- Email: [email protected]
A Novel Technique for Natural Orifice Endoscopic Full-Thickness Colon Wall Resection: An Experimental Pilot Study
A Novel Technique for Natural Orifice Endoscopic Full-Thickness Colon Wall Resection: An Experimental Pilot Study Erwin Rieder, MD, Danny V Martinec, BS, Christy M Dunst, MD, FACS, Lee L Swanström, MD, FACS Natural orifice endoscopic full-thickness colon resection attempts to overcome the need for invasive surgery in selected colorectal indications. Because basic technical requirements have not been met so far, the aim of this study was to develop a novel technique for endolumenal colon-wedge resection addressing current shortcomings. STUDY DESIGN: Endoscopic full-thickness colon resection was attempted in a human cadaver model (n ⫽ 2), explanted porcine colon stumps (n ⫽ 10), and 3 acute pig models. A hypothesized colon lesion was created and retracted into an endoscopic clip closure system (ECCS). Initially used endoscopic graspers (n ⫽ 2) were replaced by a T-tag suture approach for retraction (n ⫽ 13). T-type anchors were deployed circumferentially to the lesion, which simultaneously marked resection margins. The clip was then applied for pre-resection tissue closure. The inverted tissue was excised by snare resection and was removed together with the sutures. Air leak-pressure of tissue closure was tested. RESULTS: Endoscopic full-thickness colon resection was achieved in 14 of 15 attempts. The mean diameter (⫾SD) of resected animal specimen, including the predetermined margins, was 26 ⫾ 4 mm. Using the T-tag sutures for retraction, the defined lesion was neither touched by an endoscopic grasper nor compromised by puncturing the center. Leak pressure tests revealed a significantly higher air pressure resistance of the pre-resection ECCS closure (61 ⫾ 5 mmHg) compared with the hand-sewn control (26 ⫾ 7 mmHg). CONCLUSIONS: A novel endoscopic technique for full-thickness colon wall resection using tissue anchors for traction and an ECCS for pre-resection tissue closure appears to address several fundamental surgical principles. However, further studies are necessary before initial clinical application. (J Am Coll Surg 2011;213:422–429. © 2011 by the American College of Surgeons) BACKGROUND:
be readily apparent. Current techniques, such as endoscopic piecemeal resection, are controversial due to high recurrence rates or inability to assess the resection margin.3 More recent methods, such as endoscopic submucosal dissection in the colon, are not yet widely used because of their difficulty to master.4 Consequently, surgery was and still is the basic treatment for certain malignant or “difficult” polyps. An endoscopic en-bloc full-thickness resection (eFTR) of the colonic wall together with an easy and reliable method of closure could extend the indication of appropriate endoscopic treatment. Additionally, eFTR would be a tremendous benefit for selected patients by avoiding a large organ resection. So far, several approaches or devices for endoscopic gastrointestinal wall resection have been described with variable success,5-7 however, most would be limited in a clinical setting due to technical reasons or difficulty with providing a reliable closure. Others have described more sophisticated procedures involving resection and secondary closure of the iatrogenic colon perfora-
Colorectal cancer remains among the leading causes of cancer deaths in the United States.1 Colonoscopic screening and endoscopic snare resection of colonic polyps have become standard techniques to reduce the incidence of colorectal carcinoma;2 polyp excision, however, may be challenging for several reasons. In addition, a considerable number of polyps harbor early malignancies that may not Disclosure Information: Dr Swanström receives honoraria for his work as a consultant for Ethicon Endo-Surgery. Ethicon Endo-Surgery, Cincinnati, OH provided the TAS, and Aponos Medical, Kingston, NH as well as OVESCO Endoscopy AG, Tübingen, Germany provided the clip closure systems. The companies were not involved in study design or data acquisition and interpretation. All other authors have nothing to disclose. Received January 24, 2011; Revised April 18, 2011; Accepted May 4, 2011. From the Minimally Invasive Surgery Program, Legacy Health (Rieder, Swanström), and Gastrointestinal and Minimally Invasive Surgery Division, The Oregon Clinic (Martinec, Dunst, Swanström), Portland, OR. Correspondence address: Erwin Rieder, MD, Minimally Invasive Surgery Program, Legacy Health, 1040 NW 22nd Ave, Suite 560, Portland, OR 97210. email: [email protected]
© 2011 by the American College of Surgeons Published by Elsevier Inc.
422
ISSN 1072-7515/11/$36.00 doi:10.1016/j.jamcollsurg.2011.05.015
Vol. 213, No. 3, September 2011
Rieder et al
Endoscopic Full-Thickness Colon Resection
423
Abbreviations and Acronyms
ECCS ⫽ endoscopic clip closure system eFTR ⫽ endoscopic full-thickness colon resection GI ⫽ gastrointestinal
tion.8 This would expose the patient to risk of fecal contamination and subsequent septic complications as well as the possibility of tumor spillage. The aim of this study was to develop a straightforward method for eFTR addressing current technical shortcomings. The novel eFTR technique that we have developed within this feasibility study for the first time appears to meet fundamental surgical principles. These include en-bloc tissue resection with defined tumor margins, eliminating the need to grasp the lesion itself. Simultaneously, any temporarily colonic wall perforation or mandatory secondary closure is avoided.
METHODS Inanimate and animate models
Freshly explanted distal porcine colons (up to 60 cm from the anus) obtained from a local slaughterhouse and used within 3 hours of excision, an acute porcine animal model, and a human cadaver were used for development and evaluation of a novel technique for natural orifice eFTR. The acute model was conducted on female pigs (weight range 45 to 53 kg) under a protocol approved by the Legacy Health Institutional Animal Care and Use Committee. General anesthesia was induced with Telazole (6 to 8 mg/ kg) and atropine (0.06 mg/kg), and endotracheal intubation was performed. Isoflurane (1.5% to 3%) inhalation anesthesia was maintained throughout the operation. Cardiac monitoring, pulse oximetry, end tidal CO2, and blood pressure monitoring ensured a normal physiologic response to the anesthetic agent and CO2 insufflation. The colon was preoperatively cleaned using a tap water enema. After the procedure the animals were euthanized with sodium pentobarbital (80 mg/kg). Devices for eFTR
For all experiments, a standard therapeutic dual-channel endoscope (GIF 2T 160, Olympus) was used. Closure of the colonic wall was performed by an endoscopic clip closure system (ECCS), which was either a prototype clip closure system (ECCS I: closure device: Padlock-G, delivery system: Lock-it, Aponos Medical) or an over-the-scope clip system (ECCS II: OTSC System, Ovesco GmbH). For traction of the intestinal wall as well as specimen removal, T-tag anchoring sutures (Tissue Apposition System [TAS], Ethicon EndoSurgery), delivered by a tissue puncturing
Figure 1. The T-tag suture device consists of a hollow needle to deliver a threaded tag through the tissue. The handle includes the needle exposure mechanism and the T-tag suture release button (right hand thumb).
