Commercial systems for fluconazole susceptibility testing of yeasts: comparison with the broth microdilution method

Fluconazole susceptibility was tested in 100 clinical yeast isolates (65 Ca ndida albicans, 13 C. glabrata, 8 C. tropicalis, 7 C. parapsilosis, 3 Sacch aromyces cerevisiae, 1 each of C. krusei, C. lusitaniae, Cryptococcus neofo rmans, Rhodotorula glutinis) and two control strains (Candida krusei ATCC(R ) 6258, C. parapsilosis ATCC(R) 22019) using broth microdilution (reference method), disk diffusion, Etest strips, Sensititre YeastOne, Candifast, Int egral System Yeasts. Using M27-A breakpoints, isolates were classified as s usceptible (81%), susceptible-dose dependent or Resistant with broth diluti on. Rates of concordance with the reference method were good for Sensititre YeastOne, Etest and disc-diffusion (81.2%-94.7%) but very low for the Cand ifast(R) (3.1%) and Integral System (16.6%), which classified most suscepti ble isolates as resistant. Lack of standardisation (inoculum, medium compos ition) and non-objective interpretation schemes may be the cause of their p oor performance. Sensititre YeastOne, Etest and disc-diffusion are potentia lly useful for fluconazole antifungal susceptibility testing of yeasts in c linical laboratories. (C) 2000 Elsevier Science Inc. All rights reserved.