|Home | About | Journals | Submit | Contact Us | Français|
The in vitro susceptibilities of 36 clinical isolates of Madurella mycetomatis, the prime agent of eumycetoma in Africa, to ketoconazole, itraconazole, fluconazole, voriconazole, amphotericin B, and flucytosine were determined by the Sensititre YeastOne system. This system appeared to be a rapid and easy test, and by use of hyphal suspensions it generated results comparable to those of a modified NCCLS method. After 10 days of incubation, the antifungal activities of ketoconazole (MIC at which 90% of isolates were inhibited [MIC90], 0.125 μg/ml), itraconazole (MIC90, 0.064 μg/ml), and voriconazole (MIC90, 0.125 μg/ml) appeared superior to those of fluconazole (MIC90, 128 μg/ml) and amphotericin B (MIC90, 1 μg/ml), with MICs in the clinically relevant range. All isolates were resistant to flucytosine (all MICs above 64 μg/ml). Based on the relatively broad range of MICs obtained for the antifungal agents, routine testing of M. mycetomatis isolates for susceptibility to antifungal agents seems to be relevant to adequate therapeutic management.
Eumycetoma is a chronic disease endemic in many tropical and subtropical countries and is characterized by tumefaction, draining sinuses, and the presence of grains (8, 9, 15, 21, 24). The lesion initiates in the subcutaneous tissue and spreads to destroy the surrounding skin, the muscles, and the bones (8, 15). Eumycetoma can be caused by a variety of microorganisms, but the causative organism most frequently encountered in Sudan is the fungus Madurella mycetomatis (9, 10, 15, 24). Mycetoma caused by M. mycetomatis is usually treated with extensive surgery, combined with prolonged antifungal therapy (10, 15, 21). Surgery for mycetoma ranges from local excision of distinct lesions to amputation of the infected limb (8, 10). In clinical trials a number of antifungal agents, including itraconazole and ketoconazole, are being used for the treatment of eumycetoma, with various degrees of success (9, 16, 29). Limited data are available regarding the antifungal susceptibility of M. mycetomatis (2, 25, 31). Determination of the in vitro susceptibility of the nonsporulating M. mycetomatis is troublesome. In the method for testing the susceptibility of spore-forming filamentous fungi according to the NCCLS guidelines (M38-A), a conidial suspension is used as an inoculum (28). For nonsporulating fungal species, a hyphal suspension has to be used (2). In a recently published study by Ahmed et al., the hyphal inocula of M. mycetomatis were standardized by sonicating the fungus to obtain a homogeneous suspension (2). Using the modified NCCLS method, the authors obtained reproducible amphotericin B data for 90% of the M. mycetomatis isolates tested (2). However, for itraconazole, visual reading of the gradually decreasing antifungal activity was not always accurate. That is why Ahmed et al. used the 2,3-bis(2-methoxy-4-nitro-5-sulfophenyl)-5-[(phenylamino)carbonyl]-2H-tetrazo-lium hydroxide (XTT) assay to facilitate the reading of end points. In this assay, antifungal activity can be quantified by relating the viable fungal mass to levels of fungal mitochondrial dehydrogenase activity (18, 26). The XTT assay appeared appropriate and accurate for determination of the susceptibility of M. mycetomatis to antifungal agents (2). Unfortunately, the modified NCCLS method combined with the viability-based XTT assay is both time-consuming and laborious and therefore is not easily implemented to screen M. mycetomatis isolates routinely for their susceptibilities to a large number of antifungal agents.
A less laborious method, which recently became available, is the Sensititre YeastOne system (Trek Diagnostic Systems, Ltd., East Grinstead, England). This system is a commercial microdilution method that uses the oxidation-reduction indicator Alamar blue to determine in vitro susceptibilities to amphotericin B, fluconazole, itraconazole, ketoconazole, flucytosine, and voriconazole. The test has been used successfully to assess the susceptibilities of various yeasts and filamentous fungi, including Aspergillus spp. (4, 7, 23, 27). In the present study, we evaluated the in vitro activities of a number of antifungal agents against 36 clinical isolates of M. mycetomatis by using the commercial Sensititre YeastOne system. In addition, we compared this assay with the modified NCCLS method and the viability-based XTT assay.
A total of 36 clinical isolates of M. mycetomatis were included in this study. Thirty-four of these strains were obtained from patients seen in the Mycetoma Research Centre, University of Khartoum, Khartoum, Sudan. Two additional strains were obtained from patients originating in Mali. The strains were isolated from biopsy specimens and were maintained on Sabouraud dextrose agar (Difco Laboratories, Paris, France). The strains were previously identified by morphology and PCR-restriction fragment length polymorphism (2, 3, 5).
For the modified NCCLS method and the XTT assay, ketoconazole and itraconazole were obtained from Janssen Pharmaceutical Products (Beerse, Belgium), fluconazole and voriconazole were obtained from Pfizer BV (Capelle aan de Ijsel, The Netherlands), and amphotericin B was obtained from Bristol-Myers Squibb (Woerden, The Netherlands). Flucytosine was obtained from ICN Pharmaceuticals Holland BV, Zoetermeer, The Netherlands.
