A clinically relevant reinterpretation of disk approximation testing for inducible AmpC in Pseudomonas aeruginosa.

Abstract

Introduction: Current antimicrobial susceptibility methods fail to capture the variability of AmpC induction pathways in Pseudomonas aeruginosa. This leads to preventable errors in ceftazidime susceptibility reporting. We introduce a reinterpretation for the disk approximation test, based on the categorical change in ceftazidime susceptibility upon imipenem exposure.

Materials and methods: A total of 73 ceftazidime-susceptible P. aeruginosa isolates were evaluated. Automated broth microdilution test was the reference standard. AmpC inducibility was conventionally defined as a ≥5 mm flattening of the ceftazidime inhibition zone adjacent to imipenem. Our novel criterion for inducibility was a change in ceftazidime category from susceptible to nonsusceptible. The probability of error in ceftazidime reporting was evaluated with three parameters. ROC curve, comparative and regression analyses were performed.

Results: The reference method showed an overall error probability of 35.6%. The conventional 5 mm cutoff interpretation reduced this rate to 5.4%, but generated a significant proportion of false positives and negatives. Regression analyses showed that the flattening effect is a strong predictor of the categorical change. No association was found between basal ceftazidime susceptibility and AmpC inducibility.

Conclusion: Our reinterpretation criterion for disk approximation test turns a phenotypic assay into a clinically relevant diagnostic tool to avoid errors in ceftazidime susceptibility reports. Moreover, this interpretation is readily applicable in routine microbiology laboratories, and has the potential to be directly useful for antimicrobial stewardship efforts.