Comparison of Two Culture Methods for Use in Assessing MicrobialContamination of Duodenoscopes

Michaela A. Gazdik,a Jana Coombs,a John P. Burke,a,b Bert K. Lopansria,b

Department of Clinical Epidemiology and Infectious Diseases, Intermountain Medical Center, Murray, Utah, USAa; Division of Infectious Diseases, University of Utah,Salt Lake City, Utah, USAb

Recent outbreaks of carbapenem-resistant Enterobacteriaceae infections associated with duodenoscopes used for endoscopicretrograde cholangiopancreatography have highlighted the challenge of cleaning and high-level disinfection of these instru-ments. The Food and Drug Administration has suggested that duodenoscope surveillance by microbiological culturing, alongwith strict adherence to reprocessing protocols, may help reduce the risk of duodenoscope-associated infection transmission.We developed and validated an effective, user-friendly duodenoscope sampling and culture protocol and compared its perfor-mance to the interim Centers for Disease Control and Prevention–recommended guidelines. Our protocol resulted in a 65% re-covery rate for Gram-negative organisms, demonstrating a 2-fold increased recovery rate compared to the CDC method. Theimplementation of this protocol may increase the feasibility of duodenoscope surveillance for microbiology laboratories andendoscopy departments.

Endoscopic retrograde cholangiopancreatography (ERCP) isused as a diagnostic and therapeutic tool for many pancreatic

and biliary diseases, providing a less invasive treatment optionthan surgery. There are currently more than 500,000 ERCP pro-cedures performed in the United States annually (1). Duodeno-scopes used for ERCP have been linked to serious postprocedureinfections, including at least 11 clusters of antibiotic-resistant En-terobacteriaceae infections over the past 7 years (2–11). The Foodand Drug Administration (FDA) has received notification of 142cases of patient infection or exposure from reprocessed duodeno-scopes since 2010 (12).

Epidemiological investigation of recent ERCP-associated anti-biotic-resistant Enterobacteriaceae outbreaks revealed that manywere not caused by reprocessing deficiencies or detectable devicedefects (8, 11, 13). Instead, these clusters have been attributed tothe complex design of the duodenoscope, which includes an in-strument/suction channel with a cantilevered elevator mechanismthat allows for the manipulation of equipment during procedures(12, 13). This elevator mechanism contains microscopic crevicesthat can retain organic debris and fluid despite brushing andcleaning prior to high-level disinfection. As a result, the FDA hasrecommended supplemental measures to the manufacturers’ rec-ommendations as a means to help reduce the risk of infectiontransmission. These include microbial culturing, ethylene oxidesterilization, use of liquid chemical sterilant processing, and re-peat high-level disinfection (14).

Performing surveillance cultures of endoscopes is a controver-sial topic in endoscopy infection control but is promoted by somegastroenterological societies as a quality-control marker of theadequacy of cleaning and disinfection in an endoscopy unit and tomonitor the integrity of an endoscope (15–21). The Gastroenter-ological Society of Australia and the European Society of Gastro-enterology (ESGE)–European Society of Gastroenterology andEndoscopy Nurses and Associates (ESGENA) have published for-mal guidelines for endoscope surveillance (18, 21). Additionally,the Emergency Care Research Institute recently published a high-priority hazard report recommending culturing of duodeno-scopes as a key step in reducing carbapenem-resistant Enterobac-

teriaceae infections (22), and the FDA has recommendedsupplemental measures to enhance reprocessing with microbio-logical culture as one of the suggested measures (14). While theCenters for Disease Control and Prevention (CDC) published in-terim guidelines for duodenoscope culturing in 2015 (23), thesehave not been rigorously investigated, and the performance char-acteristics are unknown.

We developed and validated an alternate duodenoscope sam-pling and culture protocol and compared its performance to theCDC-recommended method. Our primary objectives were to de-velop an effective, user-friendly method and to understand limi-tations in sample collection and culture.

