Endoscopy-associated transmission of carbapenemase-producing Enterobacteriaceae: return of 5 years’ experience

The positive risk–benefit profile of endoscopic interventions has been clearly established, and the risk of nosocomial infection from endoscopes that have been appropriately reprocessed is assumed to be very low [1]. However, despite the absence of recognized reprocessing lapses, Epstein et al. described an endoscopyassociated transmission of NewDelhi-metallo-β-lactamase-producing Escherichia coli [2]. We have previously witnessed an outbreak of Klebsiella pneumonia carbapenemase-2-producing K. pneumoniae (KPC-2-Kp), the source of which was a contaminated endoscope [3]. During the outbreak, 13 patients were directly contaminated via the endoscope (including four infections). Of note, one of the patients, who had a history of recurrent cholangitis related to a biliary stent, remained colonized by the acquired KPC-2-Kp for more than 5 years. He regularly underwent endoscopic retrograde cholangiopancreatography (ERCP) and unclogging of the stent, and died following KPC-2-Kp septicemia. In light of this outbreak, new local guidelines for endoscope reprocessing were established at our hospital: 1) the duodenoscopes used for ERCP were all replaced with a newer model, which incorporated a newly designed elevator channel with no valves, leading to a more efficient disinfection; 2) all duodenoscopes were tested three times per year instead of only once; 3) one duodenoscopewas dedicated to patients who were colonized with carbapenemase-producing Enterobacteriaceae (CPE); and 4) this duodenoscope was sequestered and tested for any bacterial contamination (after its reprocessing) after each endoscopic procedure performed in CPE-colonized patients. As pointed out by Rutala et al., it seems reasonable to reinforce microbiological surveillance of duodenoscopes to assess microbial contamination [4]. However, questions remain, including the most appropriate cutoff measure to define proper disinfection and the actions to be undertaken in case of positive results [4]. Currently, our cutoff is<5 colony-forming units, according to French guidelines (http://www.sante.gouv.fr/IMG//pdf/microbio_endoscopes.pdf). In addition, even if the threshold is not reached, antibiograms are performed on all cultured bacteria. Identification of any multidrug-resistant bacteria, including CPE, extended spectrum-β-lactamase-producing Enterobacteriaceae, or glycopeptide-resistant Enterococcus, results in a repeat disinfection procedure and microbiological retesting. If two successive decontamination procedures fail, the duodenoscope is withdrawn and the reprocessing procedure is audited. The use of this algorithm has resulted in no further transmission being detected. However, it implies a systematic screening of all patients undergoing duodenoscopy, which is difficult to implement. Finally, Rutala et al. encouraged clinicians to report any case of infectious disease related to endoscopy [4]. It is very likely that many outbreaks have been missed in the past because of multi-susceptible bacterial pathogens belonging to the gut flora [5]. However, with the emergence of highly resistant strains, this phenomenon is becoming more visible [2,3 ,5].