Introduction
Endoscopic full-thickness resection (EFTR) is indicated for upper gastrointestinal
subepithelial tumors originating in the muscularis propria (MP), especially < 3-cm
gastrointestinal stromal tumors (GISTs) suitable for removal via the mouth [1]. A recent study showed excellent outcomes of laparoscopic endoscopic cooperative
surgery, which involves endoscopic navigation and tumor resection followed by laparoscopic
suture closure [2]. Several unique techniques derived from endoscopic submucosal dissection have been
further developed as minimally invasive endoluminal surgeries (i. e., “no-scar surgeries”
without laparoscopic assistance). These techniques involve submucosal tunneling with
endoscopic resection (STER) [3], exposed EFTR followed by endoscopic closure [4], and advanced non-exposed EFTR [5]
[6]
[7]
[8]. However, when a GIST is buried deep in the MP or exhibits extraluminal growth,
STER in narrow working spaces might not easily achieve complete resection of GISTs
with an intact capsule. Thus, exposed EFTR can more reasonably achieve a negative
surgical margin. Because a larger perforation occurs in exposed EFTR, insufflation
leakage increases the difficulty of maintaining a sufficient operative field, managing
intraoperative bleeds, and achieving secure endoscopic full-thickness closure (EFTC)
of the defect, which migrates toward the peritoneal cavity. This study aimed to evaluate
the safety and feasibility of a new exposed EFTR.
Patients and methods
Four 11–to 13-month-old female beagles were enrolled in this study from December 2019
to March 2020. Exposed EFTR was performed for virtual GISTs of approximately 2-cm
diameter at four locations in the upper stomach where GISTs often occur [2]: the greater and lesser curvatures and anterior and posterior walls. The dogs were
fasted for 24 hours before the intervention. Procedures were performed under general
anesthesia with intubation. The dogs received an antibiotic and proton pump inhibitor
once daily for 2 days postoperatively. After the absence of significant symptoms,
such as loss of appetite, tarry stool, and ptyalism were confirmed in all dogs during
the 2 days of fasting, a solid diet was started on postoperative day 3. This study
was approved by the institutional review board based on the Regulations and Guidelines
of Animal Experiments of Kagawa University (Registration No. A248, Approval No. 20619).
Experimental equipment
The following equipment was used: gastroscope (Q260J; Olympus, Tokyo, Japan), over-the-scope
clip (OTSC) system and Twin Grasper (TG) forceps (Ovesco Endoscopy GmbH, Tübingen,
Germany), hemoclips (HX-610–090L; Olympus), endoscopic variceal ligation device (MD-48720U;
Sumius, Tokyo, Japan), loop cutter (FS-5L-1; Olympus), endoscopic knives [DualKnife
(KD-650 L); Olympus and ITknife2 (KD-611 L); Olympus], hemostatic forceps (FD-410
LR; Olympus), glycerol for injection (Chugai Pharmaceutical Co., Ltd., Tokyo, Japan),
carbon dioxide insufflation device (UCR; Olympus), and electrosurgical unit (VIO 300
D; Erbe Elektromedizin, Tübingen, Germany). The settings for EFTR were as follows:
cut mode, Endocut I, effect 2, duration 3, interval 2 for mucosal incision; coagulation
mode, Swift Coag, effect 3, 80 W for submucosal trimming and full-thickness resection;
and Soft Coag, effect 6, 80 W for hemostasis.
Video 1 This video shows a novel therapeutic strategy of exposed EFTR using clip-line traction
followed by closure using an O-ring and OTSC in the stomach.
Traction-assisted EFTR
Step 1: Groove After marking around the lesion, whole-circumferential mucosal incision and submucosal
trimming were performed until the muscle layer appeared([Fig. 1a]).