MATERIALS AND METHODSSoiling of the ERCP duodenoscope. Two Olympus TJF duodenoscopes(model 160VF and model Q180V) were used for culture validation testingin the laboratory. Duodenoscopes were reprocessed between tests by theendoscopy department at LDS Hospital in Salt Lake City, Utah, followingstandard operating procedures. Artificial test soil plus mucin (ATS-M)(Healthmark) was rehydrated with sterile water following manufacturerspecifications. A mixed culture ATS-M inoculum was prepared with ap-proximately 5 � 108 CFU/ml of each of the following clinical isolates:Escherichia coli, Klebsiella pneumoniae, Pseudomonas aeruginosa, and En-terococcus faecium. These organisms were selected because they are con-sidered high-concern or indicator organisms. Serial dilutions of 10�2,10�4, 10�5, and 10�6 of ATS-M inoculum were generated and, along withthe original 5 � 108 concentration, used to artificially contaminate theduodenoscopes. The elevator mechanism at the distal end of the endo-

Received 12 October 2015 Returned for modification 28 October 2015Accepted 17 November 2015

Accepted manuscript posted online 18 November 2015

Citation Gazdik MA, Coombs J, Burke JP, Lopansri BK. 2016. Comparison of twoculture methods for use in assessing microbial contamination of duodenoscopes.J Clin Microbiol 54:312–316. doi:10.1128/JCM.02754-15.

Editor: D. J. Diekema

Address correspondence to Michaela A. Gazdik, michaela.gazdik@imail.org.

Copyright © 2016, American Society for Microbiology. All Rights Reserved.

crossmark

312 jcm.asm.org February 2016 Volume 54 Number 2Journal of Clinical Microbiology

scope was partially raised to a 90-degree angle. Then, 100 �l of ATS-Minoculum was spotted above and below the elevator mechanism and al-lowed to dry for 30 min, based on manufacture specifications (24). Twomethods of endoscope sampling were evaluated for percent recovery andease of use. Duodenoscopes were also cultured before inoculation andafter reprocessing as controls.

Modified ESGE protocol for duodenoscope sampling. We used amodified version of the ESGE guidelines (18) but sampled only the instru-ment channel (Fig. 1). Briefly, 5 ml of sterile saline (0.9%) was flushedthrough the instrument channel via the instrument port and collected in asterile 15-ml conical tube. The elevator mechanism and lens face werethen vigorously swabbed with a Floqswab (COPAN), being sure to swabboth over and under the elevator mechanism as it was set in its open andclosed positions. The swab was broken off below the handling point intothe 5 ml of sterile saline previously flushed through the instrument chan-nel and vortexed for 2 min in 10-s bursts. Samples were diluted and platedin triplicate onto tryptic soy agar (TSA) with 5% sheep blood and incu-bated at 35°C for 24 h. Five experiments (three with model 160VF and twowith model Q180V) were performed for each dilution, for a total of 25samplings using this collection method.

Interim CDC protocol for duodenoscope sampling. The CDC in-terim sampling and quantitative culturing guidelines were followed, andthree experiments (two with model 160VF and one with model Q180V)were performed for each dilution, for a total of 15 samplings. Briefly, the

elevator mechanism and lens face at the distal end of the duodenoscopewas brushed with a duodenoscope channel-opening brush (OlympusMH-507) with the mechanism in both the lowered and raised positions.The brush was placed into sterile phosphate-buffered saline with 0.02%Tween 80 (PBST; Teknova P3875) and vortexed for 2 min in 10-s bursts.Then, 50 ml of sterile water was flushed through the instrument channelvia the instrument port and collected in a sterile 50-ml conical tube. Thebrush and channel fluids were processed separately by centrifuging at4,200 � g for 15 min and resuspended in 1 ml of either PBST or sterilewater, respectively. Samples were diluted and plated in triplicate onto TSAwith 5% sheep blood and incubated at 35°C for 24 h. CFU counts from thebrush and fluid were combined to calculate the CFU per scope.

