Risk of Transfusion in Isolated Coronary Artery Bypass Graft: Models Developed From The Society of Thoracic Surgeons Database

This study utilized data from 1,266,545 consecutive isolated coronary artery bypass graft (CABG) operations (2008-2016) in The Society of Thoracic Surgeons Adult Cardiac Surgery Database to develop risk assessment tools to predict the likelihood of perioperative transfusions and the expected volume of blood products required.  The study found that 657,821 patients (51.9 percent) received transfusions of red blood cells (RBC), fresh frozen plasma (FFP), cryoprecipitate, and/or platelets. The researchers established a rapid risk score based on five key variables: age, body surface area, sex, preoperative hematocrit, and the use of an intra-aortic balloon pump.  The full model showed C-statistics indicating good predictive performance: 0.785 for any blood product, 0.815 for RBC, 0.707 for FFP, and 0.699 for platelets. The rapid risk assessments also demonstrated effective predictive capability, with C-statistics of 0.752, 0.785, 0.670, and 0.661, respectively. Calibration plots indicated strong alignment between observed and expected transfusion risks, with discrepancies of less than 10.8 percent across different risk percentiles.  Overall, the study concluded that these well-calibrated risk assessment tools can guide surgeons in understanding the risks associated with transfusions and the anticipated need for blood products, ultimately helping to optimize patient outcomes and improve resource management in clinical settings.