Changes in Muscle Strength Following Semaglutide Treatment in Adults: A Systematic Review With Meta‐Analysis of Handgrip Strength

Aims To systematically review and quantitatively synthesize within‐group changes in skeletal muscle strength following semaglutide treatment in adults. Materials and Methods This PRISMA 2020 systematic review was registered in PROSPERO (CRD420251185935). MEDLINE, Embase, CENTRAL, LILACS and Google Scholar were searched through 20 August 2026, without language restrictions. Eligible studies included adults treated with semaglutide reporting quantitative skeletal muscle strength outcomes, including handgrip and knee extension strength. Randomized, non‐randomized and uncontrolled before‐after studies were included. The primary meta‐analysis pooled within‐group changes in handgrip strength from baseline to the longest available follow‐up of semaglutide‐only arms using a random‐effects model. Knee extension strength was summarized descriptively. Results Sixteen reports represented eleven studies, all assessing handgrip strength; one small randomized trial also assessed knee extension strength. Ten studies (660 semaglutide‐treated participants) contributed to the primary meta‐analysis of uncontrolled within‐group changes, showing a nonsignificant pooled handgrip strength change (−0.16 kgf; 95% confidence interval: −1.98 to 1.65), with substantial heterogeneity ( I 2 = 90.8%) and a 95% prediction interval from −5.96 to 5.63 kgf. Duration‐stratified and sensitivity analyses were also nonsignificant. Individual study findings were heterogeneous. Certainty was very low. Conclusions Available evidence is insufficient to determine whether semaglutide is associated with changes in handgrip strength because the principal synthesis was based on uncontrolled within‐group changes, with very low certainty and significant heterogeneity. These findings do not establish strength preservation or a causal treatment effect. More robust comparative studies with standardized strength assessment and longer follow‐up are needed.