Medical versus surgical management of Staphylococcus aureus prosthetic valve endocarditis

PURPOSE: The study's purpose was to identify prognostic factors associated with mortality in Staphylococcus aureus prosthetic valve endocarditis and to determine whether these factors influenced decisions to treat medically versus surgically. We also analyzed whether there was a subset of patients who were cured with medical therapy alone. SUBJECTS AND METHODS: A retrospective review of patients with S aureus prosthetic valve endocarditis was performed. Demographic and clinical data were collected from existing medical records. Severity of illness was classified using American Society of Anesthesiologists (ASA) score. Impact of treatment on in-hospital mortality was assessed using multiple logistic regression analysis. RESULTS: Fifty-five patients met the Duke criteria for definite S aureus prosthetic valve endocarditis. Twenty-three patients were treated medically, and 32 patients hail surgical intervention. Overall mortality was 36% (28% in the surgical group and 48% in the medical group). ASA score IV (P < .001) and older age (P = .014) were significant risk factors of mortality. Patients with ASA score IV (P = .037) and multiple prosthetic valves (P = .013) were less likely to undergo surgery. Medically treated patients were older compared with those in the surgical group (median age 66 vs 55 years, P = .04). All 4 patients aged less than 50 years in the medically treated group survived. CONCLUSION: Mortality was generally higher in the medically treated patients with S aureus prosthetic valve endocarditis. Multivariable analysis showed that ASA class IV and bioprosthetic valves were independent predictors of mortality. A subset of medically treated patients characterized by age less than 50 years, ASA score III, and without cardiac, central nervous system, or systemic complications were cured without surgical intervention. (c) 2006 Elsevier Inc. All rights reserved.