Construction of a Risk Prediction Model for the Development of Pulmonary Infection in Elderly Patients with Atrial Fibrillation Combined with Sarcopenia

doi:10.3969/j.issn.1009-0126.2025.01.023

Abstract：Objective: To explore the risk factors for pulmonary infection in elderly patients with atrial fibrillation combined with sarcopenia, and to construct a risk prediction model. Methods: A retrospective study was conducted on 461 elderly patients with atrial fibrillation combined with sarcopenia admitted to Changan District Hospital in Beijing from January 2023 to May 2024. The patients were divided into an infected group (n=230) and a non-infected group (n=231) based on whether they had pulmonary infection. Univariate and multivariate logistic regression analyses were used to identify the risk factors for pulmonary infection in these patients, and a risk prediction model for pulmonary infection was established. The predictive value of the risk prediction model was analyzed using ROC curves. Results: The infected group showed significantly higher age, coronary heart disease, anemia, SARC-F score, serum creatinine, and paroxysmal atrial fibrillation ratio compared to the non-infected group, while the ADL score and albumin (ALB) were significantly lower in the infected group (P<0.05, P<0.01). Univariate logistic regression analysis showed that age, ADL score, SARC-F score, ALB, serum creatinine, coronary heart disease, and anemia were significant factors influencing pulmonary infection in elderly patients with atrial fibrillation combined with sarcopenia (P<0.05, P<0.01). Multivariate logistic regression analysis showed that ADL score, SARC-F score, and ALB were independent risk factors for pulmonary infection (OR=1.477, 95%CI: 1.239–1.760, P=0.000; OR=0.989, 95%CI: 0.982–0.995, P=0.001; OR=0.933, 95%CI: 0.896–0.973, P=0.001). ROC curve analysis showed that the combination of ADL score, SARC-F score, and ALB had an area under the curve (AUC) of 0.829 for predicting pulmonary infection in elderly patients with atrial fibrillation combined with sarcopenia, with an optimal cutoff value of 0.681, sensitivity of 87.2%, and specificity of 80.9%. Conclusion: The risk prediction model constructed by combining ADL score, SARC-F score, and ALB has good predictive value for pulmonary infection in elderly patients with atrial fibrillation combined with sarcopenia.