Int J Clin Exp Med 2018;11(12):13083-13091www.ijcem.com /ISSN:1940-5901/IJCEM0077944

Original ArticlePreliminary study on endoscopic trans-esophageal submucosal tunneling surgery: a new therapeutic approach

Qianqian Chen*, Ying Xiong*, Xiaobin Zhang, Ningli Chai, Enqiang Linghu

Department of Gastroenterology and Hepatology, Chinese PLA General Hospital, Beijing 100853, China. *Equal contributors and co-first authors.

Received April 15, 2018; Accepted November 10, 2018; Epub December 15, 2018; Published December 30, 2018

Abstract: Background: Natural orifice transluminal endoscopic surgery (NOTES) provides minimally invasive alterna-tive access to the peritoneal cavity, avoiding abdominal wall incisions. The current study presents a new therapeutic approach, endoscopic trans-esophageal submucosal tunneling surgery (EESTS), aiming to protect overlying mucosa to the maximum extent. Some preliminary explorations were carried out for preoperative localization, surgical po-sitioning, incision approaches of the muscularis propria, and differentiation of endoscopic and standard anatomic images. Methods: In this study, 27 porcine corpses were tested. The developed method was divided into 4 parts. In part 1, 6 pigs were randomly divided into two groups: methylene blue solution group (ML-group) and control group (CT-group). The duration of operation starting from the creation of the tunnel incision up to the entering of the endo-scope into abdominal cavity was recorded. In part 2, 9 pigs were randomly divided into three groups, L-group: fixed in the left-lateral position, S-group: fixed in the supine position, and RR-group: fixed in the raised right shoulder posi-tion. Difficulties related to the operation and endoscopic view were also recorded. In part 3, 9 pigs were randomly divided into three groups: transverse full-thickness incision group (T-group), longitudinal full-thickness incision (L-group), and progressive longitudinal full-thickness incision group (PL-group). In part 4, EESTS was performed to record and differentiate endoscopic and standard anatomical images. Results: In part 1, duration of the operation in the ML-group (21.67 ± 2.08 minutes) was shorter than that in the CTL-group (15.00 ± 3.00 minutes). In part 2, the RR-group presented with a better entrance site, shorter duration of operation (14.7 ± 1.5 minutes), straight tunnel, appropriate endoscopic vision, and easier operation. In part 3, the PL-group with a 2-cm incision length had proper flexibility of the endoscope and a straight tunnel, which could also be used for future operations. In part 4, the abdominal aorta, left hepatic lobe, inferior vena cava, splenic vein, gastric fundus, and spleen pancreas were observed under different endoscopic conditions. Conclusion: The raised right shoulder position, preoperative local-ization, and progressive longitudinal full-thickness incision were optimally achieved. In addition, endoscopic images were recorded in all positions. This represents a proper basis for future surgeries.

Keywords: Endoscopic trans-esophageal submucosal tunneling surgery, endoscopic tunneling technique, full-thickness resection, gastrointestinal tract

Introduction

Regarding endoscopy, several studies have concentrated on treating deeper diseases, even diseases outside the gastric wall, with extremely minimally invasive methods [1-3]. Natural orifice transluminal endoscopic surgery (NOTES) consists of several new endoscopic and surgical entryways into the abdominal cav-ity. It has been basically regarded as one of the potential paths for flexible endoscopy beyond the gastrointestinal wall [4, 5]. Many novel diag-nostic approaches have been proposed for ani-

mal models, demonstrating appropriate re- sults. These include transgastric peritoneosco-py, gastrojejunostomy, transvaginal cholecysec-tomy liver resection, and lymph node biopsy [6-10]. Therefore, surgery from intramural to extramural diseases may be a trend and NOTES may overcome some serious problems in gastroenterology.

