TECHNICAL REVIEW

Colonic polypectomy (with videos) Nicholas G. Burgess, MBChB, BSc, FRACP, Farzan F. Bahin, MBBS, FRACP, Michael J. Bourke, MBBS, FRACP Sydney, New South Wales, Australia

The role of colonoscopic polypectomy in the prevention of were searched for English-language manuscripts by using colorectal cancer is now well-established. Resection of adeno- the keywords “colonoscopy” AND “colon OR colonic OR matous colonic polyps reduces colorectal cancer incidence1 colorectal” AND “polypORneoplasmORneoplasia” AND and mortality.2 Despite this proven effectiveness, polyp “polypectomy OR resection.” Studies were included for resection techniques and outcomes remain understudied, review if they were well-designed, prospective trials or and the potential for improvement in efficacy and safety is systematic reviews. Where these were not available, high. Advances in endoscopic and electrosurgical evidence was assessed from large, prospective, observa- technology have allowed the role of endoscopic resection tional trials and case series or reports from recognized to expand, reducing the reliance on surgery for early cancer experts. Each study was assessed according to the and larger or more complex lesions. The optimum GRADE guidelines (Table 1.), and, where sufficient resection technique for any given polyp is quick, ensures evidence was available, a recommendation was made complete adenoma removal, and minimizes adverse events. based on an overall review of the strength of the Variations in polyp size, morphology, histology, and evidence.3 Two reviewers independently reviewed the location mean that there cannot be a “one-size-fits-all” available evidence for each practice point, then arrived approach to resection technique and that polypectomy at a consensus agreement. Each section of the review must be tailored to the characteristics of the lesion, based contains a summary box of the key findings, an on the best available evidence. This review will focus on the indication of the strength of the evidence supporting technical aspects of endoscopic resection in the colon, that finding, and an indication of whether the highlight areas in which evidence is lacking, and comment statement is strong or weak, based on the available on future directions in research. evidence. Existing guidelines were reviewed by searching the National Guideline Clearinghouse for GRADING OF EVIDENCE evidence, specifically on the technical aspects of colon polyp resection. Polyp surveillance strategies and the In conducting this review, electronic databases, associated evidence are beyond the scope of this including MEDLINE/PubMed and the Cochrane Library, review. Before-resection endoscopic assessment of the target lesion is a pivotal component of polypectomy Abbreviations: APC, argon plasma coagulation; ASGE, American Society practice but is covered extensively elsewhere so is not for Gastrointestinal Endoscopy; CT, Computed tomography; DP, dimin- 4 utive polyp; EMR, Endoscopic mucosal resection; ESD, endoscopic sub- addressed in this review. Polypectomy is a technical mucosal dissection; EUS, Endoscopic ultrasound; PRR, polyp retrieval skill, so there are several aspects that are difficult to rate; SC-1, first surveillance colonoscopy; SC-2, second surveillance colo- objectively study. There is increasing evidence that noscopy; SSP, sessile serrated polyp. technique has a strong influence on polypectomy DISCLOSURE: N. Burgess was supported by a grant from the Westmead outcomes. As a result, we have provided technical tips Medical Research Foundation. F. Bahin was supported by a grant from to guide endoscopists on achieving efficient and safe the National Health and Medical Research Council of Australia. Neither excision of lesions. These are not primarily evidence entity influenced preparation, review, or approval of the manuscript. All based but are derived from the literature and our own other authors disclosed no financial relationships relevant to this article. experience, observation, and insights into the understanding of best-practice polypectomy. This video can be viewed directly A total of 2547 articles were identified by using from the GIE website or by using the search strategy. Titles were initially screened, the QR code and your mobile de- and 630 articles were deemed relevant for review. vice. Download a free QR code Article abstracts were individually evaluated for inclusion. scanner by searching “QR Scanner” A total of 410 were excluded.Thecompletetextsof in your mobile device’s app store. the remaining 220 were obtained for the articles that Copyright ª 2015 by the American Society for Gastrointestinal Endoscopy were deemed potentially relevant. In addition, a manual 0016-5107/$36.00 recursive search of the reference sections of the http://dx.doi.org/10.1016/j.gie.2014.12.027 selected studies was performed to identify other www.giejournal.org Volume 81, No. 4 : 2015 GASTROINTESTINAL ENDOSCOPY 813 Colonic polypectomy Burgess et al

TABLE 1. GRADE system for rating the quality of evidence for effectiveness of complete resection for this type of guidelines3 forceps. Hot biopsy forceps techniques may provide an advan- Quality of tage over cold biopsy forceps polypectomy because of evidence Definition Symbol the penumbra of thermal ablation created beyond the High quality Further research is very unlikely to change 4444 focus of tissue excision. However, rates of incomplete our confidence in the estimate of effect. resection range from 17% to 22%,12,13 and the thermal ef- Moderate Further research is likely to have an 444B fect results in difficulty with histologic interpretation of quality important impact on our confidence in the the resected specimen.14-17 It is recommended that hot estimate of effect and may change the estimate. biopsy forceps are used with a low power, peak voltage limited current, and thought that brief application may Low quality Further research is very likely to have an 44BB important impact on our confidence in reduce thermal injury; however, in practice the majority the estimate of effect and is likely to of studies are performed with a coagulating current, change the estimate. and application time inherently varies.18,19 Considerable Very low Any estimate of effect is very uncertain. 4BBB lateral and deep thermal injury can occur with hot biopsy quality forceps. An in vivo porcine study showed that residual target tissue beneath the hot biopsy forceps ulcer occurred in 15%, lateral mucosal injury was unpredict- able, and transmural injury occurred in nearly a third of potentially relevant articles. The full manuscripts of cases.15 Observational studies have demonstrated a high thesearticleswerereviewedindetailtopreparea delayed bleeding rate associated with hot biopsy standardized evidence table and generate focused forceps use.19,20 A survey of hot biopsy forceps use by recommendations. U.S. endoscopists in 13,081 procedures suggested high rates of delayed bleeding (0.41%) and perforation (0.05%), particularly in the right side of the colon.21 REMOVAL OF DIMINUTIVE AND SMALL Unquestionably, removal of DPs should be safe. POLYPS Because of its ineffectiveness and high adverse event rates, hot biopsy forceps cannot be recommended for Diminutive polyps (DP) are defined as polyps %5mm removal of DPs. in size and are extremely common, occurring in 60% to Cold-snare polypectomy has emerged in recent years 70% of patients undergoing screening or surveillance colo- as a safe and efficient way to remove diminutive and noscopy.5 They have a low prevalence of advanced small (%9 mm) polyps, with lower rates of incomplete histologic features.5,6 Although resection of DPs by using resection than biopsy techniques22 and few adverse cold biopsy forceps is attractive, based on ease, accessi- events.23-25 The excision should include a 1- to 2–mm bility, and safety, this method has been associated with sig- rim around the polyp to ensure complete removal nificant rates of incomplete resection. Complete resection (Videos 1 and 2; available online at www.giejournal. by cold biopsy forceps occurred in only 39% when the ad- org). Intraprocedural bleeding is common; however, equacy of resection was assessed by performing EMR of this is typically self-limited and does not result in clini- the polypectomy site.7 Another study examined the DP cally significant bleeding even in anticoagulated pa- cold biopsy forceps site at endoscopy 1 to 4 weeks after tients.26 Cold-snare polypectomy is safe for lesions up the index resection. Residual adenoma was present in to 9 mm in size, with negligible rates of bleeding.23 29%.8 In contrast to this, a similar recent study has Cold-snare polypectomy of lesions O6mminsizeis shown that when performed in conjunction with associated with a residual protrusion in the center of chromoendoscopy and with careful washing and post- the defect (Fig. 1). This is not associated with short- resection examination, complete resection was achieved term sequelae; however, the protrusion may contain in 90% of DPs and 100% of polyps !3 mm in size.9 The muscularis mucosa, indicating incomplete mucosal key concern regarding cold biopsy forceps is that a resection. This aspect requires further research. Early number of bites may be required to remove the polyp, follow-up should be considered when the polyp contains and immediate bleeding and traumatic disruption of the high-grade dysplasia.27 A2003surveyofU.S. mucosal surface may interfere with the assessment of endoscopists showed that 80% used hot snare the adequacy of resection after that.10 A median of 2 polypectomy for lesions 7 to 9 mm in size28; however, bites typically is required for complete resection with it is likely that in contemporary practice this has standard-size biopsy forceps,7,9 and although using decreased substantially. No published studies have “jumbo” forceps has been shown to reduce the number examined the rate of incomplete resection in cold- of bites required for visual eradication of a polyp,11 no snare polypectomy of small polyps, although some pre- studies have been performed to examine the liminary data suggest that incomplete resection rates

814 GASTROINTESTINAL ENDOSCOPY Volume 81, No. 4 : 2015 www.giejournal.org Burgess et al Colonic polypectomy

Figure 1. A, A 3-mm polyp in the distal sigmoid colon. The lesion is oriented at the 6 o’clock position. B, A 15-mm snare is applied over the polyp, ensuring a rim of normal tissue around. In one movement, the snare is fully closed with the application of downward pressure to ensure successful tissue capture. C, The mucosal defect has been expanded by water jet irrigation. This distends the defect and everts the edges to facilitate visual inspection for residual polyp. Any bleeding typically ceases within 1 minute. In this case, there remains a central stalk. This finding has been associated with the resection of lesions O6 mm in size and does not appear to have short-term clinical sequelae.

22,29 are minimal. The complete adenoma resection TABLE 2. Questions in removal of diminutive (% 5 mm) and small 30 (CARE) study highlighted that residual adenoma is (6-9 mm) polyps common after hot snare polypectomy; the incomplete Is there an optimal technique for cold biopsy forceps removal of resection rate was 6.8% for polyps 6 to 9 mm, 17.8% polyps !3 mm? for polyps 10 to 20 mm, and 31.0% for sessile serrated What is the best research model for assessing complete resection? polyps (SSPs). Polypectomy also was highly operator (EMR? biopsy?) dependent, with a 3.4-fold variation in incomplete resec- What is the rate of incomplete resection with cold snare polypectomy? tion rates (6.5%-22.7%) between endoscopists, suggest- What is the optimal technique for cold snaring to ensure complete ing that technique is central to outcome. Prospective resection? fi randomized comparisons are required to assess the ef - Is there an optimum snare type for cold snare polypectomy (wire size cacy and safety of cold-snare polypectomy versus hot or stiffness)? snare polypectomy, particularly in lesions 6 to 9 mm When should hot snare polypectomy be used for small (!10 mm) (Table 2). polyps? Surveillance strategies currently are based on polyp his- Which key aspects of endoscopists technique influence incomplete 31 tology, so effective polyp retrieval after resection is resection? imperative. Biopsy techniques provide high polyp Can methods be developed to include complete resection as a quality 22,32 retrieval rates ([PRR]; 95%-100%) because the tissue metric in colonoscopy practice? is captured within the forceps jaws and simply can be with- drawn through the endoscope. Snare techniques require a second technique for post-resection retrieval. Described techniques include resecting the polyp, removing the polyp retrieved at the end of the procedure. However, if snare, then suctioning the polyp into the working channel multiple polyps are to be resected, the snare is removed by using a filter trap (PRR 88%-100%)22,24,25,32 or ensnaring and a trap used. Lost polyps may require colon washing the polyp, pulling it into the colonoscope channel, and and suctioning for retrieval.33 Failed polyp retrieval in then resecting it while suctioning (PRR 98%).24 The snare one retrospective study with an overall PRR of 94% was may be left in place to occlude the channel and the associated with a PRR of 87% for smaller polyps 1 to www.giejournal.org Volume 81, No. 4 : 2015 GASTROINTESTINAL ENDOSCOPY 815 Colonic polypectomy Burgess et al

