Impact of subclinical hypothyroidism on endocrine features in patients with polycystic ovary syndrome

Abstract

Background

Polycystic ovary syndrome (PCOS) and subclinical hypothyroidism (SCH) are both common endocrine disorders. This study investigates the impact of SCH on the endocrine features of patients diagnosed with PCOS.

Methods

This retrospective study included 124 women diagnosed with PCOS between January 2020 and November 2022. Participants were divided into two groups: those with PCOS alone (n = 93) and those with both PCOS and SCH (n = 31).Clinical parameters (age, body mass index, blood pressure, age at menarche) and endocrine markers were collected. Hormonal measurements included follicle-stimulating hormone (FSH), luteinizing hormone (LH), estradiol (E2), prolactin (PRL), testosterone (TES), progesterone (PRG), TSH, free triiodothyronine (FT3), FT4, and Anti-Müllerian hormone (AMH). Thyroid autoantibodies, including antithyroid peroxidase (TPO-Ab) and antithyroglobulin antibody (TG-Ab), were also assessed. Antral follicle count (AFC) was evaluated using transvaginal ultrasound.

Results

The study found that patients with PCOS and SCH exhibited significantly higher levels of TSH and PRL compared to those with PCOS alone. However, there were no significant differences in PRG, LH, FSH, TES, and E2 levels between the groups. The PCOS group showed a moderate positive correlation between TSH and AMH, whereas the PCOS with SCH group demonstrated a negative correlation between TSH and FSH, albeit not statistically significant.

Conclusions

Subclinical hypothyroidism in women with PCOS is associated with altered thyroid–prolactin axis activity and a disrupted correlation between TSH and AMH, while gonadotropin levels and ovarian morphology appear unaffected. These findings warrant confirmation through prospective studies.

Introduction

Polycystic ovary syndrome (PCOS) is a heterogeneous endocrinopathy affecting 6–12% of women of reproductive age, characterised by oligo- or anovulation, clinical and/or biochemical hyperandrogenism, and polycystic ovarian morphology on ultrasound per the Rotterdam Criteria [1, 2]. Beyond its hallmark androgen excess and elevated LH-to-FSH ratio, PCOS is frequently accompanied by insulin resistance and dyslipidaemia, contributing to an adverse metabolic milieu that predisposes affected women to type 2 diabetes and cardiovascular disease [3]. Understanding the dual endocrine–metabolic nature of PCOS is essential for elucidating its pathogenesis and guiding comprehensive management strategies [4].

The 2023 International Evidence-based Guideline for PCOS emphasises the importance of excluding thyroid dysfunction during diagnosis, recommending measurement of TSH to rule out overt and subclinical hypothyroidism prior to confirming PCOS under the revised Rotterdam criteria (TSH reference range 0.4–4.0 mIU/L) [5]. Women with PCOS face a two- to three-fold higher risk of infertility even when ovulatory, attributable in part to altered oocyte competence and embryo developmental potential, as well as compromised endometrial receptivity characterised by dysregulated sex-steroid receptor expression, impaired glucose uptake, and chronic low-grade inflammation [6]. Moreover, PCOS is associated with increased obstetric complications—including gestational diabetes, pre-eclampsia, and preterm birth—likely driven by the interplay of hyperandrogenism, insulin resistance, and metabolic derangements [7]. Experimental and clinical data further implicate suboptimal oocyte competence—variable across PCOS phenotypes and metabolic profiles—as a key determinant of subfertility and ART outcomes [8]. Taken together, these findings underscore the need to clarify how even mild thyroid impairment (SCH) may exacerbate the endocrine-metabolic and reproductive abnormalities inherent to PCOS [9].

Despite extensive research on PCOS, the data remain limited regarding the impact of SCH in this population [10]. Few studies have specifically compared endocrine outcomes in PCOS women with and without SCH, and existing reports have yielded conflicting results on how mild thyroid dysfunction modulates gonadotropin secretion, steroidogenesis, and ovarian morphology [11, 12]. Moreover, the potential disruption of the thyroid–prolactin axis and its downstream effects on ovarian reserve markers such as AMH and AFC have not been systematically evaluated in PCOS cohorts. This study aimed to investigate, in a retrospective cohort of PCOS patients with and without subclinical hypothyroidism, the impact of SCH on thyroid–prolactin axis function, sex‐steroid hormone profiles, and ovarian reserve markers, with the goal of refining clinical screening and optimizing individualised management strategies for reproductive and metabolic outcomes.

