Endoloop-assisted polypectomy for large peduncolated colorectal polyps

Surg Endosc (2006) 20: 1257–1261 DOI: 10.1007/s00464-005-0713-5  Springer Science+Business Media, Inc. 2006

Endoloop-assisted polypectomy for large pedunculated colorectal polyps P. Katsinelos,1 J. Kountouras,2 G. Paroutoglou,1 A. Beltsis,1 G. Chatzimavroudis,3 C. Zavos,2 I. Vasiliadis,1 T. Katsinelos,1 B. Papaziogas3 1 2 3

Department of Endoscopy and Motility Unit, Central Hospital, Ethnikis Aminis 41, 546 35 Thessaloniki, Greece Second Medical Clinic, Aristotle University of Thessaloniki, Ippokration Hospital, Thessaloniki, Greece Second Surgical Clinic, Aristotle University of Thessaloniki, Thessaloniki, Greece

Received: 19 October 2005/Accepted: 14 March 2006/Online publication: 20 July 2006

Abstract Background: The use of an endoloop may minimize the risk for bleeding after endoscopic polypectomy of large colorectal polyps. This study aimed to assess the safety and efficacy of colonoscopic ligation of the stalk of large pedunculated polyps by means of an endoloop technique, and to focus particular attention on the instances in which the use of this device was unsuccessful. Methods: This study retrospectively evaluated attempted endoloop endoscopic polypectomy in 33 patients (19 men and 14 women; mean age, 62.5 years) with large pedunculated polyps. Results: Application of the endoloop was impossible in four patients, and the snare became entangled with the loop in one patient. The remaining 28 patients underwent endoloop-assisted polypectomy. Bleeding occurred in four patients, either because the loop slipped of the stalk after polypectomy (2 patients) or because a thin stalk ( £ 4 mm) was transected by the loop before polypectomy (2 patients). Conclusion: Colonoscopic polypectomy with an endoloop may be safer than conventional polypectomy. The reasons for technical failure of this technique include a narrow left colon lumen, a thin stalk ( £ 4 mm), and close cutting in relation to the site of encirclement by the loop.

greater with large polyps that have thick stalks [1–3]. The other important complications, namely, postpolypectomy coagulation syndrome (0.5–1%) [4] and perforation (0.5%) [5] are less common. Bleeding can occur in the immediate postpolypectomy period, or may be delayed a few days [4] or up to 30 days [6]. Several endoscopic techniques have been developed to prevent bleeding. Injection of the stalk with epinephrine solution or sclerosants before transection is recommended to diminish the risk of postpolypectomy hemorrhage. However, epinephrine injection may prevent only procedural bleeding, whereas sclerosant injection may increase the risk of perforation [6]. More recently, mechanical hemostatic devices, such as endoloops [7–9] and clips [3], have been introduced for the prevention of early and delayed postpolypectomy bleeding. Although anecdotal evidence suggests that use of the endoloop is relatively widespread in specialist therapeutic endoscopy units, surprisingly, few published studies have described the experience with this device [7–9]. We report our experience with the use of endoloops for colonoscopic resection of large pedunculated polyps in the current series, drawing particular attention to those instances in which endoloop application was ineffective.

Key words: Colonoscopic ligation — Endoloop-assisted polypectomy — Large colorectal polyps — Large pedunculated polyps

Patients and methods

Postpolypectomy bleeding is reported to occur in 2% of all polypectomies. The risk of bleeding probably is

Correspondence to: P. Katsinelos

This study reviewed patients in whom an endoloop-assisted polypectomy for colorectal polyps was attempted between October 1999 and September 2005. A total of 33 patients (19 men and 14 women; mean age, 62.5 years; range, 41–87 years) with large pedunculated polyps were included in the study. The patients were treated in our department, which serves as a tertiary referral center for northern Greece. All the patients were identified by colonoscopy performed in our department, by colleagues in other hospitals, or by private gastroenterologists. The ethics committee of our hospital approved the initiation of this technique, and written informed consent was obtained from all the patients after

1258

Fig. 1. An endoloop for endoscopic polypectomy of large pedunculated polyps.

explanations detailing the risks and benefits of an endoloop-assisted or conventional polypectomy. The inclusion criteria required the following: • Pedunculated polyp in the colorectum with a head at least 10 mm in diameter. Because there is a tendency to either overestimate or underestimate the polyp size at colonoscopy by almost one-third, even when using biopsy forceps for comparison, we used the absolute size of the polyps head and stalk, measured at its retrieval after the resection. • Endoscopist determination that pedunculated polyp was benign in appearance and not ulcerated or indurated.

