Polycystic ovary syndrome (PCOS) is an endocrine and reproductive condition that is one of the most prevalent in young women and contributes significantly to menstrual dysfunction, anovulatory infertility, and metabolic morbidity.[1] It is estimated that between 10–13% of women of reproductive age worldwide have PCOS, and as many as 70% of these may not be diagnosed.[2] It is known as the most prevalent cause of anovulation globally and is an important factor in female infertility.[3] PCOS is a disorder that has implications beyond fertility; women with PCOS are also at increased risk for insulin resistance, obesity, impaired glucose tolerance, and type 2 diabetes mellitus.[4]

Insulin signaling deficiency and hyperinsulinism play a close role in reproductive dysfunction in PCOS.[5] Women with PCOS frequently suffer from insulin resistance, and this may happen despite their body mass index.[6] Compensatory hyperinsulinemia may lead to ovarian hyperandrogenization, subnormal follicular maturation, and disruption of normal hypothalamic–pituitary–ovarian activity, which contribute to irregular menstruation and anovulation.[7] Thus, any treatment that can increase insulin sensitivity has the potential not only to reduce metabolic parameters but to restore ovulatory function and increase the chances of conception.

The insulin sensitizing agent metformin has been used for a long time in PCOS to treat metabolic dysfunction and, in selected women, reproductive abnormalities.[8] Metformin decreases hepatic glucose production and enhances peripheral insulin sensitivity, thereby decreasing hyperinsulinemia, and may then restore menstrual cyclicity and ovulatory function.[7] The 2023 International Evidence-Based Guideline acknowledges that the use of metformin in women with PCOS and anovulatory infertility may be considered to enhance clinical pregnancy and live-birth outcomes, while other more effective ovulation-induction agents are available.[9] Gastrointestinal side effects, however, might restrict its clinical application, affecting treatment acceptance and adherence.[10]

A naturally occurring compound called myo-inositol (MI), which is a mediator in insulin signaling, has now been identified as an alternative strategy for women suffering from PCOS.[11] MI is a player in intracellular insulin action and glucose usage, as well as in granulosa cell function and follicular developmental and maturation pathways regulated by follicle-stimulating hormone (FSH).[12] Its biological activity could offer a logical explanation for its application in the treatment of PCOS, especially in women with insulin resistance and irregular periods.[13] Importantly, evidence synthesized for the international guideline in 2023 indicates that inositol has beneficial effects on some metabolic parameters and that inositol is generally tolerated less poorly than metformin in terms of gastrointestinal side effects, although data on its effects on ovulation and pregnancy are limited and uncertain.[14]

Metformin and MI are both used to treat the metabolic issues associated with PCOS, but they vary significantly in terms of their pharmacological characteristics, mechanisms of action, tolerability, and evidence.[15] Existing studies have only compared these interventions to placebo or have looked at their effect when used in addition to other forms of fertility treatment, making it hard to see if MI can offer a reproductive benefit equal to that of metformin as an alternative insulin-sensitizing therapy. The current evidence synthesis has reported no clear difference between inositol and metformin in terms of reproductive outcomes and has reported uncertainty in the available evidence. In addition, because of the lack of high-quality evidence, it is too early to recommend with confidence the specific type, dose, and combination of inositol.

In this context, a head-to-head comparison of MI and metformin in regard to clinically relevant outcomes is of special interest. While evaluating metabolic improvement alone is insufficient, a combination of evaluation of ovulation, insulin sensitivity, and actual pregnancy achievement may be more patient-centered to determine the effectiveness of treatment and whether improvement in insulin sensitivity translates into meaningful reproductive benefit. The findings have implications for the use of the metabolically active, well-tolerated MI in women with PCOS to improve reproductive and metabolic outcomes, especially those wishing to become pregnant, as this is an established insulin-sensitizing therapy. This study was thus designed to compare the efficacy of oral myo-inositol with the efficacy of metformin on ovulation, insulin resistance, and pregnancy rate in women with PCOS.