system, were used (Fig. 1). Resection of the colonic wall specimen was performed using a standard electrocautery snare (Boston Scientific) at cutting mode. An intraabdominal view in the human cadaver model as well as the acute animal model was obtained to observe the resection and was provided by standard laparoscopy (5mm, 30degree endoscope, Stryker Endoscopy). Due to the anatomy of the pig colon, laparoscopic assistance was used in the animal model just for appropriate overview and not for endoscopic resection. Experimental models Model I
The human cadaver model was placed in a Lloyd-Davis position on a standard operating room table. A modified “grasp and snare” technique (n ⫽ 2) was tested within the left colon, and served as control. In this model, for tissue traction, an endoscopic grasper (grasping forceps FG49L-1, Olympus) (and not T-tags) was used with the ECCS I for pre-resection closure. Model II
To initially avoid a large number of animals being tested, the distal 60 cm of fresh porcine colon (n ⫽ 10) were mounted onto a custom-made endoscopy model and eFTR was performed on each colonic specimen 10 cm distal to the proximal end. T-tags were used for tissue traction instead of a grasper, and both clip closure systems (ECCS I: n ⫽ 4; ECCS II: n ⫽ 6) were used for pre-resection closure of the colonic wall. The proximal ends of the colon stumps were closed using an endoscopic stapler. Model III
For the acute model, female swine were placed in a supine position on the operating table on a warm watercirculating blanket. One eFTR was attempted in each
424
Rieder et al
Endoscopic Full-Thickness Colon Resection
J Am Coll Surg
like guidewires to rapidly advance the endoscope to the location of interest. Step IV
Figure 2. Schematic illustration of the endoscopic full-thickness colon resection. (A) The hollow needle with the attached T-tag inside is used to puncture the colon wall at a predetermined distance to the lesion. (B) The T-tag suture is deployed. (C) T-tag sutures are used to draw the colon tissue into the hood of the endoscopic clip closure system, mounted onto the tip of the flexible endoscope. (D) Colon tissue is fully retracted into the hood using the T-tag suture together with suction via the endoscope. The T-tags are located extramurally. (E) Before tissue resection, the clip is applied, creating the pseudo-polyp. (F) The inverted tissue is resected using a standard endoscopic electrocautery snare.
The endoscope, together with the mounted closure device, was placed in front of the marked lesion, and along with the assistant, care was taken to keep the sutures under tension individually. Thereby appropriate extramural placement of the T-type tissue anchors was attempted. The colonic wall was retracted into the plastic hood of the closure clip system by gently pulling the T-tag sutures. Gentle scope suction could also be applied if needed, to start the lesion into the resection cap. The closure clip was fired by an external release mechanism, not before the colon tissue was completely pulled into the hood. The endoscope was then removed again (Fig. 3E). Step V
The electrocautery snare was placed through one channel, and the T-tag sutures were threaded through the open snare and subsequently through the opposite channel. Step VI
swine (n ⫽ 3) at about 40 cm from the anus. T-tags were used to mark the resection margins of the hypothesized lesion on the antimesenteric side and for tissue traction. The ECCS II was used for pre-resection tissue closure. Operative steps of colonic eFTR are described below in detail and are schematically illustrated in Figure 2. Experimental operative steps Step I
During standard colonoscopy, a location was predetermined and a “lesion” (less than 2 cm) was marked using electrocautery. Up to 4 T-tag sutures were applied circumferentially to the hypothesized lesion, marking a predetermined resection margin. Care was taken not to grasp or otherwise manipulate the defined lesion (“no-touch technique”). The endoscope was removed, leaving the T-tag sutures in place (Figs. 3A and B). Step II
The endoscopic clip closure system was mounted onto the tip of the endoscope, and all T-tag sutures were again threaded through one channel using a standard grasper or snare and a pull-through technique (Fig. 3C). Step III
The endoscope along with the closure device was inserted into the colon and directed toward the hypothesized lesion surrounded by T-tags (Fig. 3D). Sutures were used almost
The electrocautery snare (Fig. 3F) was then used to resect the inverted tissue between the clip and the T-tags. To avoid energy transmission through the clip, care was taken to place the snare precisely above the base of the clip, and energy was not applied before the appropriate snare location was ensured. The specimen was then easily removed as the T-tags were still attached to it and could follow the scope during withdrawal. The maximum diameters of the resected colon wall specimens were measured. Air leak pressure test of closure
Air leak pressure tests were performed on 12 of 13 animal experiments (ECSS I, n ⫽ 4; ECCS II, n ⫽ 8). The colon stumps were cut approximately 10 cm below the location of the resection, and the distal ends were tied onto the hose of a sphygmomanometer. Each proximal end was closed with an endoscopic stapler. The colon stumps were submerged in water and subsequently insufflated. Intracolonic pressure to achieve air leak, defined as the first sign of air bubbles from the serosal side, was recorded in mmHg. Air within the colon stump was manually deflated and each test was repeated twice. For control and basic reference value, 6 additional colon stumps were used for an open full-thickness resection (1 per specimen, resected diameter 2.5 cm) and closure was performed with extramucosal single-layer continuous 3-0 absorbable sutures. Air leakage pressure in the control group was assessed as described above.
Vol. 213, No. 3, September 2011
Rieder et al
Endoscopic Full-Thickness Colon Resection
425
Figure 3. Endolumenal view of the endoscopic full-thickness colon resection (live animal model). (A) Endoscopic view of first tissue anchor applied next to a defined lesion with a predetermined margin. (B) Endoscopic view of 4 tissue anchors circumferentially applied to a defined lesion in the porcine colon. Note that the lesion itself is neither touched nor disrupted. (C) With the endoscopic clip closure system (ECCS) II mounted, sutures are pulled through the flexible endoscope. (D) The ECCS is lined up with the colonic lesion, which is subsequently pulled into the hood without grasping the lesion itself (live animal model). (E) After the clip is fired, the inverted colon tissue with the tissue anchors attached can be seen. (F) The endoscopic electrocautery snare is placed around the base of the created pseudo-polyp for tissue resection.
Statistical analysis
Model II
Using the SPSS 17.0 software package, a single analysis of variance (ANOVA) model, was used for comparison of leak pressure between groups and results are given as mean ⫾ standard deviation. A sample size of less than 6 was calculated to be sufficient to detect the observed differences in air leak pressures between groups with alpha ⫽ 0.05 and a power of 0.8.
Our model using porcine colons mounted onto an endoscopy model revealed that T-type suture anchors (3 or 4 circumferentially to a hypothesized lesion) placed in a triangular or rectangular fashion, gave reliable pulling strength. The hypothesized lesion could be completely retracted into the clip closure system in all experiments without injuring the defined lesion itself. The elasticity of the suture provided an excellent tactile feedback during retraction of the colonic tissue. No suture was observed to pull through the colonic wall. Subsequent clip application was feasible in all (10 of 10) specimens. The snare-resected specimens could be easily removed from the colon stump with the sutures still in place without further manipulation. Mean specimen diameter was 27 ⫾ 3 mm. Procedure time was 19 ⫾ 5 minutes (Fig. 4).
RESULTS Endoscopic full-thickness resection Model I
In the human cadaver model, it was observed that the required tension for appropriate full-thickness retraction into the cap was not achieved with an endoscopic grasper. Full-thickness resection was achieved, but resulted only in small specimens. The mean diameter of the resected colon wall (n ⫽ 2) was only 7 ⫾ 1 mm. Procedure time was 24 ⫾ 4 minutes.