Fig. 2. A large pedunculated polyp with a thick stalk.

Endoscopic technique After careful preparatory cleansing of the bowel using a polyethylene glucol-electrolyte solution, all colonoscopies were performed using a standard video colonoscope with the patient under conscious sedation by administration of midazolam and meperidine. When necessary, hyoscine-n-butylbromide was administered intravenously for colon relaxation, unless contraindicated, in which case, glucagon was introduced intravenously. To minimize the degree of variability in polypectomy technique, all resections were performed by one of two experienced endoscopists (P.K. and G.P.). A detachable endoloop system is composed of an operating part and an attached loop (Fig. 1). The operating part consists of a Teflon sheath 2.5 mm in diameter and 195 cm in working length, a stainless steel coil sheath 1.9 mm in diameter, a hook wire, and a handle. Endoloops are currently available in diameters of 20 and 30 mm (MAJ340 and MAJ-254; Olympus, Tokyo, Japan). The larger loop can be opened to a size of 50 · 30 mm. Moreover, the attached loop sloughs in 4 to 7 days. Endoloops are nonconductive and consist of a heattreated elliptically shaped soft Teflon ring and a silicon rubber stopper that maintains the tightness of the loop. Before use, the loop is retracted inside the plastic sheath for insertion through the accessory channel of the colonoscope. We prefer using a large-channel therapeutic endoscope that provides adequate suction alongside the inserted snare. We maneuvered the scope so that the polyp was visualized at the 6 oÕclock position on the video screen (Fig. 2). A second assistant often was needed to hold the scope at the correct position. Patient monitoring during colonoscopy and polypectomy included pulse oximetry. After the loop had been extended and applied at the base of the stalk, it was tightened around the stalk by sliding the stopper, then detached from the operating part (Fig. 3). To ensure sufficient tightening, we observed the color of the polyp head changing to dark red after ligation. A diathermic snare then was used to sever the stalk of the polyp above the tightened loop using electrosurgical coagulation current (20–30 W; PSD-30; Olympus) (Fig. 4). After polypectomy, the polyp was captured with a basket catheter and gently extracted through the anus. The maximum diameter of the head and the stalk of the polyp was measured and recorded, then submitted for histologic evaluation. Unless hematochezia occurred, the patients left the hospital after colonoscopic polypectomy. If hematochezia occurred, the patients

Fig. 3. The endoloop has been tightened at the base of the stalk. were hospitalized until hematochezia ceased and blood tests (hemoglobin–hematocrit) became stable.

Histologic examination The polyps were pinned on a plate, fixed with 10% formalin, sliced into parallel sections, and stained with hematoxylin and eosin for histopathologic examination. Two experienced pathologists examined the sections. If cancer was detected, the decision to pursue a more aggressive treatment was based on the histologic features of the cancer, the distance from the resection margins, and the level of cancer invasion into the stalk of the polyp. In general, pedunculated polyps with carcinoma invading the head neck or stalk require no further treatment, provided that the resection margins are free of cancer, the histology is well or moderately differentiated, and there is no lymphatic or vascular invasion. Polyps with carcinoma invading the submucosa carry a significant risk of lymph node metastases, and in such cases, patients were considered as candidates for additional radical surgery. If the patients refused operation, they were followed along with the other patients.

Complications Polypectomy-induced bleeding was defined as intraprocedural (during polypectomy), early (within 24 h), or delayed (> 24 h). The diagnosis

1259 Table 1. Characteristics of 28 large pedunculated polyps removed by endoloop-assisted polypectomy Characteristics Location Descending colon Sigmoid colon Rectum Mean size (mm) Head: n (range) Stalk: n (range) Histology Tubular adenoma Adenocarcinoma Complications Intraprocedural bleeding Delayed bleeding Perforation Postcoagulation syndrome

Polyps 6 20 2 17.8 (11–26) 5.4 (3.8–7.1) 23 5 2 2 0 0

Fig. 4. After polypectomy the endoloop is left in place with no bleeding at the resection site.

of early or delayed bleeding was based on the passage of fresh blood per rectum. If necessary, bleeding was controlled by hemoclip placement or heater probe treatment. Postcoagulation syndrome, caused by transmural thermal injury with resultant serosal inflammation, was evaluated. This syndrome is characterized by localized abdominal pain, occasionally with fever. Perforation was defined as abdominal pain with fever, leucocytosis, and the presence of free air inside the abdominal cavity.

