Characterizing Acute Pulmonary Embolism After Off-Pump Coronary Artery Bypass Surgery Using a Predictive XGBoost...

Area of Science:

• Cardiovascular Surgery
• Pulmonary Medicine
• Medical Informatics

Background:

• Acute pulmonary embolism (APE) is a potential complication following off-pump coronary artery bypass grafting (OPCABG).
• Characterizing APE occurrence and developing predictive tools are crucial for improving patient prognosis.

Purpose of the Study:

• To characterize the incidence and clinical features of APE post-OPCABG.
• To develop and validate a predictive model for APE after OPCABG.

Main Methods:

• Retrospective review of 15,259 OPCABG patients.
• Statistical analysis of clinical features and development of a predictive model using LASSO and XGBoost algorithms.
• SHAP analysis to identify key prognostic factors.

Main Results:

• APE occurred in 1.49% of patients, with common symptoms including hypoxia and shortness of breath.
• The XGBoost model demonstrated high performance with 97.6% AUC, identifying postoperative D-dimer and deep vein thrombosis as significant predictors.
• Key predictors included postoperative D-dimer, deep vein thrombosis, low molecular weight heparin timing, and preoperative uric acid.

Conclusions:

• A robust predictive model for APE following OPCABG was developed.
• This model can aid in identifying at-risk patients, optimizing management, and potentially reducing healthcare costs.