Orthopedics · Medical Literature
Effects of exercise interventions on adolescents with mild to moderate scoliosis: a meta-analysis.
Abstract
Adolescent idiopathic scoliosis (AIS) is a prevalent spinal deformity that impairs physical function and quality of life. While supervised exercise is recommended for moderate cases, its effectiveness remains debated. This meta-analysis evaluates the impact of exercise therapy on Cobb angle, trunk rotation (ATR), and quality of life (QoL) in AIS patients. We systematically searched PubMed, Web of Science, EBSCO and China National Knowledge Infrastructure (CNKI) databases from inception to October 2025 for randomized controlled trials (RCTs) investigating exercise in AIS. Study quality was assessed using the PEDro scale. Data synthesis involved pooling effect sizes, conducting subgroup analyses, and assessing publication bias using Stata 18. A total of 26 studies (21 RCT and five controlled trials) were included in the meta-analysis. The exercise intervention demonstrated statistically significant improvements across multiple functional domains: reduced the Cobb angle (Hedge's = 1.22, 95% CI [0.90-1.53]), improved the ATR (Hedge's systematically searched 1.25, 95% CI [0.74-1.76]), and enhanced QoL (Hedge's systematically searched 0.73, 95% CI [0.38-1.09]). Subgroup analyses further indicated that Schroth-based training (Hedge's = 1.304) and combined approaches (e.g., exercise with orthotic management) yielded the most favorable outcomes. Importantly, both short-term (≤12 weeks, 1-2 sessions/week) and long-term (≥25 weeks, ≥5 sessions/week) intervention durations demonstrated significant clinical efficacy. Exercise therapy demonstrates significant potential as a fundamental conservative intervention for improving spinal morphology, function, and quality of life in adolescents with idiopathic scoliosis. The current evidence supports its integration into primary clinical management protocols. Future research should focus on conducting multi-center randomized controlled trials to validate long-term efficacy and to establish optimal, individualized intervention protocols.
Clinical relevance
In adolescents with mild to moderate idiopathic scoliosis, does supervised exercise therapy produce statistically and clinically meaningful improvements in Cobb angle, angle of trunk rotation (ATR), and health-related quality of life compared to control conditions? __SNAPSHOT__ {"studyDesign":"Systematic review with meta-analysis of randomized controlled trials (RCTs) and controlled trials (CTs)","setting":"Multinational; databases searched include PubMed, Web of Science, EBSCO, and China National Knowledge Infrastructure (CNKI) from inception to October 2025; specific trial settings not reported in abstract","sampleSize":"26 studies included (21 RCTs and 5 controlled trials); aggregate patient-level N not reported in abstract","population":"Adolescents diagnosed with mild to moderate idiopathic scoliosis (AIS); specific Cobb angle thresholds for inclusion and patient demographics not reported in abstract","intervention":"Exercise therapy (type, duration, frequency, and supervision details not further specified in abstract)","comparator":"Control conditions as defined by individual included studies; specific nature of comparators (e.g., usual care, bracing, no treatment, sham) not reported in abstract","primaryOutcome":"Cobb angle reduction; secondary outcomes include angle of trunk rotation (ATR) and health-related quality of life (QoL)","followUp":"Not reported in abstract","fundingSource":"Not reported"}
Key findings
Exercise therapy was associated with statistically significant improvements across all three pooled outcome domains compared to control. For Cobb angle reduction, the pooled effect size was Hedges' g = 1.22 (95% CI: 0.90–1.53); for ATR improvement, Hedges' g = 1.25 (95% CI: 0.74–1.76); and for quality of life, Hedges' g = 0.73 (95% CI: 0.38–1.09). All three confidence intervals exclude zero, indicating statistical significance; however, p-values and heterogeneity statistics (I² values) are not reported in the abstract. Effect sizes for Cobb angle and ATR exceed conventional thresholds for a 'large' effect (g > 0.8), and the QoL effect size falls in the 'medium-to-large' range. These are standardized mean differences and cannot be directly translated into absolute degree changes in Cobb angle without individual study data. Subgroup analyses were conducted but results are incompletely reported in the provided abstract. Absolute risk differences, NNT/NNH, and adverse event data are not reported.
