Which specific modes of exercise training are most effective for treating adolescent idiopathic scoliosis? A systematic review and network meta-analysis

This study aimed to compare the relative effectiveness of exercise-based interventions for adolescent idiopathic scoliosis (AIS) on spinal curvature, axial trunk rotation, and health-related quality of life. Web of Science, Cochrane Library, Embase, PubMed, and Scopus were systematically searched from inception to July 20, 2025. Additional records were identified through reference lists and gray literature screening. Eligible studies were randomized controlled trials enrolling participants with AIS that compared core stabilization exercises (CSE), physiotherapy scoliosis-specific exercises (PSSE, excluding Schroth exercises), Schroth exercises (SE, subgroup of PSSE), or conventional exercise therapy (CET) with each other or with non-exercise controls (CG). Trials were required to report outcomes for Cobb angle, axial trunk rotation (ATR), or the Scoliosis Research Society-22 (SRS-22) questionnaire. Two reviewers independently screened the studies, with disagreements resolved by consensus. Data extraction and risk of bias assessment (Cochrane RoB 2.0 tool) were independently performed by two reviewers, and any discrepancies were resolved through consultation with a third reviewer. Thirty-six trials ( n = 1,911) met the inclusion criteria. A frequentist network meta-analysis indicated that SE produced the greatest reductions in Cobb angle [SMD − 1.19 (95% CI: -1.94 to -0.43) ] and ATR [SMD − 1.37 (95% CI: -2.37 to -0.37)] compared with CET. Relative to the CG, SE also significantly improved Cobb angle [SMD − 0.89 (95% CI: -1.57 to -0.22)] and ATR [SMD − 1.29 (95% CI: -2.15 to -0.44)]. Both CSE and PSSE were superior to CET for Cobb angle improvement (SMD ≈ -0.90), though their effects did not differ significantly from SE. No exercise modality demonstrated a statistically significant advantage in SRS-22 scores. Surface under the cumulative ranking curve analysis ranked SE highest for curvature correction (94%), followed by CSE (74%) and PSSE (71%). Global and node-splitting analyses supported network consistency. SE provides the most pronounced corrective effects on spinal curvature and rotation in AIS, while CSE and PSSE represent effective alternative interventions when SE implementation is limited. Exercise modality does not appear to differentially affect health-related quality of life. Several studies suggest that exercise interventions may improve appearance; however, as these conclusions are primarily based on qualitative analysis, further high-quality research is needed to validate these effects. The study protocol was registered on the International Prospective Register of Systematic Reviews. Registration Number: CRD420251049820.