OBJECTIVE To assess the in vitro antimicrobial activity of aztreonam-avibactam against metallo-β-lactamase (MBL)-producing carbapenem-resistant Enterobacteriaceae(CRE) using different antimicrobial susceptibility testing methods, evaluate the consistency of results based on the latest breakpoints recommended by European Committee on Antimicrobial Susceptibility Testing (EUCAST), and analyze the clinical applicability.

METHODS The imipenem- or meropenem-resistant Enterobacteriaceae strains were isolated from the clinical specimens of Beijing Chaoyang Hospital from Jan. 2019 to Mar. 2023, MBL-producing CRE were selected after they were confirmed by colloidal gold immunoassay and Sanger sequencing as the study subjects. The minimum inhibitory concentrations (MICs) of single aztreonam and aztreonam-avibactam compounds were determined by using broth microdilution method. In addition, the disk diffusion method and the gradient diffusion method were employed to further detect the in vitro susceptibility to aztreonam-avibactam.

RESULTS Among the 87 strains of MBL-producing CRE that were included in the study, Escherichia coli was the most common species, accounting for 44.83%. NDM-5 was the predominant carbapenemase type, detected in 48.28% of the isolates. According to the latest EUCAST breakpoints, 6 isolates were aztreonam-avibactam-resistant strains based on broth microdilution method, all of which were E. coli, and the resistance rate was 6.90% (6/87). However, the resistance rate that was determined by the disk diffusion method was significantly higher (22.99%, 20/87). Among these, 14 strains was within the area of technical uncertainty/resistant, making result interpretation difficult. In addition, the categorical agreement between the gradient diffusion method and the broth microdilution method reached up to 98.85%.

CONCLUSIONS Aztreonam-avibactam has high antimicrobial activity against MBL-producing CRE. However, based on the latest EUCAST breakpoints, the antimicrobial susceptibility testing by the disk diffusion method results of some strains are hard to interpret. It is necessary to integrate with other methods for further validation.