Introduction
Despite high-level disinfection (HLD) protocols, several outbreaks of multidrug-resistant
bacterial contamination of duodenoscopes with fixed distal caps have been documented
over the past decade [1 ]
[2 ]. Owing to the complex design with an elevator mechanism, duodenoscopes are more
difficult to clean and disinfect than other endoscopes. This in turn can result in
potential contamination of the duodenoscope even after standard HLD. Duodenoscopes
with a fixed distal cap have a plastic or rubber cap permanently attached to the distal
end; this design limits accessibility during manual cleaning and disinfection, with
the brush being unable to access the distal end of the duodenoscope particularly behind
the elevator. The US Food and Drug Administration (FDA) currently recommends that
health care facilities and manufacturers begin to transition to duodenoscopes with
disposable components in order to reduce the risk of infection in patients [3 ].
Recently, a newly designed duodenoscope with a detachable distal cap has been developed
to make scope reprocessing easier and more effective. These devices may have an advantage
over conventional duodenoscopes, which have a fixed distal cap, by allowing the brush
to access the back of the elevator mechanism. However, no surveillance studies evaluating
contamination rates of these newly designed duodenoscopes have been conducted. Furthermore,
comparative data on the contamination rates after HLD between the two duodenoscope
designs are lacking.
The Centers for Disease Control and Prevention (CDC) recommends surveillance culture
for bacterial contamination from the elevator and the working channel [4 ], and the FDA suggests that endoscope surveillance sampling and culturing are specifically
required or regulated by the state or local authority [5 ]. As the microbiologic testing of duodenoscopes is costly and requires 72 hours for
culture, it remains to be evaluated whether this quarantine approach is practical
and financially feasible in endoscopy centers with limited budgets. Recently, several
studies have proposed the use of the adenosine triphosphate (ATP) test as an alternative
method of point-of-care testing (POCT) to evaluate bacterial contamination and organic
residue in duodenoscopes after HLD [6–16]. However, the recommended cutoff value of
ATP to determine scope contamination has not been determined [6–16].
Our study aimed to compare the proportion of potential bacterial contamination and
organic residue in the newly designed duodenoscopes with detached distal cap vs. those
with a fixed distal cap after HLD by using ATP testing as POCT and bacterial culture
to confirm bacterial contamination after ≥ 72 hours. Furthermore, we validated the
optimal cutoff value of 40 relative light units (RLUs) based on our previous reports
to ensure that the duodenoscopes were free of contamination [15 ]
[16 ].
Methods
The study was conducted at the Excellence Center of Gastrointestinal Endoscopy, King
Chulalongkorn Memorial Hospital, between December 2018 and April 2019. Following use,
all newly designed duodenoscopes with detachable distal caps (ED 580XT; Fujifilm,
Tokyo, Japan) were prospectively recruited to the study. Duodenoscopes were enrolled
and divided into two groups in a 1:1 ratio: in group A the distal cap was detached
prior to reprocessing and in group B the cap remained fixed in place.
As the study did not involve human subjects, it did not require approval from the
institutional review board. The study protocol was registered at the national clinical
registry (TCTR2019209002).
Duodenoscope reprocessing
After endoscopic retrograde cholangiopancreatography (ERCP) procedures, duodenoscope
reprocessing was performed by well-trained staff according to the manufacturer’s recommendations
and standard guidelines [5 ], including precleaning, leak testing, manual cleaning, HLD, drying, and storage.
After precleaning at the point of care, the duodenoscope was transferred to the reprocessing
area using fully enclosed and labeled containers for subsequent steps. Leak testing
was performed before formal reprocessing.
For manual cleaning, the duodenoscope was immersed in enzymatic detergent solution
(3 M Low Foam Ultra Rapid Multi-Enzyme Cleaner; 3 M, Taipei, Taiwan). In group A,
the distal cap of the duodenoscope was detached before manual cleaning ([Fig. 1a,b ]), whereas in group B, the distal cap was not detached ([Fig. 1c,d ]) in order to imitate the conventional duodenoscopes with fixed distal cap. A single-use,
manufacturer-recommended channel and elevator brush (WB1318DE; Fujifilm, Tokyo, Japan)
was used to remove visible debris from the accessible area of the duodenoscope, including
the elevator, suction port, air/water port, and instrument channel port. The elevator
system, including the guidewire-locking groove and the forceps elevator recess, was
cleaned using the brush. The elevator control lever was operated at least three times
to raise and lower the forceps elevator while in the detergent solution. In addition,
the forceps elevator recess, air/water channel, suction channel, and instrument channel
were flushed through with the detergent solution.