However, there are still several unanswered questions about the advantages and disadvan-tages of NOTES. Through gastric access, NOTES cannot be performed under a sterile environ-

Preliminary study of EESTS: a new therapeutic approach

13084 Int J Clin Exp Med 2018;11(12):13083-13091

ment [11, 12]. Thus, incidence of infections and complications of arteriovenous fistulas would be remarkable [11, 12]. Peroral endo-scopic myotomy (POEM) is the application of esophageal myotomy to NOTES through the uti-lization of the submucosal tunneling method. It can protect the overlying gastric mucosa, which can effectively resist bacterial invasion and reduce complications [13-15]. Based on this theory, the investigators originally attempted the transluminal endoscopic surgery procedure via the submucosal tunneling technique, called endoscopic trans-esophageal submucosal tun-neling surgery (EESTS).

The present study conducted a preliminary investigation on preoperative localization, sur-gical positioning, and muscularis propria, aim-ing to search for an optimal method for entrance of the endoscope into the peritoneal cavity. In addition, several approaches regarding the operation of an endoscope to reach a vision were investigated. This was based on the navi-gation system for an endoscope and the dis-crepancy between endoscopic images and standard anatomical images.

Materials and methods

Animals

For experimentation, the 14 male and 13 fe- male porcine corpses were contributed by the Animal Laboratory of Pinggu District Hospital (Beijing, China). The present study was ap- proved by the Ethics Committee of the Animal Facility of Chinese PLA General Hospital.

gastroscope (GIF-Q260J, Olympus Medical Sy- stems, Corp., Tokyo, Japan), an electric knife (KD-V451M, Olympus Medical Systems, Corp., Tokyo, Japan), and an injection needle (NM- 200L-0425, Olympus Medical Systems, Corp., Tokyo, Japan) were employed.

Study protocol

Investigating preoperative localization: EESTS, using the submucosal tunneling technique, was undertaken as follows: 1) Si pigs were fixed in the left-lateral position; 2) The methyl blue/nor-mal saline mixture with concentrations of 1:2 and 1:10000 was prepared, respectively, as shown in Figure 2A. The experimental pigs were randomly divided into two groups: methy-lene blue solution group (ML-group) and control group (CT-group); 3) In the ML-group, the methy-lene blue solution (1:2) was injected into sub-mucosal layer in the site of the right posterior wall of the cardia for localization (Figure 2B). In the CT-group, no injection was made; 4) 10-mL methylene blue solution (1:10,000) was inject-ed to form a submucosal cushion in the right posterior wall, which was 5 cm above the esophagogastric junction (EGJ) (Figure 2C); 5) An inverted T-shaped incision was created in the mucosal layer of the esophagus (Figure 2D) in both groups; 6) An endoscopy was intro-duced into the submucosal space and the sub-mucosa was gently separated from the muscu-laris propria, creating a submucosal tunnel in both groups. The dark blue localization spot in the cardia should be sought in the ML group (Figure 2E). In the CTL group, the endoscope

Figure 1. Preoperative preparation: (A) A self-made over-tube; (B) Installation and fixing the self-made over-tube.

Preoperative preparation

The porcine corpses were defrosted at a temperature of -20°C. They were defrosted for 3 days before beginning the experiment. The pigs were assigned in the supine posi-tion and positioned at the right shoulders. A self-made transparent over-tube, pre-pared using a 50-mL syringe, removing the piston and cut-ting off the end of the con-necting needle, was employed before starting the experi-ment (Figure 1). The proce-dure was performed by an experienced endoscopist. A

Preliminary study of EESTS: a new therapeutic approach

13085 Int J Clin Exp Med 2018;11(12):13083-13091

was withdrawn from the submucosal tunnel and entered into the stomach towards the gas-tric fundus, aiming to observe the tunnel posi-tion; 7) A full-thickness incision was created through the muscularis propria and serosa, as presented in Figure 2F. Next, the scope was moved into lesser sac or omental bursa, which is a potential peritoneal space within the abdo-men and part of the peritoneal cavity (Figure 2G); 8) The endoscopist carefully identified the other organs of the abdominal cavity, such as the liver, inferior vena cava, abdominal aorta, spleen, and posterior wall of the stomach. The scope was repeatedly entered through the abdominal cavity and the endoscopist contrast-ed the location of these organs; 9) Necropsy was carried out to observe the incision site,

where the scope was entered into the abdomi-nal cavity (Figure 2H); 10) The duration of the operation from submucosal tunnel creation to entrance of the scope into abdominal cavity was recorded.

