skewness and kurtosis values. Skewness and kurtosis values between − 2 and + 2 were considered indicative of acceptable normality. Descriptive statistics are presented as mean ± standard deviation (SD) for normally distributed variables and as median (minimum–maximum) for non-normally distributed variables. The primary objective of the study was to examine the associations between Mediterranean diet adherence, assessed by the Mediterranean Diet Adherence Screener (MEDAS), and inflammatory, anthropometric, and metabolic parameters, as well as the associations between C-reactive protein (CRP) levels and these parameters. Because both MEDAS scores and CRP concentrations did not meet the assumption of normal distribution, correlation analyses involving these variables were conducted using Spearman’s rank correlation coefficient. Correlation coefficients (r) were interpreted according to conventional thresholds, with values around 0.10 considered weak, 0.30 moderate, and ≥ 0.50 strong associations. All statistical tests were two-sided, and a p-value < 0.05 was considered statistically significant. In line with current recommendations, the magnitude and direction of associations were emphasized over sole reliance on p-values. Additionally, multivariable linear regression analyses were performed to evaluate whether the associations between MEDAS scores and selected outcomes (CRP, HbA1c, and HOMA-IR) remained significant after adjustment for age and BMI.

Results

Table 1 presents the mean, standard deviation (SD), median, skewness, kurtosis, minimum, and maximum values of various physical and biochemical characteristics of the study participants. These data include age, height, weight, body mass index (BMI), body fat percentage, muscle mass, lipid profile (total cholesterol, triglycerides, HDL-C, LDL-C), glycemic control indicators (HbA1C, OGTT fasting, 1-h OGTT, 2-h OGTT), insulin, and CRP levels as an indicator of inflammation.

Table 1.

Descriptive characteristics of the study participants (n = 420)

	Mean±SD	Median	Min-Max
Age (years)	28.0±4.9	28	20-45
Height (cm)	160.9±6.8	160	150-177
Weight (kg)	76.3±16.2	71.1	52.7-121.5
BMI (kg/m2)	29.5±5.9	27.6	20.5-42.3
Body Fat (%)	33.1±7.5	32.6	17.5-46.8
Muscle Mass (kg)	47.6±6.4	46.9	38.5-66.4
Total Cholesterol (mg/dL)	182.2±34.1	180.5	121.6-240
Triglyceride (mg/dL)	136.2±86.0	119	37-420
HDL-C (mg/dL)	53.8±13.8	50	25-99
LDL-C (mg/dL)	99.1±27.0	92	58-146
HbA1c (%)	5.3±0.5	5.3	3.7-7.2
CRP (mg/L)	3.4±2.6	3.1	0.2-9.2
Fasting glucose (mg/dL)	89.5±13.9	86	75-149
1-h OGTT (mg/dL)	128.5±47.1	119	61-266
2-h OGTT (mg/dL)	104.9±33.6	95.5	63-222
Insulin (µU/mL)	14.0±8.6	11.2	2.5-36.5
HOMA-IR	3.23 ± 2.27	2.57	0.50–11.01
MEDAS (score 0–14)	6.8±2.3	6	3-11

SD standard deviation, Min-Max minimum-maximum

Participant characteristics, biochemical and glycemic parameters

The mean age of the participants was 28.0 ± 4.9 years (median: 28; range: 20–45). The mean height was 160.9 ± 6.8 cm (median: 160; range: 150–177), and the mean weight was 76.3 ± 16.2 kg (median: 71.1; range: 52.7–121.5). BMI was 29.5 ± 5.9 kg/m² (median: 27.6; range: 20.5–42.3). Body fat percentage was 33.2 ± 7.5% (median: 32.6; range: 17.5–46.8), and mean muscle mass was 47.6 ± 6.4 kg (median: 46.4; range: 38.5–66.4) (Table 1).

According to the biochemical data, the mean total cholesterol level was 182.2 ± 34.1 mg/dL, triglycerides 136.2 ± 86.0 mg/dL, HDL-C 53.8 ± 13.8 mg/dL, and LDL-C 99.1 ± 27.0 mg/dL. Among glycemic parameters, the mean HbA1c level was 5.3 ± 0.5%. Mean fasting, 1-h, and 2-h OGTT glucose levels were 89.5 ± 13.9 mg/dL, 128.5 ± 47.1 mg/dL, and 104.9 ± 33.6 mg/dL, respectively. The mean insulin level was 14.0 ± 8.6 µU/mL. The mean CRP level, an indicator of inflammation, was 3.4 ± 2.6 mg/L (Table 1).