Fig. 1 Schema for traction-assisted endoscopic full-thickness resection (EFTR). a Whole circumferential mucosal and submucosal incision (groove). b Ring-anchor for subsequent closure. c Half circumferential EFTR of endpoint. d Pulley traction using clip-line in forward approach. e Simple clip-line traction in retroflexed approach. f Complete EFTR
Step 2: Ring-anchor for subsequent EFTC At both edges around the prepared two-point marking, an approximately 5-mm perforated
slit hole was created using the Dual Knife on the exposed MP. A hemoclip that captured
a handmade ring-thread (4-cm diameter with 3–0 surgical nylon) was firmly anchored
on the edge of the slit hole to grasp the full-thickness layer. The same procedure
was then performed on the opposite side([Fig. 1b]). This was performed to prepare for possible insufflation leakage during subsequent
EFTC.
Step 3: Half-circumferential EFTR of endpoint Half-circumferential EFTR was performed from the slit holes on both sides toward the
distal side([Fig. 1c]
). Careful attention was paid to avoid cutting the ring-thread.
Step 4: Traction using clip-line method Traction was applied to the lesion using a previously described clip-with-line method
[9]. Pulley traction was used in the forward approach [10] ([Fig. 1d]), while the simple clip-line method was used in the retroflexed approach ([Fig. 1e]).
Step 5: Complete EFTR EFTR was achieved by resecting the residual full-thickness layer under traction ([Fig. 1f]).
EFTC using O-ring and OTSC
Step 1: Endoscopic ligation with O-ring closure (E-LOC) As previously described [11], E-LOC was introduced to approximate large defects ([Fig. 2a]). A hemostatic forceps was used to capture the prepared ring-thread and pull it
into the cap of the endoscopic variceal ligation device ([Fig. 2b]). After the two deployed hemoclips were pulled into the cap, an O-ring was fired
around these hemoclips ([Fig. 2c]). When the tails of these hemoclips were caught in the omentum, they were released
using grasping forceps so that the hemoclips could move freely. The ring-thread was
cut with a loop cutter.
Fig. 2 Schema for endoscopic full-thickness closure (EFTC) using O-ring and OTSC. a Capture of the prepared ring-thread. b Two deployed hemoclips captured in endoscopic variceal ligation cap by pulling the
thread. c Firing of an O-ring. E-LOC diminishes insufflation leakage and approximates the large
defect. d Twin grasper (TG)-assisted OTSC closure. Easy TG maneuver to grasp full-thickness
layer. e Deployment of two OTSCs. f Mucosal closure using hemoclips. Endoscopic ligation with O-ring closure, E-LOC.
Step 2: TG-assisted OTSC closure The approximated full-thickness defect was grasped with the TG ([Fig. 2d]) and closed with two OTSCs (type gc, 10 mm) ([Fig. 2e]). The OTSCs were deployed directly around the fired O-ring to compensate for the
space between the hemoclips and OTSCs.
Step 3: Mucosal closure The exposed mucosa was closed using hemoclips, resulting in complete EFTC ([Fig. 2f]).
During the entire procedure, bleeding was managed by using hemostatic forceps. A 20G
needle (Nipro Corporation, Osaka, Japan) was percutaneously inserted to evacuate excessive
intra-abdominal air. Representative endoscopic images obtained during EFTR and EFTC
are shown in [Fig.3a], [Fig. 3b], [Fig. 3c], [Fig. 3d], [Fig. 3e], and [Fig. 3f].
Fig. 3 Representative images of EFTR and EFTC. a Forward approach: pulley traction using clip-line at greater curvature. b Retroflexed approach: simple clip-line traction at lesser curvature. c Ring-anchor prepared for EFTC in the section of EFTR. d Two deployed hemoclips captured in the ligation cap by pulling the thread. e Firing of an O-ring. f TG-assisted OTSC closure.
