- Email: [email protected]
Tu1525 “Clip and Loop” Technique for Endoscopic Resection of Large Pedunculated Polyps
Abstracts
incomplete colonoscopy during the 4-month time period. The majority of patients were over age 50 (n⫽108), undergoing colonoscopy for polyp detection (n⫽112), and were outpatients (80.1%). Procedures were performed by 22 unique endoscopists (range 1-18 incomplete procedures per physician). The most common reasons for incomplete colonoscopy were endoscope looping (n⫽47), inadequate bowel preparation (n⫽32), tortuosity (n⫽25), and inadequate sedation (n⫽8). In 23 procedure notes (19%), no recommendation was made for complete colon evaluation (CCE). Specific recommendations were enumerated in 98 procedure notes (81.0%), with radiologic testing (n⫽53) and repeat colonoscopy with extended bowel preparation (n⫽21) most common. CCE was recommended immediately in 71 patients, within 1 year in 3 patients, and after 1 year in 24 patients. Incomplete procedures in which polyps were identified were more likely to have specified recommendations for CCE (95% vs. 78.4%, p⫽0.12) and for CCE to be recommended within 1 year (73.6% vs. 56.8%, p⫽0.2) compared to procedures without polyps seen. Incomplete procedures due to poor preparation were more likely to recommend CCE utilizing endoscopy than incomplete procedures due to endoscope looping/colon angulation (75.0% vs. 7.0%, p ⬍ .0001). Procedures in which the extent was distal to the hepatic flexure were more likely to have CCE recommended within 1 year compared to extent proximal to the hepatic flexure (69.1% vs. 50.9%, p⫽.06). In total, 57/74 patients adhered to the recommendation of CCE within 12 months. Adherence was similar regardless of modality of CCE recommended, inpatient/outpatient status, polyps on initial exam, or extent of initial exam. There was a trend towards lower adherence rates in African American’s (64.3%) compared to Caucasians (83.0%, p⫽0.26). Conclusions: Colonoscopists are more likely to recommend CCE on examinations that do not reach the hepatic flexure and when polyps are seen on initial incomplete exam. However, after incomplete colonoscopy, nearly 25% of patients do not undergo evaluation of the remaining colon within 12 months despite physician recommendation.
Tu1522 Can Hyperplasic-Serrated Lesions and Adenomatous Lesions Be Differentiated During a Conventional Colonoscopy? Predictive Features Based on Endoscopic Characteristics José M. Mella, Carolina Fischer, Raquel GonzáLez, Lisandro Pereyra, Guillermo Nicolás Panigadi, Adriana Mohaidle, Pablo Luna, Sandra Lencinas, Silvia C. Pedreira, Daniel G. Cimmino, Luis A. Boerr Hospital Alemán, Buenos Aires, Argentina Introduction. Colonic adenomas are well known as colorectal cancer precursors. Although serrated lesions are currently considered to present an increased risk of cancer, they are macroscopically similar to hyperplasic polyps, and therefore are usually mistaken. Aims: To analyze the endoscopic characteristics of the most frequent colonic lesions (hyperplasic, serrated and adenomas) using white-light colonoscopy, and to find out endoscopic predictors of hyperplasic-serrated lesions, which may help to differentiate them from adenomatous lesions. Material and methods: We analyzed all resected polyps in every colonoscopy performed in our Unit between May and August 2010. The following characteristics of colonic lesions were analyzed: morphology (sessile or flat vs pedunculated), size (⬍1cm vs ⱖ1cm), localization (right vs left colon), presence of mucus covering the lesion, and some combinations of this characteristics. The diagnosis of hyperplasic, serrated and adenomatous lesions was determined histopathologically. Hyperplasic and serrated lesions were analyzed together as a composite endpoint and compared with adenomatous lesions. Endoscopic characteristics significantly related to hyperplasic-serrated lesions were identified by univariate analysis (considering significant an odds ratio (OR) ⬍or⬎1 which confidence intervals (CI) 95% were ⫽ 1). Independent predictors for this group of lesions were also analyzed using a binary logistic regression model. Results: We prospectively analyzed 256 colonic lesions. Most of them were sessile (83%) and small (less than 1 cm, 77%); 52% were in the right colon and 12% had mucus on their surface. Histopathologically, 45% were hyperplasic-serrated lesions, and 55% were adenomas. By univariate analysis, we identified the following endoscopic characteristics to be associated with the diagnosis of hyperplastic-serrated lesions: location in the right colon OR 1.77 (CI 1.07-2.93), the presence of mucus OR 6.82 (CI 2.69-17.25), a flat or sessile morphology OR 11.2 (CI 1.44-87), right colon location and having mucus OR 8.88 (CI 2.97-26), the presence of mucus in lesions ⱖ 1 cm located in the right colon OR 12.3 (CI 1.53-98). By multivariate analyses, the only endoscopic feature independently associated with hyperplasic-serrated histology was the presence of mucus covering the lesion: OR 5.31 (CI 2.04-13.85). Discussion. These endoscopic characteristics, which can be easily obtained during a white-light colonoscopy, could be useful to identify hyperplasic-serrated lesions, and to encourage the pathologists to look for serrated features.
