O7.5 Is preemptive therapy of candidemia useful? An eight years experience

Abstract

Listen

Translate article

Rationale: The candidemia preemptive therapy according to colonization index remains controversial. We report our experience using either fluconazole (susceptible Candida species) or caspofungin (fluconazole-resistant species, since 2006) in a adult Burns Intensive Care unit from 2003 to 2010. Methods: Using prospectively collected data since 2003 (S period, 286 patients) compared to those of 2001–2002 (A period, 77 patients), we determined variation in incidence of candidemia, circumstances of candidemia, attributable mortality and consequences on overall mortality, Candida species evolution. Colonization index was determined according to weekly samples from at least four sites. A cost evaluation approach was undertaken (treatment and biological tests) for the last four years. Results: The incidence of candidemia was significantly lower during S period (1.7% versus 7.8%, Fisher’s test p=0.0142). Candidemic patients had a 37% mortality. Candidemia always occurred in protocol misuse circumstances. Non candidemic patients had a 20% mortality, as overall mortality. Despite a 17% attributable mortality for candidemia, sample size was too small to show a significant mortality difference (p=0.2924). Obviously, there is no difference in overall mortality. There was no significant variation in candida species during the S period. Candida albicans remained in 2010 at his level of 2002 (70%) despite the use of fluconazole and caspofungin as indicated. During the years 2007–2010 forty two per cent of patients were treated. The additional cost spread over the entire population was approximately 30 per patient and per day. Conclusion: Preemptive therapy of candidemia appeared effective in reducing candidemia. It did not reduce significantly overall mortality. It did not induce selection of resistant Candida species. The extra cost effectiveness must be estimated according to theses results.