Introduction
Colorectal endoscopic submucosal dissection (C-ESD) is a promising, minimally invasive
treatment developed as an alternative to surgery. It allows removal of large superficial
colorectal tumors en bloc. However, this may result in the creation of a large mucosal defect, which sometimes
causes adverse events (AEs), such as bleeding, perforation, and electrocoagulation
syndrome [1]
[2]. Preventing those AEs is important for improving the safety and applicability of
C-ESD. Complete closure of such artificial mucosal defects has been reported to reduce
those complications after colorectal endoscopic resection [3]
[4]. However, complete closure of large mucosal defects after C-ESD is considered impossible
in most cases because of the limited width of the open clip. We therefore invented
a simple closure technique using clip-and-line, named “line-assisted complete closure
(LACC)” [5]. Our idea was inspired by a traction technique performed during C-ESD using clip-and-line
[6]
[7]. We herein describe the details of our closure technique, and assess the technical
feasibility of the method.
Patients and methods
Patients
Between January and February 2016, 28 patients underwent C-ESD at Osaka Medical Center
for Cancer and Cardiovascular Diseases. Of these, 11 cases were performed or supervised
by 1 endoscopic specialist (Y. T.), who invented LACC technique, and all the consecutive
11 cases underwent LACC after C-ESD and were included in this retrospective feasibility
study. Outcome measures were procedural success rate, procedure time, and post-procedural
complications, such as bleeding, perforation, and electrocoagulation syndrome. Procedural
success was defined as complete closure of the mucosal defect after C-ESD, which means
covering the mucosal defect with normal mucosa so that the submucosal layer cannot
be seen. Procedure time was measured from insertion of the first clip-and-line until
achieving or giving up complete closure. Post-ESD bleeding was defined as bleeding
requiring emergency endoscopic hemostasis or transfusion, or as a decrease in the
hemoglobin level of more than 2 g/dL following C-ESD. Post-ESD perforation was defined
as no perforation during the C-ESD procedure, followed by sudden appearance of symptoms
of peritoneal irritation after the procedure, with the presence of free air on abdominal
computed tomography (CT) or X-ray. A diagnosis of post-ESD electrocoagulation syndrome
(PECS) was made if the patients presented with localized abdominal pain and fever
(≥ 37.6 °C) or inflammatory response (leukocytosis [≥ 10,000 cells/µL], increased
C-reactive protein [≥ 0.5 mg/dl]), after C-ESD in the absence of visualized perforation
on abdominal CT. This study was approved by the ethics committee at Osaka Medical
Center for Cancer and Cardiovascular Diseases.
Line-assisted complete closure (LACC)
The LACC procedure was conducted immediately after completion of C-ESD using a colonoscope
(PCF-Q260AZI, PCF-Q260JI, or CF-Q260DI; Olympus, Tokyo, Japan) with a distal attachment
(D-201 – 13404 or D-201 – 11804; Olympus), without withdrawal and reinsertion of the
colonoscope. The LACC procedure was performed in a uniform, standardized fashion as
we previously described ([Fig. 1], [Fig. 2], [Fig. 3], [Video 1]) [5]. At first, a long nylon line (e. g. a fishing line) was tied to the teeth of an
endoclip (HX-610 – 090; Olympus), which was attached to an applicator (HX-110LR, Olympus;
[Fig. 1]). Based on a previous report, it was important not to open the endoclip fully at
this moment [6]. The endoclip-and-line was then retracted into the applicator, and the applicator
was inserted into the accessory channel. The endoclip-and-line was then placed on
the normal mucosa, 5 mm from the proximal margin of the wound ([Fig. 2a]). Another endoclip without a line was then inserted through the accessory channel
([Fig. 2b]) and anchored to the other side of the normal mucosa ([Fig. 2c]). Both sides of the wound were gathered by gently pulling the anchored line ([Fig. 2 d]). Additional endoclips were placed to achieve complete closure of the wound ([Fig. 2e]). Multiple endoclip-and-lines can be applied on demand, because LACC does not require
withdrawal and reintroduction of a colonoscope. Only 1 assistant is needed to hold
the line while clipping. The line tied to the endoclip was finally cut using scissor
forceps (FS-3L-1; Olympus, [Fig. 2f]).
