The Endoscopic Submucosal Dissection Surgery in the Whole-Course Antegrade Endoscopic Approach: A More Effective Treatment Strategy for the Siewert II/III Type Mucosal Lesions of Esophagogastric Junction

Abstract

Objective:

The purpose of this study was to investigate the safety and efficacy of endoscopic submucosal dissection (ESD) for treating cardiac mucosal lesions.

Methods:

A total of 86 patients with cardiac mucosal lesions were treated with ESD in retrograde endoscopic approach or antegrade endoscopic approach. The relationship between the two methods was analyzed according to the size, location, depth of pathological infiltration, classification, and examination results. The main evaluation indexes of intraoperative complications were operation time, bleeding, perforation, and complete resection (R0 resection).

Results:

Total R0 excision was performed in 85 patients and curative excision in 77 patients. When the diameter of lesion was 2–4 cm or >4 cm, the median treatment time in the antegrade endoscopic approach group was shorter than that in the retrograde group (P < .001, respectively). When the lesion was confined to the mucosa, the median treatment time in the antegrade endoscopic approach group was shorter than that in the retrograde group (P < .001). When the lesion was located in the posterior wall of the cardia, the average treatment time in the antegrade endoscopic approach group was shorter than that in the retrograde group (P < .05). When the lesion was located in the lesser curvature of the cardia, the average treatment time in the antegrade endoscopic approach group was shorter than that in the retrograde group (P < .001).

Conclusion:

The ESD surgery in the antegrade endoscopic approach is effective and safe for the treatment of cardiac mucosal lesions.

Introduction

Endoscopic submucosal dissection (ESD) has been the standard for precancerous gastric lesions and early gastric cancer.^1–4 The esophagogastric junction (EGJ) is an area of high incidence of gastric cancer. It is difficult to carry out the ESD surgery in this area, and it usually takes a long time to perform this type of ESD surgery. First, Siewert II/III type lesion is in a particular location, and it is often difficult to see it clearly. Second, the submucosal vascular network of the connection part of the esophagus and stomach is abundant, which makes intraoperative bleeding very easy. Therefore, many surgeons still think that it is technically challenging.^5–7

Due to the special position of Siewert II/III type lesions, the traditional ESD operation method is to adopt reverse endoscopy as the main treatment strategy, with antegrade endoscopic approach as auxiliary.⁸ Prolonged operation time will increase the incidence of intraoperative bleeding, perforation, postoperative pulmonary infection, delayed bleeding, and other related surgical complications to some extent.^9,10 In the multivariate analysis, it was found that the location of lesions and the size of specimens were the main factors to affect the operation time, whereas the influence of specimen length was more significant.^9,11–13 When reasonable surgical strategies are developed before surgery and efficient and safe operative techniques are carried out during surgery, the operation time and complications during/after surgery are reduced.

In recent years, we have studied the methods of ESD surgery for the mucosal lesions of the EGJ. Based on the practical clinical experience, traditional conventional treatment method has been improved, and a new efficient and safe method has been formed. Aimed at the new method (the ESD surgery in the whole-course antegrade endoscopic approach) for mucosal lesions of the EGJ especially Siewert II/III type mucosa lesions of the operating efficiency of related factors, we have analyzed these data in this study to develop a safe, efficient, and reasonable surgical plan for such lesions.

Patients and Methods

Patients

A total of 86 patients with Siewert II/III type gastric mucosal lesions who underwent ESD surgery at the Endoscopic Center of Zhejiang Cancer Hospital in China between June 2010 and December 2019 were included in this study. This study was approved by the ethics committee of the Cancer Hospital affiliated to the University of Chinese Academy of Sciences (Zhejiang Cancer Hospital). All patients signed informed consent. Inclusion criteria were as follows: (1) patients with Siewert II/III type gastric mucosal lesions.

Siewert II type gastric mucosal lesion is defined as a tumor with an epicenter located within 1 cm proximal and 2 cm distal to the anatomical EGJ.¹¹ Siewert III type gastric mucosal lesion is defined as a tumor with an epicenter located within 2 cm proximal and 5 cm distal to the anatomical EGJ.¹¹ (2) Patients with flat adenoma or polyp of EGJ, which was confirmed by biopsy as early differentiated adenocarcinoma or high-grade intraepithelial neoplasia (HGIN), consistent with the indications of gastric ESD, or mucosal lesions suspected of tumor suggested by endoscopy. (3) Endoscopic ultrasound and computed tomography showed that the tumor was confined to the mucosa or superficial submucosa. (4) No history of taking anticoagulant drugs or antiplatelet drugs within 7 days before surgery.⁸

Patients were divided into two groups. In the first group, from June 2010 to June 2016, 36 patients were treated by the traditional ESD operation method, which was to adopt reverse endoscopy as the main treatment strategy, with antegrade endoscopic approach as auxiliary. In the second group, from July 2016 to June 2019, 50 patients were treated by improved treatment strategy, which was the ESD surgery in the whole-course antegrade endoscopic approach, without reversing endoscopy.

