Discrepancy Between Radiographic Correction and Cosmetic Improvement in Adolescent Idiopathic Scoliosis: The Role of Extremely Low BMI

Study Design Retrospective matched cohort study. Objectives To investigate the association between body mass index (BMI) and postoperative cosmetic improvement in adolescent idiopathic scoliosis (AIS). Methods Among 737 female AIS patients, the 10th percentile of the raw BMI distribution was used to define “extremely low BMI”. Lenke type 1 patients with extremely low BMI were matched 1:1 to non-extremely low BMI patients by age, apical vertebra, and preoperative Cobb angle. Sex, Risser sign category, curve pattern and surgical approach were standardized through the inclusion criteria. Pre- and 1-year postoperative radiographic and cosmetic parameters and SRS-22 scores were compared using paired analyses accounting for matched-pair design. The primary outcome was the between-group difference in postoperative cosmetic improvement; all other analyses were prespecified as secondary or exploratory. Results The BMI threshold for extremely low BMI was 15.57 kg/m 2 . Radiographic correction was comparable between cohorts. However, the non-extremely low BMI group improved more in shoulder area indices (SAI1/SAI2), shoulder (SA) and axilla (AA) angles, right-left waist angle difference (RLWAD), and posterior superior iliac spine (PSIS) angle (all P < 0.05). The extremely low BMI group had lower SRS-22 self-image (3.3±0.6 vs. 4.2±0.8) and mental health (3.4±0.4 vs. 4.1±0.3) scores (both P ≤ 0.010). Fisher’s r-to-z test showed significantly weaker or directionally different radiographic–cosmetic correlations in the extremely low BMI group (P<0.05). Conclusions Extremely low BMI was associated with poorer cosmetic improvement, worse self-image, and weaker radiographic–cosmetic correlations, suggesting a wider discrepancy between radiographic correction and cosmetic improvement.