Journal of Hospital Infection 83 (2013) 22e29
Available online at www.sciencedirect.com
Journal of Hospital Infection
journal homepage: www.elsevierhealth.com/journals/jhin
Current limitations about the cleaning of luminal
endoscopes
R. Hervé*, C.W. Keevil
Environmental Healthcare Unit, Centre of Biological Sciences, University of Southampton, Southampton, UK
A R T I C L E
I N F O
Article history:
Received 1 September 2011
Accepted 22 August 2012
Available online 23 October
2012
Keywords:
Luminal endoscope
Decontamination
Prion
CJD
S U M M A R Y
Background: The presence and potential build-up of patient material such as proteins in
endoscope lumens can have significant implications, including toxic reactions, device
damage, inadequate disinfection/sterilization, increased risk of biofilm development and
potential transmission of pathogens.
Aim: To evaluate potential protein deposition and removal in the channels of flexible
luminal endoscopes during a simple contamination/cleaning cycle.
Methods: The level of contamination present on disposable endoscopy forceps which
come into contact with the lumen of biopsy channels was evaluated. Following observations in endoscopy units, factors influencing protein adsorption inside luminal endoscope
channels and the action of current initial cleaning techniques were evaluated using
a proteinaceous test soil and very sensitive fluorescence epimicroscopy.
Findings: Disposable endoscope accessories appear to be likely to contribute to the
contamination of lumens, and were useful indicators of the amount of proteinaceous soil
transiting through the channels of luminal endoscopes. Enzymatic cleaning according to
the manufacturer’s recommendations and brushing of the channels were ineffective at
removing all proteinaceous residues from new endoscope channels after a single
contamination. Rinsing immediately after contamination only led to a slight improvement
in decontamination outcome.
Conclusion: Limited action of current decontamination procedures and the lack of
applicable quality control methods to assess the cleanliness of channels between patients
contribute to increasing the risk of cross-infection of potentially harmful micro-organisms
and molecules during endoscopy procedures.
ª 2012 The Healthcare Infection Society. Published by Elsevier Ltd. All rights reserved.
Introduction
Complex instruments such as flexible luminal endoscopes
present various materials, cavities and hinges allowing the
deposition of contaminants. The adsorption and aggregation of
* Corresponding author. Address: Environmental Healthcare Unit,
Centre of Biological Sciences, University of Southampton, Southampton SO16 7PX, UK. Tel.: þ44 (0) 2380592034; fax: þ44 (0)
2380 595159; FAO Dr R. Hervé.
E-mail address: [email protected] (R. Hervé).
patient tissue proteins in endoscope lumens could have significant implications. Such contamination may lead to toxic reactions and device damage, provide a favourable background for
colonization by micro-organisms and biofilm formation, and
subsequently increase the risk of inadequate disinfection/sterilization.1 Protein contamination has emerged as a potential
threat with identification of the misfolded prion protein (PrPSc)
as the most likely ‘non-conventional’ infectious agent of
transmissible spongiform encephalopathies (TSEs).2 Identified
from sheep scrapie, PrPSc is also found in TSEs affecting humans,
including the variant and sporadic forms of Creutzfeldt-Jakob
0195-6701/$ e see front matter ª 2012 The Healthcare Infection Society. Published by Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.jhin.2012.08.008
R. Hervé, C.W. Keevil / Journal of Hospital Infection 83 (2013) 22e29
disease (vCJD and sCJD, respectively). Variant CJD is characterized by a lengthy incubation period and wide tissue distribution of infectivity in asymptomatic carriers. It is therefore
difficult, at present, to quantify the risk of re-using surgical
devices that have been used on patients at risk of CJD.3e6
Endoscopes are fragile, expensive pieces of equipment.