MICs were determined independently in duplicate by the colorimetric Sensititre YeastOne method (Trek Diagnostic Systems, Ltd.), the XTT assay, and the modified NCCLS method (2). M. mycetomatis was cultured for 10 days at 37°C in RPMI 1640 medium supplemented with l-glutamine (0.3 g/liter) and 20 mM morpholinepropanesulfonic acid (MOPS). The mycelia were harvested by a 5-min centrifugation at 2,158 × g and were washed with sterile saline. To homogenize the inoculum, the mycelia were sonicated for 20 s at 28 μm (Soniprep, Beun de Ronde, The Netherlands). For the Sensititre test, the final inoculum was prepared from the homogenized fungal suspension mixed with the YeastOne broth (Trek Diagnostic Systems) to obtain a transmission of 70% at 660 nm (Novaspec II; Pharmacia Biotech). One hundred microliters of this suspension was applied to the Sensititre YeastOne plate, and the plate was incubated for 7 days at 37°C. MIC end points were determined visually at the first blue well for amphotericin B and the first purple well for the other antifungal agents. Drug concentrations used in this test ranged from 0.008 to 16 μg/ml for itraconazole, ketoconazole, voriconazole, and amphotericin B; from 0.125 to 256 μg/ml for fluconazole; and from 0.032 to 64 μg/ml for flucytosine.
The XTT assay and the modified NCCLS method were performed according to the method of Ahmed et al. (2). Briefly, in the XTT assay, MIC end points were determined by adding the substrate XTT, which is enzymatically converted to the colored formazan in viable cells (18). The extinction of the supernatant was measured spectrophotometrically at 450 nm. The MIC end points were defined as the lowest concentrations resulting in 80% or more reduction in viable fungal mass (2). In the modified NCCLS method, the MIC end points were determined visually as the lowest concentrations that prevented any discernible fungal growth (2, 28). Drug concentrations used in both tests ranged from 0.016 to 8 μg/ml for itraconazole, ketoconazole, voriconazole, and amphotericin B; from 0.25 to 128 μg/ml for fluconazole; and from 0.125 to 64 μg/ml for flucytosine.
For a total of 36 M. mycetomatis isolates, the susceptibilities to ketoconazole, itraconazole, fluconazole, voriconazole, amphotericin B, and flucytosine were determined by using the Sensititre system, the NCCLS method, and the XTT assay. For two strains, MICs could not be determined by using the Sensititre system because these strains did not grow in the presence of Alamar blue. The MIC distribution for the other 34 isolates is shown in Fig. Fig.1.1. The MICs of the different antifungal agents appeared variable and dependent on the M. mycetomatis isolate. The majority of the M. mycetomatis strains showed high susceptibilities to ketoconazole, itraconazole, and voriconazole. Ketoconazole MICs ranged from <0.016 to 1 μg/ml, and 0.125 μg of ketoconazole/ml was needed to inhibit 90% of the isolates. The isolates were even more susceptible to itraconazole. For this antifungal agent, a MIC range of <0.016 to 0.5 μg/ml was found, and only 0.064 μg/ml was needed to inhibit 90% of the strains. The antifungal activity of voriconazole was similar to that of ketoconazole, with MICs ranging from <0.016 to 1 μg/ml, and a concentration of 0.125 μg/ml was needed for inhibition of 90% of the isolates. The only azole for which a high range of MICs (0.25 to >128 μg/ml) was found was fluconazole. A concentration of 16 μg of fluconazole/ml was needed to inhibit 90% of the isolates. The isolates were less susceptible to the non-azole antifungal agents. Amphotericin MICs were between <0.016 and 4 μg/ml, and a concentration of 2 μg/ml was required to inhibit 90% of the strains. All M. mycetomatis isolates were resistant to flucytosine.
To determine the accuracy of the Sensititre test, all M. mycetomatis strains were investigated by three methods: the modified NCCLS method, the XTT assay, and the Sensititre method. The percentages of agreement in experimental outcomes for each antifungal agent are shown in Table Table1.1. It is concluded from these data that the reproducibility of the Sensititre test was good. Reproducible results, differing by no more than a one-step dilution, were obtained for more than 90% of the strains with all the antifungal agents except ketoconazole (Table (Table1).1). For ketoconazole, only 88.2% reproducibility was found, which is still very high. When the Sensititre system was compared to the NCCLS method, identical MICs or MICs differing by a single dilution step were obtained in 88.2 to 100% of the cases. This was comparable to the level of agreement found between the NCCLS method and the XTT assay (85.3 to 100.0%). A somewhat lower level of agreement was found between the Sensititre system and the XTT assay. In this case, levels of agreement were still high for the non-azole antifungal agents (82.4% for amphotericin B and 100.0% for flucytosine) but marginally lower for the azoles. For fluconazole, the agreement was as low as 67.6%, while for the other azoles, agreement ranged from 70.6 to 91.2%. Overall, the XTT assay resulted in relatively higher MICs than the Sensititre method, with a two- or three-step dilution difference.