Endoscopic resection

All patients underwent ESD surgery under tracheal intubation anesthesia. ESD was performed with single-channel endoscope (GIF-Q260J; Olympus, Tokyo, Japan), and it included main accessories including Dual knife (Olympus 650 type) and Insulated-tip (IT2) knife (Olympus). ESD surgical procedures included labeling, injection, circumferential incision, submucosal dissection, and hemostasis. A sodium hyaluronate mixture, including epinephrine (0.01 mg/ml) and a small amount of methylene blue, was used for injection. Hemostasis forceps (FD-410LR; Olympus) was used for endoscopic hemostasis. The bleeding or exposed blood vessels were prophylactic hemostasis after dissection. ERBE-VIO 200D electrosurgical unit was used for all ESD surgery.

ESD was performed using a typical sequence involving marking, mucosal incision, and submucosal dissection with simultaneous hemostasis. The marking was performed on the borders of the lesions identified by conventional endoscopy or chromoendoscopy. After marking, sodium hyaluronate mixture was injected submucosally around the lesion to lift it. A small incision was made with a Dual knife (Olympus), followed by a circumferential mucosal incision outside the marking using an IT2 knife or Dual knife. Submucosal dissection was performed with an IT2 knife and/or Dual knife. If necessary, hemostatic procedures were performed during and aftersubmucosal dissection. In most cases, we only use a Dual knife to perform ESD surgery, including marking, mucosal incision, and submucosal dissection with simultaneous hemostasis.

Observational indexes

The operation time of ESD treatment was defined as the time from the marked lesion to complete dissection and resection of the specimen, including the time of intraoperative hemostasis treatment, but excluding the time of wound prevention and hemostasis treatment after dissection and resection of the lesion. Complications during and after surgery were recorded, including abdominal pain, perforation, and bleeding. The patients were hospitalized for observation for 2–4 days. Oral intake was generally restored 24 hours after surgery, and endoscopic follow-up was conducted 1–3 months after surgery. Delayed complications were also recorded.

Histopathologic analysis

Specimens were cut at 2–3 mm intervals, and the curative effect of ESD specimens was evaluated by histopathology. R0 resection was defined as total resection without tumor at the horizontal and basal resection margins. R0 resection without vascular invasion and deep submucosal infiltration (<500 μm) was considered as curative resection.⁸

Statistical analysis

SPSS 20.0 (International Business Machines, Corp., Armonk, NY, USA) was used to perform statistical analysis. The measurement data of normal distribution are represented by mean ± standard deviation. Since the rank-sum test showed that the operation time was skewed, median (25th percentile, 75th percentile) was adopted to represent it, and the median of operation time was used as the cutoff point for analysis in regression. The chi-square test and the rank-sum test were used to perform univariate analysis on the relevant influencing factors, such as complete removal rate. P < .05 was considered to be statistically significant.

Results

General clinical characteristics

A total of 86 patients with esophageal and gastric junction mucosal lesions, including hyperplastic polyps, precancerous mucosal lesions, and early gastric cancer, were included in this study. There were 67 males and 19 females. The average age was 66.8 years. There were no statistical differences in age, sex, height, weight, and obesity index between antegrade endoscopic approach group (traditional group) and retrograde group (improvement group).

All patients completed ESD treatment, with an average operative time of 60.9 minutes. The average operative time was 79.4 minutes in the traditional group and 46.8 minutes in the improvement group. In addition, it was significantly shorter in the improvement group than in the traditional group (median time: 43 minutes versus 73 minutes, P < .05). The average length diameter of the specimens was 30.8 mm. The proportions of the most severe lesions distributed in all directions were: 42 cases (48.8%) in the lesser curvature, 37 cases (43.0%) in the posterior wall, 5 cases (5.8%) in the greater curvature, and 2 cases (2.3%) in the anterior wall. R0 whole-block resection was performed in 85 patients (98.8%) and therapeutic resection in 77 patients (89.5%). For postoperative pathology, there were 9 patients (10.4%) with low-grade intraepithelial neoplasia, 18 patients (20.9%) with HGIN, 53 patients (61.6%) with adenocarcinoma, and 7 patients (8.1%) with hyperplastic polyps. Among the 53 patients with adenocarcinoma, there were 34 patients (39.5%) with mucosal carcinoma,10 patients (11.7%) with superficial submucosal carcinoma, 7 patients (8.1%) with deep submucosal carcinoma, 1 patient (1.1%) with muscle layer invasion, and another patient (1.1%) with muscle layer invasion accompanied by vascular tumor thromboembolism (Table 1).

