PCOS: What a New Randomized Trial Shows About Combining Metformin, Inositol, and Diet

This article is for educational purposes intended for healthcare practitioners and informed readers. It does not constitute medical advice. Any change to metformin dosing or the addition of a supplement or dietary protocol should be made under the guidance of a treating practitioner.

Why This Trial Is Worth Reading

Naturopathic PCOS protocols routinely combine multiple interventions at once: an insulin sensitiser, a low-glycaemic diet, sometimes a botanical, sometimes all three together. The evidence base for each individual component is reasonably solid, but the evidence for stacking them, whether combining interventions actually produces additive benefit rather than the same benefit several ways over, is thinner. Trials that randomise patients to single interventions and a combined arm, in the same study, under the same conditions, are uncommon in this literature.

A trial published in July 2026 in Clinical Endocrinology did exactly that. It randomised 192 women with PCOS to metformin, inositol, calorie-restricted diet, or all three combined, and measured metabolic and reproductive outcomes across all four arms after 12 weeks. This article reviews what the trial found, how confident that finding should make a practitioner, and where it fits alongside the existing evidence covered in our PCOS naturopathic protocol guide.

The Trial: Design and Population

The study, "Multimodal Therapy With Metformin, Inositol and Dietary Restriction Improves Insulin Resistance and Endocrine Outcomes in Women With Polyendocrine Metabolic Ovarian Syndrome: A Randomized Controlled Trial," was conducted by Irfan, Saleem, and Irfan and published in Clinical Endocrinology. The authors use the label "polyendocrine metabolic ovarian syndrome (PMOS)" rather than the more familiar PCOS, but the diagnostic criteria applied are the standard 2003 Rotterdam criteria (at least two of: oligo/anovulation, clinical or biochemical hyperandrogenism, or polycystic ovarian morphology), so this is the same condition covered in our existing PCOS protocol, not a distinct diagnosis.

Women aged 18 to 35 were randomised to one of four parallel arms of 48 each:

• Metformin: 1500 to 2000 mg/day
• Inositol: myo-inositol 2 g plus D-chiro-inositol 50 mg, twice daily (a 40:1 myo-inositol to D-chiro-inositol ratio, the same physiological ratio discussed in our existing protocol)
• Diet: 1200 to 1500 kcal/day, low-glycaemic carbohydrates, lean protein, high fibre
• Combination: all three interventions together

Of the 192 women randomised, 186 completed the 12-week trial; six were lost to follow-up. The design was open-label, meaning participants and investigators knew which arm they were in, which matters for interpreting the self-reported adherence and menstrual cycle outcomes discussed below.

What the Trial Found

Across every outcome measured, the combination arm produced the largest improvement, and the differences between arms were statistically significant.

Insulin resistance (HOMA-IR): metformin reduced HOMA-IR by 0.92, inositol by 1.01, diet by 1.74, and the combination arm by 2.64 (95% CI −2.82 to −2.46, p<0.001), roughly a 44 percent larger reduction than diet alone and close to three times the reduction seen with metformin alone.

Weight and BMI: the combination arm lost 7.1 kg and 2.8 BMI points on average, more than either metformin (3.2 kg) or inositol (3.8 kg) alone, and somewhat more than diet alone (5.4 kg).

Testosterone and LH:FSH ratio: testosterone fell by 24.1 ng/dL in the combination arm versus 12.4 ng/dL with metformin, 15.1 ng/dL with inositol, and 18.6 ng/dL with diet alone. The LH:FSH ratio, a marker of the gonadotrophin dysregulation described in our PCOS pathophysiology overview, fell furthest in the combination arm as well (−0.68 versus −0.22 to −0.44 in the single-intervention arms).

Menstrual cycle regularisation: 85.4 percent of women in the combination arm achieved a regular 21 to 35 day cycle in the final four weeks of the trial, compared with 72.9 percent on diet alone, 64.6 percent on inositol alone, and 39.6 percent on metformin alone.

Hirsutism: the Ferriman-Gallwey hirsutism score improved in all arms, but the between-group differences were not statistically significant (p=0.08), the one outcome in the trial where combination therapy did not clearly separate from the single-intervention arms.