Fig. 1 A newly designed duodenoscope with detachable distal cap (ED 580XT; Fujifilm, Tokyo,
Japan). a,b The distal cap was detached before manual cleaning to allow the brush to access the
back of the elevator (white arrow). c,d The distal cap was not detached before manual cleaning in order to imitate the conventional
duodenoscope with a fixed distal cap.
Following manual cleaning, the duodenoscope was placed in an automated endoscope reprocessor
(OER-AW; Olympus Medical Systems Corp., Tokyo, Japan) using a high-level disinfectant
(Acecide; Olympus Medical Systems Corp.). The total cycle reprocessing time was approximately
20 minutes. After HLD, the duodenoscope was rinsed and flushed with filtered water
to remove disinfectant solution and sampling procedures were then performed.
Sampling protocol
All sampling procedures were performed by staff who had received training and were
certified from the ATP manufacturer according to the CDC protocol [5 ]. These were the same staff who performed manual cleaning and they were therefore
not blinded to group allocation. After a single round of HLD, a sample was collected
from the elevator site using swab rotation, with the cap removed in group A and the
cap in place in group B, and submitted for culture. Then, the sample was rapidly evaluated
for bacterial contamination and organic residue by analyzing the RLU with a luminometer
(3 M, Maplewood , Minnesota, USA). The ATP test (Clean-Trace Surface ATP; 3 M) was used to evaluate
the ATP-containing specimen obtained from the elevator.
The gold standard to detect any bacterial growth in samples was by culture for at
least 72 hours. A sterile swab was used to collect the sample from the elevator by
rotating it at the elevator mechanism in the raised/open position and in the lowered/closed
position. Then, the collected sample was spread on to a blood agar plate and cultured
for at least 72 hours. Culture samples from the elevator were kept on cold pack and
promptly transferred to the laboratory. In the laboratory, culture samples from the
elevator on the blood agar plate were incubated at 37 °C. Initial and final culture
results were reported at 72 hours and at 7 days after receiving the specimen, respectively.
If the culture result was positive for bacterial contamination, the bacterial species
were identified and counted (colonies forming units [CFU]/mL), and antibiotic sensitivity
testing was performed. Contamination of the duodenoscope was defined as any growth
of pathogenic organisms such as Enterobacteriaceae species, Staphylococcus aureus , Streptococcus viridans group, and Enterococcus or the growth of more than 10 CFU of nonpathogenic organisms such as coagulase-negative
Staphylococcus , Bacillus , and Diphtheroid
[17 ]
[18 ].
Management
Previous full publications were based on a higher cutoff value (200 RLU) and the specimens
were obtained from the suction channel rather than from the elevator; the results
from those studies showed that contamination was present [7 ]
[8 ]. Subsequent studies using the cutoff value of 200 RLU still showed the inconsistent
relationship between ATP and culture results [10,12–14]. Based on these previous data,
the 2018 CDC guidelines postulated the ATP test as a useful marker for the scope cleaning
process but its sensitivity was not sufficient to be used as a marker for the adequacy
of the HLD process [5 ]. In real clinical practice, we require a rapid screening test to identify potentially
contaminated duodenoscopes that need repeat HLD before they can return to clinical
use; therefore, the test must achieve a perfect negative predictive value (NPV) of
100 %. Our previous data [15 ]
[16 ] on the ATP test demonstrated an area under the receiver operating characteristic
curve of 0.85 (95 % confidence interval [CI] 0.78–0.90) ([Fig. 2 ]); the ATP threshold of 40 RLU had a sensitivity of 100 % (95 %CI 93.3 – 100) and
NPV of 100 % (95 %CI 90.3 – 100), whereas the ATP threshold of 200 RLU had sensitivity,
specificity, positive predictive value, and NPV of 86.8 % (95 %CI 74.7 – 94.5), 59.9 %
(95 %CI 51.5 – 67.9), 43.8 % (95 %CI 34.1 – 53.8), and 92.6 % (85.4 – 97.0), respectively
([Table 1 ]). Furthermore, our additional tests [15 ]
[16 ] in the validated group confirmed that at the ATP threshold of 40 RLU, all culture
samples were negative for pathogenic bacterial contamination (100 % sensitivity and
100 % NPV).