Model III
Natural orifice eFTRs were performed in an acute animal model (n ⫽ 3) with a laparoscopic overview. After successful T-tag placement at the antimesenteric side (3 of 3) and clip
426
Rieder et al
Endoscopic Full-Thickness Colon Resection
J Am Coll Surg
served at a mean of 10 ⫾ 6 mmHg using the prototype ECCS I and at 61 ⫾ 5 mmHg using the ECCS II (p ⬍ 0.01). The hand sewn full-thickness resection revealed a mean air pressure resistance of 26 ⫾ 7 mmHg, which was significantly lower than that achieved by the ECCS II closure (p ⬍ 0.01).
Figure 4. (A) Area of endoscopic full-thickness colon resection and the endoscopic clip closure system (ECCS) I placed for pre-resection closure (ex vivo model). (B) Excised full-thickness colon wall specimen with 4 T-tag sutures still in place around the hypothesized lesion (ex vivo model). Note that the hypothesized lesion is neither disrupted nor perforated.
closure before the resection (2 of 3), full-thickness specimens of 22 ⫾ 1 mm (not pinned out) were harvested in 2 of the 3 animals. No bleeding was observed. Appropriate bowel preparation was not possible in the third animal. Due to severe fecal congestion, the clip could not be applied after pulling the colon tissue into the application cap. Because further attempts to clear the porcine colon were not effective, the intervention had to be abandoned. Procedure time (n ⫽ 2) was 33 ⫾ 4 minutes. No T-tag suture was observed to rip through the tissue and no technical problems with the T-tag sutures during excision were observed. No energy transmission through the clip during amputation of the specimen was observed. Macroscopically, the edges of the specimen appeared to be clearly cut with the T-tag sutures in place. Air leak pressure of closure
After pre-resection closure and subsequent tissue snare resection of the porcine colon tissue, air leaks were ob-
DISCUSSION It is widely recognized that the ability to perform reliable full-thickness resection and secure closure of the gastrointestinal (GI) tract would be a desirable adjunct to endoscopic GI treatment. The ideal form of this resection would be inexpensive, quick, applicable to the full colon and perhaps other areas, would provide adequate resection size, not injure external organs, provide a secure non-narrowing closure, avoid peritoneal contamination, and adhere to established oncologic principles of GI surgery. This feasibility study demonstrates that a new technique for eFTR, using T-tag anchoring sutures with an endoscopic clip closure system, appears to accomplish most of these goals. Our approach uses 3 new devices basically designed to close enterotomies as part of natural orifice translumenal endoscopic surgery (NOTES) procedures. Two are large, “over-the-scope” clips (Padlock-G and OTSC) and the third is part of a closure system using T-tags (TAS). The use of suture anchors applied around a hypothesized colonic wall lesion, instead of grasping or penetrating the tumor,9 respects several surgical principles and also appears to be more accurate. This is an important aspect when dealing with potentially dysplastic tissue. It has also been shown that more than 25% of colorectal neoplasms with benign histology on initial biopsy and unresectable with standard techniques might harbor invasive adenocarcinoma.10 Although hypothetical, puncturing a cancerous polyp could also carry the risk of tumor cell exfoliation into the peritoneal cavity, converting a stage I cancer to stage IV before surgical resection. The method described within this study allows accurate tissue traction around the circumference of a potential lesion without the need to manipulate or penetrate the tumor tissue itself. Additionally, the T-tags can be applied at a predetermined distance to the tumor before retraction into the clip application cap. So clear tumor margins can be predetermined before tissue retraction and clip application. It is commonly accepted that 2 mm of clear margins is adequate for a complete excision of early cancerous lesions.11 Although several techniques and devices for endoscopic gastrointestinal full-thickness resection have been described, none satisfies basic technical requirements so
Vol. 213, No. 3, September 2011
far. The first endolumenal system for transmural resection of colorectal tissue was described in 2001.5,12 Unfortunately, device length and size limited its use to the lower left side of the colon. Another study described the use of an endoscopic grasper to retract the colonic wall through a polypectomy snare with subsequent resection, which, in contrast to a suction-resection technique, resulted in no injury to adjacent organs, or mesentery.13 Unfortunately, this study was not designed to close the defect after resection. Based on these early reports in the pig model, we attempted to prove the principles of a modified “grasp-and-snare” technique within a human cadaver model, but adding a simultaneous closure with a clip device designed. However, it was observed that even an aggressive endoscopic grasper could not sufficiently grasp and simultaneously pull the colonic wall into the clip application cap. Furthermore, it is obvious that any grasper used forcefully also could tear tumor tissue, leading to less accurate histologic analysis or a non en-bloc resection. Therefore this method was abandoned. Although it has been previously observed that entry into the peritoneum during transanal endoscopic microsurgery excision does not necessarily lead to intraabdominal sepsis or abscess and can be managed using endoscopy,14 we consider intestinal wall closure before resection to be an important safety feature of eFTR. Additionally, free perforation into the peritoneal cavity results in decreased endoluminal visibility due to collapse of the colon lumen. Therefore, any technique that calls for excision and then secondary closure appears to be concerning.15 A recent study using standard endoscopic mucosal clips to close the colon wall after fullthickness resection found closure difficult and results underwhelming because less than half of excisions could be closed with the first attempt.9 Recent reports of the experimental and clinical use of an over-the-scope clip system for endoscopic closure of colonic perforation showed favorable results.16,17 It has been observed that resection techniques using blind traction can result in incorporation of adjacent structures or bowel loops,18 which is certainly a concern regarding this approach. Using T-tags for tissue closure, others have observed only 2 inadvertently placed T-tags without complications in a large series, which was related to a refined delivery technique. It was noted that the exposed needle was withdrawn by 0.5 cm after puncturing the colon wall and before release of the T-tag.8 We also found the smallest needle exposure possible along with a desufflated colon to be sufficient for barely puncturing the colon wall. Paying attention to slowly deploying the T-tag, rather than quickly punching down the
Rieder et al
Endoscopic Full-Thickness Colon Resection
427
