Inositol and Metformin: What the Trials Compared

Metformin is a prescription medicine the NHS lists for polyendocrine metabolic ovarian syndrome (PMOS, previously called PCOS); inositol is a food supplement with no authorised health claim in Great Britain. In the ten head-to-head trials the 2023 international guideline review pooled, fasting insulin, glucose, testosterone and BMI did not differ, hirsutism and waist-to-hip ratio improved more on metformin, and stomach side effects were far less common on myo-inositol.^{1, 2, 3, 4} Nobody changes a prescription on that basis without their prescriber. One spoke of the Wellgard inositol guide.

Evidence at a glance

• Trials: 10 head-to-head trials pooled in the 2024 review written for the international PCOS guideline; 6 trials of inositol added to metformin, 388 women, pooled in 2025.^{4, 9}
• Largest single trial: 192 women aged 18 to 35, four groups, 12 weeks (2026).¹⁰
• Strongest measured outcome: stomach side effects, odds 0.09 times those on metformin in six trials of 382 women, moderate certainty.⁴
• Not tested: no comparison longer than six months; no measurement of blood glucose when inositol is added to prescribed metformin outside a trial.⁴

Sources checked on 17 October 2026.

What is metformin, and why is it compared with inositol?

The NHS describes metformin as a medicine for type 2 diabetes, gestational diabetes and the symptoms of PMOS, "only available with a prescription"; its PMOS page lists it as "a type 2 diabetes medicine to help tiredness and weight problems".^{2, 3} Inositol is a sugar-like molecule the body makes and food supplies, sold as a food supplement with no authorised claim on the Great Britain register.

Both have been studied for the insulin measures that run high in PCOS, hence the head-to-head trials.⁴ The 2023 guideline's position: inositol "could be considered", with "limited clinical benefits including in ovulation, hirsutism or weight", and "metformin should be considered over inositol for hirsutism and central adiposity, noting that metformin has more gastrointestinal side effects than inositol" (recommendations 4.7.1 and 4.7.2).¹

What did the head-to-head trials find?

The guideline review screened every trial for research integrity, assessing 43 and leaving out 13.⁴ Ten of them compared myo-inositol directly with metformin over three to six months, in India, Italy, Iran, Denmark and Bosnia and Herzegovina.⁴

"No difference" means the pooled interval crossed zero: for regular periods the odds ratio was 1.85, interval 0.68 to 5.01.⁴ In the two trials at low risk of bias, 100 women each, fasting insulin was lower on metformin, regular menstruation more likely on myo-inositol (odds ratio 2.84, interval 1.26 to 6.37), and stomach side effects did not differ.⁴ The review calls the evidence "limited and inconclusive".⁴

Which trials were retracted, and which are counted?

A 2018 trial of 60 women that compared myo-inositol with metformin on mental-health measures was retracted in 2024; a 2017 trial of 60 women from the same group, comparing the two on clinical and metabolic measures, was placed under an expression of concern in 2023.^{5, 6} Neither is counted here or in the guideline review's pooled set, so every figure in the table stands on trials with no notice against them as of 17 October 2026.⁴ Two earlier poolings whose included sets could not be checked against those notices (six trials, 355 women, 2019; eight studies, 1,088 women, 2023) reached the same broad conclusion and are not used for figures.^{7, 8}

Which has fewer side effects?

In six trials of 382 women, stomach side effects were far less common on myo-inositol than on metformin: odds ratio 0.09, interval 0.02 to 0.37, moderate certainty.⁴ Two qualifications: 23 of the review's 29 trials did not report side effects at all, and in the two low-risk trials the difference disappeared; it calls metformin's stomach effects "generally mild and self-limiting".⁴ The NHS lists metformin's common side effects as feeling or being sick, diarrhoea, tummy pain, loss of appetite, a change in taste and vitamin B12 deficiency, says most "should ease after a couple of weeks", and asks anyone bothered by them to speak to a pharmacist or doctor.³ What the inositol trials reported is on the inositol side effects page.

What happens when inositol is added to metformin?

A 2025 pooling of six trials, 388 women, three to six months, compared metformin plus inositol with metformin alone: a regular cycle was 1.56 times as likely on the combination (interval 1.01 to 2.41), the hirsutism score 0.97 points lower and the LH/FSH ratio lower; acne, BMI, fasting glucose and HOMA-IR did not differ significantly.⁹ The guideline review had one such trial, 50 women over four months, with no difference in BMI, weight or fasting insulin, and calls any benefit of the combination unclear.⁴

The largest single trial, 2026, randomised 192 women aged 18 to 35 with PMOS to metformin, inositol (myo- with D-chiro-inositol), a calorie-restricted diet, or all three, for 12 weeks.¹⁰ All four groups improved; the combination had the largest falls in HOMA-IR and BMI, and cycle regularisation was reported by 85.4% on the combination, 72.9% on diet, 64.6% on inositol and 39.6% on metformin, which the authors call early improvement rather than sustained normalisation.¹⁰ Diet in PCOS is covered in food for PCOS.

Can you take inositol with metformin?

