Authors: Halis Doğukan Ozkan, Merve Ayas Ozkan, Ruken Dayanan, Dilara Duygulu Bulan, Ahmet Arif Filiz, Yaprak Engin-Ustun
Categories: Research, Polycystic ovary syndrome, Inflammation, Biomarkers, Chronic disease
Source: BMC Women's Health
Authors: Halis Doğukan Ozkan, Merve Ayas Ozkan, Ruken Dayanan, Dilara Duygulu Bulan, Ahmet Arif Filiz, Yaprak Engin-Ustun
Polycystic ovary syndrome (PCOS) is a common endocrine-metabolic disorder in women of reproductive age, often associated with low-grade chronic inflammation. This study aimed to investigate the relationship between composite inflammatory markers—Neutrophil-to-Lymphocyte Ratio (NLR), Systemic Immune Inflammation Index (SII), Systemic Inflammation Response Index (SIRI), and Aggregate Index of Systemic Inflammation (AISI)—and clinical phenotypes of PCOS, including oligo-/amenorrhea, hyperandrogenism, and polycystic ovarian morphology (PCOM).
In this retrospective cross-sectional study, women with PCOS were categorized into subgroups based on clinical phenotypes. Composite inflammatory markers were calculated from complete blood count parameters, and their association with clinical, hormonal, and ultrasonographic findings was analyzed. ROC analysis assessed the discriminatory value of each marker.
Inflammatory markers were significantly higher in the oligo-/amenorrhea and PCOM groups. In oligo-/amenorrhea, NLR (2.42 vs. 1.96, p = 0.028), AISI (282.16 vs. 188.73, p = 0.005), SII (664.59 vs. 519.86, p = 0.008), and SIRI (0.97 vs. 0.70, p = 0.005) were elevated. In PCOM, NLR (2.85 vs. 2.25, p = 0.025), AISI (323.15 vs. 262.78, p = 0.049), SII (738.52 vs. 641.27, p = 0.030), and SIRI (1.10 vs. 0.90, p = 0.046) were also higher. AISI showed the best discrimination for oligo-/amenorrhea (AUC = 0.652), while NLR was most predictive for PCOM (AUC = 0.617).
Composite inflammatory markers are elevated in specific PCOS phenotypes, especially oligo-/amenorrhea and PCOM, and may reflect low-grade inflammation. These markers may help identify patients at higher cardiometabolic risk and guide preventive strategies.
Polycystic ovary syndrome (PCOS) is one of the most common hormonal-metabolic disorders in women of reproductive age [1]. Increased hair growth, oligomenorrhea or amenorrhea and polycystic appearance of the ovaries are the main symptoms [2, 3]. The most widely accepted diagnostic criteria are the Rotterdam criteria. PCOS is diagnosed when two of the following criteria are Oligo-anovulation, hyperandrogenism and polycystic appearance of the ovaries [4]. The exact etiology is unknown, but lifestyle, genetic predisposition and environmental factors are thought to play a role [2, 5, 6]. Recent studies have shown that inflammatory mechanisms also play a role in the pathogenesis of PCOS [7–9]. Obesity and insulin resistance are conditions associated with higher levels of inflammation and may exacerbate the inflammatory process in PCOS [9, 10]. In addition, low-grade inflammation has been reported to be associated with cardiovascular disease, type 2 diabetes, and dyslipidemia [11–13].
Low grade inflammation is the presence of persistent mild immune activation in the body without obvious signs of disease. The most commonly used markers for this condition are C-reactive protein (CRP), interleukin (IL)−6, leukocyte and platelet counts [14–16]. Recently, composite inflammatory markers calculated using complete blood cell count have also gained popularity. In various fields such as gynecological, obstetric, and metabolic diseases, the Systemic Inflammation Response Index (SIRI), Systemic Immune Inflammation Index (SII), and Neutrophil-Lymphocyte Ratio (NLR), as well as the Total Systemic Inflammation Index (AISI), have become widely used composite markers [17–21].
Recent studies investigating the relationship between PCOS and inflammation have often focused on the etiology [17, 18]. Cytokines and interleukins have frequently been used as inflammatory markers, and studies involving composite inflammatory markers are limited in the literature [6]. The association between inflammatory markers and clinical symptoms and findings of PCOS has not been adequately studied. In this study, we aimed to investigate the association between composite markers reflecting low-grade inflammation (SIRI, SII, AISI, NLR) and clinical symptoms and findings in women with PCOS. In doing so, we aimed to emphasize that women with clinically apparent findings in the presence of inflammation may have a higher risk for metabolic disease and that such findings could inform follow-up strategies.
