- Email: [email protected]
Outcome of Endoscopic Resection of Large Colonic Polyps by Endoscopic Mucosal Resection or Endoscopic Submucosal Dissection in Patients with Inflammatory Bowel Disease
AGA Abstracts
EMR, 4 required ESD, and 4 had hybrid ESD-EMR. Only 6 patients had previous ESD or EMR. Thirty-eight patients (70%) had en-bloc resection while the other 30% underwent piecemeal resection. Argon-plasma coagulation was used in 35 cases, hot snare in 24 and hot biopsy in 8 cases. Twenty-five cases (46%) had clips applied to prevent bleeding. Lowgrade dysplasia was found in 33 polyps, serrated adenoma in 15 polyps, tubular adenoma in 14 polyps, tubulovillous adenoma in 9 polyps and hyperplastic in 5polyps. Adverse events were seen in only one patient who developed post-procedure fever, but did not require hospitalization or antibiotics. Only 9 patients needed colonic resection within 12 months of the procedure. A recurrence was seen in 5 cases, with 3 patients requiring EMR, 1 requiring ESD, and one who needed both. For the recurrent polyps, en-bloc resection was done in all 5 patients. There were 4 low-grade dysplastic polyps and 1 serrated polyp found. There were no adverse events amongst the recurrent cases. Conclusion: ESD and EMR are effective and safe therapies for removal of large colonic polyps (>2cm). Recurrence (which was endoscopically resectable) occurred in only a quarter of patients, and there were no major adverse events. Thus, these techniques are effective alternatives to colectomy for patients with IBD and dysplastic polyps.
Mo1807 CROHN'S DISEASE DIAGNOSIS AFTER PROCTOCOLECTOMY AND ILEAL POUCH ANAL-ANASTOMOSIS FOR ULCERATIVE COLITIS Julian Hercun, Carole Richard, Ramses Wassef, Raymond Lahaie, Pierre Poitras Introduction: Total proctocolectomy and ileal pouch anal-anastomosis (IPAA) is considered a curative procedure for ulcerative colitis (UC). However, signs of Crohn's disease (CD) can develop postoperatively in some cases. Our aim was to identify the prevalence of signs suggestive of CD in UC patients with IPAA as well as potential predictive factors present at time of surgery. Methods: We reviewed the files of 302 patients with an IPAA performed between 1985 and 2014 at the CHUM Hôpital Saint Luc in Montreal. We included the 163 cases with a minimal postoperative follow-up of 5 years in our analysis. The preoperative diagnosis was UC in 145 cases and indeterminate colitis (IC) in 18 cases. There were no cases of CD. Results: 1) Signs of CD were noted in 35 patients. The diagnosis of CD was based on intestinal inflammation (proximal to the ileal pouch) in 19 cases, on fistulising perianal disease in 11 cases, and a combination of both in 5 cases. 2) The development of CD was related to the duration of the postoperative evolution: the total number of CD patients identified increased from 12 within the first 5 years, to 26 within the first 10 years and to 35 beyond 10 years of follow-up. 3) When comparing IPAA patients with or without CD, the following predictive factors for CD were observed: active tobacco smoking at time of surgery (5 % UC vs 22 % CD p=0.006), preoperative steroid treatment (86 vs 100% p= 0.014), the clinician's suspicion of possible CD (7 vs 26 % p=0.004), mouth ulcers (12 vs 27% p=0.093). Endoscopic or histological appearance were not different in the two groups. Pouchitis was observed more frequently in the CD group (63 vs 26 % p<0.001) 4) Disease was controlled by 5-ASA in 3 of 15 patients (20 %), by immunomodulators in 7 of 19 patients (37%), and by TNF-inhibitors in 13 of 17 patients (76%). Surgical treatment was performed in 13 patients (7 patients needed pouch removal); in 9 cases (82%) with fistulising disease (with pouch removal in 36% of cases), while in only 2 cases (11%) in the intestinal inflammation group. Conclusion: Proctocolectomy and IPAA offers a cure for UC in a majority of cases. However, the possibility of CD undiagnosed pre-procedure and revealed postoperatively should not be underestimated as it can appear many months/years after surgery with its frequency increasing over time. In our study, endoscopic and clinical characteristics at time of surgery failed to differentiate apparent UC patients from those later identified as CD. Therapeutic options for postoperative CD of the ileal pouch are identical to those available for treatment of typical CD.
