Bingbing Xiang,1 Jingyuan Zhang,2 Chaoyi Deng,1 Han Yang,1 Liu Qian,1 Wensheng Zhang1,3 

1Department of Anesthesiology, West China Hospital, Sichuan University, Chengdu, 610041, People’s Republic of China; 2Department of Anesthesiology, The Second Affiliated Hospital of Kunming Medical University, Kunming, 650000, People’s Republic of China; 3Laboratory of Anesthesia and Critical Care Medicine, National-Local Joint Engineering Research Centre of Translational Medicine of Anesthesiology, West China Hospital, Sichuan University, Chengdu, 610041, People’s Republic of China

Correspondence: Wensheng Zhang, Department of Anesthesiology, West China Hospital, Sichuan University, Chengdu, 610041, People’s Republic of China, Email zhang_ws@scu.edu.cn

Background: Postoperative pneumonia is one of the most common complications following hip arthroplasty in older adults. It often results in delayed recovery, prolonged hospital stays, and increased perioperative mortality rates.
Objective: To analyze the risk factors for postoperative pneumonia in older adults undergoing hip arthroplasty and develop a nomogram-based prediction model using perioperative variables.
Methods: A retrospective analysis was performed on 308 older adults who underwent hip arthroplasty. Relevant clinical data were collected and recorded. Univariate and multivariate logistic stepwise regression analyses were conducted to identify the risk factors for postoperative pneumonia in this population. A risk prediction model for postoperative pneumonia was then developed and visualized using a nomogram.
Results: Among the 308 older adults, 46 developed postoperative pneumonia, with an incidence rate of approximately 14.94%. Multivariate logistic regression analysis revealed that American Society of Anesthesiologists (ASA) classification, intensive care unit (ICU) admission, preoperative anemia, creatine kinase-MB (CKMB), brain natriuretic peptide (BNP), and postoperative aspartate aminotransferase (AST) were independent risk factors for postoperative pneumonia in elderly patients (P< 0.05). The final prediction model for postoperative pneumonia was: P = 1 / [1 + e^(− 3.690 + 0.982×ASA + 0.982×ICU + 0.806×Preoperative Anemia + 1.494×CKMB + 0.843×BNP + 0.917×Postoperative AST)], with Hosmer-Lemeshow χ² = 5.989 (P = 0.541). Receiver operating characteristic curve analysis showed an area under the curve of 0.792 (95% CI: 0.761– 0.823). The Brier score of the calibration curve was 0.103 (close to 0), and decision curve analysis indicated that the threshold probability of the model ranged from 0.01 to 0.8, with net benefits greater than 0 across all probabilities, suggesting the model has good accuracy and clinical utility.
Conclusion: We identified six important predictors—ASA grade, ICU admission, preoperative anemia, CKMB, BNP, and postoperative AST levels—and developed a risk prediction model for postoperative pneumonia following hip arthroplasty in older adults, providing a valuable reference for its prevention in this population.

Keywords: arthroplasty, replacement, hip, pneumonia, risk factors, prediction algorithms, aged