Endoscopic Full Thickness Resection (EFTR) of Colonic Lesions

Title:

Endoscopic full thickness resection (EFTR) of colonic lesions

Authors:

Andrew Currie, MRCS1

RacheleTarquini, MD1

Adela Brigic, MRCS, MD(Res)1

Robin H. Kennedy, FRCS, MS1,2

1Department of Surgery,St. Mark’s Hospital and Academic Institute, Harrow, Middlesex, UK

2Department of Surgery and Cancer, Imperial College London, London, UK.

Corresponding Author
Professor Robin Kennedy
St. Mark’s Hospital and Academic Institute
Watford Road
London, HA1 3UJ, UK
Email:

Conflict of interest: RHK has received research grants from Ethicon Endosurgery and Olympus. AC has received research funding from the Barcapel Foundation, the Wates Foundation and the SavvasRegas Foundation.Other authors report no funding or conflicts of interest.

Abstract

Background: The introduction of bowel cancer screening programs has led to an increasing incidence of complex colonic polyps and early colonic cancer requiring colectomy. Traditional radical colonic resection risks substantial morbidity and there is a need for alternative approaches. This review summarizes the published methods of colonic endoscopic full-thickness resection (EFTR),examining data on feasibility and safety.Preclinical research reported on three EFTR techniques using endoscopic stapling devices, T-tags or compression device closure for defect closure before or after specimen resection. A total of 103 procedures were performed in 99 porcine models, with an overall success rate of87% (90/103 procedures). The intraoperative complication rate was 19% (19/90 procedures). When bowel closure was performed after resection, rather than before it, it more commonly resulted in failure to close the defect and a high incidence of abnormal findings at post-mortem examination. Clinical experience involved five studies reporting EFTR in 38 patients, with three of these using compression device pre-resection closure and one, post-resection closure. EFTR was completed in 33/38 without assistance. Only three patients had complications reported. Lateral margin clearance was variably reported and complete full-thickness resection was achieved in only 12/17 patients. The technique of EFTR is evolving, with only limited clinical evidence to date, but currently pre-resection closure methods seem advisable. Significant technological challenges remain, including reproducible lateral margin clearance. before colonic EFTR can be recommended.

Key words:

Endoscopy, colonic polyp, minimally-invasive surgery, preclinical model, laparoscopic surgery

Introduction:

The advent of bowel cancer screening programmes internationally has resulted in more early stage colonic cancers being identified, along with an increase in the volume of complex colonic polyps1, 2. To tackle this challenge, advanced polypectomy techniques such as endoscopic mucosal resection (EMR) and endoscopic submucosal dissection (ESD), have been developed to effect complete lesion excision without compromising the integrity of the bowel wall. For larger colonic lesions however, EMR is usually performed piecemeal.This means that accurate histological assessment of the lesion is compromised, and that there is a substantial risk of recurrencewhich requires intensive surveillance and, often, further intervention3, 4. In contrast, ESD provides an en bloc resection of the specimen but complications (mainly perforation and bleeding) are considerably more frequent3, 5. With these challenges in mind, a significant proportion of patients with large colonic polyps, and most patients with early malignancy, are referred for surgery6. However, even when performed laparoscopically and within an optimized enhanced recovery after surgery programme, colectomy carries significant risks of patient morbidity and mortality7, 8.

In the rectum, this challenge has been addressed by full-thickness resection techniques. Transanal Endoscopic MicroSurgery (TEMS) has evolved as an organ-preserving approach for patients with low-risk early rectal cancer and those with significant co-morbidity. Systematic reviews have confirmed comparable survival between TEMS and radical resection for T1 and T2 rectal cancer9. Whilst there is a substantial risk of local recurrence, the procedure is associated with rapid patient recovery and excellent post-operative functional results10. In the colon most polyps and early malignancy will occur outwith the reach of TEMS or other transanal techniques. As described in a systematic review by Brigic and colleagues11, recent improvements in endoscopic technology have allowed exploration of the possibility of endoscopic full-thickness resection (EFTR) in the colon.

This article summarizes the current evidence base for colonic EFTR and assesses the available data regarding feasibility and safety in experimental models. The available data on the evolution of colonic EFTR into clinical practice will then be examined. We will also evaluate the potential of EFTR and explorethe challenges that need to be overcome to allow further development.