This study was designed as a retrospective cross-sectional study conducted between January 2021 and January 2023 at Etlik Zübeyde Hanım Women’s Diseases and Obstetrics Education and Research Hospital in patients diagnosed with PCOS. It included 315 PCOS patients aged 18–35 years. Ethical approval was obtained from the local ethics committee (Decision No. 2024/15–23 dated December 9, 2024) and the study was conducted in accordance with the Declaration of Helsinki.
A total of 315 patients diagnosed with PCOS according to the Rotterdam criteria were included in the study (presence of two of the following Hyperandrogenism, oligo-anovulation and polycystic ovarian morphology) [4]. Exclusion criteria included diseases that could cause increased androgen levels (androgen-producing tumors, congenital adrenal hyperplasia, hyperprolactinemia, Cushing’s disease, etc.), inflammatory diseases, infections, use of antibiotics, diseases affecting platelet count, diabetes, pregnancy and patients with incomplete data. Patients who met the inclusion and exclusion criteria were divided into subgroups based on the presence of oligo-/amenorrhea, hyperandrogenism and PCO morphology (PCOM). All women diagnosed with PCOS according to the Rotterdam criteria, who presented to our clinic during the study period and had complete clinical, hormonal, and ultrasonographic data available, were included in the study. No additional filtering or selection was applied beyond the predefined exclusion criteria stated above.
All patient data was retrieved retrospectively from the hospital information system. Transvaginal ultrasound findings, a complete blood count (lymphocyte count, neutrophil count, platelet count, monocyte count), a hormone profile (follicle stimulating hormone (FSH), luteinizing hormone (LH), Estradiol (E2), anti-Müllerian hormone (AMH), total testosterone, sex hormone-binding globulin (SHBG) and 17-hydroxyprogesterone) was obtained from fasting blood samples taken in the morning on the 2nd and 4th day of menstruation. and day 4 of the menstrual cycle. Hyperandrogenism was diagnosed clinically and biochemically (Ferriman-Gallway score > 8, total testosterone > 2.3, FAI > 5) [20, 21]. Oligomenorrhea was defined as a cycle length of more than 35 days, PCOM, an antral follicle count of more than 12 or an ovarian volume of more than 10 cm³ [4]. All hormonal, complete blood count, and ultrasound parameters were determined from fasting blood samples and ultrasound examinations performed on the same day.
Systemic inflammatory markers were calculated using the parameters of the complete blood count obtained from all study participants. NLR was calculated by dividing the neutrophil count by the lymphocyte count. SIRI was calculated by dividing the product of the neutrophil and monocyte count by the lymphocyte count. SII was calculated by dividing the product of platelet and neutrophil count by the lymphocyte count. The Aggregate Index of Systemic Inflammation was calculated by dividing the product of the neutrophil, monocyte and platelet count by the lymphocyte count. All values were calculated using absolute cell counts (×10³/µL).
Statistical analysis was performed using IBM SPSS Statistics version 22.0 (IBM Corporation, Armonk, NY, USA). The Kolmogorov-Smirnov test was used to evaluate conformity to normal distribution. Descriptive statistics of continuous variables are presented as mean ± standard deviation for normally distributed data and as median (interquartile range) for non-normally distributed data. Categorical variables were compared using the uncorrected Pearson chi-squared test or Fisher’s exact test, as appropriate. Continuous variables were analyzed using the two-tailed independent samples t-test for normally distributed data and the Mann-Whitney U test for non-normally distributed data. Receiver operating characteristic (ROC) curve analysis was used to calculate and compare the areas under the curve (AUC) and to determine optimal cut-off values according to the Youden index. Multivariate logistic regression analysis adjusted for BMI was performed to assess the independent associations between inflammatory markers and the presence of oligo-/amenorrhea. For multiple comparisons across several variables, the false discovery rate (FDR) correction method was applied to control for type I error. Effect sizes (r) were calculated for each comparison to quantify the magnitude of the observed differences. A p-value < 0.05 was considered statistically significant for all tests.
