Poster Presentation Australian Society for Microbiology Annual Scientific Meeting 2019

Comparison of CDS disc diffusion and Thermo Fisher SensititreTM broth microdilution methods using Enterobacteriaceae strains with established ISO broth dilution MICs (#141)

Pratibha Malini James 1 , Dianne Rafferty 1 , Julie Allerton 1 , Sydney Bell 1
  1. CDS Reference laboratory, Department of Microbiology, NSW Health Pathology, St George Hospital, Sydney, NSW, Australia


Antibiotic resistance in Enterobacteriaeceae family is on the rise and accurate reporting of the antimicrobial susceptibility is important as it has a direct impact on patient response to treatment. A study was undertaken using seventeen members of Enterobacteriaceae family against eleven antibiotics belonging to six classes. The Calibrated Dichotomous Susceptibility (CDS) disc diffusion method and the Thermo Fisher SensititreTM micro broth dilution methods were compared with the MIC results of the gold standard, ISO broth dilution (performed by two independent laboratories attached to the United Kingdom National External Quality Assessment Scheme). SensititreTM GNX2F plates were used for the study and the antibiotics which were common to all the three methods were cefotaxime, ceftazidime, piperacillin- tazobactam, amikacin, gentamicin, tobramycin, ciprofloxacin, ertapenem, imipenem, meropenem and tigecycline. Categorical agreement (susceptible or resistant) was used to calculate the concordance with CDS testing. Essential agreement was used for broth dilutions, with consensus defined as MICs that were within ± 1 doubling dilution of the MIC determined by the ISO broth dilution. A discordance was categorised as a minor error when a susceptible isolate was reported as intermediate/resistant and a major error when an intermediate/resistant isolate was reported as susceptible. The CDS test correlated well with the reference standard with no major errors. The SensititreTM test correlated well for all antibiotics except with cefotaxime, where there was a major error.