Integration of Surgical and Coagulation Risk Factors for Predicting Pulmonary Embolism in Thoracic Surgery: A Multicenter Study
Introduction:
Pulmonary embolism (PE) is a critical and often preventable complication in thoracic surgery, with an incidence of 1-5%. It poses a significant burden on patients and healthcare systems, with mortality rates as high as 30%. Standard treatment includes anticoagulant therapy and respiratory support, but delayed diagnosis due to nonspecific symptoms can lead to poor outcomes. This study aims to integrate surgical characteristics and coagulation markers to predict postoperative PE risk in thoracic surgery patients.
Methodology:
The study included 977 patients undergoing pulmonary surgery from three centers between 2019 and 2024. Data was collected on demographics, clinical factors, and coagulation parameters. A nomogram was constructed using multivariate logistic regression to identify independent predictors of PE. The model was validated through internal and external validation using bootstrapping and multicenter datasets.
Results:
Seven independent risk factors were identified: advanced age, upper lobe lesion location, open thoracotomy, prolonged surgical duration, increased intraoperative blood loss, elevated postoperative D-dimer, and fibrinogen levels. The nomogram demonstrated excellent discrimination (AUC 0.94-0.97) and good calibration across validation sets. The model identified high-risk patients with PE probabilities >15%, who had older age, more frequent upper lobe involvement, longer operative times, and higher postoperative D-dimer and fibrinogen levels.
Discussion:
This study highlights the importance of integrating surgical and coagulation risk factors for PE prediction in thoracic surgery. D-dimer and fibrinogen emerged as key biomarkers, reflecting hypercoagulability. The model's strong discriminatory power suggests its potential for early risk identification and personalized preventive strategies. Further research is needed to validate and refine the model for clinical application.
Conclusion:
The integrated risk prediction model for postoperative PE in thoracic surgery shows promise for improving patient outcomes. With further validation, it may become a valuable tool for clinical decision-making and personalized care.