Materials and methods

Study design

A comprehensive retrospective analysis was undertaken at our institution to investigate the impact of SCH on the endocrine characteristics of patients diagnosed with PCOS. This study spanned from January 2020 through November 2022. The cohort was divided into two groups: patients with concurrent SCH and PCOS formed the observation group, while those with PCOS alone constituted the control group. The exclusion criteria include: patients with a history of thyroidectomy or radioactive iodine treatment; those receiving thyroid hormone replacement or antithyroid medication; individuals with known pituitary or adrenal disorders; pregnant or breastfeeding women; and patients with other significant systemic illnesses such as severe liver or renal disease. Informed consent was obtained from all subjects. The study's protocols and methodology were rigorously reviewed by our hospital's ethics committee (Approval No. 20240723083), ensuring compliance with all applicable guidelines and the Declaration of Helsinki's principles for human research. All methods adhered to strict ethical standards, and data confidentiality was maintained by anonymizing personal identifiers before analysis to safeguard participant privacy.

Diagnosis criteria for polycystic ovary syndrome and subclinical hypothyroidism

The Rotterdam Criteria, established in 2003 by the European Society for Human Reproduction and Embryology (ESHRE) and the American Society for Reproductive Medicine (ASRM), serve as the foundation for diagnosing PCOS. A diagnosis of PCOS necessitates the presence of at least two of the following three criteria: Oligo-ovulation or Anovulation, characterised by irregular or absent ovulation, often evidenced by menstrual cycles exceeding 35 days or fewer than eight cycles per year; Clinical or Biochemical Hyperandrogenism, where elevated androgen levels are detected through blood tests or manifested as hirsutism, acne, or androgenic alopecia; and Polycystic Ovaries on Ultrasound, indicated by the presence of 12 or more follicles each measuring 2–9 mm in diameter in each ovary and/or an increased ovarian volume.

SCH is diagnosed through thyroid function tests, primarily focusing on two key parameters [13]: Elevated TSH Levels, where TSH levels are above the upper limit of the laboratory reference range, typically below 10 mIU/L, indicating a mild thyroid function impairment; and Normal Free Thyroxine (Free T4) and/or Triiodothyronine (T3) Levels, where, despite the elevated TSH, the levels of these hormones remain within the normal laboratory range, suggesting a compensated state of thyroid function.

Assessment of hormonal profiles and antral follicle count in polycystic ovary syndrome and subclinical hypothyroidism

Venous blood samples were collected in the early morning on an empty stomach from participants on days 2 to 5 of their menstrual cycle, or from amenorrheic patients when ultrasound examination confirmed the absence of dominant follicles and physiological cysts. Hormonal levels including serum Follicle Stimulating Hormone (FSH), Luteinizing Hormone (LH), Estradiol (E2), Prolactin (PRL), Testosterone (TES), and Progesterone (PRG) were measured using the Siemens fully automated chemiluminescence immunoassay system from Germany, along with the corresponding reagents. Quality control was maintained using Bio-Rad control products.

The concentrations of Thyroglobulin Antibody (TG-Ab) and Thyroid Peroxidase Antibody (TPO-Ab) were also determined. The presence of thyroid autoantibodies (TAA) was defined as TPO-Ab > 35 U/ml and/or TG-Ab > 40 U/ml. Serum Anti-Müllerian Hormone (AMH) levels were assessed using the Roche Cobas e411 electrochemiluminescence immunoassay system and its associated reagents, with quality control ensured through Roche control products. In addition, Antral Follicle Count (AFC) was measured utilizing the Siemens X-100 colour Doppler ultrasound diagnostic system.