release button, might add another important safety feature when using T-type tissue anchors. Additionally, emerging tools such as endoscopic helical graspers could eventually overcome some of these issues. However, the feasibility of a purely endoscopic approach has to be addressed in further studies, and later clinical studies should rather be performed under laparoscopic assistance to maximize safety. Favorable clinical data on the use of T-tag sutures in the colon for closure under laparoscopic control were presented recently.19 After resection, the ECCS II closure appeared to be secure because it revealed higher air leak pressure resistance compared with the hand-sutured closure used in this study. Colon closure by surgical staplers would have been another method in surgery; this has been shown to be as secure as open suture closure.20,21 Although the optimal in vivo leak pressure is unknown, a distension pressure of 25 cm saline (18 mmHg) for intraoperative integrity testing of colorectal anastomosis has been suggested.22 Our demonstrated ECCS II closure test of 60 mmHg would therefore appear to be more than sufficient, which has also been confirmed in a published survival animal model.23 These results are also comparable to colotomy closure leak pressures achieved with a prototype endoscopic stapler.21 Whether the prototype ECCS I closure would observe higher leak pressure resistance in the human colon has to be evaluated. Earlier clinical case series have already described the use of the ECCS II for treatment of colon perforations and hemostasis throughout the colon; they have reported that the majority of clips could not be found after 3 weeks. On the other hand, re-examinations within only 4 days found the clip in place, without any signs of inflammation or necrosis.24 This indicates that the clip itself gets passed with the stool in the long-term. In case of a lesion not fully resected, an additional endolumenal resection attempt could therefore become possible within a few weeks. Our study does have some weaknesses; our model did not have an actual mucosal lesion, just a hypothetical one. Actual lesions may occupy more volume or prove more difficult to pull into the resection cap. It is hoped that a larger clip applicator with this specific use in mind will be able to compensate for this. In this study the ECCSs mounted onto the endoscope had a maximum inner diameter of 11 mm, which, along with the thickness of the colon wall, limits the amount of tissue that can be retracted into the application cap and therefore the size of the resected full-thickness tissue. The live animal model resulted in a specimen diameter of 22 mm with the T-tags still in place. This would allow the lesion
428
Rieder et al
Endoscopic Full-Thickness Colon Resection
to be about 18 mm with clear margins of 2 mm when the small clip size is used. Because the specimens were not pinned out, this diameter might have been slightly underestimated. A plastic cap with a larger dimension might make larger specimens feasible. We do not know so far whether resection of larger specimens will result in narrowing of the colon lumen, and this will also be the aim of further studies before initial clinical trials. Additionally, eFTR on the mesenteric side of the colon in particular has to be evaluated regarding potential difficulties, and the response of the surrounding tissue, such as bleeding or ischemia, has to be documented. However, endoscopic full-thickness colon wall resection would especially be an appealing option to overcome invasive surgery in patients with incompletely resected polyps tethered to the colon from scar tissue. Furthermore, a full-thickness specimen might allow superior pathologic evaluation, which, in particular, could be useful in the management of early submucosal cancer for a more patient- tailored approach or for patients not suitable for surgery. Although still theoretical, the combination of translumenal sentinel lymph node staging along with endolumenal submucosal resection has already been described in an animal model.25
CONCLUSIONS In conclusion, we demonstrate within this experimental pilot study that a novel approach for colon wall eFTR is feasible using fairly simple tools designed for natural orifice translumenal endoscopic surgery. Using T-type tissue anchors for both marking of resection margins and tissue traction, and over-the-scope clip (ECCS) designs, we were able to reliably draw the lesion and defined margins into a clip closure system, preclose the colon wall, and then resect beyond the T-tags with a standard cautery snare. Several technical shortcomings of previously demonstrated eFTR methods were addressed, and closure was demonstrated to be secure. Although there are certainly needs for further refinements of the devices used in order to increase the size and volume of lesions that can be resected, as well as additional experimental studies to confirm the technique’s safety, it does seem to be a promising natural orifice endoscopic intervention, which, in the near future could help to overcome the need for invasive surgery in selected patients. Author Contributions
Study conception and design: Rieder, Martinec, Dunst, Swanström Acquisition of data: Rieder, Martinec, Swanström
J Am Coll Surg
Analysis and interpretation of data: Rieder, Martinec, Dunst, Swanström Drafting of manuscript: Rieder, Martinec, Swanström Critical revision: Rieder, Martinec, Dunst, Swanström
REFERENCES 1. American Cancer Society: Cancer Facts and Figures 2010. Atlanta GACS, 2010. Available at: http://www.cancer.org. Accessed June 8, 2011. 2. Winawer SJ, Zauber AG, Ho MN, et al. Prevention of colorectal cancer by colonoscopic polypectomy. The National Polyp Study Workgroup. N Engl J Med 1993;329:1977–1981. 3. Ramirez M, Schierling S, Papaconstantinou HT, Scott Thomas J. Management of the malignant polyp. Clin Colon Rectal Surg 2008;21:286–290. 4. Tanaka S, Oka S, Chayama K. Colorectal endoscopic submucosal dissection: present status and future perspective, including its differentiation from endoscopic mucosal resection. J Gastroenterol 2008;43:641–651. 5. Rajan E, Gostout CJ, Burgart LJ, et al. First endoluminal system for transmural resection of colorectal tissue with a prototype full-thickness resection device in a porcine model. Gastrointest Endosc 2002;55:915–920. 6. Fritscher-Ravens A, Cuming T, Jacobsen B, et al. Feasibility and safety of endoscopic full- thickness esophageal wall resection and defect closure: a prospective long-term survival animal study. Gastrointest Endosc 2009;69:1314–1320. 7. Elmunzer BJ, Trunzo JA, Marks JM, et al. Endoscopic fullthickness resection of gastric tumors using a novel grasp-andsnare technique: feasibility in ex vivo and in vivo porcine models. Endoscopy 2008;40:931–935. 8. Raju GS, Malhotra A, Ahmed I. Colonoscopic full-thickness resection of the colon in a porcine model as a prelude to endoscopic surgery of difficult colon polyps: a novel technique (with videos). Gastrointest Endosc 2009;70:159–165. 9. von Renteln D, Schmidt A, Vassiliou MC, et al. Endoscopic full-thickness resection and defect closure in the colon. Gastrointest Endosc 2010;71:1267–1273. 10. Brozovich M, Read TE, Salgado J, et al. Laparoscopic colectomy for apparently benign colorectal neoplasia: A word of caution. Surg Endosc 2008;22:506–509. 11. Cooper HS, Deppisch LM, Gourley WK, et al. Endoscopically removed malignant colorectal polyps: clinicopathologic correlations. Gastroenterology 1995;108:1657–1665. 12. Schurr MO, Raestrup H, Arezzo A, et al. Full thickness resection device (FTRD) for endoluminal removal of large bowel tumours: development of the instrument and related experimental studies. Min Invas Ther & Allied Technol 2001;10:9. 13. Ahmed I, Shibukawa G, Groce R, et al. Study of full-thickness endoluminal segmental resection of colon in a porcine colon model (with videos). Gastrointest Endosc 2007;65:696–702. 14. Gavagan JA, Whiteford MH, Swanström LL. Full-thickness intraperitoneal excision by transanal endoscopic microsurgery does not increase short-term complications. Am J Surg 2004; 187:630–634. 15. Raju GS, Ahmed I, Shibukawa G, et al. Endoluminal clip closure of a circular full-thickness colon resection in a porcine model (with videos). Gastrointest Endosc 2007;65:503–509. 16. Parodi A,Repici A, Pedroni A, et al. Endoscopic management of
Vol. 213, No. 3, September 2011
17. 18. 19. 20. 21.
GI perforations with a new over-the-scope clip device. Gastrointest Endosc 2010;72:881–886. Guarner-Argente C, Cordova H, Martinez-Palli G, et al. Yes, we can: reliable colonic closure with the Padlock-G clip in a survival porcine study. Gastrointest Endosc 2010;72:841–844. von Renteln D, Schmidt A, Vassiliou MC, et al. Endoscopic closure of large colonic perforations using an over-the-scope clip: a randomized controlled porcine study. Endoscopy 2009;41:481–486. Delaney CP, Champagne BJ, Marks JM, et al. Tissue apposition system: new technology to minimize surgery for endoscopically unresectable colonic polyps. Surg Endosc 2010;24:3113–3118. Lustosa SA, Matos D, Atallah AN, Castro AA. Stapled versus handsewn methods for colorectal anastomosis surgery. Cochrane Database Syst Rev 2001:CD003144. Ryou M, Fong DG, Pai RD, et al. Transluminal closure for NOTES: an ex vivo study comparing leak pressures of various gastrotomy and colotomy closure modalities. Endoscopy 2008;40:432–436.