Tu1523 Evaluation of Subepithelial Lesion of the Appendix by Endoscopic Ultrasound Tae Hee Lee, Jin-Oh Kim, Hyun Gun Kim, Wan Jung Kim, Won Young Cho, Bong Min Ko, Joo Young Cho, Joon Seong Lee, Moon Sung Lee Institute for Digestive Research, Soonchunhyang University, Seoul, Republic of Korea Background: The use of through-the-scope (TTS) miniprobe catheter endoscopic ultrasound is a valuable technique for evaluating subepithelial lesions in the proximal colon. Few reports include the evaluation of the appendix by EUS. Objective: To evaluate the performance characteristics of EUS in the diagnosis of periappendiceal subepithelial lesions. Methods: Retrospective case series in a single academic medical center. Adult patients referred for evaluation of subepithelial lesions of the appendix identified by colonoscopy between January, 2004 to July, 2010. Data were abstracted from an endoscopic ultrasound database for all patients undergoing miniprobe endoscopic ultrasound examination of the appendix. Results: Fifteen cases were identified. Nine (60%) patients were male. EUS corrected presumptive endoscopic diagnosis in two cases(13%). A total of two cases were identified as being extrinsic compression of the small bowel. Presumptive diagnosis became to be certain in 10 cases(67%) after EUS exam. Surgery was needed in 5 cases on the basis of the presumptive EUS diagnosis. A total of 3 cases were confirmed histologically by surgery. Conclusions: EUS evaluation can improve diagnostic certainty of subepithelial lesions of the appendix. Furthermore it may assist in the selection of patients who may benefit from surgical management in the patients with subepithelial lesions of the appendix
Tu1524 The Clinical Significance and Etiology of Lymphoid Follicular Proctitis Junghyun Lee1, Byung Kook Kim1, So Dug Lim2, Kim Jeong Hwan1, Lee Sun-Young1, In Kyung Sung1, Hyung Seok Park1, Choon-Jo Jin1, Shim Chan Sup1 1 Digestive Disease Center, Konkuk University Medical Center, Seoul, Republic of Korea; 2Department of pathology, Konkuk University Medical Center, Seoul, Republic of Korea Background/aims: Lymphoid follicular proctitis is an uncommon inflammatory condition confined to the rectal mucosa. The aim of this study was to identify disease specific clinical, endoscopic, and histopathological findings, to aid the effective differential diagnosis of lymphoid follicular proctitis in the general population. Methods: This was a retrospective analysis of patients from our database who undergo colonoscopy and sigmoidoscopy by two expert endoscopists at Konkuk university hospital from May 2009 to July 2010. Diagnosis of lymphoid follicular proctitis was only based on endoscopic finding that is a striking finely granular pattern without erosions or ulcerations on the rectal mucosa. And then the clinical features, endoscopic findings and pathologic findings of 12 patients with lymphoid follicular proctitis were reviewed. Results: One thousand thirty nine patients underwent colonoscopy or sigmoidoscopy for 14 months. One hundred and three cases were excluded due to previous large bowel surgery, pancolitis, and a prior history of inflammatory bowel disease. Twenty patients(0.72%) were diagnosed as lymphoid follicular proctitis among the 1636 examinations. A retrospective review of lymphoid follicular proctitis showed the commonest cause is Chlamydia trachomatis infection(7 patients, 0.43%). This was followed by taking lipid lowering agent in 0.12%(2 patients), unknown etiology in 0.12%(2 patients), and allergic disorder in 0.06%(1 patient). Chlamydia proctitis is the most common cause of the lymphoid follicular proctitis in general population. Chlamydia proctitis is significantly related with pelvic inflammatory disease (PID) and/or Fitz-Hugh Curtis (FHC) syndrome. Six of 7 patients with chlamydia proctitis were successfully treated by oral azithromycin. Two patients with non-chlamydia proctitis taking lipid lowering agent were healed after switching to other lipid lowering agent. Conclusion: Lymphoid follicular proctitisis a rare disorder in non-homosexual person. Chlamydia infection is the most common cause of lymphoid follicular proctitis, especially in women of childbearing age who have PID and/or FHC syndrome.