Fig. 2 Receiver operating characteristic (ROC) curve of adenosine triphosphate test from
the elevator compared with culture results (AUROC = 0.851) [16 ].
Table 1
Performance characteristics of adenosine triphosphate test from the elevator sites
compared with cultures.
ATP cutoff, RLU
Sensitivity, % (95 %CI)
Specificity, % (95 %CI)
PPV, % (95 %CI)
NPV, % (95 %CI)
40
100 (93.3–100)
24.5 (178–32.3)
32.3 (25.2–40.1)
100 (90.3–100)
50
98.1 (89.9–100)
35.4 (27.7–43.7)
35.4 (27.7–43.7)
98.1 (89.9–100)
60
98.1 (89.9–100)
41.5 (33.4–49.9)
37.7 (29.6–46.3)
98.4 (91.3–100)
70
96.2 (87–99.5)
46.3 (38–54.7)
39.2 (30.8–48.2)
97.1 (90.1–99.7)
80
96.2 (87–99.5)
46.9 (38.7–55.3)
39.5 (31–48.5)
97.2 (90.2–99.7)
90
94.3 (84.3–98.8)
46.9 (38.7–55.3)
39.1 (30.6–48.1)
95.8 (88.3–99.1)
100
94.3 (84.3–98.8)
46.9 (38.7–55.3)
39.1 (30.6–48.1)
95.8 (88.3–99.1)
150
88.7 (77–95.7)
53.7 (45.3–62)
40.9 (31.8–50.4)
92.9 (85.3–97.4)
200
86.8 (74.7–94.5)
59.9 (51.5–67.9)
43.8 (34.1–53.8)
92.6 (85.4–97)
ATP, adenosine triphosphate; RLU, relative light units; CI, confidence interval.
In the current study, in group A, if the ATP result passed the threshold after HLD,
then the duodenoscope was returned to the clinic ready for use without waiting for
the culture results. If the ATP levels were ≥ 40 RLU, then the duodenoscope was returned
for another cycle of HLD until the ATP level was lower than the threshold before it
could be used for the next patient. For all group B duodenoscopes, regardless of whether
they met the ATP benchmark after initial HLD, repeat cap-off cleaning based on the
manufacturer’s reprocessing instructions was performed before they were returned to
clinical use in patients. When there was disagreement between the POCT and the gold
standard sample culture, for instance when a duodenoscope with an ATP test < 40 RLU
was used in a patient but a positive culture result was subsequently recorded, then
the patient was tracked for clinical evidence of duodenoscope-related infection until
Day 90 after the procedure.
Statistical analysis
We assumed that the newly designed duodenoscope with the detachable distal cap would
have a contamination rate of 1 % compared with conventional fixed-cap duodenoscopes,
which have a contamination rate of 16 % [19 ]. The sample sizes for the two groups were calculated using a two independent proportions
formula with a difference in contamination rate of 15 % between the two groups. A
sample size of 54 in each group would give 80 % power to detect this difference at
a 2-sided significance level of 5 %.
Continuous variables were presented as the median (range) for non-normal distributions.
Categorical variables were reported as numbers and percentages. Using the threshold
of ATP < 40 RLU, we calculated the number and percent of duodenoscopes with organic
residue requiring an additional cycle of HLD to ensure 100 % NPV, and the relative
risk between groups. A two-sided P value of < 0.05 was considered to be significant. Statistical analyses were performed
using SPSS version 23 for Windows software (IBM Corp., Armonk, New York, USA).
Results
During the 5-month study period, 108 used newly designed duodenoscopes (54 in each
group) were enrolled into the study ([Fig. 3 ]). The indications for ERCP were not different between the groups: bile duct stones
(36 [66.7 %] vs. 30 [55.6 %]; P = 0.23), malignant biliary stricture (15 [27.8 %] vs. 20 [37.0 %]; P = 0.30), benign biliary stricture (2 [3.7 %] vs. 3 [5.6 %]; P = 0.65), and pancreatic duct stones (1 [1.9 %] vs. 1 [1.9 %]; P > 0.99) in group A vs. group B, respectively. No patient with a history of carbapenem-resistant
Enterobacteriacae infection was found during the study period.
Fig. 3 Flow chart of the study. ATP, adenosine triphosphate; RLU, relative light units;
HLD, high-level disinfection.