They cannot be autoclaved at temperatures sufficient to
destroy micro-organisms or eliminate the infectivity of potentially adsorbed prions. Despite presenting difficulties to clean,
endoscopes are subjected to repetitive use and rapid turnaround. This involves manual cleaning, leak testing and
washing in automatic endoscope reprocessors (AERs). Current
methods available to clinical settings for the rapid assessment
of instrument cleanliness lack sensitivity and are not applicable
to endoscope channels.7 Most microbiological contamination
tests still rely on sufficient recovery after swabbing.8
Manufacturers give specific recommendations for the
decontamination of flexible endoscopes. Current evidence
suggests that the prevalence of vCJD is low; nevertheless,
there is a thriving market for AERs and associated chemistries,
some designed to eliminate prion infectivity where this is
a concern.9 However, there is a lack of independent scientific
evidence about the action of these products on protein removal
or the ‘inactivation’ of human prion strains. Recommended
‘best practices’ for the reprocessing of flexible endoscopes
(such as ISO 15883 Parts 1 and 4) are based on current evidence
and available technologies.10,11 Due to the limited follow-up of
elective patients, variable incubation periods and potential
lack of reports from patients, it is impossible to maintain
a comprehensive record of iatrogenic infections through
endoscopy procedures. Most cross-infections will remain
unnoticed because they involve common innocuous commensals. However, it is well documented that endoscope channels
represent ideal niches for a range of micro-organisms, which
may include occasional pathogens with serious consequences in
immunocompromised patients.12 Proteinaceous contaminants
are likely to offer the basic substrate for adhesion and
embedding of bacteria, viruses and, in extreme cases, parasites, and should therefore be a prime target during endoscope
reprocessing.
Episcopic differential interference contrast/epi-fluorescence
(EDIC/EF) microscopy is a very sensitive technique developed in
the authors’ laboratory for the rapid quantification of microbial
and proteinaceous contamination on surfaces.13,14 Based on
observations in local endoscopy units, this study applied EDIC/EF
microscopy to evaluate the protein adsorption inside working
channels of flexible endoscopes.
Methods
Endoscope channels and accessories
Endoscope channels were kindly provided by a manufacturer. Single-use endoscope biopsy forceps and single-use
brushes for manual cleaning of the particular channels examined were a kind gift from various endoscopy units.
Soil
Edinburgh soil (Medisafe; Bishops Stortford, UK) was resuspended in reverse-osmosis (RO) distilled water and stored
23
at 4 C, following the manufacturer’s recommendations. The
suspension was further diluted (10% or 1% v/v) in RO water as
required. A protein assay (Bradford method, Bio-Rad, Hemel
Hempstead, UK) was performed using a range of test soil dilutions (in triplicate) to determine the protein concentration.
Contamination of endoscope channels
New endoscope biopsy and air/water channels were cut into
short sections (20 cm) and sealed to a syringe. Test soil was
pumped through the whole length of the tubing, followed by air
to remove excess soil, to simulate the passage of tissues.
Adsorbed soil was then left in contact with the luminal surface
while the channel was placed in a dry incubator at 37 C for
10 min (according to the test soil manufacturer’s recommendations e identified as ‘dry condition’ later in the text);
alternatively, the channel was filled with de-ionized water
(‘wet condition’) until further cleaning.
Decontamination
An enzymatic cleaner used in a number of endoscopy units
was employed for the decontamination of soiled channels,
following the manufacturer’s recommendations. Volumes were
measured accurately and temperature (37 C) was monitored.
Contact time was 5 min for all channels (including brushing
when applied).
Disposable single-use endoscope brushes were also used
according to current practice. With the channel immersed in
the enzymatic solution, the brush head was inserted at one
end, pushed and pulled several times through the whole length
of the channel, then pushed completely out at the other end.
Channels were finally rinsed by flushing de-ionized water, and
excess water was removed by flushing air through the channel
(using new syringes for each stage) based on standard practice.
Staining and microscopy
Proteinaceous deposits on endoscope accessories and
inside channel lumens were detected using SYPRO Ruby
(excitation: 470 nm; emission: 618 nm; Molecular Probes,
Eugene, OR, USA) and an EDIC/EF microscope (Best Scientific,
Wroughton, UK) as described elsewhere.14 After cleaning,
a new syringe was sealed to the channel to allow gentle
injection of the dye inside the lumen. After 15 min, the syringe
was replaced by a new syringe containing de-ionized water,
and the unbound dye inside the channel was washed off by
gentle flushing. Air was then pumped through once to remove
excess water. The tubing was cut in half longitudinally (using
a new scalpel blade for each channel) and left to dry for up to
1 h (in the dark) before examining the lumen. Staining was
quantified from 10 representative fields of view for each
channel examined, using Image Pro software (MediaCybernetics, Silver Spring, MD, USA).