Rieder et al
Endoscopic Full-Thickness Colon Resection
429
22. Gilbert JM, Trapnell JE. Intraoperative testing of the integrity of left-sided colorectal anastomoses: a technique of value to the surgeon in training. Ann R Coll Surg Engl 1988;70: 158–160. 23. Schurr MO, Hartmann C, Ho CN, et al. An over-the-scope clip (OTSC) system for closure of iatrogenic colon perforations: results of an experimental survival study in pigs. Endoscopy 2008; 40:584–588. 24. Kirschniak A, Kratt T, Stüker D, et al. A new endoscopic overthe-scope clip system for treatment of lesions and bleeding in the GI tract: first clinical experience. Gastrointest Endosc 2007;66: 162–167. 25. Cahill RA, Askuma M, Perretta S, et al. Supplementation of endoscopic submucosal dissection with sentinel node biopsy performed by natural orifice transluminal endoscopic surgery (NOTES). Gastrointest Endosc 2009;69:1152–1160.
be readily apparent. Current techniques, such as endoscopic piecemeal resection, are controversial due to high recurrence rates or inability to assess the resection margin.3 More recent methods, such as endoscopic submucosal dissection in the colon, are not yet widely used because of their difficulty to master.4 Consequently, surgery was and still is the basic treatment for certain malignant or “difficult” polyps. An endoscopic en-bloc full-thickness resection (eFTR) of the colonic wall together with an easy and reliable method of closure could extend the indication of appropriate endoscopic treatment. Additionally, eFTR would be a tremendous benefit for selected patients by avoiding a large organ resection. So far, several approaches or devices for endoscopic gastrointestinal wall resection have been described with variable success,5-7 however, most would be limited in a clinical setting due to technical reasons or difficulty with providing a reliable closure. Others have described more sophisticated procedures involving resection and secondary closure of the iatrogenic colon perfora-
Colorectal cancer remains among the leading causes of cancer deaths in the United States.1 Colonoscopic screening and endoscopic snare resection of colonic polyps have become standard techniques to reduce the incidence of colorectal carcinoma;2 polyp excision, however, may be challenging for several reasons. In addition, a considerable number of polyps harbor early malignancies that may not Disclosure Information: Dr Swanström receives honoraria for his work as a consultant for Ethicon Endo-Surgery. Ethicon Endo-Surgery, Cincinnati, OH provided the TAS, and Aponos Medical, Kingston, NH as well as OVESCO Endoscopy AG, Tübingen, Germany provided the clip closure systems. The companies were not involved in study design or data acquisition and interpretation. All other authors have nothing to disclose. Received January 24, 2011; Revised April 18, 2011; Accepted May 4, 2011. From the Minimally Invasive Surgery Program, Legacy Health (Rieder, Swanström), and Gastrointestinal and Minimally Invasive Surgery Division, The Oregon Clinic (Martinec, Dunst, Swanström), Portland, OR. Correspondence address: Erwin Rieder, MD, Minimally Invasive Surgery Program, Legacy Health, 1040 NW 22nd Ave, Suite 560, Portland, OR 97210. email: [email protected]
© 2011 by the American College of Surgeons Published by Elsevier Inc.
422
ISSN 1072-7515/11/$36.00 doi:10.1016/j.jamcollsurg.2011.05.015
Vol. 213, No. 3, September 2011
Rieder et al
Endoscopic Full-Thickness Colon Resection
423
Abbreviations and Acronyms
ECCS ⫽ endoscopic clip closure system eFTR ⫽ endoscopic full-thickness colon resection GI ⫽ gastrointestinal
tion.8 This would expose the patient to risk of fecal contamination and subsequent septic complications as well as the possibility of tumor spillage. The aim of this study was to develop a straightforward method for eFTR addressing current technical shortcomings. The novel eFTR technique that we have developed within this feasibility study for the first time appears to meet fundamental surgical principles. These include en-bloc tissue resection with defined tumor margins, eliminating the need to grasp the lesion itself. Simultaneously, any temporarily colonic wall perforation or mandatory secondary closure is avoided.
METHODS Inanimate and animate models
Freshly explanted distal porcine colons (up to 60 cm from the anus) obtained from a local slaughterhouse and used within 3 hours of excision, an acute porcine animal model, and a human cadaver were used for development and evaluation of a novel technique for natural orifice eFTR. The acute model was conducted on female pigs (weight range 45 to 53 kg) under a protocol approved by the Legacy Health Institutional Animal Care and Use Committee. General anesthesia was induced with Telazole (6 to 8 mg/ kg) and atropine (0.06 mg/kg), and endotracheal intubation was performed. Isoflurane (1.5% to 3%) inhalation anesthesia was maintained throughout the operation. Cardiac monitoring, pulse oximetry, end tidal CO2, and blood pressure monitoring ensured a normal physiologic response to the anesthetic agent and CO2 insufflation. The colon was preoperatively cleaned using a tap water enema. After the procedure the animals were euthanized with sodium pentobarbital (80 mg/kg). Devices for eFTR
For all experiments, a standard therapeutic dual-channel endoscope (GIF 2T 160, Olympus) was used. Closure of the colonic wall was performed by an endoscopic clip closure system (ECCS), which was either a prototype clip closure system (ECCS I: closure device: Padlock-G, delivery system: Lock-it, Aponos Medical) or an over-the-scope clip system (ECCS II: OTSC System, Ovesco GmbH). For traction of the intestinal wall as well as specimen removal, T-tag anchoring sutures (Tissue Apposition System [TAS], Ethicon EndoSurgery), delivered by a tissue puncturing
Figure 1. The T-tag suture device consists of a hollow needle to deliver a threaded tag through the tissue. The handle includes the needle exposure mechanism and the T-tag suture release button (right hand thumb).
system, were used (Fig. 1). Resection of the colonic wall specimen was performed using a standard electrocautery snare (Boston Scientific) at cutting mode. An intraabdominal view in the human cadaver model as well as the acute animal model was obtained to observe the resection and was provided by standard laparoscopy (5mm, 30degree endoscope, Stryker Endoscopy). Due to the anatomy of the pig colon, laparoscopic assistance was used in the animal model just for appropriate overview and not for endoscopic resection. Experimental models Model I
The human cadaver model was placed in a Lloyd-Davis position on a standard operating room table. A modified “grasp and snare” technique (n ⫽ 2) was tested within the left colon, and served as control. In this model, for tissue traction, an endoscopic grasper (grasping forceps FG49L-1, Olympus) (and not T-tags) was used with the ECCS I for pre-resection closure. Model II
To initially avoid a large number of animals being tested, the distal 60 cm of fresh porcine colon (n ⫽ 10) were mounted onto a custom-made endoscopy model and eFTR was performed on each colonic specimen 10 cm distal to the proximal end. T-tags were used for tissue traction instead of a grasper, and both clip closure systems (ECCS I: n ⫽ 4; ECCS II: n ⫽ 6) were used for pre-resection closure of the colonic wall. The proximal ends of the colon stumps were closed using an endoscopic stapler. Model III
For the acute model, female swine were placed in a supine position on the operating table on a warm watercirculating blanket. One eFTR was attempted in each
424
Rieder et al
Endoscopic Full-Thickness Colon Resection
J Am Coll Surg
like guidewires to rapidly advance the endoscope to the location of interest. Step IV
Figure 2. Schematic illustration of the endoscopic full-thickness colon resection. (A) The hollow needle with the attached T-tag inside is used to puncture the colon wall at a predetermined distance to the lesion. (B) The T-tag suture is deployed. (C) T-tag sutures are used to draw the colon tissue into the hood of the endoscopic clip closure system, mounted onto the tip of the flexible endoscope. (D) Colon tissue is fully retracted into the hood using the T-tag suture together with suction via the endoscope. The T-tags are located extramurally. (E) Before tissue resection, the clip is applied, creating the pseudo-polyp. (F) The inverted tissue is resected using a standard endoscopic electrocautery snare.
