Keywords Endoscopy Lower GI Tract - Colorectal cancer - Endoscopic resection (polypectomy,
ESD, EMRc, ...)
Introduction
Endoscopic treatment is a minimally invasive and widely used modality for superficial
colorectal lesions. The European Society of Gastrointestinal Endoscopy guidelines
state that polypectomy or endoscopic mucosal resection is the strongly recommended
treatment for most superficial colorectal tumors [1 ]. However, local recurrence is found in approximately 12% to 21% of cases in which
resection, including piecemeal resection, is performed [2 ]
[3 ]. Moreover, accurate pathological diagnosis is unattainable in the specimens obtained
by piecemeal resection, owing to the burning effect and inadequate orientation of
multiple specimens. Endoscopic submucosal dissection (ESD) achieves secure en bloc
resection regardless of specimen size and can overcome the abovementioned limitations
of endoscopic resection [4 ]. However, performing ESD requires the expertise of an experienced endoscopist; therefore,
it is not generally used in several countries, except Japan and other Asian countries
[5 ]
[6 ]
[7 ]
[8 ]
[9 ]. We believe that if the technical hurdles of ESD are reduced, it would become more
popular and benefit patients worldwide.
Recently, the effectiveness of underwater techniques for endoscopic procedures has
been reported [10 ]. In colorectal ESD, usefulness of the underwater techniques has been described and
it eliminates the gravity effect, utilizes buoyancy as a natural traction, and creates
a magnified view owing to the difference in photorefraction between water and air
[11 ]
[12 ]
[13 ]. In addition, we previously reported on the water pressure method (WPM), characterized
by not only underwater conditions but also use of active water pressure via the water-jet
function, which widens the submucosa more, helps the endoscope get beneath the lesion,
and enables submucosal dissection in narrow spaces [14 ]
[15 ]
[16 ]. WPM has been reported to overcome difficult situations such as fibrosis in colorectal
ESD [17 ]. Thus, we hypothesized that WPM would reduce technical difficulty and allow novice
endoscopists to perform colorectal ESD. This study aimed to elucidate the usefulness
of colorectal ESD with WPM by novices.
Materials and methods
Study design
This was an ex vivo study using an excised bovine rectum in our institution. Three
novices with no experience in clinical cases of colorectal ESD were enrolled. Each
novice performed conventional ESD (C-ESD) and ESD with WPM (WPM-ESD) 10 times; they
alternated between C-ESD and WPM-ESD to eliminate any bias introduced owing to the
learning effect ([Fig. 1 ]). Treatment outcomes were then compared between the two groups. This study used
only excised bovine rectum; thus, Investigational Review Board approval and written
consent were not required.
Fig. 1 Flowchart of this study.
Animal training model and target lesions
A frozen bovine rectum was used for the experiment. The overtube was attached to the
bovine anus, and the blind end was made by tying a string approximately 40 to 50 cm
from the anus. Mesh copper was wrapped around the rectum and connected to an electrical
surgical unit (ESU). The animal model was placed in a box and fixed to prevent it
from moving. The lesion was created by marking a 2-cm diameter plastic disk placed
on the gravitational side within 30 cm of the anus.
Conventional method
C-ESD was performed according to the following procedure: 1) initial mucosal incision
of the proximal side; 2) creation of a mucosal flap by dissection of the proximal
submucosa; 3) full circumferential incision; and 4) submucosal dissection of the residual
area [18 ].
Water pressure method
WPM was performed according to the following procedure: 1) aspiration of intraluminal
air in the working space and irrigation of normal saline through the water-jet function
(creating underwater conditions); 2) initial mucosal incision and trimming of the
distal side to create an endpoint of submucosal dissection; 3) full circumferential
incision; 4) creation of a mucosal flap by dissection of the proximal submucosa with
application of active water pressure from the water-jet function to widen the submucosal
layer; 5) dissection of both lateral edges of the submucosa using active water pressure;
and 6) submucosal dissection of the central area ([Fig. 2 ] and [Video 1 ]).
Fig. 2 Water pressure method. a The target lesion was created by marking around a 2 cm diameter plastic disc. b The lesion was on the gravitational side where the water was pooling. c Initial mucosal incision and trimming of the distal side were performed to create
endpoints. d A full circumferential incision was made and a mucosal flap was created using water
pressure. e One-sided submucosal dissection was performed using active water pressure. f Submucosal dissection of the other side was performed using active water pressure.
g Submucosal dissection of the residual central area was performed. h En bloc resection.
