Ethical approval

After obtaining informed consent from all study groups, blood samples were collected. The study protocol was reviewed and accepted by the Middle Technical University (MTU)/Medical Ethics Committee in Al Zafraniya, Baghdad, Baghdad Governorate, P.C.: 10074. The Ethics Committee Reference (Nr.37) was approved on 27/10/2025. The procedures were performed in strict obedience to the ethical standards set by the WMA Declaration of Helsinki, ensuring the dignity, rights, and privacy of all participants were protected.

Study design and setting

The current case-control study was carried out at the Yarmouk Teaching Hospital/Infertility and IVF Center, Baghdad, Iraq. The case-control design was selected due to its efficiency in studying rare outcomes and its ability to assess associations between exposures (inflammation and HCMV infection) and disease presence (PCOS).

Sample size determination

G*Power version 3.1 [11] was used to conduct sample size calculation to ensure sufficient statistical power (80%) and to detect a medium effect size (Cohen’s d = 0.5) at a significance level of α = 0.05. Based on the foregoing, a minimum study population of n = 160 was required. Eventually, the sample size was increased to n = 168 participants to account for probable dropouts and incomplete data.

Participants and eligibility criteria

A total of (n = 168) women aged from 18 to 40 years were recruited and grouped into four categories:

• Obese women with PCOS (n = 44).

• Normal-weight women with PCOS (n = 40).

• Obese women without PCOS (n = 38).

• Healthy controls (n = 46).

PCOS was detected according to [12], requiring as a minimum two of the following:

• Oligo- or anovulation.

• Clinical and/or biochemical signs of hyperandrogenism; the morphology of Polycystic ovarian was as follows (≥ 12 follicles measuring (2–9 mm) or ovarian volume > 10 cm³ with ultrasound).

Exclusion criteria were as follows: individuals diagnosed with chronic illnesses and diseases including cardiovascular disease, diabetes mellitus, individuals suffering from current infections, those individuals on hormonal therapy within the last three months, and those individuals with a family history of diabetes or PCOS.

Confounding variables and bias mitigation

Structured questionnaire was used to reveal probable confounders and were further adjusted in statistical analysis, e.g., dietary habits and physical activity. Bias in measurement was further diminished by blinding the lab workers who carried out biochemical and immunological analyses to participant groups.

Data collection and clinical assessment

Clinical and physical checkups such as weight, height, and body mass index (BMI), were checked by trained healthcare specialists by means of calibrated equipment. BMI was derived from weight (kg) divided by height squared (m²) and categorized based on Asia-Pacific criteria [13]:

• Normal weight: 18.5–22.9 kg/m².

• Overweight: 23–24.9 kg/m².

• Obesity: ≥ 25 kg/m².

Waist circumference and blood pressure were also measured to justify cardiovascular risk factors.

Sample collection and laboratory analysis

An amount of 10 mL of venous blood samples was drawn between 8:00 and 10:00 AM to examine the daily disparities in hormone and cytokine levels. Samples were then centrifuged at 3,000 rpm for 10 minutes. Serum aliquots were stored at −20°C until further use.

Biochemical and hormonal assays

• Lipid profiles and glucose: estimated via enzymatic colorimetric methods (Roche Diagnostics, Switzerland) following guidelines of Clinical and Laboratory Standards Institute (CLSI).

• Hormone evaluation: follicle-stimulating hormone (FSH), luteinizing hormone (LH), prolactin (PRL), and testosterone levels were detected by enzyme-linked fluorescent assay (ELFA) using the MiniVidas^® system [14].

• Insulin Resistance: examined using the homeostatic model assessment for insulin resistance (HOMA-IR) formula [15].

HOMA-IR = [Fasting insulin (μU/mL) × Fasting glucose (mg/dL)]/405

Inflammatory marker analysis

Serum TNF-α and IL-18 concentrations were determined using commercial enzyme-linked immunosorbent assay (ELISA) kits following the manufacturer's protocol. Sensitivities of 2.3 pg/mL for TNF-α and 0.31 pg/mL for IL-18, with intra- and inter-assay coefficients of variation < 10%, were documented using this method.

HCMV serology

OnSite CMV IgG/IgM Rapid Test (CTK Biotech Inc., 2021) was utilized to measure HCMV IgG and IgM antibodies with a sensitivity of 96% and specificity of 98% [16]. Positive findings were confirmed via ELISA.