Tu1525 “Clip and Loop” Technique for Endoscopic Resection of Large Pedunculated Polyps Pietro Fusaroli, Stefania Ferri, Stefania Fiorentini, Francesca Serotti, Mariangela Tattini, Rossella Callegari, Giancarlo G. Caletti University of Bologna; Dept. of Gastroenterology, Castel S. Pietro Terme, Italy Background: Colonoscopic polypectomy of large pedunculated polyps may be associated with complications such as bleeding. Use of a detachable snare (endoloop) may reduce the risk of bleeding but its application may be
AB436 GASTROINTESTINAL ENDOSCOPY Volume 73, No. 4S : 2011
www.giejournal.org
Abstracts
cumbersome due to entrapment in the head of the polyp and risk of slipping off immediately after polypectomy. The latter is particularly frequent there is not enough room for endoloop application due to a short polyp stalk. We report about the use of a simple and safe method combining the use of clips and endoloop. Methods: Patients with large pedunculated polyps (⬎3 cm) were included. We used a prototype therapeutic sigmoidoscope (Olympus, Tokyo, Japan), 80 cm long, equipped with two working channels (2.8 mm and 3.7 mm, respectively). One working channel has an up/down elevator. Using a “twohanded” approach, 2 clips were initially placed at the base of the stalk, on two opposite sides. In order to fix it securely, the endoloop was then placed at the base of the stalk below the two clips. To facilitate the procedure the endoloop was opened through one channel, the polyp was grabbed through the other channel by a tripod and then pulled inside the endoloop. Thus, the endoloop was subsequently firmly secured at the base of the stalk. If necessary, the same procedure was repeated to capture the polyp within the diathermic snare for resection. The procedures were performed under conscious sedation using meperidine and midazolam. Results: Twenty-five pedunculated polyps in 24 patients (13 F, 11 M; median age 59) were resected with the “clip and loop” technique. It was possible to place an endoloop successfully below the clips at the basis of the stalk and perform a resection in all cases. No major complication occurred. Particularly, in no case the endoloop fell off after resection. In comparison to the previous cases performed with the endoloop only, the success rate in terms of correct and durable endoloop placement was higher (100% vs. 84%). Conclusions: Application of endoloop below the clips is feasible and safe and prevents slipping off of the endoloop after polypectomy. Use of a combination of clips and endoloop is an effective prophylactic measure to prevent bleeding after polypectomy of large pedunculated polyps, also when the stalk is short.
Tu1526 Endoscopic Submucosal Dissection (ESD) up to 11 cm for 54 Large Laterally Spreading Tumors or Local Recurrencies in the Recto-Sigmoid Juergen Hochberger, Guenter Wilhelms, Martin Froelich, Detlev Menke, Edris Wedi, Karl-Friedrich Buerrig, Elena Kruse Gastroenterology, Intervent. Endoscopy, St. Bernward Hospital, Hildesheim, Germany Piecemeal-EMR of lateral spreading tumors (LST) in the colo-rectum over 2 cm is associated with a recurrence rate of about 15% (3-32%). Adenomata have a sizedependent progressive risk of malignant transformation especially over 2 cm. ESD provides an ’en bloc’ specimen allowing a precise vertical and lateral histopathological evaluation and promises a low recurrence rate (0-3%). However, ESD experience in the ’Western World’ is limited compared to published large Japanese series.Patients and Method: March 28, 2006 to November 11, 2010 52 patients were treated with 54 ESDs (1x 3ESDs): 71 y (4990 y); 19 f, 34 m; ASA 2-4 with a flat or elevated polyp in the recto-sigmoid over 2 cm. In 1 patient 3 lesions were resected in 2 ESD sessions. 11/52 patients (21%) had a previous local treatment (7x EMR; 4x surgical full-thickness resection). Location: 31/54 (57%) in the distal, 16 (30%) in the middle and 7(13%) in the proximal rectum or sigma. Results: 46 of 54 lesions could be resected macroscopically ’en bloc’ (85%); histologically a clear adenoma/tumor-free margin was present in 34/54 (63%). In 5 ESDs (9%) a clear statment was not possible due to termal artefacts. The macro-pathological size of the formalinfixed specimen pinned on kork ranged from 2.2 cm to 11 cm for the largest diameter (48/54 specimen ⱖ3 cm)Histopathology revealed foci of adenocarcinoma in 6/54 (11%), foci of a neuroendocrine tumor in 1 (1.85%; net), lymphoma in 1. In 46/54 specimen histology showed tubulovillous (44) or villous (2) adenomata (37% LG-IEN; 24/44⫽ 44% HG-IEN). 32/54 (59%) of all lesions showed HG-IEN or cancer/lym/net. However a referral bias seems probable.Median follow-up was 301 days (20-1,708 days). Two patients died unrelated to the resection.4 pT1 cancers underwent surgery due L⫹; V⫹/G3; R1 (2 cases). The neuroendocrine tumor due to V⫹, R1 G3. All other cases were followed conservatively by endoscopy and telephone interviews. Complications: 3 perforations due to wall damage during or following ESD (5.6 %) managed conservatively. In one case ESD was accomponied by a rare colonic methan gas explosion and the patient operated for safety reasons (prior ESD specimen R0). 5 minor bleedings (9%) occured, no patient needed blood transfusion (0%). Proven recurrence macroscopically and in local biopsies was 0%. In one case (1.85%) a secondary 6 mm lesion close to the scar was found after 12 month and resected (R0, LGD tv adenoma). Conclusions: ESD up to 11 cm for large mucosal lesions or local recurrencies after EMR/ surgical full-thickness resection in the rectosigmoid is feasable in the western endoscopic world. A rate of almost 2/3 of HGIEN or early malignancies in our series highlights the importance of a histopathological ’en bloc’ specimen for rectal lesions ⬎3 cm. The low local recurrence rate of 0-3% underlines the potential value of the technique.