Outcome measures
The primary outcome measure was the complete resection rate. Complete resection was
defined as a negative lateral margin and macroscopic or histological confirmation
of a full-thickness wall. The secondary outcome measures were technical success, intraoperative
or postoperative complications, maximum resection size, and 1-month survival. An intraoperative
complication was defined as spurting bleeding or other organ injury. The procedure
times of the total operation, EFTR, and EFTC were measured. The total procedure time
was defined as the time from marking the dots to complete closure.
Results
Complete resection and technical success were achieved in all four locations ([Fig. 4a], [Fig. 4b], [Fig. 4c], [Fig. 4d]). Intraoperative bleeding occurred in one case during full-thickness pre-cutting.
No postoperative complications occurred. The median maximum resection size (synonymous
with the perforated defect size) was 27.5 (range, 25–30) mm. The 1-month survival
rate was 100 %. The median procedure time of the total operation, EFTR, and EFTC was
76 (49–92), 37 (24–51), and 35.5 (25–48) minutes, respectively. These outcomes are
summarized in [Table 1]. Endoscopic examination on postoperative day 30 showed no anastomotic leakage. The
macroscopic and histological findings of the resected specimens and anastomoses are
shown in [Fig. 5a], [Fig. 5b], [Fig.5c], [Fig. 5d], [Fig. 5e], and [Fig. 5f].
Fig. 4 Anastomotic site of exposed EFTR at four locations in the upper stomach. a Greater curvature. b Lesser curvature. c Anterior wall. d Posterior wall.
Table 1
Outcomes of exposed EFTR and EFTC.
|
Case no.
|
Location of upper stomach
|
Complete resection
|
Technical success
|
Intraoperative complications
|
Postoperative complications
|
Procedure time, min
|
Resected maximum size
|
1-month survival
|
|
Total
|
Resection
|
Closure
|
|
|
|
1
|
Greater curvature
|
Yes
|
Yes
|
None
|
None
|
75
|
45
|
30
|
30
|
Yes
|
|
2
|
Lesser curvature
|
Yes
|
Yes
|
Bleeding[1]
|
None
|
77
|
29
|
48
|
25
|
Yes
|
|
3
|
Anterior wall
|
Yes
|
Yes
|
None
|
None
|
49
|
24
|
25
|
30
|
Yes
|
|
4
|
Posterior wall
|
Yes
|
Yes
|
None
|
None
|
92
|
51
|
41
|
25
|
Yes
|
1 Spurting bleeding was endoscopically managed using hemostatic forceps.
Fig. 5 Macroscopic and histological findings from resected specimen and anastomosis on postoperative
Day 30. a Mucosal side of resected specimen (30 × 25 mm) obtained by complete resection. b Serosal side of full-thickness specimen. c Histological examination of full-thickness specimen with muscle and serosa (hematoxylin
and eosin staining). d Anastomotic mucosal surface showing sustained OTSC closure. e Anastomotic serosal surface showing whitish healing scar (red arrows) without fistula.
f Histological examination of the anastomotic site compensated by massive fibrotic
tissues between resected layers (yellow arrows).
Discussion
This is the first animal experimental study to demonstrate the procedural strategy
of exposed EFTR using clip-line traction and novel EFTC technique using an O-ring
and OTSC in the stomach. Exposed EFTR has several advantages, including achieving
simple whole-circumferential resection and intact tumor resection.
Our exposed EFTR technique involves whole-circumferential submucosal dissection and
full-thickness muscle-serosa resection. The two-step procedure can coagulate exposed
vessels identified in gaps between muscle layers. Importantly, technical tricks are
required to manage thick vessels on the serosal side and resect the lesser or greater
curvature adhered to the serosa. We performed traction using the clip-line method
[9]
[10] to successfully manage these situations. EFTR was completed without severe complications
in all four locations of the upper stomach within an acceptable procedure time (median
of 76 min for total operation and 37 min for EFTR). In this small case series, we
found no difference in technical difficulty among the four upper stomach locations.
