Correlation between the E test and the CLSI M-38 A microdilution method to determine the activity of amphotericin B, voriconazole, and itraconazole against clinical isolates of Aspergillus fumigatus.

The in vitro activities of amphotericin B, itraconazole, and voriconazole against 283 clinical isolates of Aspergillus fumigatus were studied by comparing the E test method with the reference procedure NCCLS (CLSI) M-38A. The methods were considered to agree when the results of the MICs by E test were within +/-2 dilutions of the MICs obtained by CLSI M-38 A. Agreement of readings at 24, 48, and 72 h of incubation was as follows: amphotericin B: 4.2%, 98%, and 40.6%; itraconazole: 92.5%, 100%, and 89.9%; and voriconazole: 98.9%, 100%, and 99.7%. The correlation between methods to classify the strains as "susceptible" or "resistant" was very good for itraconazole and voriconazole after 48 h of incubation, but 23% of the strains were incorrectly classified by E test for amphotericin B. The E test is an efficacious method for antifungal susceptibility testing in A. fumigatus for itraconazole and voriconazole when the plates are read after 48 h of incubation. The use of the E test to study the activity of amphotericin B should be avoided.