Logistic or additive EuroSCORE for high-risk patients?

Area of Science:

• Cardiovascular Surgery
• Medical Statistics
• Risk Stratification

Background:

• Cardiac surgery involves inherent risks, necessitating accurate patient risk assessment.
• The European System for Cardiac Operative Risk Evaluation (EuroSCORE) is widely used for this purpose.
• Two versions exist: the additive EuroSCORE and the full logistic EuroSCORE.

Purpose of the Study:

• To compare the predictive accuracy of the full logistic EuroSCORE against the standard additive EuroSCORE.
• To determine superiority in predicting mortality for high-risk cardiac surgical patients.

Main Methods:

• Applied both additive and logistic EuroSCORE models to 14,799 European cardiac surgery patients.
• Focused analysis on 4,293 high-risk patients (additive EuroSCORE ≥ 6).
• Evaluated absolute prediction accuracy and discrimination using the area under the ROC curve.

Main Results:

• Actual mortality was 4.72%. The logistic model predicted 4.84%, while the additive model predicted 4.21%.
• In high-risk patients (actual mortality 11.18%), the logistic model predicted 11.23%, significantly closer than the additive model's 7.83%.
• Discrimination was comparable: Area under ROC curve was 0.783 for additive and 0.785 for logistic.

Conclusions:

• The additive EuroSCORE remains a valuable bedside tool for European cardiac surgery risk assessment.
• The logistic EuroSCORE demonstrates superior risk prediction, particularly for high-risk patients.
• The logistic model is recommended for institutions focused on risk stratification research and development.