www.giejournal.org
Tu1527 The Detection of Polyps and Advanced Lesions in HIV Infected Individuals During Colonoscopy Alexander Schlachterman, Ayaz Matin, David Oustecky, Scott Naples, Daniel A. Ringold Medicine, Drexel University College of Medicine, Philadelphia, PA Background: As HIV infected patients live longer, colonoscopies for both screening and symptomatic patients are performed more often. Compared to controls, non-AIDS defining malignancies are more prevalent in HIV patients. It is not known if the development of colonic neoplasia occurs at a higher rate or at a younger age in HIV infected patients. Methods: We conducted a retrospective study to investigate the prevalence of polyps in HIV patients detected on colonoscopy from 2000-10 at an inner city tertiary care center. These patients underwent colonoscopy either for screening, surveillance or symptoms. Age, sex, symptoms and CD4 count were recorded. Size and histology of both benign and neoplastic polyps were collected. Advanced neoplasms were defined as an adenoma ⬎10mm or any lesion with high-grade dysplasia(HGD), villous histology or carcinoma. A mixed control group without HIV was used for comparison. Results: A total of 319 HIV patients underwent colonoscopy; the mean age of these patients was 51.5 y and 92 were over 50y. There were 209 men and 110 women. 31 of 209 men and 14 of 110 women had polyps(17.4% vs. 12.7%, p⫽ 0.73). The overall neoplastic polyp detection rate in HIV patients was 6.6%, compared to 27.9% in the control group(p⫽⬍0.01 ). Advanced neoplastic lesions were seen in 11 HIV patients(3.5%). In the subgroup of those patients with neoplastic lesions, 52.4%(11/21) had advanced neoplasms, including 4 cancers and 1 HGD. In comparison, the control group had 15/48 advanced neoplasms(31.3%, p⫽0.11). A total of 93 polyps were found in 47 individuals. Of those individuals with any lesion found on colonoscopy, 32% were women and 68% were men(p⫽0.74). 23.4% of neoplasms were found to be advanced neoplasms in the HIV patients compared to 18.3% in the control group(p⫽0.33). The rate of neoplasia detection among patients over 50(n⫽148, mean⫽ 56.7y) was 22.9%. The rate of neoplasia detection in the under 50 group (n⫽137, mean⫽ 43 y) was 9.4%. The preparation was poor for 75 of the 319(23.5%) HIV patients that had colonoscopy in comparison to 9 of 172 (5.2%) patients without HIV (p⬍0.0001). In a subgroup analysis of patients followed in an HIV specialty care clinic, only 29 of 503 patients(5.7%) over 50 received screening colonoscopy. Conclusions: The frequency of polyp detection in HIV patients is not greater than in the control population. However, a greater proportion of advanced neoplasms were seen in HIV infected patients and HIV infected men had more polyps compared to women, although this was not statistically significant. Amongst HIV patients, a low rate of screening and a high rate of poor preparation were observed in our study. Both of these observations are potential areas for quality improvement.
Tu1528 Missed Colon Cancers: Difference Between Right vs TransverseLeft Colon. Five Year Survey in a Community Hospital Francisco Igea, Sergio Maestro, Antonio Perez Millan, Henar Nuñez, Rosa Eva Madrigal, Esther Saracibar, Javier Barcenilla Complejo Hospitalario de Palencia, Palencia, Spain Background: Missed colon cancers are nowadays a great concern in gastrointestinal endoscopy. Colonoscopy seems to be less protective in rigth colon cancers that in left-transverse side. Some authors denied protective effect of colonoscopy in right colon cancers (1). Aims: To determine if there is any difference between right and Left colon in terms of missed colon cancers. To determine if there is any prottective effect of colonoscopy against right colon cancer by comparing frequence of previous colonoscopy with a healthy control group. Methods: We reviewed all colon cancers diagnosed in our hospital between Jan 2005-Jan 2010. Right was defined by location of tumour between appendix and hepatic angle ( both included), all others were considered lefttransverse. All previous colonoscopies during a 10 year period before diagnosis in those patients were extracted from our database.Our hospital is the only public institution able to do colonoscopy in our province and all endoscopic procedures are collected in a computorized data base since 1995. A control group of patients with a”clean colonoscopy”⫽(No cancer) during 2009 were matched by sex and age 2:1 with the study group. We extracted also all previous colonoscopies in this control group. Missed colon cancer was defined by a previous colonoscopy done within the 3 years period before diagnosis of cancer as in Ontario group(2) Results: 726 colon cancers were found. 416 males/310 females.206 were Right, 510 Left & 10 sinchronic. 1454 controls were extracted. Frequence of previous colonoscopies are summarized in table 1. 2,5% of the patients had a missed colon cancer. There were 4,6% in the right colon vs 1,7% in left transverse P⬍0,05. Previous colonoscopy rate in healthy controls was significantly higher 10,2% p⬍0,05 Conclusions: 1) Missed colon cancers are considerably more frequent in right colon (more than double) 2) It is mandatory to make changes in our practice to improve detection of flat/minimal lesions in right colon 3) Although previous colonoscopy was more frequent in right colon cancer group than in other locations there is still a difference with healthy controls. Protective effect of colonoscopy seems reasonable also in this