Colorectal ESD with water pressure method.Video 1
Device and equipment
ESD was performed using a therapeutic endoscope with water-jet function (GIF-H290T;
Olympus Medical Systems, Tokyo, Japan). A short-type small-caliber transparent hood
(DH-28GR; Fujifilm, Tokyo, Japan) was attached to the tip of the endoscope. We used
a 1.5-mm DualKnife J (Olympus Medical Systems, Tokyo, Japan) and submucosal injection
of 10% glycerin solution (Glycerol; Chugai Pharmaceutical Co., Ltd., Tokyo, Japan).
We used VIO 300D (ERBE Elektromedizin, Tübingen, Germany) with Endocut I (effect 2,
cut interval 2, cut duration 2) for mucosal incision and swift coagulation (effect
3, 30 W) for submucosal dissection.
Measured outcomes
We collected data on the following characteristics of enrolled novices: age, years
since graduation from a medical university, whether they were board-certified fellows
of the Japan Gastroenterological Endoscopy Society (JGES), number of esophagogastroduodenoscopy
(EGD) experiences, number of colonoscopy (CS) experiences, and number of ESD experiences.
The primary endpoint of this study was procedure time. The secondary endpoints were
as follows: activation time of ESU during the procedure, interval time during the
procedure, time required for the endoscope to enter the submucosa, en bloc resection
rate, perforation rate, whether the resected specimen had an incision inside the markings,
and the area of the resected specimen (area [cm2 ] = semi-major axis [cm] × semi-minor axis [cm] × π). Procedure time was defined as
the time from initiation of mucosal incision to completion of lesion removal. Procedure
time was divided into activation time of the ESU and interval time at which the ESU
was not activated. Activation time of ESU was measured by VIO DOKU (ERBE Elektromedizin,
Tübingen, DEU), which is a program for maintenance and inspection of the VIO 300D.
The time required for the endoscope to enter the submucosa was defined as the time
from the end of the proximal mucosal incision to complete visualization of the submucosa
by capturing the proximal edge of the lesion using the upper rim of the distal attachment.
Statistical analysis
Fisher’s exact test was used to analyze categorical data. Quantitative data were compared
using the Mann-Whitney U test. Statistical significance was set at P <0.05. All statistical analyses were performed using JMP software (version 16.0.0;
SAS Institute, Cary, North Carolina, United States).
Results
Characteristics of the novices
Characteristics of the three enrolled novices are presented in [Table 1 ]. Their ages were 30 to 33 years, and time since graduation from medical university
ranged from 6 to 9 years. No endoscopists were board-certified fellows of the JGES.
The number of CS experiences ranged from 750 to 2000. None of the participating endoscopists
had any experience in performing colorectal ESD in humans, as mentioned in the inclusion
criteria.
Table 1 Characteristics of the three enrolled novices.
Novice A
Novice B
Novice C
EGD, esophagogastroduodenoscopy; CS, colonoscopy; ESD, endoscopic submucosal dissection.
Age, years
31
33
30
Years since graduation
7
9
6
Board-certified Fellow
No
No
No
Number of EGD experiences
4000
2500
2300
Number of CS experiences
2000
1500
750
Number of esophageal ESD experiences
6
0
0
Number of gastric ESD experiences
35
17
0
Number of colorectal ESD experiences
0
0
0
Treatment outcomes
Each novice performed C-ESD and WPM-ESD (10 times each). The median procedure time
was significantly shorter in the WPM-ESD group than in the C-ESD group (54 minutes;
interquartile range [IQR] 42–67 vs. 45 minutes; IQR, 34–55, P =0.035), and the WPM-ESD group showed a reduction of approximately 17% in the median
procedure time ([Fig. 3 ]). There was no significant difference in the ESU activation time during the procedure
between the two groups (2.8 minutes; IQR, 2.2–3.1 vs. 3.0 minutes; IQR, 2.6–3.5, P =0.11). The interval time during the procedure was significantly shorter in the WPM-ESD
group than in the C-ESD group (52 minutes; IQR, 40–65 vs. 42 minutes; IQR, 32–52,
P =0.030). In addition, the time required for the endoscope to enter the submucosa was
significantly shorter in the WPM-ESD group than in the C-ESD group (8.0 minutes; IQR,
6.0–10 vs. 5.0 minutes; IQR, 3.0–6.0, P <0.001) ([Fig. 4 ]).