ATP testing
Overall, after a single round of HLD, median ATP value in group A was significantly
lower than that of group B (45.2 RLU [range 9–209] vs. 141.0 RLU [range 19–653]; P < 0.001) ([Table 2 ]). Using 40 RLU as the ATP cutoff value, the proportion of potential bacterial contamination
and organic residue in group A was significantly lower than that of group B (20 /54
[37.0 %] vs. 41/54 [75.9 %]; P < 0.001; relative risk ratio 0.49, 95 %CI 0.33–0.71). Of the 108 used duodenoscopes,
34 in group A and 13 in group B with ATP levels of < 40 RLU after one cycle of HLD
were returned to the clinic ready for use. The remaining duodenoscopes with ATP values
of ≥ 40 RLU after the initial HLD (group A = 20, group B = 41) underwent a second
cycle of HLD until the duodenoscopes passed the ATP benchmark. All 61 duodenoscopes
passed the ATP threshold after the second cycle of HLD and were returned to clinical
use.
Table 2
Comparison of rapid adenosine triphosphate testing and culture results between the
duodenoscope with detached (group A) vs. fixed (group B) distal cap after high-level
disinfection.
Group A
Group B
P value
ATP, median (range), RLU
45.2 (9–209)
141.0 (19–653)
< 0.001
ATP ≥ 40 RLU, n/N (%)
20/54 (37.0)
41/54 (75.9)
< 0.001
Positive cultures, n/N (%)
0/54 (0)
1/54 (1.9)
0.32
0/54 (0)
0/54 (0)
N/A
ATP, adenosine triphosphate; RLU, relative light units; N/A, not applicable.
Culture results
In group A, all 54 culture samples were negative for pathogenic and nonpathogenic
bacterial contamination after HLD ([Table 2 ]). In group B, one sample taken after the first HLD was found to have nonpathogenic
bacterial contamination (coagulase-negative Staphylococcus ) with an ATP value of 137 RLU from the same duodenoscope. Pathogenic bacterial contamination
confirmed by culture was not documented in any of the group B cultures. There was
no disagreement between POCT and the culture result (i. e. no patient developed clinical
evidence of duodenoscope-related infection).
Discussion
Several studies on microbiological surveillance after duodenoscope reprocessing have
shown pathogenic bacterial contamination in duodenoscopes with a fixed distal cap
[1, 20–23]. Recently, the position statement of the European Society of Gastrointestinal
Endoscopy and European Society of Gastroenterology Nurses and Associates recommended
the use of duodenoscopes with a detachable distal cap in order to make it easier to
clean difficult-to-reach areas such as the elevator [24 ]. In 2015, the FDA letter stated that infectious outbreaks of multidrug-resistant
bacterial contamination of duodenoscopes were reported in duodenoscopes with a fixed
distal cap [25 ]. Another nationwide study by Rauwers et al. evaluated the prevalence of bacterial
contamination of 10 different duodenoscope types and designs, including the elevator
channel and the distal cap, and found that the contamination was not associated with
one particular duodenoscope type (P ≥ 0.20) [26 ]. This led manufacturers to design duodenoscopes with detachable caps in order to
reduce the risk of contamination at the elevator site. Our study demonstrated that
the duodenoscope with the detachable distal cap had a significantly reduced amount
of organic material, as tested by ATP, from 76 % to 37 % (P < 0.001) when compared with duodenoscopes with the fixed distal caps. More importantly,
bacterial culture confirmed that none of the duodenoscopes with the detachable distal
cap were contaminated, whereas bacterial contamination from skin and membrane flora
was still detected in one sample from a duodenoscope with a fixed distal cap.
Contamination of the duodenoscope usually develops during ERCP, as many of the indications
for ERCP are related to infection such as acute cholangitis and acute cholecystitis.
Technically, the elevator system of the duodenoscope has a high risk of trapping contaminated
debris and stone fragments. In our study, all specimens were obtained from the elevator
site after standard reprocessing of duodenoscopes that had been used for standard
ERCP performed in real patients. Although we showed a statistically significantly
higher proportion of ATP ≥ 40 RLU in the duodenoscopes with the fixed distal cap compared
with those with detachable caps, none of them were found to be contaminated with pathogenic
bacteria. The only concern is that nonpathogenic bacteria were found in one duodenoscope
with a fixed distal cap (P = 0.32), whereas all the detachable duodenoscopes had negative bacterial cultures
([Table 2 ]). The coagulase-negative Staphylococcus detected in group B is part of the normal flora of human skin (nonpathogenic bacterium).