Determining sampling efficiency. The four bacterial species in theATS-M inoculum are phenotypically distinct, enabling all four species tobe identified morphologically on TSA with 5% sheep blood and countedindividually. The CFU/ml count from all duodenoscope-collected fluidwas determined for each of the four bacterial species and used to calculatethe CFU per scope. Additionally the ATS-M inoculum was diluted, platedonto TSA with 5% sheep blood, and incubated at 35°C for 24 h to obtainthe accurate CFU/ml. This value was used to calculate the CFU spotted onthe duodenoscope. The CFU per scope was compared to CFU spotted todetermine sampling efficiency. ATS-M without bacteria was cultured as anegative control. Student’s t test was used to compare sampling efficiencybetween sampling methods.

FIG 1 Pictorial diagram illustrating the MEP sampling method.

Comparison of Duodenoscope Culture Methods

February 2016 Volume 54 Number 2 jcm.asm.org 313Journal of Clinical Microbiology

RESULTS

To compare the sampling efficiency of the modified ESGE proto-col (MEP) and CDC protocol, we calculated that the distal end ofduodenoscopes had been spotted with an average of 10, 100, 103,105, and 107 CFU of mixed culture. Sampling duodenoscopes witheither the MEP or CDC sampling method indicated a similar per-cent recovery when comparing the 160VF and Q180V duodeno-scope models (data not shown). The similarity in recovery is un-surprising, as both sampling methods focus on the instrumentchannel and distal end elevator, which are structurally similar be-tween models. Sampling was not performed on the open guidewire channel which is found in the 160VF model but not Q180Vmodel. Data from the 160VF and Q180V were combined for com-parison of the MEP and CDC sampling methods.

Ten CFU/scope was found to be below the level of accuratedetection for both protocols, as the agar plates frequently showedno growth after sampling (CDC, 4 of 9; MEP 5 of 15). At 100CFU/scope, the percent recovery of Gram-negative organisms wassignificantly enhanced (P � 0.0009) using the MEP (64.1%) com-pared to the CDC protocol (32.9%) but was equivalent for Gram-positive organisms, with approximately 60% recovery for bothmethods. At 107 CFU/scope, the recovery of Gram-negative or-ganism increased for both protocols; however, a statistical differ-ence (P � 0.03) between percent recovery with MEP (80.6%) andCDC (47.4%) remained. Again, recovery of Gram-positive organ-isms remained consistent for both protocols at approximately60% (Fig. 2).

Focusing on endoscopes soiled with 100 CFU, we examined thepercent recovery of individual bacterial species. Of the four bac-terial species, recovery of P. aeruginosa was greatest using MEP,followed by similar rates of both K. pneumoniae and E. faecium. E.coli had the lowest recovery rate of the four organisms tested (Ta-ble 1). Using the CDC protocol, the percentage of P. aeruginosarecovered was highest of the spotted Gram-negative organisms

but was 34% less than the MEP protocol. Recovery of K. pneu-moniae and E. coli was lowest using the CDC protocol. The percentrecovery of E. faecium was similar by both protocols (Table 1).

We observed that the brush recommended by the interim CDCprotocol did not completely reach the entire surface of the elevatorin either the raised or the lowered position. To visually comparethe efficacy of the brush versus the Floqswab (Copan) at samplingthe elevator mechanism, we coated the distal end of an ERCPduodenoscope with GlitterBug potion (Brevis) before samplingand after application of the swab or brush. The residual GlitterBugcoating the elevator mechanism was qualitatively lower by visualinspection after using the flocked swab compared to the brush(Fig. 3). The flocked swab also proved to be simple to snap off intothe collection vial, minimizing risk of contaminating the culture.

DISCUSSION

The American Society of Microbiology Public and Scientific Af-fairs Board Committee on Laboratory Practices has recom-mended that clinical microbiology labs do not perform routinesurveillance of duodenoscopes due to a lack of validation andinappropriateness of techniques, equipment, and expertise for aclinical microbiology lab (25). Due to uncertainties surroundingthe optimal culturing methods, this study compared the perfor-mance of a modified ESGE sampling and culturing method to theCDC interim protocol. While both methods demonstrated qual-itative recovery of spiked organisms, the MEP method demon-strated several advantages over the interim CDC sampling proto-col and may increase the feasibility of obtaining qualitativeduodenoscope surveillance cultures at endoscopy departments.

