Novel risk factors and personalized risk calculator for predicting proximal junctional kyphosis after adult spinal deformity surgery.
Discover the novel risk factors and personalized calculator that predict spinal deformity surgery complications, empowering doctors to tailor treatment and prevent debilitating outcomes.
This study used machine learning to identify 7 modifiable risk factors for proximal junctional kyphosis (PJK) after adult spinal deformity surgery, including pelvic compensation and preoperative lower limb function. The random forest model achieved high predictive performance (AUC 0.847), enabling a personalized risk calculator to guide individualized perioperative management and prevent this prevalent, detrimental complication. Limitations include the retrospective design and single-center data, warranting further validation.