Tolerability: metformin had the highest adverse event rate (18.7 percent, mostly gastrointestinal), followed by the combination arm (16.7 percent), inositol (10.4 percent), and diet alone (8.3 percent). The between-group difference was not statistically significant (p=0.29), and there were no serious adverse events or treatment discontinuations.

Reading This Trial Honestly

This is a genuinely useful trial for naturopathic practice because it directly tests the working assumption behind multi-intervention PCOS protocols, that combining an insulin sensitiser, inositol, and dietary change outperforms any one of them alone, rather than assuming it from separate single-arm literature. On every outcome except hirsutism, the combination arm won, and the size of the advantage over diet alone (the strongest single intervention) was clinically meaningful, not marginal.

Several limitations temper how far that result should be generalised. It was conducted at a single centre, which limits confidence that the effect sizes will reproduce elsewhere. It was open-label, which is a real source of bias for a self-reported outcome like menstrual cycle regularity and for a behaviourally demanding intervention like calorie restriction, participants who knew they were in the most intensive arm may have tried harder, reported more favourably, or both. Follow-up was 12 weeks, too short to confirm that the menstrual improvements represent sustained ovulatory normalisation rather than an early, transient response; the authors themselves describe the menstrual findings as "early improvements rather than sustained normalisation." The trial did not report fertility outcomes (confirmed ovulation, pregnancy) or any cardiometabolic endpoint beyond the 12-week window, and a four-arm, single-centre trial of this size, while adequately powered for the primary metabolic endpoints, is not the last word on a condition this heterogeneous.

It is also worth noting what the trial does not test: it compares metformin, inositol, and diet against each other and in combination, but it does not include a placebo or no-treatment arm, so the absolute effect of any single intervention against doing nothing cannot be read directly from this study. For that, the placebo-controlled inositol literature summarised in our existing PCOS protocol guide remains the more relevant source.

Practical Implications for Naturopathic Practice

The trial supports a specific, moderate conclusion: for women with PCOS able to sustain a calorie-restricted diet, adding inositol at the 40:1 physiological ratio and, where clinically appropriate, metformin produces measurably better metabolic and cycle outcomes at 12 weeks than any single intervention alone, with no meaningful increase in adverse events from combining them. That is consistent with, and now has direct trial support for, the layered approach already described in our protocol guide. It does not establish that combination therapy improves fertility, sustains menstrual regularity beyond 12 weeks, or outperforms single interventions in hirsutism, and practitioners should be clear with patients that this is one open-label trial from a single centre, not a settled question. Diet alone, notably, outperformed either metformin or inositol alone on most outcomes in this trial, a reminder that dietary intervention deserves to be presented as a primary therapeutic lever, not an adjunct to supplementation, consistent with the dietary strategy already emphasised in our PCOS protocol.

Clinical Summary

A 2026 randomized controlled trial of 192 women with PCOS (Rotterdam criteria) found that combining metformin, inositol (40:1 myo-inositol to D-chiro-inositol), and a calorie-restricted diet over 12 weeks produced significantly greater improvements in insulin resistance, weight, testosterone, LH:FSH ratio, and menstrual cycle regularity than any of the three interventions used alone, with no significant increase in adverse events. Hirsutism improved in all arms without a significant between-group difference. The trial was single-centre, open-label, and limited to 12 weeks, with no fertility or long-term cardiometabolic data, so the findings support combination therapy as a reasonable, evidence-consistent approach rather than a definitively proven one. Diet alone was the strongest single intervention across most outcomes, reinforcing dietary change as a foundational, not supplementary, part of PCOS management.

References

• Irfan NA, Saleem S, Irfan A. "Multimodal Therapy With Metformin, Inositol and Dietary Restriction Improves Insulin Resistance and Endocrine Outcomes in Women With Polyendocrine Metabolic Ovarian Syndrome: A Randomized Controlled Trial." Clinical Endocrinology. 2026. DOI: 10.1111/cen.70188. PMID: 42517336. Full text: PMC13532120.

This article is intended for educational purposes and professional practice reference. It does not constitute individual medical advice. Any change to metformin dosing, or the addition of inositol or a calorie-restricted diet, should be made in consultation with a qualified practitioner.