Sampling using the MEP led to a 2-fold increase in recovery ofGram-negative organisms compared to the CDC protocol. Thishigher recovery rate is likely due to the use of the flocked swab forcollecting samples from under and around the elevator mecha-nism as opposed to use of the Olympus cleaning brush. Flockedswabs are designed to capture and release organisms, and they areable to fit under the raised elevator mechanism more efficientlythan the recommended cleaning brush. The bristles do not extendaround the head of the brush, limiting complete sample collectionin tight spaces. The brush is able to fit more effectively into theinstrument channel for which it was designed. However, the ele-vator mechanism has been identified as the likely structure forreprocessing failures, making it more likely than the channel toharbor organisms after an unsuccessful cleaning and reprocessing.Olympus has recent designed a smaller brush for improved eleva-tor cleaning to address this problem.

FIG 2 Percent of microorganisms recovered after sampling duodenoscopes. A mixed culture of E. coli, K. pneumoniae, P. aeruginosa, and E. faecium was spottedonto the distal end of a duodenoscope at a concentration of 100 CFU (A) or 107 CFU (B). Duodenoscopes were sampled with the MEP (black bar) and CDC(white bar) protocols, and the percentage of organisms recovered was calculated. *, P � 0.05; **, P � 0.001.

TABLE 1 Percent recovery of individual species after inoculating aduodenoscope with 100 CFU of each species in a mixed culture

Organism

% recovery � SD by protocol:

PMEP CDC

P. aeruginosa 80.3 � 23.5 46.2 � 12.6 0.04E. coli 46.0 � 13.0 25.6 � 7.8 0.07K. pneumoniae 66.0 � 9.7 32.1 � 3.2 0.001E. faecium 67.2 � 15.6 60.2 � 4.2 0.73

Gazdik et al.

314 jcm.asm.org February 2016 Volume 54 Number 2Journal of Clinical Microbiology

During the evaluation, we also observed instances in the CDCprotocol where contamination could be introduced into the sam-pling fluid. The use of a 50-ml flush for the instrument channelrequires two people to complete the protocol, as a 60-ml syringe istoo cumbersome to use while also holding the scope and collectingthe fluid. The large volume combined with the need to coordinatecollection between two individuals may increase the possibility ofspillage or overflow as the fluid is being passed into the channel.An advantage of small flush volume of the MEP is that a 5-mlsyringe allows for single-person collection when necessary, withminimal contamination risk.

Another potential limitation to the CDC protocol that we ob-served was handling of the Olympus brush following use, as theentire brush is dropped into the collection fluid, including thehandle that had been held during collection. The protocol requiresthat sterile gloves must be worn; however, the collector has likelybeen handling the collection cup, endoscope, and other items andmay introduce contaminants into the sample fluid from the brushitself. The flocked swab prevents this possibility, as it is broken offinto the sample fluid below the handling point. This also allows forthe use of nonsterile gloves, which is more realistic for completingsampling in an endoscopy department instead of in a dedicatedmicrobiology laboratory.

There are also limitations to the MEP and this study. This in-vestigation did not include soiling of the instrument channel,which could also contain microbial contamination postproce-dure. The larger flush volume used in the CDC protocol may bemore efficient at removing contamination from the channel.Therefore, the CDC protocol may yield a higher percent recoveryunder actual in-use conditions than we report here. The ESGE-ESGENA guidelines suggest neutralizer as an additive to the sterilesaline to neutralize any chemicals remaining from reprocessingthat may inhibit microbial growth. Further testing should be doneto determine if the addition of neutralizer increases percent recov-

