Background: The increasing frequency of systemic fungal infections and the emergence of secondary resistance to antifungals in the lasts years, has stimulated the use of methods for antifungal susceptibility testing. Etest(r) is an easily performed quantitative method that has a good agreement with the broth microdilution reference method (NCCLS), if appropriate media are used. Aim: To compare the susceptibility to Amphotericin B (AmB) and Fluconazole (Flu) of 22 opportunistic yeast isolates (C albicans (7), C tropicalis (9), C parapsilosis (3) and Cryptococcus neoformans (3) by Etest ®., using three different media and to choose the best medium for each tested drug. The studied media were RPMI 1640, Casitone (Cas) and Sabouraud. Results: The interpretation of minimal inhibitory concentration (MIC) endpoints on Sabouraud was difficult for AmB. The American Type Culture Collection (ATCC) strains MICs were out of the acceptable range in this medium. RPMI and Cas were suitable media to test AmB and Flu, but best endpoints were obtained for AmB in RPMI and Flu in Cas. Conclusions: The use of appropriate media for each antifungal drug optimizes the MIC readings by Etest(r). AmB should be tested in RMPI and Flu in Cas. Sabouraud must not be used (Rev Méd Chile 2003; 131: 299-302).