Few studies have systematically standardised and evaluated matrix-assisted laser desorption ionization time-of-flight mass spectrometry (MALDI-TOF MS) for identification of yeasts from bloodstream infections. This is rapidly becoming pertinent for early identification of yeasts and appropriate antifungal therapy. We used 354 yeast strains identified by polymerase chain reaction (PCR) sequencing for standardisation and 367 blind clinical strains for validation of our MALDI-TOF MS protocols. We also evaluated different sample preparation methods and found the on-plate formic acid extraction method as most cost- and time-efficient. The MALDI-TOF assay correctly identified 98.9% of PCR-sequenced yeasts. Novel main spectrum projections (MSP) were developed for Candida auris, C. viswanathii and Kodamaea ohmeri, which were missing from the Bruker MALDI-TOF MS database. Spectral cut-offs computed by receiver operating characteristics (ROC) analysis showed 99.4% to 100% accuracy at a log score of ≥ 1.70 for C. tropicalis, C. parapsilosis, C. pelliculosa, C. orthopsilosis, C. albicans, C. rugosa, C. guilliermondii, C. lipolytica, C. metapsilosis, C. nivariensis. The differences in the species-specific scores of our standardisation and blind validation strains were not statistically significant, implying the optimal performance of our test protocol. The MSPs of the three new species also were validated. We conclude that MALDI-TOF MS is a rapid, accurate and reliable tool for identification of bloodstream yeasts. With proper standardisation, validation and regular database expansion, its efficiency can be further enhanced.
Keywords: Bloodstream infection; Candida; matrix-assisted laser desorption ionization time-of-flight mass spectrometry; rapid diagnosis; yeast.
Copyright © 2014 European Society of Clinical Microbiology and Infectious Diseases. Published by Elsevier Ltd. All rights reserved.