Polyendocrine metabolic ovarian syndrome (PMOS), formerly called polycystic ovary syndrome (PCOS), is a common and complicated endocrine-metabolic condition in women of reproductive age.[1] It is a complex interplay of hyperandrogenism, ovulatory dysfunction, insulin resistance, and abnormal ovarian follicular development, resulting in significant reproductive and metabolic morbidity.[2]  Insulin resistance is a key pathophysiological aspect of PMOS and is associated with reproductive dysfunction through compensatory hyperinsulinemia and changes in ovarian production of androgens and in the process of follicular maturation.[3] Notably, metabolic derangements can happen even in healthy-weight women, which means that insulin resistance is not only related to overweight.[4]

In parallel to these metabolic changes, a glycoprotein released mainly by granulosa cells of small growing ovarian follicles, anti-Müllerian hormone (AMH), has received considerable attention as a marker of ovarian follicular activity.[5] In women suffering from PMOS, AMH is often high because there are more small antral follicles and a disruption in the development of follicles.[6] The AMH level in serum was accepted as an alternative method to ultrasound to define polycystic ovarian morphology in adult females in the 2023 international guidelines, but not has an independent diagnostic test.[7] Therefore, the assessment of AMH in serum in addition to insulin resistance might give additional insights into the reproductive and metabolic phenotypes of PMOS and the effects of therapeutic intervention.[3]

Metformin has been used to correct metabolic derangements, which are mainly seen in PMOS, by inhibiting the production of glucose from the liver and by enhancing insulin sensitivity.[8] It has been recommended internationally for its therapeutic benefit, especially in women with metabolic risk factors, and gastrointestinal side effects can impact tolerability and adherence.[9] Meanwhile, myo-inositol, a naturally occurring inositol isomer crucial to insulin signaling intracellularly and physiological functions in the ovary, has been identified as a potential insulin-sensitizing intervention.[10] It is also proposed to affect glucose metabolism, insulin signaling, and ovarian function and has therefore led to growing interest in its application among females with PMOS.[11] A systematic review and meta-analysis in 30 trials and 2,230 subjects found that inositol showed potential for improvement in selected metabolic parameters and had fewer gastrointestinal adverse events than metformin.[12]

The clinical management of PMOS is a specific clinical challenge as the concurrent changes in metabolic functions may not be reflected by the parallel changes in ovarian endocrine function. In spite of the close relationship of both myo-inositol and metformin with insulin resistance pathways, the relative relationship between them and AMH, however, is not well defined and does not allow us to draw a firm conclusion whether or not the observed metabolic improvement is accompanied by measurable changes in ovarian follicular activity. This distinction is significant from a clinical standpoint because the clinical benefit of insulin sensitivity should not be the only criterion used for therapeutic selection, but the overall reproductive and endocrine status of women with the disease should be taken into account. Comparing these interventions can help contribute to more individualised and evidence-based PMOS management. Hence, the present study seeks to assess the differences in the association of myo-inositol versus metformin with anti-Müllerian hormone (AMH) levels and insulin resistance (IR) in women with PMOS, which might help guide therapy decisions.