A prospective, randomized, open-label comparative study was conducted in the Department of Obstetrics and Gynecology of Creek General Hospital, Karachi, for 6 months, from 1st November 2025 to 30th April 2026. Participants were monitored over the course of the treatment period to evaluate changes in ovulation, insulin resistance, and pregnancy outcomes. The sample size was determined as per OpenEpi version 3.01 (for comparing two independent proportions). The calculation was based on the ovulation rates previously reported, showing an ovulation rate of 73.9% for the treatment group that also contained ovulation-inducing myo-inositol vs. an ovulation rate of 44.0% in the metformin group.[16] The desired sample size per group was 42 with an alpha level of 5%, a 95% confidence level, an allocation ratio of 1:1, and an 80% statistical power. An anticipated 10% loss to follow-up was allowed, resulting in 47 participants per group. A consecutive sampling technique was used for the recruitment of participants. Women who met the inclusion criteria at the study setting during the study were approached for participation. Consecutive eligible women who gave written informed consent were enrolled until the desired sample size was reached. After enrollment, participants were randomized using a computer-generated randomization sequence in a 1:1 ratio to either the myo-inositol group or the metformin group. The Rotterdam criteria for the diagnosis of PCOS included women aged 18–40 years. The participants presented with documented ovulatory dysfunction either as oligomenorrhea or amenorrhea and were looking to conceive. Women who were interested in the study, followed the treatment as prescribed, were subjected to biochemical and ultrasonographic evaluation, and were willing to attend for follow-up were included. Participants who met all other study criteria, with insulin resistance or metabolic features of PCOS, were eligible to participate. Pregnant and lactating women were not included. Patients with type 1 or 2 diabetes mellitus, thyroid dysfunction, hyperprolactinemia, congenital adrenal hyperplasia, Cushing syndrome, androgen-secreting tumors, or other forms of endocrine abnormalities leading to menstrual or ovulatory abnormalities were excluded. Other exclusion criteria were women with marked liver or renal disease, contraindications or known sensitivity to metformin or to myo-inositol, and women taking any hormonal contraceptive, insulin-sensitizing drug or other medication for the treatment of PCOS. The patients previously treated with ovarian surgery or assisted reproductive treatment were excluded from the study period. Final analysis also excluded women who failed to meet requirements for treatment and follow-up. Demographic and clinical data were collected in a structured data collection proforma after informed consent was obtained. Data on age, length of infertility period, menstrual pattern, parity, previous pregnancy, family history of diabetes mellitus, and relevant medical history were recorded. Baseline anthropometric measures were taken, including weight, height, BMI, and waist circumference. A detailed menstrual history was obtained to establish the regularity of the menstrual cycles and initial ovulatory function. Baseline biochemical measurements (fasting blood glucose and fasting serum insulin) were obtained in all participants. Insulin resistance was evaluated by the homeostatic model assessment of insulin resistance (HOMA-IR), which was derived from fasting insulin and glucose levels. The study protocol and other biochemical and hormonal investigations, which were deemed clinically appropriate, were undertaken, including serum testosterone, luteinizing hormone, and follicle-stimulating hormone. Two groups were randomly formed. Oral myo-inositol 2 g twice daily was given to the myo-inositol group, and oral metformin 500 mg thrice daily was given to the metformin group, for the predetermined study duration. Previous comparative trials of PCOS used myo-inositol-based regimens of around 4 g/day and metformin regimens of 1.5–2 g/day. Adherence to treatment and side effects were evaluated at follow-up visits. Treatment was conducted with serial follicular monitoring by pelvic ultrasonography to determine the status of the ovary with respect to ovulation. Appropriate follicular development followed by evidence of follicular rupture was detected as being a sign of ovulation. When available, serum progesterone was also measured during the mid-luteal phase to aid in confirming ovulation. Documented ovulatory cycles during the treatment period were recorded. Fasting blood glucose, fasting serum insulin, and HOMA-IR were reassessed at the end of the treatment period to evaluate insulin resistance. Participants who were trying to get pregnant during the study period were evaluated for pregnancy. Pregnancy was confirmed by ultrasonography after performing a pregnancy test in cases with delayed period or clinical suspicion of pregnancy. Clinical pregnancy rate was defined as the number of participants with confirmed intrauterine pregnancy divided by the total number of participants followed up. Data were coded, entered, cleaned, and analyzed in SPSS version 26.0. Normality was checked for continuous variables by using the Shapiro–Wilk test. Continuous variables were presented as mean ± SD, and non-normally distributed variables as a median and interquartile range. Categorical variables were described in terms of frequencies and percentages. The baseline characteristics were compared between myo-inositol and metformin groups to determine comparability. Normally distributed continuous variables were compared using the independent-samples t-test, and non-normally distributed continuous variables were compared using the Mann–Whitney U test. Categorical variables such as the rate of ovulation and pregnancy were compared between the two treatment groups using the chi-square test and Fisher's exact test. The changes in fasting glucose, fasting insulin, and HOMA-IR from baseline to post-treatment within and between the groups were compared using the appropriate paired and independent statistical tests depending on the data distribution. Effect estimates with 95% confidence intervals were reported if available. A p-value ≤0.05 was considered statistically significant.

