Assessment of the Phoenix system and EUCAST breakpoints for antimicrobial susceptibility testing against contemporary isolates expressing relevant resistance mechanisms

Abstract number: P577

Objective: To evaluate the accuracy of the PHOENIX system (BD, USA) in combination with the EUCAST breakpoints for antimicrobial susceptibility testing (AST) against contemporary clinical isolates expressing relevant resistance mechanisms.

Methods: A total of 393 isolates were included: i) 200 fresh consecutively recovered isolates (51 Enterobacteriaceae, EB; 40 P. aeruginosa, PA; 30 S. aureus, SA; and 29 Enterococcus spp., E) and ii) 193 stored (-80°C) isolates with characterized resistance mechanism [96 extended-spectrum-(ESBL) and 50 metalo-(MBL) b-lactamase producing EB, 10 MBL-producing PA, 20 methicillin-resistant SA, MRSA, and 17 vancomycin-resistant E, VRE]. Comparator AST results were either routine or historical data obtained following CLSI guidelines (broth microdilution). Discrepancies were resolved using Etest. Quality control ATCC strains were used to assure MIC reproducibility. MBL- and ESBL-coding genes and mecA and van determinants were confirmed by PCR and sequencing. PHOENIX performance was assessed based on MICs concordances (essential agreement, EA, ±1 log2 dilution) and percentage of interpretive category error rates (ER) (minor, mi; major, M; very major, VM).

Results: An overall EA of 96% (3,283 organism-antimicrobial combinations) and global 2.4% mi, 1.2% M and 1.1% VM errors were observed. Among fresh isolates, EA was: 97% (97% EB; 94% PA; 98% SA; and 98% E). Global ER in these isolates were: 0.8% mi; 0.6% M; and 2.3% VM, being this latter value mainly due to discrepancies in the combination gentamicin-PA. In isolates with characterized resistance mechanisms, EA was: 95% (99% ESBL-EB; 94% MBL-EB; 100% MBL-PA; 97% SARM; and 91% EVR) Overall ER in this subset of isolates were: 3.5% mi; 2.0% M; and 0.8% VM. In these isolates, mi errors were mainly due to discrepancies in ciprofloxacin and ESBL-EB as well as in imipenem, meropenem and ciprofloxacin and MBL-EB whereas M errors were mainly attributed to amoxicillin–clavulanic acid-ESBL-EB discrepancies.

Conclusion: This study indicates reliable AST results of PHOENIX system and accurate interpretive categorization when using EUCAST breakpoints in contemporary isolates expressing relevant resistance mechanisms. Interpretive category discrepancies were mainly due to particular organism-antimicrobial combinations due to specific breakpoint definition such as gentamicin and PA or heterogeneous expression of resistance mechanisms such MBL affecting carbapenems in EB.

Session Details

Date: 10/04/2010
Time: 00:00-00:00
Session name: Abstracts 20th European Congress of Clinical Microbiology and Infectious Diseases
Location: Vienna, Austria, 10 - 13 April 2010
Presentation type:
Back to top