DEFINING CONDITIONS FOR MICROBROTH ANTIFUNGAL SUSCEPTIBILITY TESTS - INFLUENCE OF RPMI AND RPMI-2-PERCENT GLUCOSE ON THE SELECTION OF END-POINT CRITERIA

We have compared amphotericin B, flucytosine, ketoconazole and flucona zole susceptibilities of 40 clinical isolates of Candida albicans by b roth microdilution in two different media: RPMI 1640 (RPMI) and the sa me medium supplemented with 18 g of glucose per L (RPMI-2% glucose). P reparation of media, drugs and inocula, as well as incubation temperat ure, followed the recommendations of the National Committee for Clinic al Laboratory Standards (Villanova, PA, USA) antifungal agent working group for broth macrodilution tests with antifungal agents. Microtitre plates were agitated for 5 min before spectrophotometric readings wer e performed with an automatic plate reader. MIC endpoints were defined in three different ways: (i) MIC-100% for amphotericin B and flucytos ine; (ii) MIC-80% for azole-drugs and also for flucytosine; (iii) IC1/ 2 for atole-drugs. The MIC endpoints were compared between the two dif ferent media in order to ascertain which was the best criterion. For a mphotericin B, 100% of strains had a maximum difference of a twofold d ilution in both media. For flucytosine, MIC values were very similar i n both media, irrespective of the MIC endpoint chosen, MIC-100% or MIC -80%. For atole-drugs the (MIC-80%)(50) and (MIC-80%)(90) in RPMI were higher than those in RPMI-2% glucose. However, (IC1/2)(50) and (IG(1/ 2))(90) were identical in both media as well as (MIC-80%)(50) and (MIC -80%)(90) in RPMI-2% glucose. The limited growth of yeasts in RPMI pre cludes the selection of an atole-MIC-80% endpoint, although the MIC de termination was performed with an objective turbidimetric method (spec trophotometric reading plus mathematical MIC calculation). The use of RPMI-2% glucose produces the same MIC whatever method was used, IC1/2 or MIC-80%. However, some minor discrepancies can be expected between IC1/2 and MIC-80% when strains with higher ''trailing effect'' are bei ng tested. Therefore, we recommended IC1/2 in RPMI-2% glucose as the m ethod of choice for MIC calculation, until more studies correlating in -vitro results with clinical outcome have been performed.