[Evaluation of the BD Phoenix ID Yeast System for the Species Identification of Clinical Yeast-Like Organisms].

ABSTRACT

Most fungi isolated from patients with deep-seated mycosis are yeast-like organisms such as Candida and Cryptococcus. As their respective susceptibilities to antifungal agents can vary depending on the species, rapid identification is important for the administration of appropriate antifungal therapy. The aim of this study was to evaluate the performance of a new automated identification panel, Phoenix Yeast ID (Becton, Dickinson Diagnostics, USA) as well as the time required for identification. The identification results of 106 isolates generated by this system were then compared with those of the API 20C AUX system (SYSMEX bioMérieux Co., Ltd. Japan). Among the 106 isolates, the identification agreement between the two yeast panels was 97/106 (91.5%). Of the 9 (8.5%) discrepant identifications, 5 identification using the Phoenix Yeast ID system and 1 identification using the API 20C AUX system agreed with the genotypic identification. Genotypic identification did not agree with the Phoenix Yeast ID or API 20C AUX findings for the remaining 3 discrepant identifications. Approximately 60% of the C. albicans, C. tropicalis, and C. parapsilosis isolates were identified within 4 hours. In total, about 90% of the 4 major Candida sp. (C. albicans, C. tropicalis and C. glabrata) were identified within 8 hours. In conclusion, the Phoenix Yeast ID findings agreed well with the API 20C AUX findings. Genotypic identification of the discrepant identifications confirmed most of the Phoenix Yeast ID panel identifications. As approximately 80% of the major Candida sp. could be identified within 8 hours using the Phoenix Yeast ID identification system, our results suggest that this system is a clinically useful addition to commercially available yeast identification panels. The Phoenix Yeast ID system showed excellent concordance with genotypic identification for the classification of organisms with discrepant API 20C AUX findings.