TABLE 3. Evidence-based practice points PEDUNCULATED POLYPS

Strength of Quality of Pedunculated polyps make up approximately one third Recommendation recommendation evidence of all polyps in the colon, are predominantly located distal Cold snare polypectomy should be the Strong 444B to the transverse colon, and are typically adenomatous.34 primary modality used for resection About 75% are O10 mm,34 and they may grow to of diminutive polyps. substantial size, developing large feeding blood vessels in Cold biopsy forceps resection should Strong 444B the stalk.35 Large, submucosal lesions such as lipomas % be reserved for polyps 3 mm, may become pedunculated. Both types of large, large capacity or jumbo forceps should be used, and careful visual pedunculated lesions may be mobile and result in colon inspection should be used after obstruction, and they may develop significant ulceration resection to ensure complete and bleeding. removal. Adverse events rarely are encountered with polypec- Hot biopsy forceps should be avoided. Strong 444B tomy of pedunculated polyps up to 20 mm in the distal co- 34 Small, 6-9 mm polyps can be resected Weak 4BBB lon, and these can be resected safely by observing a few by cold snare polypectomy or hot basic principles (Table 7). Increased bleeding rates are snare polypectomy because the noted when stalk diameter exceeds 5 mm.35 However, optimum technique is not defined. the size threshold for the prophylactic application of mechanical measures to prevent bleeding is unknown. Detachable nylon loops applied to ligate the stalk below the resection point (Polyloop; Olympus, Tokyo, Japan) TABLE 4. Technical tips to achieve complete excision in removal of can be effective in preventing postpolypectomy bleeding. diminutive polyps Bleeding rates of 2.7% with nylon loop application in Stiff, thin, wire snares (%0.3-mm diameter) probably are more effective polyps with heads O20 mm in size compared to 2.9% at cold snare polypectomy. with adrenaline stalk injection alone and 15.1% in the 36 Position the polyp at 6 o’clock in the endoscopic field, and stabilize the control group have been described. Polyps !20 mm endoscope position. showed only a trend toward the same effect. The If the polyp is flat, consider directly suctioning the polyp to elevate it combination of nylon loops with epinephrine or nylon from the surrounding mucosa.140 loops with clipping of the resected stalk tip in polyps Deploy the snare over the polyp, ensuring a 1-2 mm margin of normal O20 mm has been shown to be more effective than tissue around the polyp. epinephrine alone (Fig. 2).37,38 Ensure that the snare is parallel to the mucosal surface; a tangential Detachable nylon loop application can be problematic excision risks leaving residual polyp. because the loops are relatively floppy and may be diffi- Apply firm, downward pressure by using the up/down wheel to anchor cult to apply where there is limited space in the colon. En- the snare on normal mucosa. circling polyps with very large heads may be impossible, Suction gas to decrease colon wall tension and facilitate polyp capture. and nylon loops may loosen or detach after placement. Close the snare fully in one slow, continuous movement, observing the An alternative is to apply clips. This technique has been polyp to ensure appropriate tissue capture and completely resect reported in small studies to reduce bleeding. One ran- the polyp. domized study has shown similar bleeding rates when If the snare stalls, maintain continued full closure. clip placement was compared with nylon loop application In cases of sustained stalling, loosen and close again. It may be in polyps with stalks O5 mm in diameter. However, the possible to amputate the polyp against the end of the end of the study was underpowered to demonstrate the non- colonoscope. inferiority of clips.39 Clip application is more costly than Avoid the use of electrocautery because sustained stalling may nylon loops,38 and the total occlusion of feeding vessels indicate muscularis propria capture. by clips may not be assured in very large stalks. Clips Carefully inspect the post-resection defect to ensure complete are likely to be most effective in small stalks in which 1 resection. Liberal water pump irrigation into the defect expands the or 2 clips can be applied or as an alternative if a nylon submucosa, everting the edges of the excision and facilitating inspection, and it may contribute to tamponade of small vessels in loop cannot be placed. Epinephrine volume reduction 40 the event of bleeding to improve visualization. has been described in a small case series and involves injection of dilute epinephrine (1:10000 solution) into the head and stalk. After a period of 3 to 5 minutes, substantial reduction in the size of the polyp has been 5 mm and 81% when cold-snare polypectomy was used.32 reported, permitting standard snare resection. The The optimal retrieval strategy is not defined. However, optimum strategy for prophylactic treatment of high retrieval rates are apparent with all techniques pedunculated polyp stalks has not been defined, and (Tables 3 and 4). the risk of bleeding likely exists along a continuum of

816 GASTROINTESTINAL ENDOSCOPY Volume 81, No. 4 : 2015 www.giejournal.org Burgess et al Colonic polypectomy increasing stalk and polyp sizes. An arbitrary distinction of TABLE 5. Questions in removal of pedunculated polyps R R pedicle diameter 5mmoraheadsize 20 mm, based Is there an optimum current type for removal of small, pedunculated on existing studies, should cause the endoscopist to polyps? consider prophylactic treatment. What is the optimum method for prevention of bleeding in large, Lipomas in the colon may grow to substantial size, her- pedunculated polyps (efficacy and cost effectiveness)? niate into the colon lumen, and resemble a pedunculated adenoma. Typically, the distinction is clear because the mucosa overlying the lesion is normal, and the lesion is soft and easily compressible (pillow sign). Uncommonly, TABLE 6. Evidence-based practice points ulceration and bleeding can occur because of trauma and ischemia, and the lesion may be mistaken for a malig- Strength of Quality of nancy.41 EUS may be helpful in discerning the Recommendation recommendation evidence submucosal etiology.42 Colon lipomas are benign, and Prophylactic measures to prevent Weak 44BB malignancies mimicking lipoma, such as liposarcoma, are postpolypectomy bleeding should be rare.43 Resection should be considered only if the lesion used in pedunculated polyps when stalks exceed 5 mm in diameter or is symptomatic or is causing bleeding, colon obstruction, head size is greater than 20 mm. or intussusception. In these settings, snare resection Volume reduction with epinephrine Weak 4BBB should be attempted with caution because substantial may be used as a method of thermal energy is required to transect the thick stalk, reducing pedunculated polyp size particularly if it contains fat (which conducts heat before resection. poorly). In addition, a large lesion may invaginate the Resection of lipomas by using a Weak 44BB muscularis propria into the stalk. In either case, the risk detachable nylon loop–assisted of deep thermal injury or perforation is high.44 techniques is recommended to avoid Detachable nylon loop ligation of the mucosal “waist” of perforation. the pedunculated lipoma may allow subsequent snare Resection of lipomas should be Strong 444B resection by excluding the muscularis mucosa from the performed only if the lesion is causing or is likely to cause stalk (or cinching it together) and ligating blood symptoms (bleeding, 45,46 vessels. Use of a nylon loop to safely cause gradual intussusception, obstruction) self-amputation of the lipoma has been described.44 Both because the risk of perforation is are acceptable techniques; however, we caution that high, whereas the risk of malignant these should be used only by experienced endoscopists, transformation is very low. preferably in tertiary-care referral units (Tables 5 and 6).

EMR AND ADVANCED POLYPECTOMY dictive of underlying submucosal invasive cancer,49,50 so accurate description is essential. Granular laterally EMR by using a flexible endoscope was first described in spreading tumor lesions have a reported incidence of sub- the 1970s,47 and the fundamental elements remain the mucosal invasion of 3.2% to 7.0%, whereas non-granular same today. Fluid is injected into the submucosal space laterally spreading tumor lesions have a reported incidence to expand this layer, separating the mucosal lesion from of submucosal invasion of 14.0% to 15.3%.49,50 the underlying muscularis propria. The lesion is then The submucosal injectate is a critically important resected by snare electrocautery. The technique is component of successful EMR. Normal saline solution intended to reduce the risk of inadvertent thermal injury has traditionally been used for submucosal elevation. How- or mechanical resection of the muscularis propria and ever, it is limited by a nonsustained, diffuse mucosal lift. lower the risk of perforation (Video 3, available online at Randomized controlled trials comparing saline solution www.giejournal.org). EMR of flat or sessile lesions up to with alternative solutions have shown that a colloidal solu- 25 mm in size has become routine for appropriately tion, succinylated gelatin (Gelofusine [B.Braun, Crissier, trained and experienced endoscopists. Polyps are Switzerland], an inexpensive plasma volume expander), re- classified according to the Paris criteria48 as minimally sults in fewer injections, fewer resection pieces, and elevated !2.5 mm above the surrounding mucosa (0-IIa), improved procedure time in lesions O20 mm,51 and flat (0-IIb), depressed (0-IIc), excavated (0-III), sessile 0.13% hyaluronic acid improves complete resection and O2.5 mm (0-Is), or combinations of these (ie, 0-IIa þ prolongs mucosal elevation in lesions !20 mm.52 Both Is). Laterally spreading lesions O10 mm have been termed appear to be safe. However, Gelofusine is derived from laterally spreading tumors. Their surface morphology may bovine protein and is not universally available, and be classified as granular or non-granular (smooth sur- hyaluronic acid is expensive, which may limit uptake. A faced). In larger polyps, morphology rather than size is pre- number of other solutions have been investigated as an www.giejournal.org Volume 81, No. 4 : 2015 GASTROINTESTINAL ENDOSCOPY 817 Colonic polypectomy Burgess et al

TABLE 7. Technical tips for removal of pedunculated polyps Position the patient so that the polyp hangs in a dependent manner. This may require the patient to be rolled into a supine or right-lateral position. Dependency elongates the stalk and facilitates snare placement. In the event of immediate postpolypectomy bleeding, blood streams away from the nondependent bleeding point, and endoscopic access for hemostasis is optimized. Similarly, in the unlikely event of a perforation, the risk of leakage of bowel content is minimized. Align the polyp mucosal attachment point at 6 o’clock in the endoscopic view. For polyps with a pedicle diameter O5 mm or a head size O20 mm, consider prophylactic detachable nylon loop placement or endoscopic clips. Deploy the snare midway between the mucosal attachment point and the head. In cases where malignant head infiltration is suspected, consider application closer to the mucosal wall. Apply the snare to resistance. Use conventional low-power coagulation current to maximize coagulation while closing the snare in a controlled manner to transect the stalk. If the snare stalls, consider options that include removing the snare by fully opening it and gently passing the colonoscope 5-10 cm proximal to the polyp. In cases of sustained stalling without evidence for muscularis propria entrapment, consider the use of pure-cut or blended electrocautery to complete the resection.

