Introduction
Colonoscopic polypectomy (CP) is one of the most important endoscopic interventions
as it is associated with a reduction in incidence of colorectal cancer [1]
[2]. Low-volume endoscopists have higher rates of complications related to CP [3]. Achievement of technical competence with CP requires a long learning curve and
a dedicated training program, and above all, appropriate simulators for training [4]
[5]
[6]. However, the number of training simulators available for colonoscopic polypectomy
is small [6]
[7]
[8]
[9]. Their functional parts are made from raw pig intestine. This material offers limited
advantages for polypectomy such as submucosal injection [10]. Moreover, to use these ex vivo animal platforms, special animal use endoscopes,
organ preparation, extensive setup, and disposal processes are required. Because of
the quickly degrading material, a continuous cool chain is required [10]. Finally, the training possibilities in ex vivo simulators can be considered limited.
To improve endoscopy training during courses, we developed two novel simulators to
allow for training of colonoscopic polypectomy, the magnetic system-based simulator
(MSPS) and the simulator for polypectomy with high frequency current (HFPS). They
offer sessile and pedunculated polyps, and perfect ergonomic training stations in
combination with endoscopy towers.
This study aimed to describe and establish face, content, and preliminary assessment
of construct validity for these two novel mechanical simulators and to assess their
perceived utility as training and assessment tools.
Methods
Simulators
The MSPS and HFPS presented, including the polyps, were developed and custom-made
by one of the authors (EF). The size of the simulators is 24 cm x 13 cm, excluding
the detachable intubation tube. The weight of the individual simulators, including
the colon modules, is 560 g. The housing consists of a cylindrical part, which can
be screwed off, and a base module ([Fig. 1a, ]
[Fig. 1b, ]
[Fig. 1c]).
Fig. 1 a The housing of the simulator (below) consists of a cylindrical part and the base module
to which it is screwed on. The simulator is fixed to an Olympus trolley with a custom-made
mount. The endoscope is inserted through the intubation tube. The endoscopic image
shows the colon for MSPS equipped with a pedunculated polyp. The arrangement of the
simulator and endoscopy trolley makes a perfectly ergonomic training station. b The cylindrical part of its housing has been screwed off, exposing the interior of
the base module. The base module consists of the white lid of the housing, reinforced
with two round, green metal plates and the cylindrical white holder for the colon
module. The colon module consists of the rubber colon inserted into its Plexiglas
sleeve, which is pushed onto the holder. The base module shown is the one used for
the HFPS. The wiring connects the polyp sockets, which can be seen sticking out of
the colon wall. c Inner and outer aspects of the HFPS base module (green, in the foreground). In the
background are the three different colon modules. Left: HFPS colon module with the
wiring for the polyps. The black plug fits into the inner part of the main socket
(left base module). The outer part (right base module) of the main socket serves the
connection to the electrosurgical unit needed for HF polypectomy.Middle: MSPS colon
module for pedunculated polyps. The visible stalk belongs to one of the stalked polyps
inside.Right: MSPS colon module for sessile polyps. The two magnets fixed to the outside
of the colon keep the two sessile polyps inside the colon in position.All the colon
modules are easily interchangeable in the same simulator.
The base module is composed of the lid of the cylindrical part, which is reinforced
with aluminum discs, a cylindrical holder for the colon module, and the intubation
tube. The colon module consists of a custom-made cylindrical rubber colon, measuring
198 x 53 mm, which is inserted into a sleeve made of Plexiglas ([Fig. 1b, ]
[Fig. 1c]). The colon is provided in two versions, with ([Video 1,]“02:16”) and without ([Video 1,]”00:22”) haustration, and has holes for fixing the polyps.
Video 1 (00:00–00:22) MSPS mounted on the trolley. Rotation of the simulator, outside and
inside view. (00:22–03:50) Polypectomy of different-shaped polyps with the principle
of self-repair used in MSPS. (03:50–05:39) Polypectomy of pedunculated polyps with
HF current.
The colon module containing the polyps can be easily exchanged after unscrewing the
cylindrical housing ([Fig. 1b]).
