This study aimed to develop a method for rapid and accurate identification of yeasts obtained in the clinic, especially from immunocompromised patients, in order to provide a timely and appropriate antifungal therapy. A total of 112 Candida isolates were analyzed in this study; 28 of them were used to validate the PCR-HRM method in species identification in a blinded manner. Strains were identified by conventional techniques that use VITEK 2 YST cards and Matrix-assisted laser desorption/ionization time of flight mass spectrometry (MALDI-TOF-MS). These methods were compared to the newly developed technique based on real-time polymerase-chain reaction high-resolution melting (PCR-HRM). Discordant results were resolved with internal transcribed spacer (ITS) gene sequencing, the "golden standard" used to evaluate the reliability of all methods in identifying yeasts at the species level. PCR-HRM sensitivity was assessed with the isolated strains. VITEK 2, MALDI-TOF-MS, and PCR-HRM accurately identified 89.2% (74/83), 97.6% (81/83), 100% (83/83) of the isolates, respectively. PCR-HRM detection limit was 1fg/μl of yeasts. In validation assays, a 100% accuracy rate was achieved by the use of PCR-HRM. Therefore, the PCR-HRM method is a rapid, sensitive, and specific diagnostic approach, which provides a cost-effective and more suitable alternative for yeast identification in a clinical laboratory. Future research is needed for automation of data acquisition.
Keywords: Candida; Polymerase-chain reaction high-resolution melting analysis (PCR-HRM); Rapid identification.
© 2015 by the Association of Clinical Scientists, Inc.