A surveillance of nosocomial candida infections: epidemiology and influences on mortalty in intensive care units

INTRODUCTION: It was aimed to investigate the frequency of Candida infections (CI) in the intensive care units (ICU), to determine typing of candida to evaluate risk factors associated with CI and mortality, and to evaluate influence of CI on mortality. METHODS: The prospective cohort study was carried out between Jan 1, 2009 and Dec 31, 2010 in ICUs, and the patients were observed with active surveillance. VITEK 2 Compact System (BioMerieux, France) kits were used for the identification of isolates from various clinical samples. RESULTS: A total of 2362 patients had enrolled for 16135 patients-days into the study. During the study, 63 (27,5%) of patients developed 77 episodes of CI were observed. Of the patients; 54% were male, 46% were female. Duration of hospitalization (OR = 1,03, p = 0,007), hyperglycemia (OR = 17,93, p = 0,009), and co-infections (OR = 3,98, p = 0,001) were identified as independent risk factors for CI. The most common infections were bloodstream (53%). 77 of 135 candida strains was isolated as causative pathogens. C. albicans (63,6%) was the most frequent species. Overall mortality rate was 78%. The rates of mortality attributable to CI and candidemia were 27%, and 18,3% respectively. Species- specific mortality rates of C.albicans and C.tropicalis were determined as 12%. High APACHE II scores (OR = 1,37; p = 0,002), and the use of central venous catheter (OR = 9,01; p = 0,049) were assigned as independent risk factors for mortality. CONCLUSION: CI is an important problem in our hospital. CI and associated mortality can be prevented by controlling of risk factors. Updating of epidemiological data is required for successful antifungal treatment.