Predicting risk factors for postoperative pneumonia in patients with lower limb fractures based on logistic regression model

Abstract

Abstract Purpose: Lower limb fracture is a frequent cause of hospitalization, and postoperative pneumonia is an important marker of hospital cost and quality of care provided. As an extension of traditional statistical methods, machine learning provides the possibility of accurately predicting the postoperative pneumonia. The aim of this paper is to retrospectively identify predictive factors of postoperative pneumonia by using multivariate logistic regression model. Methods: The incidence and admission of postoperative pneumonia in patients with lower limb fractures in the Second Affiliated Hospital of Nanchang University from 2017 to 2023 were retrospectively analysed. Patients who developed postoperative pneumonia during hospitalisation were defined as the pneumonia group, and those who did not develop postoperative pneumonia were defined as the no pneumonia group. Then logistic regression model of the postoperative pneumonia was developed and evaluated. Results: The incidence of postoperative pneumonia was 6.44%, and the AUC values was 0.821, indicating that the module could predict the occurrence of postoperative pneumonia to a large extent. Sex, age, smoking history, alcohol consumption history, operation time, cerebrovascular disease, hypertension, diabetes, fracture type, surgical grade, globulin ratio, platelets, and C-reactive protein were identifed as signifcant factors for postoperative pneumonia. Conclusions: Our proposed model corresponding to the predictors is designed to be convenient for clinical use. This model offers promising potential as a tool for the prevention and treatment of postoperative pneumonia in patients with lower limb fractures. Adopting appropriate health management methods may reduce the risk of postoperative pneumonia in patients with lower limb fractures.