[Colorimetric broth microdilution for antifungal susceptibility testing].

A colorimetric broth microdilution modification of the National Committee for Clinical Laboratory Standards (NCCLS) for antifungal susceptibility testing was developed and evaluated. The test method modification includes; air-dried microdilution trays, in which serial two-fold dilutions of three antifungal agents, amphotericin B (AMPH), flucytosine (5-FC) and fluconazole (FCZ) were first prepared and evaporated, then reconstituted by adding 100 microliters of yeast inocula, and an oxidation-reduction color indicator (sodium resazurin, Sigma) added to RPMI 1640 medium buffered to pH7.0 with 0.165M morpholinepropanesulfonic acid. The trays were incubated in air at 35 degrees C and were inspected after 24 and 48hr of incubation. The MICs were defined as the lowest concentration of the respective agents; no color change (blue) for AMPH, and slight color change (blue to purple) for 5-FC and FCZ. The MICs of AMPH, 5-FC and FCZ were determined for four reference strains and 100 clinical isolates, in comparison with the NCCLS macrodilution method. The four reference strains were tested seven times each by macrodilution method (MACRO) and 14 times each against all three antifungal agents by microdilution method (MICRO). Overall, 100% of MICs determined by MACRO, 96% of those determined at 24hr incubation by MICRO, and 93% of those determined at 48hr incubation by MICRO fell within the 3-log2 dilutions, although 20 to 33% were out of the acceptable MIC ranges of the NCCLS M27-P proposal. Excellent reproducibility in determining growth endpoint by color change was also demonstrated, giving 99 to 100% agreements in duplicated dilutions. When the MICs determined for 100 clinical isolates against three antifungal agents were compared, those determined at 24hr incubation by MICRO gave 61.7% of agreement within the 3-log2 dilutions of the NCCLS MACRO, and those determined at 48hr incubation by MICRO were 86.2%. The MICRO read at 24hr trended to the lower MICs, except for 5-FC. While the MICs of MICRO read at 48hr were mostly comparable (78 to 97% agreement) to those of MACRO, especially against 5-FC (97%) and AMPH (85%), but the discrepant MICs of Candida tropicalis against FCZ were noted. With these results, it can be concluded that the resazurin colorimetric broth microdilution method is easy to perform and highly precise, and may provide MICs comparable to those determined by the reference NCCLS macrodilution method.