The endoscope, together with the mounted closure device, was placed in front of the marked lesion, and along with the assistant, care was taken to keep the sutures under tension individually. Thereby appropriate extramural placement of the T-type tissue anchors was attempted. The colonic wall was retracted into the plastic hood of the closure clip system by gently pulling the T-tag sutures. Gentle scope suction could also be applied if needed, to start the lesion into the resection cap. The closure clip was fired by an external release mechanism, not before the colon tissue was completely pulled into the hood. The endoscope was then removed again (Fig. 3E). Step V
The electrocautery snare was placed through one channel, and the T-tag sutures were threaded through the open snare and subsequently through the opposite channel. Step VI
swine (n ⫽ 3) at about 40 cm from the anus. T-tags were used to mark the resection margins of the hypothesized lesion on the antimesenteric side and for tissue traction. The ECCS II was used for pre-resection tissue closure. Operative steps of colonic eFTR are described below in detail and are schematically illustrated in Figure 2. Experimental operative steps Step I
During standard colonoscopy, a location was predetermined and a “lesion” (less than 2 cm) was marked using electrocautery. Up to 4 T-tag sutures were applied circumferentially to the hypothesized lesion, marking a predetermined resection margin. Care was taken not to grasp or otherwise manipulate the defined lesion (“no-touch technique”). The endoscope was removed, leaving the T-tag sutures in place (Figs. 3A and B). Step II
The endoscopic clip closure system was mounted onto the tip of the endoscope, and all T-tag sutures were again threaded through one channel using a standard grasper or snare and a pull-through technique (Fig. 3C). Step III
The endoscope along with the closure device was inserted into the colon and directed toward the hypothesized lesion surrounded by T-tags (Fig. 3D). Sutures were used almost
The electrocautery snare (Fig. 3F) was then used to resect the inverted tissue between the clip and the T-tags. To avoid energy transmission through the clip, care was taken to place the snare precisely above the base of the clip, and energy was not applied before the appropriate snare location was ensured. The specimen was then easily removed as the T-tags were still attached to it and could follow the scope during withdrawal. The maximum diameters of the resected colon wall specimens were measured. Air leak pressure test of closure
Air leak pressure tests were performed on 12 of 13 animal experiments (ECSS I, n ⫽ 4; ECCS II, n ⫽ 8). The colon stumps were cut approximately 10 cm below the location of the resection, and the distal ends were tied onto the hose of a sphygmomanometer. Each proximal end was closed with an endoscopic stapler. The colon stumps were submerged in water and subsequently insufflated. Intracolonic pressure to achieve air leak, defined as the first sign of air bubbles from the serosal side, was recorded in mmHg. Air within the colon stump was manually deflated and each test was repeated twice. For control and basic reference value, 6 additional colon stumps were used for an open full-thickness resection (1 per specimen, resected diameter 2.5 cm) and closure was performed with extramucosal single-layer continuous 3-0 absorbable sutures. Air leakage pressure in the control group was assessed as described above.
Vol. 213, No. 3, September 2011
Rieder et al
Endoscopic Full-Thickness Colon Resection
425
Figure 3. Endolumenal view of the endoscopic full-thickness colon resection (live animal model). (A) Endoscopic view of first tissue anchor applied next to a defined lesion with a predetermined margin. (B) Endoscopic view of 4 tissue anchors circumferentially applied to a defined lesion in the porcine colon. Note that the lesion itself is neither touched nor disrupted. (C) With the endoscopic clip closure system (ECCS) II mounted, sutures are pulled through the flexible endoscope. (D) The ECCS is lined up with the colonic lesion, which is subsequently pulled into the hood without grasping the lesion itself (live animal model). (E) After the clip is fired, the inverted colon tissue with the tissue anchors attached can be seen. (F) The endoscopic electrocautery snare is placed around the base of the created pseudo-polyp for tissue resection.
Statistical analysis
Model II
Using the SPSS 17.0 software package, a single analysis of variance (ANOVA) model, was used for comparison of leak pressure between groups and results are given as mean ⫾ standard deviation. A sample size of less than 6 was calculated to be sufficient to detect the observed differences in air leak pressures between groups with alpha ⫽ 0.05 and a power of 0.8.
Our model using porcine colons mounted onto an endoscopy model revealed that T-type suture anchors (3 or 4 circumferentially to a hypothesized lesion) placed in a triangular or rectangular fashion, gave reliable pulling strength. The hypothesized lesion could be completely retracted into the clip closure system in all experiments without injuring the defined lesion itself. The elasticity of the suture provided an excellent tactile feedback during retraction of the colonic tissue. No suture was observed to pull through the colonic wall. Subsequent clip application was feasible in all (10 of 10) specimens. The snare-resected specimens could be easily removed from the colon stump with the sutures still in place without further manipulation. Mean specimen diameter was 27 ⫾ 3 mm. Procedure time was 19 ⫾ 5 minutes (Fig. 4).
RESULTS Endoscopic full-thickness resection Model I
In the human cadaver model, it was observed that the required tension for appropriate full-thickness retraction into the cap was not achieved with an endoscopic grasper. Full-thickness resection was achieved, but resulted only in small specimens. The mean diameter of the resected colon wall (n ⫽ 2) was only 7 ⫾ 1 mm. Procedure time was 24 ⫾ 4 minutes.
Model III
Natural orifice eFTRs were performed in an acute animal model (n ⫽ 3) with a laparoscopic overview. After successful T-tag placement at the antimesenteric side (3 of 3) and clip
426
Rieder et al
Endoscopic Full-Thickness Colon Resection
J Am Coll Surg
served at a mean of 10 ⫾ 6 mmHg using the prototype ECCS I and at 61 ⫾ 5 mmHg using the ECCS II (p ⬍ 0.01). The hand sewn full-thickness resection revealed a mean air pressure resistance of 26 ⫾ 7 mmHg, which was significantly lower than that achieved by the ECCS II closure (p ⬍ 0.01).
