INTRODUCTION AND PURPOSE
Detection of beta-lactam resistance in Escherichia coli and Klebsiella pneumoniae strains is clinically relevant. Moreover, it is important to differentiate between extended-spectrum beta-lactamase (ESBL) production and other mechanisms of resistance to avoid inadequate treatment of infection caused by these strains. The aim of this study was to compare the performance of the Vitek 2 and BD Phoenix automated systems for confirmatory testing of ESBL production.
MATERIAL AND METHODS
A total of 193 clinical isolates of phenotypically confirmed ESBL producers (174 E. coli and 19 K. pneumoniae) were assayed by the Vitek 2 and BD Phoenix systems using AST-N058 cards and UNMIC/ID-62 panels, respectively. The double-disk synergy test and the Etest were used as phenotype reference methods. Twelve strains characterized by genotyping were used as positive and negative controls.
RESULTS
In the clinical isolates, the sensitivity of the tests was 99.5% for Vitek and 95.3% for Phoenix. There were no significant differences between the 2 systems in the control strains. Execution of the expert system raised the sensitivity of Phoenix to 100%. However, the Vitek 2 expert system considered the results obtained in 7 strains with ESBL-positive tests to be incoherent.
CONCLUSION
Confirmatory testing for ESBL production with the Vitek 2 system (AST-N058 card) showed higher sensitivity than the Phoenix (UNMIC-ID 62 panel) system. Nevertheless, the performance of the expert systems in the 2 automated tests was similar for ESBL detection in E. coli and K. pneumoniae.