To the Editor—Flexible endoscopes may become heavily contaminated with blood, secretions, and microorganisms during use. Over the last several years an increasing number of cases have been reported in which patients have been exposed to infectious microorganisms by contaminated gastrointestinal (GI) endoscopes. Reference Kovaleva, Peters, van der Mei and Degener1
The complete and accurate reprocessing of flexible endoscopes is a multistep procedure involving manual cleaning followed by high-level disinfection (HLD) and active drying before storage. 2 Because almost all reported outbreaks are related to breaches in reprocessing techniques, it is crucial that endoscope cleaning, disinfection, and drying are performed according to a strict protocol. Manual cleaning is a critical reprocessing step, removing >99% of the bioburden from the endoscope prior to automated reprocessing. Reference Garces3 Several different types of brushes are used as an essential accessory to routine manual cleaning, but very little is known about the differences in the performance of these brushes. Therefore, we compared the performance of different brushes widely used for manual cleaning of flexible endoscopes: (1) a PULL THRU brush (Medivators, Minneapolis, MN), (2) Push&Pull brush (with sweeper) (Endoss, The Netherlands), and (3) a double-cleaning brush (DCB; Nova LightSystems, France).
For the endoscope model, we used new polytetrafluoroethylene (PTFE) tubes soiled with a pool of 4 positive hemocultures. The method was based on the method described by Cattoir et al, Reference Cattoir, Vanzieleghem and Florin4 and detailed method description can be found in the Supplementary Material (online). Mean adenosine triphosphate (ATP) values and mean culture yield after 7 days (colony-forming units [CFU]/100 mL) obtained after using different brushing techniques on soiled PTFE tubes are presented in Table 1. To our knowledge, this is the first study to compare the effectiveness of different types of brushes in manual cleaning of surrogate endoscope channels by means of microbiological culture and ATP values.
Table 1. Mean ATP Results and Culture Yield After Different Brushing Techniques Performed on an Endoscope Model a
![](https://static.cambridge.org/binary/version/id/urn:cambridge.org:id:binary:20220802084039404-0470:S0899823X21001707:S0899823X21001707_tab1.png?pub-status=live)
Note. RLU, relative light units; SD, standard deviation; NA, not applicable.
a In a pairwise comparison of the ATP values (Kruskall-Wallis test P < .05 was considered statistically significant) only the difference between the Push&Pull and DCB subgroups was retained as statistically significant (P = .009).
b Two positive control tubes were not submerged in the water and not brushed.
c Two positive control tubes were submerged in water and flushed with water but not brushed.
d Correlation between mean ATP value and mean yield of culture (Spearman correlation coefficient)
e Log10 reduction of ATP caused by brushing was calculated as log10 reduction of mean ATP value of the brushed PTFE tubes to the positive controls not flushed with water and not brushed.
Our in vitro experiments revealed that, for soiled PTFE tubes, ATP values of samples cleaned with the Push&Pull brush were significantly lower than those cleaned with the DCB and lower (but not significantly) than those cleaned with a PULL THRU brush. A literature search showed that there is no confirmed threshold for ATP value, so difficulties remain in interpreting this parameter.
There was no correlation between ATP measurements and culture results (Table 1). This result corresponds to the findings of Batailler et al, Reference Batailler, Saviuc, Picot-Gueraud, Bosson and Mallaret5 who concluded that ATP cannot be used as an alternative to microbiological tests for monitoring endoscope reprocessing. The Aquasnap Total test (Hygiena, Camarillo, CA) detects ATP from bacteria in addition to all other biological sources (organic residues) of ATP. Because CFU and relative light unit (RLU) values are determined using different test methods and measure different substances, we expect that RLU values do not consistently correlate with CFUs.
Culture results showed that mean yield from soiled PTFE tubes was lowest when the tube was cleaned with the Push&Pull brush. However, differences in mean CFU count did not reach statistical significance. The difference in CFU count can be presumed to not be clinically relevant either. Recovery in CFU/100 mL was calculated from the mean yield per brush type relative to the positive control, which did not undergo any reprocessing. These results are consistent with previous findings in literature that manual cleaning removes >99% of the bioburden from the endoscope. Reference Garces3 Our study confirms the paramount importance of manual cleaning.
A fourth type of brush is a simple brush (Olympus). The design of the simple brush is basically a half DCB (with the brush part only on 1 side). Because the DCB had already showed inferior results to the 2 other brushes, we did not think it would be useful to include these in a further comparison.
In 2007, Charlton also compared the cleaning efficacy of different brushes. Efficacy was tested by applying a simulated blood soil to a lumen and comparing the weight difference before and after cleaning. In contrast to our study, which examined microbiological growth. Charlton concluded that the PULL THRU device was found to offer a consistently significant improvement in soil removal. Reference Charlton6 He hypothesized this is because the wiper element design of the PULL THRU device provides a complete circumferential seal in the lumen channel. Reference Charlton6
In Belgium, the authorities have issued guidelines for endoscope reprocessing. 2 The guidelines state that the brushes used for manual cleaning are preferably single use, without other specifications. The European guidelines recommend brushing of all accessible channels using flexible, purpose-designed brushes. The size and type of cleaning brush must be matched appropriately to the size and type of endoscope channels to ensure contact with channel walls. In our study, PTFE tubes close to the actual size of endoscope channels were used. To ensure maximum effectiveness of cleaning and to avoid tissue carryover, the European Society of Gastrointestinal Endoscopy (ESGE) and the European Society of Gastroenterology and Endoscopy Nurses and Associates (ESGENA) recommend the use of single-use brushes because they have undamaged bristles without any tissue remnants from previous examinations. Reference Beilenhoff, Neumann, Rey, Biering, Blum and Schmidt7 In the United States, the guidelines state to flush and brush all accessible channels with a brush appropriate for the size of the endoscope channel to remove all organic (eg, blood, tissue) and other residues. They also recommend that cleaning items should be disposable or thoroughly cleaned and disinfected or sterilized between uses. Reference Petersen, Chennat and Cohen8
This study has several limitations. Our manually soiled PTFE tubes did not contain a biofilm in the sense of Michelle Alfa’s definition. Reference Alfa and Howie9 The “buildup biofilm” in endoscopes develops as a result of cyclical exposure to wet and dry phases during use and reprocessing. Because certain cleaning devices may compound the accretion of residual soil by causing surface abrasion or grooving of the lumen wall, additional research to address this issue could yield helpful insights. Reference Beilenhoff, Neumann, Rey, Biering, Blum and Schmidt7 Furthermore, the conclusion of our study is not necessarily generalizable for several reasons. Experiments were only performed at a single hospital site by 1 person, and only endoscope models were used. A future larger study could include different types of endoscopes from different manufacturers in a real-life hospital setting.
Acknowledgments
Sofie Colman performed the calculations and wrote the manuscript. All authors discussed the results and commented on the manuscript.
Financial support
No financial support was provided relevant to this article.
Conflicts of interest
All authors report no conflicts of interest relevant to this article.
Supplementary material
To view supplementary material for this article, please visit https://doi.org/10.1017/ice.2021.170