Clinical data collection

Clinical and demographic data were extracted from patients’ medical records at their initial clinical visit. This included age, body mass index (BMI), age at menarche, and blood pressure. BMI was calculated using measured height and weight (weight in kilograms divided by height in meters squared). Systolic and diastolic blood pressures were measured in a seated position after at least 5 min of rest using an automated sphygmomanometer, as part of routine clinical evaluation. All measurements were performed by trained nursing staff in accordance with hospital protocols.

Statistical analysis

Quantitative data in this study were represented as mean ± standard deviation (x̄ ± s). To compare groups, one-way analysis of variance (ANOVA) was employed. This was followed by specific post-hoc tests, namely the Student–Newman–Keuls (S–N–K) and the Least Significant Difference (LSD) tests, to determine significant differences between the groups. The S–N–K test, used for multiple comparisons, is effective in controlling the overall Type I error rate, while the LSD test, offering more power, is used when variances are equal and sample sizes are similar. The choice between these post hoc tests was determined based on the data's normality and variance homogeneity. In addition, Pearson regression analysis was utilised to explore and quantify the correlations between various variables, providing insights into the linear relationships within the dataset. This comprehensive statistical approach ensures a thorough and accurate analysis, crucial for validating the study's findings. All hypotheses were two-tailed, and a p value threshold of less than 0.05 was applied to ascertain statistical significance.

Results

Comparative clinical characteristics of polycystic ovary syndrome and polycystic ovary syndrome with subclinical hypothyroidism

In our study, the analysis of clinical parameters between the PCOS group (n = 93) and the PCOS with SCH group (n = 31) demonstrated no significant statistical differences, indicating comparability between the groups. The mean age was similar in both groups, with the PCOS group averaging 28.45 ± 3.57 years and the PCOS + SCH group at 28.17 ± 3.45 years. Body Mass Index (BMI) comparisons also showed minimal variation, being 23.31 ± 3.50 kg/m² in the PCOS group and 23.56 ± 3.67 kg/m² in the PCOS + SCH group. Systolic and diastolic blood pressure measurements were closely aligned, with the PCOS group recording 120.5 ± 13.2 mmHg and 76.2 ± 8.4 mmHg, respectively, and the PCOS + SCH group showing 121.8 ± 14.1 mmHg and 76.9 ± 8.7 mmHg. In addition, the age of menarche was comparable between the two groups, being 13.49 ± 2.42 years in the PCOS group and 13.71 ± 2.53 years in the PCOS + SCH group (Table 1). These findings suggest that the general clinical characteristics of patients with PCOS are consistent, regardless of the coexistence of SCH.

Table 1.

Baseline characteristics of PCOS and PCOS with SCH Groups

Parameter	PCOS Group (n = 93)	PCOS + SCH Group (n = 31)
Age (years)	28.45 ± 3.57	28.17 ± 3.45
BMI (kg/m2)	23.31 ± 3.50	23.56 ± 3.67
Systolic BP (mmHg)	120.5 ± 13.2	121.8 ± 14.1
Diastolic BP (mmHg)	76.2 ± 8.4	76.9 ± 8.7
Menarche Age (years)	13.49 ± 2.42	13.71 ± 2.53

Hormonal differences in polycystic ovary syndrome and polycystic ovary syndrome with subclinical hypothyroidism

In the study comparing PCOS patients with and without SCH, significant hormonal differences were noted. The PCOS + SCH group showed a marked increase in TSH levels (6.90 ± 3.37 mU/L) compared to the PCOS group (2.49 ± 0.88 mU/L, P < 0.05), indicating notable thyroid dysfunction. Additionally, PRL levels were significantly higher in the PCOS + SCH group (16.11 ± 5.45 ng/ml) compared to the PCOS group (13.22 ± 5.87 ng/ml, P < 0.05). However, PRG, LH, FSH, TES, and E2 levels showed no significant differences between the two groups. FT4 levels were slightly lower in the PCOS + SCH group (1.08 ± 0.17 ng/dl) versus the PCOS group (1.12 ± 0.14 ng/dl, P < 0.05), but FT3 levels and AFC remained similar across groups. These results highlight a complex hormonal interaction in PCOS, especially when accompanied by SCH, and underscore the potential clinical management and treatment implications for these patients In terms of thyroid autoimmunity, TAA positivity was observed in 11 out of 31 patients (35.5%) in the PCOS + SCH group and in 10 out of 93 patients (10.8%) in the PCOS group. This difference was statistically significant (P < 0.05), indicating a higher prevalence of thyroid autoimmunity among PCOS patients with concomitant SCH (Table 2).

