False resistance to imipenem observed for clinical Clostridium difficile isolates using Etest is partly explained by late sporulation

Introduction and Purpose

Carbapenems are frequently used in treatment of critically ill hospital patients who are at high risk of Clostridium difficile infection (CDI).
The resistance rate in C. difficile, measured by minimal inhibitory concentration (MIC), is generally low using agar dilution method (imipenem 4-8, meropenem 1-4 and ertapenem 4-8 mg/L), and presumably the risk of carbapenems to incite CDI is low.
In contrast, using the Etest (AB bioMerieux, Solna, Sweden), a high level of imipenem resistance (97% of 606 isolates, MIC >32 mg/L) was observed in 1993-2008 in Sweden, which required further evaluation.

 

Methods

The Etest was performed on 20 toxigenic C. difficile isolates, belonging to 10 different PCR-ribotypes, recovered consecutively during January 2008.
Recommended IsoSensitest agar media and three different concentrations of C. difficile inoculum (Mc Farland 1, 2 and 4) were used. The MIC was read after 24 and 48 hours of anaerobic incubation at 37ºC.
Imipenem, ertapenem and meropenem were tested and finally agar dilution was performed for imipenem according to CLSI to confirm our result.
Susceptible controls C. difficile (ATCC 9689), Bacteroides fragilis (ATCC 25285) and C. perfringens (ATCC13124) were included.

Results

All isolates had low MICs (1.5-4 mg/L) determined by Etest
intersection after 24 hours (Table 1). When read according to the manual after 48 hours all but one isolate had acquired highly resistant microcolonies scattered in the inhibition zone (>32) (Fig 1, Table 1) and 7/20 (35%) without visible double zone.
Microcolonies sub-cultured from the resistant zone were not  homogeneously resistant, instead, these gave similar results when recultured four times, i.e. homogenous growth of susceptible bacteria and resistant microcolonies after 48 hours.
The results were similar for all the different ribotypes.

Diluted concentrations of the inoculum (Mc Farland 4 to 2) still read ± one step of MIC on the E-test strip and lower concentration (Mc Farland 1) could not be read safely due to non-confluence distant from zone of resistance.
Agar dilution in Brucella agar, however, displayed susceptible MICs at expected levels to imipenem (1-4 mg/L).
Using the Etest, four isolates were highly resistant to ertapenem while all were susceptible to meropenem according to CLSI guidelines. No double zones or microcolonies were seen when testing these latter carbapenems.

Klicka för större bild

Conclusion

Etest using imipenem for C. difficile may show false resistance due to a resistant sub-population within the inhibition zone of the Etest strip.
A possible explanation might be late germination in inhibition zone in combination with degradation of imipenem and/or excessive inoculum rich in spores. Heteroresistance is less likely since sub-culturing from resistant zones still present a majority of susceptible isolates.
Nevertheless, caution should be made evaluating Etest results for imipenem when testing C. difficile and possibly if encountering unexpected imipenem Etest related resistance in other sporulating pathogens.