A meta-analysis of medical versus surgical therapy for Candida endocarditis

Abstract

Objectives: The optimal management of Candida infective endocarditis (IE) is unknown.

Methods: We reviewed all 879 cases of Candida IE reported from 1966-2002 in the peer-reviewed literature to better understand the role of medical and surgical therapies. This review included 163 patients from 105 reports that met our inclusion criteria: 31 cases treated with antifungal monotherapy, 25 cases treated with medical antifungal combination therapy, and 107 cases treated with adjunctive surgical plus medical antifungal therapy. We also used meta-analytic techniques to evaluate 22 observational case-series (72 patients) of the 105 reports with two or more patients with definite Candida IE.

Results: We found that in patients who underwent adjunctive surgery there was a lower reported proportion of deaths [prevalence odds ratio (POR)=0.56; 95% confidence interval (CI)=0.16, 1.99)]. Higher mortality was noted in patients treated prior to 1980 (POR=2.03; 95% CI=0.55, 7.61), treated with antifungal monotherapy (POR=1.49; 95% CI=0.39, 5.81), infected with Candida parapsilosis (POR=1.51; 95% CI=0.41, 5.52), or with left-sided endocarditis (POR=2.36; 95% CI=0.55, 10.07).

Conclusions: Medical antifungal therapy of Candida IE is poorly characterized, and recent antifungal developments lend promise for those patients who cannot undergo surgery.