Table 2.

Comparison of basal endocrine hormones of PCOS and PCOS with SCH Groups

Hormone	PCOS Group (n = 93, x̄ ± s)	PCOS + SCH Group (n = 31, x̄ ± s)	P value
TSH (mU/L)	2.49 ± 0.88	6.90 ± 3.37	< 0.05
TES (ng/dl)	34.03 ± 18.04	34.30 ± 13.54	0.897
PRL (ng/ml)	13.22 ± 5.87	16.11 ± 5.45	< 0.05
PRG (ng/ml)	0.40 ± 0.20	0.41 ± 0.23	0.675
LH (U/L)	11.52 ± 5.83	9.36 ± 4.72	0.126
FT4 (ng/dl)	1.12 ± 0.14	1.08 ± 0.17	< 0.05
FT3 (pg/ml)	4.22 ± 0.91	4.17 ± 0.85	0.861
FSH (U/L)	5.92 ± 1.78	6.45 ± 1.46	0.267
E2 (pg/ml)	55.38 ± 29.67	53.49 ± 20.36	0.088
AFC	30.58 ± 7.58	30.69 ± 9.19	0.376
TAA positivity	10 (10.8%)	11 (35.5%)	< 0.05

TSH Thyroid-Stimulating Hormone, FT3 Free Triiodothyronine, FT4 Free Thyroxine, FSH Follicle-Stimulating Hormone, LH Luteinizing Hormone, E2 Estrogen, TES Testosterone, PRL Prolactin, PRG Progesterone, AMH Anti-Müllerian Hormone, AFC Antral Follicle Count, TAA Thyroid autoantibody

Analysis of thyroid-stimulating hormone correlation with hormonal profiles in polycystic ovary syndrome and polycystic ovary syndrome with subclinical hypothyroidism

Our study examined the correlation between TSH levels and various reproductive hormones, along with ovarian reserve markers, in patients with PCOS and those with PCOS complicated by SCH. The analysis revealed distinct correlation patterns within these groups. In the PCOS group, a moderate positive correlation was observed between TSH and AMH, indicating a potential link between thyroid function and ovarian reserve. However, other hormones such as FSH, LH, E2, TES, PRL, and PRG showed no significant correlation with TSH levels. In contrast, the PCOS with SCH group demonstrated a negative correlation between TSH and FSH, though not statistically significant. Similar to the PCOS group, correlations between TSH and other hormones were generally weak and not statistically significant, suggesting that SCH may not significantly impact the overall hormonal interplay in PCOS patients (Table 3).

Table 3.

Correlation of TSH with reproductive hormones and ovarian reserve in PCOS and PCOS with SCH Groups

Hormone	PCOS Group (n = 93) PCOS + SCH Group (n = 31)
	r value	P value	r value	P value
FSH	0.137	0.315	− 0.269	0.237
LH	− 0.057	0.633	− 0.02	0.891
E2	− 0.116	0.413	− 0.068	0.658
TES	0.061	0.412	0.053	0.734
PRL	− 0.071	0.294	− 0.057	0.654
PRG	− 0.13	0.482	− 0.048	0.76
AMH	0.318	0.054	0.299	0.286
AFC	− 0.052	0.686	0.084	0.63

TSH Thyroid-Stimulating Hormone, FSH Follicle-Stimulating Hormone, LH Luteinizing Hormone, E2 Estrogen, TES Testosterone, PRL Prolactin, PRG Progesterone, AMH Anti-Müllerian Hormone, AFC Antral Follicle Count