Statistics
Quantitative data were expressed as mean standard error
of the mean from a least five separate experiments, and
differences in contamination were assessed using analysis of
variance and t-test. A value of P 0.05 was considered to
indicate significance.
24
R. Hervé, C.W. Keevil / Journal of Hospital Infection 83 (2013) 22e29
Results
Protein contamination on endoscope channels and
accessories
The biopsy forceps examined after clinical use carried
several micrograms of proteins (Figure 1AeE). This provided an
estimation of the amount of residual proteins remaining on the
surface of endoscopy instrumentation after clinical use.
Two channel types were identified (Figure 1F), both PTFElike according to the manufacturer. The first series of biopsy
channels observed were made of two visible layers: a white
rough opaque layer on the outside and a smoother translucent
layer on the luminal side (‘O-type channels’). Air/water
channels and more recent biopsy channels appeared to be
made of a homogenous translucent material (‘T-type channels’). The plastic materials observed generated some background fluorescence that did not affect the sensitivity of the
method. The protein concentration in the test soil used was
approximately 420 mg/mL (as determined by the Bradford
method); therefore, the 10% and 1% dilutions corresponded to
42 mg/mL and 4.2 mg/mL of protein, respectively.
Soil applied for only a few seconds led to the adsorption of
relatively large quantities of proteins on the luminal surface of
working channels (Figure 2). Protein adsorption in T-type
channels was significantly less than that of O-type channels for
10% and 100% test soil. Test soil diluted at 1% (v/v) produced
a similar level of micro-contamination (under 10 ng/mm2) in
both types (Figure 2, left micrograph). The 10% dilution
revealed some characteristics of protein deposition (Figure 2,
middle micrograph). Proteins aggregated after partial drying
(white arrows), with some aggregates apparently displaced by
the flow, leaving longitudinal marks (green arrow). Undiluted
soil left a homogenous film, with drying cracks appearing after
partial dehydration (Figure 2; right micrograph).
Enzymes, brushes, and dry vs wet issue in luminal
endoscope decontamination
The initial binding described above (no washing) was used as
a control. Enzymatic cleaning had a significant effect against
gross contamination (above 10 ng/mm2; Figure 3). Enzymatic
cleaning of 10% or 1% test soil in T-type channels had limited
effect (Figure 3, 1% and magnified 10%), indicative of a basal
level of strongly adsorbed proteinaceous micro-contamination.
The O-type channels showed less residual contamination than
the T-type channels after cleaning, suggesting different physical characteristics for the two types (lesser but stronger
retention in T-type channels). Using this particular test soil and
enzymatic cleaner, keeping the channels saturated with water
only improved decontamination results slightly with the 10%
mixture in T-type channels (Figure 3). There was also an
improvement when keeping the 1% test soil wet in the T-type
channels, although this was not statistically significant due to
relatively high variability in micro-contamination levels.
Micro-droplets of water remained on the luminal surface of
channels despite the forced passage of air and extended drying
while the channel was cut open (Figure 3, picture), confirming
that these channels are liable to retain humidity.
Specifically designed brushes were evaluated on 100% test
soil, which seems to be representative of soil levels following
endoscopic procedures. Non-brushed controls were included in
the same experiments for statistical comparison, and the
outcome was consistent with the previous series of experiments (data not shown). Brushing improved cleaning outcome
in T-type channels, regardless of whether they were partially
dried or kept wet (Figure 4). No improvement was observed in
O-type channels, where brushing actually increased spreading
and enhanced retention in wet conditions, indicating that
micro-contamination adsorbed to these channels is highly
resistant to removal. In both types of channels, brushes
produced smearing marks of residual micro-contamination
(Figure 4 pictures, white arrows).
