Detection of plasmid-mediated AmpC lactamases in Enterobacteriaceae lacking chromosomal ampC genes using the Rosco double disc synergy test and the Etest

Abstract number: P2065

Adler H., Hohler D., Bruderer T., Frei R.

Objectives: to evaluate 2 commercial tests namely the Rosco double disc synergy test (Rosco Diagnostica, Taastrup, Denmark) and the Etest (AB bioMerieux, Solna, Sweden) to detect plasmid-mediated AmpC b-lactamases in Enterobacteriaceae lacking chromosomal ampC genes.

Methods: A total of 48 isolates of Enterobacteriaceae (41 Klebsiella pneumoniae, 4 Klebsiella oxytoca, 2 Salmonella Newport, 1 Proteus mirabilis) were studied. They comprised 12 isolates with plasmid-mediated AmpC b-lactamases of the CIT-, DHA-, and FOX-type, and 12 isolates producing extended-spectrum b-lactamases (ESBLs) of the CTX-M and SHV-type. Except for ESBL-producers only cefoxitin-resistant isolates were included in the study. All isolates were tested for AmpC b-lactamase production with the following tests: 1) the Rosco double disc synergy test which is based on the synergy between ceftazidime or cefoxitin and cloxacillin, and ceftazidime+clavulanic acid or cefotaxime+clavulanic acid and boronic acid, 2) a new Etest for AmpC detection containing cefotetan/cefotetan + cloxacillin (a MIC ratio of 8 or a deformation of the ellipse was interpreted as positive, a MIC ratio of <8 or a cefotetan MIC of 0.5 mg/l was interpreted as negative).

Results: The Rosco synergy test detected 11 of 12 AmpC producers. In these isolates, at least 3 combinations of antibiotics showed a synergy phenomenon. One isolate harbouring an ampC gene produced double inhibition zones that were not interpretable and was thus omitted from the statistical evaluation. Of 36 isolates without AmpC, 34 (94%) showed no synergy phenomenon with any combination of antibiotics, while 2 showed a synergy phenomenon with only one combination. When interpreting only 2 or more positive synergy tests as indicative of an AmpC b-lactamase, sensitivity and specificity of the test was 100%. The Etest detected 10 of 12 AmpC producers. The MIC ratios of 3 isolates (2 isolates harbouring ampC and one without ampC) were not interpretable and thus omitted from the statistical evaluation. One AmpC-negative isolate gave a false positive result with a MIC ratio of >64. Sensitivity and specificity of the Etest were 100% and 97%, respectively.

Conclusions: Both the Rosco double disc synergy test and the Etest proved to be useful tools to detect plasmid-mediated AmpC b-lactamases