The simulators can be mounted onto endoscopy towers ([Fig. 1a]) with custom-made holders, compatible with various endoscopy trolleys. They allow
for fixation of the simulators at the appropriate height, and an easy 360-degree rotation
for changing the position of the polyps ([Video 1]).
Specially designed transport cases have been made to facilitate packaging and shipping
of the simulators to training course locations ([Fig. 2]). They measure 67 x 46 x 30 cm, accommodating up to three simulators each, including
the accessories needed.
Fig. 2 Transport case packed with three simulators, MSPS/HFPS, plus accessories. Please
take note the extraordinarily small size of the simulators, which facilitates packaging
and transport.
Specific features of the MSPS
The MSPS colon module can be equipped with pedunculated and sessile rubber polyps
of various shapes and sizes ([Fig. 3]), not all of which are lifelike. The small spherical heads of the pedunculated polyps
have a diameter of 12 mm and 14.5 mm. The large head of the cylindrical spiky polyp
is 26 x 36 mm. The stalk of the polyps has a diameter of 4 mm, and is up to 60 mm
long. The width of the spiky worm polyp is 16 mm and its working length is 12 cm.
The stalk of the polyps is divided into two parts, each of which contains a magnet
that keeps them together. For attaching the polyps to the colon, the stalks of the
pedunculated ones are threaded through the holes of the colon from the inside out.
The length of the stalks inside the colon can be adjusted, depending on how far the
stalks are pulled out. They are held in place by the elasticity of the colon wall,
since the diameter of the holes is slightly smaller than the diameter of the stalks.
A successful polypectomy is carried out once the snare has been precisely directed
to the point where the two magnets meet. Once the snare is closed for the cut, the
two magnets are separated by the traversing snare ([Video 1, ]“00:22”), which when set free indicates a correct polypectomy. The magnets will snap
back instantly, “repairing” the cut stalk ([Video 1, ]“00:22–03:49”). If the stalk is pulled apart by incautiously pulling the snare, without
having it placed correctly at the gap, it can be endoscopically repositioned by the
trainee ([Video 1, ]“00:22).
Fig. 3 Pedunculated and sessile polyps for MSPS. All polyps contain magnets that self-repair
after “polypectomy.” Some of the polyps are not lifelike, e. g. the large spiky polyp
and the spiky worm polyp. They were created to make polypectomy more challenging,
and to train extraordinary skills in young colonoscopists. Lifelikeness is not considered
as a value of its own.
The sessile polyps are fixed to the colon with magnets. One of them is integrated
into the center of the polyp and the other one onto the outer surface of the colon
wall ([Fig. 1c]). The diameters of the polyps vary between 7 mm and 24 mm and the circumferences
are smooth or star-like. Large star-like polyps are more difficult to grasp in one
piece. Cutting is achieved by closing the snare, which then slides through the gap
between the polyp and colon wall, setting the snare free and indicating a successful
polypectomy ([Video 1,]”03:07”). The two magnets keep the polyp in its original position, both during and
after the cut.
Specific features of the HFPS
This simulator is only equipped with pedunculated polyps ([Fig. 4a]), which can be cut by HF-current. They are a hybrid of two different materials.
The reusable rubber heads of the polyps are identical to the ones used in the MSPS
described above. The stalk is made of material called DCM, which is made from an innocuous
material, developed specially for this simulator, from processed animal material available
in regular commercial stores. Less than 1 g of this material is needed per polyp. The
polypectomy carried out on the DCM material with HF current is quite realistic. The
lightened coagulation area and smoke obscuring the view during the cut ([Video 1,]”03:50”) closely resemble the polypectomy in patients. DCM is a convenient alternative
to slaughterhouse material used for ex vivo simulators, as it does not require an
uninterrupted cool chain. It is stable up to 2 to 3 days at room temperature. The
stalk has a usable cutting length of up to 35 mm, and a diameter of 4.5 mm. A socket
is integrated at the free end of the stalk for connecting the polyp to the surgical
HF generator ([Fig. 4a, ]
[Fig. 1b]). For training courses, the colon modules are preloaded with the needed polyps (up
to six per module) ([Fig. 4b]). More polyps can be installed as well, reducing the distance between the polyps.
