This meta-analysis evaluates the efficacy of acupuncture in improving glucose metabolism in patients with metabolic disorders. We comprehensively searched PubMed, Web of Science, Excerpta Medica Database (Embase), Chinese National Knowledge Infrastructure (CNKI), Wanfang, and other databases from January 2017 to October 2025 for randomized controlled trials (RCTs) comparing acupuncture to control interventions. Fourteen RCTs involving 1179 participants were included. Pooled random-effects analyses demonstrated that acupuncture was associated with greater improvements in several glycemic parameters compared with control groups: homeostasis model assessment of insulin resistance (HOMA-IR; MD -0.79, 95% CI -1.30 to -0.28, I²=91%), glycosylated hemoglobin A1c (HbA1c; MD -0.68%, 95% CI -1.02 to -0.34, I²=88%), fasting plasma glucose (FPG; MD -0.61 mmol/L, 95% CI -0.82 to -0.41, I²=95%), 2-hour Postprandial Plasma Glucose (P2hPG; MD -0.94 mmol/l, 95% CI -1.49 to -0.39, I²=92%) and fasting insulin (FINS; MD -3.27 mU/L, 95% CI -5.59 to -0.95). However, these pooled estimates were accompanied by substantial to considerable heterogeneity, and findings varied across stratified analyses based on control type. Subgroup analyses suggested that manual acupuncture might be more effective. The overall certainty of evidence was moderate, with heterogeneity and potential risks of selection and detection bias, identified as major limitations. Adverse events were infrequent and mild. In conclusion, this meta-analysis indicates a potential role for acupuncture in managing glucose metabolism, but the high heterogeneity and methodological limitations preclude definitive conclusions. Future large-scale, rigorously designed RCTs with homogeneous comparators and standardized protocols are needed to establish its efficacy and optimal role in clinical practice. • An updated meta-analysis on acupuncture for glucose metabolism. • Shows potential improvements in key markers like HOMA-IR. • Manual acupuncture appears more effective than other forms. • High heterogeneity indicates need for cautious interpretation.
Jiang et al. (Sun,) studied this question.