Failure of Vitek2 to reliably detect vanB-mediated vancomycin resistance in Enterococcus faecium.

OBJECTIVES: The increasing prevalence of VRE necessitates their reliable detection, especially for low-level resistance mediated by vanB in Enterococcus faecium. In this prospective study we analysed if vanB-mediated vancomycin resistance can be reliably detected by Vitek2. METHODS: One thousand, three hundred and forty-four enterococcal isolates from routine clinical specimens were tested by Vitek2 (bioMérieux, Nürtingen, Germany). Additionally, a bacterial suspension (with a turbidity equivalent to that of a 0.5 McFarland standard) was inoculated on chromID VRE screening agar (bioMérieux) and incubated for 48 h. If vancomycin tested susceptible by Vitek2 but growth was detected on the screening agar, PCR for vanA/vanB was performed (GeneXpert vanA/B test, Cepheid, Frankfurt, Germany). For isolates that tested susceptible to vancomycin by Vitek2 but were vanA/B positive, MICs were determined before and after cultivation in broth with increasing concentrations of vancomycin. RESULTS: One hundred and fifty-six out of 491 E. faecium were VRE and were predominantly vanB positive (81.0%). Of these, Vitek2 did not identify 14 as VRE (sensitivity 91.0%). By broth microdilution 9/14 isolates demonstrated high MICs (≥32 mg/L) and 5/14 showed low vancomycin MICs, which did not increase despite vancomycin exposure. Three of the 14 isolates demonstrated growth on chromID VRE; after vancomycin exposure seven additional isolates were able to grow on chromID VRE. CONCLUSIONS: Vitek2 fails to detect vanB-mediated vancomycin resistance consistently, especially, but not limited to, low-level resistance. As this may lead to treatment failure and further dissemination of vanB VRE, additional methods (e.g. culture on VRE screening agar or PCR) are necessary to reliably identify vanB-positive enterococci in clinical routine.