Analysis of Factors Associated with Length of Stay
Mo1805 SYSTEMATIC REVIEW AND META-ANALYSIS: RISK OF COLORECTAL CANCER IN PATIENTS OF ULCERATIVE COLITIS AND AN ASIAN PERSPECTIVE Sawan Bopanna, Ashwin Ananthakrishnan, Saurabh Kedia, Vijay Yajnik, Vineet Ahuja Introduction: The increased risk of colorectal cancer (CRC) in ulcerative colitis (UC) is well known.Risk of sporadic colorectal cancer in Asian populations is considered low and risk estimates of UC related CRC from Asia vary. This meta-analyses is an Asian perspective on the risk of UC related CRC. Methods: We searched MEDLINE/EMBASE for terms related to CRC in UC from inception to June 2016. Search for published articles was done country wise for all countries in Asia. We included 48 studies with information on the prevalence and cumulative risk of CRC at various time points. Stratification according to region within Asia was done. A random-effects meta-analysis was performed to calculate the pooled prevalence rates as well as cumulative risk at 10,20 and 30 years of disease. Results: Our analysis included a total of 25399 patients of ulcerative colitis with a total of 338 reported colorectal cancers. Using pooled prevalence estimates from various studies, the overall prevalence was 0.85 [95% CI 0.65-1.04]. At 10, 20 and 30 years of disease, the risk for CRC were 0.02%, 4.87% and 13.7 %. Subgroup analysis by stratifying the studies according to region or period of study did not reveal any significant difference. Conclusion: In a metaanalysis of 48 studies from Asian countries we found the risk of UC associated CRC to be similar to that in the West. Adherence to screening is therefore necessary. Larger population based prospective studies would be required for better estimates of the risk.
Mo1808 A RETROSPECTIVE ANALYSIS OF CLOSTRIDIUM DIFFICILE INFECTION IN PATIENTS WITH ULCERATIVE COLITIS Hui Xu, Yue Li, Tao Xu, Ji Li, Hong Yang, Hong Lv, Jiaming QIan Background Many reports have documented the increasing impact of Clostridium difficile infections (CDI) in patients with inflammatory bowel disease (IBD) in the latest years. To determine the prevalence, risk factors, clinical characteristics and prognosis of CDI in hospitalized ulcerative colitis (UC) patients, we conducted this retrospective analysis. Methods Patients with UC, hospitalized from January 2010 to December 2015 at the department of gastrointestinal in PUMCH, China were objects of this study. For all the patients suspected of CDI, stool samples were tested for toxins A and B of Clostridium difficile (CDAB) with enzyme-linked immuno sorbent assay (ELISA). Clinical data of CDAB positive patients were collected. Controls were CDAB negative patients by matching age, gender and the year CDAB tested at 1:2 ratios. Logistic regression was used to reveal the risk factors of CDI. Results In a total of 421 in-patients with UC, 34 (8.08%) were CDAB positive and diagnosed as CDI. 68 CDAB negative patients were matched. Univariate analyses revealed that risk factors for CDI were: antibiotic exposure within 3 months prior to CDAB test (P=0.004), prior hospitalization within 1 month (P=0.025), systemic use of steroids (P=0.002), and dose of steroids used in CDI patients was higher than non-CDI patients (P=0.001). At the meanwhile, the study found a correlation between active cytomegalovirus (CMV) infection and CDI in UC patients (P=0.001). On logistic regression analyses, active CMV infection had a significant difference between CDI and non-CDI patients (OR 13.502, 95%CI: 1.307~139.512, P=0.029). However, the severity of UC (evaluated on clinical criteria and endoscopic scoring system), distribution of UC, disease course, duration of disease, history of smoking and alcohol use, combination of diabetes, history of surgery, 5-aminosalicylic acid (5-ASA), proton pump inhibitor (PPI), immunosuppressants except steroids, infliximab, parenteral nutrition within 1 month didn't increase the risk of CDI in UC patients. Clinical features of CDI patients in UC had no significant differences from non-CDI patients, such as body mass index (BMI), defecation frequency, toxic megacolon incidence rate, leucocyte and neutrocyte level of peripheral blood (P≥0.05). CDI didn't increase the subsequent colectomy rate in this study. Conclusions The complication of IBD by C. difficile infection has received increasing attention. This retrospective study found that recent usage of antibiotic,
