Berberine vs inositol
These get compared because both are sold for insulin resistance, especially to women with PCOS. The papers are about different questions. Berberine has type 2 diabetes glucose meta-analyses. Inositol has PCOS reviews that do not agree with each other. We did not read a human trial of one against the other.
| Berberine | Inositol | |
|---|---|---|
| Where the trials cluster | Type 2 diabetes glucose markers | Polycystic ovary syndrome |
| Head-to-head in humans | We did not read a trial of berberine versus inositol. A network meta-analysis included both and did not conclude that one beats the other. | |
| Guideline tone | Glucose meta-analyses, with the limits on the metformin page | The 2023 PCOS guideline review called the evidence limited and inconclusive |
Inositol, and why the reviews disagree
The review written for the 2023 international PCOS guideline (Fitz 2024, PMID 38163998, doi:10.1210/clinem/dgad762) included 30 trials and 2,230 women, with 19 pooled. It found possible metabolic benefits and a possible ovulation signal for D-chiro-inositol, no clear effect on other outcomes, and fewer gut side effects than metformin. The conclusion is blunt: evidence for inositol in PCOS is limited and inconclusive.
Greff 2023 (PMID 36703143) is more enthusiastic and should be read next to Fitz, not instead of it. Twenty-six trials, 1,691 women (806 inositol, 311 placebo, 509 metformin). Versus placebo, the chance of a regular cycle was 1.79 times higher (95% CI 1.13 to 2.85), and the authors said inositol was non-inferior to metformin on that outcome. Versus placebo they also report mean differences of −0.45 for BMI (−0.89 to −0.02), −0.41 for free testosterone (−0.69 to −0.13), −20.39 for total testosterone (−40.12 to −0.66), −0.69 for androstenedione (−1.16 to −0.22), −3.14 for glucose (−5.75 to −0.54), and −2081.05 for insulin AUC (−2745.32 to −1416.78). The abstract does not print the units on those laboratory mean differences. Sex-hormone-binding globulin rose (mean difference 32.06, 1.27 to 62.85).
Kelly 2025 (PMID 39331347) asked a narrower question: metformin plus inositol versus metformin alone. Six trials, 388 women, 3 to 6 months. Cycles were more regular (risk ratio 1.56, 1.01 to 2.41). The modified Ferriman-Gallwey score mean difference was −0.97 (95% CI −1.53 to −0.40). LH/FSH ratio fell by 0.13. Acne, BMI, fasting glucose, and HOMA-IR were not significant.
Berberine is a different literature
The glucose figures for berberine are on berberine vs metformin: Xie 2022 and the Yin pilot. Zhao 2021 (PMID 34407851) put metformin, thiazolidinediones, inositol, and berberine into one network of 22 PCOS trials and 1,079 women. The result the authors highlight is myo-inositol plus D-chiro-inositol, and metformin plus a thiazolidinedione, not a verdict that berberine beats inositol. The menstrual-frequency odds ratio for the inositol combination versus metformin was 14.70 with a confidence interval from 2.31 to 93.58. That interval is too wide to treat as a precise ranking.
Ge 2026 (PMID 41761674) combined inositol and berberine in a mouse model of PCOS. That is preclinical. It is not evidence that women should buy both.
The PCOS-specific page already on this site is inositol vs berberine for PCOS. This page does not copy numbers from it that were not re-read here.
Sources
- Fitz V, et al. Inositol for PCOS, for the 2023 guideline. PMID 38163998. DOI 10.1210/clinem/dgad762.
- Greff D, et al. Inositol in PCOS. PMID 36703143.
- Kelly FA, et al. Metformin plus inositol versus metformin. PMID 39331347.
- Zhao H, et al. Network meta-analysis of oral insulin sensitizers in PCOS. PMID 34407851.
- Yin J, et al. Berberine pilot. PMID 18442638.
- Ge J, et al. Inositol plus berberine in a mouse PCOS model. PMID 41761674.