Introduction
Endoscopic submucosal dissection (ESD) was developed initially for resecting superficial
gastrointestinal tumors, and its application was recently expanded to other purposes;
perioral endoscopic myotomy (POEM) [1], antireflux mucosectomy (ARMS) [2], and endoscopic antralplasty [3]. These reports suggest that ESD techniques may improve serious gastrointestinal
disturbances by different mechanisms, but utilization of ESD is currently very limited.
Herein, we report a case with obstruction due to severe flexure of the gastric conduit
after esophagectomy for esophageal cancer. The flexure was straightened and the obstruction
successfully abolished by ulcer scar formation following a mucosectomy by ESD on the
oral-side mucosa of the flexure. This is the first case report of the effectiveness
of mucosectomy by ESD for an obstruction due to flexure of the gastric conduit after
esophagectomy.
Case report
A 70-year-old-man underwent an esophagectomy and posterior mediastinal reconstruction
for esophageal cancer. The esophageal cancer was curatively resected. Meals were started
on postoperative day (POD) 11, but the patient repeatedly vomited after drinking water
or eating meals. Changing the patient’s posture had no effect. An upper gastrointestinal
series revealed stagnation of the contrast agent at a flexure of the gastric conduit
( [Fig.1a]). Endoscopic examination revealed an obstruction due to the flexure ([Fig. 2a]). The flexure was located at about 5 cm oral side from the pylorus ring. Endoscopic
balloon dilation was attempted twice (POD 60, POD 67) ([Fig. 2b]), but was ineffective because the obstruction was not due to a small lumen diameter,
but rather to severe flexure of the gastric conduit ([Fig. 2c]). The patient could not be discharged from the hospital due to repeated vomiting
even after ingesting only a small amount of fluid or food, and thus required total
parental nutrition. Further surgery was considered, but was thought to be extremely
risky due to difficulties in reoperating after posterior mediastinal reconstruction.
We therefore attempted ESD to treat the severe flexure of the gastric conduit on POD112.
A piece of the mucosa 8 cm in diameter was resected with ESD using the ITknife 2 (Olympus,
Tokyo, Japan) and the clip-with-line method [4] and an artificial ulcer was created at the oral side of the flexure ([Fig. 3a, b, c, d]). A delayed perforation (~5 mm in diameter) observed at the oral edge of an artificial
ulcer on post-ESD Day 2 ([Fig. 3e]) was conservatively treated by packing 10 15 × 7-mm polyglycolic acid sheets (Neoveil;
Gunze Co., Kyoto, Japan) into the hole, applying a solution containing fibrinogen
(Solution A of the Beriplast P Combi-Set; CSL Behring Pharma, Tokyo, Japan) with a
spray tube, and then applying a solution containing thrombin (Solution B of the Beriplast
P Combi-Set [[Fig. 3f]]) with another spray tube to fix the sheets. Endoscopic examination on post-ESD
Day 10 revealed that the perforation was closed, the gastric conduit flexure straightened
due to ulcer scarring, and obstruction at the flexure opened over time ([Fig. 1b], [Fig. 3 g], [Fig. 4]). Meals were restarted on post-ESD Day 18 and the patient was able to eat without
vomiting; the patient was discharged from the hospital on post-ESD Day 35 (POD147).
Obstruction at the gastric conduit flexure was completely opened 3 months after ESD
([Fig. 1c], [Fig. 3h]) and remained open 1 year later ([Fig. 3i]).
Fig. 1 Upper gastrointestinal study (x-ray) before and after endoscopic submucosal dissection
(ESD). a Postoperative Day 35. Upper gastrointestinal study revealed stagnation of the contrast
agent at a flexure of the gastric conduit. b One month after ESD, the gastric conduit flexure had straightened. Contrast agent
smoothly flowed at the flexure. c Three months after ESD, the flexure was obscure. Red arrows point to flexure of the
gastric conduit.