Table 1.

Pathological Results of the Two Groups

Group	LGID	HGID	M	SM1	≥SM2	Hyperplastic polyp	Total cases
Traditional inverted endoscopy group	4	9	12	5	2	4	36
Improved straight endoscopy group	5	9	22	5	6	3	50
Total (%)	9 (10.4)	18 (20.9)	34 (39.5)	10 (11.7)	8 (9.3)	7 (8.1)	86

HGID, high-grade intraepithelial neoplasia; LGID, low-grade intraepithelial neoplasia; M, the carcinoma is confined to the mucosa; SM1, the carcinoma is confined to the superficial submucosa; ≥SM2, the invasion depth of tumor tissue was deeper than the deep submucosa.

There was 1 patient (1.1%) with postoperative black stool hemorrhage in the traditional group, whereas there was no obvious upper gastrointestinal hemorrhage, such as hematemesis and black stool, in the improvement group. There were no perforation patients in the two groups. Postoperative epigastric pain occurred in 10 patients (11.7%), including 6 patients in the traditional group and 4 patients in the improvement group. There were no other serious postoperative complications, such as acute pancreatitis, pneumonia, acute myocardial infarction, or cerebral hemorrhage in either group. There were no significant differences in postoperative complications between the two groups.

Analysis of influencing factors of operation time

According to three indicators, including the specific distribution location of lesions, the size of specimens, and the invasion depth of lesions, we subdivided the data into three groups for statistical analysis one by one. Table 2 shows the comparison of the lesion location of EGJ and the operation time between the traditional retrograde endoscopy group and the improved straight endoscopy group. The mean time of ESD treatment in the small curvature and posterior wall of EGJ in the improvement group was significantly shorter than that in the traditional group (P < .05), suggesting that the ESD surgery in the whole-course antegrade endoscopic approach was significantly more efficient than that in the traditional retrograde endoscope. The number and proportion of mucosal lesions in the large curved side and the anterior wall of EGJ were relatively small, and there were only 7 patients (8.1%) in the two combinations, which could not be compared statistically.

Table 2.

Comparison of Operation Time and Lesion Location of Esophagogastric Junction Between the Two Groups

Group	Lesser curvature	Posterior wall	Greater curvature	Antetheca	Total cases
Traditional inverted endoscopy group	17	15	3	1	36
Operation time (minutes), $\bar{X}$ ± S	76.82 ± 41.67	78.8 ± 36.51	87.67 ± 22.7	108	—
Improved straight endoscopy group	25	22	2	1	50
Operation time (minutes), $\bar{X}$ ± S	39.24 ± 11.61	55.72 ± 20.35	45.50 ± 2.50	43	—
t-Value	3.625	2.223	—	—	—
P value	.002	.038	—	—	—

$\bar{X}$ ± S, mean ± standard deviation.

Table 3 shows the comparison of the operation time and the longest diameter of specimens after ESD operation between the traditional group and the improvement group. As the scope of the lesion expands, the treatment time of ESD was also increased correspondingly. When the lesion was small and the length diameter of the surgical specimens was <2 cm, the treatment time of the two groups was very close, and there was no advantage in the improvement group. With the expansion of the lesion range, the efficiency advantage in the improvement group was reflected. When the lesion length was 2–4 cm, the median time of ESD treatment in the improvement group was significantly shorter than that in the traditional group (P < .05). When the lesion scope is >4 cm, the median time in the two groups were 70 and 133 minutes (P < .001), indicating that the efficiency advantage in the improvement group was particularly obvious.

Table 3.