Mediterranean diet adherence

The mean of the MEDAS total scores was 6.8 ± 2.3, the median was 6, the skewness value was 0.42, and the kurtosis value was − 0.683. The minimum value was 3 and the maximum value was 11. A total score of 9 and above indicates that the individual has strict compliance with the Mediterranean diet, a score of 7 and above indicates an acceptable degree of compliance with the Mediterranean diet, and a total score below 7 indicates that there is no compliance with the Mediterranean diet. In this study, it was found that PCOS patients generally did not comply with the Mediterranean diet. In the study population, 55% of the patients had MEDAS scores below 7, indicating poor adherence to the Mediterranean Diet, whereas 45% achieved scores of 7 or higher, reflecting at least an acceptable level of compliance.

Correlations between MEDAS and anthropometric/biochemical parameters

Table 2 shows the correlations between Mediterranean diet adherence (MEDAS score) and anthropometric and biochemical parameters. MEDAS scores were weakly negatively correlated with BMI (r = − 0.228, p < 0.001), body fat percentage (r = − 0.134, p = 0.006), CRP levels (r = − 0.097, p = 0.047), and 1-hour OGTT glucose levels (r = − 0.139, p = 0.004). In contrast, weak positive correlations were observed between MEDAS scores and HDL-C (r = 0.157, p = 0.001), HbA1c (r = 0.122, p = 0.012), fasting glucose (r = 0.188, p < 0.001), 2-hour OGTT glucose (r = 0.224, p < 0.001), and insulin levels (r = 0.124, p = 0.011).

Table 2.

Spearman correlation between MEDAS score and anthropometric and biochemical parameters

Variable	MEDAS
	r	p
BMI (kg/m2)	−0.228	< 0.001*
Body Fat (%)	−0.134	0.006*
Muscle Mass (kg)	−0.087	0.065
Total Cholesterol (mg/dL)	0.025	0.616
Triglyceride (mg/dL)	−0.056	0.254
HDL-C (mg/dL)	0.157	0.002*
LDL-C (mg/dL)	−0.039	0.423
HbA1c (%)	0.122	0.012*
CRP (mg/L)	−0.097	0.047*
Fasting glucose (mg/dL)	0.188	< 0.001*
1-h OGTT (mg/dL)	−0.139	0.004*
2-h OGTT (mg/dL)	0.224	< 0.001*
Insulin (µU/mL)	0.124	0.011*
HOMA-IR	0.126	0.010*

*p < 0.05

Correlations between CRP and anthropometric/biochemical parameters

The associations between CRP (C-reactive protein) levels and anthropometric and biochemical parameters were examined using Spearman correlation analysis. The findings are presented in Table 3. A moderate, positive, and statistically significant relationship was observed between CRP and BMI (r = 0.501; p < 0.001) as well as body fat percentage (r = 0.522; p < 0.001). In contrast, the associations between CRP and both insulin (r = 0.042; p = 0.350) and HbA1c (r = 0.060; p = 0.259) were positive but weak and not statistically significant. In summary, CRP levels demonstrated a particularly significant and moderate correlation with obesity indicators such as BMI and body fat percentage, whereas no significant associations were found with glycemic parameters including insulin and HbA1c.

Table 3.

Spearman correlation between CRP levels and anthropometric and metabolic parameters

	BMI (kg/m2)		Body Fat (%)		Insulin (µU/mL)		HbA1c (%)		HOMA-IR
	r	p	r	p	r	p	r	p	r	p
CRP	0.501*	< 0.001	0.522 *	< 0.001	0.042	0.350	0.060	0.259	0.183*	< 0.001

Values indicated with * were analyzed with Spearman correlation coefficient.*, p < 0.05; Spearman correlation coefficients

Multivariable regression analyses

To further evaluate whether the observed associations remained independent of age and BMI, multivariable linear regression analyses were performed. After adjustment, MEDAS scores remained significantly associated with CRP (β = −0.268, p < 0.001), HbA1c (β = 0.263, p < 0.001), and HOMA-IR (β = 0.097, p = 0.037). BMI was independently associated with HbA1c (β = 0.241, p < 0.001) and HOMA-IR (β = 0.398, p < 0.001), whereas age was inversely associated with HOMA-IR (β = −0.236, p < 0.001). The regression models explained 6.8%, 9.7%, and 12.4% of the variance in CRP, HbA1c, and HOMA-IR, respectively (Table 4).

