Session: Poster Session: Clinical Mycology I
Monday, October 27, 2008: 12:00 AM
Room: Hall C
Background: Candida peritonitis is frequent but information on Candida spp., their in vitro susceptibility, antifungal strategies and patient (pt) outcome is still scarce. Methods: AmarCand is a prospective, multicenter study in 271 adult ICU pts with proven invasive Candida infection who received systemic antifungal therapy (2005-6). Of them, 93 pts (54 ICUs) had peritonitis ± candidemia. Results: Of the peritonitis pts (50 M, 63 ± 12 years, SAPS II 50 ± 16, SOFA score 10 ± 4), 73 had nosocomial infection, 53 had concomitant bacterial infection, and 26 had candidemia. Candida species were C. albicans (58% of 108 isolates), C. glabrata (20%), C. krusei (8%), C. kefyr (5%), C. parapsilosis (3%), C. tropicalis (3%), or another species (3%). 27% of isolates were fluconazole-R/S-DD (C. albicans 9%, C. glabrata 64%, C. krusei 100%). Empiric antifungal treatment was started 1 day (median) after diagnosis, with fluconazole (77% of pts), caspofungin (13%), voriconazole (3%), any amphotericin B formulation (2%), or a combination (4%). Three days (median) after susceptibility results, treatment was modified in 25 pts (susceptibility results 64% and/or clinical reasons 56%). Most commonly, fluconazole was replaced by caspofungin (n=9) or voriconazole (n=3). The case fatality ratio in ICU was 38% and was similar taking into account the species (C. albicans 35%, C. glabrata 32%), fluconazole susceptibility, time to treatment start, candidemia, nosocomial / community acquired infection, or concomitant bacterial infection. Conclusions: A high proportion of fluconazole R/S-DD strains was cultured leading to discuss the first line treatment of Candida peritonitis. Mortality was high despite early treatment and no single factor was associated with death. The results confirm that Candida peritonitis is a complex issue. Improvement of its prognosis remains a challenge.