Fig. 1 A long nylon line was tied to the arm of an endoclip.
Fig. 2 Schematic diagram of the line-assisted complete closure. a The endoclip and line were placed on the normal mucosa, 5 mm from the proximal margin
of the wound. b Another endoclip without a line was then inserted into the accessory channel. c The line was anchored by another endoclip to the other side of the mucosa. d Both sides of the wound were gathered by pulling the anchored line. e Additional endoclips were placed to achieve complete closure. f The line attached to the endoclip was cut using scissor forceps.
Fig. 3 Endoscopic images of line-assisted complete closure. a A large mucosal defect after endoscopic submucosal dissection. b The endoclip and line were placed on the proximal margin of the wound. c The line was anchored by another endoclip to the other side of the mucosa, and both sides of the wound were gathered by pulling the anchored line. d Additional endoclips were placed to achieve complete closure. e, f The line attached to the endoclip was cut using scissor forceps.
Video 1: Line-assisted complete closure technique for a large mucosal defect after
endoscopic submucosal dissection.
Results
Patient characteristics and outcomes of LACC are shown in [Table 1]. Two lesions were located in the cecum, 4 in the ascending colon, 2 in the transverse
colon, 2 in the sigmoid colon, and 1 in the rectum. C-ESD was successfully accomplished
without perforation in all cases. Median size of the resected specimen was 36 mm (range
30 – 72). Procedural success was achieved in 10 of 11 cases (91 %). These 10 cases
required a median of 9 endoclips (range 6 – 12) for complete closure. Median procedure
time for LACC was 14 minutes (range 6 – 22). No complications after the procedure
were observed in any of the cases. LACC failed in 1 case, in which the lesion was
located at a flexure of the sigmoid colon.
Table 1
Patient characteristics and outcomes of line-assisted complete closure.
|
Patient
|
Sex
|
Age, years
|
Location
|
Size of resected specimen, mm
|
Procedural success
|
Number of clips used, n
|
Procedure time, min
|
Post-ESD bleeding
|
Post-ESD perforation
|
Post-ESD coagulation syndrome
|
|
1
|
Female
|
52
|
Ascending colon
|
30
|
Yes
|
8
|
16
|
No
|
No
|
No
|
|
2
|
Male
|
44
|
Rectum
|
45
|
Yes
|
10
|
22
|
No
|
No
|
No
|
|
3
|
Male
|
78
|
Sigmoid colon
|
30
|
Yes
|
7
|
14
|
No
|
No
|
No
|
|
4
|
Female
|
80
|
Cecum
|
40
|
Yes
|
9
|
13
|
No
|
No
|
No
|
|
5
|
Female
|
56
|
Transverse colon
|
32
|
Yes
|
9
|
11
|
No
|
No
|
No
|
|
6
|
Female
|
74
|
Transverse colon
|
35
|
Yes
|
7
|
18
|
No
|
No
|
No
|
|
7
|
Male
|
67
|
Cecum
|
36
|
Yes
|
8
|
7
|
No
|
No
|
No
|
|
8
|
Female
|
72
|
Ascending colon
|
72
|
Yes
|
10
|
20
|
No
|
No
|
No
|
|
9
|
Male
|
41
|
Ascending colon
|
35
|
Yes
|
12
|
6
|
No
|
No
|
No
|
|
10
|
Female
|
72
|
Ascending colon
|
40
|
Yes
|
6
|
7
|
No
|
No
|
No
|
|
11
|
Female
|
56
|
Sigmoid colon
|
38
|
No
|
4
|
19
|
No
|
No
|
No
|
ESD, endoscopic submucosal dissection
Discussion
In the current study, our newly developed closure technique, LACC, was successfully
carried out on most of the patients with an acceptable procedure time. We could manage
to close a large mucosal defect of up to 70 mm in diameter using this technique. Our
results indicate that LACC is a feasible technique for closing large mucosal defects
after C-ESD.