ery, which would be necessary for scopes contaminated with verylow levels of organisms. Additionally, this study was completed bytrained microbiologists in a laboratory environment using loanedduodenoscopes that were not in clinical circulation. Samplingmay be more inefficient if performed by nonmicrobiologicallytrained individuals, such as endoscopy technicians in an endos-copy department. It is imperative that any individual who is sam-pling a duodenoscope for culture be trained in proper techniques.Finally, because the duodenoscopes used for our study were infre-quently used clinically, they may lack potential biofilm. The lack ofbiofilm may not fully simulate conditions of duodenoscopes inactual use, as incomplete removal of biofilm would create condi-tions for bacterial to thrive despite high-level disinfection. Soilingof scopes used in routine clinical practice, or longer ATS-M inoc-ulum drying times, could be used in future studies to address thislimitation. We elected to use loaned scopes for this study, as theinstruments in clinical use were not routinely available for ourexperiments.

In conclusion, agencies such as the FDA and American Societyfor Gastrointestinal Endoscopy have recommended that imple-mentation of microbial culturing can supplement a strict, robustadherence to duodenoscope reprocessing in order to help reducethe risk of infection transmission by duodenoscope (14, 26). Basedon our results, a microbial surveillance program utilizing the MEPwould recover organisms from contaminated duodenoscopes atan approximate 65% recovery rate with a limit of detectionaround 10 CFU/scope. While this surveillance can detect repro-cessing failures, it is important to note that, with only a 65% re-covery rate, certain scopes may still be contaminated below thelimit of detection. Thus, duodenoscope cultures should not be thesole basis when assessing risk of post-ERCP infections. Improvedrecovery could be tested by including neutralizer in the saline flushor by increasing the saline flush to 10 to 20 ml. Due to limitationsof microbial sampling, risk assessment should also include active

FIG 3 Visualization of sampling coverage on the distal end of a duodenoscope. (A) Distal end of the duodenoscope was coated in GlitterBug and viewed inambient light (no UV), under UV light (before sampling), and then after sampling (after sampling) with either a flocked swab (top panel) or Olympus brush(bottom panel). (B) Visual comparison of the brush and flocked swab in ambient light (top panel) and then again after sampling under UV light to observecollected GlitterBug (bottom panel). UV, ultraviolet.

Comparison of Duodenoscope Culture Methods

February 2016 Volume 54 Number 2 jcm.asm.org 315Journal of Clinical Microbiology

clinical surveillance of patients undergoing these procedures toidentify the likelihood that a duodenoscope will become colonizedwith a high-concern organism. Assessment of certain patient fac-tors, such as recent clinical culture results collected preprocedure,in addition to postprocedure clinical monitoring could serve asadditional methods to more rapidly identify high risk situations.Periodic culture surveillance of duodenoscopes and reprocessingtraining can serve as routine tests for breakdown in cleaning pro-cedures and ongoing quality indicators on processing these com-plex devices.

ACKNOWLEDGMENTS

We thank Preston Dahlgren and the dedicated endoscopy technicians atthe LDS Hospital for loaning us the duodenoscopes and reprocessing theinstruments after each experiment.

FUNDING INFORMATIONThis study was funded by the Intermountain Healthcare Department ofClinical Epidemiology and Infectious Diseases and the IntermountainHealthcare Surgical Services Clinical Program. This research received nospecific grant from any funding agency in the public, commercial, ornot-for-profit sector.

REFERENCES1. Anderson MA, Fisher L, Jain R, Evans JA, Appalaneni V, Ben-

Menachem T, Cash BD, Decker GA, Early DS, Fanelli RD, Fisher DA,Fukami N, Hwang JH, Ikenberry SO, Jue TL, Khan KM, Krinsky ML,Malpas PM, Maple JT, Sharaf RN, Shergill AK, Dominitz JA. 2012.Complications of ERCP. Gastrointest Endosc 75:467– 473. http://dx.doi.org/10.1016/j.gie.2011.07.010.

2. Doherty DE, Falko JM, Lefkovitz N, Rogers J, Fromkes J. 1982. Pseu-domonas aeruginosa sepsis following retrograde cholangiopancreatogra-phy (ERCP). Dig Dis Sci 27:169 –170. http://dx.doi.org/10.1007/BF01311712.

