Background: To evaluate a new semi-automatized system (ATB-Fungus, Bio Mérieux, Spain) for antifungal in vitro susceptibility testing in yeast.
Methods: We studied the susceptibility of 91 yeast strains (22 Candida albicans, 18 Candida parapsilopsis, 18 Candida krusei, 17 Candida tropicalis, 11 Candida glabrata, 5 Candida guillermondii) against amphotericin B, flucytosine, ketoconazole, miconazole, econazole and nystatin, by means of "ATB-fungus" and agar diffusion and only against systemic usage antifungal agents by agar-dilution method.
Results: The correlation between "ATB-fungus" and diffusion method and agar dilution was 100% for amphotericin B, except for Candida krusei and C. tropicalis strains. This correlation rate was lower for fluocytosine and ketoconazole, without major differences between species. There is an absolute concordance between the system under evaluation and agar diffusion method regarding nystatin. For the two azole drugs remaining (miconazole and econazole) the correlation found for Candida glabrata was 100% and for C. parapsilopsis 95% with agar diffusion method. Candida glabrata was the single species that showed better correlation with all three methods, while major differences were seen with C. krusei and C. tropicalis.
Conclusions: Although "ATB-fungus" seems to be a nearly well standardized method for antifungal susceptibility testing, more studies are required for assessing its reproducibility and also to determine when, how and for which antifungal agents would be a valid procedure.