Discussion
Materials and protein adsorption
Endoscopy forceps are likely vectors of contamination for
the whole length of the working channel, and protein deposition is likely to occur during the extraction of biopsy specimens. This study examined single-use forceps; however,
re-usable forceps are also available. This study measured
significant amounts of residual proteins adsorbed in endoscope
channels made of two different materials after a single
exposure and a short contact time. One material showed
reduced adsorption characteristics; however, very resilient
micro-contamination was observed in all cases. To the
authors’ knowledge, both types of material are in current use,
and T-type biopsy channels were introduced more recently.
The manufacturer did not indicate if O-type channels were
being systematically replaced by T-type channels worldwide,
or if other types of channels are being introduced. It is the
authors’ understanding that channel replacement is not
a compulsory event during routine maintenance. Consequently, an endoscope may be fitted with the same channel for
many months or years. Despite the passage of dry air, water
micro-droplets remained present within the channels. Adsorbed proteins may contribute to the retention of water. This
might explain the relatively small differences observed
between the ‘dry’ and ‘wet’ conditions. The fact that working
channels retain water is likely to contribute to biofilm
formation.
Limitations of enzymatic cleaning and brushing
Previously, the authors reported on the levels of soil present
on re-usable stainless steel surgical instruments deemed to be
clean by current control methods, and the influence of
humidity on instrument cleaning.15,16 The present study used
new endoscope channels and a common enzymatic cleaner for
flexible luminal endoscopes according to recommendations.
This failed to completely remove test soil adsorbed after only
a few seconds, either partially dried or kept wet, under
favourable laboratory conditions. Temperature and concentration, two important parameters for such processes, may be
poorly controlled in some busy endoscopy units.
Detergents contained in cleaning solutions are likely to be
more stable than enzymes, particularly when stored at room
temperature. Detergents may also work at a wider range of
temperatures (preferably ‘warm’ water), although concentration and time of contact should also be properly
R. Hervé, C.W. Keevil / Journal of Hospital Infection 83 (2013) 22e29
25
Figure 1. Montage of approximate positions showing proteins deposited on (A) needle, (B) open articulation, (C) hinge and (D) axis of
endoscope biopsy forceps; (E) outside view of closed cups. Bars are 100 mm. (F) Macroscopic external view and microscopic view of the
cross-section of opaque and transparent biopsy channels.
26
R. Hervé, C.W. Keevil / Journal of Hospital Infection 83 (2013) 22e29
180
***
160
Initial adsorption (ng/mm2)
140
120
*
100
**
80
60
40
20
*
0
1%
10%
100%
Figure 2. Initial protein adsorption from three concentrations of test soil after one single short contact in O-type (grey bars) and T-type
(white bars) channels. Values are mean standard error of the mean from at least five separate experiments. Statistical significance is
reported as follows: comparison with previous lower concentration, *P 0.05, ** P 0.01, ***P 0.001; comparison of T-type with O-type
channels for the same soil concentration, --P 0.01, ---P 0.001. Micrographs showing SYPRO-Ruby stained residual proteins from 1%,
10% and 100% test soil, respectively, detected in T-type channels using episcopic differential interference contrast/epi-fluorescence. In
the middle micrograph showing 10% soil residues, the green arrow points at a smear mark left by soil flow running longitudinally inside the
channel. White arrows point at protein aggregates remaining after drying. Bars are 100 mm.
controlled. However, detergents alone may displace and
solubilize the contamination without inactivating noxious
agents that may be present. Displacement of soil and
potential cross-contamination of instruments reprocessed in
shared AERs (and automatic washer disinfectors) is a potential issue in clinical facilities. Thorough rinsing is also essential. Although the present study did not aim to evaluate
potential adsorption of cleaning chemistries themselves, the
authors have reported this issue in other publications,
particularly relating to salts and the quality of rinsing water in
automatic washer disinfectors (which would also apply to
AERs).17 No protein adsorption was observed after a single
wash cycle with the enzymatic solution tested. This does not
preclude the possibility of this occurring after several cycles
and/or when protein deposits are already present. In addition
to the variable detergent and enzyme content, cleaning
solutions are also characterized by their pH, which is often
alkaline. If applied frequently, strongly alkaline products are
likely to be detrimental to instrument surfaces. Luminal
endoscopes rely on chemical decontamination alone, and
there is currently limited scientific evidence regarding the
cleaning efficacy and material compatibility of commercial
products.