Recently Ahmed et al. reported two reproducible assays for measurement of the susceptibility of M. mycetomatis isolates to antifungal agents: an adapted protocol based on the NCCLS (M38-A) guidelines and a viability-based XTT assay for facilitating end point reading (2). Both test systems appeared reproducible and sensitive but were also time-consuming. For routine use, a system for testing susceptibility to antifungal agents should be cheap, fast, and easy to interpret. Recently, the YeastOne Sensititre system for determination of the susceptibilities of several yeast and fungal species, such as Candida spp., Crytococcus spp., and Aspergillus spp., to antifungal agents has been introduced (4, 6, 20, 23, 27, 30). In this system, the MIC end points can be determined visually because of the dye Alamar blue, which is converted from blue to red when fungal growth occurs (4, 6, 20, 23, 27, 30). Various studies show that the MICs obtained for several yeasts by the Sensititre system were in good agreement with those obtained by the NCCLS method (4, 6, 20). However, for Aspergillus spp. there was less agreement between the two methods (23, 27). To investigate the value of the Sensititre system for M. mycetomatis isolates, in the present study this system was compared to the modified NCCLS method and the XTT assay (2). Good agreement was found between MICs obtained by the Sensititre method and the modified NCCLS method. Overall, the MICs obtained by the Sensititre method were equal to, or 1 dilution lower than, the MICs obtained by the modified NCCLS method. Findings of lower MICs with the Sensititre system have also been reported for other fungal species, such as Cryptococcus neoformans and Aspergillus spp. (6, 27). Although we found lower MICs, the difference was not statistically significant. The discrepancies between the Sensititre system and the XTT test were higher. This may be explained by the fact that the indicator systems in the two tests were different (17, 21). The discrepancies between the Sensititre system and the XTT test were highest for the azoles. This suggests that the trailing end point effect, usually obtained with azoles, is measured more effectively with XTT than with Alamar blue.
In the present study, the antifungal susceptibilities of 36 M. mycetomatis isolates to ketoconazole, itraconazole, fluconazole, voriconazole, amphotericin B, and flucytosine were determined by using the Sensititre method. For two of the strains MICs could not be determined, because these strains did not grow in the test medium in the presence of Alamar blue. Jahn et al. encountered the same problem when testing isolates of Aspergillus fumigatus. They found strain-dependent differences which could not easily be explained (17). In the present study, both ketoconazole and itraconazole appeared to be very effective at inhibiting the M. mycetomatis strains. For both antifungal agents, only low concentrations were needed to inhibit 90% of the clinical isolates: 0.125 and 0.064 μg/ml, respectively. The MICs found for these two antifungal agents correlate with attainable levels in serum (5). Ketoconazole was one of the first antifungal agents used in the treatment of eumycetoma caused by M. mycetomatis; more recently, itraconazole has been used as well (21, 22, 24). Although some clinical studies showed that ketoconazole and itraconazole resulted in complete cure, the clinical response to these agents is often poor (2, 21, 22, 24). This may be partially explained by the observed variation in MICs for the M. mycetomatis isolates. Fluconazole was less effective than ketoconazole and itraconazole at inhibiting fungal growth. Fluconazole MICs were high for two isolates (>64 μg/ml), while those for the other isolates ranged from 0.125 to 16 μg/ml. Although these MICs are high, they still correlate with physiologically attainable levels in serum (5, 11). Voriconazole, a relatively new azole that is highly effective against aspergillosis, showed similarly high antifungal activity against the M. mycetomatis strains compared with ketoconazole and itraconazole (13). Amphotericin B appeared to be less effective than ketoconazole, itraconazole, and voriconazole at inhibiting M. mycetomatis. This observation is in accordance with the study performed by Ahmed et al. (2). They also found that amphotericin B was less effective than itraconazole at inhibiting M. mycetomatis; for 33% of the isolates, amphotericin B MICs exceeded the attainable peak levels of the drug in plasma (2). Of all the antifungal agents tested in the present study, flucytosine was the least effective. Even at high concentrations, no fungal inhibition was noticed. The M. mycetomatis isolates appeared to be resistant to flucytosine, which is also the case for many other filamentous fungi (1, 12, 14, 19, 28).
In conclusion, the Sensititre YeastOne system is an appropriate system for determination of the susceptibility of M. mycetomatis strains to antifungal agents. The fungus was highly susceptible to ketoconazole, itraconazole, and voriconazole, moderately susceptible to fluconazole and amphotericin B, and resistant to flucytosine. The differences in MICs observed for the different M. mycetomatis isolates suggest that the introduction of routine testing of the susceptibility of M. mycetomatis isolates to antifungal agents is important for adequate therapeutic management.
We gratefully acknowledge Ilbert Borgsteijn (MCS Diagnostics, Swalmen, The Netherlands) for providing the Sensititre YeastOne test.
MCS Diagnostics had no involvement in the design of the present study, the interpretation of the data, or the preparation of the present paper.