However, the procedure time may have been influenced by the severity of adhesion between
the omentum and serosa because dissection of the severely adhered lesser omentum took
the longest time in the posterior wall. A large-scale study showed the superiority
of traction-assisted EFTR (n = 64) over non-assisted EFTR (n = 128) in terms of the
total procedure time (44.2 vs. 54.2 min) and complication rate (3.1 % vs. 12.5 %)
[12]. However, the description of the circumferential resection order was unclear. In
our previous experience, the endpoint of the distal side was invisible, resulting
in excessive resection and tumor damage. Thus, half-circumferential EFTR of the endpoint
on the distal side should be attempted before use of traction. Moreover, use of traction
has been reported with the retroflexed approach alone. The forward approach is needed
for the greater curvature.
EFTC is challenging for large perforated defects. The keys to success include maintenance
of a sufficient working space and achievement of inverted “serosa–serosa apposition.”
The OTSC, a full-thickness clipping device, enables serosa–serosa apposition [13]. However, even with TG assistance, OTSC closure sometimes fails because of a difficult
TG maneuver and insufflation leakage. Thus, our E-LOC technique was used during EFTR
[11]. E-LOC reduced insufflation leakage and enabled separation of a large defect into
two small slits resembling sunglasses. The TG was then easily maneuvered to grasp
the full-thickness layer, succeeding in serosa–serosa apposition. Traction and E-LOC
also are mandatory to maintain an adequate operation field, which is influenced by
insufflation leakage. A case series using OTSC before resection showed a high technical
success rate in 16 cases [14]; however, full-thickness resection accounted for only 69 % of cases. One clinical
report revealed excellent outcomes of EFTR followed by OTSC closure [4]. Perforated defects of < 2 cm were successfully closed with single OTSC in all 23
cases. Thus, our method has potential to achieve OTSC closure for larger lesions because
the median maximum defect size was 27.5 mm. This finding also shows that the procedure
may be indicated for the lesion with maximum size of 25 mm. Meanwhile, centered hemoclips
with O-ring may have a possibility of clip falling. These hemoclips dropped off in
one of all cases on postoperative Day 30, while OTSCs remained in all cases. A previous
study reported that inverted serosal apposition provided a more durable and reliable
repair than everted mucosal apposition [15]. Thus, our procedure, which aimed to grasp a full-thickness layer using the TG and
achieve its inverted closure, called for serosa–serosa apposition. Inverted full-thickness
closure probably compresses the gap between the hemoclips and OTSCs, even if the hemoclips
drop off at an early stage. The OTSCs were also deployed directly around the fired
O-ring to compensate for the gap. A large-scale study should be conducted to evaluate
the durability of this EFTC method.
Non-exposed EFTR is an advanced procedure in which closure is performed before resection
without communication between the intra-lumens and extra-lumens. This may diminish
the risk of postoperative peritonitis and broaden the indications to include early-stage
gastric cancers with no lymph node metastasis and with severe submucosal fibrosis.
Novel procedures using full-thickness suturing devices are still experimental in ex
vivo [5]
[6], in vivo [7]
[8], and clinical settings [16].
This study had several limitations. First, the number of animals was small to minimize
animal suffering. Second, the indication of the procedure was limited to < 2-cm lesions
in the upper stomach. This procedure may be contraindicated for lesions located in
cardia or pyloric ring with a narrow lumen because of the potential risk of endoluminal
stenosis from OTSC deployment. Third, the clinical indication is limited to GIST without
ulcers because exposed EFTR is associated with a risk of peritoneal tumor seeding.
Fourth, a leak pressure test was omitted because no leaks were confirmed in previous
ex vivo experiments. Fifth, the use of two OTSCs and the TG forceps is expensive,
thus, this method should be further simplified to reduce the cost when considering
the clinical application. Further investigations are needed.
Conclusions
This experimental study demonstrated that clip-line traction, E-LOC, and OTSC can
be used to achieve exposed EFTR.