Assessment of surgical positioning: EESTS was conducted, using the submucosal tunnel-ing technique, as follows: 1) Nine pigs were ran-domly divided into three groups: left-lateral position (L-group), supine position (S-group), and raised right shoulder position (RR-group) (Figure 3A-C); 2) Steps in the section, “Inve- stigating the preoperative localization” (steps 3-9) were implemented; 3) Challenges related to the operation and endoscopic vision were recorded.

Figure 2. Operation procedure to study the preoperative localization: (A) Methylene blue solution (1:2) in the right and the methylene blue solution (1:10000) was in the left; (B) A 0.5-mL methylene blue solution (1:2) was injected into submucosal layer at the site of the right posterior wall of the cardia for localization; (C) The methylene blue solu-tion (1:10000) was injected into the submucosal layer to create a submucosal cushion; (D) An inverted T-shaped incision was established; (E) A deep blue preoperative mark was observed underneath the cardia, as shown by the yellow arrow; (F) The muscularis propria and serosa were incised; (G) The lesser omental sac was bluntly dissected using an electric knife; (H) A site where the endoscope was entered into the abdominal cavity was observed.

Figure 3. Investigating surgical positioning: (A) Left-lateral position; (B) Supine position; (C) Raised right shoulder position.

Preliminary study of EESTS: a new therapeutic approach

13086 Int J Clin Exp Med 2018;11(12):13083-13091

Incision approaches of muscularis propria: EESTS was conducted, using the submucosal tunneling technique, as follows: 1) Nine pigs were randomly divided into three groups: trans-verse full-thickness incision (T-group), longitudi-nal full-thickness incision (L-group), and pro-gressive longitudinal full-thickness incision (PL-group) (Figure 4A-C); 2) The pigs were fixed in the raised right shoulder position; 3) The steps in the section, “Investigating the preop-erative localization” (steps 3-6) were imple-mented; 4) In the T-group, the muscularis pro-pria and serosa were incised in a transverse way using an electric knife. The incision leng- th of the gastric parietal muscle wall was 1.5 cm, as depicted in Figure 4A. In the L-gro- up, the incision was made in a longitudinal manner. The incision length of the gastric pari-etal muscle wall was 1.5-2.0 cm (Figure 4B). In the PL-group, the incision was made in the progressive longitudinal approach. In addition, an incision was progressively made from 1-2 cm in the oral margin of the preoperative lo- cation to the distal margin. The depth of inci-sion was from the circular muscle layer to the longitudinal muscle layer. A full-thickness incision was made by incising the serosa. The scope was entered through the lesser

Differences between endoscopic and standard anatomical images: EESTS was performed, using the submucosal tunneling technique, as follows: 1) Three pigs were fixed in the raised right shoulder position; 2) Blue methylene solu-tion (1:2) was used for intraoperative localiza-tion; 3) A full-thickness incision was made in a progressive longitudinal manner; 4) Endoscopic and standard anatomical images were record-ed and differentiated.

Statistical analysis

Statistical analyses were performed using SPSS statistical software version 22.0 (SPSS, Inc.; Chicago, IL, USA). Student’s t-test was used for continuous variables and the results are presented as mean ± standard deviation (SD). P≤0.05 indicates statistical significance.

