Abstract
C5 palsy is a significant yet poorly understood complication following cervical posterior surgery. Currently, few predictive models exist for estimating the preoperative risk of C5 palsy. This study aimed to develop and internally validate such a predictive model. We included patients aged 18 years or older who underwent cervical laminoplasty for cervical spondylosis or ossification of the posterior longitudinal ligament at a single institution. Demographic and radiographic data were collected. Key radiographic parameters included the C2-7 Cobb angle, C7 slope, presence of ossification of the posterior longitudinal ligament, anterior projection of the superior articular process of C5, and the width of the intervertebral foramen at C4/5, measured on computed tomography. Logistic regression with optimism adjustment was used to develop the model. A total of 180 patients were analyzed, with 18 cases (10.0%) of C5 palsy. Logistic regression identified width of the intervertebral foramen, C7 slope, age, and sex as significant predictors. The model demonstrated an area under the curve of 0.797 (95% confidence interval: 0.695-0.900) and a Brier score of 11.7%. After internal validation using bootstrapping, the optimism-adjusted area under the receiver operating characteristic curve was 0.736 (95% confidence interval 0.629-0.830). Final regression coefficients were 0.054 for C7 slope, -0.039 for age, -1.161 for female sex, and -0.721 for width of the intervertebral foramen. In conclusion, we developed and internally validated a preoperative prediction model for C5 palsy following double-door laminoplasty. Predictors such as width of the intervertebral foramen, C7 slope, age, and sex may aid in risk assessment and surgical planning.