Stefano Negrini, Silvia Minozzi, Josette Bettany‐Saltikov, Nachiappan Chockalingam, Theodoros B. Grivas, Tomasz Kotwicki, Toru Maruyama, Michele Romano, Fabio Zaina

Idiopathic scoliosis is a three-dimensional deformity of the spine. The most common form is diagnosed in adolescence. While adolescent idiopathic scoliosis (AIS) can progress during growth and cause a surface deformity, it is usually not symptomatic. However, in adulthood, if the final spinal curvature surpasses a certain critical threshold, the risk of health problems and curve progression is increased. To evaluate the efficacy of bracing for adolescents with AIS versus no treatment or other treatments, on quality of life, disability, pulmonary disorders, progression of the curve, and psychological and cosmetic issues. We searched CENTRAL, MEDLINE, EMBASE, five other databases, and two trials registers up to February 2015 for relevant clinical trials. We also checked the reference lists of relevant articles and conducted an extensive handsearch of grey literature. Randomized controlled trials (RCTs) and prospective controlled cohort studies comparing braces with no treatment, other treatment, surgery, and different types of braces for adolescent with AIS. We used standard methodological procedures expected by The Cochrane Collaboration. We included seven studies (662 participants). Five were planned as RCTs and two as prospective controlled trials. One RCT failed completely, another was continued as an observational study, reporting also the results of the participants that had been randomized.There was very low quality evidence from one small RCT (111 participants) that quality of life (QoL) during treatment did not differ significantly between rigid bracing and observation (mean difference (MD) -2.10, 95% confidence interval (CI) -7.69 to 3.49). There was very low quality evidence from a subgroup of 77 adolescents from one prospective cohort study showing that QoL, back pain, psychological, and cosmetic issues did not differ significantly between rigid bracing and observation in the long term (16 years).Results of the secondary outcomes showed that there was low quality evidence that rigid bracing compared with observation significantly increased the success rate in 20° to 40° curves at two years' follow-up (one RCT, 116 participants; risk ratio (RR) 1.79, 95% CI 1.29 to 2.50). There was low quality evidence that elastic bracing increased the success rate in 15° to 30° curves at three years' follow-up (one RCT, 47 participants; RR 1.88, 95% CI 1.11 to 3.20).There is very low quality evidence from two prospective cohort studies with a control group that rigid bracing increases the success rate (curves not evolving to 50° or above) at two years' follow-up (one study, 242 participants; RR 1.50, 95% CI 1.19 to 1.89) and at three years' follow-up (one study, 240 participants; RR 1.75, 95% CI 1.42 to 2.16). There was very low quality evidence from a prospective cohort study (57 participants) that very rigid bracing increased the success rate (no progression of 5° or more, fusion, or waiting list for fusion) in adolescents with high degree curves (above 45°) (one study, 57 adolescents; RR 1.79, 95% CI 1.04 to 3.07 in the intention-to-treat (ITT) analysis).There was low quality evidence from one RCT that a rigid brace was more successful than an elastic brace at curbing curve progression when measured in Cobb degrees in low degree curves (20° to 30°), with no significant differences between the two groups in the subjective perception of daily difficulties associated with wearing the brace (43 girls; risk of success at four years' follow-up: RR 1.40, 1.03 to 1.89). Finally, there was very low quality evidence from one RCT (12 participants) that a rigid brace with a pad pressure control system is no better than a standard brace in reducing the risk of progression.Only one prospective cohort study (236 participants) assessed adverse events: neither the percentage of adolescents with any adverse event (RR 1.27, 95% CI 0.96 to 1.67) nor the percentage of adolescents reporting back pain, the most common adverse event, were different between the groups (RR 0.72, 95% CI 0.47 to 1.10). Due to the important clinical differences among the studies, it was not possible to perform a meta-analysis. Two studies showed that bracing did not change QoL during treatment (low quality), and QoL, back pain, and psychological and cosmetic issues in the long term (16 years) (very low quality). All included papers consistently showed that bracing prevented curve progression (secondary outcome). However, due to the strength of evidence (from low to very low quality), further research is very likely to have an impact on our confidence in the estimate of effect. The high rate of failure of RCTs demonstrates the huge difficulties in performing RCTs in a field where parents reject randomization of their children. This challenge may prevent us from seeing increases in the quality of the evidence over time. Other designs need to be implemented and included in future reviews, including 'expertise-based' trials, prospective controlled cohort studies, prospective studies conducted according to pre-defined criteria such as the Scoliosis Research Society (SRS) and the international Society on Scoliosis Orthopedic and Rehabilitation Treatment (SOSORT) criteria. Future studies should increase their focus on participant outcomes, adverse effects, methods to increase compliance, and usefulness of physiotherapeutic scoliosis specific exercises added to bracing.