Results

Preoperative localization

In the ML-group, 3 endoscopes were entered in the abdominal cavity through the same route from the posterior wall of gastric cardia to the lesser omental sac. In the CTL-group, 3 endo-

Figure 4. Investigation of incision approaches of the muscularis propria: (A) Transverse full-thickness incision; (B) Longitudinal full-thickness incision; (C) Progressive longitudinal full-thickness incision.

Table 1. Comparing incision sites and duration of operation between ML- and CT-groups (mean ± SD)Group Duration of operation Penetration siteCT 21 minutes Upper part of the lesser omental sac

24 minutes Not in the lesser omental sac20 minutes Posterior lower part of the lesser omental sac

ML 15 minutes Middle and upper part of the lesser omental sac12 minutes Middle part of the lesser omental sac18 minutes Middle part of the lesser omental sac

omental sac. The incision length of the gastric pari-etal muscle wall was 2.0 cm, as shown in Figure 4C; 5) Status of the endo-scope, operational chal-lenges, flexibility of the endoscope, and experi-ences and difficulties re- lated to endoscopic vi- sion were recorded.

Preliminary study of EESTS: a new therapeutic approach

13087 Int J Clin Exp Med 2018;11(12):13083-13091

scopes were entered in the abdominal cavity through different routes. One of these was not entered into the lesser omental sac (Table 1), however. Duration of operation in the ML-group (21.67 ± 2.08 minutes) was shorter than that in the CTL-group (15.00 ± 3.00 minutes), with P = 0.034 (Table 1).

ris propria

Factors including incision length of gastric pari-etal muscle wall, flexibility of the endoscope, straightness of the tunnel, and practicability of further operations were assessed and differen-tiated. Results revealed that the progressive

Figure 5. Assessment of surgical positioning: (A) In the L-group, the tail of the injection needle was at the eight o’clock position; (B) In the L-group, the submucosal cushion was at the three o’clock position; (C) In the L-group, liquid accumulated in the left wall, as illustrated by the yellow arrow, and could not obscure the vision of the sub-mucosal cushion; (D) In the S-group, the posterior wall of the esophagus was at the six o’clock position; (E) In the S-group, liquid accumulated in the posterior wall, which seriously affected endoscopic vision; (F) In the RR-group, endoscopic vision was clear, because the liquid in the esophagus did not inundate the entrance of the tunnel.

Table 2. Investigation of 3 operative positions

Group Entrance site Duration of operation (mean ± SD)

Straightness of tunnel

Endoscopic vision Operation difficulty

L-group Three o’clock 18.0 ± 2.0 minutes No Excellent DifficultS-group Six o’clock 16.3 ± 0.6 minutes Yes Medium A little difficultRR-group Five-six o’clock 14.7 ± 1.5 minutes Yes Good Easy

Table 3. Assessment and differentiation of 3 full-thickness incision ways

Group Incision length

Flexibility of endoscope

Straightness of tunnel

Practicability of further operation

T-group 1.5 cm Bad No NoL-group 1.5 cm Medium No No

2 cm Good Yes YesPL-group 2 cm Good Yes Yes

Surgical positioning

Factors including entrance of the tun-nel, duration of operation (P = 0.242), straightness of the tunnel, endoscop-ic vision, and operation difficulty were investigated, finding that the raised right shoulder position was optimal (Figure 5; Table 2).

Incision approaches for the muscula-

Preliminary study of EESTS: a new therapeutic approach

13088 Int J Clin Exp Med 2018;11(12):13083-13091

longitudinal full-thickness incision was optimal (Table 3).

