• Kun Luo, Yong-Quan Huang, Liang-Bo Zhu, Xin-Rong Gan, Yu Zhang, Shi-Ning Xiao, Rong-Ping Zhou, Jiang-Wei Chen, Jia-Ming Liu, and Zhi-Li Liu.
• Medical Innovation Center, the First Affiliated Hospital of Nanchang University, Nanchang, PR China; Institute of Spine and Spinal Cord, Nanchang University, Nanchang, PR China.
• World Neurosurg. 2023 Jun 19.

ObjectiveTo identify the risk factors for developing postoperative pulmonary infection in patients with acute cervical spinal cord injury (CSCI), and to develop a nomogram prediction model.MethodsPatients with CSCI who were admitted to 3 different medical centers between July 2011 and July 2021 were included in this study. All patients underwent cervical spine surgery. Data for patients admitted to the first 2 centers were included in a training set to establish the nomogram prediction model, and data for patients admitted to the third center were included in a validation set to externally verify the efficacy of the prediction model. For the training set, patients were divided into an infected group and a noninfected group (control group). Independent risk factors for postoperative pulmonary infection in patients with CSCI were identified by univariate and multivariate logistic regression analyses. Additionally, a nomogram prediction model was developed and validated based on the risk factors.ResultsA total of 689 patients were enrolled, including 574 for the training set and 115 for the validation set. Of the patients included for the training set, 144 developed pulmonary infection, with an incidence of 25.09%; 40 patients included for the validation set developed pulmonary infection (34.78%). Multivariate logistic regression analysis showed that age, American Spinal Injury Association grade, steroid pulse, high-level injury, smoking, multistage surgery, and operation duration were risk factors for the development of postoperative pulmonary infection in patients with CSCI. The area under the curve of the receiver operating characteristic curve of the model built by the training set was 0.905, and that of the receiver operating characteristic curve of the verification set was 0.917. The decision curve indicated that the model was in the range 1%-100%, and the predicted net benefit value of the model was high.ConclusionsAge, American Spinal Injury Association grade, steroid pulse, CSCI site, smoking history, number of surgical levels, and surgical duration are correlated with the development of postoperative pulmonary infection in patients with CSCI. The risk prediction model of postoperative pulmonary infection has a good prediction efficiency and accuracy.Copyright © 2023 Elsevier Inc. All rights reserved.

5 keywords...

hide…