TABLE 8. Questions in EMR and advanced polypectomy TABLE 9. Evidence-based practice points Can the submucosal injectate constituents be altered to improve EMR Strength of Quality of outcomes or reduce bleeding (eg, by incorporating a hemostatic Recommendation recommendation evidence agent)? Does electrosurgical current choice influence the frequency or extent Patients undergoing EMR of Strong 44BB of recurrence? lesions R20 mm in size should have scheduled follow-up within 6 months What is the optimal adjuvant thermal modality for preventing to assess for residual or recurrent recurrent polyp? disease, which occurs in 15%-30%. What is the optimal adjunctive thermal modality for treating residual A chromic dye should be incorporated Strong 44BB or recurrent polyp? into the submucosal injection Can intervals to SC-1 be lengthened, based on histology findings of the solution to facilitate identification of initial EMR (ie, absence of high grade dysplasia)? fluid cushion extent and lesion Subsequent to a clear SC-1, what is the optimal time interval for margins and to identify deep mural surveillance? injury. 44BB What is the optimal technique of scar examination for the detection or Solutions other than saline solution Weak exclusion of recurrence? may be considered for submucosal injection because they may provide a SC-1, First surveillance colonoscopy. more sustained mucosal lift and reduce the number of snare resection pieces. alternative to saline solution. However, their applicability Thermal ablative treatment should be Strong 44BB has been limited because of equivalent or inferior applied only after all efforts have performance to existing solutions, occurrence of local been made to completely excise or tissue reactions, expense, or availability.53-57 The effects remove visible residual adenoma. of epinephrine use in the submucosal injectate are unclear. Thermal ablation of significant 58,59 residual adenoma is an ineffective, Two studies reported low adverse event rates. Howev- single-session treatment committing er, two underpowered studies indicated that it had no ef- the patient to more intensive follow- fect on the rate of early or delayed bleeding.36,60 A meta- up. analysis reported that epinephrine may reduce early Adjuvant thermal ablative treatment Weak 44BB bleeding.61 No studies have been performed in the may be applied to the margins of the setting of large polyp EMR, so this may be an area for endoscopic mucosal defect after future study. Incorporation of a biologically inert blue complete snare resection to prevent recurrence. dye (indigo carmine or methylene blue) is now a common practice in EMR. The dye avidly and selectively stains the loose areolar tissue of the submucosa, resulting in a homogeneous blue post-resection plane areas of concern for deeper resection during the proce- where non-staining residual adenoma and muscularis prop- dure can be irrigated with the dye-containing injectate to ria exposed by deeper resection are easily identified. Any confirm the presence or absence of submucosa and inform

818 GASTROINTESTINAL ENDOSCOPY Volume 81, No. 4 : 2015 www.giejournal.org Burgess et al Colonic polypectomy

TABLE 10. Technical tips for EMR Ensure that the endoscope is straight without a loop in the insertion tube. Position the patient so that the endoscopic view is optimized, preferably with any fluid pool opposite the lesion. This may require right-lateral or supine positioning. Carefully examine the lesion for evidence of submucosal invasion and consider aspects that may increase difficulty: submucosal fibrosis, ileocecal valve or appendiceal involvement, difficult positioning. If the lesion is complex, consider referral to a tertiary-care endoscopic resection service. Align the area for resection at 6 o’clock in the endoscopic view. Begin submucosal injection Deploy the needle tip, and prime the injector needle. Gently touch the lesions surface with the needle tip. Begin injection just before needle puncture of the mucosa. When submucosal lifting is confirmed, lift and manipulate the needle and colonoscope (by using the up/down wheel while gently pulling back on the needle) while continuing infiltration to control the direction and form of the submucosal cushion to optimize elevation and access. Apply the snare to ensure that a rim of normal tissue is captured in addition to the polyp. Apply firm, downward pressure by using the up/down wheel to anchor the snare. Suction gas to decrease colon wall tension and facilitate tissue capture. Use a 3-stage snare closure technique Initially, close until the target tissue is seated within the snare, and the loop of the snare has just started to enter the snare sheath. There may be the sensation of a small “jolt” experienced by the endoscopist within the snare catheter at that point. Aspirate gas again (sometimes even to the point of complete lumen collapse for lesions resistant to snare capture), while pushing down firmly and closing the snare to resistance. Re-insufflate; confirm that the target tissue and margins are ensnared. Excessive puckering may indicate MP capture. Move the snare sheath back and forth to assess mobility. The ensnared tissue should move independently of the colon wall. Fixation may indicate MP capture. If there is concern at this point for MP capture or inaccurate snare placement, the snare may be released and reapplied. Alternatively, gentle release of resistance while elevating the tissue with the up wheel may allow excess tissue or entrapped MP to be excluded. If the endoscopist is confident that snare capture is optimal, then the snare should be closed fully and tightly (Video 3, available online at www. giejournal.org). With full snare closure, by using fractionated current, minimal electrocautery should be required to completely resect the tissue. Resection should be complete in 1 to 3 pulses, 1 to 2 seconds. After resection, irrigate the defect and then carefully inspect for evidence of deep mural injury. For piecemeal resection, continue the resection in a sequential manner, aligning the snare with the edge of the advancing mucosal defect. Snare capture should incorporate the resected mucosal edge and submucosa to avoid leaving tissue islands.

decisions on prophylactic clip closure of possible muscula- TABLE 11. Questions in difficult lesions 62 ris propria injury (Fig. 3). Which polyps are best handled in specialty centers? The dye delineates the transition from adenoma to Which features of difficult lesions predict EMR failure? normal mucosa at the margins of lesions so that the full extent of adenomatous change, particularly in inconspic- Will the development of tertiary-care referral EMR networks improve patient outcomes? uous flat (Paris 0-IIb48) or serrated lesions, is more easily defined. This allows the extent of the submucosal cushion and the safe resection margin to be more easily discerned (Fig. 4). Retroflexion is a useful adjunctive technique used by Controlling the form and extent of the submucosal many endoscopists to improve access for endoscopic cushion allows accurate positioning for snare resection. resection. Lesions on the proximal aspect of folds or partic- Uncontrolled and excessive submucosal injection may ularly the anterior or medial wall of the cecum and lift the lesion into an unfavorable position, and a large, ascending colon may necessitate retroflexion for removal. tense cushion may limit the purchase of the snare, Small case series have shown that the technique is safe particularly in lesions O40 mm. Beginning injection to and effective.64-66 Retroflexion-related perforation has elevate and resect the most difficult and inaccessible been reported, so caution is necessary.67 Retroflexion portion of the lesion first, molding the cushion using with a colonoscope is relatively contraindicated in the subtle catheter and endoscope movements, and lifting left side of the colon because of the narrower lumen. If larger lesions in sections to optimize snare purchase im- necessary, a gastroscope may overcome this challenge; it proves resection.63 also improves maneuverability in retroflexion.68 www.giejournal.org Volume 81, No. 4 : 2015 GASTROINTESTINAL ENDOSCOPY 819 Colonic polypectomy Burgess et al

TABLE 12. Evidence-based practice points TABLE 15. Question in endoscopic submucosal dissection Which lesions should be selected for resection en bloc by endoscopic Strength of Quality of submucosal dissection? Recommendation recommendation evidence

Submucosal tattoo should be used to Strong 444B mark difficult or high-risk lesions for subsequent surgical or endoscopic treatment. TABLE 16. Evidence-based practice points 44BB Submucosal tattoo marking should be Strong Strength of Quality of 2 to 3 cm from lesions where Recommendation recommendation evidence subsequent endoscopic treatment is planned because of the risk of Endoscopic submucosal dissection can Weak 44BB submucosal fibrosis. be considered for the resection of colon lesions; however, local Resection of difficult lesions should be Strong 44BB experience and availability and performed at a tertiary-care center procedural risks compared with risks with experience in endoscopic and benefits of surgery must be resection. weighed carefully.

TABLE 13. Questions in polypectomy and EMR adverse events Which patients and polyps require prophylactic treatment to prevent position or behind a fold. Inject-and-resect EMR and cap- bleeding? assisted EMR both involve lifting the lesion by submucosal Which electrosurgical current is safest? injection. However, in cap-assisted EMR, the lesion is aspi- rated into a specially designed cap that has an inbuilt gutter What is the optimal prophylactic treatment to prevent delayed postpolypectomy bleeding? containing a polypectomy snare. This technique has been used successfully by experienced operators.69,70 However, the risk of perforation is high when the cap is filled with tissue in the thin-walled colon, and the requirement for TABLE 14. Evidence-based practice points specialized equipment when standard inject-and-resect EMR is equally effective have limited its uptake. Strength of Quality of Recommendation recommendation evidence Underwater EMR is a technique that eschews submuco- sal injection based on the EUS observation that water im- 444B Carbon dioxide insufflation should be Strong mersion maintains the involutions of the mucosa and used in preference to air for fl advanced resection procedures in submucosa and oats these layers away from the deeper 71 the colon. muscularis propria layer. The lesion is resected by Deep mural injury identified by the Strong 44BB snare with electrocautery, and in a small series by a target sign should be closed with single operator, resection outcomes were good, with clips. rates of bleeding similar to those of inject-and-resect 71 A microprocessor controlled or blended Weak 44BB EMR. The limitations of the technique are that the non- current could be considered for EMR lifting sign for submucosal invasive cancer is lost, and the because it may reduce delayed outcomes of perforation with a water-filled bowel are un- bleeding in comparison to a pure- known. The technique needs comparative assessment, coagulation current. ideally by a multicenter, randomized, controlled trial in a 44BB Prophylactic endoscopic clip placement Weak large cohort with a range of operators, before it can be may be considered for preventing delayed bleeding in patients applied widely. receiving antiplatelet or Recurrent or residual adenoma at the first surveillance anticoagulant medications, with colonoscopy (SC-1) (typically 3-6 months) is reported in polyps O10 mm in size. 10% to 30% of large, prospective series of EMR outcomes and is a limitation of the technique.72,73 It is associated with larger lesions and the use of thermal ablative thera- pies in which snare resection has been incomplete.49 Other alterations to the standard EMR technique Optimal piecemeal resection involves complete removal include cap-assisted EMR and underwater EMR. The use of the entire lesion by snare, preferably including a 1- to of a standard clear plastic cap to maintain separation 2–mm margin of normal tissue (Fig. 5).74 There is little between the endoscope tip and colon wall to assist in visu- definitive evidence to guide the ideal interval to SC-1. Given alization of lesions is common and may be particularly use- the rates of early recurrence after piecemeal EMR for lesions ful for inject-and-resect EMR when the lesion is in a difficult O 20 mm, review within 6 months is prudent. Follow-up

