To the Editor—Duodenoscopes used for endoscopic retrograde cholangio-pancreatography (ERCP) have complex designs that make reprocessing challenging. Infections have been linked to manual cleaning of the scope especially its forceps elevator. Other factors that contribute to infections include use of unsterile water and inappropriate storage of scopes. Reference Kovaleva, Peters, van der Mei and Degener ^{1 ,} Reference Muscarella ² Despite duodenoscope reprocessing procedures exceeding manufacturer’s recommendations, high-concern organisms such as Klebsiella spp and Pseudomonas spp have been implicated in clinical infections.Reference Epstein, Hunter, Arwady and Tsai ^{3 ,} Reference Wendorf, Kay and Baliga ^{4 ,} Reference Aumeran, Poincloux and Souweine ⁵ Media reports of high-concern organisms, such as carbapenem-resistant Enterobacteriaceae (CRE) and extended-spectrum β-lactamase (ESBL) outbreaks linked to duodenoscopes, have heightened awareness regarding reprocessing procedures.Reference Epstein, Hunter, Arwady and Tsai ^{3 ,} Reference Wendorf, Kay and Baliga ⁴ Infections from duodenoscopes have been linked to positive cultures isolated from urine, blood, abscesses, and stool.Reference Kovaleva, Peters, van der Mei and Degener ^{1 ,} Reference Muscarella ² Mortality associated with contaminated duodenoscopes is ~16% with all organisms and 56% with CRE.Reference Wendorf, Kay and Baliga ⁴ These mortality rates emphasize the need for optimal reprocessing practices. The World Health Organization emphasizes team-based collaborations, such as multidisciplinary teams (MDTs), to improve communication among healthcare workers. ⁶ Many studies have shown the benefits of MDTs in reducing nosocomial infections like bloodstream infections. Reference Walz, Ellison and Mack ^{7 –} Reference Miller, Simmons, Dale, Stachowiak and Stibich ⁹ Multidisciplinary teams are effective at reducing infection rates through rapid identification of breakdowns in the process.Reference Walz, Ellison and Mack ⁷ We studied the impact of creating a MDT with clear roles and real-time huddles to optimize our scope-reprocessing practices.

This retrospective study was conducted at a tertiary-care academic medical center with 401 beds. We aimed to evaluate the impact of an MDT with clear roles on the reprocessing of duodenoscopes. Reprocessing Olympus TJF-Q180V duodenoscopes along with surveillance cultures of the duodenoscope tip (including forceps elevator) were evaluated during the baseline period (January 2016 through June 2016) and during our intervention period (September 2016 through July 2017). An MDT was created in July 2016 composed of representatives from the endoscopy center, the sterile processing department (SPD), the infection prevention department (IP), as well as hospital leadership. We utilized a responsibility assignment matrix (RAM) to outline responsibilities of team members (Table 1). The results of surveillance cultures were grouped based on risk to humans, as defined by Centers for Disease Control and Prevention, as low- and high-concern bacteria. ⁶

TABLE 1 Responsibility Assignment Matrix (RAM) Implemented as a Part of Our Intervention

NOTE. R, responsible; I, informed; C, consulted; GI, gastroenterology; CSPD, central sterile processing department; IP, infection prevention; ATP, adenosine triphosphate; HLD, high-level disinfection; ETO, ethylene oxide.

The reprocessing of duodenoscopes at our center starts with bedside manual cleaning followed by repeat manual cleaning (within an hour) in the sterile processing department. To detect any residual biological material, adenosine triphosphate (ATP) testing is then performed on 5 spots of the duodenoscope: surface, 3 channels, and the elevator. If ATP levels are <100 relative light units (RLU), the duodenoscope undergoes high-level disinfection (HLD). If the duodenoscope fails ATP testing, the duodenoscope is recleaned following manufacturer’s instructions and undergoes ATP retesting. After HLD with an automated endoscopic reprocesser, a total of 10 duodenoscopes undergo surveillance cultures every month. Duodenoscopes that are cultured are sent through HLD again. All duodenoscopes are then sterilized with ethylene oxide (ETO) prior to use. We have 24 scopes, and 92% of the scopes undergo surveillance cultures in a span of 4 months. Our monthly surveillance cultures represent efficacy of manual cleaning.

During the baseline period (January 2016 through June 2016), scopes were cultured after 9.4% of the procedures (n=267). During the intervention period (September 2016 through July 2017), cultures were obtained after 20.3% of the procedures (n=492; P<.05). During our baseline period, 10 of 25 cultures were positive (40%). During the intervention period, 4 of 100 cultures were positive (4%; P<.05). We reduced our culture positivity by 36% by increasing the efficacy of our manual cleaning. Culture positivity is the ratio of positive cultures divided by number of scopes cultured. Compliance with the policy for obtaining cultures increased from 41.7% during the baseline period to 90.9% during the intervention period. Compliance was defined as a ratio of number of cultures obtained and number of cultures expected to be obtained during a defined period. Our compliance with the policy for obtaining cultures increased by 49.2%. Our compliance with the policy for manual cleaning within 1 hour of bedside cleaning increased from 38.5% (47 of 122 cultures) in the baseline period to 50.8% (375 of 738 cultures; P<.05) in the intervention period. Improvement in compliance with other steps in the process was not statistically significant.

By establishing clear responsibilities with RAM (Table 1) and emphasizing real-time huddles (when scope cultures are positive), we reduced the rate of culture positivity significantly from 40% to 4%. We attribute our success to weekly meetings of MDT members from infection prevention and central sterile processing, which created a high level of engagement. We also developed a process of sending a notification (blast) page to all team members when a scope culture was positive. When a blast page was issued, all team players huddled within hours and conducted a root cause analysis. We used a shared database to track each endoscope from the time of use on the first patient to the subsequent patient. The data included in this database were added by different team members. An action plan was created, and a communication was sent to everyone on the RAM within 24 hours of each meeting.

This study has several limitations. It was a single-center experience, which reduces its generalizability. We did not have a control group, which reduces our confidence that these results were due to the intervention. We relied on capturing compliance based on documentation by personnel for most of the processes in our protocol, which allows for human error.

In conclusion, by emphasizing principles of accountability (RAM) and effective communication (real-time huddles), we were able to show improved efficacy of manual cleaning of endoscopes, which was indicated by reduction in the rate of culture positivity.