A prognostic nomogram integrating CTPA with clinical and hematological parameters for pulmonary embolism

Abstract

Purpose

Pulmonary embolism (PE) is a life-threatening condition requiring precise risk stratification for optimal management. This study aimed to develop and validate a prognostic nomogram that incorporated computed tomography pulmonary angiography (CTPA), blood biomarkers, and clinical characteristics to predict adverse outcomes in PE patients.

Methods

A retrospective cohort study was conducted on patients diagnosed with PE from October 2017 to October 2024. Patients were divided into two groups based on adverse outcomes within six months: good prognosis and poor prognosis. CTPA measurements included the right ventricular to left ventricular ratio (RV/LV), pulmonary artery diameter, azygos vein diameter, coronary sinus diameter, and Qanadli score. Hematological parameters analyzed consisted of a comprehensive panel of biomarkers, including D-dimer, troponin I, and inflammatory markers like HsCRP and immature granulocyte count (IG). Clinical assessments recorded the presence of chest pain and measured pulmonary artery systolic pressure.

Results

The cohort included 513 patients (410 with good prognosis and 103 with poor prognosis). Multivariate analysis identified five independent predictors of poor prognosis: chest pain (OR = 1.962, p = 0.024), RV/LV ratio (OR = 4.920, p = 0.001), troponin I (OR = 178.133, p < 0.001), HsCRP (OR = 1.015, p = 0.044), and IG (OR = 10.966, p = 0.038). The nomogram demonstrated excellent discrimination (AUC = 0.809) and calibration. External validation (n = 102; 82 with good prognosis and 20 with poor prognosis) confirmed its robustness (AUC = 0.814).

Conclusion

The novel nomogram that integrated CTPA, hematological, and clinical parameters offered a clinically applicable tool for accurately stratifying risk. This may help guide decisions regarding patient disposition, monitoring intensity, and the use of advanced therapies in high-risk PE patients.