Construction and verification of risk prediction model of pulmonary embolism in ICU patients with COPD in acute exacerbation based on age, SAPSII score, braking state and mechanical ventilation

Hong Li Qian Ning Ya Liu Yamei Pang Sifang Feng ^*

• Department of Respiratory and Critical Care Medicine, First Affiliated Hospital of Xi’an Jiaotong University, Xi'an, Shaanxi, China

Objective: To assess the risk of pulmonary embolism (PE) in ICU patients with acute exacerbation of COPD, using age, SAPS II score, immobilization status, and mechanical ventilation as factors, and to develop a PE risk prediction model.: A retrospective study of 220 ICU patients with acute COPD exacerbation between March 2017 and March 2024. Patients were categorized into PE-positive and PE-negative groups. A logistic regression model was constructed based on clinical characteristics to identify risk factors for PE. Results: Among 220 patients, 50 developed PE. Logistic regression identified age, SAPS II score, immobilization of ≥7 days, and invasive mechanical ventilation as significant predictors of PE (P < 0.05). The multifactorial prediction model had an AUC of 0.829 (95% CI: 0.744-0.914), with sensitivity of 77.81% and specificity of 70.63%. Conclusions: A PE prediction model based on age, SAPS II score, immobilization of ≥7 days, and mechanical ventilation was developed. This model effectively identifies high-risk patients and aids in early intervention for PE in ICU patients with acute COPD exacerbation.