Practice implications
This meta-analysis of 26 controlled studies reports large standardized effect sizes favoring exercise therapy over control conditions for Cobb angle, ATR, and QoL in adolescents with mild to moderate AIS. However, several factors limit immediate clinical translation. First, the abstract is truncated and subgroup analysis results—critical for understanding which exercise types, durations, or patient subgroups drive the pooled benefit—are not available for review. Second, the inclusion of five non-randomized controlled trials alongside 21 RCTs introduces heterogeneity in study quality and potential for selection bias. Third, heterogeneity statistics (I²) are not reported, so the degree of between-study variability is unknown; large effect sizes with wide CIs (particularly ATR: 0.74–1.76) raise concern for meaningful heterogeneity. Fourth, Hedges' g does not translate directly into clinically meaningful degree changes in Cobb angle; the minimum clinically important difference for Cobb angle is generally cited as 5°, and whether the pooled effect achieves this threshold cannot be determined from the abstract alone. Fifth, the inclusion of CNKI-indexed Chinese-language trials, while broadening scope, is a population with known susceptibility to publication bias and reporting inconsistencies. Adolescents with severe scoliosis (Cobb angle >45°), those who are skeletally mature, and those from non-Asian populations may not be well represented. The findings support exercise as a beneficial conservative intervention for mild-to-moderate AIS, consistent with SOSORT guidelines, but the magnitude of effect should be interpreted cautiously pending full-text review. NOTE: This summary is based solely on the author-provided abstract; the full manuscript was not retrievable at the time of analysis, and subgroup data, heterogeneity metrics, risk-of-bias results, and publication bias assessments cannot be evaluated. __LIMITATIONS__ 1. Abstract truncation: Subgroup analysis findings and statistical heterogeneity data (I²) are incompletely reported, preventing full assessment of the evidence base. 2. Study design heterogeneity: Inclusion of five non-randomized controlled trials alongside RCTs may inflate pooled effect estimates. 3. Publication bias: Inclusion of CNKI-indexed trials raises concern for language and publication bias; Stata-based publication bias testing is referenced but results not reported in abstract. 4. Clinical interpretability: Hedges' g effect sizes cannot be converted to absolute Cobb angle changes without patient-level data; clinical significance relative to minimum clinically important differences is unknown. 5. Exercise heterogeneity: The abstract does not specify whether pooled studies used Schroth, SEAS, yoga, core stabilization, or other modalities, limiting exercise-type-specific conclusions. 6. Follow-up duration: Not reported; durability of benefit after cessation of supervised exercise is unknown. 7. Sample size and aggregate N: Total patient numbers across studies not reported, limiting power calculations and precision assessment. 8. Blinding: Patient and assessor blinding for exercise trials is inherently limited; performance bias likely present. 9. Generalizability: Potential overrepresentation of Asian (particularly Chinese) populations. 10. Funding and conflict of interest: Not reported. __BOTTOM_LINE_CATEGORY__ potentially_practice_informing __BOTTOM_LINE__ This meta-analysis of 26 controlled studies reports large pooled effect sizes favoring exercise therapy for Cobb angle reduction, trunk rotation improvement, and quality of life in adolescents with mild to moderate idiopathic scoliosis. However, the abstract is truncated, heterogeneity statistics and subgroup results are unavailable, and Hedges' g values cannot be directly equated to clinically meaningful degree changes in Cobb angle. These findings are consistent with—and incrementally supportive of—current SOSORT guideline recommendations for scoliosis-specific exercise in mild-to-moderate AIS, but full manuscript review is required before further practice changes are considered.