Table 4.

Multivariable linear regression analyses adjusted for age and BMI

Dependent Variable	MEDAS β	MEDAS p	Age β	Age p	BMI β	BMI p	Adjusted R²
CRP	-0.268	< 0.001	0.071	0.180	-0.102	0.056	0.068
HbA1c	0.263	< 0.001	-0.012	0.819	0.241	< 0.001	0.097
HOMA-IR	0.097	0.037	-0.236	< 0.001	0.398	< 0.001	0.124

Models were adjusted for age and BMI. β values represent standardized regression coefficients. All variance inflation factor (VIF) values were below 2, indicating no evidence of multicollinearity.

Discussion

This study aimed to assess adherence to the Mediterranean diet among women with polycystic ovary syndrome (PCOS) and to examine the associations between dietary compliance and biochemical as well as anthropometric parameters.

As a result of the study, it was observed that the participants were overweight with an average BMI of 29.53. The average fat ratio was 33.20%. Although it varies according to age, it is important for health that the total body fat ratio in women should not exceed 30% [23]. High insulin in PCOS may contribute to excess adiposity by limiting fat mobilization and oxidation, thus making weight maintenance and weight loss a significant challenge. Obesity, especially abdominal or ectopic adiposity, may contribute to the severity of symptoms and progression of PCOS-related comorbidities [24]. A study examining subcutaneous and visceral adiposity, which is the strongest sign of sexual dysmorphism, have shown that women with PCOS have more adipose tissue mass in the abdomen, waist and upper arm regions compared to women in the control group [25]. The findings of this study also support the existing literature in this respect.

According to biochemical data, the mean total cholesterol was 182.2 mg/dL. The mean triglyceride level was 136.2 ± 86.08 mg/dL. HDL-C was 53.8 ± 13.8 mg/dL and LDL was 99.1 ± 27.0 mg/dL. A meta-analysis of 61 observational studies showed that women with PCOS had higher LDL-C, HDL-C, and TG levels and lower HDL-C levels compared to the control group [26]. It is suggested that PCOS poses an increased risk of obesity and insulin resistance-related dyslipidemia [27]. Although there are few studies examining the prevalence of dyslipidemia in older women with PCOS, the available evidence suggests that dyslipidemia is common in young women with PCOS and probably persists after menopause [28, 29]. In our study, the mean age was 28 years and the healthier blood lipid results may be due to the younger age of the patients.

According to international organizations such as the American Diabetes Association (ADA), fasting plasma glucose ≥ 126 mg/dL, incidental plasma glucose + diabetes symptoms ≥ 200 mg/dL, 2nd hour plasma glucose ≥ 200 mg/dL on the OGTT, and HbA1c ≥ 6.4% lead to a diagnosis of diabetes [30]. Again, for prediabetes, many scientific authorities consider a fasting glucose of 100–125 mg/dL as “impaired fasting glucose” and a 2nd hour plasma glucose of 140–199 mg/dL and HbA1c of 5.7–6.4% on the OGTT to indicate prediabetes [31–33]. The mean HbA1c of the participants in this study was 5.3%. In the OGTT results, the fasting glucose mean was 89.5 mg/dL; the 1-h OGTT mean was 128.5 mg/dL; the 2-h OGTT mean was 104.9 mg/dL; and all parameters were in contrast with the prior literature. Contrary to our findings, it has been reported in the literature that the prevalence of impaired glucose tolerance and DM2 is increased in women with PCOS independent of BMI [34]. In a study conducted in Denmark, it was reported that the risk of DM2 was 4 times higher among women with PCOS and DM2 was diagnosed 4 years earlier in women with PCOS compared to women in the control group [35]. It has been reported that nearly one-third of women diagnosed with PCOS exhibit impaired glucose tolerance, while approximately 10% develop type 2 diabetes mellitus [36]. Moreover, findings from a decade-long follow-up study indicate that the age-adjusted prevalence of type 2 diabetes among women with PCOS in their 40s and 50s reaches 40%, representing a 6.8-fold increase compared with age-matched women in the general population [37]. Again, PCOS has been found to be associated with insulin resistance and hyperinsulinemia. Insulin level lower than 10 µIU/mL as ideal for fasting insulin, but 2–25 µIU is considered normal [34]. In our study, the mean fasting insulin level was found to be 14.0 µU/mL. The participants could not be classified as hyperinsulinemic; however, their fasting insulin concentrations were above the ideal range.