Statistical analysis

Data analysis was performed using IBM SPSS Statistics (Version 26.0; IBM Corp. Armonk, NY, USA). Continuous variables were expressed as mean ± standard deviation (SD), whereas categorical variables were summarized using frequencies and percentages. To evaluate group-wise differences across the four cohorts for continuous parametric variables presented, a one-way Analysis of Variance (ANOVA) was executed. Upon identifying significant main effects (p < 0.05), Tukey’s post-hoc test was applied to perform pairwise multiple comparisons among the groups (obese with PCOS, normal-weight with PCOS, obese controls, and healthy controls). Tukey’s test was specifically selected to strictly control the family-wise type I error rate during multiple testing, thereby confirming specific inter-group statistical differences. Student’s t-test was utilized for two-group comparisons where applicable. Relationships between cytokines, biochemical markers, and HCMV seropositivity were evaluated using Pearson’s correlation coefficient. Furthermore, multivariate regression models were constructed to adjust for potential confounding covariates, such as age and BMI. Statistical significance was defined as p < 0.05 (two-tailed), with corresponding 95% confidence intervals (CIs) provided for key outcomes. Participant flow, eligibility assessment, exclusion criteria (n = 32 excluded out of n = 200 assessed), and final allocations across the four groups (total n = 168; obese with PCOS: n = 44, normal-weight with PCOS: n = 40, obese controls: n = 38, healthy controls: n = 46) are detailed in the CONSORT-style flow chart (Figure 1).

Display full size

Figure 1. Flow diagram of participant recruitment demonstrating eligibility assessment, follow-up, analysis phases, exclusion criteria, recruitment, randomization, and group allocation. PCOS: polycystic ovary syndrome.

Participant characteristics

A total of n = 168 participants were eligible and met the inclusion criteria and enrolled in the study out of n = 200 women as an initial number for participants. Baseline characteristics of the study population are illustrated in Table 1.

Among age groups (p = 0.11; 95% CI: –1.5 to 2.3 years), there were no statistically significant differences.

As anticipated, BMI was significantly higher in the obese PCOS group (mean: 29.4 ± 3.2 kg/m²) compared to the normal-weight PCOS group (21.2 ± 1.9 kg/m²; p < 0.001, Cohen’s d = 2.76), indicating a large size effect. Systolic and diastolic blood pressures were elevated in obese groups; however, it did not reach clinical significance (p = 0.06).

Primary outcomes: inflammatory markers and HCMV seropositivity

As demonstrated in Figure 2, pro-inflammatory cytokine levels were remarkably higher in the PCOS cohorts than in the healthy control groups.

Display full size

Figure 2. Serum concentrations (pg/mL) of tumor necrosis factor-alpha (TNF-α) and interleukin-18 (IL-18) across the different study groups. Bars represent the mean values ± standard deviation (SD). Statistical significance lines with asterisks indicate pairwise post-hoc comparisons based on Tukey’s test following One-way ANOVA (*** p < 0.001, ** p < 0.01). PCOS: polycystic ovary syndrome.

• TNF-α levels: Obese women with PCOS showed the highest TNF-α concentrations (25.14 ± 3.7 pg/mL; 95% CI: 23.8–26.5) compared to normal-weight PCOS individuals (13.23 ± 0.25 pg/mL; p < 0.001, Cohen’s d = 3.10) and controls (3.03 ± 0.38 pg/mL; p < 0.001, Cohen’s d = 5.62), indicating large effect sizes.

• IL-18 levels: The highest IL-18 concentrations were found in the obese PCOS group (1.10 ± 0.5 pg/mL; 95% CI: 0.98–1.22), significantly exceeding normal-weight PCOS (0.56 ± 0.06 pg/mL; p = 0.003) and control participants (0.35 ± 0.08 pg/mL; p < 0.001).

Secondary outcomes: biochemical and hormonal parameters

Correlation and regression analyses

A robust positive association was demonstrated using Pearson’s correlation analysis between TNF-α and insulin resistance, using HOMA-IR (r = 0.67; p < 0.001), signifying the role of systemic inflammation in deteriorating insulin sensitivity. Furthermore, a significant correlation was found between IL-18 levels and BMI (r = 0.52; p = 0.004), underscoring the role of adiposity in the increase in inflammatory cytokine.

Inflammatory profile and metabolic interplay in PCOS

In this study, we investigated a previously unexplored relationship between PCOS and HCMV infection. Among Iraqi women with PCOS, we looked at metabolic variables, inflammatory markers (TNF-α and IL-18), and HCMV seropositivity. Our findings indicate that TNF-α and IL-18 levels are much greater in women with PCOS, particularly those who are obese. In addition, we also discovered that compared to healthy women, women with PCOS exhibited greater frequencies of HCMV IgG and IgM seropositivity. Interestingly, HCMV seropositivity was associated with increased TNF-α levels even after adjusting for factors including insulin resistance and BMI. These findings imply that viral infections and chronic inflammation may be important factors in PCOS development.

Participants with PCOS and those with obesity exhibited much higher levels of the pro-inflammatory cytokines TNF-α and IL-18. TNF-α levels in obese PCOS patients (25.14 ± 3.7 pg/mL) were about eight times higher than in healthy controls (3.03 ± 0.38 pg/mL, p < 0.001). The rise in TNF-α and IL-18 is consistent with previous research, e.g., [3, 4]. that identified chronic low-grade inflammation as a major characteristic of PCOS. Insulin resistance may have resulted from high TNF-α levels interfering with insulin receptor signalling, e.g., [5]. That might be because of how the body handles fats and glucose; IL-18 also contributes to metabolic issues [17]. In this study, we also discovered a high positive correlation between TNF-α and HOMA-IR, which lends credence to the theory that systemic inflammation exacerbates metabolic problems in PCOS.

