Development and Validation of a Clinical Prediction Model for Postoperative Pneumonia in Elderly Patients with Femoral Neck Fracture

Objective: This study aimed to investigate the risk factors associated with postoperative pneumonia in elderly patients undergoing hip replacement for femoral neck fracture and to develop and validate a risk prediction model for such complications. Methods: A retrospective cohort of elderly patients (January 2022–September 2024) who underwent hip replacement for femoral neck fracture at the Qingpu Branch of Zhongshan Hospital, Fudan University, was analyzed. A total of 27 parameters were evaluated, including age, Modified 5-Factor Frailty Index (MFI-5), and anesthesia scores. The cohort was randomly divided into a derivation cohort (75.8%) and a validation cohort (24.2%). LASSO regression was applied to select optimal predictors through 10-fold cross-validation. Logistic regression identified significant factors for postoperative pneumonia, and a nomogram was constructed. Model accuracy was assessed using calibration curves, while predictive performance was evaluated via area under the receiver operating characteristic curve (AUC) and decision curve analysis (DCA). Results: Among 418 included patients, 317 in the derivation cohort were stratified into pneumonia (n=83) and non-pneumonia (n=234) groups. Multivariate logistic regression identified age >65 years, smoking history, hemiarthroplasty, American Society of Anesthesiologists (ASA) grade ≥3, and MFI-5 >3 as independent risk factors (all P<0.05). The nomogram demonstrated excellent consistency and accuracy. AUC values for the derivation and validation cohorts were 0.872 (95% CI: 0.82496–0.91904) and 0.895 (95% CI: 0.82444–0.96556), respectively. Calibration curves showed minimal deviation (mean absolute error: 0.022 vs. 0.039). ROC and DCA confirmed robust predictive efficacy. Conclusion: Advanced age (>65 years), smoking, hemiarthroplasty, ASA grade ≥3, and MFI-5 >3 are significant risk factors for postoperative pneumonia in elderly femoral neck fracture patients. The developed nomogram provides a practical tool for preoperative risk stratification, enabling clinicians to optimize perioperative management and improve patient outcomes.