Comparison of the sensititre yeastone and CLSI M38-A2 microdilution methods in determining the activity of amphotericin B, itraconazole, voriconazole, and posaconazole against Aspergillus species

[[abstract]]This study compared the YeastOne and reference CLSI M38-A2 broth microdilution methods for antifungal susceptibility testing of Aspergillus species. The minimum inhibitory concentrations (MICs) of antifungal agents were determined for 100 Aspergillus isolates, including 54 A. fumigatus (24 TR34/L98H isolates), 23 A. flavus, 13 A. terreus, and 10 A. niger isolates. The overall agreement (within two two-fold dilutions) between the two methods was 100%, 95%, 92%, and 90% for voriconazole, posaconazole, itraconazole, and amphotericin B, respectively. The voriconazole geometric mean (GM) MICs were nearly identical for all isolates using both methods, whereas the itraconazole and posaconazole GM MICs obtained using the YeastOne method were approximately one dilution lower than those obtained using the reference method. In contrast, the amphotericin B GM MIC obtained using the YeastOne method was 3.3-fold higher than that observed using the reference method. Of the 24 TR34/L98H A. fumigatus isolates assayed, the categorical agreement (classified according to the CLSI epidemiological cutoff values) was 100%, 87.5%, and 83.3% for itraconazole, voriconazole, and posaconazole, respectively. For four A. niger isolates, the itraconazole MICs were > 8 mug/mL using the M38-A2 method due to trailing growth, whereas the corresponding itraconazole MICs obtained using the YeastOne method were all </= 0.25 mug/mL without trailing growth. These data suggested that the YeastOne method can be used as an alternative for azole susceptibility testing of Aspergillus species and for detecting TR34/L98H A. fumigatus but that this method fails to detect A. niger isolates exhibiting trailing growth with itraconazole. Additionally, for isolates with azole MICs approaching or at susceptibility breakpoints or with high amphotericin B MICs detected using the YeastOne method, further MIC confirmation using the reference CLSI method is needed