This in turn may represent contamination from skin or scope contamination during transportation.
Duodenoscopes are complex reusable devices that are composed of hard-to-clean components,
particularly the elevator mechanism. Previous studies have demonstrated higher ATP
levels in samples from the elevator compared with samples from the working channel
of conventional duodenoscopes with a fixed distal cap [6 ]
[7 ]
[12 ]
[14 ]
[27 ]. In the study by the Indiana group focusing on the impact of double HLD, the water
irrigated from both working channels and the elevator of conventional duodenoscopes
with fixed distal caps was randomly selected for bacterial culture. All 120 initial
cultures from the working channels were found to be negative, whereas all positive
cultures were from the elevator [28 ]. Furthermore, the study by Verfaillie et al. reported on a large outbreak of VIM-2-producing
Pseudomonas aeruginosa , which was linked to the use of duodenoscopes with a fixed distal cap and sealed
elevator wire channel port (TJF-Q180V; Olympus) [23 ]. The TJF-Q180V duodenoscopes were evaluated by an expert from Delft University of
Technology and a delegation from Eramus MC at Olympus (Zoeterwoude, The Netherlands);
it was found that the fixed distal cap precluded cleaning and disinfection [23 ]. Eventually, the high monthly number of newly infected patients with VIM-2-producing
P. aeruginosa during the outbreak period (7.5 cases during January to April 2012) decreased to
1.8 cases after the TJF-Q180V device was withdrawn from clinical use [23 ]. These data suggest that bacterial contamination mainly originated in the elevator
mechanism of used duodenoscopes.
The elevator is a complex area of the duodenoscope in terms of cleaning; despite brushing
and adequate HLD, these areas had persistent bacterial contamination [23 ]
[27 ]
[29 ]
[30 ]. Results from our study suggest that the detachable distal cap allows the area behind
the elevator to be cleaned with the brush during manual cleaning, potentially reducing
the contamination rate of the duodenoscopes. Our study also suggests that this new
duodenoscope with a detachable cap should undergo careful manual cleaning using specific
cleaning devices, such as brushes, with special attention paid to the elevator mechanism.
In addition, brushing while raising and lowering the forceps elevator, particularly
at the back of the elevator, should be an important part of this process.
Although the FDA recommends a gradual transition to duodenoscopes with disposable
components in order to reduce the risk of infecting patients, it is important to perform
post-market surveillance using a larger volume of tests than our study in order to
evaluate and confirm the lower contamination rate with these newly designed duodenoscopes
before conventional duodenoscopes with fixed distal caps are withdrawn from the market.
The limitations of our study include the lack of blinding in personnel who performed
the manual cleaning and sampling of the duodenoscopes; however, this would have been
impossible given the need to detach the cap from group A duodenoscopes. Second, the
duodenoscopes with detachable caps used in the study had been developed by a single
company. Additional studies are warranted to test other duodenoscopes with detachable
caps made by other companies. In addition, our group B duodenoscopes were not conventional
fixed-cap duodenoscopes but the same duodenoscopes as those used in group A but with
the cap left in place during cleaning; thus, genuine randomized head-to-head comparisons
of duodenoscopes with detachable vs. fixed distal caps are needed to confirm these
results.
In conclusion, use of a newly designed detachable duodenoscope significantly reduced
contamination with residual organic material, as confirmed by ATP testing and bacterial
culture. The detachable cap provided easier access to the back of the elevator during
manual cleaning with brushes. The ATP test with a cutoff of 40 RLU was the practical
POCT to ensure cleanliness of duodenoscope reprocessing without waiting for the bacterial
culture result.
Acknowledgment
This research was funded by the intramural funding from the King Chulalongkorn Memorial
Hospital, Thai Red Cross, Bangkok, Thailand, and was supported by Fujifilm, Tokyo,
Japan and the Center of Excellence for Innovation and Endoscopy in Gastrointestinal
Oncology, Faculty of Medicine, Chulalongkorn University.
An earlier version of these data were accepted as an oral presentation at the Asian
Pacific Digestive Week, 2019, Kolkata, India, and has been selected for an oral presentation
in the ASGE Presidential Plenary session at the annual meeting of Digestive Diseases
Week, 2020, Chicago, Illinois, USA.