820 GASTROINTESTINAL ENDOSCOPY Volume 81, No. 4 : 2015 www.giejournal.org Burgess et al Colonic polypectomy should not be too soon (within 3 months) because this these methods is unclear. The use of APC to ablate visible does not allow an opportunity for microscopic residual dis- residual adenoma has been examined in small studies after ease to regenerate and declare itself for treatment. Recur- EMR of sessile lesions O20 mm in size. These studies show rence is typically diminutive and easily managed that although recurrence is reduced, the effect is unreli- endoscopically at this point.49,72 The lack of data means able, and adenoma persists in 14% to 50% of cases.77-79 that the U.S. Multi-Society Task Force on Colorectal Cancer In more recent observational studies, the use of APC as is unable to make an evidence-based statement but recom- an adjunctive treatment for residual adenoma is strongly mends follow-up within 1 year for flat and sessile polyps associated with polyp recurrence.49,80 Prophylactic treat- O15 mm if there is any question about incomplete resec- ment of completely resected adenoma to prevent recur- tion.31 Given recent evidence of residual polyp even in rence is likewise poorly studied. A small trial randomizing lesions deemed to have been completely resected,30 this 21 patients to APC of the edges and base of sessile lesions area urgently requires further study. Second surveillance O15 mm or to no treatment showed that recurrence at 3 colonoscopy (SC-2) review timing is equally undefined. months was significantly reduced to 9% in the APC group. In patients with no recurrence at SC-1 after resection of However, the recurrence rate in the untreated group was flat or sessile lesions O20 mm, recurrence at SC-2 (12 64%, well in excess of typical recurrence rates for months subsequent) is seen in 4.0%. In the patients who EMR.49,78 The variation in results leaves much to be resolved do have recurrence at SC-1 that is endoscopically treated, in this area, because, although APC may reduce some SC-2 recurrence is 20.2%.72 Follow-up at 1 year is therefore recurrence, the technique is neither complete nor consis- essential in this group because the majority of recurrence tent. Criticisms of APC include that it is poorly controllable, remains endoscopically treatable. There are no data on that the ablation depth varies, and that it may leave patchy whether routine biopsy of the polypectomy site in the areas of residual mucosa explaining the recurrence that absence of endoscopic recurrence is required. Prospective occurs despite treatment. Despite this incomplete study to stratify the risk of recurrence for varying polyp evidence, APC is used widely, because there are few other sizes, types, and resection techniques is required. studied thermal ablation modalities available. Research Electrocautery is a key component to snare resection. into alternative techniques is ongoing (Fig. 6, Table 9). However, there is very little evidence to guide practice. Complete histologic assessment is possible only if all There is considerable heterogeneity in the reporting of polyp fragments are retrieved. Retrieving polyp specimens electrocautery equipment and settings. Currents are often is uncomplicated in the distal colon because large pieces described as pure cut (lower peak voltage and continuous can be suctioned directly onto the endoscope tip and waveforms producing a concentrated, local cutting effect then withdrawn for retrieval, and smaller pieces often with little tissue coagulation), pure coagulation (high can be suctioned into the working channel and caught in peak voltage and intermittent waveforms producing a a filter trap. In the proximal colon, repeated endoscope wide coagulation effect with little cutting), or blended intubation may be time consuming or difficult, and (on the spectrum between the two).18 Newer, deformable retrieval nets can be used (Table 10).81,82 microprocessor controlled currents fractionate cutting and coagulating phases and adjust output based on tissue impedance restricting deep tissue injury. In DIFFICULT LESIONS: SUBMUCOSAL FIBROSIS comparison to low-power coagulating currents, the use AND APPENDICEAL AND ILEOCECAL VALVE of microprocessor controlled current results in a crisp LESIONS resection margin more easily interpretable for residual dis- ease. This type of clean excision may improve the technical The majority of colorectal polyps are simple to resect, ease of piecemeal EMR, avoiding tissue islands. However, are in easily accessible locations, and lift well with submu- the penumbra of thermal mucosal injury is small or absent, cosal injection. A small number, however, are difficult or, so the risk of marginal recurrence may be increased. Use of as some authors have termed them, defiant polyps.80 a microprocessor controlled current is associated with Sessile lesions O20 mm in size are associated with reduced rates of delayed bleeding75,76 (see Polypectomy higher adverse event rates,34 and endoscopists must and EMR Adverse Events). Restricted deep injury theoreti- decide whether the resources available to them (ancillary cally may reduce serositis, post-polypectomy pain, and staff, equipment, time, and endoscopic skill) are deep mural injury or perforation, although these aspects sufficient to remove the entire lesion safely and manage have not been well-studied (Table 8). potential adverse events. If not, tertiary-care center referral Diminutive areas of adenoma may be resected with the should be considered strongly (Table 11). use of a monofilament microsnare. However, in some in- Endoscopic treatment of advanced lesions in a tertiary-care stances tiny areas equivocal for adenoma may persist. center setting is substantially more cost effective and likely These areas often are treated with thermal ablationdtypi- safer than surgery.83,84 General endoscopists may be comfort- cally argon plasma coagulation (APC), snare tip soft coagu- able performing en bloc (single piece) or oligo-piecemeal (2-3 lation, or hot biopsy forcepsdbut the effectiveness of piece) resection, but lesions that are large and laterally www.giejournal.org Volume 81, No. 4 : 2015 GASTROINTESTINAL ENDOSCOPY 821 Colonic polypectomy Burgess et al

TABLE 17. Suggested approach to endoscopic treatment of colorectal polyps Diminutive polyps 1-5 mm CSP Cold biopsy forceps only for polyps !3 mm, in difficult position for CSP Small polyps 6-9 mm CSP or hot snare polypectomy Flat and sessile polyps 10-25 mm EMR with blended or microprocessor controlled current Sessile polyps O15 mm (non-granular) or O25 mm (granular) Referral to tertiary-care polypectomy service EMR with microprocessor controlled current Endoscopic submucosal dissection for lesions with a moderate risk of submucosal invasion in the rectum or low sigmoid colon* Pedunculated polyps R10 mm Snare resection with blended current Prophylactic mechanical pretreatment (polyp head O20 mm or stalk O5 mm) (endoscopic clip or detachable nylon loop) For giant, pedunculated polyps for which detachable nylon loop placement is not possible, consider volume reduction with epinephrine or consider referral to a tertiary-care polypectomy service. Lipoma Resection is not required unless symptomatic (bleeding/pain/obstruction). Consider referral to tertiary-care polypectomy service. EUS assessment Endoloop snare resection CSP, Cold snare polypectomy. Treatment options should be considered carefully in any polyp with features suggestive of submucosal invasive cancer (Paris 0-IIc component,48 Kudo Pit Pattern V, non-lifting, Sano III). This may include surgical evaluation. Lesions with features of deep submucosal invasion (combinations of the above, Sano IIIb) or obvious cancers should not be considered for endoscopic treatment. *Endoscopic submucosal dissection is highly dependent on local availability and expertisedhigh-volume centers with low adverse event rates may consider resection of lesions outside these parameters. spreading, require extensive piecemeal resection, are in for subsequent endoscopic resection, and injections difficult colon locations or behind folds are handled should be placed at least 2 to 3 cm from the lesion. Saline more safely in tertiary-care referral units (Fig. 7).20,80,84 solution preinjection may reduce local reactions or A key factor contributing to difficulty with resection is peritoneal injection.92 Lesions involving the appendiceal submucosal fibrosis. Fibrosis adheres the mucosa and sub- orifice or ileocecal valve are difficult to resect and mucosa to the underlying muscularis propria, resulting in historically have been seen as a contraindication to incomplete separation of the layers with snare polypec- resection. However, high rates of success have been tomy and areas of non-lifting with submucosal injection. reported in tertiary-care level advanced endoscopy units, As a result, the lesion is at high risk of incomplete resec- and referral to an expert center should be considered before tion, and the muscularis propria may be injured, resulting surgery.93 Endoscopists should make a careful assessment of in deep mural injury. In addition, previous attempts at lesions before undertaking any resection, assessing the risk resection, or even injudicious biopsy, may cause fibrosis, of invasive cancer, and if the lesion is amenable to EMR, leading to increased rates of subsequent EMR failure.49 taking care not to undertake maneuvers that may Risk of submucosal invasive cancer can be estimated compromise attempts at endoscopic resection. accurately from the gross morphology and surface Endoscopic resection by experienced operators at tertiary- pattern of the lesion, and biopsies often are not required care centers is highly effective, so there are no colon- if the lesion is amenable to EMR and at very low risk of location or polyp factors that completely preclude excision. submucosal invasive cancer.49 Tattoo marking of the The only true marker that a lesion should not be endoscop- mucosa, is an effective method for surgical lesion ically resected is the presence of submucosal invasive can- location,85,86 but infiltration of the ink particles into the cer, and lesions with overt evidence of cancer (Kudo pit submucosa underlying the polyp can cause adverse pattern V, a Paris 0-IIc component,48 non-lifting, a Sano III events87-90 and may result in a fibrotic reaction, increasing vascular pattern, or combinations of these) should be exten- the risk of muscularis propria injury at EMR.91 sively photographed and discussed in a multidisciplinary Endoscopists should exercise care when marking a polyp setting with a view to surgical resection (Table 12).

822 GASTROINTESTINAL ENDOSCOPY Volume 81, No. 4 : 2015 www.giejournal.org Burgess et al Colonic polypectomy

Figure 2. A, A large, pedunculated polyp in the distal sigmoid colon. Maximum head diameter 30 mm, stalk diameter 10 mm. B and C, A flexible nylon loop is applied, encircling the stalk in its lower half. Note the cyanosis of the stalk indicating ischemia. D, A 30-mm snare is applied in the upper half of the stalk, and resection is completed by using a low-power forced coagulation current to ensure adequate hemostasis. E, The resulting stalk stump displays complete resection without coagulation injury to the nylon loop or colon wall. F, Examination of the resection specimen shows complete resection of the entire adenomatous head and a large central feeding artery. Histopathology of the resected specimen showed tubulovillous adenoma with low-grade dysplasia.

POLYPECTOMY AND EMR ADVERSE EVENTS tomy as increasing lesion size, pedunculated polyps, poor bowel preparation, and the use of a pure-cut current.94 Adverse events are uncommon with removal of DPs, but Laterally spreading tumors also were noted to cause high they increase as the size and complexity of endoscopic resec- risk, and in an Australian prospective multicenter study tion escalates.20,34 Consensus on the definition of intrapro- of 1172 patients with large sessile and flat cedural bleeding is lacking. A practical definition of the lesions R20 mm (mean size 35.5 mm), intraprocedural requirement for intervention to control bleeding is objective bleeding was encountered in 11.3% and was associated and reflects the severity of bleeding as well as representing with increasing size, villous or tubulovillous histology, cost in terms of interruption to the procedure and equip- Paris type 0-IIaþIs lesions,48 and study centers ment use. Intraprocedural bleeding is reported in up to contributing !75 cases.75 Intraprocedural bleeding also 2.8% of patients undergoing standard polypectomy in non- was associated with delayed bleeding (odds ratio [OR] specialty settings94 and 11% of patients undergoing EMR of 2.16; P Z .016) and recurrence (relative risk 1.68; P Z sessile or flat lesions O20 mm in size.75 .011) at SC-1. Intraprocedural bleeding typically is A large, prospective, South Korean study identified risk controlled by using thermal modalities, although brisk or factors for intraprocedural bleeding in standard polypec- persistent bleeding unresponsive to these methods can www.giejournal.org Volume 81, No. 4 : 2015 GASTROINTESTINAL ENDOSCOPY 823 Colonic polypectomy Burgess et al

Figure 3. A, Nearing completion of a 60-mm, Paris 0-IIaþIs, laterally spreading tumor resection in the distal descending colon, the defect base is noted to have nonstaining areas (B) raising concern for deep mural injury. C, Topical application of a chromic dye (methylene blue) via the injector sheath with the needle retracted highlights the distinction between submucosa and muscularis propria and demonstrates a ring of nonstaining tissue consistent with a site of muscularis propria excision or injury “target sign” inside the outer normal cautery ring of mucosal incision. Target signs generally are aligned trans- versely to the long axis of the colon as this one is. Presumably, the inner circular muscle fibers splay after injury as shown here. D, After resection of the adjacent adenoma, the injury is subsequently fully closed with endoscopically placed clips everting the edges of the muscle wound into the lumen. It is not necessary to close the entire mucosal defect. require endoscopic clip application. If bleeding occurs part risk factors are increasing lesion size,34,76,97,100,101 large way through a resection, avoiding clip placement is recom- sessile polyps,102 proximal colon location,34,76,100 and thie- mended because it may interfere with the completion of nopyridine or anticoagulant use.96,103,104 Increasing age the procedure. The use of a voltage-limited micropro- has not been shown to predict bleeding but may be asso- cessor controlled current to apply thermal energy is rec- ciated with a requirement for transfusion.102 Although ommended to avoid deep thermal injury. The ideal hypertension was associated with bleeding in one modality for control is quick, avoids changing devices, is study,99 the majority of lesions in this cohort were accurate, localized, and safe. Snare tip soft coagulation resected by hot biopsy forceps, and other studies failed fits all of these criteria and has been shown to be an effec- to verify the association.76,97 Electrosurgical current selec- tive method of intraprocedural bleeding control.95 tion has been shown to influence bleeding. However, The technique involves protruding the snare tip 2 to few studies have compared current types. A retrospective 3 mm beyond the sheath and then lightly touching the study of postpolypectomy bleeding in 1485 snare resec- tip directly onto the bleeding point. Vigorous washing tions that used either a pure-coagulating or blended cur- with a foot pedal–controlled water pump is essential to rent showed that, although bleeding rates did not differ, clear the visual field and pinpoint the bleeding vessel. bleeding was immediate in all cases that used a blended Direct washing also infiltrates the submucosal tissue, current, and bleeding was delayed in all cases that used providing some tissue elevation and potentially producing pure-coagulation current.105 a tamponade effect. Kim et al76 showed in a retrospective case control study Postpolypectomy bleeding is challenging to study that bleeding rates were significantly reduced by use of a because event rates are low in general endoscopic practice, microprocessor controlled current for EMR but not for large numbers of patients need to be enrolled to accurately hot biopsy forceps removal of small polyps. A large, study risk factors, and definitions of bleeding vary. The ma- prospective, multicenter, observational study of wide- jority of retrospective studies of bleeding estimate the risk field EMR reported that clinically significant delayed to be 0.5% to 2%,96-99 and the most consistently reported bleeding was more common when a pure-coagulation or