Brushing had some beneficial effect, which was enhanced at
the macroscopic level by partial drying and subsequent
aggregation of proteins. However, brushing appeared to be
liable to spread micro-contamination. Micro-contamination
being the most challenging target, cleaning outcome could
improve through better brush design.
Test soils are not fully representative of ‘real tissue’.
Nevertheless, a properly normalized soil appeared to be suitable to achieve the aim of this study, which was to compare the
R. Hervé, C.W. Keevil / Journal of Hospital Infection 83 (2013) 22e29
180
30
Residual soil (ng/mm2)
Residual soil (ng/mm2)
160
140
120
100
80
60
40
*
20
***
*
20
15
10
0
0
100
5
90
80
4
70
60
50
40
30
20
10
***
***
+
*
25
*
***
5
***
Residual soil (ng/mm2)
Residual soil (ng/mm2)
27
***
3
+
2
1
***
***
0
0
Control
Dried
Wet
7
1%
Residual soil (ng/mm2)
6
5
4
3
2
*
1
*
**
0
Control
Dried
Wet
Figure 3. Effect of enzymatic washing on different soil concentrations kept wet or partially dried in O-type (grey bars) and T-type (white
bars) channels. Magnified scales are shown for 100% and 10% data. Values are mean standard error of the mean from at least five
separate experiments. Statistical significance is reported as follows: comparison with unwashed control, *P 0.05, ** P 0.01,
***P 0.001; comparison of T-type vs O-type channels for the same condition, -P 0.05, --P 0.01, ---P 0.001; comparison of dried
with wet condition for the same channel type and soil concentration, þP 0.05. Picture shows residual water micro-droplets inside
a working channel after cleaning and passage of forced air (bar is 100 mm).
effect of different parameters on the initial cleaning of
lumens. This study shows that new channels adsorb significant
quantities of proteins after a single test contamination.
Rigorous cleaning only reduces residual proteins to levels
undetectable by methods currently employed in clinical
settings. This could be critical in countries with a population at
risk of vCJD, given that prions represent the fraction of
proteinaceous contamination that is most resistant to decontamination.15 The relevance of animal infectivity assays to
extrapolate the potential ‘inactivation’ of CJD itself by any
chemistry, apart from (possibly) strong alkaline solutions, is
arguable.18 In addition, there is evidence that residual infectivity may remain until all individual prion molecules have been
degraded.19 The risk of iatrogenic transmission of vCJD in
countries where the disease has been observed remains
unclear. Precautionary principles apply in these countries,
leading to the quarantining of identified suspect instruments
despite the claims from AERs and/or cleaning chemistry
28
R. Hervé, C.W. Keevil / Journal of Hospital Infection 83 (2013) 22e29
A
30
B
***
Residual soil (ng/mm2)
25
20
15
10
5
+++
*
+++
***
0
Dried
Wet
Figure 4. (A) Residual soil after enzymatic wash with brushing. Values are mean standard error of the mean from at least five separate
experiments. Statistical significance is reported as follows: comparison of brushing with no brushing, *P 0.05, ***P 0.001; comparison
of T-type with O-type channels for the same condition, ---P 0.001; comparison of dry with wet conditions, þþþP 0.001.
(B) Representative pictures of brushing marks observed in each channel type. Bars are 100 mm.
manufacturers. The amount, type and potential infectivity of
residual contamination in endoscopes in current use in various
parts of the world remain difficult to ascertain in the absence
of systematic surveys. With currently available decontamination technologies, endoscopes present a favourable field for
iatrogenic cross-infections, and incidents are likely to occur
locally and sporadically.
In addition to strict adherence to current recommendations
and good clinical practice, improvement of decontamination
procedures, equipment and chemistries for the reprocessing
of flexible endoscopes could lower the risk of cross-infection
with various pathogens. While limiting potential instrument
damage, more effective cleaning would reduce concerns and
the need for quarantining instruments when this is applied,
with all the associated delays and costs.
Acknowledgements
The authors wish to thank all clinical staff in various
endoscopy units who collaborated in this study.
Conflict of interest statement
None declared.