Differences between endoscopy and standard anatomical images

The endoscope was entered in the abdominal cavity in a straight direction through a series of surgical procedures, such as inverted T-shaped incision, progressive longitudinal full-thickness incision, and blunt dissection of the lesser omental sac (Figure 6A-C). After dissecting the porcine corpses, the endoscope was penetrat-

ed through the abdominal cavity from the pos-terior gastric wall to the lesser omental sac (Figure 6D). The abdominal aorta was observed in the status of the endoscope, as shown in Figure 6E. The left hepatic lobe and inferior vena cava were observed by twisting the body of the endoscope to the right (clockwise) or left (counter-clockwise) direction (Figure 6F). The posterior gastric wall and spleen were observed by continuously turning the endoscope upward (Figure 6G). The pancreas was observed by inserting the endoscope along the path of the spleen, as illustrated in Figure 6H. Endoscopic

Figure 6. Study of the differences between endoscopic and standard anatomical images: (A) Inverted T-shaped inci-sion; (B) Progressive longitudinal full-thickness incision; (C) An endoscope penetrated into the abdominal cavity in the direct status; (D) An endoscope penetrated into the abdominal cavity from the posterior gastric wall, as shown by the yellow arrow, indicating endoscopic light; (E) An endoscopic vision of the abdominal aorta was observed in the endoscopic direct status (yellow arrow); (F) The endoscopic vision of the left hepatic lobe (red arrow), inferior vena cava (yellow arrow), and splenic vein (white arrow) was observed by turning the endoscope to the left; (G) The endoscopic vision of the splenic vein (yellow arrow), gastric fundus (white arrow), tail of spleen (red arrow), and inferior vena cava (green arrow) was observed by twisting the body of the endoscope to the right (clockwise) or left (counter-clockwise) direction; (H) The endoscopic vision of the pancreas and spleen, shown by the yellow and red arrows, respectively, was observed by inserting the endoscope along the path of the spleen; (I) Standard anatomical images of the left hepatic lobe (denoted by “a”), spleen (denoted by “b”), and splenic vein (denoted by “c”) were observed after dissecting the porcine corpse.

Preliminary study of EESTS: a new therapeutic approach

13089 Int J Clin Exp Med 2018;11(12):13083-13091

vision was obscured when the body of the endoscope was twisted to the right or left, which led to great difficulties in observation and operation. After dissecting the porcine corpses, standard anatomical images and loca-tions of the endoscope were found, as shown in Figure 6I.

Discussion

EESTS is a technique that utilizes a flexible endoscope through the gastric wall and to organs surrounding the stomach in the abdomi-nal cavity. It cannot be performed by strict aseptic manipulation, however, and is closed completely by endoscopic clips, leading to a high ratio of infection and other complications. Therefore, finding an effective approach to pro-tect the integrity of the intestinal mucosa has become increasingly significant [16, 17]. The submucosal tunneling technique creates a sub-mucosal tunnel between mucous layer and intrinsic myometrium, allowing the endoscope to be enter into the extraluminal cavities and finally closed completely [18-20]. In the present study, the EESTS strategy was proposed using the submucosal tunneling technique. In addi-tional, it was revealed that the proposed approach could maintain the integrity of the overlying mucosa, minimizing contamination from intra-luminal contents and bacteria.

The abdominal aorta is located at the lesser curvature of the posterior wall of the gastric cardia. If NOTES is performed through the car-dia, it may result in severe complications, including diaphragm injury, mediastinal emphy-sema, or pneumothorax. Thus, the lesser cur-vature of the posterior wall of the gastric fun-dus was selected as the endoscopic access route to the abdominal cavity. However, a flexi-ble endoscope could not maintain a straight state after perforating due to the large angle between the gastric fundus and esophagus. Hence, entering an endoscope into the abdomi-nal cavity through the same endoscopic access route would be a technical challenge. In the present study, blue methylene solution (1:2) was injected into the submucosal layer in the site of the right posterior wall of the cardia, appearing to be an effective method for surgi-cal localization. In the ML-group, the endoscope could be entered in the abdominal cavity through the same endoscopic access route.

