There is an increase in the prevalence of resistance to extended-spectrum cephalosporins among members of the Enterobacteriaceae in both the hospital and community setting. Such cephalosporin resistance is often mediated by the production of extended-spectrum beta-lactamases (ESBLs). It may also be due to AmpC cephalosporinase produced as a result of the acquisition of a plasmid-mediated ampC gene in Escherichia coli and Klebsiella pneumoniae, or alternatively the hyper-production of a chromosomally encoded AmpC enzyme in E. coli which is otherwise expressed at low levels.1
AmpC beta-lactamases are also present on the chromosome of many members of the Enterobacteriaceae, including Enterobacter cloacae and Citrobacter freundii, and can be induced or expressed at high levels due to a mutation that leads to depression of the chromosomal ampC gene (Fig 1). Enterobacteriaceae can also be co-producers of AmpC and ESBL enzymes, although this is rare.
Many laboratories routinely test for ESBLs but do not test for AmpC resistance mechanisms2 because the tests available are time-consuming or difficult to perform and there is no approved standard method for their detection. However, there is a need for a simple diagnostic test for the detection
of these enzymes as chromosomal ampC genes are being disseminated on plasmids, mirroring the early dispersion and evolution of ESBLs.3
Such beta-lactamases can be distinguished from ESBLs by their ability to hydrolyse cephamycins and oxyiminocephalosporins, resulting in restricted therapeutic options and treatment failure.2 Kohner et al.3 described the proportion of ESBLs, AmpCs and concomitant ESBL and ampC genes typically found in clinical isolates (Fig 2).
The D69C AmpC identification test from Mast is based on a combination disc method, which is easy to perform and interpret. It incorporates cefpodoxime as the screening agent, benzo(b)thiophene-2-boronic acid (BZBTH2B) as the AmpC inhibitor, and clavulanic acid as an inhibitor against the activity of ESBLs, which can mask the effect of the AmpC inhibitor.4 This study aims to validate D69C MASTDISCS before commercial release.
Isolates, controls and other considerations
Sixty-five non-replicate cephalosporin-resistant clinical isolates were used in this study (AmpC-producing isolates [n=58], AmpC/ESBL-producing isolates [n=4], KPC-producing isolates [n=2], K1-producing isolate [n=1]).These comprised Escherichia coli (n=11), K. pneumoniae (n=2) Enterobacter cloacae (n=15), E. aerogenes (n=14), C. freundii (n=12), Morganella morganii (n=7), K. oxytoca (n=1) and Serratia marcescens (n=3).
AmpC beta-lactamase gene pools (DHA-1, DHA-2 [DHA]; LAT-1 to LAT-4, CMY-2 to CMY-7, BIL-1 [CIT]; MIR-1T, ACT-1 [EBC]; FOX-1 to FOX5b [FOX]; MOX-1, MOX-2, CMY-1, CMY-8 to CMY-11 [MOX]; ACC [ACC]) were confirmed by multiplex polymerase chain reaction (PCR) assay using the method described by Pérez-Pérez and Hanson5 (Figure 3 shows an example of the gel electrophoresis previously carried out, and Figure 4 shows the distribution of gene pools).
All isolates were screened with Mast D66C (cefpodoxime 10 μg and cefpodoxime 10 μg/clavulanic acid 1 μg) ESBL identification discs and Mast FOX10C (cefoxitin 10 μg) susceptibility discs using the British Society for Antimicrobial Chemotherapy (BSAC) standardised susceptibility method. Blunting of the cefpodoxime/clavulanic acid zone was considered positive for inducible AmpC.
One AmpC control organism was used, K. pneumoniae NEQAS specimen 9215 (from distribution 2454), which produced a plasmid-mediated CIT-type (CMY-like enzymes derived from C. freundii) AmpC beta-lactamase. Four ESBL controls were used (Escherichia coli NCTC 13351 [TEM-3], E .coli NCTC 13352 (TEM-10), E. coli NCTC 13353 [CTX-M-15] and K. pneumoniae ATCC 700603 [SHV-18]). Eight sensitive controls were used (E. coli [n=4], Enterbacter cloacae [n=1], K. pneumoniae [n=2], S. marcescens [n=1]).
A pre-production batch of D69C AmpC identification discs was manufactured at Mast Diagnostics (discs A, B and C). Disc A contained cefpodoxime 10 μg (CPD10), disc B contained cefpodoxime 10 μg (CPD10) plus clavulanic acid 1 μg (CLAV1) and disc C contained cefpodoxime 10 μg (CPD10), clavulanic acid 1 μg (CLAV1) and BZBTH2B 64 μg (BA64).
Plates of Mast isotonic sensitivity test agar (DM604) were inoculated with the test and control organisms to give semiconfluent growth using BSAC methodology. Discs A, B and C were placed on the agar surface and the plates were incubated overnight at 35–37°C. Zones of inhibition were then measured.
A ≥5 mm increase in zone diameter around disc C compared to discs A and B was considered positive for AmpC production. Zones within 2 mm of each other were considered negative.
Test results
Of the 65 cephalosporin-resistant clinical isolates tested, 57 were identified as positive for AmpC production using D69C (Fig 5). Phenotypically, ampC genes were inducible on 13 of these isolates.
Five isolates were recorded as false negatives (E. cloacae [n=2], E. aerogenes [n=3]), all of which possessed intrinsic ampC genes, and one was an AmpC and ESBL co-producer. Three of these were inducible strains. There were two KPC producers and a K1 producer, which were interpreted as negative for AmpC production.
There were no false-positive results and all susceptible control strains (Fig 6) and ESBL control strains were identified as negative for AmpC production (Table 1).
Convenient and effective detection
Given the fact that AmpC resistance is increasing, so also is the need for an effective diagnostic test. In this study, D69C AmpC identification discs proved a reliable indicator of the AmpC cephalosporinase enzyme. Overall sensitivity was 92% and specificity was 100%.
Organisms which produce plasmid-mediated AmpC beta-lactamase are responsible for spreading resistance to other organisms in outbreak situations in hospital environments.6 In the present study, 100% of organisms with plasmid acquisition were detected.
The AmpC-producing isolates not detected gave a susceptible phenotype, showing susceptibility to cefpodoxime. Three of these were strains with an inducible AmpC. Inducible strains can be difficult to detect unless cefoxitin is used as an indicator. This can be a problem in E. cloacae and E. aerogenes infections where isolates originally susceptible to fourth-generation cephalosporins become resistant during therapy.
In conclusion, D69C AmpC identification discs are a simple, cost-effective, convenient and accurate means of detecting AmpC production for organisms exhibiting plasmid-mediated AmpC ?-lactamase and those which have chromosomally encoded AmpC beta-lactamase production.