Comparison of Matrix-assisted Laser Desorption Ionization Time-of-flight Mass Spectrometry (MALDI-TOF MS) and the Vitek 2 System for Routine Identification of Clinically Relevant Bacteria and Yeast.

OBJECTIVE: In this study, we compared the observed agreement and correlation of the Vitek 2 system with the biomarker-based MALDI-TOF MS identification results of bacteria and yeast on a routine basis. METHODS: Clinical isolates collected from two years were included. Isolates were identified using the Vitek 2 system and MALDI-TOF MS. The percent of observed agreements and the kappa coefficient (kappa) with its corresponding 95% interval confidence were calculated between both results. When species-level biotyper identifications matched a member of a group, complex, or one of the species of a slashing call, the identification was considered correct for agreement calculations. RESULTS: The 4,238 recruited isolates included 2,669 gram-negative bacteria, 1,479 gram-positive bacteria, and 90 yeast. Among gram-negative bacteria, the most frequent species identified were Escherichia coli (kappa=0.983), Acinetobacter baumannii complex (kappa=0.979), Klebsiella pneumoniae (kappa=0.972), and Pseudomonas aeruginosa (kappa=0.970). Among Staphylococcal species, Staphylococcus aureus was the most frequently species detected (kappa=0.986), followed by S. epidermidis (kappa=0.904). For enterococcal species, Enterococcus faecalis (kappa=0.882) and Enterococcus faecium (kappa=0.849) were the most frequently detected. For yeasts, the more common species were Candida albicans (kappa=0.888), followed by Candida tropicalis (kappa=0.946) and Candida glabrata (kappa=1.000). CONCLUSIONS: According to our results, when antimicrobial susceptibility tests are performed using Vitek 2 cards, the most common pathogens are correctly identified for the most frequent clinical isolates.