The study was of the observational type and was done at the Institute of Allied Health Sciences, Faisalabad Medical University, Faisalabad, Pakistan. The study period was six months from February 2026 to July 2026. The sample size was estimated using the OpenEpi framework for comparing two independent means with 95% confidence level, 80% statistical power, and equal allocation ratio (1:1). An RCT by Ravn et al. (2022) was deemed relevant for the selection of study outcomes, as it compared the effects of myo-inositol versus metformin in women with PCOS and included measurements of serum AMH and HOMA-IR.[13] With a standardised mean difference of 0.70 between the treatment groups, the minimum sample size was estimated to be around 33 subjects per group. After adjustment for 10% anticipated incomplete data, the target sample size was 74 participants, comprising 37 women receiving myo-inositol and 37 women receiving metformin. The study subjects were recruited using non-probability consecutive sampling. Women diagnosed with PMOS who visited the participating clinical services within the study period were screened for inclusion/exclusion criteria based on pre-established criteria. The number of participants was determined by the number necessary to obtain the desired sample size, and those eligible were consecutively enrolled. Patients were divided into two groups based on the treatment that was prescribed. Group A consisted of women who were treated with myo-inositol therapy and Group B consisted of women who were given metformin therapy. However, allocation of treatment was not investigator-directed randomization, but rather based on current clinical prescriptions. The inclusion criteria were a diagnosis of PMOS established by the modified Rotterdam diagnostic criteria and age between 18 and 40 years. To confirm diagnosis, three of the following three criteria, after exclusion of other causes, were required: clinical or biochemical hyperandrogenism, irregular menstrual cycles indicative of ovulatory dysfunction, and polycystic ovarian morphology seen on ultrasonography or, as appropriate in older females, on the basis of serum AMH in 2023 international evidence-based guidelines. Women who had been treated with either myo-inositol or metformin were eligible as part of their management as documented by their treatment history and biochemical values. All participants gave informed consent in writing and were willing to participate in the necessary clinical and biochemical evaluations. Pregnant and lactating women, pre-existing type 1 or type 2 diabetes mellitus, or women with any other endocrine disorders which might independently affect ovarian function, serum AMH concentrations, or insulin sensitivity were excluded. These disorders were uncontrolled thyroid disease, hyperprolactinemia, Cushing's syndrome, congenital adrenal hyperplasia and androgen-secreting tumors. Additionally, women who had severe hepatic or renal dysfunction, history of ovarian surgery or early onset of menopause were excluded. Hormonal contraceptive treatment, systemic corticosteroid therapy, fertility-related hormonal therapy and other insulin-sensitising drugs that could affect the outcomes of the study were excluded for those treatments that overlapped with the defined time period of the assessment. Women receiving both myo-inositol and metformin simultaneously were not included to avoid the confounding influence of combined treatment. Screening of the participants was done on the basis of the eligibility criteria and then data were collected with a structured data collection proforma. Information on age, marital status, and menstrual history, history of PMOS, family history of metabolic disorders, reproductive history, and current medications used was obtained. Anthropometric measurements were made using standard methods and body mass index (BMI) was determined as weight in kg divided by the square of the height in meters. They were divided into two groups: those that were prescribed myo-inositol and those that were prescribed metformin. Differences in treatment exposure were accounted for by documenting information related to the prescribed dosage, duration of treatment, adherence, and any adverse effects reported. Blood samples were drawn from the veins after an 8-12 hour fast. Blood was processed per standard laboratory procedure and serum was separated by centrifugation and used for biochemical analysis. The laboratories employed used an enzyme-linked immunoassay (ELISA) or another validated immunoassay to measure serum AMH concentrations. An immunoassay was used to measure fasting serum insulin, and an enzymatic method was used to measure fasting blood glucose. HOMA-IR was determined as an indicator of insulin resistance. Insulin resistance was found to be higher due to higher HOMA-IR. Biochemical parameters were noted and compared between the two groups of myo-inositol and metformin. Association of treatment type with serum AMH concentration and HOMA-IR was assessed, taking into account the important demographic and anthropometric factors. Standardized procedures were used for laboratory assessments to minimize laboratory variability. The data collected were then entered and analyzed with the Statistical Package for the Social Sciences (SPSS) version 26.0. Prior to analysis, data were checked for completeness and consistency, as well as for possible outliers. Normality of continuous variables was tested with the Shapiro–Wilk test and graphing. The quantitative variables such as age, BMI, serum AMH, fasting blood glucose, fasting insulin, and HOMA-IR were presented as either mean ± SD. Categorical variables were reported in terms of frequencies and percentages. The independent-samples t-test was used to compare the means of normally distributed continuous data. The chi-square test and fisher's exact test was used to compare categorical variables. Pearson's or Spearman's correlation analysis was used to assess the association between serum AMH levels and HOMA-IR, depending on the distribution of the variables. The independent association of treatment type with AMH and HOMA-IR was assessed by using multivariable linear regression analysis, adjusting for age, BMI, and treatment duration. The results of the study were reported as adjusted regression coefficients and 95% confidence intervals. A p-value < 0.05 was regarded as statistically significant.

A total of 74 women diagnosed with polyendocrine metabolic ovarian syndrome (PMOS) were included in the analysis, with 37 participants in each treatment group. The mean age was 27.84 ± 5.21 years in the myo-inositol group and 28.46 ± 5.67 years in the metformin group. There were no significant differences between the groups with respect to age, body mass index (BMI), duration of PMOS, marital status, menstrual history, family history of metabolic disorders, or reproductive history (p > 0.05), suggesting a similar baseline profile (Table 1).

Table 1. Comparison of Demographic, Anthropometric, and Clinical Characteristics between

Study Groups (n = 74)

The mean treatment duration was 3.82 ± 1.21 months in the myo-inositol group and 4.06 ± 1.34 months in the metformin group, with no significant difference in treatment duration or medication adherence (p > 0.05). However, the frequency of reported adverse effects was significantly higher in the metformin group (37.8%) than in the myo-inositol group (13.5%; p = 0.017). Women on metformin were more likely than women not taking the drug to experience gastrointestinal upset, nausea, and diarrhea (Table 2).

Table 2. Comparison of Treatment-Related Characteristics between Myo-Inositol and Metformin

Groups (n = 74)

Biochemical analysis showed that the serum anti-Müllerian hormone (AMH) was significantly lower in the myo-inositol group than in the metformin group ( p = 0.005). Similarly, women who were given myo-inositol had significantly lower fasting blood glucose, fasting serum insulin, and HOMA-IR values of 2.44 ± 0.91 and 3.27 ± 1.12, respectively (p = 0.001) (Table 3).

Table 3. Comparison of Serum AMH and Metabolic Parameters between Myo-Inositol and Metformin Groups

(n = 74)

The correlation analysis showed that in both treatment group 1 and treatment group 2, there was a positive, statistically significant correlation between serum AMH levels and HOMA-IR. The correlation was slightly stronger in the metformin group (r = 0.43; p = 0.008) than in the myo-inositol group (r = 0.38; p = 0.020). A moderate positive correlation was also found among all participants, r = 0.46; p < 0.001 (Table 4).

