RESUMO
INTRODUCTION: The increasing clinical and microbiologic resistance of Candida spp. isolates to several antifungal agents is becoming a serious problem. It is now reasonable to propose the use of antifungal susceptibility testing in Candida spp. isolates from patients who have failed conventional therapy, before the selection of an empirical therapy. METHODS: One hundred and fifty eight isolates of Candida spp. were evaluated simultaneously by broth microdilution (NCCLS standard) and well diffusion testing (WD), a diffusion method similar to disc diffusion. RESULTS: According to the Wilcoxon Signed Ranks test performed, there was no significant difference (p>0.05) between both methodologies for all antifungal agents tested (fluconazole, itraconazole, posaconazole, caspofungin and amphotericin B, with C. tropicalis, C. krusei, C. dubliniensis, C. guillermondii, C. parapsilosis, C. albicans and C. glabrata). A significant difference was observed when comparing well diffusion with NCCLS for fluconazole WD 80% (p=0.008) in C. glabrata, as well as WD 80% (p=0.002) and WD 50% (p=0.002) in C. albicans. CONCLUSIONS: The well diffusion test is simple, easy to reproduce, inexpensive, easy both to read and interpret, and has a good correlation to the reference NCCLS microdilution test and may represent an alternative method for antifungal drug susceptibility testing of Candida spp., mainly in laboratories with few resources.