During the courses, the modules are kept in airtight bags prior to their use to prevent
the polyp stalks from drying out. When the cylindrical enclosure of the simulator
has been unscrewed, the modules with the cut polyps can be replaced with fresh ones
([Video 1,]”05:02”).
Fig. 4 a Pedunculated hybrid polyps for the HFPS. The reusable heads made of rubber are fixed
onto the DCM stalks, a new material which can be cut with HF-current. At the ends
of the polyps, sockets are integrated for the connection to the surgical HF generator.
The wiring of the polyps for connection to the HF electrosurgical generator is shown
in Fig. 1b and Fig. 1c. b For training courses, the readily exchangeable colon modules can be equipped with
six (or even more) polyps.
Endoscopic equipment
A standard colonoscope (CF-H180, Olympus Germany, Hamburg, Germany) was used. In MSPS,
polypectomy was done using a monofilament oval polypectomy snare (SD-990-25, Olympus
Germany, Hamburg, Germany). In HFPS model, polypectomy was done with a braided electrosurgical
polypectomy snare (SD-210U-25, Olympus Germany, Hamburg, Germany) and with the use
of a electrosurgical generator (VIO 300 D, Erbe Elektromedizin, Tübingen, Germany).
Set-up for evaluation
Two pedunculated polyps were fixed in the MSPS model. The first proximal polyp (P1)
had a smooth surface and a diameter of 12 mm, the second distal polyp (P2) had a spiky
one and the diameter of the head was 26 x 36 mm. In the HFPS model, one polyp was
fixed distally.
Study design
A total of 27 participants were recruited for this study. There were 10 novices, seven
intermediate-level, and 10 advanced endoscopists. No participant had used the MSPS
or the HFPS simulators before the study. Participant demographic information, including
the endoscopic experience and the number of colonoscopic and polypectomy procedures,
were collected. Each participant received an orientation including explanation and
inspection of the simulators. Moreover, they were informed about the endpoint of the
study, which was defined as the time needed for completion of each task. Afterwards,
each participant received an individual brief demonstration of the simulators and
operation of the colonoscope and the endoscopic instruments. Finally, each participant
was asked to complete the two simulation models, starting with MSPS with which the
participant was asked to perform two polypectomies for the two fixed polyps starting
with the first polyp (P1) and then continuing with the second polyp (P2). The time
needed for each polypectomy was recorded. Afterwards, the participant moved to the
HFPS and asked to perform a polypectomy using high-frequency current and then to retrieve
the polyp outside the simulator. The time needed for polypectomy and the polyp retrieval
was recorded. Five minutes were allowed at maximum to complete the tasks in each simulator
([Fig. 5]).
Fig. 5 Study design.
Outcome variables and data collection
The primary outcome of this study was to establish construct validity by investigating
whether MPSP and HFPS simulators can distinguish between novices, intermediates, and
experts. The preliminary assessment of construct validity was based on the median
time needed to complete the tasks. This concept has been used in previous pilot studies
[11].
Secondary endpoints included establishment of face and content validity. To assess
the face validity, the non-expert (novice and intermediate) and expert participants
filled out an evaluation questionnaire regarding credibility of each simulator as
an adjunct to colonoscopy training after completing the previously described tasks.
To assess content validity the expert participants completed a questionnaire grading
different aspects of the simulators’ realism and their usefulness for training after
completing the previously described tasks. The questionnaires were adapted from previous
published studies and the following criteria were scored using a 7-point Likert scale
(1 = poor, 7 = excellent) [12]
[13]
[14].
Face validity questionnaire: (1) Practicing with this simulator improved my specific
skills; (2) I would like to continue exercising with this simulator in my further
training; (3) training with this simulator reduces the risks for patients in my opinion;
and (4) I would recommend this simulator.
Content validity questionnaire: Realism: (1) The models are easy to handle (preparation,
down time, complexity during the course); (2) Practicing with the model is advantageous
during a training program; (3) Models can be easily integrated into a training program;
(4) Models display enough realism for the practiced procedure; and (5) Training with
the models leads to a direct improvement of the trainee.
