Pure NOTES Rectosigmoid Resection: Transgastric Endoscopic IMA Dissection and Transanal Rectal Mobilization in Animal Models

Introduction

Natural orifice translumenal endoscopic surgery (NOTES^®; American Society for Gastrointestinal Endoscopy [Oak Brook, IL] and the Society of American Gastrointestinal and Endoscopic Surgeons [Los Angeles, CA]) is an evolving technique that represents the future of minimally invasive surgery. Since the report of a successful liver biopsy using a transgastric endoscopic approach appeared in 2004, ¹ NOTES has been the focus of not only small end-organ procedures, such as appendectomies and cholecystectomies,^2–6 but also large organ procedures involving the colon.^7–9 Whiteford et al. ⁷ described NOTES sigmoidectomy using transanal endoscopic microsurgery (TEM) in human cadavers; they showed that adequate colon and mesenteric dissection by laparoscopic tools could be achieved transanally, but with some technical challenges. The acute angle created by the sacral promontory is the major anatomic obstacle limiting instrument access, and for more extensive resections, flexible and longer instruments would be required. The techniques of endoscopic intraluminal surgery and TEM have not yet been perfected so as to replace laparoscopy in terms of creating reliable traction, dissecting major vessels, or managing unexpected operative events. ¹⁰ There are many potential barriers to the clinical application of NOTES, such as intestinal closure or suturing, anastomotic devices, or a multitasking platform.^2,11

We investigated whether endoscopic procedure of major vessels is feasible using a pure NOTES approach. In the present study we aimed to determine if endoscopic dissection of the inferior mesenteric artery (IMA) is feasible and safe in pure NOTES rectosigmoidectomy assisted by transanal rectal mobilization in animal models.

Materials and Methods

All animal experiments were performed in the Animal Laboratory of Konkuk University, Seoul, Korea, from December 2010 to November 2011. The study protocol was approved by the Institutional Animal Care and Use Committee of Konkuk University (protocol number KU11078). NOTES rectosigmoidectomy was performed on 10 live animals (2 pigs weighing 35–40 kg each and 8 dogs weighing 25–30 kg each).

Results

A hybrid method was performed for the first 3 cases, and a pure method was used for the other 7 cases, 1 of which was converted to a hybrid procedure because a Veress needle was required to control intra-abdominal pressure. Adequate anatomic exposure around the IMA was achieved by transanal insertion of a circular stapler and traction of the mesocolon. Endoscopic dissection of the IMA was successful in all cases. However, minor bleedings occurred in 3 cases, of which 1 came from lymph nodes near the IMA and 2 were due to inadequate clipping.

Overall, the mean operation time was 180.5 minutes (range, 145–210 minutes). The mean time from dissection and clipping to the division of the IMA was 36.7 minutes (range, 25–45 minutes). A submucosal tunnel 4 cm long was created to secure closure but resulted in a longer procedure time and mucosal edema followed by rather difficult endoscopic clipping. A SILS port was used for rectal mobilization in 3 cases. Dissection under direct visualization was easily performed given the length of the rectum was short and straight. However, a SILS port was useful for monitoring intra-abdominal pressure or dissecting the mesorectum endoscopically. The mean length of the resected specimen was 11.2 cm (range, 9–17 cm). Closing the seromuscular layer outside was attempted by clipping endoscopically via the anus after clipping the mucosa inside perorally. However, the seromuscular layer was too thick to be picked up with a clip. There were no intraoperative complications or episodes of hemodynamic instability.

Discussion

The future of NOTES has yet to be determined—will it remain just a maneuver of intraperitoneal biopsy, or will it develop into a universal, minimally invasive technique for performing transluminal surgery? The clinical application of NOTES to large organs, as well as to the appendix and gallbladder, hinges on many barriers, including successful suturing/closure, dissection, traction, and ligation. Many companies are currently trying to develop new types of endoscopes and instruments to overcome these challenges.