The mean age of patients with and without oligo-/amenorrhea was similar (24.3 ± 4.9 vs. 25.3 ± 6.5 years, adjusted p = 1.000), as was the BMI (26.1 ± 9.2 vs. 24.8 ± 3.2 kg/m², adjusted p = 1.000). Women with oligo/amenorrhea had significantly higher total testosterone (0.40 vs. 0.26 ng/mL, adjusted p = 0.023, r = 0.186), FAI (7.33 vs. 2.89, adjusted p = 0.029, r = 0.148), AMH (5.60 vs. 3.90 ng/mL, adjusted p = 0.023, r = 0.159) than the women without oligo-/amenorrhea. The inflammatory markers were also significantly higher in the oligo-/amenorrhea AISI (282.16 vs. 188.73, adjusted p = 0.020, r = 0.259), SII (664.59 vs. 519.86, adjusted p = 0.019, r = 0.278) and SIRI (0.97 vs. 0.70, adjusted p = 0.011, r = 0.297). Further clinical and laboratory findings are summarized in Table 1. In a multivariate logistic regression model adjusted for BMI, AISI (adjusted OR: 1.074, p = 0.046), SII (adjusted OR: 1.069, p = 0.048), and SIRI (adjusted OR: 1.943, p = 0.022) remained significantly associated with oligo-/amenorrhea, whereas NLR demonstrated a borderline association (adjusted OR: 0.060, p = 0.060) (Table 2).
Table 1Descriptive and comparative analysis of demographic data between PCOS patients according to oligomenorrhea or amenorrheaAbsent n = 46 (14.6%)Present n = 267 (85.4%)r (effect size)P-valueAdjusted p-valueAge (y)25.3 ± 6.524.3 ± 4.90.0690.246^b^1.000Body mass index (kg/m²)24.8 (3.2)26.1 (9.2)0.0570.310^a^1.000FGS6.0 (4.0)9.5 (7.0)0.0470.406^a^1.000Total Testosterone (ng/mL)0.26 (0.26)0.40 (0.28)0.186**<**0.001^a^0.023DHEAS (µg/dL)200.0 (103.0)239.5 (134.0)0.0790.164^a^1.000SHGB (nmol/L)55.0 (33.1)48.8 (27.8)0.0290.607^a^1.000Androstenedione (nmol/L)2.30 (1.87)3.17 (1.56)0.0720.201^a^1.000FAI2.89 (10.0)7.33 (8.0)0.1480.009^a^0.029FSH (mIU/mL)7.41 (4.18)6.01 (1.90)0.0360.520^a^1.000LH (IU/L)9.05 (4.16)8.57 (5.68)0.0890.114^a^1.000Estradiol (pg/mL)26.7 (15.5)35.8 (21.5)0.0900.112^a^1.000T4 (mcg/dL)15.0 (2.8)15.1 (2.9)0.0570.314^a^1.000TSH (mIU/L)1.80 (0.50)2.24 (2.01)0.0630.266^a^1.000AMH (ng/mL)3.90 (5.08)5.60 (4.71)0.1590.005^a^0.023D21-24 progesterone (ng/mL)0.8 (10.0)0.39 (1.0)0.1630.004^a^0.09217-OH progesterone (ng/mL)0.41 (0.64)0.46 (0.47)0.0100.858^a^1.000Prolactin (µg/L)23.0 (8.6)15.1 (14.2)0.0330.556^a^1.000Right ovary volume16.6 (18.2)10.5 (5.3)0.1360.016^a^0.368Left ovary volume13.3 (8.1)10.5 (5.1)0.0630.263^a^1.000NLR1.96 (1.00)2.42 (1.41)0.1240.028^a^0.071AISI188.73 (180.89)282.16 (178.63)0.2590.005^a^0.020SII519.86 (281.26)664.59 (347.89)0.2780.008^a^0.019SIRI0.70 (0.53)0.97 (0.60)0.2970.005^a^0.011Data are expressed as n (%), mean ± SD or median (interquartile range) where appropriate. A p value of < 0.05 indicates a significant difference and statistically significant p-values are in bold. P-values were adjusted using the false discovery rate (FDR) method. ^a^: Mann Withney-U test, ^b^: Student t-test, ^c^: Fisher’s exact test, PCOS Polycystic ovary syndrome, FGS Ferriman-Gallwey score, DHEAS Dehydroepiandrosterone sulfate, SHGB Sex hormone binding globulin, FAI Free Androgen Index, FSH Follicle stimulating hormone, LH Luteinizing hormone, T4 Thyroxine, TSH Thyroid stimulating hormone, AMH Anti-Mullerian Hormone. NLR Neutrophil-to-Lymphocyte Ratio, AISI Aggregate Index of Systemic Inflammation, SII Systemic Immune-Inflammation Index, SIRI Systemic Inflammation Response Index