Preclinical studies

The preclinical colonic EFTR work to date has described full thickness wall resection and closure of the subsequent defect using acute and survival porcine models. The reported approaches used either a “pre-resection” closure method, in which the bowel wall was plicated and anastomosed prior to resection, or a “post-resection” closure, where colotomy and specimen resection preceded defect closure.

Procedural methods

The methodsemployed for colonic EFTR varied significantly between the preclinical studies and a brief summary is presented in Table 1. Schurret al12and Rajanet al13 reported usinga Full Thickness Resection Device (FTRD), which includes a hollow flexible shaft with a resection head, an endoscope within the central channel of the device (outer diameter 9.8 mm) alongside graspers for manipulating tissue under endoscopic vision (Figure 1). Procedurally, the resection target was maneuvered into the FTRD resection chamber using either traction or suction and the device fired to create a stapled full thickness resection.

Rajuet al14 described a post-resection closure method involving colonic opening (mesenteric or anti-mesenteric side) using a combination of an insulated needle-knife and snare, followed by closure of the resulting defect with multiple deployments of Tissue Apposition System (TAS) sutures (Ethicon, Endo-surgery Inc, Cincinnati, Ohio)(Figure 2).

A number of studies describedvariations of a grasp-and-snare technique15-18. In the first study from von Rentelnet al16, both pre- and post-resection closure methods using an Over-The-(endo)Scope Clip (OTSC) (Ovesco Endoscopy, Tubingen, Germany) were described(Figure 3). A therapeutic double channel gastroscope (2T160, Olympus, Hamburg, Germany) was used to introduce a tissue closure clip, which deployed three needles at its tip to grasp the bowel wall. Traction was used to create a pseudopolyp. Eight procedures were performed in four animals in which the base of the pseudopolypunderwent pre-resection closure withendoloop ligation (HX-400U-30, Olympus) prior to snare resection (2.5 cm snare, SD-990, Olympus). An OTSC, was then loaded onto a transparent 14-mm cap at the end of the endoscope and applied at the base of the endoloop. A further 20 procedures were performed in 10 animals with defect closure subsequent to the resection (post-resection closure method). Following resection, the edges of the colotomy were manipulated into the cap using a twin grasper (Ovesco, Endoscopy) and one or more OTSCs were deployed to close the defect.

Two subsequent studies describe a modification of this pre-resection closure method15, 17. In these papers, a transparent cap was placed at the end of a single-channel endoscope (EG-2940, Pentax, Hamburg, Germany), preloaded with an OTSC and an electrosurgical snare. Grasping forceps (FG-42-L, Olympus, Hamburg, Germany) were used to manipulate the target area into the cap andthe OTSC was deployed creating a pseudopolyp. The specimen was then snare-resected taking tissue superficial to the clip. In the second study, this equipment had been developed into a single device (Ovesco FTRD (Ovesco endoscopy, Tubingen, Germany)) (Figure 4).

Riederet al18 described a similar pre-resection closure method using the same endoscopic closure system (OTSC, Ovesco Endoscopy, Tubingen, Germany) mounted on a dual channel gastroscope (GIF-2T-160, Olympus). All procedures were performed with laparoscopic overview to ensure safety. In an attempt to improve excision accuracy, anchoring TAS sutures were placed around the target lesion to aid suction into the cap. This was followed by OTSC deployment at the base of the inverted fold,before snare-resection of the pseudopolypsuperficialto the clip.

Feasibility

Resection and defect closure

In the literature, 90/102 (88%) preclinical cases hada completedEFTRprocedure. Of the 12 cases in which failure occurred, these were due to difficulties in defect closure14-16.Raju et al14 reported TAS sutures snapping in 1/20 (5%) animals with a resultant 3 to 4cm colonic wall defect. Von Rentlen et al reported failing to close the defect in 11/20 (55%) animals using the OTSC system, mainly in those with defects of greater than 3cm16.

