Epidemiology and Prognostic Factors of Non-albicans Candida species Candidemia: A Multicenter Study

BACKGROUND: The incidence of Non-albicans Candida (NAC) fungemia has increased over the past decades with high mortality rates. However, the epidemiology and prognostic factors have seldom been investigated between species of NAC. METHODS: Patients with NAC fungemia between 2011 and 2014 from five tertiary hospitals in Taiwan were enrolled. The epidemiology data and factors associated with mortality including antifungal agents were collected by a standardized case-record form. A multivariate regression model was applied to analysis the factors associated with mortality. RESULTS: In total, 611 non-duplicated patients were enrolled. Candida tropicalis (n = 245, 42.3%) was most common followed by Candida glabrata (n = 213, 34.9%), Candida parapsilosis (n = 106, 17.3%) and others (n = 47, 7.7%). The overall 30-day mortality of all NAC candidemia was 47.7%. C. tropicalis infection had higher 30-day mortality (54.6%) than C. glabrata (42.8%) and C. parapsilosis (36.8%) (P < 0.05). In general, Charlson Comorbidity Index (CCI), liver cirrhosis, double lumen use, and recent steroid exposure predicted a poor prognosis. Instead, central line infection was a protective factor (OR 0.42; 95% CI 0.24–0.71; P = 0.001) because removal of central line was a most effective method for infection source control. In individual species of NAC, patients with C. parapsilosis infection took advantage from favorable host factors including younger age, lower CCI, fewer steroid exposure and more from central line infection than other two species. On the other hand, though the host factors were similar between C. glabrata and C. tropicalis infection, patients with C. grabrata infection took benefit from more echinocandin or high dose fluconazole (≥10 mg/kg/day) use, which was associated lower mortality than those with usual dose fluconazole (6–10 mg/kg/day). However, the echinocandin or high dose fluconazole did not improved outcome of C. tropicalisi infection. CONCLUSION: The epidemiology and prognostic factors were different among NAC species. Risk assessment and therapeutic strategy should be individualized according to species when facing the rising threat of NAC infection. DISCLOSURES: All authors: No reported disclosures.