Impact of initial empirical antifungal agents on the outcome of critically ill patients with invasive candidiasis: analysis of the China-SCAN study.

Abstract

The effect of different empirical antifungal agents on the clinical outcome of critically ill patients with invasive candidiasis (IC) has not been fully elucidated. In this study, 136 patients with proven IC who received empirical therapy in the China-SCAN multicentre study were retrospectively analysed. Initial empirical antifungal monotherapy consisted of a triazole [fluconazole (n = 61), voriconazole (n = 20) or itraconazole (n = 12)] or an echinocandin (n = 43). Hospital mortality as the primary outcome and global responses (clinical and microbiological) were assessed. The results indicated that rates of hospital mortality (P = 0.006) and intensive care unit (ICU) mortality (P = 0.011) were significantly lower in patients treated with an echinocandin compared with those receiving fluconazole, voriconazole or itraconazole. Multivariate regression analysis indicated that the type of antifungal agent used in empirical therapy was an independent predictor of hospital mortality (P = 0.033). Initial empirical echinocandin treatment was associated with decreased hospital mortality compared with fluconazole [odds ratio (OR) = 0.22, 95% confidence interval (CI) 0.06-0.85; P = 0.028], voriconazole (OR = 0.11, 95% CI 0.02-0.56; P = 0.008) or itraconazole (OR = 0.12, 95% CI 0.02-0.72; P = 0.020). Similar findings were observed for the clinical success endpoint. This study demonstrated that the initial empirical antifungal agent was an independent predictor of hospital mortality in critically ill patients with IC. Empirical therapy with an echinocandin was associated with decreased hospital mortality and greater clinical success than empirical therapy with fluconazole, voriconazole or itraconazole.