Follow-up evaluation At discharge from the hospital, the patients and their relatives were instructed to contact the endoscopic team in case of any adverse event. To monitor complications, patients were contacted by telephone at the end of the first and second week after the procedure and advised to undergo surveillance at 6 months.

Results In 4 of the 33 patients, application of the endoloop was not possible because of a narrow sigmoid colon lumen and polyp head size. These polyps were treated with piecemeal resection after injection of the stalk with 50% dextrose plus epinephrine (1:10000) solution. In another patient, the polypectomy snare became entangled with the endoloop in a way precluding transection of the peduncle, and the polyp was excised with piecemeal resection. These five patients were excluded from the study. In the remaining 28 patients, the endoloop was correctly placed, and cutting of the polyp head was possible. The characteristics of the 28 polyps are summarized in Table 1, and the outcome is depicted in Fig. 5. The mean duration of the procedure was 17.8 min (range, 11–27 min). The polyp head was 20 mm or larger in 21 lesions (63.8%), the largest being 28 mm. The diameter of the stalk was 5 mm or larger in 23 lesions (89.1%), the largest being 7.1 mm. Intraprocedural bleeding occurred in two patients because the thin stalk ( £ 4 mm) was transected with the

Fig. 5. Outcome of endoloop-attempted polypectomy of 33 large pedunculated polyps.

tightening of the endoloop. In both cases, hypertonic dextrose 50% plus epinephrine (1:10000) solution was injected initially around the residual stalk followed by placement of three and four clips (HX-6UR-1; Olympus) on the stalks, respectively. In two other patients, the endoloop slipped off after polypectomy and showed delayed bleeding: the one patient on day 6 and the other on day 7. They both were treated with hemoclipping. In the five cases of carcinoma, the cancer was limited to the mucosal layer in four lesions, whereas the depth of invasion was submucosal in the fifth patient. In this patient, a subsequent colectomy with lymph node dissection was performed because of possible lymph node metastases. Follow-up evaluation was available for a median of 32.4 months (range, 3–68 months). Follow-up endoscopy was performed for 27 patients. No recurrence of polyp nor appearance of cancer was observed.

Discussion Postpolypectomy hemorrhage frequently is an unpredictable event that creates great anxiety and, like all complications, is best avoided. Bleeding after colono-

1260

scopic polypectomy has been reported in 0.3% to 6.1% of cases, depending on the numbers of patients and the types of resected polyps studied [10]. It is encountered mainly with removal of pedunculated polyps because of a large artery in the stalk [10, 11]. In earlier studies, bleeding usually occurred immediately after snaring, whereas in more recent reports, delayed hemorrhage accounts for about 65% of postpolypectomy bleeding, and may occur as long as 3 to 4 weeks after polypectomy, although it most often occurs within the first 24 h. This is possibly attributable to the change from blended to coagulation current [12]. New strategies aimed at preventing bleeding after polypectomy include the endoloop technique, first introduced by Pontecorvo and Pesce [13] and developed by Hachisu [14], for endoscopic ligation of the stalk of large pedunculated polyps. Moreover, a significant advantage of using loop ligation for large pedunculated polyps with a thick stalk is the ability to use greater electrosurgical generator power, including cutting or blended current for polypectomy, without fear of precipitating hemorrhage. Despite the evidence that the endoloop is widely used prophylactically for polypectomy of large pedunculated polyps in therapeutic endoscopic practice, a literature search yielded only a few published reports on the use of this technique [7–9, 15–22]. However, the endoloop technique is very useful for removal of large pedunculated lesions [19]. Rey and Marek [7], in their preliminary results from the use of endoloop for prevention of postpolypectomy bleeding in 15 patients with large pedunculated polyps, reported no blood flow immediately after polypectomy, and no delayed bleeding at a 1-month follow-up assessment. In a randomized trial of endoloop application involving 89 patients with pedunculated colorectal polyps 1 cm in diameter, postpolypectomy bleeding occurred in none of the 47 patients assigned to the endoloop arm, and in only 5 of the 42 patients (12%) in the control group (1 immediate and 4 delayed). Moreover, the use of an endoloop reduced the duration of hospitalization after polypectomy [8]. Matsushita et al. [9] reported their experience using the endoloop for colonoscopic polypectomy of 20 large pedunculated polyps. In three patients, the loop slipped off after polypectomy because the polyps were cut close to the site of encirclement, but no bleeding ensued. Bleeding occurred in four cases because the loop transected the 4-mm-thick stalk (1 case), because the loop slipped (1 case), and because the loop was insufficiently tightened (2 cases). A few case reports described use of the endoloop for other gastointestinal polyps including a large juvenile polyp [17], a submucosal colonic lymphangioma [20], large colonic lipomas [18, 21], and bleeding mesenteric vessels [22]. Our study is the second largest in the number of patients with large pedunculated polyps for whom the endoloop was used to ligate the stalk prophylactically before resection. One important point of the current series is that it showed the advantages and disadvantages of the device. In four patients, the polyps were located in the sigmoid colon with a narrow lumen be-

