Standard conditions are not available for evaluating the minimum fungicidal concentrations (MFCs) of antifungal agents. This multicenter collaborative study investigated the reproducibility in three laboratories of itraconazole, posaconazole, ravuconazole, voriconazole, and amphotericin B MFCs for 15 selected isolates of Aspergillus spp. After MIC determinations for the 15 isolates in each center by the NCCLS M38-A broth microdilution method with four media, standard RPMI 1640 (RPMI), RPMI with 2% dextrose, antibiotic medium 3 (M3), and M3 with 2% dextrose, MFCs were determined for each isolate-medium-drug combination. MFCs were defined as the lowest drug dilutions that yielded <3 colonies (approximately 99 to 99.5% killing activity). The highest reproducibility (96 to 100%) was for amphotericin B MFCs with the four media. Although reproducibility was more variable and medium dependent for the azoles (91 to 98%), agreement was good to excellent for itraconazole, ravuconazole, and voriconazole MFCs with RPMI and M3 (93 to 98%). For posaconazole, the agreement was higher with M3 media (91 to 96%) than with RPMI media (91%). These data extend the refinement of testing guidelines for susceptibility testing of Aspergillus spp. and warrant consideration for introduction into future versions of the M38 document. The role of the MFC under these standardized testing conditions as a predictor of clinical outcome needs to be established in clinical trials.Aspergillus fumigatus and other Aspergillus spp. are responsible for the majority (85 to 90%) of clinical manifestations of severe infections caused by the filamentous fungi (moulds), especially in the immunocompromised host (4). The increased incidence of fungal infections and the development of new antifungal agents have underscored the importance of the laboratory's role in the selection and monitoring of antifungal therapy. The National Committee for Clinical Laboratory Standards (NCCLS) Subcommittee on Antifungal Susceptibility Tests has developed a reproducible reference testing procedure for the antifungal susceptibility testing of moulds (the M38-A document [18]). The recommendations described in the M38-A document for determination of MICs include the use of the standard RPMI 1640 broth (RPMI), which contains 0.2% dextrose (18). However, the document does not describe testing conditions for determination of minimum fungicidal (or lethal) concentrations (MFCs). Whether MICs are the best in vitro predictors of in vivo or clinical response to antifungal therapy is uncertain. Although standard conditions are not available for determination of fungicidal activities for either yeasts or moulds, it has been demonstrated that MFCs may be better predictors than MICs of therapeutic failure of amphotericin B in trichosporonosis (26, 27) and candidemia (20). The fungicidal activities of the new triazoles have also been evaluated during the last few years by nonstandardized methods (3,7,11,13,14,16,17,21,24,25).The purpose of this collaborative study was to investigate the in...