Evaluation of the Becton Dickinson Phoenix Yeast Panel and the RapID Yeast Plus System in the Identification of Yeast and Yeast-Like Organisms
Michelle L Oros, Shobha Parajuli, Raquel Deleon-Gonsalves, Raghava Potula, Allan L Truant. Temple University Hospital, Philadelphia, PA
Background: The frequency of opportunistic yeast and yeast-like infections continues to grow as the prevalence of immunocompromised individuals rises from transplantation, HIV/AIDS, neoplastic diseases, and immunosuppressive therapies. These infections are associated with high morbidity and mortality rates, especially as the diversity of the organisms continues to expand and drug resistance becomes more prevalent. Therefore, the ability to make a rapid and accurate diagnosis is essential. In 2011, a new automated system, Becton Dickinson (BD) Phoenix Yeast Panel, was introduced. It uses fluorogenic and chromogenic substrates for yeast identification. Manufacturing studies claim greater than 95% accuracy for all the 64 yeast and yeast-like isolates. This study compares the new methodology with a common manual qualitative micromethod, RapID Yeast Plus System, which uses conventional and chromogenic substrates for identification. To our knowledge, this is the first known independent study to evaluate the BD Phoenix Yeast Panel.
Design: Clinical yeast isolates collected at a single institution from March to August 2012 were plated to Sabouraud-dextrose Agar and incubated at 30°C for a maximum of 48 hours. Cultures were suspended into Phoenix or RapID yeast inoculation broth after 24 and 48 hours, respectively and processed. In addition to this, ChromAgar analysis and germ tube test were performed for additional phenotypic characterization. Discrepant results were retested with the API 20C AUX and results were obtained after 72 hours of incubation.
Results: A total of 133 isolates were obtained from 14 different clinical species with the four most common species being C. albicans (27%), C. glabrata (26%), C. tropicalis (20%), and C. parapsilosis (13%). The Phoenix and RapID systems agreed for 127 isolates (95%). Of the 6 discrepancies, API 20C AUX agreed with the Phoenix for 5 isolates and the RapID Yeast Panel for 1 isolate. The 5 misidentified by the RapID yeast system were all of the same species, C. parapsilosis and overall the RapID panel only identified 12 out of 17 C. parapsilosis isolates (70%). Overall, no statistically significant difference was observed between the BD Phoenix and RapID Yeast Plus systems (P=0.09, Chi Square test).
Conclusions: The two identification systems produce similar results. However in this study, the BD Phoenix yeast panel appears to identify C. paraspilosis more accurately. Molecular confirmation of the isolates and correlation of discordant results would provide further explanation for discrepancy.
Monday, March 4, 2013 9:30 AM
Poster Session I Stowell-Orbison/Surgical Pathology/Autopsy Awards Poster Session # 246, Monday Morning