TEXT

Top Abstract Text References

In the era of increasing invasive, often life-threatening candidal infections, rapid identification of pathogenic yeasts and detection of polyfungal specimens in the laboratory is mandatory (1). Chromogenic media have shown better detection rates of yeasts in mixed cultures than traditional media and allow direct and more rapid identification of Candida albicans and also other species (2-7, 9). Candida ID (bioMérieux, Marcy l'Etoile, France) is a new differential medium for the direct identification of C. albicans. It is intended to improve differentiation between mixed cultures, specificity for direct identification of C. albicans producing blue colonies, as well as classification in a group of four Candida species (C. tropicalis, C. guilliermondii, C. kefyr, and C. lusitaniae) which produce pink colonies. Also, CHROMagar Candida (CHROMagar, Paris, France) is a chromogenic medium allowing selective isolation of yeasts and simultaneous identification of C. albicans, C. tropicalis, and C. krusei (2, 6, 7, 9).

The purpose of this study was to evaluate the performance of Candida ID directly from clinical specimens in comparison to CHROMagar Candida.

A total of 596 clinical specimens (297 respiratory samples, 151 ear, nose, or throat specimens, 64 vaginal swabs, 44 stool specimens, 19 urine specimens, and 21 miscellaneous samples such as swabs from wounds, eye, and skin) were investigated in order to determine the effect of Candida ID on the growth of Candida species and to assess the quality of performance in comparison with CHROMagar and the routinely used Sabouraud glucose agar (SGA).

Each nonliquid sample was suspended in 1 ml of 0.85% physiologic saline, and then 0.01 ml of this suspension was plated onto Candida ID, CHROMagar, and SGA (Oxoid, Basingstoke, United Kingdom) containing gentamicin (20 mg/liter) and chloramphenicol (50 mg/liter). Equivalent amounts of each liquid specimen were applied directly to the media. After inoculation, the cultures were incubated in air at 35°C and inspected after 24, 48, and 72 h. All yeast isolates observed on the chromogenic media were judged by colony morphology and pigmentation according to the manufacturers' instructions. In addition, colonies of each medium were identified by using the commercial ATB ID 32C (API, bioMérieux, Marcy l'Etoile, France) and by their micromorphology on rice extract agar (Becton Dickinson and Co., Sparks, Md.). Isolates suspected of being C. dubliniensis were also subcultured at 42°C. Candida ID agar and CHROMagar Candida were provided as ready-to-use agar plates, which were stored at 4°C and used within 4 weeks. The performance of both chromogenic media was analyzed for C. albicans in terms of sensitivity [true positives × 100/(true positives + false negatives)] and specificity [true negatives × 100/(true negatives + false positives)].

Of the 596 clinical specimens, 469 had positive cultures yielding one or several yeast strains. A total of 424 showed growth on all media used, 10 on SGA and CHROMagar, 10 on SGA and Candida ID, and 5 on CHROMagar and Candida ID. Twenty specimens grew on one medium only (9 on SGA, 5 on CHROMagar, and 6 on Candida ID). Thus, the overall sensitivity was 96.6% for SGA, followed by Candida ID with 94.9% and CHROMagar with 94.7%. A total of 403 cultures showed growth of one species, 60 showed growth of two species, and 6 showed growth of three species.

Distribution of the colony colors on Candida ID and CHROMagar within each yeast species after 24 and 48 h is shown in Table 1. After 72 h only minor changes were observed (on CHROMagar, six additional green-colored strains of C. albicans and one pink strain of C. glabrata, on Candida ID one more blue-colored strain of C. albicans). Sensitivity and specificity for C. albicans were very high with both chromogenic media, specificity always being 100%. After 24 h, the sensitivity of Candida ID was 96.8%, whereas CHROMagar showed a sensitivity of only 49.6%. After 48 h the respective percentages were 99.7 and 98.9%. After 72 h both chromogenic media showed a sensitivity of 100%. In particular, detection of C. albicans after 24 h of incubation was easier on Candida ID than on CHROMagar. This is because the blue color of C. albicans was better developed on Candida ID after 24 h than the green color on CHROMagar after the same time period.

In contrast to those of Freydière et al. (3) and Fricker-Hidalgo et al. (5), both our isolates of C. dubliniensis developed dark bluish-green colonies on both chromogenic media and could therefore be differentiated from C. albicans. However, it has to be emphasized that the differential character of the darker colonies appeared only after 48 h of incubation and, on Candida ID in one case, only after 72 h. As has been shown for CHROMagar (8), dark bluish-green coloration may be taken as an indication of the presence of C. dubliniensis but may not be used as a sole criterion for identification. We assume that this applies to Candida ID also.

In concordance with other authors (3, 5) we observed not only the species described by the manufacturer (C. kefyr, C. lusitaniae, C. tropicalis, and C. guilliermondii) but also several strains of C. glabrata, C. parapsilosis, and Rhodotorula mucilaginosa growing as pink colonies. On the other hand, our isolates of C. kefyr never developed pink coloration at any time.

Table 2 shows that CHROMagar and Candida ID detected mixtures of yeast species in 53 samples (80.3%) and SGA only in 20 samples (30.3%). Out of 12 polyfungal cultures, 9 were only detected on CHROMagar and 3 of them only on Candida ID. The misidentification of these mixtures may be due to the small inoculum, as only a few colonies grew on each medium. The combination of C. parapsilosis and Saccharomyces cerevisiae was the only one which would have been overlooked using only Candida ID because both of them grew as white inconspicuous colonies. The other mixtures, which were only detected on either CHROMagar or Candida ID, only grew as a mixture on the respective medium and would otherwise have been recognized because of the different coloration and texture of colonies.

In conclusion, the results of our study show that Candida ID is adequately sensitive to grow most of the important yeasts. In more than 50%, identification of C. albicans was shown to be more rapid with Candida ID than with CHROMagar. On the other hand, Candida ID is not as good as CHROMagar in detecting polyfungal specimens, though it can be assumed that most of the mixed cultures will be detected. Nevertheless, Candida ID seems to be a valuable medium for improving and accelerating the detection and identification of C. albicans.