Figure 4. (A) Area of endoscopic full-thickness colon resection and the endoscopic clip closure system (ECCS) I placed for pre-resection closure (ex vivo model). (B) Excised full-thickness colon wall specimen with 4 T-tag sutures still in place around the hypothesized lesion (ex vivo model). Note that the hypothesized lesion is neither disrupted nor perforated.
closure before the resection (2 of 3), full-thickness specimens of 22 ⫾ 1 mm (not pinned out) were harvested in 2 of the 3 animals. No bleeding was observed. Appropriate bowel preparation was not possible in the third animal. Due to severe fecal congestion, the clip could not be applied after pulling the colon tissue into the application cap. Because further attempts to clear the porcine colon were not effective, the intervention had to be abandoned. Procedure time (n ⫽ 2) was 33 ⫾ 4 minutes. No T-tag suture was observed to rip through the tissue and no technical problems with the T-tag sutures during excision were observed. No energy transmission through the clip during amputation of the specimen was observed. Macroscopically, the edges of the specimen appeared to be clearly cut with the T-tag sutures in place. Air leak pressure of closure
After pre-resection closure and subsequent tissue snare resection of the porcine colon tissue, air leaks were ob-
DISCUSSION It is widely recognized that the ability to perform reliable full-thickness resection and secure closure of the gastrointestinal (GI) tract would be a desirable adjunct to endoscopic GI treatment. The ideal form of this resection would be inexpensive, quick, applicable to the full colon and perhaps other areas, would provide adequate resection size, not injure external organs, provide a secure non-narrowing closure, avoid peritoneal contamination, and adhere to established oncologic principles of GI surgery. This feasibility study demonstrates that a new technique for eFTR, using T-tag anchoring sutures with an endoscopic clip closure system, appears to accomplish most of these goals. Our approach uses 3 new devices basically designed to close enterotomies as part of natural orifice translumenal endoscopic surgery (NOTES) procedures. Two are large, “over-the-scope” clips (Padlock-G and OTSC) and the third is part of a closure system using T-tags (TAS). The use of suture anchors applied around a hypothesized colonic wall lesion, instead of grasping or penetrating the tumor,9 respects several surgical principles and also appears to be more accurate. This is an important aspect when dealing with potentially dysplastic tissue. It has also been shown that more than 25% of colorectal neoplasms with benign histology on initial biopsy and unresectable with standard techniques might harbor invasive adenocarcinoma.10 Although hypothetical, puncturing a cancerous polyp could also carry the risk of tumor cell exfoliation into the peritoneal cavity, converting a stage I cancer to stage IV before surgical resection. The method described within this study allows accurate tissue traction around the circumference of a potential lesion without the need to manipulate or penetrate the tumor tissue itself. Additionally, the T-tags can be applied at a predetermined distance to the tumor before retraction into the clip application cap. So clear tumor margins can be predetermined before tissue retraction and clip application. It is commonly accepted that 2 mm of clear margins is adequate for a complete excision of early cancerous lesions.11 Although several techniques and devices for endoscopic gastrointestinal full-thickness resection have been described, none satisfies basic technical requirements so
Vol. 213, No. 3, September 2011
far. The first endolumenal system for transmural resection of colorectal tissue was described in 2001.5,12 Unfortunately, device length and size limited its use to the lower left side of the colon. Another study described the use of an endoscopic grasper to retract the colonic wall through a polypectomy snare with subsequent resection, which, in contrast to a suction-resection technique, resulted in no injury to adjacent organs, or mesentery.13 Unfortunately, this study was not designed to close the defect after resection. Based on these early reports in the pig model, we attempted to prove the principles of a modified “grasp-and-snare” technique within a human cadaver model, but adding a simultaneous closure with a clip device designed. However, it was observed that even an aggressive endoscopic grasper could not sufficiently grasp and simultaneously pull the colonic wall into the clip application cap. Furthermore, it is obvious that any grasper used forcefully also could tear tumor tissue, leading to less accurate histologic analysis or a non en-bloc resection. Therefore this method was abandoned. Although it has been previously observed that entry into the peritoneum during transanal endoscopic microsurgery excision does not necessarily lead to intraabdominal sepsis or abscess and can be managed using endoscopy,14 we consider intestinal wall closure before resection to be an important safety feature of eFTR. Additionally, free perforation into the peritoneal cavity results in decreased endoluminal visibility due to collapse of the colon lumen. Therefore, any technique that calls for excision and then secondary closure appears to be concerning.15 A recent study using standard endoscopic mucosal clips to close the colon wall after fullthickness resection found closure difficult and results underwhelming because less than half of excisions could be closed with the first attempt.9 Recent reports of the experimental and clinical use of an over-the-scope clip system for endoscopic closure of colonic perforation showed favorable results.16,17 It has been observed that resection techniques using blind traction can result in incorporation of adjacent structures or bowel loops,18 which is certainly a concern regarding this approach. Using T-tags for tissue closure, others have observed only 2 inadvertently placed T-tags without complications in a large series, which was related to a refined delivery technique. It was noted that the exposed needle was withdrawn by 0.5 cm after puncturing the colon wall and before release of the T-tag.8 We also found the smallest needle exposure possible along with a desufflated colon to be sufficient for barely puncturing the colon wall. Paying attention to slowly deploying the T-tag, rather than quickly punching down the
Rieder et al
Endoscopic Full-Thickness Colon Resection
427
release button, might add another important safety feature when using T-type tissue anchors. Additionally, emerging tools such as endoscopic helical graspers could eventually overcome some of these issues. However, the feasibility of a purely endoscopic approach has to be addressed in further studies, and later clinical studies should rather be performed under laparoscopic assistance to maximize safety. Favorable clinical data on the use of T-tag sutures in the colon for closure under laparoscopic control were presented recently.19 After resection, the ECCS II closure appeared to be secure because it revealed higher air leak pressure resistance compared with the hand-sutured closure used in this study. Colon closure by surgical staplers would have been another method in surgery; this has been shown to be as secure as open suture closure.20,21 Although the optimal in vivo leak pressure is unknown, a distension pressure of 25 cm saline (18 mmHg) for intraoperative integrity testing of colorectal anastomosis has been suggested.22 Our demonstrated ECCS II closure test of 60 mmHg would therefore appear to be more than sufficient, which has also been confirmed in a published survival animal model.23 These results are also comparable to colotomy closure leak pressures achieved with a prototype endoscopic stapler.21 Whether the prototype ECCS I closure would observe higher leak pressure resistance in the human colon has to be evaluated. Earlier clinical case series have already described the use of the ECCS II for treatment of colon perforations and hemostasis throughout the colon; they have reported that the majority of clips could not be found after 3 weeks. On the other hand, re-examinations within only 4 days found the clip in place, without any signs of inflammation or necrosis.24 This indicates that the clip itself gets passed with the stool in the long-term. In case of a lesion not fully resected, an additional endolumenal resection attempt could therefore become possible within a few weeks. Our study does have some weaknesses; our model did not have an actual mucosal lesion, just a hypothetical one. Actual lesions may occupy more volume or prove more difficult to pull into the resection cap. It is hoped that a larger clip applicator with this specific use in mind will be able to compensate for this. In this study the ECCSs mounted onto the endoscope had a maximum inner diameter of 11 mm, which, along with the thickness of the colon wall, limits the amount of tissue that can be retracted into the application cap and therefore the size of the resected full-thickness tissue. The live animal model resulted in a specimen diameter of 22 mm with the T-tags still in place. This would allow the lesion
