Matters arising: statistical overfitting and outcome discordance undermine the claimed superiority of T2 fusion for shoulder balance in adolescent idiopathic scoliosis

A recent study by Li et al. evaluated shoulder balance in Lenke type 2 adolescent idiopathic scoliosis and concluded that the ratio of proximal to main thoracic curve (PTC/MTC) correction independently predicts postoperative shoulder imbalance (PSI), further suggesting that fusion to T2 provides superior shoulder balance compared with fusion to T3/4. A re-examination of the study’s methodology indicates that these conclusions are compromised by two principal limitations: (1) statistical overfitting and quasi-separation within the multivariable logistic regression model used to identify predictors of PSI, and (2) lack of alignment between radiographic findings and patient-reported outcomes. These methodological and interpretive issues are well recognized in the clinical and statistical literature, and addressing them is essential for ensuring valid inference and accurate clinical interpretation.