TEXT

Top Abstract Text References

Oropharyngeal candidiasis (OPC) is the most common fungal infection in patients infected with the human immunodeficiency virus (HIV), occurring in up to 90% of these patients (6). For several years, OPC has been treated effectively with azole antifungal compounds. Fluconazole (FLCZ), one of the most commonly used azoles, has been found to be orally active, weakly toxic, extremely effective, and well tolerated, and treatment with this compound is simple (7). Unfortunately, extensive use of the drug for treatment or prophylaxis has led to treatment failure for an increasing number of patients (12). FLCZ resistance was found in 43% of HIV-infected patients (17). Itraconazole (ITCZ) may serve as an effective antifungal agent in patients with OPC nonresponsive to FLCZ (1, 14, 16), because use of amphotericin B is limited due to its toxicity and lack of patient compliance. Furthermore, many works suggest that the results from in vitro susceptibility testing of FLCZ, and perhaps ITCZ, have some utility in predicting the clinical outcome of patients with AIDS-associated OPC (5, 9, 11).

During the last 10 years, considerable progress has been made in the standardization of antifungal susceptibility testing. A reference broth dilution method for determining the MIC, approved standard M27-A, was proposed by the National Committee for Clinical Laboratory Standards (NCCLS) (13). The M27-A reference standard document provides guidelines for establishing interpretative breakpoints that indicate reduced susceptibility and resistance to these two widely used antifungal agents (FLCZ and ITCZ). However, this method is time-consuming and difficult to use as a routine technique. Thus, new methods that are easier to perform have recently been developed (2, 3, 8, 10). Fungitest (Sanofi-Pasteur, Marnes la Coquette, France), presented in the form of a microplate, is one of them. The Fungitest kit was evaluated by Davey et al. (4) with different clinical isolates. Their results indicated that few strains were resistant to FLCZ (6.5%) and ITCZ (5%). In this study, we focused on strains from HIV-infected patients, because of the existence of a high level of azole agent resistance among HIV strains. We compared the results obtained, for FLCZ and ITCZ, by Fungitest and the NCCLS microdilution plate method.

Over the course of 1 year, 118 clinical isolates from HIV-infected patients suffering from OPC were tested prospectively by these two tests. The strains were obtained from 50 patients after 85 samplings by washing patient mouths with 10 ml of sterile 0.9% NaCl solution. This work included 76 Candida albicans isolates, 21 Candida glabrata isolates, 21 other Candida isolates (9 of which were Candida krusei, 7 of which were Candida tropicalis, 3 of which were Candida kefyr, 1 of which was Candida lusitaniae, and 1 of which was Candida inconspicua). C. albicans was identified by conventional methods (determination of assimilation profiles by the Auxacolor kit [Sanofi-Pasteur] and chlamydospore production), but these tests did not allow the detection of Candida dubliniensis. Two reference strains, Candida parapsilosis (ATCC 22019) and C. krusei (ATCC 6258), were incorporated in this study as quality control isolates.

Fungitest was used to study the growth of yeast in the presence of two of the six antifungal agents present and at two different concentrations. Growth assessment was based on color change of a redox indicator. The breakpoint concentrations used in this kit were 8 and 64 µg/ml for FLCZ and 0.5 and 4 µg/ml for ITCZ. The obtained results allowed us to classify the strains into (i) inhibited strains, which exhibited no growth at the two antifungal concentrations; (ii) intermediate strains, which exhibited growth only at the low antifungal concentration; and (iii) noninhibited strains, which exhibited growth at the two antifungal concentrations.

Fungitest was performed according to the manufacturer's instructions, and the broth microdilution method was performed according to the guidelines of NCCLS document M27-A (13). The NCCLS microdilution test was performed in 96-well microplates. Yeast suspensions was prepared in RPMI 1640 medium buffered with MOPS (morpholinepropanesulfonic acid) buffer to get a final inoculum concentration of about 0.5 × 10³ to 2.5 × 10³ organisms/ml. The microplates were incubated at 37°C and read after 48 h. The MICs were determined by spectrophotometric reading at a wavelength of 492 nm as the lowest concentration at which there was 80% growth inhibition compared with that in a drug-free control.

The breakpoints established for FLCZ and ITCZ according to the NCCLS guidelines do not correspond to breakpoints selected for Fungitest. Here, the Fungitest breakpoints (8 and 64 µg/ml for FLCZ and 0.5 and 4 µg/ml for ITCZ) were applied to the MIC method to permit comparisons. The strains are considered resistant with an FLCZ MIC of 64 µg/ml and an ITCZ MIC of 4 µg/ml. The reference strains showed the susceptibility patterns expected with the two techniques: FLCZ MICs of 4 to 8 µg/ml and ITCZ MICs of 0.125 to 0.25 µg/ml for C. parapsilosis; FLCZ MICs of 16 to 32 µg/ml and ITCZ MICs of 0.25 to 0.50 µg/ml for C. krusei.

The results with FLCZ and ITCZ for clinical isolates are presented as Tables 1 and 2, respectively. In our tested population, 31 isolates (26%) of Candida spp. were FLCZ resistant (MIC 64 µg/ml) and 39 isolates (33%) were ITCZ resistant (MIC 4 µg/ml).

The MIC of FLCZ for 64 of the 66 inhibited strains (97%) was 8 µg/ml. For eight of nine of the noninhibited strains (88%) MICs were 64 µg/ml. However, for only 8 of 22 intermediate strains of C. albicans, 6 of 16 strains of C. glabrata, and 4 of 5 strains of the other species, MICs were between 8 and 64 µg/ml. For more than half (53%) of the intermediate strains (12 of 22 strains of C. albicans, 10 of 16 strains of C. glabrata, and 1 of 5 strains of the other species) MICs were 64 µg/ml.

For 55 of the 65 inhibited strains (84%) ITCZ MICs were 0.5 µg/ml. For only two (C. glabrata) of the four noninhibited strains, MICs were 4 µg/ml. For only 9 of 24 intermediate strains of C. albicans, 1 of 16 strains of C. glabrata, and 5 of 10 strains of the other species, ITCZ MICs were between 0.5 and 4 µg/ml. For the 35 intermediate and misclassified strains, 33 (15 of 24 strains of C. albicans, 15 of 16 strains of C. glabrata, and 3 of 10 strains of the other species) MICs were 4 µg/ml.

In tests with the two antifungal agents, a majority of C. glabrata strains (10 of 16 with FLCZ, 15 of 16 with ITCZ) were intermediate strains by Fungitest but were resistant strains by the MIC method. The difference between the results of the two techniques can be explained by the difficult growth of C. glabrata on a majority of the commonly used incubation media.

Furthermore, only 8 of the 31 strains (26%) resistant by the MIC method to FLCZ and 2 of the 39 strains (5%) resistant to ITCZ were determined to be noninhibited strains by Fungitest. The results of this test are dependent on the yeast species tested and antifungal agents used.

These results confirm the difficulty presented by Davey et al. (4) of determining the susceptibilities of strains by Fungitest. In conclusion, the results of Fungitest are in good agreement with those of the NCCLS microdilution plate method with inhibited strains (97% with FLCZ and 84% with ITCZ), but it is better to verify the results with intermediate strains by another method. The Fungitest method with broth medium has the advantages of being easy to read (with its colored indicator) and easy to use. Nevertheless, the major inconvenience of Fungitest is its limited ability to detect resistant strains, particularly those resistant to ITCZ.