cause of diverticulosis, making application of the endoloop impossible. These polyps were excised with piecemeal polypectomy after injection of 50% dextrose plus epinephrine (1:10000) solution in the stalk. This appears to be a safe and practical technique for endoscopic resection of large sessile colorectal polyps [23]. In contrast to the conventional metal snare, endoloops have limited expansible force and stiffness because of their thin nylon composition. Therefore, it may be challenging to place the endoloop in circumstances offering limited space for it to open fully, either due to the large size of the target lesion or due to the narrow luminal diameter. This problem is rarely encountered in the right colon, but can cause significant technical difficulty in the left colon, particularly in the setting of severely stenotic diverticular disease. When the endoloop cannot be maneuvered over the polyp, then, before polypectomy, we recommend either injection of dilute epinephrine (1:10000) plus dextrose 50% into the stalk or clipping of the stalk with long clips to diminish the risk of bleeding. In two patients, the stalks (£ 4 mm) were transected when the endoloop was fully tightened. We believe that pedunculated polyps with thin stalks (£ 4 mm) are a contraindication for this technique, whereas pedunculated polyps with stalks 5 mm or thicker are well-suited lesions. Comparable conclusions also have been reported by others [9]. In two of our patients, the endoloop slipped off after polypectomy, with delayed bleeding occurring on days 6 and 7, respectively. These bleeding events coincided with the fact that the attached loop usually sloughs in 4 to 7 days. The bleeding was successfully treated with the application of clips. For safe transection of large pedunculated polyps, an adequate distance from the ligated loop must be kept during cutting to avoid slippage of the loop. We also treated one patient in whom the polypectomy snare became entangled with the endoloop in a way that precluded transection of the peduncle. The polypÕs head was excised by piecemeal resection. Considering the aforementioned findings, the following conclusions can be drawn about avoiding postpolypectomy bleeding by means of an endoloop technique. The type of lesion chosen for endoloop application appears to be essential. Furthermore, underlying colonic disease and polyps with extremely thin stalks and semipedunculated lesions are inappropriate for endoloop placement. If the peduncle is 4 mm or less in diameter, it can be transected by a sufficiently tightened loop. It is doubtful that a loop is useful in this situation. Moreover, for broad semipedunculated lesions, closure of the loop beneath the planned resection margin may not be possible. In summary, the results of this study suggest that although colonoscopic resection of large pedunculated polyps using an endoloop is safe and effective, the problems encountered with this method show some of the limitations encountered with endoscopic therapy in preventing postpolypectomy bleeding when currently available instruments are used.