Table 4. Correlation between Serum Anti-Müllerian Hormone (AMH) and Insulin Resistance (HOMA-IR)

Among Women with PMOS (n = 74)

After adjusting for age, BMI, and treatment duration, multivariable linear regression analysis revealed that myo-inositol treatment was independently associated with lower serum AMH levels ( p = 0.009) and lower HOMA-IR ( p = 0.003) than metformin treatment. BMI was also associated with a positive increase in HOMA-IR (p < 0.001), but not with either outcome (Table 5).

Table 5. Multivariable Linear Regression Analysis of Factors Associated With Serum AMH and

HOMA-IR among Women with PMOS (n = 74)

This analysis showed that women who received myo-inositol and metformin for polyendocrine metabolic ovarian syndrome (PMOS) had different levels of serum anti-Müllerian hormone (AMH) and insulin resistance. In the myo-inositol group, lower mean concentrations of AMH and HOMA-IR were found suggesting a possible correlation between the choice of treatment and reproductive-metabolic characteristics. The results, however, were not congruent with Ravn et al. (2022) who performed a randomized controlled trial using 45 women with a difference between myo-inositol and metformin after 6 months treatment (p = 0.09). The effect of metformin on fasting glucose and body weight were more beneficial.[13] Likewise, Fatima et al. (2023) found no significant difference between myo-inositol and metformin with regard to fasting insulin, fasting glucose, or HOMA-IR in the meta-analysis (n = 1,088).[14] These differences point out the potential for variations in participant characteristics, treatment, and study design and could affect the observed metabolic results. The lower insulin resistance in the myo-inositol group was partly in line with what was reported by Gudović et al. (2024), who studied 60 women with PCOS and normal BMI with insulin resistance. Both myo-inositol and metformin treatment resulted in considerable improvement in insulin-related parameters, with treatment effects being modestly different for each parameter.[15] Similarly, Bahadur et al. (2021) found that insulin and other hormonal or metabolic markers improved when using a combination of metformin and inositol, compared to metformin alone.[16] However, no significant between-group difference was observed in HOMA-IR. The results provide further evidence for the metabolic relevance of inositol-preparations, and do not prove them to be superior to metformin. In terms of ovarian endocrine function, results showed significantly lower levels of AMH in the serum of women taking myo-inositol. However, Nazirudeen et al. (2023) in their randomized controlled trial with 53 women did not find any significant difference between the two groups in terms of hormonal and metabolic parameters, including AMH.[17] However, the combination group had enhanced quality of life and menstrual regularity. Moreover, in a systematic review and meta-analysis of 30 trials and 2,230 patients, Fitz et al. (2024) found that the evidence for the metabolic and reproductive benefits of inositol was still limited and inconclusive.[12] The observations suggest a cautious interpretation of the differences in serum AMH since the lower AMH concentrations may not be indicative of better ovarian function or fertility. Kriplani et al. (2025) did a multi-center randomized trial with 196 women suffering from PCOS and also tested for the potential association of treatment type and insulin resistance. There was improvement in HOMA-IR in 75.0% of those receiving combined metformin and myo-inositol and 60.67% of those receiving only metformin (p = 0.049).[18] A meta-analysis of six randomized controlled trials of 388 women by Kelly et al. (2025), however, showed no statistically significant difference in HOMA-IR between the two treatment groups (combination therapy vs. metformin monotherapy; p = 0.25).[19] The analysis also showed that there was a positive correlation between serum AMH and HOMA-IR, indicating that higher insulin resistance levels could be associated with changes in ovarian follicular activity. The cross-sectional association, however, was only an association and not a causal relationship. Irfan et al. (2026) in a randomized controlled trial study with 192 women, reported better metabolic and endocrine outcomes after metformin and inositol based treatment, dietary restriction and the combination of both, respectively.[20] Combined intervention showed the highest decrease in HOMA-IR, highlighting the combined therapeutic potential of the different interventions for metabolic control. Their treatment, however, involved dietary modification and formulations of a combination of inositols, which do not allow direct comparisons with the present treatment groups. Limitations This present study had some limitations. First, the size of the study population and the single-center design might have restricted the generalizability of the results. Second, there is the potential for selection bias and residual confounding because of the observational design and the lack of a random allocation of the treatment. Thirdly, no baseline or follow-up biochemistry testing was performed which may have allowed the evaluation of changes in the serum AMH and insulin resistance levels as a result of treatment. In addition, the different durations of treatment, compliance with treatment, dietary intake and exercise patterns may have affected the results seen. More multicenter prospective studies are needed to confirm these results.

The findings suggested that women receiving myo-inositol had lower serum AMH concentrations and HOMA-IR values than those receiving metformin. Myo-inositol was also linked to a lower incidence of gastrointestinal side effects. A positive correlation was observed between serum AMH and insulin resistance, highlighting the relationship between ovarian endocrine and metabolic abnormalities in PMOS.