824 GASTROINTESTINAL ENDOSCOPY Volume 81, No. 4 : 2015 www.giejournal.org Burgess et al Colonic polypectomy

Figure 4. A, A 30-mm, Paris 0-IIb Kudo III, nongranular laterally spreading polyp in the mid-ascending colon. The lesion is difficult to discern from the surrounding normal mucosa by high-definition white-light endoscopy. B, The margins remain indistinct when narrow-band imaging is used. C, After sub- mucosal injection with a combination of succinylated gelatin (Gelofusine, B. Braun, Crissier, Switzerland), indigo carmine, and epinephrine, the margins of the defect are clearly demarcated. D, E, As the submucosal injectate is expanded, the margins become clearer again, revealing a subtle transition be- tween adenoma and normal tissue. F, Now fully appreciated, the distinction is clear, and resection can proceed, ensuring a clear margin of normal mucosa. blended current was used in comparison to a micropro- coagulation currents may produce excessive deep-tissue cessor controlled current (n Z 1172, OR 2.03; P Z .038) injury resulting in delayed bleeding. Further controlled tri- and that 69% of bleeding occurred within 48 hours.75 als are warranted, based on this limited evidence, because Use of a microprocessor controlled current (Endocut) it is likely that selecting the correct electrosurgical current was reported to improve the quality of histologic for the specific resection task may significantly improve specimens in comparison to use of a blended current.106 outcomes and reduce risk. Pure-cut current was associated with an acceptable postpo- Prophylactic endoscopic clip placement appears ineffec- lypectomy bleeding rate of 1.1% in a large, retrospective tive at reducing delayed bleeding in unselected polyps ! study. However, 12% of patients had prophylactic Endo- 10 mm in size.108 However, studies examining this group loops or clips applied. have low adverse event rates, so are typically These findings are consistent with evidence in animal underpowered. A meta-analysis and cost-effectiveness study models107 showing that pure-cut currents result in little showed that for polyps O10 mm in size, treatment was cost deep-tissue injury, but they may not provide sufficient effective only for those patients receiving antiplatelet or anti- coagulation to prevent immediate bleeding, and pure- coagulant agents.109 Endoscopic clips were reported to be www.giejournal.org Volume 81, No. 4 : 2015 GASTROINTESTINAL ENDOSCOPY 825 Colonic polypectomy Burgess et al

Figure 5. A, A 30-mm, Paris 0-IIaþIs, granular laterally spreading tumor in the ascending colon. B, With the lesion in a favorable position oriented at 6o’clock in the endoscopic view, the lesion is injected with a solution containing a combination of succinylated gelatin (Gelofusine), indigo carmine, and adrenaline. C, A 20-mm snare is applied, ensuring that a wide margin of normal mucosa is captured in addition to the adenoma. D, E, The lesion is sequentially snare resected, and the snare wire is carefully aligned with the edge of the advancing mucosal defect to avoid leaving islands of residual adenoma within the defect. F, A small area of residual tissue is noted at the proximal margin (arrow). G, This is resected with a small, stiff, thin, wire snare. Complete snare resection must be the goal, avoiding ablation of residual tissue except in instances where snare resection is impossible or unsafe. H, The final defect without evidence of residual adenoma.

826 GASTROINTESTINAL ENDOSCOPY Volume 81, No. 4 : 2015 www.giejournal.org Burgess et al Colonic polypectomy

Figure 6. A, A 30-mm, Paris 0-IIaþIs, granular laterally spreading polyp in the rectosigmoid junction. B, The lesion is resected piecemeal by sequential inject-and-resect EMR. C, Careful attention is paid to ensure that all visible polyp is completely removed. D, The margins of the defect are treated by snare tip soft coagulation. Prophylactic adjuvant ablation by using a thermal modality may reduce polyp recurrence, although the optimum technique is yet to be defined. effective at reducing bleeding in an uncontrolled Pain after polypectomy or EMR often is related to persis- retrospective study of 524 unselected polyps O20 mm in tent colon distension and tension on the postresection size.110 These findings require assessment in randomized defect or wound. Carbon dioxide is now the preferred insuf- controlled trials because there are technical and cost flation gas for colonoscopy because it significantly reduces barriers to clipping of very large lesions (Table 13). postprocedural pain in comparison to air, reduces pneumo- Perforation is an uncommon adverse event after poly- peritoneum in the event of perforation, and is of minimal pectomy, but it can have devastating clinical sequelae. If additional cost.113,114 Water immersion colonoscopy re- it or the warning signs are not recognized during the pro- duces patient discomfort associated with intubation, and, cedure, fecal peritoneal contamination may occur, signifi- in conjunction with carbon dioxide, is an effective way to cantly raising the risk of surgery.111 Inspection of the reduce procedural pain.115,116 Postpolypectomy syndrome post-EMR defect is an integral part of EMR. Signs indicating refers to serositis induced by polypectomy and is thought injury to the muscularis propria may be identified before to be related to transmural thermal injury or inadvertent they become frank perforation. Incorporation of a blue peritoneal injection. Typically, patients present with abdom- chromic dye into the submucosal injectate allows the spec- inal pain and, in some cases, symptoms and signs indistin- imen target sign (a pale disk on the underside of the re- guishable from colon perforation, requiring hospital sected specimen) or the defect target sign (concentric admission and CT imaging. In a large, retrospective study, pale rings within the resection defect) to be identified postpolypectomy syndrome occurred in 2.9% of patients and prophylactically closed with clips to reduce the risk and was severe enough to require hospitalization in 0.7 in of perforation (Fig. 8).112 These characterized signs 1000. Large lesion size and nonpolypoid morphology pre- represent only part of the spectrum of deep mural injury, dicted postpolypectomy syndrome.117 It has been and a range of injuries exist spanning non-concerning reported in 1% to 3.6% after EMR.118,119 Outcomes are exposure of uninjured muscularis propria to frank perfora- good with conservative management (Table 14).117 tion with fecal contamination. Structured appraisal of the risk and approach to management is applicable to all sub- SERRATED POLYPS AND POLYPECTOMY mucosal lift polypectomy and allows endoscopists greater confidence in their assessments of patient disposition after In addition to being difficult to detect, SSPs are more polypectomy. likely to be incompletely resected than conventional www.giejournal.org Volume 81, No. 4 : 2015 GASTROINTESTINAL ENDOSCOPY 827 Colonic polypectomy Burgess et al

Figure 7. A, A circumferential, Paris 0-IIaþIs, granular laterally spreading polyp at the rectosigmoid junction. B, The lesion is sequentially removed by inject-and-resect EMR. C, D, The resection defect is bland, with no areas of deep mural injury or bleeding. Stenosis may be possible as the defect heals, so close follow-up is required, and dilatation may be considered. Lesions such as this are challenging to remove safely and successfully and ideally should be handled in a specialist tertiary-care endoscopy center. adenomas. The CARE study demonstrated that 31% of SSPs injector that can be used to infiltrate the submucosal had remnant tissue in the resection defect compared with space, allowing concurrent lifting and dissection of the sub- 7.2% of conventional adenomas, and in lesions O10 mm in mucosal layer. The technique now has a central role in the size, residual tissue remained in 47.5%.30 SSPs may have management of early esophageal and gastric cancer. ESD indistinct margins, and smaller lesions may prove difficult has the ability to achieve en bloc or R0 resection. This to entrap with the snare because of their flat nature. has 3 potential advantages over piecemeal EMR: (1) More SSPs also may contain dysplastic foci within the lesion, accurate histologic assessment. This may be of benefit with an endoscopic appearance indistinguishable from with subtle submucosal invasion that could be missed or conventional adenomas, and the surrounding serrated misinterpreted by the histopathologist when multiple component may be overlooked and incompletely piecemeal specimens are assessed. In practice, this has resected if this is not recognized.120 The incorporation of not proven to be clinically relevant.121 (2) Reduced a contrast dye into the submucosal injectate allows clear recurrence. Recent, long-term outcomes from large, pro- demarcation of the edges of the lesion, and EMR with a spective, EMR series show that recurrence usually is dimin- chromo-injectate is therefore recommended for all flat utive and is treated easily in follow-up, with no impact on lesions R10 mm in size in the proximal colon. Lesions ! the likelihood of long-term remission.72 (3) The possibility 10 mm should be inspected carefully with narrow-band im- of cure in low-risk submucosal invasive cancer. Low risk as- aging and high-definition white light endoscopy. There is sumes !1000 mm depth of invasion from the muscularis currently no evidence to guide the electrocautery choice mucosae and absence of lymphovascular involvement, tu- for serrated lesions, so this is an area for future research. mor budding, or poor differentiation. Such cases are infre- quent, based on large Japanese and Korean series to date, ENDOSCOPIC SUBMUCOSAL DISSECTION AND making up 6% to 12% of patients treated with ESD.122–124 HYBRID PROCEDURES Thus, in most cases, an up-front ESD strategy to treat large, flat, and laterally spreading polyps offers little additional Endoscopic submucosal dissection (ESD) is a technique benefit to the patient but exposes them to increased pro- pioneered in Japan for the treatment of early gastric can- cedural risks and the health service to increased costs, cer. It involves the use of specially engineered electrosur- decreased colonoscopy capacity, and mandatory multiple- gical needles or knives, often with a high-pressure day hospital admission for these patients (Table 15).

828 GASTROINTESTINAL ENDOSCOPY Volume 81, No. 4 : 2015 www.giejournal.org Burgess et al Colonic polypectomy

Figure 8. A, A mucosal defect after resection of a colonic, laterally spreading polyp. The loose areolar connective tissue is homogeneously stained by the indigo carmine dye. An area in the center of the image is poorly staining, suggesting fibrosis or potential deeper resection. B, After topical submucosal chromoendoscopy, this area is homogeneously stained, confirming resection in the submucosal plane. C, A defect target sign in the base of an EMR defect. The concentric rings indicate injury to the muscularis propria. D, The resected specimen displays the specimen target sign. E, A prominent defect target sign. F, Clip closure of the defect ensures that any risk of subsequent complete perforation is avoided.