Recent evidence in the literature has focused on low-grade chronic inflammation as a potential cause of the long-term consequences of PCOS. Elevated CRP levels, inflammatory cytokines (such as interleukin-1 (IL-1), tumor necrosis factor-alpha (TNF-α), interleukin-6 (IL-6) and increased leukocyte count indicate low-grade chronic inflammation in PCOS [38]. The mean CRP level was 3.4 mg/L (slightly elevated). The data obtained are consistent with the presence of a low-grade chronic inflammatory state of the PCOS patients [39].

In this study, the mean MEDAS total score was 6.8. Scores below 7 are indicative of low compliance with the Mediterranean diet. Weak but statistically significant negative correlations were observed between MEDAS scores and both BMI and body fat percentage. Similarly, Barrea et al. [40] reported lower adherence to the Mediterranean diet among women with PCOS compared with controls.

Contrary to the literature, our study also found a negative but not statistically significant correlation between MEDAS scores and muscle mass. The timing of bioelectirical impedance analyses measurement has not been mentioned in most studies on the subject. Increases in extracellular fluid and water retention may temporarily affect the accurate assessment of the body composition of women, especially during the menstrual cycle [41]. Although previous studies examining the possible effect of the menstrual cycle on body composition have not reached consistent results [41, 42], there may be deviations in muscle mass measurements because the measurements were performed between the second and fifth days of menstruation. However, our study did not control for menstrual cycle-related fluid fluctuations beyond noting the measurement window, which should be acknowledged as a limitation. Future studies should consider controlling for the cycle phase or conducting repeated measurements to minimize this potential source of bias.

Consistent with our findings, Mei et al. demonstrated that a low-calorie Mediterranean dietary pattern resulted in more pronounced improvements in anthropometric outcomes than a low-fat diet in overweight women with PCOS [43]. In another study conducted in 2023, a Mediterranean diet was administered to one group of women with PCOS and a ketogenic diet to another group. Statistically significant differences were observed in the anthropometric and biochemical parameters of both groups after both dietary treatments (p < 0.001). However, the decrease in all parameters was significantly higher in the ketogenic diet group than in the Mediterranean diet group [29].

In this study, a weak but statistically significant positive correlation was observed between HbA1c levels and MEDAS scores. Although the Mediterranean diet is generally associated with improved glycemic control due to its high fiber content and low glycemic index [44, 45], and previous intervention studies have reported improvements in insulin resistance and fasting glucose following Mediterranean diet interventions in women with PCOS [46], the present findings demonstrated an unexpected association. This unexpected direction of association may be explained by the cross-sectional design of the study and the lack of control for potential confounding factors. It is also possible that some participants with previously identified metabolic abnormalities may have adopted healthier dietary habits before clinical evaluation, which could have contributed to the observed associations. Therefore, causal inferences cannot be drawn from the present findings. Furthermore, because MEDAS is a brief dietary adherence screener rather than a comprehensive dietary assessment tool [18], it does not assess total energy intake, macronutrient distribution, glycemic load, or portion sizes. Therefore, individuals with higher MEDAS scores may still have dietary characteristics not captured by MEDAS that could influence glycemic control. In addition, residual confounding by unmeasured factors such as physical activity, socioeconomic status, and other lifestyle-related variables cannot be excluded. Therefore, the observed associations should not be interpreted as evidence of a detrimental effect of the Mediterranean diet on glycemic outcomes.

Evidence from observational and meta-analytic studies indicates that elevated serum CRP concentrations in PCOS are not fully explained by adiposity alone and may persist even after BMI matching or statistical adjustment for BMI [47, 48].