1261

References 1. Akahoshi K, Kojima H, Fujimaru T, Kondo A, Kubo S, Furuno T, Nakanishi K, Harada N, Nawata H (1999) Grasping forceps assisted endoscopic resection of large pedunculated GI polypoid lesions. Gastrointest Endosc 50: 95–98 2. Brandimarte G, Tursi A (2001) Endoscopic snare excision of large pedunculated colorectal polyps: a new, safe, and effective technique. Endoscopy 33: 854–857 3. Cipolletta L, Bianco MA, Rotondano G, Catalano M, Prisco A, De Simone T (1999) Endoclip-assisted resection of large pedunculated colon polyps. Gastrointest Endosc 50: 405–406 4. Rex DK, Lewis BS, Waye JD (1992) Colonoscopy and endoscopic therapy for delayed postpolypectomy hemorrhage. Gastrointest Endosc 38: 121–129 5. Vernava AM III, Longo WE (1996) Complication of endoscopic polypectomy. Surg Oncol Clin North Am 5: 663–673 6. Singaram C, Torbey CF, Jacoby RF (1995) Delayed postpolypectomy bleeding. Am J Gastroenterol 90: 146–147 7. Rey JF, Marek TA (1997) Endoloop in the prevention of the postpolypectomy bleeding: preliminary results. Gastrointest Endosc 46: 387–389 8. Iishi H, Tatsuta M, Narahara H, Iseki K, Sakai N (1996) Endoscopic resection of large pedunculated colorectal polyps using a detachable snare. Gastrointest Endosc 44: 594–597 9. Matsushita M, Hajiro K, Takakuwa H, Kusumi F, Maruo T, Ohana M, Tominaga M, Okano A, Yunoki Y (1998) Ineffective use of a detachable snare for colonoscopic polypectomy of large polyps. Gastrointest Endosc 47: 496–499 10. Rosen L, Bub DS, Reed JF III, Nastasee SA (1993) Hemorrhage following colonoscopic polypectomy. Dis Colon Rectum 36: 1126– 1131 11. Van Gossum A, Cozzoli A, Adler M, Taton G, Cremer M (1992) Colonoscopic snare polypectomy: analysis of 1,485 resections comparing two types of current. Gastrointest Endosc 38: 472–475 12. Fruhmorgen P, Demling L (1979) Complications of diagnostic and therapeutic colonoscopy in the Federal Republic of Germany: results of an inquiry. Endoscopy 11: 146–150

13. Pontecorvo C, Pesce G (1986) The ‘‘safety snare’’: a ligatureplacing snare to prevent haemorrhage after transection of large pedunculated polyps. Endoscopy 18: 55–56 14. Hachisu T (1991) A new detachable snare for hemostasis in the removal of large polyps or other elevated lesions. Surg Endosc 5: 70–74 15. Pietropaolo V, Bogliolo G, Lande A, Gerardi A, Ierfone N, Giannacco A, Colona R, Pontone P (1999) Endoscopic ligation in the prevention of hemorrhage after polypectomy: our experience. G Chir 20: 51–52 16. Pai KP, Morgan RH (1995) Endoloop: a helpful new device for transanal rectal polypectomy. Br J Surg 82: 748 17. Matsushita M, Hajiro K, Okazaki K, Takakuwa H, Nishio A (2000) A large juvenile polyp in a 1-year-old child safely removed by colonoscopic polypectomy with a detachable snare. Gastrointest Endosc 52: 118–120 18. Murray MA, Kwan V, Williams SJ, Bourke MJ (2005) Detachable nylon loop–assisted removal of large clinically significant colonic lipomas. Gastrointest Endosc 61: 756–759 19. Bourke MJ, Williams SJ (2003) Clips, loops, and bands: applications in the colon. In: Waye JD, Rex DK, Williams CB (eds). Colonoscopy principles and practice. Blackwell Publishing: Oxford, UK, pp 287–294 20. Ajiki T, Nakamura T, Kubota S, Kim YS, Kuroda Y (2000) Successful endoscopic treatment colon lymphangioma with a ligating device. Gastrointest Endosc 52: 800–802 21. Kim CY, Bandres D, Lok Tio T, Benjamin SB, Al-Kawas FH (2002) Endoscopic removal of large colonic lipomas. Gastrointest Endosc 55: 929–931 22. Hepworth CC, Kadirkamanathan SS, Gong F, Swain CP (1998) A randomised controlled comparison of injection, thermal, and mechanical endoscopic methods of haemostasis on mesenteric vessels. Gut 42: 462–469 23. Katsinelos P, Kountouras J, Paroutoglou G, Zavos C, Rizos C, Beltsis A (2006) Endoscopic mucosal resection of large sessile colorectal polyps with submucosal injection of hypertonic 50percent dextroseepinephrine solution. Dis Colon Rectum 49: (in press)