Mo1806 OUTCOME OF ENDOSCOPIC RESECTION OF LARGE COLONIC POLYPS BY ENDOSCOPIC MUCOSAL RESECTION OR ENDOSCOPIC SUBMUCOSAL DISSECTION IN PATIENTS WITH INFLAMMATORY BOWEL DISEASE Siddhant Yadav, Edward V. Loftus, Louis M. Wong Kee Song, Nayantara Coelho-Prabhu Background/Aim: Patients with chronic colitis due to inflammatory bowel disease (IBD) have an increased risk of colorectal cancer. There is increasing evidence to suggest that endoscopic removal of polyps by endoscopic mucosal resection (EMR) and endoscopic submucosal dissection (ESD) can prevent or lower the cancer risk, but data are scarce about removal of large polyps. We sought to describe our experience with EMR and ESD in IBD patients with polyps >20 mm in size. Method: We performed a retrospective chart review of patients with IBD and colonic polyps >20 mm in size who underwent EMR and/or ESD at our institution between January 1st, 2012 and June 2nd, 2016. Patient demographics, IBD characteristics, polyp morphologic features and histology, resection techniques, adverse events and lesion recurrence were noted. Result: Fifty-four patients (30 men) with IBD were included in the study. Thirty-one (57%) patients had ulcerative colitis, 16 (30%) had Crohn's disease and 7 (13%) had indeterminate colitis. Colonic involvement was found in 89% of cases. Polyps were located in the ascending colon (39%), transverse colon (33%) and sigmoid colon (28%). Twelve patients (22%) had scarring noted during colonoscopy. According to the Paris classification, 54% of patients had polypoid sessile (Is) polyps, while 39% of patients had non-polypoid (IIa, IIb, IIc) polyps. Forty six patients underwent an
AGA Abstracts
S-786
EMR, 4 required ESD, and 4 had hybrid ESD-EMR. Only 6 patients had previous ESD or EMR. Thirty-eight patients (70%) had en-bloc resection while the other 30% underwent piecemeal resection. Argon-plasma coagulation was used in 35 cases, hot snare in 24 and hot biopsy in 8 cases. Twenty-five cases (46%) had clips applied to prevent bleeding. Lowgrade dysplasia was found in 33 polyps, serrated adenoma in 15 polyps, tubular adenoma in 14 polyps, tubulovillous adenoma in 9 polyps and hyperplastic in 5polyps. Adverse events were seen in only one patient who developed post-procedure fever, but did not require hospitalization or antibiotics. Only 9 patients needed colonic resection within 12 months of the procedure. A recurrence was seen in 5 cases, with 3 patients requiring EMR, 1 requiring ESD, and one who needed both. For the recurrent polyps, en-bloc resection was done in all 5 patients. There were 4 low-grade dysplastic polyps and 1 serrated polyp found. There were no adverse events amongst the recurrent cases. Conclusion: ESD and EMR are effective and safe therapies for removal of large colonic polyps (>2cm). Recurrence (which was endoscopically resectable) occurred in only a quarter of patients, and there were no major adverse events. Thus, these techniques are effective alternatives to colectomy for patients with IBD and dysplastic polyps.
Mo1807 CROHN'S DISEASE DIAGNOSIS AFTER PROCTOCOLECTOMY AND ILEAL POUCH ANAL-ANASTOMOSIS FOR ULCERATIVE COLITIS Julian Hercun, Carole Richard, Ramses Wassef, Raymond Lahaie, Pierre Poitras Introduction: Total proctocolectomy and ileal pouch anal-anastomosis (IPAA) is considered a curative procedure for ulcerative colitis (UC). However, signs of Crohn's disease (CD) can develop postoperatively in some cases. Our aim was to identify the prevalence of signs suggestive of CD in UC patients with IPAA as well as potential predictive factors present at time of surgery. Methods: We reviewed the files of 302 patients with an IPAA performed between 1985 and 2014 at the CHUM Hôpital Saint Luc in Montreal. We included the 163 cases with a minimal postoperative follow-up of 5 years in our analysis. The preoperative diagnosis was UC in 145 cases and indeterminate colitis (IC) in 18 cases. There were no cases of CD. Results: 1) Signs of CD were noted in 35 patients. The diagnosis of CD was based on intestinal inflammation (proximal to the ileal pouch) in 19 cases, on fistulising perianal disease in 11 cases, and a combination of both in 5 cases. 2) The development of CD was related to the duration of the postoperative evolution: the total number of CD patients identified increased from 12 within the first 5 years, to 26 within the first 10 years and to 35 beyond 10 years of follow-up. 3) When comparing IPAA patients with or without CD, the following predictive factors for CD were observed: active tobacco smoking at time of surgery (5 % UC vs 22 % CD p=0.006), preoperative steroid treatment (86 vs 100% p= 0.014), the clinician's suspicion of possible CD (7 vs 26 % p=0.004), mouth ulcers (12 vs 27% p=0.093). Endoscopic or histological appearance were not different in the two groups. Pouchitis was observed more frequently in the CD group (63 vs 26 % p<0.001) 4) Disease was controlled by 5-ASA in 3 of 15 patients (20 %), by immunomodulators in 7 of 19 patients (37%), and by TNF-inhibitors in 13 of 17 patients (76%). Surgical treatment was performed in 13 patients (7 patients needed pouch removal); in 9 cases (82%) with fistulising disease (with pouch removal in 36% of cases), while in only 2 cases (11%) in the intestinal inflammation group. Conclusion: Proctocolectomy and IPAA offers a cure for UC in a majority of cases. However, the possibility of CD undiagnosed pre-procedure and revealed postoperatively should not be underestimated as it can appear many months/years after surgery with its frequency increasing over time. In our study, endoscopic and clinical characteristics at time of surgery failed to differentiate apparent UC patients from those later identified as CD. Therapeutic options for postoperative CD of the ileal pouch are identical to those available for treatment of typical CD.
