Validation of automated bacterial suspension preparation by Colibri® and plate streaking by WASP® for antibiotic disk diffusion susceptibility testing.

Background
One of the most crucial responsibilities of clinical microbiology laboratories involves conducting precise and fast antimicrobial susceptibility testing (AST) on bacterial isolates, necessary to guide antibiotic therapy. Standardized disk diffusion, a manual AST method, consumes a significant amount of time and is error-prone. Total lab automation in microbiology should enable a lower workload, high traceability, and standardization in AST. Therefore, we examined the concordance at the categorical level between the manual reference method and a new automated approach for bacterial suspension preparation and plate streaking in AST.
Material and methods
In this study, we validated the automated bacterial suspension preparation by Colibri® and plate streaking by WASP® for antibiotic disk diffusion susceptibility testing. Fifty-three non-duplicate bacterial strains, derived from a variety of different bacterial species, encompassing key known resistance mechanisms and comprising both Gram-positive (N=16) and Gram-negative (N=37) strains, were tested. Both the manual (reference) and the automated (Colibri® with WASP®) method for AST preparation and plate streaking, used the Radian® in-line carousel and expert system for antibiotic susceptibility interpretation. European Committee on Antimicrobial Susceptibility Testing (EUCAST) clinical breakpoints (version 13.1) were used to interpret susceptibility results. The categorization of agreement follows the ISO-20776-2 standard, which includes the identification of minor (mES), major (MEs), and very major errors (VMEs).
Results
The overall categorical agreement between the two compared methods was 96.5% (635/658). We identified 2.6% (17/658) mES, 1.7% (5/658) MEs, and 0.3% (1/658) VMEs. The sole VME occurred during the examination of piperacillin/tazobactam in Escherichia coli. However, it is noteworthy that the zone diameter in the manual method measured was 19 mm, placing it within the “area of technical uncertainty”.
Conclusion
The combination of Colibri® and WASP® appears to be a compelling automated tool for the automated preparation of bacterial suspensions and plate streaking in AST, with an accuracy that is equal to the reference method. Furthermore, it enables the optimization of hands-on time and standardization of (pre-) analytical procedures.