Table 2Multivariate logistic regression analysis including BMI, NLR, AISI, SII, and SIRI for prediction of oligo/amenorrhea in PCOS patientsVariableaOR (95% CI)p valueBMI1.099 (0.979–1.234)0.109NLR0.060 (0.003–1.127)0.060AISI1.074 (0.949–1.091)0.046SII1.069 (0.999–1.082)0.048SIRI1.943 (1.260–2.390)0.022*NLR *Neutrophil-to-Lymphocyte Ratio, *AISI *Aggregate Index of Systemic Inflammation, *SII Systemic Immune-Inflammation Index, SIRI Systemic Inflammation Response Index, aOR adjusted Odds ratio, CI *confidence interval
By hyperandrogenism status, there was no significant difference in age between women with and without hyperandrogenism (23.5 ± 5.1 vs. 24.8 ± 5.1 years, adjusted p = 0.120). Women with hyperandrogenism had higher total testosterone (0.40 vs. 0.19 ng/mL, adjusted p = 0.005, r = 0.180), FAI (8.35 vs. 2.90, adjusted p = 0.005, r = 0.178), and Ferriman–Gallwey scores (10.0 vs. 6.0, adjusted p = 0.005, r = 0.185) compared to those without hyperandrogenism. SHBG levels were significantly lower in the hyperandrogenism group (45.0 vs. 60.7 nmol/L, adjusted p = 0.005, r = 0.192). No significant differences were observed in inflammatory markers or ovarian volumes between the groups. Remaining findings comparing women with and without hyperandrogenism are summarized in Table 3.
Table 3Descriptive and comparative analysis of demographic data between PCOS patients according to hyperandrogenismAbsent n = 84 (%26.3)Present n = 231 (%73.7)r (effect size)P-valueAdjusted p-valueAge (y)23.5 ± 5.124.8 ± 5.10.1040.064^b^0.120Body mass index (kg/m²)26.4 (5.0)24.6 (8.6)0.0610.281^a^0.497FGS6.0 (3.0)10.0 (5.0)0.185**<0.001^a^0.005Total Testosterone (ng/mL)0.19 (0.27)0.40 (0.21)0.180<0.001^a^0.005DHEAS (µg/dL)206 (84)249 (124)0.1150.042^a^0.120SHGB (nmol/L)60.7 (73.0)45.0 (25.2)0.192<0.001^a^0.005Androstenedione (nmol/L)2.71 (1.40)3.10 (1.60)0.0340.544^a^0.640FAI2.9 (2.0)8.35 (6.0)0.178<**0.001^a^0.005FSH (mIU/mL)5.63 (3.09)6.07 (1.75)0.0330.555^a^0.636LH (IU/L)7.16 (4.12)8.68 (5.24)0.1330.018^a^0.068Estradiol (pg/mL)36.2 (25.8)35.07 (23.3)0.1300.018^a^0.070T4 (mcg/dL)16.1 (2.7)15.0 (2.9)0.0860.125^a^0.314TSH (mIU/L)1.79 (1.17)2.20 (1.68)0.0780.168^a^0.386AMH (ng/mL)4.65 (4.84)5.70 (5.20)0.0380.502^a^0.640D21-24 progesterone (ng/mL)0.35 (2.00)0.50 (1.00)0.0230.685^a^0.75017-OH progesterone (ng/mL)0.40 (0.54)0.44 (0.42)0.0400.473^a^0.640Prolactin (µg/L)13.5 (14.1)17.7 (13.9)0.0070.898^a^0.898Right ovary volume11.5 (7.7)10.55 (5.87)0.0330.557^a^0.640Left ovary volume10.8 (5.0)10.5 (5.5)0.0730.198^a^0.414NLR2.33 (1.30)2.30 (1.48)0.0480.396^a^0.607AISI288.13 (191.51)266.95 (170.75)0.0500.379^a^0.598SII662.67 (334.2)643.05 (377.70)0.0160.781^a^0.816SIRI0.99 (0.58)0.92 (0.62)0.0650.246^a^0.471Data are expressed as n (%), mean ± SD or median (interquartile range) where appropriate. A p value of < 0.05 indicates a significant difference and statistically significant