Detection of resistance to beta-lactamase inhibitors in strains with CTX-M beta-lactamases: a multicenter external proficiency study using a well-defined collection of Escherichia coli strains

Abstract

Under the auspices of the Spanish Society for Infectious Diseases and Clinical Microbiology Quality Control program, 14 Escherichia coli strains masked as blood culture isolates were sent to 68 clinical microbiology laboratories for antimicrobial susceptibility testing to β-lactam antibiotics. This collection included three control strains (E. coli ATCC 25922, an IRT-2 producer, and a CMY-2 producer), six isogenic strains with or without the OmpF porin and expressing CTX-M β-lactamases (CTX-M-1, CTX-M-15, and CTX-M-14), one strain carrying a double mechanism for β-lactam resistance (i.e., carrying CTX-M-15 and OXA-1 enzymes), and four strains carrying CTX-M variants with different levels of resistance to β-lactams and β-lactam-β-lactamase inhibitor (BLBLI) combinations. The main objective of the study was to ascertain how these variants with reduced susceptibilities to BLBLIs are identified in clinical microbiology laboratories. CTX-M variants with high resistance to BLBLIs were mainly identified as inhibitor-resistant TEM (IRT) enzymes (68.0%); however, isogenic CTX-M mutant strains with reduced susceptibilities to BLBLIs and cephalosporins were mainly associated with extended-spectrum β-lactamase production alone (51 to 80%) or in combination with other mechanisms (14 to 31%). Concerning all β-lactams tested, the overall interpretative discrepancy rate was 11.5%, of which 38.1% were the consequence of postreading changes in the clinical categories when a resistance mechanism was inferred. Therefore, failure to recognize these complex phenotypes might contribute to an explanation of their apparent absence in the clinical setting and might lead to inadequate drug treatment selection. A proposal for improving recognition is to adhere strictly to the current CLSI or EUCAST guidelines for detecting reduced susceptibility to BLBLI combinations, without any interpretative modification.

FIG 1

Resistance mechanisms inferred by participating laboratories. Shown are control strains (A), wild-type ESBL-CTX-M-harboring strains (B), and CTX-M variants with reduced susceptibility to BLBLIs (C). Lowercase letters represent the other resistance mechanisms: a, TEM-1/2, OXA-1/IRT, TEM hyperexpression, and porin alteration; b, chromosomal AmpC/TEM hyperexpression. In the CTX-M variant-harboring strains, centers reported ESBL phenotypes despite showing BLBLI resistance.