In a study conducted on adolescent PCOS group [49], a significant and positive correlation was found between CRP and BMI (r = 0.308, p < 0.01). The results of the study indicated that CRP levels alone may not be associated with PCOS in the Indian adolescent group but may instead be associated with fat mass in this subgroup. Similarly, in another study [50], this time in non-obese adolescent PCOS women, a strong positive correlation was found between CRP levels and BMI (p = 0.001). In a study examining obesity and inflammatory biomarkers in women with PCOS, neither IL-6 nor CRP levels differed between PCOS patients and age- and BMI-matched controls. However, multivariate analyses revealed that BMI was the strongest determinant of IL-6 and CRP levels, indicating that inflammation was related to adiposity rather than PCOS-specific endocrine abnormalities [51]. In addition, several studies have reported that inflammatory and metabolic parameters are significantly higher in adolescents with PCOS compared with controls, even after adjustment for BMI, with more pronounced alterations observed in overweight individuals [52, 53]. Therefore, adolescents with PCOS should be encouraged to maintain a healthy lifestyle and ideal body weight to reduce future metabolic risk. Moreover, CRP and cystatin C have been proposed as promising biomarkers for predicting future metabolic complications in this population [52]. Consistent with the existing literature, our study demonstrated a positive, moderate, and statistically significant correlation between CRP and BMI (r = 0.501, p < 0.001), as well as between CRP and body fat percentage (r = 0.522, p < 0.001).

This study has several limitations that should be acknowledged. Dietary assessment was based solely on MEDAS scores, as detailed dietary intake data were not available, which may have limited a more comprehensive evaluation of participants’ nutritional status. Furthermore, MEDAS is a brief dietary adherence screener and does not evaluate total energy intake, macronutrient distribution, glycemic index, glycemic load, portion sizes, or overall dietary quality. In addition, due to the cross-sectional design, causal relationships between Mediterranean diet adherence and metabolic or anthropometric outcomes cannot be established. Furthermore, important lifestyle-related factors such as physical activity, smoking status, and socioeconomic status were not assessed and may have influenced the observed associations. Therefore, residual confounding cannot be excluded. In addition, information regarding PCOS phenotypes, hyperandrogenism severity, menstrual characteristics, and fertility status was not available and therefore could not be included in the analyses. In addition, participants were recruited from a specialized PCOS outpatient clinic, which may limit the generalizability of the findings to the broader population of women with PCOS. Future prospective studies with more detailed dietary and lifestyle assessments are warranted to further clarify these relationships.

Conclusions

In conclusion, women with PCOS in this study showed generally low adherence to the Mediterranean diet. Several statistically significant but weak correlations were observed between MEDAS scores and selected anthropometric and metabolic parameters. Although several associations reached statistical significance, the observed effect sizes were generally weak, suggesting limited clinical relevance. In contrast, CRP levels were moderately associated with BMI and body fat percentage, highlighting the relationship between adiposity and low-grade inflammation in women with PCOS.

These findings should be interpreted with caution due to the retrospective cross-sectional design and the use of a brief dietary screening tool. Nevertheless, promoting healthy dietary patterns such as the Mediterranean diet alongside weight management strategies may be associated with more favorable metabolic profiles in women with PCOS. Further well-designed prospective studies are needed to better clarify the role of dietary patterns in modulating metabolic and inflammatory outcomes in this population.

Authors’ contributions

Conceptualization, K.Y. and A.K.A.; methodology, A.K.A. and M.C.İ.; software, K.Y. and A.K.A.; validation, A.K.A., M.C.İ., and K.Y.; formal analysis, A.K.A. and Ç.P.F; investigation, K.Y.; resources, A.K.A. and K.Y.; data curation, A.K.A., M.C.İ., Y.E.Ü.; writing—original draft preparation, A.K.A. and K.Y.; writing—review and editing, A.K.A., K.Y., Ç.P.F.; visualization, M.C.İ.; supervision, M.C.İ. and Y.E.Ü. All authors have read and agreed to the published version of the manuscript.

Funding

Not applicable.

Data availability

The datasets used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Declarations

Ethics approval and consent to participate

This study was conducted in accordance with the Declaration of Helsinki, and approval for the retrospective study titled “Determination of Mediterranean Diet Adherence in PCOS Patients and Its Relationship with Biochemical and Anthropometric Parameters” was obtained from the Ethics Committee of Etlik Zübeyde Hanım Women’s Health Training and Research Hospital (approval number: 12–37, dated 18 November 2024). Due to the retrospective design of the study and the use of anonymized data, the requirement for informed consent was waived by the ethics committee.

Consent for publication

Not applicable.

Competing interests

The authors declare no competing interests.