Analysis of Risk Scores to Predict Mortality in Patients Undergoing Cardiac Surgery for Endocarditis

Area of Science:

• Cardiology
• Cardiac Surgery
• Medical Statistics

Background:

• Clinical practice faces uncertainty in selecting appropriate risk scores.
• Accurate prediction of mortality is crucial for patients undergoing cardiac surgery for active infective endocarditis (IE).

Purpose of the Study:

• To evaluate the predictive performance of logistic EuroSCORE, EuroSCORE II, and several IE-specific scores (STS-IE, PALSUSE, AEPEI, EndoSCORE, RISK-E).
• To identify the optimal risk score for hospital mortality prediction in patients with active IE undergoing cardiac surgery in Southern Brazil.

Main Methods:

• Retrospective cohort study of 107 adult patients undergoing cardiac surgery for active IE (2007-2016).
• Risk scores assessed using calibration (observed/expected mortality ratio) and discrimination (area under the ROC curve [AUC]).
• Statistical significance determined by DeLong test (p < 0.05).

Main Results:

• Overall hospital mortality was 29.0%.
• PALSUSE score showed the best calibration (O/E ratio: 1.01).
• Logistic EuroSCORE exhibited the highest discrimination (AUC: 0.77), significantly outperforming other scores including EuroSCORE II and IE-specific scores.

Conclusions:

• The logistic EuroSCORE appears to be the most effective predictor of mortality risk in this cohort, balancing calibration and discrimination.
• There is a need for local validation of IE-specific scores to enhance preoperative surgical risk assessment.