Safety

Intraoperative complications

Reported complications included both clinical problems and equipment-related difficulties. The overall procedural complication rate was XX (range), in animals where EFTR was completed(Table 2). Rajanet al13 used 10 FTRD devices for 8 procedures as staple-failure occurred in two animals due to the incorrect deployment of a safety ‘lock-out’ mechanism. This was relative easily resolved by device replacement which permitted procedures to be completed successfully. Incomplete resection occurred in another animal leaving a narrow tag of tissue attached at the end of the resection line, which required further electrosurgical snare excision. Haemostatic complications were noted in some studies. Rajanet al13 reported transientbleeding at the resection site in one animal and Schurr et al12observed small haematomas (2 – 20 mm in diameter) abutting the staple line in five animals treated with the FTRD.

Von Rentelnet al16reported complications in 6/9 (67%) animalsusing the post-resection closure method involving the OTSC system. In one case, an adjacent small bowel injury occurred when the OTSC anchor pins were placed through the colonic wall. Bowel obstruction occurred in 5 animals – three through the creation of a stenotic anastomosis and in two through the incorporation of small bowel into the OTSC closure. Fewer complications (in 2/8 animals) were observed in the pre-resection closure method in this study; they included lumen obstruction following OTSC application (n = 1) and endoloop rupture during snare resection (n = 1). The same research group subsequently published a modified method of pre-resection closure that they concluded was more successful, reporting complications in only 2/8 animals15.Both cases were a result of incomplete OTSC deployment – one animal had attempted clip closure of the defect, but this led to bowel obstruction and the animal was euthanized. In the other animal, clip closure rescue was adequate and resulted in a patent lumen.

Schurret al12 described manipulating tissue into the FTRD using either tissue graspers or suction. Incorporation of small bowel into the FTRD closure was reported when suction was used for tissue manipulation. Similar findings were reported by another group investigating closure of colonic perforation using an OTSC19. When the procedural techniques were modified to use traction rather than suction to position the colonic wall, accidental incorporation of adjacent organs was avoided, in both studies.

Anastomotic integrity assessment

An assessment of anastomotic integrity was undertaken in four studies12, 14, 16, 17. Rajuet al14 used intra-luminal methylene blue dye and detected a resection line leak in one animal following a survival study. The other three studies reported bursting pressures following acute studies12, 16. Schurret al12 reported pressures of greater than 38 mm Hg for FTRD stapled anastomosis, whilst values presented by von Renteln et al16 differed significantly between pre- and post-resection closure methods. Use of the endoloop/OTSC combination for pre-resection closure generated higher mean bursting pressures (76.6 mmHg) compared to OTSC post-resection closure (29.2 mmHg). In an assessment of the Ovesco FTRD, the bursting pressure analysis on three animals reported that pressures over 40mmHg could be applied without anastomotic compromise.

Evaluation of the excised specimen

The nature of the excised specimen was assessed in five studies12, 13, 15, 17, 18. Variation in the size of the excised specimen was noted with a mean or median diameter of between 1.7 cm and3.6 cm12-16, 18. Clearance-margin assessment was reported in only two studies13, 18. Riederet al18 placed T-tags in each quadrant around the lesion in order to pullthe target area into the OTSC cap. All resected specimens contained the T-tag sutures. Rajan et al13 who used tissue manipulators to pull the demarcated area of colon into the head of the FTRD, found that although the diathermy markings were evident in all specimens, the mucosal surface was wider (mean 3.8 [SD 0.9] cm) than the serosal surface (mean 2.41 [SD 0.6] cm). This implies a tangential endoscopic excision and raises concerns as to the adequacy of the longitudinal margin clearance. Most other studies did not report on clearance of the longitudinal margins.

Follow-up in pre-clinical Studies

Post-mortem examination was performed by all research groups to evaluate efficacy in preclinical models.13, 14. In the study by Rajanet al13staple-line ulceration was reported in 3/4 animals terminated at 14 days, but normal anastomoses were seen in animals terminated at 28 days. Rajuet al14, who used a post-resection closure method, reported a high incidence of abnormal findings on post-mortem: 16/19 (84%). These included abscesses in 5 cases and significant adhesions in 6 cases. A small anastomotic leak was also identified using methylene blue in one animal which hadrecovered poorly after the procedure.

Clinical studies.