Comparison of Operation Time and Lesion Scope (The Longest Diameter of the Endoscopic Submucosal Dissection Specimen) Between the Two Groups

Group	<2 cm	2–4 cm	>4 cm	Total cases
Traditional inverted endoscopy group (cases)	8	19	9	36
Operation time (minutes), P50 (P25, P75)	34.5 (33, 54)	68 (53, 87)	133 (121, 136)	—
Improved straight endoscopy group (cases)	15	25	10	50
Operation time (minutes), P50 (P25, P75)	35 (32, 43)	45 (38, 48)	70 (48, 86)	—
Z value	−0.586	−3.462	−3.679	—
P value	.558	.001	<.001	—

P50 (P25, P75), median (25th percentile, 75th percentile).

According to the postoperative pathological results, Table 4 showed the comparison of the operation time and the lesion invasion depth between the traditional group and the improvement group. It was found that the median time in the improvement group limited to mucosal lesions was significantly shorter than that in the traditional group (P < .001), suggesting that ESD treatment in the improvement group was obviously efficient. The number and proportion of superficial submucosal lesions and lesions with deeper invasion rather than the deep submucosal layer were relatively small. Although there were no significant differences, the improvement group also had an advantage in comparison with the traditional group.

Table 4.

Comparison of Operation Time and Lesion Invasion Depth Between the Two Groups

Group	M	SM1	≥SM2	Total cases
Traditional inverted endoscopy group (cases)	29	5	2	36
Operation time (minutes), P50 (P25, P75)	78 (45, 115)	56 (35, 110)	76 (65, 87)	—
Improved straight endoscopy group (cases)	39	5	6	50
Operation time (minutes), P50 (P25, P75)	43 (34, 46)	49 (48, 65)	55 (43, 98)	—
Z value	−3.978	−0.210	−0.671	—
P value	<.001	.834	.502	—

M refers to lesions confined to the mucosal layer (including low-grade intraepithelial neoplasia, high-grade intraepithelial neoplasia, carcinoma of the mucous membrane, and hyperplastic polyp). SM1 refers to the cancer tissue limited to the superficial submucosal layer 500 μm. ≥SM2 refers to the cancer extends beyond the submucosal layer 500 μm. P50 (P25, P75), median (25th percentile, 75th percentile).

Analysis of influencing factors of R0 removal rate

It is worth noting that, in addition to operating time, complete removal rate (R0) is also a major reference index for evaluating ESD efficiency.⁸ Many cancer patients already have distant micrometastasis during surgery. Complete R0 resection may also lead to disseminated metastasis, so there are still some patients with postoperative recurrence and metastasis. In this study, 85 patients (98.8%) underwent whole resection of R0 and 77 patients (89.5%) underwent curative resection. In the traditional group, 36 patients (100%) of R0 were resected as a whole, among which 2 patients invaded the submucosal deep layer, 1 patient of tumor cells invading the mucosal muscular layer had blood vessel invasion, and 33 patients (91.7%) were cured. Among the 50 cases in the improvement group, 1 patient showed postoperative pathological findings of intramuscular propria invasion, positive cutting margin, and whole R0 resection in 49 patients (98.0%). In addition, 5 patients of tumor cells invaded the deep submucosal layer, and 44 patients (88.0%) were cured. Univariate analysis by chi-square test showed no significant effect of the two methods on the R0 resection rate.

Discussion

With the development of endoscopic diagnosis, the detection rate of early gastric lesions, including precancerous lesions and early cancer, is increasing steadily. ESD has become one of the first-line treatment options due to its minimally invasive nature and high cut rate. ESD is effective for the radical treatment of early gastric cancer confined to the mucosa. At the same time, in the stomach, the esophageal gastric junction of the side of the mucosal precancerous lesions in early cancer is the high incidence area.^1–4 Due to the abundant vascular network in the submucosa of the connection of esophagus and stomach, bleeding is very easy to occur during ESD surgery. Especially Siewert II/III lesion location, ESD surgery becomes more difficult. Therefore, many endoscopic surgeons found that it was difficult and technically challenging, and often requiring relatively long operation time.

Treatment strategy in Siewert I type lesions was the same with that in esophageal mucosal lesions. This operation method was basically used to look directly under the mirror. However, due to the special location, II/III type esophagus stomach connection department in the traditional operation method is mainly taken inversion of endoscopic treatment strategies. When Siewert II/III stomach esophagus connection of mucosal lesions prolong, operation time will be increased to a certain extent in intraoperative bleeding, perforation, postoperative pulmonary infection, late-onset hemorrhage, and related surgical complications.^9,10 Previous reports have shown that the size, location, infiltration depth, submucosal fibrosis, perforation, and ulceration of gastric mucosal lesions are factors contributing to the long operation time.^9,11–15 In the multi-factor analysis, it was found that the location of lesions and the size of specimens were the main factors influencing the operation time, while the length and diameter of specimens had a significant influence, and the same conclusion was also reached in previous studies.^9,12–15

