Influence of test conditions on antifungal time-kill curve results: proposal for standardized methods

This study was designed to examine the effects of antifungal carryover, agitation, and starting inoculum on the results of time-kill tests conducted with various Candida species. Two isolates each of Candida albicans, Candida tropicalis, and Candida glabrata were utilized. Test antifungal agents included fluconazole, amphotericin B, and LY303366. Time-kill tests were conducted in RPMI 1640 medium buffered with morpholinepropanesulfonic acid (MOPS) to a pH of 7.0 and incubated at 35 degrees C. Prior to testing, the existence of antifungal carryover was evaluated at antifungal concentrations ranging from 1x to 16x MIC by four plating methods: direct plating of 10, 30, and 100 microl of test suspension and filtration of 30 microl of test suspension through a 0.45-microm-pore-size filter. Time-kill curves were performed with each isolate at drug concentrations equal to 2 x MIC, using a starting inoculum of approximately 10(5) CFU/ml, and incubated with or without agitation. Last, inoculum experiments were conducted over three ranges of starting inocula: 5 x 10(2) to 1 x 10(4), >1 x 10(4) to 1 x 10(6), and >1 x 10(6) to 1 x 10(8) CFU/ml. Significant antifungal carryover (>25% reduction in CFU/milliliter from the control value) was observed with amphotericin B and fluconazole; however, carryover was eliminated with filtration. Agitation did not appreciably affect results. The starting inoculum did not significantly affect the activity of fluconazole or amphotericin B; however, the activity of LY303366 may be influenced by the starting inoculum. Before antifungal time-kill curve methods are routinely employed by investigators, methodology should be scrutinized and standardized procedures should be developed.