Results indicated that the duration of operation in the ML-group was shorter than that in the CTL-group. Accordingly, this was carefully in- vestigated. First, in the ML-group, localization of the methylene blue could be used to navi-gate the endoscope to enter in the tunnel, cre-ating a straight tunnel. When a blue mark was observed, a full-thickness incision could be made at this site, allowing the endoscope to enter the abdominal cavity. In the CTL-group, the observation of the direction of the tunnel by repeatedly withdrawing back the endoscope from tunnel when creating a tunnel was very time-consuming. Second, the methylene blue solution was injected in the same site for local-ization, allowing for easy observation in the gastric fundus near the cardia. This resulted in the entrance of the endoscope into the abdomi-nal cavity through the same endoscopic access route. In conclusion, methylene blue solution (1:2) for surgical localization provides an appro-priate basis for further investigation.

However, another problem presented, causing tunnel establishment to be difficult. The poste-rior wall of the esophagus was located at the three o’clock position. To solve this problem, surgical positioning was further investigated by differentiating the left-lateral, supine, and raised right shoulder positions. Surgery in the L-group was the most difficult, because the channel of the used endoscope was located at the eight o’clock position, causing the head of the used endoscope accessory to be at the six o’clock position. Simultaneously, the posterior wall of the esophagus located at the three o’clock position was not an optimal site for this surgery. The endoscope should be turned left by 90° to maintain the posterior wall of esoph-agus on the six o’clock position during the pro-cedure. This resulted in difficult operations. Creating a tunnel and no straight tunnel was time consuming, making the surgery more dif-ficult. This position presents an advantage, including appropriate endoscopic vision. In the S-group, the posterior wall of the esophagus was at the six o’clock position and had poor endoscopic vision, due to fluid accumulation in the tunnel entrance. In the RR-group, the right upper limb and shoulder of the pigs were raised to 15-20°. The posterior wall of the esophagus was at the five to six o’clock position. This group had the easiest operation procedure, the short-est duration of operation, and optimal endo-

Preliminary study of EESTS: a new therapeutic approach

13090 Int J Clin Exp Med 2018;11(12):13083-13091

scopic vision. In conclusion, the raised right shoulder position was the most appropriate surgical positioning.

In the present study, the best endoscopic access route to the abdominal cavity, when using the submucosal tunneling technique, was successfully determined by EESTS. Moreover, it was revealed that the endoscope would become inflexible when passing through the transverse incision to the abdominal cavity. In this case, if the endoscope was continuously entered in the abdominal cavity, the endoscope would turn in any direction, making subsequent surgeries extremely difficult. Therefore, another strategy was developed to explore an optimal incision approach for the muscularis propria. Three groups were introduced: transverse full-thickness incision, longitudinal full-thickness incision, and progressive longitudinal full-thick-ness incision groups. In the T-group, bad flexi-bility of the endoscope was due to the limited incision length of 1.5 cm and the block of the muscularis in the oral side, which caused that the endoscope to enter in the abdominal cavity at a certain angle. In the L-group, the flexibility of the endoscope was related to the limited incision length. Moreover, the limited incision length of 1.5 cm that belonged to the L-group had the same effect as that in the T-group. However, an incision length of 2 cm would pro-vide a wide entrance. In this case, if the endo-scope was entered through the abdominal cav-ity, the endoscope would turn in any direction, leading to another problem (greater injury). This would be detrimental to postoperative recov-ery, the sphincter of cardia dysfunction after the long incision of myotomy, and gastroesoph-ageal reflux disease. Transverse and longitudi-nal full-thickness incisions were both in one-off total myotomy approaches, which could easily injure large vessels in the spatium intermuscu-lare and extra lumen. However, a progressive longitudinal full-thickness incision could over-come these problems by performing the inci-sion from shallow to deep. The incision by layer made hemostasis (electrocoagulation) easy in the operation, maintaining a clear endoscopic vision. Furthermore, an incision with a length of 2 cm provided enough space to enter the endo-scope into the abdominal cavity in a straight state, as well as keeping the flexibility of the endoscope. In conclusion, a progressive longi-tudinal full-thickness incision presents several advantages.