Discussion

SCH and PCOS are two endocrine disorders that have shown a notable association, particularly with an increased incidence of SCH in women with PCOS. This relationship suggests a potential interplay between thyroid dysfunction and the pathophysiology of PCOS. SCH, characterised by elevated serum TSH levels with normal thyroid hormone concentrations, may influence the hormonal milieu of PCOS, potentially exacerbating its symptoms like menstrual irregularities and metabolic disturbances [14]. The innovative aspect of exploring the link between SCH and PCOS lies in the potential for new therapeutic strategies. Understanding this relationship could lead to more targeted treatment approaches, addressing not only the reproductive and metabolic symptoms of PCOS but also the underlying thyroid dysfunction [15, 16]. Clinically, this could improve patient outcomes by providing a more comprehensive management plan for women with PCOS, particularly those also suffering from SCH. This dual-focus approach underscores the importance of routine screening for thyroid function in women diagnosed with PCOS, ensuring a holistic treatment strategy.

The notable increase in TSH levels in the PCOS with SCH group compared to the PCOS group points towards a significant thyroid dysfunction in the former. SCH, characterised by elevated TSH levels, may exacerbate the endocrine imbalance inherent in PCOS. The increased TSH levels could be indicative of a compensatory mechanism in response to suboptimal thyroid function, potentially impacting metabolic and reproductive processes. Thyroid hormones play a critical role in regulating menstrual cycles and ovulation, and disturbances in these hormones may exacerbate PCOS symptoms such as irregular menstruation and ovulatory dysfunction. The elevated PRL levels in the PCOS with SCH group further complicate the hormonal milieu [17]. Hyperprolactinaemia, even of a mild degree, can interfere with gonadotropin-releasing hormone (GnRH) secretion, potentially leading to disruptions in menstrual and ovulatory cycles. This could manifest as further irregularities in menstrual cycles and reduced fertility in women with PCOS and SCH, making the management of these patients more challenging [18].

The positive correlation between TSH and AMH observed in the PCOS group is intriguing. AMH, a marker of ovarian reserve, is typically higher in PCOS due to an increased number of antral follicles. The correlation with TSH may suggest that even subclinical variations in thyroid function could influence ovarian reserve. This finding could have implications for fertility management in PCOS, as it hints at a possible link between thyroid health and ovarian response, particularly in the context of assisted reproductive technologies [19]. The absence of significant correlations between TSH and most reproductive hormones in the PCOS with SCH group is noteworthy. It suggests that SCH may not significantly alter the hormone profile in PCOS beyond the changes attributed to PCOS itself. This finding could imply that the primary endocrine disturbances in PCOS, such as hyperandrogenism and insulin resistance, may overshadow the subtler hormonal influences of SCH.

Furthermore, the significantly higher prevalence of TAA positivity observed in the PCOS + SCH group compared to the PCOS-only group highlights the potential contribution of autoimmune thyroiditis to the pathophysiology of reproductive endocrine disturbances in this population. Thyroid autoimmunity—particularly the presence of TPO-Ab and TG-Ab—has been shown to exert both direct and indirect effects on ovarian function through immune-mediated mechanisms [20]. These may include localised inflammation, altered cytokine profiles, and cross-reactivity between thyroid and ovarian antigens. Several studies have proposed that autoimmune thyroid dysfunction may impair folliculogenesis, disrupt ovulatory cycles, and affect oocyte quality, even in the absence of overt hypothyroidism [21, 22]. In the context of PCOS, which is already characterised by an abnormal follicular environment, the presence of thyroid autoimmunity may further exacerbate reproductive dysfunction and contribute to subfertility [23]. Therefore, routine screening for thyroid autoantibodies in women with PCOS—particularly those exhibiting subclinical hypothyroidism—may aid in the early identification of individuals at increased risk of ovarian dysfunction and may inform more tailored therapeutic strategies, including the potential benefit of levothyroxine in TAA-positive individuals.