Volume 73, No. 4S : 2011
GASTROINTESTINAL ENDOSCOPY
AB437
incomplete colonoscopy during the 4-month time period. The majority of patients were over age 50 (n⫽108), undergoing colonoscopy for polyp detection (n⫽112), and were outpatients (80.1%). Procedures were performed by 22 unique endoscopists (range 1-18 incomplete procedures per physician). The most common reasons for incomplete colonoscopy were endoscope looping (n⫽47), inadequate bowel preparation (n⫽32), tortuosity (n⫽25), and inadequate sedation (n⫽8). In 23 procedure notes (19%), no recommendation was made for complete colon evaluation (CCE). Specific recommendations were enumerated in 98 procedure notes (81.0%), with radiologic testing (n⫽53) and repeat colonoscopy with extended bowel preparation (n⫽21) most common. CCE was recommended immediately in 71 patients, within 1 year in 3 patients, and after 1 year in 24 patients. Incomplete procedures in which polyps were identified were more likely to have specified recommendations for CCE (95% vs. 78.4%, p⫽0.12) and for CCE to be recommended within 1 year (73.6% vs. 56.8%, p⫽0.2) compared to procedures without polyps seen. Incomplete procedures due to poor preparation were more likely to recommend CCE utilizing endoscopy than incomplete procedures due to endoscope looping/colon angulation (75.0% vs. 7.0%, p ⬍ .0001). Procedures in which the extent was distal to the hepatic flexure were more likely to have CCE recommended within 1 year compared to extent proximal to the hepatic flexure (69.1% vs. 50.9%, p⫽.06). In total, 57/74 patients adhered to the recommendation of CCE within 12 months. Adherence was similar regardless of modality of CCE recommended, inpatient/outpatient status, polyps on initial exam, or extent of initial exam. There was a trend towards lower adherence rates in African American’s (64.3%) compared to Caucasians (83.0%, p⫽0.26). Conclusions: Colonoscopists are more likely to recommend CCE on examinations that do not reach the hepatic flexure and when polyps are seen on initial incomplete exam. However, after incomplete colonoscopy, nearly 25% of patients do not undergo evaluation of the remaining colon within 12 months despite physician recommendation.
Tu1522 Can Hyperplasic-Serrated Lesions and Adenomatous Lesions Be Differentiated During a Conventional Colonoscopy? Predictive Features Based on Endoscopic Characteristics José M. Mella, Carolina Fischer, Raquel GonzáLez, Lisandro Pereyra, Guillermo Nicolás Panigadi, Adriana Mohaidle, Pablo Luna, Sandra Lencinas, Silvia C. Pedreira, Daniel G. Cimmino, Luis A. Boerr Hospital Alemán, Buenos Aires, Argentina Introduction. Colonic adenomas are well known as colorectal cancer precursors. Although serrated lesions are currently considered to present an increased risk of cancer, they are macroscopically similar to hyperplasic polyps, and therefore are usually mistaken. Aims: To analyze the endoscopic characteristics of the most frequent colonic lesions (hyperplasic, serrated and adenomas) using white-light colonoscopy, and to find out endoscopic predictors of hyperplasic-serrated lesions, which may help to differentiate them from adenomatous lesions. Material and methods: We analyzed all resected polyps in every colonoscopy performed in our Unit between May and August 2010. The following characteristics of colonic lesions were analyzed: morphology (sessile or flat vs pedunculated), size (⬍1cm vs ⱖ1cm), localization (right vs left colon), presence of mucus covering the lesion, and some combinations of this characteristics. The diagnosis of hyperplasic, serrated and adenomatous lesions was determined histopathologically. Hyperplasic and serrated lesions were analyzed together as a composite endpoint and compared with adenomatous lesions. Endoscopic characteristics significantly related to hyperplasic-serrated lesions were identified by univariate analysis (considering significant an odds ratio (OR) ⬍or⬎1 which confidence intervals (CI) 95% were ⫽ 1). Independent predictors for this group of lesions were also analyzed using a binary logistic regression model. Results: We prospectively analyzed 256 colonic lesions. Most of them were sessile (83%) and small (less than 1 cm, 77%); 52% were in the right colon and 12% had mucus on their surface. Histopathologically, 45% were hyperplasic-serrated lesions, and 55% were adenomas. By univariate analysis, we identified the following endoscopic characteristics to be associated with the diagnosis of hyperplastic-serrated lesions: location in the right colon OR 1.77 (CI 1.07-2.93), the presence of mucus OR 6.82 (CI 2.69-17.25), a flat or sessile morphology OR 11.2 (CI 1.44-87), right colon location and having mucus OR 8.88 (CI 2.97-26), the presence of mucus in lesions ⱖ 1 cm located in the right colon OR 12.3 (CI 1.53-98). By multivariate analyses, the only endoscopic feature independently associated with hyperplasic-serrated histology was the presence of mucus covering the lesion: OR 5.31 (CI 2.04-13.85). Discussion. These endoscopic characteristics, which can be easily obtained during a white-light colonoscopy, could be useful to identify hyperplasic-serrated lesions, and to encourage the pathologists to look for serrated features.