A total of 94 women with PCOS were recruited into the study, and 47 women were assigned to each treatment group. There was no difference in the mean age between the two groups of myo-inositol and metformin (p=0.741). There were no significant differences between the two groups in terms of duration of infertility, parity, BMI or waist circumference, menstrual pattern, or family history of DM, indicating similar baseline demographic and clinical characteristics (Table 1).

Table 1. Baseline demographic and clinical characteristics of participants (n=94)

Also, baseline biochemical and hormonal parameters were similar in both groups. Mean fasting blood glucose was 92.4 ± 8.7 mg/dL in the myo-inositol group and 93.1 ± 9.1 mg/dL in the metformin group (p=0.709), while median HOMA-IR was 3.42 (2.72–4.36) and 3.51 (2.79–4.42), respectively (p=0.764). There were no significant differences in the baseline levels of fasting insulin, testosterone, LH, FSH, or LH/FSH ratio (Table 2).

Table 2. Baseline biochemical and hormonal characteristics of participants

After the treatment, both groups showed significant improvement in the metabolic parameters. HOMA-IR decreased from 3.42 (2.72–4.36) to 2.58 (1.91–3.42) in the myo-inositol group and from 3.51 (2.79–4.42) to 2.76 (2.04–3.61) in the metformin group (both p<0.001). There were also significant decreases in BMI, fasting glucose, fasting insulin, and waist circumference in both groups, but these differences between the two groups were not significant (Table 3).

Table 3. Comparison of anthropometric and metabolic parameters before and after treatment

Treatment resulted in notable improvements in hormones. Testosterone decreased from 71.4 ± 18.6 to 61.2 ± 15.7 ng/dL in the myo-inositol group (p<0.001) and from 73.1 ± 19.2 to 65.8 ± 16.9 ng/dL in the metformin group (p=0.002). There was also a significant decrease in LH and LH/FSH ratio in both groups, while FSH did not change significantly. There were no significant differences in the hormones between the groups (Table 4)

Table 4. Comparison of hormonal parameters before and after treatment

Myo-inositol induced ovulation in 35 women (74.5%), while metformin induced ovulation in 29 women (61.7%). However, the percentage of women who ovulated had no significant difference between groups (p=0.186) (Table 5).

Table 5. Ovulatory response after treatment

The rate of clinical pregnancy among women in the myo-inositol group was 17 (36.2%), and among women in the metformin group, 11 (23.4%) (p=0.176). Spontaneous conception occurred in 34.0% and 21.3% of women, respectively, while pregnancy loss occurred in one participant in each group (Table 6).

Table 6. Comparison of pregnancy outcomes

In women taking myo-inositol, improvement in HOMA-IR was seen in 36 (76.6%), and in 32 (68.1%) taking metformin (p=0.354). The rate of improvement was 66.0% in treated children and 55.3% in the non-treated group, but there was no significant difference between them (Table 7).

Table 7. Treatment response according to insulin resistance

Treatment adherence was very good in both groups, with metformin-treated women showing significantly more GIT side effects than the myo-inositol-treated women (p=0.018). The number of women who discontinued treatment due to side effects was 8.5% in the metformin group and 2.1% in the myo-inositol group (p=0.164) (Table 8).

