The need for a specific risk prediction system in native valve infective endocarditis surgery.

ABSTRACT

The need for a specific risk score system for infective endocarditis (IE) surgery has been previously claimed. In a single-center pilot study, preliminary to future multicentric development and validation, bivariate and multivariate (logistic regression) analysis of early postoperative mortality predictors in 440 native valve IE patients were performed. Mathematical procedures assigned scores to the independent predictors emerged (AUC of the ROC curve 0.88). Overall mortality was 9.1%. Six predictors were identified and assigned scores, including age (5-13 points), renal failure (5), NYHA class IV (9), critical preoperative state (11), lack of preoperative attainment of blood culture negativity (5), perivalvular involvement (5). Four risk classes were drawn ranging from "very low risk" (≤5 points, mean predicted mortality 1%), and to "very high risk" (≥20 points, 43% mortality). IE-specific risk stratification models are both needed, as disease-specific factors (e.g., cultures, abscess), beside the generic ones (e.g., age, renal impairment) affect mortality, and feasible.