Tu1523 Evaluation of Subepithelial Lesion of the Appendix by Endoscopic Ultrasound Tae Hee Lee, Jin-Oh Kim, Hyun Gun Kim, Wan Jung Kim, Won Young Cho, Bong Min Ko, Joo Young Cho, Joon Seong Lee, Moon Sung Lee Institute for Digestive Research, Soonchunhyang University, Seoul, Republic of Korea Background: The use of through-the-scope (TTS) miniprobe catheter endoscopic ultrasound is a valuable technique for evaluating subepithelial lesions in the proximal colon. Few reports include the evaluation of the appendix by EUS. Objective: To evaluate the performance characteristics of EUS in the diagnosis of periappendiceal subepithelial lesions. Methods: Retrospective case series in a single academic medical center. Adult patients referred for evaluation of subepithelial lesions of the appendix identified by colonoscopy between January, 2004 to July, 2010. Data were abstracted from an endoscopic ultrasound database for all patients undergoing miniprobe endoscopic ultrasound examination of the appendix. Results: Fifteen cases were identified. Nine (60%) patients were male. EUS corrected presumptive endoscopic diagnosis in two cases(13%). A total of two cases were identified as being extrinsic compression of the small bowel. Presumptive diagnosis became to be certain in 10 cases(67%) after EUS exam. Surgery was needed in 5 cases on the basis of the presumptive EUS diagnosis. A total of 3 cases were confirmed histologically by surgery. Conclusions: EUS evaluation can improve diagnostic certainty of subepithelial lesions of the appendix. Furthermore it may assist in the selection of patients who may benefit from surgical management in the patients with subepithelial lesions of the appendix
Tu1524 The Clinical Significance and Etiology of Lymphoid Follicular Proctitis Junghyun Lee1, Byung Kook Kim1, So Dug Lim2, Kim Jeong Hwan1, Lee Sun-Young1, In Kyung Sung1, Hyung Seok Park1, Choon-Jo Jin1, Shim Chan Sup1 1 Digestive Disease Center, Konkuk University Medical Center, Seoul, Republic of Korea; 2Department of pathology, Konkuk University Medical Center, Seoul, Republic of Korea Background/aims: Lymphoid follicular proctitis is an uncommon inflammatory condition confined to the rectal mucosa. The aim of this study was to identify disease specific clinical, endoscopic, and histopathological findings, to aid the effective differential diagnosis of lymphoid follicular proctitis in the general population. Methods: This was a retrospective analysis of patients from our database who undergo colonoscopy and sigmoidoscopy by two expert endoscopists at Konkuk university hospital from May 2009 to July 2010. Diagnosis of lymphoid follicular proctitis was only based on endoscopic finding that is a striking finely granular pattern without erosions or ulcerations on the rectal mucosa. And then the clinical features, endoscopic findings and pathologic findings of 12 patients with lymphoid follicular proctitis were reviewed. Results: One thousand thirty nine patients underwent colonoscopy or sigmoidoscopy for 14 months. One hundred and three cases were excluded due to previous large bowel surgery, pancolitis, and a prior history of inflammatory bowel disease. Twenty patients(0.72%) were diagnosed as lymphoid follicular proctitis among the 1636 examinations. A retrospective review of lymphoid follicular proctitis showed the commonest cause is Chlamydia trachomatis infection(7 patients, 0.43%). This was followed by taking lipid lowering agent in 0.12%(2 patients), unknown etiology in 0.12%(2 patients), and allergic disorder in 0.06%(1 patient). Chlamydia proctitis is the most common cause of the lymphoid follicular proctitis in general population. Chlamydia proctitis is significantly related with pelvic inflammatory disease (PID) and/or Fitz-Hugh Curtis (FHC) syndrome. Six of 7 patients with chlamydia proctitis were successfully treated by oral azithromycin. Two patients with non-chlamydia proctitis taking lipid lowering agent were healed after switching to other lipid lowering agent. Conclusion: Lymphoid follicular proctitisis a rare disorder in non-homosexual person. Chlamydia infection is the most common cause of lymphoid follicular proctitis, especially in women of childbearing age who have PID and/or FHC syndrome.
