Mass spectrometry MALDI-TOF for identification of the Candida parapsilosis complex

IRANZO A; GARCIA, GUILLERMO; PEMÁN, JAVIER; CANTON, EMILIA; LOPEZ-HONTANGAS, JL

Valencia

Congreso; Trends in Medical Mycology; 2011

European Confederation of Medical Mycology and EORTC Infectious Diseases Group

Yeast species identification is performed in most laboratories using selective and identification media such as Cromagar or by automated methods based on biochemical tests. These methods require more than 48h for colony growth and subsequent identification. Mass spectrometry MALDI-TOF (MS), a novel identification method based on the detection of microorganism-specific proteins, significantly shortens the time of identification, thus allowing a more rapid decision to start antifungal treatments, which is very useful to clinicians and beneficial to patients. The aim of this study was to evaluate the ability of mass spectrometry to identify species of the group psilopsis. To this end, we have used molecularly identified isolates of Candida parapsilosis, Candida orthopsilosis and Candida metapsilosis from the strain collection of our hospital (Hospital Universitario La Fe, Spain). A total of 285 strains of the C. parapsilosis complex (215 C. parapsilosis, 64 C. orthopsilosis and 6 C. metapsilosis) previously identified by PCR using the method described by Tavanti et al. were used to compare with MS identification. All strains were isolated from blood cultures of our patients, collected during the period 1995-2007. For MS identification of each isolate, an extraction using formic acid and acetonitrile was made. This extract was analysed in a microflex mass spectrometer (Bruker Daltonik GmbH, Bremen) and the spectra obtained were compared with the MALDI Biotyper 2.0 database. MS identifications were in agreement with PCR for the 6 C. metapsilosis (100%), 204 C. parapsilosis (94,8%) and 42 C. orthopsilosis (65,6%). We obtained no reliable identification for 5 C. parapsilosis (2.3%) and 1 C. orthopsilosis (1.5%). The score range and mean average obtained for correct identifications were for C. parapsilosis from 1.782 to 2.374, mean average 2.143; for C. orthopsilosis 1.862 to 2.3, mean average 2.109; and for C. metapsilosis 2.054 to 2.279, mean average 2.16. MS correctly identified 252 isolates (overall agreement, 88.4%). MS discrepancies with PCR identification were: 6 isolates identified as C. parapsilosis by PCR were identified as C. orthopsilosis by MS; and 21 isolates identified as C. orthopsilosis by PCR were identified as C. parapsilosis by MS. There was no misidentification in C. metapsilosis. Although few data were obtained for C. metapsilosis, MS seems to be a sensitive and specific method to identify this species. The percentage of correct identification for C. orthopsilosis was the lowest and always misidentified the isolates as C. parapsilosis. We believe that in the near future MS will further improve the identification of this group, particularly C. orthopsilosis, as the quality of the databases available improves.