Individualizing Brace Treatment for Adolescent Idiopathic Scoliosis. Skeletal maturity, bracing strategies and prognostic modeling

Abstract

Nighttime bracing has been proposed as an alternative to full-time brace regimens for patients with adolescent idiopathic scoliosis (AIS) as it may reduce the burden of prolonged brace wear while maintaining comparable treatment effectiveness. Comparisons among various brace devices and regimens remain challenging, and treatment outcomes are often difficult to predict, since they are influenced by multiple factors, including skeletal maturity, brace type, and patient compliance. Aim The aim of this thesis was to compare the effectiveness of the two most commonly used brace devices in Sweden at a single institution, to develop a new maturity scale that would link pelvic and hand radiograph-based maturity assessments, and to develop a predictive model to estimate the risk of failure in nighttime brace treatment. Based on these findings, the thesis sought to contribute to more nuanced counseling regarding brace treatment. Study population The study population consisted of patients with AIS treated with either a full-time or nighttime brace between 2003 and 2023 at Sahlgrenska University Hospital, Gothenburg, Sweden. Among these, only patients who underwent nighttime brace treatment were included in the development of the prediction model. Methods In Study I, the effectiveness of the two treatment regimens was compared for the overall cohort as well as for selected subgroups. In Study II, a novel maturity scale, Pelvic Maturity Indicators (PMI)—an extension of the Risser system—was developed and validated against Sanders Maturity Scale (SMS). External validation of the PMI method was performed in Study III using inter-instrument agreement analysis in randomized, blinded surveys. Inter- and intra-observer agreement was assessed among raters with varying levels of experience. A multivariable prediction model for treatment outcome following nighttime brace treatment was developed in Study IV. The most influential predictors identified in the final model were subsequently incorporated into a regression equation and implemented in the Nighttime Brace Calculator. Results Full‑time bracing was more effective than nighttime bracing in preventing curve progression despite lower compliance and less in‑brace correction. Similar results were observed in pre‑menarcheal females and in patients treated for thoracic curves. In contrast, no differences in outcomes were found among post‑menarcheal females or patients with lumbar curves. The PMI derived from pelvic radiographs demonstrated substantial agreement with the SMS derived from hand radiographs. These results were successfully validated in an independent, blinded, randomized survey. The PMI scale demonstrated strong inter- and intra-rater reliability. The statistical model predicting the likelihood of a curve reaching the surgical threshold in patients treated with nighttime bracing demonstrated good discrimination between low- and high-risk predictions. Curve magnitude, in-brace correction, skeletal maturity, and treatment compliance were retained as significant predictors in the model. Conclusion Full-time bracing demonstrated overall superior effectiveness, although nighttime bracing may be appropriate for patients who have already passed the most vulnerable phase of scoliosis progression. The PMI effectively captures the period of rapid curve deterioration and provides a link between pelvic- and hand-derived skeletal maturity assessments. The Nighttime Brace Calculator can support clinical decision‑making by identifying the most appropriate treatment based on individual patient characteristics.