Antifungal susceptibility testing and candidaemia at a tertiary referral hospital

Candidaemia is associated with a high mortality. We have reviewed cases of candidaemia over a 2-year period at a tertiary referral hospital in association with the introduction of routine antifungal susceptibility testing. The aim of the study was two fold; firstly to establish the typical profile of a patient who might experience a Candida bloodstream infection and secondly, to evaluate methods of antifungal susceptibility testing. In 2008-2009, 31 patients with candidaemia were retrospectively identified using the Laboratory Information Systems (Apex). Clinical data were obtained by chart review. Antifungal susceptibility testing to fluconazole and voriconazole was carried out on 20 of the clinical isolates using three different methods. These isolates were also sent to the mycology reference laboratory at Bristol and results were compared. The male-to-female ratio was 2.1:1 with an age range from 6 weeks to 89 years. Candida albicans was the predominant species (n= 17). Patients were predominantly general surgical (39%), oncology (16%) and urology (13%). Identified risk factors included treatment with broad-spectrum antimicrobial agents (89%), central venous catheters (CVCs) (89%), and surgery during the current admission (54%). The crude mortality rate (death prior to discharge) was 42%. Only 1 of the 20 isolates tested, a Candida glabrata, tested resistant to fluconazole. Of 3 antifungal susceptibility test systems evaluated (VITEK 2, TREK Sensititre YeastOne and CLSI disk diffusion); the VITEK 2 system was considered most appropriate for routine use in our laboratory. Retrospective review of therapy identified 7 patients treated with echinocandins in whom susceptibility testing indicated that fluconazole could have been used with significant reduction in cost of therapy.