ESD is time consuming, with a steep learning curve. the considerable advantage of lymph node staging. Risks of perforation in the colon have been reported at Careful discussion with the patient is necessary and must 3% to 10.7%.122,125 Costs and inpatient length of stay also take into account local experience and alternative are greater.126 treatment options (Table 16).121 Hybrid procedures combining ESD and EMR have been described127 but are not widely practiced. Risks of ESD perforation are significantly lower in the rectosigmoid ANTITHROMBOTICS AND ANTIBIOTICS colon, and adverse events associated with the distal colon and rectal surgery can be considerable, so ESD as The periprocedural management of antithrombotic a primary approach for lesions in this area may be agents should be individualized in respect of the patients warranted. By contrast, ESD of early invasive lesions in cardiovascular risk and the postpolypectomy bleeding the proximal colon is associated with greater technical risk in order to minimize bleeding while maintaining suffi- difficulty, longer procedure time, and higher risks of cient protection from thromboembolic events. The emer- perforation,128 whereas a right hemicolectomy often can gence of novel oral anticoagulants (direct thrombin or be performed easily and safely laparoscopically and has factor Xa inhibitors) and increasingly potent www.giejournal.org Volume 81, No. 4 : 2015 GASTROINTESTINAL ENDOSCOPY 829 Colonic polypectomy Burgess et al

Figure 9. A, A 70-mm, hemi-circumferential, Paris 0-IIaþIs,48 granular laterally spreading polyp in the rectum involving the anorectal junction. B, Resection in the rectum may require a combination of techniques including a transparent cap or a gastroscope in retroflexion. C, Resection close to the dentate line can produce pain because of the somatic innervation, so a long-acting local anaesthetic (ropivacaine 0.5%) is incorporated in the sub- mucosal injectate along with succinylated gelatin (Gelofusine), indigo carmine, and epinephrine. D, The final resection defect. The venous drainage of the anorectal region flows directly into the systemic circulation. Antibiotic prophylaxis is not required for EMR elsewhere in the colon. However, in this circumstance it is warranted to prevent bacteremia. thienopyridines (prasugrel, ticagrelor) in addition to a wide to suggest that in EMR aspirin increases the risk of range of existing agents means that managing periproce- bleeding of sessile lesions O20 mm in size.134 Bleeding dural antithrombotic agents is complex and requires the appears to be increased in patients receiving dual endoscopist to have a clear understanding of the antith- antiplatelet agents135,136 and warfarin96 and is likely rombotic indication, onset and offset, variation with renal elevated in patients receiving novel oral anticoagulants, or hepatic function, and reversal agents. The detailed based on indirect evidence.137,138 If antithrombotic agents management of antithrombotics is beyond the scope cannot be discontinued, hemostatic clips may mitigate the of this review, and guidelines making up evidence-based risk of bleeding.139 The approach to low-risk polypectomy recommendations are available from the American in patients receiving antithrombotics is evolving and re- Society for Gastrointestinal Endoscopy (ASGE),129 quires further research. European Society for Gastrointestinal Endoscopy,130 and Antibiotics are not required for routine polypectomy. British Society of Gastroenterology.131 The most recent However, patients undergoing resection of large lesions guideline updates from these societies occurred in 2011. from the anorectal junction should have intraprocedural There have been significant changes in the field since antibiotics because the blood supply for this region drains then.132 The current ASGE guidelines class polypectomy directly into the common iliac veins, circumventing the as a high-risk procedure. However, there is increasing evi- portal circulation and the reticuloendothelial system, which dence that resection of sessile polyps !10 mm is low has a major role in sequestering pathogens (Fig. 9).68 risk.34,97 There also is evidence that cold snare polypec- tomy of lesions!10 mm is safe,23-25 even in anticoagulated Summary patients,26 although adequately powered studies are Colonoscopic polypectomy necessitates an adaptable required to further define this. Aspirin use does not approach to differing lesion morphologies, pathobiology, lo- influence bleeding event rates irrespective of polyp cations, and risks (Fig. 10). Colonoscopists need to be able to size,96-98,133 and the ASGE and European Society for select the most appropriate technique, electrosurgical Gastrointestinal Endoscopy recommend its continuation settings, and ancillary equipment for the lesion they face for all conventional polypectomy. There is some evidence and should have an array of techniques available to control

830 GASTROINTESTINAL ENDOSCOPY Volume 81, No. 4 : 2015 www.giejournal.org Burgess et al Colonic polypectomy

Figure 10. A, B, C, Three examples of large, hemi-circumferential, laterally spreading polyps. A Paris 0-IIaþIs, granular rectal lesion (A), a Paris 0-IIa, non- granular lesion in the ascending colon (B), and a Paris 0-IIaþIs, granular lesion in the sigmoid colon (C). D, E, F, The lesions are resected by sequential piecemeal EMR. G, H, I, Each lesion is fully snare resected, ensuring no residual adenoma within the defect or at the margins. adverse events when they occur (Table 17). Polypectomy ACKNOWLEDGMENTS is swiftly moving from a heuristically learned skill with a minimal evidence base to a well-defined technique with The authors would like to acknowledge the assistance of a growing evidence base for many of its core compo- Nicholas Tutticci, MBBS, FRACP, for providing the cold nents. Despite this, there are several areas that remain snare polypectomy video footage and Rebecca Sonson, without evidence, and few recommendations in this re- RN, for video editing. view are based on high-quality, definitive evidence. Well-designed studies are required to improve safety and efficacy. Serrated polyp resection, electrocautery choice, and the optimal resection of small to medium– REFERENCES sized (6-20 mm) polyps are among the areas that ur- 1. Winawer SJ, Zauber AG, Ho MN, et al. Prevention of colorectal cancer gently require further investigation. Implementation of by colonoscopic polypectomy. The National Polyp Study Workgroup. wide-scale screening programs in many countries on N Engl J Med 1993;329:1977-81. the basis of the now well-established efficacy of 2. Zauber AG, Winawer SJ, O’Brien MJ, et al. Colonoscopic polypectomy screening and polypectomy in the reduction of colorectal and long-term prevention of colorectal-cancer deaths. N Engl J Med 2012;366:687-96. cancer incidence and mortality means that more patients 3. Guyatt GH, Oxman AD, Vist GE, et al. GRADE: an emerging consensus than ever before will be exposed to the benefits and the on rating quality of evidence and strength of recommendations. BMJ risks of this procedure. 2008;336:924-6. www.giejournal.org Volume 81, No. 4 : 2015 GASTROINTESTINAL ENDOSCOPY 831 Colonic polypectomy Burgess et al

4. Holt BA, Bourke MJ. Wide field endoscopic resection for advanced 26. Horiuchi A, Nakayama Y, Kajiyama M, et al. Removal of small colo- colonic mucosal neoplasia: current status and future directions. Clin rectal polyps in anticoagulated patients: a prospective randomized Gastroenterol Hepatol 2012;10:969-79. comparison of cold snare and conventional polypectomy. Gastroint- 5. Gupta N, Bansal A, Rao D, et al. Prevalence of advanced histological est Endosc 2014;79:417-23. features in diminutive and small colon polyps. Gastrointest Endosc 27. Tutticci NJ, Burgess NG, Pellise M, et al. Prospective characterization 2012;75:1022-30. of the mucosal defect after cold snare polypectomy [abstract]. Gas- 6. Hassan C, Pickhardt PJ, Kim DH, et al. Systematic review: distribution trointest Endosc 2014;79:AB541. of advanced neoplasia according to polyp size at screening colonos- 28. Singh N, Harrison M, Rex DK. A survey of colonoscopic polypectomy copy. Aliment Pharmacol Ther 2010;31:210-7. practices among clinical gastroenterologists. Gastrointest Endosc 7. Efthymiou M, Taylor AC, Desmond PV, et al. Biopsy forceps is inade- 2004;60:414-8. quate for the resection of diminutive polyps. Endoscopy 2011;43: 29. Urquhart P, Brown GJ. Su1426 The effectiveness of cold snare poly- 312-6. pectomy for the removal of small sessile colonic polyps [abstract]. 8. Woods A, Sanowski RA, Wadas DD, et al. Eradication of diminutive Gastrointest Endosc 2012;75:AB328. polyps: a prospective evaluation of bipolar coagulation versus con- 30. Pohl H, Srivastava A, Bensen SP, et al. Incomplete polyp resection dur- ventional biopsy removal. Gastrointest Endosc 1989;35:536-40. ing colonoscopydresults of the Complete Adenoma Resection 9. Jung YS, Park JH, Kim HJ, et al. Complete biopsy resection of dimin- (CARE) study. Gastroenterology 2012;144:74-80.e1. utive polyps. Endoscopy 2013;45:1024-9. 31. Lieberman DA, Rex DK, Winawer SJ, et al. Guidelines for colonoscopy 10. Tolliver KA, Rex DK. Colonoscopic polypectomy. Gastroenterol Clin surveillance after screening and polypectomy: a consensus update by North Am 2008;37:229-51, ix. the US Multi-Society Task Force on Colorectal Cancer. Gastroenter- 11. Draganov PV, Chang MN, Alkhasawneh A, et al. Randomized, ology 2012;143:844-57. controlled trial of standard, large-capacity versus jumbo biopsy for- 32. Komeda Y, Suzuki N, Sarah M, et al. Factors associated with failed ceps for polypectomy of small, sessile, colorectal polyps. Gastrointest polyp retrieval at screening colonoscopy. Gastrointest Endosc Endosc 2012;75:118-26. 2013;77:395-400. 12. Ellis K, Schiele M, Marquis S, et al. Efficacy of hot biopsy forceps. Cold 33. Waye J. It ain’t over ’til it’s over: retrieval of polyps after colonoscopic micro-snare and micro-snare with cautery techniques in the removal polypectomy. Gastrointest Endosc 2005;62:257-9. of diminutive colonic polyps [abstract]. Gastrointest Endosc 1997;45: 34. Heldwein W, Dollhopf M, Rösch T, et al. The Munich Polypectomy AB107. Study (MUPS): prospective analysis of complications and risk fac- 13. Peluso F, Goldner F. Follow-up of hot biopsy forceps treatment tors in 4000 colonic snare polypectomies. Endoscopy 2005;37: of diminutive colonic polyps. Gastrointest Endosc 1991;37: 1116-22. 604-6. 35. Dobrowolski S, Dobosz M, Babicki A, et al. Blood supply of colorectal 14. Cohen J, Savides T, Jensen D, et al. Randomized single blind study polyps correlates with risk of bleeding after colonoscopic polypec- of jumbo bipolar hot biopsy forceps versus cold techniques for tomy. Gastrointest Endosc 2006;63:1004-9. removal of diminutive colon polyps. Gastrointest Endosc 1995;41: 36. Di Giorgio P, De Luca L, Calcagno G, et al. Detachable snare versus 375. epinephrine injection in the prevention of postpolypectomy 15. Metz AJ, Moss A, McLeod D, et al. A blinded comparison of the safety bleeding: a randomized and controlled study. Endoscopy 2004;36: and efficacy of hot biopsy forceps electrocauterization and conven- 860-3. tional snare polypectomy for diminutive colonic polypectomy in a 37. Kouklakis G, Mpoumponaris A, Gatopoulou A, et al. Endoscopic resec- porcine model. Gastrointest Endosc 2013;77:484-90. tion of large pedunculated colonic polyps and risk of postpolypec- 16. Goldstein NS, Watts JC, Neill JS, et al. The effect of electrothermal tomy bleeding with adrenaline injection versus endoloop and cautery-assisted resection of diminutive colonic polyps on histopath- hemoclip: a prospective, randomized study. Surg Endosc 2009;23: ologic diagnosis. Am J Clin Pathol 2001;115:356-61. 2732-7. 17. Mönkemüller KE, Fry LC, Jones BH, et al. Histological quality of polyps 38. Luigiano C, Ferrara F, Ghersi S, et al. Endoclip-assisted resection of resected using the cold versus hot biopsy technique. Endoscopy large pedunculated colorectal polyps: technical aspects and 2004;36:432-6. outcome. Dig Dis Sci 2010;55:1726-31. 18. Morris ML, Tucker RD, Baron TH, et al. Electrosurgery in gastrointes- 39. Ji J-S, Lee S-W, Kim TH, et al. Comparison of prophylactic clip and en- tinal endoscopy: principles to practice. Am J Gastroenterol doloop application for the prevention of postpolypectomy bleeding 2009;104:1563-74. in pedunculated colonic polyps: a prospective, randomized, multi- 19. Weston AP, Campbell DR. Diminutive colonic polyps : histopathology, center study. Endoscopy 2014;46:598-604. spatial distribution, concomitant significant lesions, and treatment 40. Hogan RB. Epinephrine volume reduction of giant colon polyps facil- complications. Am J Gastroenterol 1995;90:24-8. itates endoscopic assessment and removal. Gastrointest Endosc 20. Chukmaitov A, Bradley CJ, Dahman B, et al. Association of polypec- 2007;66:1018-22. tomy techniques, endoscopist volume, and facility type with colonos- 41. Meghoo CA, Cook PR, McDonough CA, et al. Large colonic lipoma copy complications. Gastrointest Endosc 2013;77:436-46. with mucosal ulceration mimicking carcinoma. Gastrointest Endosc 21. Wadas DD, Sanowski RA. Complications of the hot biopsy forceps 2003;58:468-70. technique. Gastrointest Endosc 1988;34:32-7. 42. Kawamoto K, Yamada Y, Utsunomiya T, et al. Gastrointestinal submu- 22. Lee CK, Shim J-J, Jang JY. Cold snare polypectomy vs. Cold forceps cosal tumors: evaluation with endoscopic US. Radiology 1997;205: polypectomy using double-biopsy technique for removal of diminu- 733-40. tive colorectal polyps: a prospective randomized study. Am J Gastro- 43. Choi YY, Kim YJ, Jin SY. Primary liposarcoma of the ascending colon: a enterol 2013;108:1593-600. rare case of mixed type presenting as hemoperitoneum combined 23. Repici A, Hassan C, Vitetta E, et al. Safety of cold polypectomy for with other type of retroperitoneal liposarcoma. BMC Cancer !10mm polyps at colonoscopy: a prospective multicenter study. 2010;10:239. Endoscopy 2012;44:27-31. 44. Kaltenbach T, Milkes D, Friedland S, et al. Safe endoscopic treatment 24. Deenadayalu VP, Rex DK. Colon polyp retrieval after cold snaring. of large colonic lipomas using endoscopic looping technique. Dig Gastrointest Endosc 2005;62:253-6. Liver Dis 2008;40:958-61. 25. Ichise Y, Horiuchi A, Nakayama Y, et al. Prospective randomized com- 45. Murray MA, Kwan V, Williams SJ, et al. Detachable nylon loop assisted parison of cold snare polypectomy and conventional polypectomy removal of large clinically significant colonic lipomas. Gastrointest En- for small colorectal polyps. Digestion 2011;84:78-81. dosc 2005;61:756-9.