Various clip closure techniques have been reported to date, for example, a method
using the “8-ring” [8], and techniques using clips and line named “loop clip” [9]
[10] and “slip knot clip suturing” [11]. In contrast with our technique, the “8-ring” technique requires the special device
which is not always available. The “loop clip” technique requires only clips and line,
which is the same as our technique; however, it needs one more step to make a nylon
loop for preparation compared with our simple technique. In addition, because our
technique can pull the line, we can change a field of vision for additional clipping.
The “slip knot clip suturing” method is similar to our technique; however, making
a slip knot loop is a complicated and time-consuming process compared with our way
of clip-and-line preparation ([Fig. 1]). The authors also reported a simpler technique later, which is very similar to
our technique, almost at the same time we first reported about LACC [5]
[12]. As the later report was just a case report and their outcomes were never disclosed,
we consider that this is the first report to assess the feasibility of this simple
closure technique.
The over-the-scope clip (OTSC) system (Ovesco Endoscopy, Tübingen, Germany) is now
commercially available and may be another option for closing large artificial mucosal
defects of the gastrointestinal tract [13]
[14]. However, it is not always applicable to colorectal lesions because to apply this
method, the colonoscope needs to be withdrawn after C-ESD to mount the applicator
cap of the OTSC system. In contrast, LACC technique does not require withdrawal and
reinsertion of the colonoscope after C-ESD, which is the biggest advantage of LACC,
especially for proximal colonic lesions or patients with difficult colonoscopic intubation.
In this study, we successfully closed all the mucosal defects in proximal colon (2
in cecum and 4 in ascending colon), and there were no technical difficulties in performing
LACC in such locations. However, we failed to achieve complete closure in 1 case,
in which the lesion was located at the flexure of the sigmoid colon. This was because
the maneuverability and stability of the colonoscope were poor in that particularly
narrow and winding location of the colon. In the failed case, instability of the colonoscope
caused too much traction to the clip-and-line device, which consequently result in
clip detachment. Applying a shielding method using polyglycolic acid sheets (Neoveil;
Gunze Co., Kyoto, Japan) with fibrin glue (Beri-plast P 3-mL Combi-Set; CSL Behring
Pharma, Tokyo, Japan) might be easier than clip closure in such difficult locations
[15]
[16]. Our method achieves a tight closure by gathering the mucosa directly by pulling
the line, even in cases in the rectum, which is thought to have less wall mobility.
Of course, the procedure for a large lesion was a bit difficult and required more
time. However, by using multiple clip-and-line devices, we also managed to close a
40-mm mucosal defect resulting from gastric ESD [17].
We did not experience any post-ESD complications in the 10 patients in whom we achieved
complete closure of the large mucosal defect. Although some studies showed that complete
closure of large artificial mucosal defects reduces complications after colorectal
endoscopic resection [3]
[4], further investigation is required to clarify whether complete closure really results
in a clinical benefit. We previously reported that rates of delayed perforation and
bleeding after C-ESD were not high (1.0 % and 3.0 %, respectively), however, incidence
of PECS, which required a longer fasting period and hospital stay after C-ESD, was
9.5 % [2]. Therefore, we consider that preventing PECS is an important issue, and complete
closure by LACC has the potential to solve the problem. Of course, taking into account
cost-effectiveness, it might be reasonable to apply the technique only to high-risk
cases for PECS, such as female patients with large tumor located in the ascending
colon or cecum [2]. In fact, we have started a prospective, randomized controlled trial to elucidate
whether LACC is effective in reducing incidence of PECS in such a high-risk population.
This feasibility trial is limited by its small sample size and retrospective design;
thus, we cannot state that it yielded clinical benefits at the moment. However, the
high procedural success rate and relatively short procedure time showed its technical
feasibility. Also, the optimal line for this technique should be considered. LACC
requires additional clipping through an endoscopic channel with an inserted line for
clip-and-line device. Because a thick line would disturb repeat insertion of the endoclip
applicator, dental floss is not an optimal line for LACC. Because the line used in
LACC does not suture colonic mucosa directly, we believe that a fishing line would
be acceptable, but using a surgical suture thread would warrant the safety of the
procedure.
Conclusion
In conclusion, LACC is feasible for complete closure of large mucosal defects after
C-ESD. We consider that this simple method is valuable and should be further assessed
to determine its clinical benefit.
