A precise blood transfusion evaluation model for aortic surgery: a single-center retrospective study

Area of Science:

• Cardiovascular Surgery
• Medical Informatics
• Hematology

Background:

• Cardiac aortic surgery frequently necessitates substantial intraoperative blood transfusions.
• Accurate prediction of blood transfusion volume before surgery remains a clinical challenge.

Purpose of the Study:

• To develop a precise machine learning model for predicting intraoperative red blood cell transfusion volume in cardiac aortic surgery.
• To identify significant risk factors influencing transfusion requirements.

Main Methods:

• Retrospective analysis of 3,654 patients undergoing cardiac aortic surgery.
• Development and evaluation of 13 machine learning models, focusing on decision tree algorithms.
• Utilized interpretability technology to analyze risk factor influence.

Main Results:

• An extremely randomized forest model demonstrated the highest accuracy, with a mean absolute error of 1.17 U and R² of 0.50.
• Identified six key risk factors: erythrocytes, operation duration, body weight, sex, red blood cell count, and D-dimer.
• Established protective thresholds for prediction based on these factors.

Conclusions:

• The novel machine learning model effectively aids clinicians in calculating precise blood transfusion volumes.
• Optimizes blood resource utilization and enables personalized intraoperative blood management.
• Provides critical insights into risk factors for timely clinical interventions.