Despite all the concern about the emergence of CRE pathogens worldwide, it is quite difficult to obtain hard quantitative, prospectively collected incidence figures.
A recent article by Robert [1] and colleagues provides useful data from France. Using a practical definition of carbapenem-non-susceptibility (ertapenem > 0.5, imipenem, meropenem > 2, doripenem > 1), they analyzed the antibiograms of > 133,000 isolates of enterobacteriaceae from 71 French laboratories between Nov. 2011 – April 2012.
The main results were as follows: 0.6% of all isolates met above CRE definitions, with Enterobacter cloacae being the predominant pathogen (58%). Expressed differently, 8.2% of all Enterobacter cloacae isolates tested were resistant to at least 1 carbapenem. There were 5.6% ESBL producers overall, and 2/3 of all CRE were ESBL producers.
Of note, resistance to carbapenem antibiotics is multifactorial and only approx. 10% of these isolates produced carbapenemase by genomic testing. Most were from the OXA-48 group, only 2 were NDM isolates.
The authors note that their incidence figure of 0.6% for France is lower than what has been reported for Belgium (3.5%) but this may be due to methodological differences.
Of course, definition for what constitutes “CRE resistance” is key here as it can affects incidence figures drastically. Remember, break points for carbapenem antibiotics have changed significantly in the past 2 years. CLSI and FDA are supposed to be in sync but are not always, and EUCAST also has definitions that are ever so slightly different from EMA.
Vive la difference? Not to worry: ID clinicians can handle small difference in breakpoints and live with the existing wobble of 1-2 dilution steps. It is actually quite amazing that the differences are so small.
So let’s not fuss too much over lack of harmonization: by the time harmonization occurs, the resistance situation will have changed anyway. Call it the microbe’s revenge or the “ICH uncertainty principle’ à la Heisenberg.
Reference:
[1] Robert. J Antimicrob Chemother 2014; 69: 2706