832 GASTROINTESTINAL ENDOSCOPY Volume 81, No. 4 : 2015 www.giejournal.org Burgess et al Colonic polypectomy

46. Wild D, Fiore J, Guelrud M. Successful endoscopic resection of a giant 65. Rex DK, Khashab M. Colonoscopic polypectomy in retroflexion. Gas- colonic lipoma causing intussusception. Gastrointest Endosc 2008;68: trointest Endosc 2006;63:144-8. 774-5; discussion 775. 66. Pishvaian AC, Al-Kawas FH. Retroflexion in the colon: a useful and safe 47. Longacre TA, Fenoglio-Preiser CM. Mixed hyperplastic adenomatous technique in the evaluation and resection of sessile polyps during co- polyps/serrated adenomas: a distinct form of colorectal neoplasia. lonoscopy. Am J Gastroenterol 2006;101:1479-83. Am J Surg Pathol 1990;14:524-37. 67. Quallick MR, Brown WR. Rectal perforation during colonoscopic retro- 48. Paris Workshop Participants. The Paris endoscopic classification of su- flexion: a large, prospective experience in an academic center. Gas- perficial neoplastic lesions: esophagus, stomach, and colon: trointest Endosc 2009;69:960-3. November 30 to December 1, 2002. Gastrointest Endosc 2003;58: 68. Holt BA, Bassan MS, Sexton A, et al. Advanced mucosal neoplasia of S3-43. the anorectal junction: endoscopic resection technique and out- 49. Moss A, Bourke MJ, Williams SJ, et al. Endoscopic mucosal resection comes (with videos). Gastrointest Endosc 2014;79:119-26. outcomes and prediction of submucosal cancer from advanced 69. Conio M, Blanchi S, Repici A, et al. Cap-assisted endoscopic mucosal colonic mucosal neoplasia. Gastroenterology 2011;140:1909-18. resection for colorectal polyps. Dis Colon Rectum 2010;53:919-27. 50. Uraoka T, Saito Y, Matsuda T, et al. Endoscopic indications for endo- 70. Conio M, Blanchi S, Filiberti R, et al. Cap-assisted endoscopic mucosal scopic mucosal resection of laterally spreading tumours in the color- resection of large polyps involving the ileocecal valve. Endoscopy ectum. Gut 2006;55:1592-7. 2010;42:677-80. 51. Moss A, Bourke M, Metz A. A randomized, double-blind trial of succi- 71. Binmoeller KF, Weilert F, Shah J, et al. “Underwater” EMR without sub- nylated gelatin submucosal injection for endoscopic resection of mucosal injection for large sessile colorectal polyps (with video). Gas- large sessile polyps of the colon. Am J Gastroenterol 2010;105: trointest Endosc 2012;75:1086-91. 2375-82. 72. Moss A, Williams SJ, Hourigan LF, et al. Long-term adenoma recur- 52. Yoshida N, Naito Y, Inada Y, et al. Endoscopic mucosal resection with rence following wide-field endoscopic mucosal resection (WF-EMR) 0.13% hyaluronic acid solution for colorectal polyps less than 20 mm: for advanced colonic mucosal neoplasia is infrequent: results and a randomized controlled trial. J Gastroenterol Hepatol 2012;27: risk factors in 1000 cases from the Australian Colonic EMR (ACE) 1377-83. study. Gut 2015;64:57-65. 53. Fasoulas K, Lazaraki G, Chatzimavroudis G, et al. Endoscopic 73. Knabe M, Pohl J, Gerges C, et al. Standardized long-term follow-up mucosal resection of giant laterally spreading tumors with submu- after endoscopic resection of large, nonpedunculated colorectal le- cosal injection of hydroxyethyl starch: comparative study with sions: a prospective two-center study. Am J Gastroenterol 2013;109: normal saline solution. Surg Laparosc Endosc Percutan Tech 183-9. 2012;22:272-8. 74. Nanda KS, Bourke MJ. Endoscopic mucosal resection and complica- 54. Katsinelos P, Kountouras J, Paroutoglou G, et al. Endoscopic mucosal tions. Tech Gastrointest Endosc 2013;15:88-95. resection of large sessile colorectal polyps with submucosal injection 75. Burgess NG, Metz AJ, Williams SJ, et al. Risk factors for intraprocedural of hypertonic 50 percent dextrose-epinephrine solution. Dis Colon and clinically significant delayed bleeding after wide-field endoscopic Rectum 2006;49:1384-92. mucosal resection of large colonic lesions. Clin Gastroenterol Hepatol 55. Arantes V, Albuquerque W, Benfica E, et al. Submucosal injection of 2014;12:651-61. e1-3. 0.4% hydroxypropyl methylcellulose facilitates endoscopic mucosal 76. Kim JH, Lee HJ, Ahn JW, et al. Risk factors for delayed post- resection of early gastrointestinal tumors. J Clin Gastroenterol polypectomy hemorrhage: a case-control study. J Gastroenterol Hep- 2010;44:615-9. atol 2013;28:645-9. 56. Hurlstone DP, Fu KI, Brown SR, et al. EMR using dextrose solution 77. Zlatanic J, Waye JD, Kim PS, et al. Large sessile colonic adenomas: use versus sodium hyaluronate for colorectal Paris type I and 0-II lesions: of argon plasma coagulator to supplement piecemeal snare polypec- a randomized endoscopist-blinded study. Endoscopy 2008;40:110-4. tomy. Gastrointest Endosc 1999;49:731-5. 57. Hiraoka S, Kato J, Fujiki S, et al. The presence of large serrated polyps 78. Regula J, Wronska E, Polkowski M, et al. Argon plasma coagulation af- increases risk for colorectal cancer. Gastroenterology 2010;139: ter piecemeal polypectomy of sessile colorectal adenomas: long-term 1503-10, 1510.e1-3. follow-up study. Endoscopy 2003;35:212-8. 58. Dobrowolski S, Dobosz M, Babicki A, et al. Prophylactic submucosal 79. Brooker JC, Saunders BP, Shah SG, et al. Treatment with argon plasma saline-adrenaline injection in colonoscopic polypectomy: prospective coagulation reduces recurrence after piecemeal resection of large randomized study. Surg Endosc 2004;18:990-3. sessile colonic polyps: a randomized trial and recommendations. Gas- 59. Hsieh YH, Lin HJ, Tseng GY, et al. Is submucosal epinephrine injection trointest Endosc 2002;55:371-5. necessary before polypectomy? A prospective, comparative study. 80. Buchner AM, Guarner-Argente C, Ginsberg GG. Outcomes of EMR of Hepatogastroenterology 2001;48:1379-82. defiant colorectal lesions directed to an endoscopy referral center. 60. Lee S-H, Chung I-K, Kim S-J, et al. Comparison of postpolypectomy Gastrointest Endosc 2012;76:255-63. bleeding between epinephrine and saline submucosal injection for 81. Miller K, Waye JD. Polyp retrieval after colonoscopic polypectomy: use large colon polyps by conventional polypectomy: a prospective of the Roth Retrieval Net. Gastrointest Endosc 2001;54:505-7. randomized, multicenter study. World J Gastroenterol 2007;13: 82. Diehl DL, Adler DG, Conway JD, et al. Endoscopic retrieval devices. 2973-7. Gastrointest Endosc 2009;69:997-1003. 61. Li LY, Liu QS, Li L, et al. A meta-analysis and systematic review of pro- 83. Ahlenstiel G, Hourigan LF, Brown G, et al. Actual endoscopic phylactic endoscopic treatments for postpolypectomy bleeding. Int J versus predicted surgical mortality for treatment of advanced Colorectal Dis 2011;26:709-19. mucosal neoplasia of the colon. Gastrointest Endosc 2014;80: 62. Holt BA, Jayasekeran V, Sonson R, et al. Topical submucosal chro- 668-76. moendoscopy defines the level of resection in colonic EMR and 84. Swan MP, Bourke MJ, Alexander S, et al. Large refractory colonic may improve procedural safety (with video). Gastrointest Endosc polyps : Is it time to change our practice? A prospective study of 2013;77:949-53. the clinical and economic impact of a tertiary referral colonic mucosal 63. Soetikno R, Kaltenbach T. Dynamic submucosal injection technique. resection and polypectomy service (with videos). Gastrointest Endosc Gastrointest Endosc Clin N Am 2010;20:497-502. 2009;70:1128-36. 64. Hurlstone DP, Sanders DS, Thomson M, et al. “Salvage” endoscopic 85. Arteaga-González I, Martín-Malagón A, Fernández EML-T, et al. The mucosal resection in the colon using a retroflexion gastroscope use of preoperative endoscopic tattooing in laparoscopic colorectal dissection technique: a prospective analysis. Endoscopy 2006;38: cancer surgery for endoscopically advanced tumors: a prospective 902-6. comparative clinical study. World J Surg 2006;30:605-11. www.giejournal.org Volume 81, No. 4 : 2015 GASTROINTESTINAL ENDOSCOPY 833 Colonic polypectomy Burgess et al

