Radiographic quantification of active self-correction during physiotherapeutic scoliosis-specific exercises and its potential prognostic value in adolescent idiopathic scoliosis

No previous study has radiographically quantified in-exercise correction during physiotherapeutic scoliosis-specific exercises (PSSE) or examined its potential prognostic value for structural outcomes in adolescent idiopathic scoliosis (AIS). This study aimed to quantify correction achieved during active self-correction (ASC) in PSSE and determine its association with structural outcomes. This prospective longitudinal prognostic study with blinded outcome assessment included patients with AIS with a major thoracic curve who completed a 6-month supervised and home-based PSSE program. Full-spine radiographs were obtained in natural posture (Pre), and during active self-correction (ASC) after a 2-week familiarization period, and after completion of the 6-month PSSE program in natural standing posture (Post). Cobb angle, apical vertebral rotation (AVR), angle of trunk rotation (ATR), self-perceived appearance, quality of life, self-efficacy, and adherence were assessed. Associations between in-exercise correction and post-treatment structural changes were analyzed using correlation and stepwise regression. Cobb angle differed significantly across conditions, with values of 37.57° at Pre, 28.82° during ASC, and 35.82° at Post, corresponding to 27.32% correction during ASC. AVR changed significantly, with the greatest improvement during ASC. Self-efficacy was associated with adherence. Post-treatment Cobb angle change was associated with adherence, ATR change, and ASC correction magnitude. In stepwise regression, in-exercise Cobb correction from Pre to ASC was the only variable retained in the final regression model (adjusted R²=0.309). Radiographically quantified ASC during PSSE was associated with Cobb angle change at 6-month follow-up. In-exercise ASC correction was the only variable retained as a significant predictor of follow-up Cobb angle change. Clinical Research Information Service (CRIS), Republic of Korea, ID: KCT0010485. Prospectively registered on 20 May 2025, prior to participant enrollment.