Subsequently, the surgery was performed with the help of the methylene blue solution (1:2) for localization in the raised right shoulder position and the use of a progressive longitudinal full-thickness incision. This caused the endoscope to enter in the abdominal cavity through a route starting from the posterior wall of the gastric cardia to the middle part of the lesser omental sac. Endoscopic and standard anatomical images were differentiated to define landmarks during the operation.

Conclusion

To reduce the abdominal complications of NOTES, a novel strategy was proposed, EESTS. This strategy used the submucosal tunneling technique. This approach could protect overly-ing mucosa to the maximum extent. In the pres-ent study, some preliminary explorations were performed, finding that the raised right shoul-der position for surgery, methylene blue solu-tion (1:2) for preoperative localization, and pro-gressive longitudinal full-thickness incisions were optimal. Afterward, differences between endoscopic and standard anatomical images were differentiated.

Acknowledgements

We would like to thank professors LINGHU Enqiang and CHAI Ningli for providing technical support for this study. We acknowledge the financial support provided by Chinese PLA Ge- neral Hospital (Grant No. 2012FC-TSYS-3035).

Disclosure of conflict of interest

None.

Abbreviations

NOTES, Natural orifice transluminal endoscopic surgery; EGJ, Esophagogastric junction; POEM, Peroral endoscopic myotomy.

Address correspondence to: Drs. Enqiang Linghu and Ningli Chai, Department of Gastroenterology and Hepatology, Chinese PLA General Hospital, Beijing 100853, China. Tel: +86-010-55499305; E-mail: linghuenqiang@sina.vip.com (EQLH); Tel: +86-010-55937485; E-mail: csxlily@163.com (NLC)

References

[1] Kantsevoy SV and Armengol-Miro JR. Endo-scopic suturing, an essential enabling technol-

Preliminary study of EESTS: a new therapeutic approach

13091 Int J Clin Exp Med 2018;11(12):13083-13091

ogy for new NOTES interventions. Gastrointest Endosc Clin N Am 2016; 26: 375-384.

[2] Lee GC and Sylla P. Shifting paradigms in mini-mally invasive surgery: applications of trans-anal natural orifice transluminal endoscopic surgery in colorectal surgery. Clin Colon Rectal Surg 2015; 28: 181-193.

[3] Cho WY, Kim YJ, Cho JY, Bok GH, Jin SY, Lee TH, Kim HG, Kim JO and Lee JS. Hybrid natural ori-fice transluminal endoscopic surgery: endo-scopic full-thickness resection of early gastric cancer and laparoscopic regional lymph node dissection--14 human cases. Endoscopy 2011; 43: 134-139.

[4] Saxena P and Khashab MA. New NOTES clini-cal training and program development. Gastro-intest Endosc Clin N Am 2016; 26: 385-400.

[5] Sumiyama K, Gostout CJ and Tajiri H. Investi-gating deeper: muscularis propria to natural orifice transluminal endoscopic surgery. Gas-trointest Endosc Clin N Am 2014; 24: 265-272.

[6] Oliveira AL, Zorron R, Oliveira FM, Santos MB, Scheffer JP, Rios M and Antunes F. Transcol-onic perirectal NOTES access (PNA): a feasibil-ity study with survival in swine model. An Acad Bras Cienc 2017; 89: 685-693.

[7] Katagiri T, Otsuka Y, Horgan S, Sandler BJ, Ja-cobsen GR, Coker AM, Tsuchiya M, Maeda T and Kaneko H. Feasibility and technique for transvaginal natural orifice transluminal endo-scopic surgery liver resection: a porcine model. Surg Laparosc Endosc Percutan Tech 2017; 27: e6-e11.

[8] Park YS, Kim SH, Ryu HY, Cho YK, Jo YJ, Son TI and Hong YO. Hybrid natural orifice translumi-nal endoscopic surgery with sentinel lymph node navigation for deep early gastric cancer in the fundic region. Clin Endosc 2016; 49: 298-302.