Five studies reporting on 38 patients provide evidence regarding the clinical efficacy of EFTR in the colon20-24 (Table 2). Four of the studies use a combination of OTSC clip and snare resection, either as separate devices or incorporated into a novel clinical FTRD device (FTRD, Ovesco, Tubingen, Germany)20, 21, 23, 24. The remaining study used a post-resection closure method, where the defect was closed with endoscopic clips (standard clip HX-600-135, Olympus; Resolution clip, Boston Scientific, Natick, Massachusetts, USA)22. This study employed laparoscopic assistance for closure in two cases.

Indications

The clinical indications for EFTR in the 17 patients reported by Fahndrich and colleagues were, further resection following endoscopic removal of polyp cancers (n=7), exicision of carcinoid (n=4) and treatment of complex or recurrent adenomas (n=6)23. A single case has been reported of planned EFTR for presumed colonic malignancy24. One series reported on EFTR for submucosal colonic lesions and the vast majority of these were benign tumours22. In the other two studies, EFTR was undertaken for complex adenomas20, 21.

Completion

Technical success (defined by the study authors) was reported in 33/37 patients. Complete resection of the target lesion (R0) resection was reported in all 33 patients across the four studies. The four patients experiencing failure underwent surgical resection. Only three of the studies undertook histological analysis to confirm all layers of the obwel were included in an EFTR specimen 20, 21, 23. When reported, the full thickness of the bowel wall was included in 12/17 procedures.

Complications

Surprisingly few problems were reported in the clinical EFTR procedures. Xu and colleagues22 reported that two of the patients who underwent EFTR without laparoscopic assistance experienced local peritonism, but this was managed conservatively. A further patient experienced bleeding from the resection site that was managed endoscopically. In the remaining three studies, no complications were reported in any patients, however, no study reported using an objective assessment for the identification of complications

Clinical Follow-up

Details of patient follow-up were reported in three studies20-22. Valli and colleagues reported three month follow-up and Schmidt et al reported that patients were followed for 6 months. The study by Xu and colleagues reported that patients did not experience recurrence of the lesion, or other problems, during follow-up, but they did not provide a surveillance time interval. No study reported a robust surveillance schedule.

Discussion

This article summarizes the experimental and clinical outcomes following colonicEFTR described to date. In keeping with a previous review11, this article finds most current experience has been developed in porcine models and although the concept appears to be feasible, further advances into clinical practice have been restricted. Advanced polypectomy techniques such as EMR and ESD are widely used for resection of complex colonic polyps however, significant recurrence rates are reported 4, 25, 26. The benefits for patients with complex colonic polyps treated endoscopically are clear, but these techniques are technically demanding and time consuming, and require extensive training and experience which is not widespread outside Japan. The ability to perform a full thickness resection of a colonic lesion would deliver a specimen superior to that achieved by ESD or EMR without the risk of residual intramural disease. In the prescence of malignancy, histopathologicalevaluation of an intact specimenwould provide a more precise risk assessment of residual disease in the bowel wall or lymph nodes and the need for further surgery could be more accurately communicated to patients. EFTR also has the potential to reduce the need for segmental colectomy, with its associated morbidity and cost. It is possible that it may be usefully combined with laparoscopic sentinel lymph node biopsy when there is a low risk of nodal positivity, in order to remove both the colonic lesion and exclude lymph nodes metastases27. The current preclinical and clinical evidence suggest EFTR may play a role in these circumstances, but further work is needed.

Although several experimental EFTR procedures have been described, problems exist with all approaches and few have progressed into clinical practice. Full thickness resection of the colonic wall was successfully performed using all techniques but secure closure remains a major challenge, with post-resection closure methods in the experimental animal often having difficulties reported. Post-mortem findings from the Raju study14 suggest that the risk of peritoneal contamination associated with this method is excessive. In addition, over 40% of polyps greater than 25mm in diameter have a malignant focus28 and therefore post-resection closure risks the seeding of malignant cells into the peritoneal cavity. The available experimental and clinical evidence suggest pre-resection closure is preferable as the risk of peritoneal contamination is avoided. Future developments of EFTR should be designed to incorporate pre-resection closure or closure that is simultaneous with resection.