Our study also confirmed those results. With the increase of the scope of our treatment, the overall treatment time also increased with the corresponding increase of the total treatment time. The treatment time of the lesser curvature with the posterior wall was also more convenient and efficient than that of the greater curvature with the anterior wall. For all the research factors included in this study, high-definition endoscopy, ultrasonic endoscopy, and magnifying endoscopy were accurately evaluated before ESD treatment, so the size of the specimen, the location of the lesion, and the depth of the invasion were all predictable before surgery. Therefore, it was necessary to adopt a reasonable treatment strategy at this time.

In our study, we found that the ESD surgery in the whole-course antegrade endoscopic approach for the Siewert II/III type mucosal lesions of EGJ was more convenient and efficient than that in the traditional ESD treatment of the retrograde endoscope, especially to the lesions in the lesser curvature and the posterior wall. The reasons were as follows. First, our experience was at the edge of gastric fundus/cardia/upper gastric body position, using traditional inversion of endoscopic operation process.

Due to the limited endoscopic equipment, it often needed to remote operation, and it was unable to fully close to see each step of the operation. It increased the operation of the probability of damage blood vessels and the stomach muscle layer, so as to prolong the operation time and corresponding risks. In the whole-course antegrade endoscopic approach, the surgeon can close watch submucosa and is easy to find the tiny blood vessels. When it was found, blood vessels can undertake preliminary hemostatic treatment and significantly reduce the probability of intraoperative bleeding. If bleeding, antegrade endoscopic approach was also easier to manage the bleeding than the state of the retrograde endoscope. Second, during the operation of retrograde endoscopy, endoscopic instruments tended to be perpendicular to the gastric wall, and endoscopy was difficult to drill into the submucosa. In the process of mucosal incision and submucosa dissection, the muscle layer of the gastric wall was easily damaged and increased the probability of bleeding. In the whole-course antegrade endoscopic approach, the endoscope can easily drill into the submucosa, and the electric knife device of the endoscope will be parallel to the mucosal layer/submucosal layer/muscle layer, which can be convenient and safe for incision and dissection.

Third, the Siewert II/III type EGJ was fit for ESD treatment. Inversion of endoscopic operation was vulnerable to the interference of respiration, heart rate. The precise positioning of endoscopy and electro tome apex was more difficult. However, during the operation in the antegrade endoscopic approach, the endoscopic body and apex can touch the gastric wall, which can significantly reduce or offset the interference of respiration and heartbeat, and can improve the accurate operation of the treatment process, so as to reflect the efficiency and safety of surgery.

Our study found that, with the enlargement of the lesion range, the treatment in the whole-course antegrade endoscopic approach had more obvious advantages. A key point of ESD treatment in the whole-course antegrade endoscopic approach was to fully attract the gas in the gastric cavity during the operation, so as to maintain the relative negative pressure or low pressure in the gastric cavity, so that the gastric wall can more easily to get close to the apex of the endoscope. When the lesion scope was large, it took a long time to perform submucosal dissection. In this case, under the condition of aspiration in the antegrade endoscopic approach, the endoscopic lens and electrotome were easy to drill into the submucosal layer for rapid and efficient dissection, which can significantly shorten the operation time in comparison with that in the reversed endoscopic state. Another key was to place the endoscopic lens in the 6–7 o'clock position of the display screen or the second and third quadrant position. According to the planar structure of the endoscopic apex, the electrosurgical instrument will extend out of the endoscopic apex just below the ChargedCoupled Device (CCD) of the endoscopic lens, so that the small distance between the electrosurgical instrument and the gastric wall can be observed more clearly during the operation. In the process of submucosal dissection, the microvessels in the submucosa can be observed clearly, and active pre-hemostasis treatment can be carried out to reduce the probability of intraoperative bleeding and shorten the time of ESD surgery.

In our study, it was found that when lesions were confined to the mucosal layer, the ESD surgery in the whole-course antegrade endoscopic approach was more safer and efficient. When the lesions invaded the submucosa and the submucosa had adhesive fibrosis, the advantage of the antegrade endoscopic approach should be more obvious because the structure of the submucosa could be clearly observed under the straight endoscopic. Dual knife was used to carefully separate the mechanically bonded areas to minimize bleeding and perforation. However, the limited number of such cases cannot be shown from the statistical results, but the relatively efficient results can be seen from the median operation time of the limited cases. It is believed that, with the accumulation of cases or the development of multicenter studies, the advantages of the ESD surgery in the whole-course antegrade endoscopic approach will also be verified.