p-values are in bold. P-values were adjusted using the false discovery rate (FDR) method. ^a^: Mann Withney-U test, ^b^: Student t-test, ^c^: Pearson chi-square, PCOS Polycystic ovary syndrome, FGS Ferriman-Gallwey score, DHEAS Dehydroepiandrosterone sulfate, SHGB Sex hormone binding globulin, FAI Free Androgen Index, FSH Follicle stimulating hormone, LH Luteinizing hormone, T4 Thyroxine, TSH Thyroid stimulating hormone, AMH Anti-Mullerian Hormone. NLR Neutrophil-to-Lymphocyte Ratio, AISI Aggregate Index of Systemic Inflammation, SII Systemic Immune-Inflammation Index, SIRI Systemic Inflammation Response Index
Comparison based on PCOM showed that patients with PCOM were slightly older (23.6 ± 4.3 vs. 24.6 ± 5.3 years, adjusted p = 0.448) and had similar BMI (25.6 vs. 24.6 kg/m², adjusted p = 0.370). Women with PCOM had significantly larger right (12.5 vs. 7.3 cm³, adjusted p = 0.011, r = 0.185) and left (12.0 vs. 7.4 cm³, adjusted p = 0.011, r = 0.185) ovarian volumes. Additional findings comparing women with and without PCOM are presented in Table 4.
Table 4Descriptive and comparative analysis of demographic data between PCOS patients according to polycystic ovary morphologyAbsent n = 50 (15.9%)Present n = 265 (84.1%)r (effect size)P-valueAdjusted p-valueAge (y)23.6 ± 4.324.6 ± 5.30.0620.219^b^0.448Body mass index (kg/m²)24.6 (13.0)25.6 (6.4)0.0790.161^a^0.370FGS9.0 (4.0)8.5 (7.0)0.0710.208^a^0.434Total Testosterone (ng/mL)0.32 (0.28)0.39 (0.30)0.0450.428^a^0.566DHEAS (µg/dL)214.0 (144.0)241.0 (126.0)0.0630.267^a^0.448SHGB (nmol/L)45.0 (23.4)55.0 (30.7)0.0190.730^a^0.810Androstenedione (nmol/L)2.60 (1.46)3.25 (1.83)0.0520.354^a^0.508FAI7.8 (13.0)7.2 (7.0)0.0090.878^a^0.878FSH (mIU/mL)7.08 (2.38)5.91 (2.05)0.0920.101^a^0.258LH (IU/L)10.70 (4.82)7.49 (4.31)0.0110.850^a^0.875Estradiol (pg/mL)34.0 (34.5)35.2 (22.8)0.0360.525^a^0.635T4 (mcg/dL)14.8 (2.0)15.7 (2.3)0.0520.353^a^0.508TSH (mIU/L)2.80 (2.50)2.00 (1.39)0.0190.740^a^0.810AMH (ng/mL)5.60 (3.76)4.89 (5.15)0.1400.013^a^0.099D21-24 progesterone (ng/mL)0.80 (4.0)0.34 (0.0)0.0960.087^a^0.25017-OH progesterone (ng/mL)0.44 (0.34)0.45 (0.53)0.0650.247^a^0.448Prolactin (µg/L)15.0 (13.4)18.5 (14.4)0.0430.443^a^0.566Right ovary volume7.3 (2.2)12.5 (6.4)0.185< 0.001^a^0.011Left ovary volume7.4 (2.6)12.0 (3.6)0.185< 0.001^a^0.011NLR2.25 (1.37)2.85 (1.55)0.1260.025^a^0.138AISI262.78 (173.67)323.15 (167.59)0.1110.049^a^0.161SII641.27 (351.70)738.52 (458.05)0.1220.030^a^0.135SIRI0.90 (0.60)1.10 (0.78)0.1120.046^a^0.159Data are expressed as n (%), mean ± SD or median (interquartile range) where appropriate. A p value of < 0.05 indicates a significant difference and statistically significant p-values are in bold. P-values were adjusted using the false discovery rate (FDR) method. ^a^: Mann Withney-U test, ^b^: Student t-test, ^c^: Fisher’s exact test, PCOS Polycystic ovary syndrome, FGS Ferriman-Gallwey score, DHEAS Dehydroepiandrosterone sulfate, SHGB Sex hormone binding globulin, FAI Free Androgen Index, FSH Follicle stimulating hormone, LH Luteinizing hormone, T4 Thyroxine, TSH Thyroid stimulating hormone, AMH Anti-Mullerian Hormone. NLR Neutrophil-to-Lymphocyte Ratio, AISI Aggregate Index of Systemic Inflammation, SII Systemic Immune-Inflammation Index, SIRI Systemic Inflammation Response Index
