Performance of disc diffusion, MIC gradient tests and Vitek 2 for ceftolozane/tazobactam and ceftazidime/avibactam susceptibility testing of Pseudomonas aeruginosa

Objectives: To assess performance of disc diffusion, gradient tests and Vitek 2 system to determine the susceptibility of clinical Pseudomonas aeruginosa strains to ceftolozane/tazobactam (C/T) and ceftazidime/avibactam (CZA). Methods: Two-hundred non-duplicate P. aeruginosa strains isolated by 47 French medical laboratories were selected to cover a wide range of C/T and CZA MICs. Performance of C/T disc (30/10 mu g, Bio-Rad), CZA discs (10/4 mu g) (Thermo Fisher and Bio-Rad), C/T and CZA gradient tests (Etest, BioMerieux; MIC Test Strip, Liofilchem), and AST-XN12 card of Vitek 2 system (BioMerieux) were compared with a broth microdilution (BMD) method (Thermo Fisher). MIC and disc results were interpreted using current EUCAST breakpoints. Results: Twenty percent and 17% of strains were resistant to C/T and CZA, respectively. All the methods tested satisfactorily determined the susceptibility of P. aeruginosa to C/T [Category Agreement (CA) >= 95%] except the disc diffusion method. Because of the high rates of Major Errors (MEs) (12.5%), this latter method tends to overestimate the resistance. For CZA, only the gradient tests yielded more than 90% of CA. The Vitek 2 system and disc diffusion misclassified 18.1%, 10.1% (disc Bio-Rad) and 11.9% (disc Thermo Fisher) of susceptible strains, respectively. Conclusions: The gradient tests (MIC Test Strip and Etest) and Vitek 2 card XN12 performed the best to determine the susceptibility of P. aeruginosa to C/T, whereas gradient tests were an acceptable alternative to BMD to assess CZA susceptibility.