Fig. 2 Endoscopic images at the flexure of the gastric conduit before and after endoscopic
balloon dilation (EBD). a Postoperative Day (POD) 60; before EBD. b EBD was repeatedly performed. c POD 67; after EBD. EBD was completely ineffective.
Fig. 3 Endoscopic images of the endoscopic submucosal dissection (ESD) (Postoperative Day
112) and after ESD. a Marking at the oral side of the flexure. b Following submucosal injection of glycerol and mucosal cutting, submucosal dissection
was performed with the ITknife 2. c Artificial ulcer after ESD. d Resected specimen measuring 80 × 80 mm. e Post-ESD Day 2; delayed perforation was identified at the oral edge of an artificial
ulcer. f The delayed perforation was treated with polyglycolic acid sheets and fibrin glue
on the same day. g One month after ESD. h Three months after ESD; obstruction at the gastric conduit flexure was completely
opened. i One year after ESD.
Fig. 4 Schema of the gastric conduit before and after endoscopic submucosal dissection (ESD)
in this case.
Discussion
Esophagectomy is a complex surgery with high risk of severe complications. Anastomotic
stricture or gastric outlet obstruction may occur after esophagectomy [5]
[6]. Treatment with repeated endoscopic balloon dilation is often required for the stricture,
and even an additional surgery may be required for the obstruction. In this case,
severe flexure of the gastric conduit caused an obstruction, i. e., stagnation of
liquids and food, and the patient repeatedly vomited. This complication seems to be
very rare, and we found no reports of a similar occurrence. The mechanism by which
severe flexure occurs following esophagectomy is unknown. An upper gastrointestinal
study showed that the flexure did not locate in the diaphragm. A computed tomography
scan (data not shown) and upper gastrointestinal study showed no herniation of the
pylorus into the thoracic cavity. Mismatching of the angle of the resection line by
the linear stapler may have caused the severe flexure.
Endoscopic balloon dilation was attempted to treat the obstruction, but it was completely
ineffective. Additional surgery was extremely risky due to the posterior mediastinal
reconstruction, and we therefore attempted another endoscopic procedure, ESD.
Mori et al. [7] and Ohara et al. [3] reported that contralateral mucosal resection was effective for the treatment of
antral deformation after ESD. Those mucosal resections utilized the power of contraction
provided by ulcer scar formation to improve the antral deformation. In our case, the
power of contraction by ulcer scar formation straightened the flexure of the gastric
conduit, successfully opening the lumen at the flexure. The schema of the gastric
conduit before and just after the ESD procedure is shown in ([Fig. 4]). This treatment may be effective for other similar obstructions due to severe flexure.
On the other hand, an artificial ulcer may cause subsequent stricture of the gastrointestinal
lumen by ulcer scar formation, which could worsen the obstruction. The gastrointestinal
lumen must have sufficient space to accommodate scarring of an artificial ulcer. We
therefore resected the mucosa not on the flexure itself, but rather on the oral side
of the flexure where the lumen was much broader. In this case, we planned the resection
outline based on our hypothesis that the edges would be drawn to the center of an
artificial ulcer by ulcer scar formation. In fact, as expected, the mucosa around
the planned resection outline was drawn to the center, and it did not seem to completely
join together due to mucosal regeneration. Further studies are needed to determine
optimal location and size of a mucosectomy for other similar situations.
On the other hand, in this case, delayed perforation was detected 2 days after ESD.
Low blood flow by cervical anastomosis via the posterior mediastinal root or bile
duct reflux may be associated with delayed perforation [8], and it was conservatively treated with polyglycolic acid sheets and fibrin glue
[9]
[10]. Polyglycolic acid sheets and fibrin glue may also be an effective option for perforation
during thoracic surgery in high risk cases.
Conclusion
This is the first case report demonstrating that mucosectomy by ESD for obstruction
due to severe flexure of the gastric conduit after esophagectomy is extremely effective.
This novel method may be a viable alternative therapy in cases in which surgery carries
additional risks. Further studies are needed to evaluate the usefulness of this technique.
