Collaborative investigation of variables in susceptibility testing of yeasts.

A multicenter study was performed to evaluate the effect of medium, incubation time (24 and 48 h), and temperature (30 and 35 degrees C) on intra- and interlaboratory variations in MICs of flucytosine, amphotericin B, and ketoconazole for yeasts. Testing was performed on coded isolates of Candida species (11 strains) and Cryptococcus neoformans (2 strains) by using a standard macrodilution protocol 11 laboratories. Four chemically defined media buffered to pH 7.0 with morpholinepropanesulfonic acid were evaluated, including buffered yeast nitrogen base, synthetic amino acid medium-fungal, RPMI 1640 medium, and high-resolution antifungal assay medium. Intralaboratory variability was less than or equal to fourfold for 97% of the replicate sets of data. The highest level of interlaboratory agreement, irrespective of antifungal agent or incubation conditions, was observed with RPMI 1640 medium. Intralaboratory variability was less than or equal to fourfold for 93% of the determinations with ketoconazole and 100% with flucytosine tested in RPMI 1640 medium at 35 degrees C for 24 h. Variability in amphotericin B results was less than or equal to fourfold for 81% of the determinations in RPMI 1640 medium at 35 degrees C for 48 h. The rank order of MICs within each antifungal test group was similar among the various laboratories and was generally in agreement with the reference rank order regardless of the test medium that we used.