Table 8. Treatment adherence and adverse effects

Polycystic ovary syndrome (PCOS) is a complex syndrome involving insulin resistance, hyperandrogenism, and ovulatory dysfunction, and so both metabolic and reproductive outcomes are of importance when assessing treatment. In the current study, 94 women with PCOS and infertility were equally divided into oral myo-inositol and metformin groups. Demographic, anthropometric, biochemical, and hormonal characteristics of the two groups were comparable at baseline, reducing the likelihood that baseline differences affected treatment outcomes. Both interventions led to an improvement in metabolic parameters, androgenic status, and ovulatory function, whereas myo-inositol resulted in a numerically higher ovulation and pregnancy rate and in significantly fewer gastrointestinal adverse effects. The present study demonstrated significant reductions in BMI, fasting glucose, fasting insulin, HOMA-IR and waist circumference following both myo-inositol and metformin treatment. Important: No between-group differences were significant. In line with the 2023 meta-analysis of Fatima et al. comparing myo-inositol and metformin, the current study showed no significant differences between them in BMI, fasting insulin, fasting glucose, or HOMA-IR. Their results are consistent with the notion that both agents could have similar metabolic effects in women with PCOS.[17] Likewise, in a systematic review including randomized and non-randomized studies, Bodepudi et al. (2023) found that the overall effect of myo-inositol and metformin was similar, with a more favourable tolerability profile for myo-inositol.[18] Our results are also similar to those of a randomized controlled trial that included 45 women diagnosed with PCOS who were assigned to receive either myo-inositol or metformin for six months (Ravn et al., 2022). The study did not show significant differences between the two interventions for HOMA-IR, but both interventions effected changes in metabolic parameters as well as cycle length improvement, which did not differ significantly between the two interventions. These differences in metabolic response may, in part, be due to their having a significantly greater mean BMI compared to the current cohort. Finally, Ravn et al. reported higher weight loss associated with treatment with metformin while our study did not demonstrate a significant difference between the weight loss seen in the two groups.[19]. The results of this study are confirmed in a further study conducted by Gudović et al. in 2024, in which they also found a negative correlation between metabolic rate and age. In 60 women with PCOS who had normal weight, myo-inositol and metformin were both effective in lowering insulin response in the oral glucose tolerance test. The similarity of metabolic response, especially for fasting insulin and HOMA-IR, is also relevant in relation to our study, which also showed significant decreases in fasting insulin and HOMA-IR with both interventions without a significant difference between them. Variation between studies, however, may be due to differences in BMI distribution, or in the insulin resistance assessment methods employed.[20] Results of the present study on hormones are also broadly in line with the previous evidence. Serum testosterone and LH were significantly lowered in both treatment groups, and the LH/FSH ratio was significantly lowered, while FSH was not significantly lowered. This indicates that enhancement of insulin sensitivity could have played a role in the dampening of hyperandrogenic and abnormal gonadotropin profile seen with PCOS. In a 2021 randomized controlled trial that evaluated myo-inositol vs. metformin over 6 months of treatment, Agarwal et al. reported that clinical, ovarian, hormonal, and metabolic parameters improved in both groups, which is in line with our findings regarding similar endocrine effects.[21] The decrease in LH/FSH ratio in our cohort is of special interest since recent data indicated that inositol-based therapy could affect regulation of the gonadotropins. Kelly et al. performed a 2025 systematic review and meta-analysis of six randomized trials that included 388 women, concluding that the combination of metformin and inositol significantly reduced LH/FSH ratio compared to metformin alone. While this study looked at combination therapy and not myo-inositol, the results do support the plausibility of an effect of inositol on gonadotropin balance.[22] The reduction of the LH/FSH ratio in our study, however, was not significantly different between the two groups, suggesting that myo-inositol and metformin may have similar effects on the endocrine parameter. In terms of reproduction, 74.5% of women who received myo-inositol ovulated, while 61.7% of those taking metformin did so. This difference was in favor of myo-inositol, but was not statistically significant. This corresponds to the 2023 international systematic review and meta-analysis conducted by Teede et al., as part of the guideline for the diagnosis and management of PCOS for which these data were gathered. However, there was no difference found in rates of ovulation between studies directly comparing myo-inositol and metformin. The authors, therefore, concluded that reproductive outcomes were likely to be similar for both interventions, albeit with limited levels of evidence.