The limitations of this study were mainly as follows. First, because our analyses were retrospective and conducted at a single center, selection and referral biases cannot be excluded. Second, the number of cases in this study was limited, especially the cases with deeper lesion invasion than the submucosa, and the cases with special positions of the anterior wall. Great bend of EGJ were very limited, and these especial cases were impossible to carry out effective statistical comparison. Third, with the accumulation of surgical doctors' experience in ESD treatment, the corresponding operation time will be relatively shortened, which will also affect the evaluation of universal applicability. It is worth noting that, in addition to operation time, the integer cut rate is also an effective indicator to evaluate the operational efficiency of ESD. Therefore, it is necessary to carry out a larger scale prospective clinical study involving different grades of hospitals and endoscopists with different operating experience.

In conclusion, through analyzing three major operating efficient indexes of ESD in mucosal lesions of EGJ, including specimen size, lesion location, and lesion invasion depth, the ESD surgery in the whole-course antegrade endoscopic approach was a more effective treatment strategy for the Siewert II/III type mucosal lesions of EGJ.

Footnotes

Disclosure Statement

No competing financial interests exist.

Funding Information

No funding was received for this article.

References

Soetikno

, Kaltenbach

, Yeh

, et al. Endoscopic mucosal re section for early cancers of the upper gastrointestinal tract. J Clin Oncol, 2005; 23:4490–4498.

Probst

, Golger

, Amholdt

, et al. Endoscopic submucosaldissection of early cancers, flat adenomas, and submucosal tumors in the gastrointestinal tract. Clin Gastroenterol Hepatol, 2009; 7:149–155.

Maple

, Abu

DBK

, Chauhan

, et al. Endoscopicsubmucosal dissection. Gastrointest Endosc, 2015; 81:1311–1325.

Kim

, Cho

, Jain

, et al. ESD around the world: Asia. Gastrointest Endosc Clin N Am, 2014; 24:283–293.

Chung

, Lee

, Iee

, et al. Therapeutic outcomes in 1000 cases of endoscopic submucosal dissection for early gastric neo plasms: Korean ESD Study Group multicenter study. Gastrointest Endosc, 2009; 69:1228–1235.

Ahn

, Choi

, et al. Procedure time of endoscopic submucosal dissection according to the size and location of early gastric cancers: Analysis of 916 dissections performed by 4 experts. Gastrointest Endosc, 2011; 73:911–916.

Nagata

, Honda

, Kobayakawa

, et al. Mycobacterium lentiflavum ileitis using aspirated intestinal fluid during endoscopy in HIV-infected patient. Dig Endosc, 2011; 23:271–272.

Japanese Gastric Cancer Association. Japanese classification of gastric carcinoma: 3rd English edition. Gastric Cancer, 2011; 14:101–112.

Goto

, Fujishiro

, Kodashima

, et al. Is it possible to predict the procedural time of endoscopie submucosal dissectioce for early gastric cancer?. J Gastrxenterol Hepatol, 2009; 24:379–383.

10.

, Yang

, Feng

, et al. Predictive factors of enlosxopie subrauscosal dissection procedure time for gastric superficial neo plasial. World J Gastroenterol, 2012; 18:7009–7014.

11.

Kurokawa

, Sasako

, Doki

. Treatment approaches to esophagogastric junction tumors. Dig Surg, 2013; 30:169–173.

12.

Lu xiaoqiang

Huang rui

, Guo meng, et al. Analysis of related factors affecting the operating efficiency of endoscopic submucosal dissection for gastric mucosal lesions. Chinese J Dig Endosc, 2018; 5:723–726.

13.

Wang

, Shen

, et al. Effect of endoscopic submucosal dissection in the treatment of esophageal and gastric junction early cancer and precancerous lesions. Chinese J Gastrointest Surg, 2016; 2:223–224.

14.

Shimizu

, Fujisaki

, Omae

, et al. Treatment outcomes of endoscopic submucosal dissection for adenocarcinoma originating from long-segment Barrett's esophagus versus short-segment Barrett's esophagus. Dig, 2018; 97:316–323.

15.

Zhang

. Comparison of clinical outcomes of ESD vs. STER for esophagus, esophagogastric junction, and gastric tumor (cancer) originating from the muscularis propria layer: A systematic review and pooled analysis. Am J Gastroenterol, 2016; 111:s159.