The complex interaction between thyroid function and PCOS pathophysiology is likely multifactorial. PCOS is associated with insulin resistance, obesity, and chronic low-grade inflammation, all of which can influence thyroid function. Conversely, thyroid hormones can impact insulin sensitivity, body weight, and inflammatory processes, creating a bidirectional relationship. In PCOS with SCH, this interplay may be further accentuated, leading to more pronounced endocrine abnormalities [24]. Clinically, these findings highlight the importance of comprehensive hormonal assessment in PCOS, particularly when symptoms suggest thyroid dysfunction. Although SCH may not drastically alter the PCOS hormone profile, its presence could potentially exacerbate symptoms and complicate management [25]. This underscores the need for a holistic approach to PCOS treatment, addressing not only the reproductive aspects but also associated comorbidities like thyroid dysfunction.

This study has several strengths. First, the inclusion criteria were rigorously applied based on the standardised Rotterdam Criteria for PCOS and well-established biochemical thresholds for SCH, ensuring diagnostic clarity and homogeneity across study groups. Second, we conducted comprehensive endocrine profiling, including a wide array of reproductive and thyroid-related hormones (TSH, FT4, FT3, LH, FSH, E2, PRL, TES, PRG), thyroid autoantibodies (TPO-Ab, TG-Ab), and ovarian reserve indicators (AMH, AFC), all assessed using validated and quality-controlled immunoassays. Third, blood sampling was standardised to the early follicular phase (or an equivalent amenorrheic window) to reduce cycle-dependent hormonal fluctuations. Finally, all assays were conducted in a single hospital-based, accredited laboratory, ensuring analytical consistency and minimizing inter-laboratory variability. Together, these methodological features enhance the internal validity and reproducibility of our findings. This study's findings are potentially constrained by several factors. Firstly, the smaller cohort size, especially in the PCOS with SCH group, could affect the robustness and wider applicability of the results. The absence of detailed information on participants'lifestyle choices and existing health conditions also presents a limitation, as these factors might significantly impact the manifestation and interrelation of SCH and PCOS. Furthermore, the reliance on hormonal measurements at a single point in time may not comprehensively represent the participants'long-term hormonal profiles, thus potentially limiting the accuracy of the observed correlations. Further research is required to unravel the precise mechanisms by which thyroid dysfunction impacts PCOS pathophysiology and vice versa. In addition, exploring the interplay between thyroid function, insulin resistance, and chronic inflammation in PCOS could offer new therapeutic targets.

Conclusions

In conclusion, subclinical hypothyroidism further destabilises the thyroid–prolactin axis in women with polycystic ovary syndrome, abolishing the normal positive relationship between TSH and anti-Müllerian hormone while leaving gonadotropin profiles and ovarian morphology essentially unaffected. Large, prospective, multicentre trials are needed to determine whether correction of SCH improves reproductive and metabolic outcomes in this high-risk population.

Author contributions

Rui Hua, Jing-Jun Li, Wei Yang, Yun Hua, and Yu-Long Ma collaborated in conceptualizing the study. Data curation was managed by Rui Hua, Wei Yang, Yun Hua, and Yu-Long Ma. Rui Hua, Jing-Jun Li, Yun Hua, and Yu-Long Ma conducted the formal analysis. The methodology was developed by Rui Hua, Jing-Jun Li, Wei Yang, and Yu-Long Ma. Resources were provided by Rui Hua, Jing-Jun Li, Wei Yang, and Yun Hua. Software handling was undertaken by Rui Hua, Jing-Jun Li, Wei Yang, Yun Hua, and Yu-Long Ma. Rui Hua was responsible for writing the original draft and also handled the review and editing of the manuscript, ensuring consistency and accuracy in the presentation of the study’s findings.

Funding

This study was supported by the National Natural Science Foundation of China (NSFC) [grant number 81960085]. Henan Medical Science and Technology Project No. 2018020931.

Data availability

The data sets generated and analyzed during this study under reasonable requirements the correspondence author can provide.

Declarations

Ethics approval and consent to participate

This study was approved by the Ethics Committee of Xinxiang Central Hospital (Approval No. 20240723083). All study procedures complied with the ethical standards of the institutional and national committees and the Helsinki Declaration. Informed consent was obtained from all participants.

Consent for publication

Written informed consent for publication was obtained from all patients in this retrospective analysis.

Competing interests

The authors declare no competing interests.