ROC analyses, as presented in Table 5, showed that inflammatory markers (NLR, AISI, SII, SIRI) moderately predicted both oligo/amenorrhea and PCOM. For oligo/amenorrhea, the optimal cut-off values NLR > 2.13 (AUC: 0.619, p = 0.030), AISI > 227.68 (AUC: 0.652, p = 0.007), SII > 588.97 (AUC: 0.642, p = 0.010) and SIRI > 0.87 (AUC: 0.650, p = 0.007) with a sensitivity of 60.9–65.5% and a specificity of 58.8–61.8%. Among these, the AISI demonstrated the highest discriminatory power (AUC: 0.652), closely followed by the SIRI (AUC: 0.650). The optimal cut-off values for PCOM NLR > 2.35 (AUC: 0.617, p = 0.018), AISI > 235.79 (AUC: 0.603, p = 0.049), SII > 671.01 (AUC: 0.614, p = 0.018) and SIRI > 0.97 (AUC: 0.603, p = 0.039), with sensitivities between 61.1 and 63.9% and specificities between 54.9 and 57.0%. In this group, NLR (AUC: 0.617) and SII (AUC: 0.614) achieved the best predictive performance. Overall, these results suggest that composite inflammatory indices make a modest but clinically relevant diagnostic contribution to the differentiation of PCOS phenotypes. Figure 1 shows the ROC curves of NLR, AISI, SII and SIRI for the prediction of oligo/amenorrhea, while Fig. 2 shows the ROC curves of these markers for the prediction of PCOM.
Table 5Evaluation of NLR, AISI, SII, and SIRI in patients with oligo-amenorrhea and polycystic ovarian morphology for prediction of outcomes using ROC analysisCut-offSensitivitySpecificityAUC%95 CIP*-valueNLR (for oligo-amenore)> 2.1362.4%58.8%0.6190.51–0.720.030AISI (for oligo-amenore)> 227.6865.5%58.8%0.6520.54–0.760.007SII (for oligo-amenore)> 588.9763.5%61.8%0.6420.53–0.750.010SIRI (for oligo-amenore)> 0.8760.9%58.8%0.6500.54–0.760.007NLR (for PCOM)> 2.3563.9%54.9%0.6170.52–0.710.018AISI (for PCOM)> 235.7961.2%56.2%0.6030.52–0.720.049SII (for PCOM)> 671.0161.1%57.0%0.6140.51–0.700.018SIRI (for PCOM)> 0.97061.1%55.4%0.6030.50–0.700.039*NLR Neutrophil-to-Lymphocyte Ratio, AISI Aggregate Index of Systemic Inflammation, SII *Systemic Immune-Inflammation Index, *SIRI Systemic Inflammation Response Index, PCOM polycystic ovary morphology, AUC Area under the curve, CI *confidence interval
Fig. 1ROC curves of NLR, AISI, SII, and SIRI for predicting oligo-amenorrhea
Fig. 2ROC curves of NLR, AISI, SII, and SIRI for predicting polycystic ovary morphology
In this study, we investigated the association between low-grade systemic inflammation and the diagnostic criteria of PCOS, namely oligo/amenorrhea, hyperandrogenism and PCOM. Our results show that women with oligo/amenorrhea have higher NLR, AISI, SII and SIRI values. Similarly, women with PCOM were found to have higher inflammatory markers. These results suggest an association between low-grade inflammation and these important clinical findings. Our study represents a unique contribution to the literature as it is one of the first studies to examine composite inflammatory markers in groups with PCOS diagnostic criteria.
