BACKGROUND: The aim of this study was to compare the BD Phoenix (Beckton Dickinson Diagnostic Systems, USA) extended-spectrum beta-lactamase (ESBL) test with the Clinical and Laboratory Standards Institute (CLSI) ESBL phenotypic confirmatory test by disk diffusion (CLSI ESBL test) in Escherichia coli, Klebsiella pneumoniae, Klebsiella oxytoca and Proteus mirabilis. METHODS: We tested 224 clinical isolates of E. coli, K. pneumoniae, K. oxytoca and P. mirabilis during May 2006 to March 2007. These isolates were examined by the Phoenix and the CLSI ESBL tests simultaneously. For the isolates showing discordant results between the two tests, boronic acid disk test was performed to differentiate AmpC beta-lactamase and ESBL. RESULTS: Among the 224 clinical isolates, 75 and 79 isolates were positive for ESBL by CLSI ESBL test and Phoenix test, respectively. Having detected 4 more isolates as ESBL-producers, Phoenix test showed a 98.2% agreement with a 100% sensitivity and 97.3% specificity compared with CLSI ESBL test. Among the four false positive isolates, three were AmpC-positive but ESBL-negative. CONCLUSIONS: The BD Phoenix ESBL test was sensitive and specific, and can be used as a rapid and reliable method to detect ESBL production in E. coli, Klebsiella species, and P. mirabilis.
BACKGROUND: The aim of this study was to compare the BD Phoenix (Beckton Dickinson Diagnostic Systems, USA) extended-spectrum beta-lactamase (ESBL) test with the Clinical and Laboratory Standards Institute (CLSI) ESBL phenotypic confirmatory test by disk diffusion (CLSI ESBL test) in Escherichia coli, Klebsiella pneumoniae, Klebsiella oxytoca and Proteus mirabilis. METHODS: We tested 224 clinical isolates of E. coli, K. pneumoniae, K. oxytoca and P. mirabilis during May 2006 to March 2007. These isolates were examined by the Phoenix and the CLSI ESBL tests simultaneously. For the isolates showing discordant results between the two tests, boronic acid disk test was performed to differentiate AmpC beta-lactamase and ESBL. RESULTS: Among the 224 clinical isolates, 75 and 79 isolates were positive for ESBL by CLSI ESBL test and Phoenix test, respectively. Having detected 4 more isolates as ESBL-producers, Phoenix test showed a 98.2% agreement with a 100% sensitivity and 97.3% specificity compared with CLSI ESBL test. Among the four false positive isolates, three were AmpC-positive but ESBL-negative. CONCLUSIONS: The BD Phoenix ESBL test was sensitive and specific, and can be used as a rapid and reliable method to detect ESBL production in E. coli, Klebsiella species, and P. mirabilis.