Activities and Ultrastructural Effects of Antifungal Combinations against Simulated Candida Endocardial Vegetations

Abstract

In vitro pharmacodynamic model (PDM) simulation of serum antifungal concentrations may predict the value of combination antifungal regimens against Candida sp. endocarditis. We investigated the effects of combinations of flucytosine (5FC), micafungin (Mica), and voriconazole (Vor) against Candida-infected human platelet-fibrin clots, used as simulated endocardial vegetations (SEVs). Single clinical bloodstream isolates of Candida albicans, Candida glabrata, Candida parapsilosis, and Candida tropicalis were used. All four isolates were susceptible to 5FC, while C. glabrata was resistant to Vor and C. tropicalis had a paradoxical resistance phenotype to Mica. The SEVs were prepared with an initial inoculum of 1 x 10(6) CFU/g of SEV and added to a PDM, which utilized yeast nitrogen broth-2% glucose and incubation at 35 degrees C and simulated antifungal pharmacokinetic profiles. Fungal densities in the SEVs were determined in quadruplicate over 72 h. Scanning electron microscopy (SEM) was used to evaluate treatment and control SEVs. Vor was the least active single agent against all Candida spp. except for C. parapsilosis, where it was comparable to Mica. In contrast, 5FC was the most active against all Candida spp. except for C. tropicalis, where it was comparable to Mica. The combination of 5FC plus Vor was superior to either agent alone against C. parapsilosis. The combination of Vor plus Mica was inferior to the use of Mica alone against C. tropicalis. The triple combination of 5FC plus Vor plus Mica was no better than single or dual agents against any of the Candida spp. The ultrastructural features of infected SEVs were unique for each Candida sp., with C. parapsilosis in particular manifesting friable biofilm clusters. In general, 5FC and Mica were superior in their rates and extents of fungal burden reduction compared to Vor against Candida-infected SEVs. Evaluation of 5FC and Mica in animal models of Candida endocarditis is warranted.

Antifungal Combinations against Simulated Candida albicans Endocardial Vegetations

“…In contrast, in vitro models provide an alternative, more rapid, and controlled environment for the assessment of antifungal combinations. My coworkers and I previously determined the activity of flucytosine (5FC), micafungin (Mica), and voriconazole as single agents and in combinations against Candida species by using an in vitro model of infective endocarditis (11). This work demonstrated the poor activity of voriconazole and the superior activity of 5FC and Mica against developing Candida biofilm.…”

“…Unlike these agents, the echinocandins target the cell wall of Candida species through inhibition of ␤-1,3-glucan synthase. A growing body of evidence suggests that echinocandins are active against Candida albicans biofilm (1, 4,7,8,11,[13][14][15]. However, cell wall integrity pathways and glucan-associated changes can also occur, which could limit echinocandin activity against Candida biofilm (10).…”

“…The log 10 CFU/g over time (0 to 48 h) was also compared between the different regimens tested and growth controls. The maximum rate of kill (K max ), area under the rate of kill curve (AURKC), and area between the treatment and control timekill curves (ABTKC) were calculated (11). The seven treatment groups were compared using one-way analysis of variance with post hoc comparisons using Bonferroni correction for multiple comparisons of significance.…”

“…A one-compartment infection model (250 ml) was utilized in duplicate as previously described (11). The experiments were conducted over 72 h, which included a 24-h biofilm development phase followed by a 48-h treatment phase.…”

“…The C. albicans strains used in this study were selected because they reflected MIC 50 s and MIC 90 s of Mica from global surveillance data (12). A single colony of C. albicans obtained from a 24-h culture on Sabouraud dextrose agar (Cole-Parmer, Vernon Hills, IL) was grown in yeast nitrogen base medium (Difco Laboratories, Detroit, MI) supplemented with 2% dextrose at 27°C for 24 h. Infected fibrin clots were prepared as previously described but at a starting inoculum of ϳ10 4 CFU/g (11). The smaller inoculum permitted growth of C. albicans to ϳ10 6 CFU/g at 24 h just prior to exposure to antifungals.…”