Fig. 3 Distribution of the procedure time between the C-ESD group (n=30) and the WPM-ESD
group (n=30). The median procedure time was significantly shorter in the WPM-ESD group
than in the C-ESD group (54 IQR, 42–67 vs. 45 IQR, 34–55 minutes, P =0.035). ESD, endoscopic submucosal dissection; WPM, water pressure method; IQR, interquartile
range.
Fig. 4 Distribution of the activation time of ESU, interval time, and time required for the
endoscope to enter the submucosa between the C-ESD group (n=30) and the WPM-ESD group
(n=30). a There was no significant difference in the ESU
activation time during the procedure between the two groups (2.8 IQR, 2.2–3.1 vs.
3.0
IQR, 2.6–3.5 minutes, P =0.11). b The
interval time during the procedure was significantly shorter in the WPM-ESD group
than
in the C-ESD group (52 IQR, 40–65 vs. 42 IQR, 32–52 minutes, P =0.030). c The time required for the endoscope to
enter the submucosa was significantly shorter in the WPM-ESD group than in the C-ESD
group (8.0 IQR, 6.0–10 vs. 5.0 IQR, 3.0–6.0 minutes, P <0.001). ESD, endoscopic submucosal dissection; WPM, water pressure method;
IQR, interquartile range.
Comparisons of other treatment outcomes between the C-ESD and WPM-ESD groups are presented
in [Table 2 ]. En bloc resection was performed in all cases. Although one perforation occurred
in the C-ESD group, there were no perforations in the WPM-ESD group. In the C-ESD
group, two resected specimens had an incision inside the markings, while in the WPM-ESD
group, none of the resected specimens were cut inside the markings ([Fig. 5 ]). The areas of the resected specimens did not differ between the two groups.
Table 2 Comparison of treatment outcomes between the C-ESD and WPM-ESD groups.
Variables
C-ESD
WPM-ESD
P value
IQR; interquartile range.
En bloc resection
Yes, N (%)
30 (100)
30 (100)
>0.99
Perforation
Yes, N (%)
1 (3)
0 (0)
>0.99
Cutting inside the markings
Yes, N (%)
2 (7)
0 (0)
0.49
Area of resected specimen
Median [IQR], cm2
12 [9.7–14]
13 [11–16]
0.14
Fig. 5 Resected specimens. a One C-ESD case with an incision
inside the marking.b No WPM-ESD cases had incisions inside the
marking. C-ESD, conventional endoscopic submucosal dissection; WPM, water pressure
method.
Discussion
This study successfully revealed that WPM-ESD significantly reduced procedure time
by 17% compared to that of the conventional method when performed by endoscopists
with no experience in colorectal ESD. To the best of our knowledge, this is the first
study to illustrate the usefulness of WPM-ESD for colorectal ESD, especially for novices.
Colorectal ESD is difficult for novice endoscopists because of several factors. First,
the colorectal wall is very thin, and careless passive energization can easily lead
to perforation. Second, the anatomical characteristics of the long luminal organs
often result in poor scope maneuverability [6 ]
[19 ]. Third, the development of this strategy is complex. Although an efficient procedure
needs to be considered by changing the patient’s position to control the direction
of gravity for good traction, the maneuverability of the scope, approachability to
the lesion, and scope angle with the muscle layer can easily worsen depending on the
patient’s body position [20 ]. The key to successful colorectal ESD is controlling these factors and performing
rapid and precise dissection.
Several useful methods have been proposed for successful colorectal ESD, such as the
mechanical traction method and the pocket-creation method (PCM) [18 ]
[21 ]
[22 ]
[23 ]
[24 ]
[25 ]
[26 ]. However, these methods still have certain issues that need to be resolved. Endoscopist
experience is required to choose the position of deployment of the traction device,
and if the position is inappropriate, the situation does not improve. In PCM, it is
difficult to create the first pocket and remove the edges in the last pocket-opening
phase. Inexperienced endoscopists sometimes lose orientation in the pocket and are
unsure of how much submucosal dissection to perform and in which direction [27 ].
In this study, we evaluated not only the total procedure time but also the time of
the various other phases to provide a detailed and objective evaluation of the WPM-ESD.
