Is bacteriologic surveillance in endoscope reprocessing stringent enough?

Endoscopes, including duodenoscopes, are medical devices that are frequently associated with outbreaks of nosocomial infections. We investigated an outbreak of multidrug-resistant PSEUDOMONAS AERUGINOSA sepsis affecting three patients after endoscopic retrograde cholangiopancreaticography (ERCP). Epidemiologic investigation supplemented by molecular typing revealed that one ERCP scope was the source of infection with P. AERUGINOSA. No contamination with this microorganism was found after screening of washer-disinfectors, connecting tubes, and environmental surfaces in the endoscopy center. PSEUDOMONAS isolates from blood and endoscope channels before gas sterilization with ethylene oxide (ETO) were characterized by molecular typing as "linked isolates". Though the current surveillance system did not prevent the infections in three patients, our microbiological surveillance protocol with routine culturing of endoscopes was helpful in detecting the source of contamination and probably avoided numerous cross-contaminations in other patients who underwent ERCP procedures with endoscopes.

Endoscope disinfection and its pitfalls--requirement for retrograde surveillance cultures.

BACKGROUND AND STUDY AIMS: Several endoscopy-related outbreaks of infection have been reported in recent years. For early recognition of inadequate disinfection of endoscopes we designed a microbiological surveillance system to evaluate the efficacy of the cleaning and disinfection procedure, and to trace disinfection problems to individual endoscopes or washer-disinfectors. METHODS: Our surveillance protocol included anterograde and retrograde sampling, a decision algorithm, genetic fingerprinting, and scanning electron microscopy. RESULTS: Over a period of 29 months we found an increasing number of patient-ready endoscopes testing positive for Candida species other than albicans, especially C. parapsilosis. These yeasts were also isolated from the washer-disinfectors. The number of positive tests for Candida species varied from 1 out of 21 to 14 out of 27 samples from nine frequently used endoscopes. The number of colony-forming units per milliliter ranged from 1 - 10 to 3000 for endoscopes and 0.002 to 0.06 for the washer disinfectors. DNA fingerprinting was not able to discriminate different strains within C. parapsilosis. CONCLUSIONS: Our protocol was able to detect a structural problem in the endoscope disinfection process. Retrograde sampling was crucial for this purpose, because it has much higher sensitivity than anterograde sampling. Endoscopes with damaged working channels are probably the source of the contamination problem with Candida species.