Background: Decision making regarding surgical intervention in native valve endocarditis (NVE) is often complex and surgery is withheld in a number of patients either because medical treatment is considered the best treatment or because the risk of operation is considered too high. The objective of this study was to investigate the outcome of surgical treatment and to validate the ability of euroSCORE to predict operative mortality in NVE patients.
Methods: Prospective cohort study including 323 consecutive NVE patients. Patients were divided into 3 groups based on treatment strategy and indication/contraindication for surgery. The additive and logistic euroSCORE was calculated and the observed and predicted mortality was compared.
Results: Cardiac surgery was associated with a good prognosis, in-hospital and after 12months, compared to conservative treatment. After adjustment for confounders surgery was associated with a survival benefit (hazard ratio (HR) 0.45, 95% CI: 0.27-0.76%; p=0.003). When propensity score was used in regression adjustment, cardiac surgery was still associated with a better outcome after 12months (HR 0.41, 95% CI: 0.25-0.68; p<0.001). Observed mortality for patients receiving surgical treatment was 11% compared to a mean logistic euroSCORE mortality of 16% (NS). The discriminating ability of euroSCORE was good, area under the ROC curve 0.74 (95% CI: 0.64-0.84; p<0.001) logistic model and 0.75 (95% CI: 0.65-0.86; p<0.001) additive model.
Conclusions: Cardiac surgery was associated with a good prognosis when indicated regardless of euroSCORE, and surgery should only be withheld after thorough consideration. EuroSCORE remains a valuable tool to identify high-risk IE patients when surgery is considered.
Copyright © 2010 Elsevier Ireland Ltd. All rights reserved.