Participants
Novices were medical students (n = 8) and medical residents (n = 2) who didn’t have
any flexible endoscopy experience before the study.
Intermediates were gastroenterology (n = 6) and surgical (n = 1) residents or fellows
who performed more than 50, but less than 200 colonoscopies.
Experts were gastroenterology (n = 7) and surgical (n = 3) staff endoscopists who
had performed more than 200 colonoscopies.
Data analysis
All analyses were performed using SPSS, Version 23. Results of Likert scaled questions
were summarized as median and range. Distributions of duration times stratified for
participants’ experience are illustrated by boxplots. To test for differences across
the study groups, Kruskal-Wallis tests were performed in a first step. If the null
hypothesis could be rejected, Mann-Whitney U tests for pairwise group comparisons
were conducted afterwards. A level of significance of 5 % was used for each test.
Results
A total of 27 /27 participants completed the task on the two simulators.
Primary outcome (construct validity)
Median time needed by the experts to complete the MSPS was 60 seconds (range, 30–150),
which was significantly faster than the intermediate subjects’ time (98 seconds (range,
81–195), P = 0.005. The median time needed by the novice participants to complete the MSPS was
significantly slower than the intermediate and expert groups (217 seconds (range,
105–300), P = 0.007 and P < 0.001 respectively).
Median time needed by the experts to complete the HFPS was 46 seconds (range, 31–105),
which was significantly faster than the intermediate subjects’ time of 78 seconds
(range, 50–140), P = 0.012). The median time needed by the novice participants to complete the HFPS
was significantly slower than the intermediate and expert groups (123 seconds [range,
100–280], P = 0.008 and P < 0.001 respectively). ([Fig. 6, ]
[Fig. 7]).
Fig. 6 Box lot showing the minimum, maximum, interquartile range and median duration of
performed polypectomy using MSPS by three different levels of participants.
Fig. 7 Box plot showing the minimum, maximum, interquartile range and median duration of
performed polypectomy using HFPS by three different levels of participants.
Secondary outcome
Face validity
All participants (10 novice, 7 intermediate, and 10 experts) filled out the first
questionnaire after completing each simulator. Median scores of face validity are
detailed in [Table 1]. Skills improvement, willingness to continue to use in training, risk reduction
for patients and finding the simulator recommendable were overall scored median values
of 6 (4–7), 5 (3–7), 6 (4–7), 6 (3–7) regarding MSPS and 6 (3–7), 6 (4–7), 6 (3–7),
7 (4–7) regarding HFPS. The first two statements were rated significantly higher by
non-experts in comparison to experts.
Table 1
Face validity: median score ratings between non-experts and experts (n = 27) regarding
the MPSP and HFPS simulators.
|
Face Validity Questionnaire
|
MPSP
|
HFPS
|
|
Median score (range)
|
Median score (range)
|
|
Non-experts
(n = 17)
|
Experts
(n = 10)
|
Overall
|
P Value
|
Non-experts
(n = 17)
|
Experts
(n = 10)
|
Overall
|
P Value
|
|
Practicing with this simulator improved my specific skills
|
6 (4–7)
|
5 (4–6)
|
6 (4–7)
|
0.020
|
6 (3–7)
|
5 (3–6)
|
6 (3–7)
|
0.015
|
|
I would like to continue exercising with this simulator in my further training
|
6 (4–7)
|
5 (3–7)
|
5 (3–7)
|
0.021
|
6 (4–7)
|
5 (4–6)
|
6 (4–7)
|
0.016
|
|
training with this simulator reduces the risks for patients
|
7 (4–7)
|
6 (4–7)
|
6 (4–7)
|
0.541
|
6 (3–7)
|
5.5 (3–7)
|
6 (3–7)
|
0.230
|
|
I would recommend this simulator
|
7 (3–7)
|
6 (4–7)
|
6 (3–7)
|
0.667
|
7 (4–7)
|
6 (4–7)
|
7 (4–7)
|
0.425
|
Median scores and range on a 7-point Likert scale (1, complete disagreement; 7 complete
agreement).
