Comparative effectiveness and safety of acupuncture vs metformin in insulin-resistant polycystic ovary syndrome women: A network meta-analysis of RCTs.

BACKGROUND

The purpose of this study was to conduct a network meta-analysis (NMA) comparing the effectiveness and safety of traditional acupuncture with metformin. This study aims to identify an effective and safe alternative treatment for women with polycystic ovary syndrome (PCOS) complicated by insulin resistance (IR).

METHODS

We searched PubMed, EMBASE, Web of Science, Cochrane, ClinicalTrials.gov, and relevant citations for eligible randomized controlled trials (RCTs) published up to August 12, 2024. Data extraction and bias assessment were performed by two independent reviewers, and Bayesian NMA was performed using the GeMTC package, with a random effects model. The NMA compared the effectiveness and safety of acupuncture with metformin which were measured using the Homeostasis Model Assessment Insulin Resistance (HOMA-IR) index, fasting plasma glucose, fasting insulin, and the incidence of adverse events (AEs).

RESULTS

We included 12 RCTs ( = 870). Four RCTs compared acupuncture with metformin, six compared metformin with control interventions (blank control or placebo), and two compared acupuncture with sham acupuncture. Acupuncture (Mean difference (MD): -0.76, 95 % credible interval (CrI): -1.42, -0.03) and metformin (MD: -0.74, 95 % CrI: -1.28, -0.17) significantly reduced the IR level (HOMA-IR). No statistical difference was observed between both treatment methods in improving glucose metabolism; however, acupuncture had fewer AEs (risk ratio: 0.13, 95 % CrI: 0.01, 0.68). Acupuncture all ranked high in terms of surface under the cumulative ranking curve across all outcomes, thus demonstrating better effectiveness and safety potential.

CONCLUSIONS

Acupuncture may be a potential alternative for PCOS women with IR, as it is equally effective, easier to perform, and relatively safer than metformin. Further trials are necessary to confirm these findings owing to the current lack of sufficient evidence.: PROSPERO, CRD42024581934.