Search

Accurate and rapid identification of yeast isolates have become important in recent years for not only antifungal susceptibility testing due to the species-specific clinical resistance breakpoints but also early initiation of appropriate antifungal therapy. In clinical microbiology laboratories species identification of yeasts is often performed with several commercial systems based on biochemical properties and rarely according to the physiological and morphological characteristics. The aim of this study was to compare the two common commercial systems, VITEK 2 YST ID Card (Vitek; bioMérieux, France) and API 20C AUX (API; bioMérieux, France) with conventional mycological methods. A total of 473 clinical yeast strains isolated from clinical specimens in different university and training/research hospitals and identified by Vitek system were included in the study. The isolates were re-identified with API and conventional methods including morphological identification in the Mycology Reference Laboratory of the Public Health Institute of Turkey. Candida dubliniensis MYA 583, Candida krusei ATCC 6258, Candida parapsilosis ATCC 22019, Candida albicans ATCC 10231 and Cryptococcus neoformans ATCC 32268 were used as quality control strains and those standard strains were studied consecutively 10 days with both of the methods. The results of identification by Vitek and API were compared with the results of conventional methods for those 473 yeast isolates [6 genus (Candida, Cryptococcus, Blastoshizomyces, Rhodotorula, Saccharomyces, Trichosporon), 17 species (5 common and 12 rarely isolated)]. The performances of the systems were better (Vitek: 95%; API: 96%) for the commonly detected species (C.albicans, C.parapsilosis, C.glabrata, C.tropicalis and C.krusei) than those for rarely detected species (Vitek: 78.4%; API: 71.6%) (p= 0.155). Misidentification or unidentification were mostly detected for C.parapsilosis (Vitek: 6/87; API: 7/87) and C.glabrata (Vitek: 9/104; API: 3/104) by both of the systems. For rarely detected yeast isolates, misidentification or unidentification were most frequently observed in species of C.pelliculosa (Vitek: 3/11; API: 6/11) and C.dubliniensis (API and Vitek: 2/5) isolates. Candida guilliermondii (API: 2/5) isolates had lower rate of identification with API compared to other species. Blastoschizomyces capitatus and Saccharomyces cerevisiae isolates could not be identified by both of the systems. As a result, the accurate diagnosis of Vitek and API systems were similar in terms of consistency (86.3%). Two systems performed well in correct identification of common clinical yeast species (at least 95%), while the identification of rare species was more challenging indicating that they require further morphological and physiological testing. The addition of morphological identification to commercial systems will be useful for accurate diagnosis and treatment of mixed infections.

Vancomycin-resistant enterococci (VRE) are important agents of hospital infections worldwide. Early recognition of VRE colonization is important in the control of hospital infections. The aim of this study was to compare a real-time PCR (Rt-PCR) system and culture methods in the detection of VRE colonization. A total of 210 perirectal swab samples obtained from the patients (142 were in internal and 68 were in surgical intensive care units) hospitalized at Ankara Training and Research Hospital, Turkey between January-September 2013 were included in the study. The samples were simultaneously evaluated with both Rt-PCR (GeneXpert®vanA/vanB, Cepheid, USA) and the culture methods. The samples were cultivated in enterococcosel agar and incubated at 370C for culture. Culture plates were evaluated for three days on a daily basis. Bacterial identification was done by conventional methods and automated Vitek 2.0 system (BioMérieux, France). Antibiotic susceptibility testing was performed by the E-test. VRE was detected in 76 (36.1%) of the samples by the Rt-PCR method; of them 70 were positive for vanA, two for vanB, and four for vanA + vanB. On the other hand, VRE was detected in 71 (33.8%) of the samples by the culture method. Out of 71 samples, colony growth was observed on the first day in 39 cases, on the second day in 29 cases, and on the third day in three cases. The two strains identified as vancomycin-sensitive enterococci by the Vitek 2 Compact system were determined as vanB positive by PCR. These samples were also confirmed as VRE by E-test. The PCR result of a sample which was found to be invalid, also yielded negative result by culture. Five out of the seven culture-negative samples were positive for vanA, and two for vanB by the GeneXpert® system. In our study, the sensitivity, specificity, positive and negative predictive values of the GeneXpert vanA/vanB PCR system were determined as 97.4%, 98.4%, 97.4%, and 97.4%, respectively. Although the GeneXpert® vanA/vanB RT-PCR method seems to be more attractive regarding the turn around time, it has a higher cost than the culture method. Thus, it was concluded that all laboratories should choose the most appropriate method for screening VRE in the hospital setting according to their own capacities.