Real-World Effectiveness of Conservative Treatment in Idiopathic Scoliosis

Background. Idiopathic scoliosis is a three-dimensional deformity of the spine affecting 2-3% of the paediatric population, mainly during adolescence (adolescent idiopathic scoliosis, AIS). If not adequately treated during pubertal growth, the curve may progress beyond critical thresholds: 30° Cobb, below which the risk of chronic back pain and progression in adulthood is very low, and 45-50° Cobb, the surgical threshold.

The current evidence on bracing in AIS is based on the BrAIST randomised controlled trial, which showed that bracing prevents surgery. However, RCTs have intrinsic limits of generalisability: (1) strict inclusion criteria, since BrAIST included only about 2% of the real clinical population; (2) a single standardised treatment, while in clinical practice treatment is personalised according to patient response and shared decision making (EBPA); (3) a focus on AIS, leaving infantile and juvenile forms poorly studied.

This study will use validated prognostic models (IS-GROWTH and BrAIST-Calc) to build an individualised control population based on each patient's predicted risk of progression without treatment. Comparing the outcome observed at skeletal maturity with the predicted risk allows the effectiveness of treatment to be measured in a real-world setting on an unprecedented scale.

Design. Retrospective, observational, non-pharmacological, non-profit study based on data collected during routine clinical practice in the ISICO institutional clinical database, active since 2003. The study is a pragmatic benchmarking of clinical practice.

Data collected. Parameters recorded every 4-6 months during treatment will be extracted: radiographic data (Cobb angle, Risser stage, sagittal parameters: kyphosis and lordosis); clinical data (angle of trunk rotation, ATR; Trunk Aesthetic Clinical Evaluation, TRACE; weight; height); quality-of-life questionnaires (SRS-22, ISYQOL, SCaRES); treatment data (brace type: Sibilla, Lapadula, Sforzesco, PASB; prescribed and actual brace wear time, self-reported or measured with thermal sensors; Scientific Exercises Approach to Scoliosis, SEAS).

Definition of success. (1) Stability or improvement of the Cobb angle (change < 5° or reduction ≥ 5°); (2) prevention of surgery, i.e. not reaching 45° (BrAIST-Calc) and 50° Cobb (IS-GROWTH); (3) low-risk target, i.e. end of growth with a curve < 30° Cobb.

Statistical analysis. Analyses will be performed mainly with R: descriptive statistics (mean and standard deviation, or median and percentiles; frequencies with 95% confidence intervals); standardised incidence ratio (SIR) of observed versus expected cases according to natural history models, with 95% confidence intervals and Poisson statistics; relative risk (RR) and number needed to treat (NNT); intention-to-treat analysis, with drop-outs considered treatment failures; per-protocol analysis. Sample size is not predetermined: the whole eligible population will be analysed to maximise statistical power in subgroup analyses.

Ethics and data protection. The study is conducted in accordance with the Declaration of Helsinki. As a retrospective analysis of routinely collected data, it involves no additional risk for participants. Given the objective impossibility of contacting all patients, a waiver of informed consent is requested from the Ethics Committee, in accordance with GDPR articles 6, 9(2)(j) and 89, article 110 of the Italian Privacy Code and the Italian Data Protection Authority provision of 9 March 2023. A Data Protection Impact Assessment (DPIA) will be performed. Data will be pseudonymised and analysed only in aggregate form.