Tu1525 “Clip and Loop” Technique for Endoscopic Resection of Large Pedunculated Polyps Pietro Fusaroli, Stefania Ferri, Stefania Fiorentini, Francesca Serotti, Mariangela Tattini, Rossella Callegari, Giancarlo G. Caletti University of Bologna; Dept. of Gastroenterology, Castel S. Pietro Terme, Italy Background: Colonoscopic polypectomy of large pedunculated polyps may be associated with complications such as bleeding. Use of a detachable snare (endoloop) may reduce the risk of bleeding but its application may be
AB436 GASTROINTESTINAL ENDOSCOPY Volume 73, No. 4S : 2011
www.giejournal.org
Abstracts
cumbersome due to entrapment in the head of the polyp and risk of slipping off immediately after polypectomy. The latter is particularly frequent there is not enough room for endoloop application due to a short polyp stalk. We report about the use of a simple and safe method combining the use of clips and endoloop. Methods: Patients with large pedunculated polyps (⬎3 cm) were included. We used a prototype therapeutic sigmoidoscope (Olympus, Tokyo, Japan), 80 cm long, equipped with two working channels (2.8 mm and 3.7 mm, respectively). One working channel has an up/down elevator. Using a “twohanded” approach, 2 clips were initially placed at the base of the stalk, on two opposite sides. In order to fix it securely, the endoloop was then placed at the base of the stalk below the two clips. To facilitate the procedure the endoloop was opened through one channel, the polyp was grabbed through the other channel by a tripod and then pulled inside the endoloop. Thus, the endoloop was subsequently firmly secured at the base of the stalk. If necessary, the same procedure was repeated to capture the polyp within the diathermic snare for resection. The procedures were performed under conscious sedation using meperidine and midazolam. Results: Twenty-five pedunculated polyps in 24 patients (13 F, 11 M; median age 59) were resected with the “clip and loop” technique. It was possible to place an endoloop successfully below the clips at the basis of the stalk and perform a resection in all cases. No major complication occurred. Particularly, in no case the endoloop fell off after resection. In comparison to the previous cases performed with the endoloop only, the success rate in terms of correct and durable endoloop placement was higher (100% vs. 84%). Conclusions: Application of endoloop below the clips is feasible and safe and prevents slipping off of the endoloop after polypectomy. Use of a combination of clips and endoloop is an effective prophylactic measure to prevent bleeding after polypectomy of large pedunculated polyps, also when the stalk is short.
Tu1526 Endoscopic Submucosal Dissection (ESD) up to 11 cm for 54 Large Laterally Spreading Tumors or Local Recurrencies in the Recto-Sigmoid Juergen Hochberger, Guenter Wilhelms, Martin Froelich, Detlev Menke, Edris Wedi, Karl-Friedrich Buerrig, Elena Kruse Gastroenterology, Intervent. Endoscopy, St. Bernward Hospital, Hildesheim, Germany Piecemeal-EMR of lateral spreading tumors (LST) in the colo-rectum over 2 cm is associated with a recurrence rate of about 15% (3-32%). Adenomata have a sizedependent progressive risk of malignant transformation especially over 2 cm. ESD provides an ’en bloc’ specimen allowing a precise vertical and lateral histopathological evaluation and promises a low recurrence rate (0-3%). However, ESD experience in the ’Western World’ is limited compared to published large Japanese series.Patients and Method: March 28, 2006 to November 11, 2010 52 patients were treated with 54 ESDs (1x 3ESDs): 71 y (4990 y); 19 f, 34 m; ASA 2-4 with a flat or elevated polyp in the recto-sigmoid over 2 cm. In 1 patient 3 lesions were resected in 2 ESD sessions. 11/52 patients (21%) had a previous local treatment (7x EMR; 4x surgical full-thickness resection). Location: 31/54 (57%) in the distal, 16 (30%) in the middle and 7(13%) in the proximal rectum or sigma. Results: 46 of 54 lesions could be resected macroscopically ’en bloc’ (85%); histologically a clear adenoma/tumor-free margin was present in 34/54 (63%). In 5 ESDs (9%) a clear statment was not possible due to termal artefacts. The macro-pathological size of the formalinfixed specimen pinned on kork ranged from 2.2 cm to 11 cm for the largest diameter (48/54 specimen ⱖ3 cm)Histopathology revealed foci of adenocarcinoma in 6/54 (11%), foci of a neuroendocrine tumor in 1 (1.85%; net), lymphoma in 1. In 46/54 specimen histology showed tubulovillous (44) or villous (2) adenomata (37% LG-IEN; 24/44⫽ 44% HG-IEN). 32/54 (59%) of all lesions showed HG-IEN or cancer/lym/net. However a referral bias seems probable.Median follow-up was 301 days (20-1,708 days). Two patients died unrelated to the resection.4 pT1 cancers underwent surgery due L⫹; V⫹/G3; R1 (2 cases). The neuroendocrine tumor due to V⫹, R1 G3. All other cases were followed conservatively by endoscopy and telephone interviews. Complications: 3 perforations due to wall damage during or following ESD (5.6 %) managed conservatively. In one case ESD was accomponied by a rare colonic methan gas explosion and the patient operated for safety reasons (prior ESD specimen R0). 5 minor bleedings (9%) occured, no patient needed blood transfusion (0%). Proven recurrence macroscopically and in local biopsies was 0%. In one case (1.85%) a secondary 6 mm lesion close to the scar was found after 12 month and resected (R0, LGD tv adenoma). Conclusions: ESD up to 11 cm for large mucosal lesions or local recurrencies after EMR/ surgical full-thickness resection in the rectosigmoid is feasable in the western endoscopic world. A rate of almost 2/3 of HGIEN or early malignancies in our series highlights the importance of a histopathological ’en bloc’ specimen for rectal lesions ⬎3 cm. The low local recurrence rate of 0-3% underlines the potential value of the technique.