Funding source
This is an independent report commissioned and funded by
the Policy Research Programme of the Department of
Health. The views expressed are not necessarily those of the
Department of Health.
References
1. Pajkos A, Vickery K, Cossart Y. Is biofilm accumulation on endoscope tubing a contributor to the failure of cleaning and decontamination? J Hosp Infect 2004;58:224e229.
2. Prusiner SB. Novel proteinaceous infectious particles cause
scrapie. Science 1982;216:136e144.
3. Axon AT, Beilenhoff U, Bramble MG, et al. Variant CreutzfeldtJakob disease (vCJD) and gastrointestinal endoscopy. Endoscopy
2001;33:1070e1080.
4. Dunstan RA, Alpers MP. Variant Creutzfeldt-Jakob disease: implications for the health care system. Aust N Z J Public Health
2005;29:308e312.
5. Lumley JS. The impact of Creutzfeldt-Jakob disease on surgical
practice. Ann R Coll Surg Engl 2008;90:91e94.
6. Pana A, Jung M. Prion diseases and iatrogenic infections I. A
review. Ig Sanita Pubbl 2005;61:325e377.
7. Lipscomb IP, Pinchin HE, Collin R, Harris K, Keevil CW. The sensitivity of approved Ninhydrin and Biuret tests in the assessment of
protein contamination on surgical steel as an aid to prevent
iatrogenic prion transmission. J Hosp Infect 2006;64:288e292.
8. Sciortino Jr CV, Xia EL, Mozee A. Assessment of a novel approach
to evaluate the outcome of endoscope reprocessing. Infect
Control Hosp Epidemiol 2004;25:284e290.
9. Garske T, Ghani AC. Uncertainty in the tail of the variant
Creutzfeldt-Jakob disease epidemic in the UK. PLoS One
2010;5:e15626.
10. ISO 15883-1. Washer disinfectors e part 1: General requirements,
terms and definitions and tests. Geneva, Switzerland:
International Organization for Standardisation. 2006.
11. ISO 15883-4. Washer disinfectors e part 4: Requirements and
tests for washer disinfectors employing chemical disinfection for
R. Hervé, C.W. Keevil / Journal of Hospital Infection 83 (2013) 22e29
12.
13.
14.
15.
thermolabile endoscopes. Geneva, Switzerland: International
Organization for Standardisation. 2008.
Nelson DB, Muscarella LF. Current issues in endoscope reprocessing and infection control during gastrointestinal endoscopy. World
J Gastroenterol 2006;12:3953e3964.
Hervé R, Collin R, Pinchin HE, Secker T, Keevil CW. A rapid dual
staining procedure for the quantitative discrimination of prion
amyloid from tissues reveals how interactions between amyloid
and lipids in tissue homogenates may hinder the detection of
prions. J Microbiol Methods 2009;77:90e97.
Lipscomb IP, Sihota AK, Botham M, Harris KL, Keevil CW. Rapid
method for the sensitive detection of protein contamination on
surgical instruments. J Hosp Infect 2006;62:141e148.
Hervé R, Secker TJ, Keevil CW. Current risk of iatrogenic
Creutzfeld-Jakob disease in the UK: efficacy of available cleaning
16.
17.
18.
19.
29
chemistries and reusability of neurosurgical instruments. J Hosp
Infect 2010;75:309e313.
Secker TJ, Herve R, Keevil CW. Adsorption of prion and tissue
proteins to surgical stainless steel surfaces and the efficacy of
decontamination following dry and wet storage conditions. J Hosp
Infect 2011;78:251e255.
Lipscomb IP, Sihota AK, Keevil CW. Comparison between visual
analysis and microscope assessment of surgical instrument cleanliness from sterile service departments. J Hosp Infect 2008;68:52e58.
Giles K, Glidden DV, Beckwith R, et al. Resistance of bovine
spongiform encephalopathy (BSE) prions to inactivation. PLoS
Pathog 2008;4:e1000206.
Muller H, Stitz L, Wille H, Prusiner SB, Riesner D. Influence of
water, fat, and glycerol on the mechanism of thermal prion
inactivation. J Biol Chem 2007;282:35855e35867.