There are numerous studies indicating that PCOS is an inflammatory disease and is associated with an immune response [7–9, 22–24]. In a meta-analysis by Gnanadass et al. showed that imbalances in various pro- and anti-inflammatory cytokines (tumor necrosis factor (TNF)-α, IL-6, IL-8, IL-10, IL-18, IL-33) and CRP are associated with ovarian dysfunction, alterations in steroidogenesis, abnormalities in follicular development, insulin resistance, obesity and cardiovascular disease risk in PCOS [7]. Similarly, Rudnicka et al. [9] reported that levels of CRP, IL-6, IL-18, TNF-α, white blood cell count, monocytochemotactic protein-1 (MCP-1) and macrophage inflammatory protein-1α (MIP-1α) were significantly higher in women with PCOS compared to healthy controls, indicating low-grade inflammation. In addition, Qu et al. reported that genes associated with immune response, such as IL6R, TLR8 and S100A9, have diagnostic value in PCOS and that there is an increased infiltration of neutrophils and dendritic cells [22]. Luan et al. [23] have emphasized that immune cells and immunoregulatory molecules play a crucial role in the pathogenesis of PCOS, with the immune system being overactivated due to low progesterone levels. This process is accompanied by an imbalance of proinflammatory factors, endothelial dysfunction and leukocytosis. These studies found that cytokines, interleukins, TNF and other markers secreted by cells of the immune system are elevated in patients with PCOS compared to healthy women. However, the use of these markers is impractical and costly.
In recent years, the use of composite inflammatory markers such as SII, SIRI and AISI, which are calculated using complete blood parameters and are more practical and cost-effective, has become increasingly widespread as indicators of inflammation. Yılmaz and colleagues demonstrated that neutrophil count and NLR were elevated in PCOS patients compared to healthy women [25]. Zorlu and colleagues also reported an increased NLR, Delta Neutrophil Index, and Platelet-Lymphocyte Ratio in women with PCOS [24]. A large-scale meta-analysis by Li and colleagues investigating the association between PCOS and NLR also reported similar findings [17]. In our study, we found that NLR, along with AISI, SII, and SIRI, were significantly elevated in patients with PCOS—particularly in those with oligo-/amenorrhea—thereby adding to the existing body of literature. In particular, the highest discriminatory power for AISI was observed in the oligo/amenorrhea group, with a cut-off value of > 227.68, AUC: 0.652 (p = 0.007), a sensitivity of 65.5% and a specificity of 58.8%. In the PCOM group, the highest discriminatory power for NLR was observed, with an AUC of 0.617 (p = 0.018) and a cut-off value of > 2.35. Our results suggest that composite inflammatory markers are significantly increased in both the oligo/amenorrhea and PCOM groups and have a more pronounced discriminatory power, especially in the oligo/amenorrhea group. These results support the possibility that low-grade inflammation may be associated with these clinical features.
Although there are many studies in the literature on the inflammatory pathophysiology of PCOS, studies examining the relationship between inflammation and PCOS diagnostic criteria are quite limited. Velez and colleagues suggest that both systemic and tissue inflammation are present in PCOS patients and may contribute to impaired follicular maturation and ovulation [26]. Our study supports the findings of Velez and colleagues and shows that inflammatory markers are elevated in patients with oligo/amenorrhea. Similarly, Siddiqui et al.‘s review emphasized that chronic low-grade inflammation, oxidative stress and immunologic dysfunction are associated with follicular developmental disorders and ovulatory dysfunction in PCOS [27]. Our results are consistent with this study and show that inflammatory markers are significantly elevated in women with oligo-/amenorrhea and polycystic ovaries. In our multivariate logistic regression analysis adjusted for BMI, AISI, SII, and SIRI remained statistically associated with oligo-/amenorrhea, whereas NLR showed a borderline association (p = 0.060). These findings suggest that the observed relationships between these inflammatory indices and oligo-/amenorrhea are largely independent of adiposity, although the effect sizes were modest, indicating limited clinical relevance.