[9] Liu L, Chiu PW, Teoh AY, Lam CC, Ng EK and Lau JY. Endoscopic suturing is superior to en-doclips for closure of gastrotomy after natural orifices translumenal endoscopic surgery (NOTES): an ex vivo study. Surg Endosc 2014; 28: 1342-1347.

[10] Zhu HY, Li F, Li KW, Zhang XW, Wang J and Ji F. Transumbilical endoscopic cholecystectomy in a porcine model. Acta Cir Bras 2013; 28: 762-766.

[11] Rezende M, Montero EF, Salomao R, Brunialti M, Rodrigues R, Gomes G, Libera AD, Ferrari A and Libera ED. Acute inflammatory response to transgastric natural orifice transluminal en-doscopic surgery peritoneoscopy: an experi-mental study in swine. Clinics (Sao Paulo) 2013; 68: 1433-1439.

[12] Cordova H, Cubas G, Boada M, Rodriguez de Miguel C, Martinez-Palli G, Gimferrer JM and Fernandez-Esparrach G. Adverse events of NOTES mediastinoscopy compared to conven-tional video-assisted mediastinoscopy: a ran-domized survival study in a porcine model. Endosc Int Open 2015; 3: E571-576.

[13] Eleftheriadis N, Inoue H, Ikeda H, Onimaru M, Maselli R and Santi G. Submucosal tunnel en-doscopy: peroral endoscopic myotomy and per-oral endoscopic tumor resection. World J Gas-trointest Endosc 2016; 8: 86-103.

[14] Wong I and Law S. Peroral endoscopic myoto-my (POEM) for treating esophageal motility dis-orders. Ann Transl Med 2017; 5: 192.

[15] Inoue H, Santi EG, Onimaru M and Kudo SE. Submucosal endoscopy: from ESD to POEM and beyond. Gastrointest Endosc Clin N Am 2014; 24: 257-264.

[16] Guarner-Argente C, Beltran M, Martinez-Palli G, Navarro-Ripoll R, Martinez-Zamora MA, Cor-dova H, Comas J, de Miguel CR, Rodriguez-D’Jesus A, Almela M, Hernandez-Cera C, Lacy AM and Fernandez-Esparrach G. Infection dur-ing natural orifice transluminal endoscopic surgery peritoneoscopy: a randomized com-parative study in a survival porcine model. J Minim Invasive Gynecol 2011; 18: 741-746.

[17] Teoh AY, Chiu PW, Chan SM, Wong TC, Lau JY and Ng EK. Direct incision versus submucosal tunneling as a method of creating transgastric accesses for natural orifice transluminal endo-scopic surgery (NOTES) peritoneoscopy: ran-domized controlled trial. Dig Endosc 2013; 25: 281-287.

[18] Kobara H, Mori H, Rafiq K, Fujihara S, Nishiya-ma N, Ayaki M, Yachida T, Matsunaga T, Tani J, Miyoshi H, Yoneyama H, Morishita A, Oryu M, Iwama H and Masaki T. Submucosal tunneling techniques: current perspectives. Clin Exp Gastroenterol 2014; 7: 67-74.

[19] Gomercic C, Vanbiervliet G, Gonzalez JM, Saint-Paul MC, Garces-Duran R, Garnier E, He-buterne X, Berdah S and Barthet M. Prospec-tive randomized comparison of endoscopic submucosal tunnel dissection and convention-al submucosal dissection in the resection of superficial esophageal/gastric lesions in a liv-ing porcine model. Endosc Int Open 2015; 3: E577-583.

[20] Kono Y, Yasuda K, Hiroishi K, Akagi T, Kawagu-chi K, Suzuki K, Yoshizumi F, Inomata M, Shi-raishi N and Kitano S. Transrectal peritoneal access with the submucosal tunnel technique in NOTES: a porcine survival study. Surg En-dosc 2013; 27: 278-285.