The feasibility of rectosigmoid resection using transrectal NOTES has been assessed in human cadavers and animal models.^7,10,12,13 The method of using transrectal TEM as a NOTES approach is quite attractive but varies from the approach in this study in terms of transgastric endoscopic IMA dissection with assistance of transanal rectal mobilization. Sylla et al. ¹³ stated that TEM is not the ideal platform for transrectal NOTES because it is difficult to manipulate the rigid proctoscope and instruments at the acute angle of the sacral promontory. They performed rectal mobilization using TEM but preserved the sigmoid mesentery to avoid bleeding. We initially began NOTES using a hybrid method and then changed to pure NOTES from the fourth case onward. We alternately used a dog and a pig for the first 4 cases simply because we wanted to know whether there are any differences between the two animal species during NOTES. We found that the pig has a thicker submucosal layer and a thinner seromuscular layer than the dog, and this made gastric endoscopic submucosal dissection much easier. We just chose a moderately difficult route for practice before applying it to a cadaver or a living human. Pneumoperitoneum was established through either an endoscope or a SILS port placed in the anus. The SILS port was useful for monitoring intra-abdominal pressure and dissecting the mesorectum endoscopically. In contrast, when pneumoperitoneum was established endoscopically, it was difficult to maintain normal values and to monitor the intra-abdominal pressure during the procedure.

In this study, endoscopic dissection of a major vessel was feasible by a pure NOTES approach. Although endoscopic dissection is a technically demanding and time-consuming procedure, it was safely performed via a magnified view with accumulated experiences of endoscopic submucosal dissection. Most important is that the operating field exposure through traction should be performed before dissection itself. Moreover, reliable methods of vessel ligation are also essential. We achieved the proper traction of the sigmoid mesocolon through transanal introduction of a circular stapler's shaft. ¹⁰ However, this was not available proximally to the sigmoid colon. Thus, in order to establish the pure NOTES procedure in the whole colon, other traction methods would be required, such as magnet traction and the dual-scope technique.^14,15

Even for a small amount of bleeding, endoscopic hemostasis is not easy to ensure unless bleeding control is promptly obtained. Subsequent hematoma surrounding the surgical site may also arrest the progress of the procedure. It is also important to note that quickly handling or changing devices via the endoscope is a difficult task. For endoscopic dissection through electrocautery, various kinds of knives are available. We prefer a Coagrasper, which allows for detailed dissection by biting and pulling little by little. Another problem is the security of endoscopic clipping. Endoscopic clips have been reported to not be reliable for vascular or ductal ligations. ³ In this study, however, endoscopic clipping of the IMA was safely performed, provided that clean dissection around the IMA was done. Otherwise, a clip became obliquely applied to the IMA or got entangled in cluttered tissues. The jaw of the endoscopic clip was not able to cover the entire IMA, leading to bleeding.

We used a transgastric route to establish a straight view toward the lower abdomen. This route also allowed for a circular stapler's insertion via the anus. However, gastric closure still remains one of the major difficulties. Even if gastric closure using endoscopic clipping is performed with air-tightness, only mucosal closure can be suspected. We do not think that clipping is the best way to close a gastrotomy, and some investigators have used more reliable methods such as a T-anchor system ¹⁰ or a detachable snare. ¹⁴ However, because the procedure of deploying T-tags is performed blind, complications such as abscess and adhesion were reported as 50% and 60%, respectively. ¹² The closure of a gastrotomy was not the main interest of this study, but we attempted additional serosal clippings by introducing another endoscope transanally before the colorectal anastomosis. However, we found that the serosal wall was not picked up by endoscopic clips because of its full thickness and stiffness. In contrast, the transvaginal or transcolonic route provides easier opening and closing of the organ wall under direct visualization. ¹⁶ For this access, however, traction should be established. External collision is estimated in the dual-scope technique or transvaginal route. Unless the sigmoid mesocolon is properly exposed by anything else, an endoscopic dissection of the vessel or mesentery might not be allowed. Moreover, in order to overcome the acute angle at the sacral promontory, a new model of transrectal approach may compensate for the limitations of current instrumentations.

Conclusions

A pure NOTES approach to rectosigmoid resection using transgastric endoscopic IMA dissection is technically feasible and relatively safe in animal models. Transanal assistance including mesocolonic traction and rectal mobilization is needed to compensate for the limitations during NOTES procedure so far. For clinical applications, the development of better adapted instruments is critical.