3. Classen DC, Jacobson JA, Burke JP, Jacobson JT, Evans RS. 1988.Serious Pseudomonas infections associated with endoscopic retrogradecholangiopancreatography. Am J Med 84:590 –596. http://dx.doi.org/10.1016/0002-9343(88)90141-6.

4. Katsinelos P, Dimiropoulos S, Katsiba D, Arvaniti M, Tsolkas P,Galanis I, Papaziogas B, Limenopoulos V, Baltajiannis S, Vasilladis I.2002. Pseudomonas aeruginosa liver abscesses after diagnostic endoscopicretrograde cholangiography in two patients with sphincter of Oddi dys-function type 2. Surg Endosc 16:1638.

5. Aumeran C, Poincloux L, Souweine B, Robin F, Laurichesse H, Baud O,Bommelaer G, Traore O. 2010. Multidrug-resistant Klebsiella pneu-moniae outbreak after endoscopic retrograde cholangiopancreatography.Endoscopy 42:895– 899. http://dx.doi.org/10.1055/s-0030-1255647.

6. Carbonne A, Thiolet JM, Fournier S, Fortineau N, Kassis-Chikhani N,Boytechev I, Aggoune M, Seguier JC, Senechal H, Tavolacci MP,Coignard B, Astagneau P, Jarlier V. 2010. Control of a multihospitaloutbreak of KPC-producing Klebsiella pneumoniae type 2 in France, Sep-tember to October 2009. Euro Surveill 15:pii�19734. http://www.eurosurveillance.org/ViewArticle.aspx?ArticleId�19734.

7. Alrabaa SF, Nguyen P, Sanderson R, Baluch A, Sandin RL, Kelker D,Karlapalem C, Thompson P, Sams K, Martin S, Montero J, Greene JN.2013. Early identification and control of carbapenemase-producing Kleb-siella pneumoniae, originating from contaminated endoscopic equipment.Am J Infect Control 41:562–564. http://dx.doi.org/10.1016/j.ajic.2012.07.008.

8. Epstein L, Hunter JC, Arwady MA, Tsai V, Stein L, Gribogiannis M,Frias M, Guh AY, Laufer AS, Black S, Pacilli M, Moulton-Meissner H,Rasheed JK, Avillan JJ, Kitchel B, Limbago BM, MacCannell D, Lon-sway D, Noble-Wang J, Conway J, Conover C, Vernon M, Kallen AJ.2014. New Delhi metallo-beta-lactamase-producing carbapenem-resistant Escherichia coli associated with exposure to duodenoscopes.JAMA 312:1447–1455. http://dx.doi.org/10.1001/jama.2014.12720.

9. Zweigner J, Gastmeier P, Kola A, Klefisch F-R, Schweizer C, HummelM. 2014. A carbapenem-resistant Klebsiella pneumoniae outbreak follow-ing bronchoscopy. Am J Infect Control 42:936 –937. http://dx.doi.org/10.1016/j.ajic.2014.04.022.

10. McCool S, Muto CA, Querry A, McGrath K, Slivka A, Zonker J,Pasculle W. High Level Disinfection (HLD) Failure in gastrointestinalscopes with elevator channels: is it time to switch to ethylene oxide (ETO)sterilization? In (ed).

11. Wendorf KA, Kay M, Baliga C, Weissman SJ, Gluck M, Verma P,D’Angeli M, Swoveland J, Kang MG, Eckmann K, Ross AS, Duchin J.2015. Endoscopic retrograde cholangiopancreatography-associatedAmpC Escherichia coli outbreak. Infect Control Hosp Epidemiol 36:634 –642. http://dx.doi.org/10.1017/ice.2015.66.

12. United States Food and Drug Administration. 2015. FDA executivesummary: effective reprocessing of endoscopes used in endoscopic retro-grade cholangiopancreatography (ERCP) procedures. http://www.fda.gov/downloads/AdvisoryCommittees/CommitteesMeetingMaterials/MedicalDevices/MedicalDevicesAdvisoryCommittee/Gastroenterology-UrologyDevicesPanel/UCM445592.pdf.