[23] Chowdhury et al. (2022) carried out another study of insulin-resistant women experiencing infertility due to PCOS in 2022 with interesting comparisons. In their study, metformin plus myo-inositol increased the ovulation rate by 73.9% compared to metformin alone (44.0%). Combination therapy also numerically increased the risk of pregnancy, though not statistically. In the current study, the rate of ovulation seen with myo-inositol alone (74.5%) was very similar to that seen with combination therapy in that study (73.9%). However, direct comparison should be done with caution because their intervention contained both metformin and myo-inositol, so it would be hard to determine if myo-inositol was equivalent to metformin alone.[16] The results of the present study were consistent with this. A total of 36.2% of women in the myo-inositol group achieved clinical pregnancy versus 23.4% in the metformin group. This was a clinically significant numerical difference, but not significant in the statistical sense. These results align with the results from the 2023 evidence synthesis that informed the international PCOS guideline, which did not show a significant difference in clinical pregnancy rates between myo-inositol and metformin in the available comparative trials. There was no significant difference between the 420 women in four trials, and no significant difference in miscarriage rates in one study. The same pregnancy loss rate (2.1% in both groups) is also consistent with the overall similar reproductive safety of the product previously reported in the literature.[24] This is the context that is provided by the evidence from assisted reproductive settings. In a systematic review and meta-analysis of randomized controlled trials (RCTs) of metformin, inositol and their combination in women with polycystic ovary syndrome (PCOS) undergoing assisted reproduction, Unanyan et al. (2022) assessed reproductive outcomes. They noted that the evidence on reproductive benefits is still murky and heavily treatment context dependent, with even greater dependency if using ART. The relatively high ovulation and pregnancy rates seen with myo-inositol in our study warrant its use particularly in women with a significant component of ovulatory dysfunction to their infertility, but further studies would be necessary to determine superiority over metformin.[25] The present results align with the existing literature, indicating that the overall effects of myo-inositol and metformin on metabolic and reproductive parameters in women with PCOS are broadly similar, with myo-inositol potentially having an advantage in terms of tolerability. Therefore, the effect on ovulation, clinical pregnancy, and HOMA-IR normalization requires careful interpretation in view of the apparent number effect in our study. The sample size of 94 participants might have restricted the power of the statistical analyses to detect moderate between-group differences. In addition, insulin resistance and the ovulation status are not the only factors that affect pregnancy; partner-related factors in fertility, tubal status, duration of infertility, frequency of intercourse, and other metabolic and reproductive parameters play a role in pregnancy. More follow-up and multicenter trials would help determine whether the numerical benefits of myo-inositol are statistically and clinically significant in terms of live birth. Finally, both oral myo-inositol and metformin medications were found to be effective for women with PCOS and infertility in alleviating metabolic and hormonal abnormalities. Both treatments showed no statistically significant improvement regarding the outcome of ovulation and/or pregnancy, HOMA-IR, or hormonal parameters. However, significantly fewer gastrointestinal adverse effects and somewhat better treatment adherence were seen with myo-inositol, which also yielded numerically higher ovulation and clinical pregnancy rates. Myo-inositol is an alternative to metformin that appears to be well tolerated and does not cause gastrointestinal intolerance as may occur with metformin; further, larger, adequately powered studies are warranted to confirm that the observed reproductive benefits are real. LIMITATIONS This study was conducted on only 94 participants, which restricted the statistical power to see small differences between the treatment groups, especially with regard to the ovulation and pregnancy outcomes. The study was done at one center, and the results might not be applicable to other centers or populations. There may also have been inadequate treatment and follow-up to evaluate long-term reproductive outcomes, such as live birth rates. In addition, female and male infertility factors might be responsible for pregnancy outcomes but were not measured. A hyperinsulinemic-euglycemic clamp was not performed, but instead a surrogate marker of insulin resistance, HOMA-IR, was used. Lastly, the study did not study the different PCOS phenotypes individually, and they might respond differently to myo-inositol and metformin.

Metabolic and hormonal parameters were significantly improved in women with PCOS and infertility after taking both oral myo-inositol and metformin. The results were not significant, but numerically, Myo-inositol increased ovulation and clinical pregnancy rates. Both treatments had similar effects on insulin resistance and hormonal disturbances. In terms of tolerability, gastrointestinal adverse effects and treatment adherence, myo-inositol was significantly better tolerated. The results indicated that myo-inositol can be a safe alternative to metformin in women with PCOS, especially those who have gastrointestinal intolerance to metformin. Comparative reproductive benefits would be better confirmed with larger, multi-center studies with longer follow-up.