IMPORTANCE Moderate-grade adolescent idiopathic scoliosis (AIS) may be treated with full-timebracing. For patients who reject full-time bracing, the effects of alternative, conservativeinterventions are unknown.

OBJECTIVE To determine whether self-mediated physical activity combined with either nighttimebracing (NB) or scoliosis-specific exercise (SSE) is superior to a control of physical activity alone (PA)in preventing Cobb angle progression in moderate-grade AIS.

DESIGN, SETTING, AND PARTICIPANTS The Conservative Treatment for Adolescent IdiopathicScoliosis (CONTRAIS) randomized clinical trial was conducted from January 10, 2013, throughOctober 23, 2018, in 6 public hospitals across Sweden. Male and female children and adolescentsaged 9 to 17 years with an AIS primary curve Cobb angle of 25° to 40°, apex T7 or caudal, and skeletalimmaturity based on estimated remaining growth of at least 1 year were included in the study. Datesof analysis were from October 25, 2021, to January 28, 2023.

INTERVENTIONS Interventions included self-mediated physical activity in combination with eitherNB or SSE or PA (control). Patients with treatment failure were given the option to transition to afull-time brace until skeletal maturity.

MAIN OUTCOMES AND MEASURES The primary outcome was curve progression of 6° or less(treatment success) or curve progression of more than 6° (treatment failure) seen on 2 consecutiveposteroanterior standing radiographs compared with the inclusion radiograph before skeletalmaturity. A secondary outcome of curve progression was the number of patients undergoing surgeryup until 2 years after the primary outcome.

RESULTS The CONTRAIS study included 135 patients (45 in each of the 3 groups) with a mean (SD)age of 12.7 (1.4) years; 111 (82%) were female. Treatment success was seen in 34 of 45 patients (76%)in the NB group and in 24 of 45 patients (53%) in the PA group (odds ratio [OR], 2.7; 95% CI, 1.1-6.6).The number needed to treat to prevent curve progression with NB was 4.5 (95% CI, 2.4-33.5).Treatment success occurred in 26 of 45 patients (58%) in the SSE group (OR for SE vs PA, 1.2; 95% CI,0.5-2.8). Up to 2 years after the primary outcome time point, 9 patients in each of the 3 groupsunderwent surgery.

CONCLUSIONS AND RELEVANCE In this randomized clinical trial, treatment with NB preventedcurve progression of more than 6° to a significantly higher extent than did PA, while SSE did not; inaddition, allowing transition to full-time bracing after treatment failure resulted in similar surgicalfrequencies independent of initial treatment. These results suggest that NB may be an effectivealternative intervention in patients rejecting full-time bracing.