Bleeding risk stratification in coronary artery surgery: the should-not-bleed score

Area of Science:

• Cardiovascular Surgery
• Transfusion Medicine
• Medical Informatics

Background:

• Allogeneic blood transfusions are common in cardiac surgery, with significant variability in rates for coronary artery bypass grafting (CABG).
• An estimated 11% of red cell resources are used for CABG patients, highlighting the need to optimize transfusion practices.
• Unnecessary transfusions contribute to clinical and economic burdens, necessitating better patient risk stratification.

Purpose of the Study:

• To develop a user-friendly application for stratifying patients based on bleeding risk.
• To identify a subgroup of CABG patients at low risk of bleeding for whom transfusions may be unnecessary.
• To present the "SHOULD-NOT-BLEED-SCORE" application for Windows-based platforms.

Main Methods:

• Developed a risk prediction model based on multivariate analysis of 1426 elective CABG patients (January 2010 - January 2018).
• Created a Windows application incorporating key variables: Age, BMI, Chronic Renal Failure, Preoperative Clopidogrel Exposure, RBC Count, Fibrinogen Level, and ASPI test AUC.
• Validated the model using Receiver Operating Curve (ROC) analysis.

Main Results:

• The SHOULD-NOT-BLEED-SCORE application was developed on the Windows platform.
• The scoring system includes variables such as age, BMI, renal failure, and preoperative lab values.
• The score demonstrated strong discriminatory performance in predicting low bleeding risk (ROC AUC 72.3%, p < 0.001).

Conclusions:

• The SHOULD-NOT-BLEED risk scoring application can aid in preoperative risk screening.
• Implementing this tool may help mitigate the clinical and economic impact of unnecessary transfusions.
• Identifying low-bleeding-risk patients allows for more judicious transfusion decisions in CABG procedures.