13. United States Food and Drug Administration. 2015. FDA safety com-munication: design of endoscopic retrograde cholangiopancreatography(ERCP) duodenoscopes may impede effective cleaning. http://www.fda.gov/MedicalDevices/Safety/AlertsandNotices/ucm434871.htm.

14. United States Food and Drug Administration. 2015. FDA safety com-munication: supplemental measures to enhance duodenoscope reprocess-ing. http://www.fda.gov/MedicalDevices/Safety/AlertsandNotices/ucm454766.htm.

15. Buss AJ, Been MH, Borgers RP, Stokroos I, Melchers WJ, Peters FT,Limburg AJ, Degener JE. 2008. Endoscope disinfection and its pitfalls:requirement for retrograde surveillance cultures. Endoscopy 40:327–332.http://dx.doi.org/10.1055/s-2007-995477.

16. Cêtre JC, Nicolle MC, Salord H, Pérol M, Tigaud S, David G, BourjaultM, Vanhems P. 2005. Outbreaks of contaminated bronchoalveolar lavagerelated to intrinsically defective bronchoscopes. J Hosp Infect 61:39 – 45.http://dx.doi.org/10.1016/j.jhin.2004.12.020.

17. Moses FM, Lee J. 2003. Surveillance cultures to monitor quality of gas-trointestinal endoscope reprocessing. Am J Gastroenterol 98:77– 81. http://dx.doi.org/10.1111/j.1572-0241.2003.07165.x.

18. Beilenhoff U, Neumann CS, Rey JF, Biering H, Blum R, Schmidt V,Committee EG. 2007. ESGE-ESGENA guideline for quality assurance inreprocessing: microbiological surveillance testing in endoscopy. Endos-copy 39:175–181. http://dx.doi.org/10.1055/s-2006-945181.

19. Systchenko R, Archetti B, Canard J, Palazzo L, Ponchon T, Rey J,Sautereau D, Endoscopy CotFSoD. 2000. Guidelines of the French Soci-ety of Digestive Endoscopy: recommendations for setting up cleaning anddisinfection procedures in gastrointestinal endoscopy. Endoscopy 32:807– 818. http://dx.doi.org/10.1055/s-2000-7710.

20. Public Health Agency of Canada. 2010. Infection prevention and controlguideline for flexible gastrointestinal endoscopy and flexible bronchos-copy. http://www.phac-aspc.gc.ca.

21. Whitby M, Fielding D, Todman M, Rapley L. 2010. Infection control inendoscopy, 3rd ed. Gastroenterological Society of Australia, Victoria, Aus-tralia.

22. ECRI Institute. 2015. ECRI institute recommends culturing duodeno-scopes as a key step to reduce CRE infections. http://www.ecri.org/cre.

23. Centers for Disease Control and Prevention. 2015. Interim samplingmethod for the duodenoscope: distal end and instrument channel. Cen-ters for Disease Control and Prevention, Atlanta, GA. http://www.cdc.gov/hai/settings/lab/lab-duodenoscope-sampling.html.

24. Healthmark Industries. Simulated-use testing: flexible endoscope. http://www.art ificial testsoi l .com/Simulated_Use_Test ing_Flexible_Endoscope.pdf.

25. PSAB Committee on Laboratory Practices. 2015. On the question ofculturing duodenoscopes. American Society of Microbiology, Washing-ton, DC. https://www.asm.org/index.php/component/content/article/98-policy/issues/93456-lp-4-15.

26. American Society for Gastrointestinal Endoscopy. 2015. Transmissionof CRE bacteria through endoscopic retrograde cholangiopancreatogra-phy (ERCP). http://www.asge.org/uploadedFiles/Publications_and_Products/ASGE_InterimGuidance_CRE_03172015.pdf.

Gazdik et al.

316 jcm.asm.org February 2016 Volume 54 Number 2Journal of Clinical Microbiology