Background: In 2010, the Clinical and Laboratory Standards Institute (CLSI) revised breakpoints for cephalosporins and carbapenems and indicated that extended-spectrum β-lactamase (ESBL) testing is no longer necessary for Enterobacteriaceae. We compared the results of the CLSI 2010 and the European Committee on Antimicrobial Susceptibility Testing (EUCAST) MIC breakpoints for Enterobacteriaceae producing ESBL and/or plasmid-mediated AmpC β-lactamase (PABL). 

Methods: A total of 94 well-characterized clinical isolates of Escherichia coli, Klebsiella oxytoca, Klebsiella pneumoniae, Proteus mirabilis, Salmonella spp., Shigella spp., Citrobacter freundii, Enterobacter aerogenes, Enterobacter cloacae, and Serratia marcescens were analyzed. Of them, 57 were ESBL producers, 24 were PABL producers, and 13 were ESBL plus PABL co-producers. Broth microdilution MIC tests were performed for cefotaxime, ceftazidime, aztreonam, cefepime, and imipenem. 

Results: Among the 94 isolates containing ESBL and/ or PABL, the number of isolates that were susceptible to cefotaxime, ceftazidime, aztreonam, cefepime, and imipenem according to the CLSI 2010 vs. the EUCAST breakpoints were 4 (4.3%) vs. 4 (4.3%); 26 (27.7%) vs. 8 (8.5%); 37 (39.4%) vs. 14 (14.9%); 71 (75.5%) vs. 31 (33.0%); and 76 (80.9%) vs. 90 (95.7%), respectively. Of the 18 isolates that were not susceptible to imipenem according to the CLSI 2010 breakpoints, 13 isolates (72.2%) were P. mirabilis.

Conclusion: The CLSI 2010 MIC breakpoints without tests to detect ESBL and/or PABL for Enterobacteriaceae could be unreliable. Thus, special tests for ESBLs and AmpC β-lactamases are required to detect the resistance mechanisms involved. (Korean J Clin Microbiol 2011;14:24-29)