Risk prediction model of pedicle screw loosening within 2 years after decompression and instrumented fusion surgery for degenerative lumbar disease.

Abstract

Pedicle screw loosening (PSL) after spinal fusion surgery is one of the most frequently reported complications and leads to poor clinical outcomes. This study aimed to develop and validate a risk prediction model for PSL within 2 years in patients undergoing lumbar instrumented fusion surgery based on their risk profiles. Retrospective, observational study. Patients who underwent lumbar instrumented fusion surgery at a single academic institution between May 2015 and February 2019. Risk assessment of PSL and development of a rating score based on patient characteristics. The demographic profiles and radiographic parameters using computed tomography were obtained. These factors were analyzed to determine possible risk factors related to postoperative PSL after 2 years. A scoring system was developed using these independent risk factors and validated using prospectively collected data from another center between May 2019 and December 2021. The occurrence of PSL within 2 years postoperation was 12.7% (40/315). PSL was significantly predicted by smoking, low Hounsfield units (HU) of the pedicle tract at the index level (P), and a low psoas-lumbar vertebral index (PLVI). The risk of PSL according to the categories of the risk score was 1.1% for those with a score of 0-1, 15.1% for a score of 2-3, and 61.5% for a score of 4-6. In validation, this model demonstrated both good discrimination and calibration results. The area under the curve was 0.887 (95% CI 0.830-0.938) for the derivation cohort and 0.835 (95% CI 0.738-0.918) for the external validation cohort. This PSL risk score, including smoking, Index P HU, and PLVI, is a novel approach to predict PSL 2 years postsurgery. This approach highlights the role of factors associated with osteoporosis and sarcopenia in the development of PSL and could aid in preoperative decision-making and surgical planning.