Studies have shown that women with PCOS have a significantly increased long-term risk of cardiovascular disease, type 2 diabetes and dyslipidemia compared to the general population [28, 29]. in et al. demonstrated that the prevalence of hypertension increased significantly with rising levels of SII, SIRI and AISI. It was emphasized that each increase in standard deviation increased the risk of developing hypertension by about 20% [30]. Additionally, SII and SIRI were shown to be associated with dyslipidemia and the associated risk of cardiovascular disease [31]. In addition, Song and colleagues reported that these indices were significantly higher in patients with type 2 diabetes compared to non-diabetics and were also associated with diabetes complications (e.g. diabetic retinopathy, peripheral arterial disease) [32]. Furthermore, there are publications in the literature suggesting that a healthy lifestyle and dietary habits may reduce SII and SIRI levels and decrease the risk of chronic diseases [33, 34]. In our study, the higher SII, SIRI and AISI levels in the oligo-/novulation group and the PCOM group suggest that these patient groups may be at increased risk of chronic diseases in the future. These results suggest that cardiovascular risks should be closely monitored in women with PCOS, especially in women with elevated inflammation levels (oligo/amenorrhea group and PCOM group). These findings highlight the potential value of patient education regarding dietary and lifestyle changes.
Studies have shown that hyperandrogenism and elevated testosterone levels are associated with alterations in hepatic lipoprotein metabolism, which may lead to an atherogenic lipid profile, endothelial dysfunction, and insulin resistance [35, 36]. hese metabolic changes have also been linked to hypertension, vascular dysfunction, and low-grade inflammation, and they often occur in the context of obesity [35, 36]. In addition, studies have shown that women with PCOS who have a hyperandrogenic phenotype, in combination with obesity and insulin resistance, have been reported to have a higher metabolic risk [37]. Based on this, it could be expected that inflammation levels would be higher in women with hyperandrogenism compared to those without. However, in our study, although the oligo-/amenorrhea group demonstrated significantly higher NLR, AISI, SIRI, and SII values than the non-oligo-/amenorrhea group, no significant differences were observed between the hyperandrogenism and non-hyperandrogenism groups. This finding may initially appear to contradict the literature. One possible explanation is that hyperandrogenism was not entirely absent in the oligo-/amenorrhea group. According to the Rotterdam criteria, PCOS is diagnosed when two of the three main features (oligo/amenorrhea, PCOM, hyperandrogenism) are present; therefore, some patients in the oligo-/amenorrhea group also had hyperandrogenism, which may have contributed to the observed increase in inflammation. In addition, the predominantly non-obese profile of our study population may have attenuated inflammatory differences between hyperandrogenism and non-hyperandrogenism groups. Further prospective studies with larger and more phenotypically diverse cohorts, employing formal PCOS phenotypic classification (phenotypes A, B, C, D), are warranted to clarify these associations.
The strongest aspect of our study is that it is one of the few studies in which inflammation-related composite markers (NLR, AISI, SII, SIRI) were evaluated in groups with PCOS diagnostic criteria including oligo-/amenorrhea, hyperandrogenism, and PCOM. In particular, separate ROC analyses were performed for each group, providing clear and clinically meaningful results regarding the discriminatory ability of these markers. However, our study also has some limitations. As the study is single center and cross-sectional, causal relationships cannot be established. Moreover, conducting the study in a single tertiary referral center may limit the ability to fully reflect the broader demographic and clinical variability of women with PCOS in the general population. Another important limitation is the lack of data on the long-term development of chronic diseases in the patients. The ROC AUC values for most markers (0.60–0.65) and the generally small-to-moderate effect sizes indicate that, although statistically significant, the observed differences are modest. This suggests that these markers are unlikely to serve as standalone diagnostic tools but may hold value when combined with other clinical and biochemical parameters. While the modest AUC values limit their direct diagnostic utility, they could still contribute to broader risk stratification frameworks in PCOS, potentially including the monitoring of long-term cardiometabolic risk-though this remains speculative and should be confirmed in prospective longitudinal studies. Overall, our findings enhance understanding of the role of inflammation in PCOS, particularly in patients with oligo-/amenorrhea, and may guide the design of future research in this area.
In conclusion, our study demonstrated that composite inflammation markers (NLR, AISI, SII, SIRI) were significantly higher in women with PCOS and oligo/amenorrhea and PCOM groups, and exhibited stronger discriminative power, particularly in the oligo/amenorrhea group. These findings suggest that low-grade inflammation may be associated with the clinical features of PCOS and that close monitoring of cardiometabolic risks in this patient group may be warranted. In particular, it is recommended that women with PCOS and elevated inflammation levels be informed about the importance of lifestyle and dietary changes and that appropriate strategies be planned for the management of cardiometabolic risks. Further support from larger, multicenter, and long-term follow-up studies is essential to establish the generalizability of these findings.