Risk Factors for Transmission of Carbapenem Resistant Enterobacteriaceae (CRE) Infection During Endoscopic Retrograde Cholangiopancreatography (ERCP): 52

Introduction: Duodenoscope associated CRE transmission during ERCP has recently been reported at several hospitals nationwide. However, the specific risk factors for this transmission remain uncertain. We aimed to identify if there are patient or procedure specific factors that are associated with an increased risk of CRE transmission after ERCP. Methods: We retrospectively identified all patients who underwent an ERCP at a single center with one of two subsequently identified CRE-colonized duodenoscopes from October 3, 2014, to January 28, 2015. We compared patients with available results who became infected with CRE (active infection or colonization) with those who did not. Relevant past medical history including previous multidrug resistant infection, cirrhosis, immunosuppression, solid organ transplantation, active cancer, pancreaticobiliary malignancy, prior antibiotic exposure, hospitalization, recent invasive procedures and the duration and number of ERCP procedures undergone were collected. In addition, intra-procedural techniques utilized during ERCP such as sphincterotomy, stent placement, stricture dilation, and cholangioscopy were analyzed. All exposed patients continue to be followed to accurately determine infection rates. Results: During the relevant time period, a total of 121 procedures were performed on 111 patients using one of the two CRE-colonized duodenoscopes. Ten of the patients underwent two ERCP procedures with either of the implicated duodenoscopes. Culture and colonization data was available for 94 (84.7%) of the 111 exposed patients. The mean age was 60.4 years (range: 7-89), and there were 64 (68.1%) men. A total of 14 (14.9%) patients became actively infected (n=8, 8.5%) or colonized (n=6, 6.4%) with CRE. No risk factors were independently associated with an increased risk of CRE infection. However, several risk factors trended towards significance including a history of cholangiocarcinoma (21.4% vs. 5.0%, p=0.065), stent placement (57.1% vs. 31.3%, p=0.074) and the use of single-operator cholangioscopy (21.4% vs. 5.0%, p=0.065). Conclusion: In patients undergoing ERCP, no clear patient or procedure risk factors were identified that increased the risk of CRE transmission. However, ongoing collection of culture data may ultimately prove several of these risk factors to be significant. Given these data, rigorous scope reprocessing protocols to prevent duodenoscope contamination are likely the best strategy to prevent future outbreaks of ERCPrelated CRE infection.