No formal interaction study was found. Insulin resistance fell in every group of the 2026 trial, so an additive effect on blood sugar is plausible; it has only been seen under clinical monitoring.¹⁰ The NHS metformin page says "there are some medicines, herbal remedies and supplements that may not mix well with metformin" and asks people to tell their doctor or pharmacist about any supplements they take.³ So anyone prescribed metformin, for PMOS or diabetes, raises inositol with their GP, diabetes team or pharmacist first, and nobody stops or reduces metformin for a supplement: the NHS says "it's important to keep taking your medicine" and that blood sugar may rise if you stop.³ The guideline encourages women taking inositol to tell their healthcare professional (4.7.3).¹

What does the guideline recommend?

The 2023 guideline recommends lifestyle intervention for all women with PCOS to improve metabolic health including central adiposity (3.1.1), and says women should be informed that metformin and active lifestyle intervention have similar efficacy (4.3.4).¹ On inositol it adds that specific types, doses or combinations "cannot currently be recommended in adults and adolescents with PCOS, due to a lack of quality evidence" (4.7.4) and that in PCOS with infertility inositol "should be considered experimental therapy" (5.8.1).¹ For ovulation induction it names letrozole as first-line (5.3.1); the NHS PMOS page names clomifene as "a medicine used to improve fertility".^{1, 2} NICE's draft guideline for the UK (consultation closed 11 August 2026; publication expected 27 January 2027) adapts the 2023 guideline, keeps metformin ahead of inositol for hirsutism and central adiposity, and would advise that "there is uncertainty about the clinical benefit of inositol for metabolic health problems"; until it is published it is a draft, not guidance.^{1, 11}

What we don't know

• No trial has compared myo-inositol with metformin for longer than six months.⁴
• Only one head-to-head trial measured depression or quality of life, in 26 women.⁴
• Whether adding inositol to metformin changes anything beyond cycle counts and a hirsutism score is unsettled.^{4, 9}
• No head-to-head trial has been run in adolescents, and the guideline recommends no type, dose or combination at any age.¹
• No study has measured blood glucose when inositol is added to prescribed metformin outside a trial.

Frequently asked questions

Is inositol a natural version of metformin?

No. One is a prescription medicine, the other a food supplement with no authorised health claim. The trials found no difference on fasting insulin, glucose, testosterone and BMI over three to six months, with metformin better for hirsutism and waist-to-hip ratio: a trial comparison, not a basis for changing a prescription.⁴

Can I switch from metformin to inositol?

That is a question for the prescribing doctor, not a blog. The NHS says to keep taking the medicine and to talk to your doctor about any problems with it.³ The guideline review's authors wrote that clinicians may consider myo-inositol in women who cannot tolerate metformin, a decision they place with the clinician and the woman.⁴

Does inositol work as well as metformin for PCOS?

On the blood measures the trials shared, myo-inositol and metformin did not differ; on hirsutism and waist-to-hip ratio metformin did better; on stomach side effects myo-inositol did.⁴ Certainty ran from very low (BMI) to moderate; the guideline prefers metformin for hirsutism and central adiposity.^{1, 4}

What about berberine?

No trial has tested berberine with inositol, and berberine has one small trial against metformin; see the berberine and inositol page and berberine vs metformin. Anyone on metformin asks their prescriber before adding either.

Sources

1. Teede HJ, et al. 2023 international evidence-based guideline for polycystic ovary syndrome. International guideline, 254 recommendations. Human Reproduction, 2023. PubMed 37580037.
2. NHS. Polyendocrine metabolic ovarian syndrome (PMOS). Page last reviewed 30 June 2026. nhs.uk/conditions/polyendocrine-metabolic-ovarian-syndrome-pmos.
3. NHS. Metformin. Page last reviewed 5 August 2026. nhs.uk/medicines/metformin.
4. Fitz V, et al. Inositol for polycystic ovary syndrome, the review informing the 2023 guideline. Systematic review, 30 trials included (19 pooled), 2,230 women, 10 of them myo-inositol against metformin. Journal of Clinical Endocrinology and Metabolism, 2024. PubMed 38163998.
5. Statement of retraction of a 2018 randomised trial, 60 women, myo-inositol against metformin on mental-health measures. Journal of Psychosomatic Obstetrics and Gynaecology, 2024. PubMed 38758364.
6. Expression of concern on a 2017 randomised trial, 60 women, myo-inositol against metformin on clinical and metabolic measures. Clinical Endocrinology, 2023. PubMed 36848124.
7. Facchinetti F, et al. Pooled analysis of 6 trials, 355 women; included set not checkable against the retractions. Gynecological Endocrinology, 2019. PubMed 30614282.
8. Fatima K, et al. Pooled analysis of 8 studies, 1,088 women; included set not checkable against the retractions. Irish Journal of Medical Science, 2023. PubMed 37148410.
9. Kelly FA, et al. Metformin with inositol versus metformin alone. Pooled analysis of 6 randomised trials, 388 women, 3 to 6 months. Endocrine, 2025. PubMed 39331347.
10. Irfan NU, et al. Multimodal therapy with metformin, inositol and dietary restriction. Randomised trial, 192 women, four groups, 12 weeks. Clinical Endocrinology, 2026. PubMed 42517336.
11. NICE. Polyendocrine metabolic ovarian syndrome (PMOS): assessment and management. Draft guideline for consultation, July 2026; consultation closed 11 August 2026; publication expected 27 January 2027. GID-NG10436, nice.org.uk/guidance/indevelopment/gid-ng10436.

Spotted something out of date? Email support@wellgard.co.uk with the page address and the sentence.