We divided the procedure time into the activation time of ESU and interval time and
found that WPM-ESD significantly reduced the median interval time by 10 minutes. This
degree of reduction in the interval time was almost the same as the degree of reduction
in the total procedure time, indicating that WPM-ESD was mainly effective in the interval
phase. The interval time (incision and dissection was not performed in this period)
refers to the time spent creating the field of view and preparing it for incision
and dissection. It was objectively shown that WPM-ESD helps create a good field of
view.
There was no significant difference in the activation time of the ESU. In a previous
report, the use of a bipolar knife was recommended in saline immersion because saline
solution is more conductive than tissue, and the use of a monopolar knife in saline
solution would dissipate the electrical energy [12 ]. The results of this study indicated that the DualKnife J did not worsen the efficiency
of the dissection to the extent that would lead to a problem. The short and thin DualKnife
J may create sufficient current density to the tissue even in saline solution and
allow for sharp dissection.
We also measured the time required for the scope to enter the submucosa, which is
the most difficult phase of colorectal ESD, and found that the WPM-ESD group was significantly
faster than the C-ESD group. Until a mucosal flap is created and the dive under the
submucosa has been performed, the visibility of the submucosa is poor, often making
it difficult for a parallel approach to the muscular layer.
Occasionally, blind dissection is crucial, and only experienced endoscopists can perform
this technique [28 ]
[29 ]. The results of this study suggest that WPM-ESD may facilitate this difficult phase,
owing to its buoyancy and the submucosal widening effect, particularly for novice
endoscopists.
In this ex vivo study, the lesion was intentionally created on the gravitational side.
As mentioned above, gravity is a factor associated with colorectal ESD. Underwater
conditions can eliminate the effects of gravity, and hence, it is partly responsible
for the shorter procedure time resulting in the simplification of colorectal ESD,
which may allow novice endoscopists to perform the steps shown in [Fig. 2 ]. In this procedure, the edges are dissected preferentially; therefore, the field
of view expands rapidly, and the orientation can be easily understood by novice endoscopists.
Furthermore, there were no perforations during WPM-ESD. This might have been due to
the effect of buoyancy, widening of the submucosa, and magnified view in the WPM,
which allows for precise, high-quality dissection. In addition, there were no cases
of resected specimen cuts inside the markings. This may be because of the precise
edge processing of the WPM, which makes it possible to safely obtain high-quality
resected specimens that allow for an accurate pathological diagnosis.
WPM-ESD has certain disadvantages in clinical settings. First, visibility can be reduced
by major bleeding or intestinal content. Second, WPM-ESD has a potential risk of causing
abdominal pollution due to the leakage of intestinal contents and tumor cells into
the abdominal cavity. Therefore, the intestinal tract should be sufficiently cleaned
before the procedure, and if perforation occurs, it should be closed as soon as possible
[20 ].
Despite these concerns, the results of this study revealed that WPM-ESD is a useful
technique that creates a good field of view and allows for precise dissection, even
by novice endoscopists; in addition, the gravity factor does not need to be considered.
It is also available at any location and time; thus, it can complement a combination
of other useful methods, such as PCM and mechanical traction [30 ]
[31 ]
[32 ]
[33 ]
[34 ]
[35 ]. In the future, WPM-ESD may reduce the technical difficulties of colorectal cases
and foster popularization of this technique worldwide.
This study had several limitations. First, it was ex vivo. These results did not consider
the various factors present in clinical ESD, such as bleeding, fibrosis, heartbeat,
peristalsis, respiratory fluctuation, maneuverability, and stool. Second, the sample
size was very small. This study includes only three novice endoscopists, which may
not be representative of the larger population of novice endoscopists. It was unclear
whether the results could be generalizable to other novice endoscopists or experienced
endoscopists. Third, a bias was introduced because the endoscopists were not blinded
to the method of the procedure. The endoscopists may have consciously or unconsciously
modified their technique or performance based on their knowledge of whether they were
performing C-ESD or WPM-ESD. Owing to these limitations, the results of our study
should be interpreted with caution. Further prospective studies in clinical settings
are needed.
Conclusions
In conclusion, WPM contributed to the creation of a good field of view and aided in
rapid, safe, and precise dissection, even for endoscopists with no experience in colorectal
ESD. WPM may reduce the technical hurdles of colorectal cases for novice endoscopists.