86. Askin MP, Waye JD, Fiedler L, et al. Tattoo of colonic neoplasms in 113 107. Chino A, Karasawa T, Uragami N, et al. A comparison of depth of tis- patients with a new sterile carbon compound. Gastrointest Endosc sue injury caused by different modes of electrosurgical current in a 2002;56:339-42. pig colon model. Gastrointest Endosc 2004;59:374-9. 87. Bang CS, Kim YS, Baik GH, et al. Colonic abscess induced by India ink 108. Shioji K, Suzuki Y, Kobayashi M, et al. Prophylactic clip application tattooing. Korean J Gastroenterol 2014;64:45-8. does not decrease delayed bleeding after colonoscopic polypectomy. 88. Alba LM, Pandya PK, Clarkston WK. Rectus muscle abscess associated Gastrointest Endosc 2003;57:691-4. with endoscopic tattooing of the colon with India ink. Gastrointest 109. Parikh ND, Zanocco K, Keswani RN, et al. A cost-efficacy decision anal- Endosc 2000;52:557-8. ysis of prophylactic clip placement after endoscopic removal of large 89. Park SI, Genta RS, Romeo DP, et al. Colonic abscess and focal perito- polyps. Clin Gastroenterol Hepatol 2013;11:1319-24. nitis secondary to India ink tattooing of the colon. Gastrointest En- 110. Liaquat H, Rohn E, Rex DK. Prophylactic clip closure reduced the risk dosc 1991;37:68-71. of delayed postpolypectomy hemorrhage: experience in 277 clipped 90. Coman E, Brandt LJ, Brenner S, et al. Fat necrosis and inflammatory large sessile or flat colorectal lesions and 247 control lesions. Gastro- pseudotumor due to endoscopic tattooing of the colon with india intest Endosc 2013;77:401-7. ink. Gastrointest Endosc 1991;37:65-8. 111. Iqbal CW, Chun YS, Farley DR. Colonoscopic perforations: a retro- 91. Moss A, Bourke MJ, Pathmanathan N. Safety of colonic tattoo with spective review. J Gastrointest Surg 2005;9:1229-35; discussion sterile carbon particle suspension: a proposed guideline with illustra- 1236. tive cases. Gastrointest Endosc 2011;74:214-8. 112. Swan MP, Bourke MJ, Moss A, et al. The target sign : an endoscopic 92. Fu KI, Fujii T, Kato S, et al. A new endoscopic tattooing technique for marker for the resection of the muscularis propria and potential identifying the location of colonic lesions during laparoscopic sur- perforation during colonic endoscopic mucosal resection. Gastroint- gery: a comparison with the conventional technique. Endoscopy est Endosc 2011;73:79-85. 2001;33:687-91. 113. Bassan MS, Holt B, Moss A, et al. Carbon dioxide insufflation reduces 93. Nanda K, Tuttucci N, Burgess N, et al. Endoscopic mucosal resection number of postprocedure admissions after endoscopic resection of of advanced mucosal neoplasia involving the ileocecal valve with large colonic lesions: a prospective cohort study. Gastrointest Endosc ileal infiltration: endoscopic features and outcome [abstract]. Gastro- 2013;77:90-5. intest Endosc 2014;79:AB547. 114. Bretthauer M, Thiis-Evensen E, Huppertz-Hauss G, et al. NORCCAP 94. Kim HS, Kim TI, Kim WH, et al. Risk factors for immediate postpolypec- (Norwegian colorectal cancer prevention): a randomised trial to tomy bleeding of the colon: a multicenter study. Am J Gastroenterol assess the safety and efficacy of carbon dioxide versus air insufflation 2006;101:1333-41. in colonoscopy. Gut 2002;50:604-7. 95. Fahrtash-Bahin F, Holt BA, Jayasekeran V, et al. Snare tip soft coagu- 115. Vemulapalli KC, Rex DK. Water immersion simplifies cecal intubation lation achieves effective and safe endoscopic hemostasis during in patients with redundant colons and previous incomplete colonos- wide-field endoscopic resection of large colonic lesions (with videos). copies. Gastrointest Endosc 2012;76:812-7. Gastrointest Endosc 2013;78:158-63.e1. 116. Lin S, Zhu W, Xiao K, et al. Water intubation method can reduce pa- 96. Hui AJ, Wong RMY, Ching JYL, et al. Risk of colonoscopic polypectomy tients’ pain and sedation rate in colonoscopy: a meta-analysis. Dig En- bleeding with anticoagulants and antiplatelet agents: analysis of dosc 2013;25:231-40. 1657 cases. Gastrointest Endosc 2004;59:44-8. 117. Cha JM, Lim KS, Lee SH, et al. Clinical outcomes and risk factors of 97. Sawhney MS, Salfiti N, Nelson DB, et al. Risk factors for severe delayed post-polypectomy coagulation syndrome: a multicenter, retrospec- postpolypectomy bleeding. Endoscopy 2008;40:115-9. tive, case-control study. Endoscopy 2013;45:202-7. 98. Yousfi M, Gostout CJ, Baron TH, et al. Postpolypectomy lower gastro- 118. Salama M, Ormonde D, Quach T, et al. Outcomes of endoscopic resec- intestinal bleeding: potential role of aspirin. Am J Gastroenterol tion of large colorectal neoplasms: an Australian experience. 2004;99:1785-9. J Gastroenterol Hepatol 2010;25:84-9. 99. Watabe H, Yamaji Y, Okamoto M, et al. Risk assessment for delayed 119. Church J. Experience in the endoscopic management of large colonic hemorrhagic complication of colonic polypectomy: polyp-related polyps. ANZ J Surg 2003;73:988-95. factors and patient-related factors. Gastrointest Endosc 2006;64: 120. Nanda KS, Tutticci N, Burgess N, et al. Caught in the act: endoscopic 73-8. characterization of sessile serrated adenomas with dysplasia. Gastro- 100. Buddingh KT, Herngreen T, Haringsma J, et al. Location in the right intest Endosc 2014;79:864-70. hemi-colon is an independent risk factor for delayed post- 121. Bourke MJ, Neuhaus H. Colorectal endoscopic submucosal dissection: polypectomy hemorrhage : a multi-center case-control study. Am J when and by whom? Endoscopy 2014;46:677-9. Gastroenterol 2011;106:1119-24. 122. Saito Y, Fukuzawa M, Matsuda T, et al. Clinical outcome of endoscopic 101. Consolo P, Luigiano C, Strangio G, et al. Efficacy, risk factors and com- submucosal dissection versus endoscopic mucosal resection of large plications of endoscopic polypectomy : ten year experience at a sin- colorectal tumors as determined by curative resection. Surg Endosc gle center. World J Gastroenterol 2008;14:2364-9. 2010;24:343-52. 102. Sorbi D, Norton I, Conio M, et al. Postpolypectomy lower GI bleeding: 123. Lee E-J, Lee JB, Lee SH, et al. Endoscopic submucosal dissection for descriptive analysis. Gastrointest Endosc 2000;51:690-6. colorectal tumorsd1,000 colorectal ESD cases: one specialized insti- 103. Witt DM, Delate T, McCool KH, et al. Incidence and predictors of tute’s experiences. Surg Endosc 2013;27:31-9. bleeding or thrombosis after polypectomy in patients receiving and 124. Niimi K, Fujishiro M, Kodashima S, et al. Long-term outcomes of endo- not receiving anticoagulation therapy. J Thromb Haemost 2009;7: scopic submucosal dissection for colorectal epithelial neoplasms. 1982-9. Endoscopy 2010;42:723-9. 104. Gandhi S, Narula N, Mosleh W, et al. Meta-analysis: colonoscopic post- 125. Kobayashi N, Yoshitake N, Hirahara Y, et al. Matched case-control polypectomy bleeding in patients on continued clopidogrel therapy. study comparing endoscopic submucosal dissection and endoscopic Aliment Pharmacol Ther 2013;37:947-52. mucosal resection for colorectal tumors. J Gastroenterol Hepatol 105. Van Gossum A, Cozzoli A, Adler M, et al. Colonoscopic snare polypec- 2012;27:728-33. tomy: analysis of 1485 resections comparing two types of current. 126. Tutticci N, Bourke MJ. Advanced endoscopic resection in the colon: Gastrointest Endosc 1992;38:472-5. recent innovations, current limitations and future directions. Expert 106. Fry LC, Lazenby AJ, Mikolaenko I, et al. Diagnostic quality of: Rev Gastroenterol Hepatol 2014;8:161-77. polyps resected by snare polypectomy: Does the type of electro- 127. Moss A, Bourke MJ, Metz AJ, et al. Beyond the snare: technically surgical current used matter? Am J Gastroenterol 2006;101: accessible large en bloc colonic resection in the West: an animal 2123-7. study. Dig Endosc 2012;24:21-9.

834 GASTROINTESTINAL ENDOSCOPY Volume 81, No. 4 : 2015 www.giejournal.org Burgess et al Colonic polypectomy

128. Tanaka S, Terasaki M, Hayashi N, et al. Warning for unprincipled 136. Feagins LA, Uddin FS, Davila RE, et al. The rate of post-polypectomy colorectal endoscopic submucosal dissection: accurate diag- bleeding for patients on uninterrupted clopidogrel therapy during nosis and reasonable treatment strategy. Dig Endosc 2013;25: elective colonoscopy is acceptably low. Dig Dis Sci 2011;56:2631-8. 107-16. 137. Connolly SJ, Ezekowitz MD, Yusuf S, et al. Dabigatran versus warfarin 129. Anderson MA, Ben-Menachem T, Gan SI, et al. Management of antith- in patients with atrial fibrillation. N Engl J Med 2009;361:1139-51. rombotic agents for endoscopic procedures. Gastrointest Endosc 138. Hernandez I, Baik SH, Piñera A, et al. Risk of bleeding with dabigatran 2009;70:1060-70. in atrial fibrillation. JAMA Intern Med 2015;175:18-24. 130. Boustière C, Veitch A, Vanbiervliet G, et al. Endoscopy and antiplatelet 139. Friedland S, Soetikno R. Colonoscopy with polypectomy in anticoagu- agents. European Society of Gastrointestinal Endoscopy (ESGE) lated patients. Gastrointest Endosc 2006;64:98-100. Guideline. Endoscopy 2011;43:445-61. 140. Pattullo V, Bourke MJ, Tran KL, et al. The suction pseudopolyp tech- 131. Veitch AM, Baglin TP, Gershlick AH, et al. Guidelines for the manage- nique: a novel method for the removal of small flat nonpolypoid le- ment of anticoagulant and antiplatelet therapy in patients undergo- sions of the colon and rectum. Endoscopy 2009;41:1032-7. ing endoscopic procedures. Gut 2008;57:1322-9. 132. Baron TH, Kamath PS, McBane RD. Management of antithrombotic therapy in patients undergoing invasive procedures. N Engl J Med 2013;368:2113-24. Received September 30, 2014. Accepted December 4, 2014. 133. Manocha D, Singh M, Mehta N, et al. Bleeding risk after invasive pro- cedures in aspirin/NSAID users: polypectomy study in veterans. Am J Current affiliations: Department of Gastroenterology and Hepatology, Med 2012;125:1222-7. Westmead Hospital; Sydney Medical School, University of Sydney, Sydney, 134. Metz AJ, Bourke MJ, Moss A, et al. Factors that predict bleeding New South Wales, Australia. following endoscopic mucosal resection of large colonic lesions. Reprint requests: Professor Michael J. Bourke, Director of Endoscopy, Endoscopy 2011;43:506-11. Department of Gastroenterology and Hepatology, Westmead Hospital, 135. Singh M, Mehta N, Murthy UK, et al. Postpolypectomy bleeding in pa- c/- Suite 106a, 151-155 Hawkesbury Road, Westmead, Sydney, New South tients undergoing colonoscopy on uninterrupted clopidogrel ther- Wales 2143 Australia. apy. Gastrointest Endosc 2010;71:998-1005.

www.giejournal.org Volume 81, No. 4 : 2015 GASTROINTESTINAL ENDOSCOPY 835