A Predictive Scoring System for Pulmonary Complications after Posterior Instrumentation and Fusion for Non-degenerative Scoliosis

Abstract Objectives The aim of this study was to identify perioperative risk factors of pulmonary complications (PC) after posterior instrumentation and fusion (PIF) for non-degenerative scoliosis and devise a predictive scoring system that could be used by spine surgeons to predict the level of risk requiring advanced measures. Patients and methods A retrospective review was performed of 703 patients who underwent PIF at our center between January 2010 and January 2018. Preoperative, peri-operative, demographic data, surgical methods, and radiographic parameters were extracted to identify the predictors of PC, and a predictive scoring system was created using multiple logistic regression and receiver operator characteristic (ROC). Results There were statistically significant differences in revision surgery, preoperative pulmonary disease, Cobb angle and thoracoplasty between the patients who suffered from PC as compared to those who did not. The predictive scoring system included four dimensions. The area under the ROC curve for the system was 0.82 for PC after PIF at the optimal cut-off point, and the sensitivity was 79.2%, which demonstrated good diagnostic accuracy. Conclusion We identified a significant relationship between perioperative parameters (revision surgery, preoperative pulmonary disease, Cobb angle and thoracoplasty) and PC after PIF for non-degenerative scoliosis and created a predictive scoring system that can be used to prepare measures to manage PC after PIF. The system was proven to be reliable in this study.