Content validity
All 10 experts filled out the second questionnaire after completing each simulator.
Ease of use, utility as a training modality, ease of integration, realism and learning
success of the trainee were scored median values of 6 (4–7), 6 (5–7), 6 (5–7), 6 (5–7)
and 6 (5–7) regarding MSPS and 6 (5–7), 6 (5–7), 6 (5–7), 6 (5–7) and 6 (4–7) regarding
HFPS ([Table 2]).
Table 2
Content validity: experts’ evaluation scores (n = 10) of the MSPS and HFPS simulators.
|
Evaluation item
|
MSPS
|
HFPS
|
|
Ease of use
|
6 (4–7)
|
6 (5–7)
|
|
Utility as a training modality
|
6 (5–7)
|
6 (5–7)
|
|
Ease of integration
|
6 (5–7)
|
6 (5–7)
|
|
Realism
|
6 (5–7)
|
6 (5–7)
|
|
Learning success of the trainee
|
6 (5–7)
|
6 (4–7)
|
Median scores and range on a 7-point Likert scale (1, complete disagreement; 7 complete
agreement).
Discussion
Achieving competence with CP and objective assessment of this technique pose a significant
challenge [15]. Traditionally, procedure numbers are used to measure the level of technical competency.
However, the number of procedures varies between different training programs and does
not correlate with the quality of performance [16]
[17]. For that purpose, some CP simulators have been developed. However, very few of
these simulators have been independently validated as training or assessment tools
[6]
[16]. Moreover, most of these simulators are available as ex vivo versions, which combine
fresh animal parts with housings made from metal and/or plastic [6]
[7]
[8]
[9]. Ex vivo organs are praised for their lifelike properties and for being realistic.
However, it is doubtful whether this assessment is justified, and whether it is a
relevant criterion at all. Cadaver colons are limp, as they lack tone and any haustration.
Moreover, their original anatomy is not accurately represented because they are detached
from their specific intraabdominal fixation, and are instead housed in metal spirals
[9] or exhaust pipings [6]. An unequivocal lifelike feature present in the decomposing ex vivo materials is
the smell.
This study describes and demonstrates the preliminary assessment construct validity
in addition to face and content validity of the two novel CP mechanical simulators,
namely MSPS and HFPS.
The MSPS can be equipped with various artificial rubber polyps. All of them are self-repairing
after the cut, and therefore, they do not need to be restored or replaced during the
training. Although some of the polyps have unrealistic shapes (e. g. spiky head polyp
and spiky worm polyp), they were purposefully created to elicit and train ultimate
skills of the trainee like passing a snare over difficult spiky structures (representing
strongly lobulated large polyp heads) or to free a snare caught in the middle of those
spikes ([Fig. 3, ]
[Video 1] “01:23–03:08”). This needs to be done carefully, otherwise the magnets holding together
the stalk will separate, representing a fragile thin stalk in a patient. The snare
then needs to be passed to a tiny specific point, exactly to the gap between the magnets
and only then can the cut be carried out ([Video 1,]”00:22–03:50”). Trying to overcome these challenges is an ideal preparation for a
real-life polypectomy.
After physical snaring, electrophysical polypectomy techniques can be practiced with
the HFPS, which is equipped with pedunculated hybrid polyps suitable for diathermic
HF-polypectomy training. The reusable rubber heads of the polyps, which are used in
MSPS as well, offer a variety of polyp shapes, which is not feasible with polyps made
from pig colons [6]. The DCM stalk, the second material used for the hybrid polyps, has a socket at
one end for an easy connection to the HF electrosurgical generator ([Fig. 1b, ]
[Fig. 1c, ]
[Fig. 4a]). DCM is a convenient material that does not require a steady cool chain, unlike
fresh pig colons, and is stable for 2 to 3 days at room temperature. This makes the
shipping of the material to the training courses less complicated. The stalks are
adjusted to be soft enough so that, if the snare is negligently closed too fast, it
causes an inadvertent cut. The consequence of this mechanical “cold” cut, which happens
inadvertently in patients as well, would be bleeding from the stalk. DCM is firm enough
to allow careful and snug grasping of the stalk with the electrosurgical snare, before
transmitting HF current for coagulation. During proper application of the current,
the DCM-stalk produces the same effects as diathermic cutting in patients: smoke emanating
from the coagulation area obscuring the view of the cutting site and a lightened area
of coagulation. After polypectomy, the polyps can be caught and retrieved. ([Video 1,]”03:50”)
The artificial colons used in MSPS and HFPS offer additional advantages over the pig
colons. They can be made in different forms, including haustra, which are absent in
ex vivo colons. Polyps located in haustrated segments add to the difficulties of carrying
out polypectomy training ([Video 1,]”01:23–03:08”).