428
Rieder et al
Endoscopic Full-Thickness Colon Resection
to be about 18 mm with clear margins of 2 mm when the small clip size is used. Because the specimens were not pinned out, this diameter might have been slightly underestimated. A plastic cap with a larger dimension might make larger specimens feasible. We do not know so far whether resection of larger specimens will result in narrowing of the colon lumen, and this will also be the aim of further studies before initial clinical trials. Additionally, eFTR on the mesenteric side of the colon in particular has to be evaluated regarding potential difficulties, and the response of the surrounding tissue, such as bleeding or ischemia, has to be documented. However, endoscopic full-thickness colon wall resection would especially be an appealing option to overcome invasive surgery in patients with incompletely resected polyps tethered to the colon from scar tissue. Furthermore, a full-thickness specimen might allow superior pathologic evaluation, which, in particular, could be useful in the management of early submucosal cancer for a more patient- tailored approach or for patients not suitable for surgery. Although still theoretical, the combination of translumenal sentinel lymph node staging along with endolumenal submucosal resection has already been described in an animal model.25
CONCLUSIONS In conclusion, we demonstrate within this experimental pilot study that a novel approach for colon wall eFTR is feasible using fairly simple tools designed for natural orifice translumenal endoscopic surgery. Using T-type tissue anchors for both marking of resection margins and tissue traction, and over-the-scope clip (ECCS) designs, we were able to reliably draw the lesion and defined margins into a clip closure system, preclose the colon wall, and then resect beyond the T-tags with a standard cautery snare. Several technical shortcomings of previously demonstrated eFTR methods were addressed, and closure was demonstrated to be secure. Although there are certainly needs for further refinements of the devices used in order to increase the size and volume of lesions that can be resected, as well as additional experimental studies to confirm the technique’s safety, it does seem to be a promising natural orifice endoscopic intervention, which, in the near future could help to overcome the need for invasive surgery in selected patients. Author Contributions
Study conception and design: Rieder, Martinec, Dunst, Swanström Acquisition of data: Rieder, Martinec, Swanström
J Am Coll Surg
Analysis and interpretation of data: Rieder, Martinec, Dunst, Swanström Drafting of manuscript: Rieder, Martinec, Swanström Critical revision: Rieder, Martinec, Dunst, Swanström
REFERENCES 1. American Cancer Society: Cancer Facts and Figures 2010. Atlanta GACS, 2010. Available at: http://www.cancer.org. Accessed June 8, 2011. 2. Winawer SJ, Zauber AG, Ho MN, et al. Prevention of colorectal cancer by colonoscopic polypectomy. The National Polyp Study Workgroup. N Engl J Med 1993;329:1977–1981. 3. Ramirez M, Schierling S, Papaconstantinou HT, Scott Thomas J. Management of the malignant polyp. Clin Colon Rectal Surg 2008;21:286–290. 4. Tanaka S, Oka S, Chayama K. Colorectal endoscopic submucosal dissection: present status and future perspective, including its differentiation from endoscopic mucosal resection. J Gastroenterol 2008;43:641–651. 5. Rajan E, Gostout CJ, Burgart LJ, et al. First endoluminal system for transmural resection of colorectal tissue with a prototype full-thickness resection device in a porcine model. Gastrointest Endosc 2002;55:915–920. 6. Fritscher-Ravens A, Cuming T, Jacobsen B, et al. Feasibility and safety of endoscopic full- thickness esophageal wall resection and defect closure: a prospective long-term survival animal study. Gastrointest Endosc 2009;69:1314–1320. 7. Elmunzer BJ, Trunzo JA, Marks JM, et al. Endoscopic fullthickness resection of gastric tumors using a novel grasp-andsnare technique: feasibility in ex vivo and in vivo porcine models. Endoscopy 2008;40:931–935. 8. Raju GS, Malhotra A, Ahmed I. Colonoscopic full-thickness resection of the colon in a porcine model as a prelude to endoscopic surgery of difficult colon polyps: a novel technique (with videos). Gastrointest Endosc 2009;70:159–165. 9. von Renteln D, Schmidt A, Vassiliou MC, et al. Endoscopic full-thickness resection and defect closure in the colon. Gastrointest Endosc 2010;71:1267–1273. 10. Brozovich M, Read TE, Salgado J, et al. Laparoscopic colectomy for apparently benign colorectal neoplasia: A word of caution. Surg Endosc 2008;22:506–509. 11. Cooper HS, Deppisch LM, Gourley WK, et al. Endoscopically removed malignant colorectal polyps: clinicopathologic correlations. Gastroenterology 1995;108:1657–1665. 12. Schurr MO, Raestrup H, Arezzo A, et al. Full thickness resection device (FTRD) for endoluminal removal of large bowel tumours: development of the instrument and related experimental studies. Min Invas Ther & Allied Technol 2001;10:9. 13. Ahmed I, Shibukawa G, Groce R, et al. Study of full-thickness endoluminal segmental resection of colon in a porcine colon model (with videos). Gastrointest Endosc 2007;65:696–702. 14. Gavagan JA, Whiteford MH, Swanström LL. Full-thickness intraperitoneal excision by transanal endoscopic microsurgery does not increase short-term complications. Am J Surg 2004; 187:630–634. 15. Raju GS, Ahmed I, Shibukawa G, et al. Endoluminal clip closure of a circular full-thickness colon resection in a porcine model (with videos). Gastrointest Endosc 2007;65:503–509. 16. Parodi A,Repici A, Pedroni A, et al. Endoscopic management of
Vol. 213, No. 3, September 2011
17. 18. 19. 20. 21.
GI perforations with a new over-the-scope clip device. Gastrointest Endosc 2010;72:881–886. Guarner-Argente C, Cordova H, Martinez-Palli G, et al. Yes, we can: reliable colonic closure with the Padlock-G clip in a survival porcine study. Gastrointest Endosc 2010;72:841–844. von Renteln D, Schmidt A, Vassiliou MC, et al. Endoscopic closure of large colonic perforations using an over-the-scope clip: a randomized controlled porcine study. Endoscopy 2009;41:481–486. Delaney CP, Champagne BJ, Marks JM, et al. Tissue apposition system: new technology to minimize surgery for endoscopically unresectable colonic polyps. Surg Endosc 2010;24:3113–3118. Lustosa SA, Matos D, Atallah AN, Castro AA. Stapled versus handsewn methods for colorectal anastomosis surgery. Cochrane Database Syst Rev 2001:CD003144. Ryou M, Fong DG, Pai RD, et al. Transluminal closure for NOTES: an ex vivo study comparing leak pressures of various gastrotomy and colotomy closure modalities. Endoscopy 2008;40:432–436.
Rieder et al
Endoscopic Full-Thickness Colon Resection
429
22. Gilbert JM, Trapnell JE. Intraoperative testing of the integrity of left-sided colorectal anastomoses: a technique of value to the surgeon in training. Ann R Coll Surg Engl 1988;70: 158–160. 23. Schurr MO, Hartmann C, Ho CN, et al. An over-the-scope clip (OTSC) system for closure of iatrogenic colon perforations: results of an experimental survival study in pigs. Endoscopy 2008; 40:584–588. 24. Kirschniak A, Kratt T, Stüker D, et al. A new endoscopic overthe-scope clip system for treatment of lesions and bleeding in the GI tract: first clinical experience. Gastrointest Endosc 2007;66: 162–167. 25. Cahill RA, Askuma M, Perretta S, et al. Supplementation of endoscopic submucosal dissection with sentinel node biopsy performed by natural orifice transluminal endoscopic surgery (NOTES). Gastrointest Endosc 2009;69:1152–1160.