HCMV seroprevalence and its immunological implications

The role of HCMV in promoting inflammation remains under discussion, but our results add to the evidence linking HCMV to PCOS. In the current study, we observed that individuals with PCOS had a much higher amount of HCMV IgG antibodies than the control group. In particular, 68% of those obese participants with PCOS tested positive for HCMV, compared to 42% of healthy participants (p = 0.03). Research by [10] findings are in line with our findings who also found that viruses were more prevalent in PCOS cases. This was also asserted by authors, e.g., [6, 8]. HCMV stays dormant in the body and reactivates in response to stress or immunosuppression.

Higher TNF-α levels were independently predicted by HCMV status, as our regression analysis demonstrates (β = 0.45, p = 0.01). This suggests that latent viral infections and the persistent low-level inflammation frequently observed in PCOS may be related. Scientifically speaking, HCMV can cause inflammation by triggering the release of several cytokines through the activation of toll-like receptors [18]. This could explain why our seropositive subjects had higher TNF-α levels. Some researchers, such as [19], suggested that chronic viral infections may affect ovarian function or metabolism via oxidative stress; however, this was not confirmed in our study. It is noteworthy to understand that our research does not allow us to draw strong conclusions about cause and effect. As [20] pointed out, results in this area can vary, which might be because of differences in viral strains or genetics. Further polymerase chain reaction (PCR) research is needed in order to clarify the role of HCMV.

Lipid metabolism and divergent clinical findings

Interestingly, despite the elevated concentrations in inflammation and insulin resistance markers, LDL and HDL cholesterol levels were similar across all groups. This finding is not in alignment with a study that found dyslipidemia is common in PCOS [21]. One explanation might be that the individuals who were younger had received a PCOS diagnosis more recently or had genetic variations unique to this population. Changes in cholesterol levels may have also been avoided by other unmeasured factors such as food or exercise.

Clinical implications and future therapeutic perspectives

Clinically, these results suggest that a team approach is the most effective way to manage PCOS. Reducing inflammation and enhancing insulin sensitivity still depend on controlling weight through diet and increased exercise. In PCOS patients with unexplained inflammation or metabolic issues that do not improve with standard therapies, screening for chronic viruses like HCMV may be helpful because viral infections may contribute to persistent inflammation. Nonetheless, it is important to consider the practicality and affordability of such screening, particularly in nations with limited healthcare resources. Antiviral therapies may be an effective addition to existing treatment regimens. However, further study is required to ascertain their long-term safety, efficacy, and benefits for PCOS.

It can be difficult to use these findings in clinical settings. For example, the expense of HCMV screening and restricted access to antiviral drugs may be barriers in some contexts, and patients’ adherence to complex treatment plans, including antiviral or anti-inflammatory therapies, can also vary. This highlights the need for patient education. Once individuals comprehend the interplay of chronic inflammation and probable viral factors in PCOS, they may act positively in their treatment and make better lifestyle choices. As suggested elsewhere by [1], patient outcomes might be improved by adopting healthier lifestyle changes such as weight loss. Our findings reveal the importance of partnership collaboration of endocrinologists, infectious disease experts, and dietitians to work as a team for better PCOS health outcomes.

Limitations section

This research has some strengths; however, a few limitations can still be noted. Firstly, a direct cause-and-effect relationship between HCMV infection and PCOS cannot be defined as a result of using case-control. Therefore, our findings should be seen as an association, rather than proof of causation. Secondly, in the current study, HCMV status was detected using IgG/IgM serology, which as a result, represents past exposure and potential latency, unlike PCR, and does not present the quantity of infectious agent. As a result, this made it even more difficult to understand the interplay of the virus and persistent inflammation. Thirdly, other possible confounding factors such as physical activity and detailed lifestyle habits in addition to the BMI and age could have been included in our model to obtain a deeper understanding.

Conclusion

The current research demonstrates a robust indication of how chronic inflammation, viral infections, and metabolic problems interact in women with PCOS. Higher levels of TNF-α and IL-18, especially in women with obesity, underscore the role of inflammation in disease development.

The strong association found here between HCMV seropositivity and elevated inflammatory parameters proposes the role that latent viral infections might play in the ongoing inflammation, especially in those individuals suffering from PCOS.

We present here a new research area with probable clinical importance. Health outcomes for patients with PCOS could be improved by monitoring metabolic and inflammatory aspects of these patients and including viral screening and targeted treatments. Further investigations might be needed to reveal the causes and to find an effective, accessible treatment plan for different PCOS patient groups.