There are only two advantageous features of ex vivo polypectomy simulators: Submucosal
injection techniques can be trained, and sessile polyps suitable for HF polypectomy
can be made from this material. However, the polyps being sutured in, made by ligation
of the mucosa, or created by injection of gelatin into the submucosa are not really
lifelike [6]
[8]
[9]. Nevertheless, there is no alternative for HF-current polypectomy of sessile polyps.
MSPS for sessile polyps offers only the physical, but not the electrophysical, part
of training.
There is no one perfect simulator available for all aspects of polypectomy training
courses. Therefore, looking for a suitable polypectomy simulator means always searching
for the best compromise. The new MSPS and HFPS offer quite a few favorable features
needed for comprehensive training. They avoid the inconveniences that are related
to the simulators based on slaughterhouse materials. Notwithstanding their very small
size, they offer excellent training possibilities, overall exceeding the currently
available simulators. They are part of a system, aimed at facilitating the establishment
of high-quality training courses. Due to their extraordinarily small size, they can
beeasily packed and shipped to training locations ([Fig. 2]). The uncomplicated set-up of the simulators on the tower through a new fixing system
provides an ergonomically perfect working station ([Fig. 1, ]
[Video 1]). The trainee can stand in front of the tower, facing the monitor, and have the
simulator in the same position as the patient. The easy and quick exchange of the
colon modules ([Video 1,]”05:01”), and the self-repairing polyps ([Video 1,]”00:22–03:10”), extend the training time for the trainees. In addition, compared
to non-haustrated colons currently available, haustrated ones are offered as well.
Effortless and swift rotation of the simulator enables easy repositioning of the polyp,
varying the challenge of snaring. It is worth pointing out that the MSPS and HFPS
can easily be reduced to one simulator. To achieve this, only the colon module needs
to be exchanged ([Fig. 1b, ]
[Fig. 1c]). This does not take more than 1 to 2 minutes. The sum of the properties mentioned
also makes the new polypectomy simulators a standardized tool for objective assessment
of endoscopic skills in all endoscopists, not just novices.
One could criticize the devices presented for not simulating negotiation of the scope
through a tortuous colon. However, advancing the scope through a difficult colon requires
a different kind of simulator, which we have also developed. These simulators have
produced favorable results in training that will be published elsewhere.
We were able to preliminary demonstrate that both simulators have high construct validity
as they can discriminate between endoscopists with different levels of experience
regarding the time needed to complete different polypectomy tasks. We were also able
to show that both simulators have excellent face validity across a range of parameters.
The simulators received favorable scores from novice and intermediate participants
regarding improving endoscopic skills, risk reduction for the patient, and use for
further training, and the majority judged the simulators as recommendable. In addition,
both simulators showed high content validity as they received high scores from experienced
endoscopists regarding ease of use, utility as a training modality, ease of integration,
realism, and learning success of the trainee. These validity tests demonstrated that
MSPS and HFPS can be used as training models for CP.
Conclusion
In conclusion, this paper presents and validates two novel simulators for colonoscopic
polypectomy, MSPS and HFPS, which are equipped with new self-repairing polyps and
hybrid polyps. They avoid the use of fresh animal organs and the inconveniences related
to them and offer, at the same time, a combination of training facilities that exceed,
by far, the status quo. This is achieved by sacrificing lifelikeness, but that actually
enhances the training outcome.