www.giejournal.org
Tu1527 The Detection of Polyps and Advanced Lesions in HIV Infected Individuals During Colonoscopy Alexander Schlachterman, Ayaz Matin, David Oustecky, Scott Naples, Daniel A. Ringold Medicine, Drexel University College of Medicine, Philadelphia, PA Background: As HIV infected patients live longer, colonoscopies for both screening and symptomatic patients are performed more often. Compared to controls, non-AIDS defining malignancies are more prevalent in HIV patients. It is not known if the development of colonic neoplasia occurs at a higher rate or at a younger age in HIV infected patients. Methods: We conducted a retrospective study to investigate the prevalence of polyps in HIV patients detected on colonoscopy from 2000-10 at an inner city tertiary care center. These patients underwent colonoscopy either for screening, surveillance or symptoms. Age, sex, symptoms and CD4 count were recorded. Size and histology of both benign and neoplastic polyps were collected. Advanced neoplasms were defined as an adenoma ⬎10mm or any lesion with high-grade dysplasia(HGD), villous histology or carcinoma. A mixed control group without HIV was used for comparison. Results: A total of 319 HIV patients underwent colonoscopy; the mean age of these patients was 51.5 y and 92 were over 50y. There were 209 men and 110 women. 31 of 209 men and 14 of 110 women had polyps(17.4% vs. 12.7%, p⫽ 0.73). The overall neoplastic polyp detection rate in HIV patients was 6.6%, compared to 27.9% in the control group(p⫽⬍0.01 ). Advanced neoplastic lesions were seen in 11 HIV patients(3.5%). In the subgroup of those patients with neoplastic lesions, 52.4%(11/21) had advanced neoplasms, including 4 cancers and 1 HGD. In comparison, the control group had 15/48 advanced neoplasms(31.3%, p⫽0.11). A total of 93 polyps were found in 47 individuals. Of those individuals with any lesion found on colonoscopy, 32% were women and 68% were men(p⫽0.74). 23.4% of neoplasms were found to be advanced neoplasms in the HIV patients compared to 18.3% in the control group(p⫽0.33). The rate of neoplasia detection among patients over 50(n⫽148, mean⫽ 56.7y) was 22.9%. The rate of neoplasia detection in the under 50 group (n⫽137, mean⫽ 43 y) was 9.4%. The preparation was poor for 75 of the 319(23.5%) HIV patients that had colonoscopy in comparison to 9 of 172 (5.2%) patients without HIV (p⬍0.0001). In a subgroup analysis of patients followed in an HIV specialty care clinic, only 29 of 503 patients(5.7%) over 50 received screening colonoscopy. Conclusions: The frequency of polyp detection in HIV patients is not greater than in the control population. However, a greater proportion of advanced neoplasms were seen in HIV infected patients and HIV infected men had more polyps compared to women, although this was not statistically significant. Amongst HIV patients, a low rate of screening and a high rate of poor preparation were observed in our study. Both of these observations are potential areas for quality improvement.
Tu1528 Missed Colon Cancers: Difference Between Right vs TransverseLeft Colon. Five Year Survey in a Community Hospital Francisco Igea, Sergio Maestro, Antonio Perez Millan, Henar Nuñez, Rosa Eva Madrigal, Esther Saracibar, Javier Barcenilla Complejo Hospitalario de Palencia, Palencia, Spain Background: Missed colon cancers are nowadays a great concern in gastrointestinal endoscopy. Colonoscopy seems to be less protective in rigth colon cancers that in left-transverse side. Some authors denied protective effect of colonoscopy in right colon cancers (1). Aims: To determine if there is any difference between right and Left colon in terms of missed colon cancers. To determine if there is any prottective effect of colonoscopy against right colon cancer by comparing frequence of previous colonoscopy with a healthy control group. Methods: We reviewed all colon cancers diagnosed in our hospital between Jan 2005-Jan 2010. Right was defined by location of tumour between appendix and hepatic angle ( both included), all others were considered lefttransverse. All previous colonoscopies during a 10 year period before diagnosis in those patients were extracted from our database.Our hospital is the only public institution able to do colonoscopy in our province and all endoscopic procedures are collected in a computorized data base since 1995. A control group of patients with a”clean colonoscopy”⫽(No cancer) during 2009 were matched by sex and age 2:1 with the study group. We extracted also all previous colonoscopies in this control group. Missed colon cancer was defined by a previous colonoscopy done within the 3 years period before diagnosis of cancer as in Ontario group(2) Results: 726 colon cancers were found. 416 males/310 females.206 were Right, 510 Left & 10 sinchronic. 1454 controls were extracted. Frequence of previous colonoscopies are summarized in table 1. 2,5% of the patients had a missed colon cancer. There were 4,6% in the right colon vs 1,7% in left transverse P⬍0,05. Previous colonoscopy rate in healthy controls was significantly higher 10,2% p⬍0,05 Conclusions: 1) Missed colon cancers are considerably more frequent in right colon (more than double) 2) It is mandatory to make changes in our practice to improve detection of flat/minimal lesions in right colon 3) Although previous colonoscopy was more frequent in right colon cancer group than in other locations there is still a difference with healthy controls. Protective effect of colonoscopy seems reasonable also in this
Volume 73, No. 4S : 2011
GASTROINTESTINAL ENDOSCOPY
AB437