Analysis of Factors Associated with Length of Stay
Mo1805 SYSTEMATIC REVIEW AND META-ANALYSIS: RISK OF COLORECTAL CANCER IN PATIENTS OF ULCERATIVE COLITIS AND AN ASIAN PERSPECTIVE Sawan Bopanna, Ashwin Ananthakrishnan, Saurabh Kedia, Vijay Yajnik, Vineet Ahuja Introduction: The increased risk of colorectal cancer (CRC) in ulcerative colitis (UC) is well known.Risk of sporadic colorectal cancer in Asian populations is considered low and risk estimates of UC related CRC from Asia vary. This meta-analyses is an Asian perspective on the risk of UC related CRC. Methods: We searched MEDLINE/EMBASE for terms related to CRC in UC from inception to June 2016. Search for published articles was done country wise for all countries in Asia. We included 48 studies with information on the prevalence and cumulative risk of CRC at various time points. Stratification according to region within Asia was done. A random-effects meta-analysis was performed to calculate the pooled prevalence rates as well as cumulative risk at 10,20 and 30 years of disease. Results: Our analysis included a total of 25399 patients of ulcerative colitis with a total of 338 reported colorectal cancers. Using pooled prevalence estimates from various studies, the overall prevalence was 0.85 [95% CI 0.65-1.04]. At 10, 20 and 30 years of disease, the risk for CRC were 0.02%, 4.87% and 13.7 %. Subgroup analysis by stratifying the studies according to region or period of study did not reveal any significant difference. Conclusion: In a metaanalysis of 48 studies from Asian countries we found the risk of UC associated CRC to be similar to that in the West. Adherence to screening is therefore necessary. Larger population based prospective studies would be required for better estimates of the risk.
Mo1808 A RETROSPECTIVE ANALYSIS OF CLOSTRIDIUM DIFFICILE INFECTION IN PATIENTS WITH ULCERATIVE COLITIS Hui Xu, Yue Li, Tao Xu, Ji Li, Hong Yang, Hong Lv, Jiaming QIan Background Many reports have documented the increasing impact of Clostridium difficile infections (CDI) in patients with inflammatory bowel disease (IBD) in the latest years. To determine the prevalence, risk factors, clinical characteristics and prognosis of CDI in hospitalized ulcerative colitis (UC) patients, we conducted this retrospective analysis. Methods Patients with UC, hospitalized from January 2010 to December 2015 at the department of gastrointestinal in PUMCH, China were objects of this study. For all the patients suspected of CDI, stool samples were tested for toxins A and B of Clostridium difficile (CDAB) with enzyme-linked immuno sorbent assay (ELISA). Clinical data of CDAB positive patients were collected. Controls were CDAB negative patients by matching age, gender and the year CDAB tested at 1:2 ratios. Logistic regression was used to reveal the risk factors of CDI. Results In a total of 421 in-patients with UC, 34 (8.08%) were CDAB positive and diagnosed as CDI. 68 CDAB negative patients were matched. Univariate analyses revealed that risk factors for CDI were: antibiotic exposure within 3 months prior to CDAB test (P=0.004), prior hospitalization within 1 month (P=0.025), systemic use of steroids (P=0.002), and dose of steroids used in CDI patients was higher than non-CDI patients (P=0.001). At the meanwhile, the study found a correlation between active cytomegalovirus (CMV) infection and CDI in UC patients (P=0.001). On logistic regression analyses, active CMV infection had a significant difference between CDI and non-CDI patients (OR 13.502, 95%CI: 1.307~139.512, P=0.029). However, the severity of UC (evaluated on clinical criteria and endoscopic scoring system), distribution of UC, disease course, duration of disease, history of smoking and alcohol use, combination of diabetes, history of surgery, 5-aminosalicylic acid (5-ASA), proton pump inhibitor (PPI), immunosuppressants except steroids, infliximab, parenteral nutrition within 1 month didn't increase the risk of CDI in UC patients. Clinical features of CDI patients in UC had no significant differences from non-CDI patients, such as body mass index (BMI), defecation frequency, toxic megacolon incidence rate, leucocyte and neutrocyte level of peripheral blood (P≥0.05). CDI didn't increase the subsequent colectomy rate in this study. Conclusions The complication of IBD by C. difficile infection has received increasing attention. This retrospective study found that recent usage of antibiotic,
Mo1806 OUTCOME OF ENDOSCOPIC RESECTION OF LARGE COLONIC POLYPS BY ENDOSCOPIC MUCOSAL RESECTION OR ENDOSCOPIC SUBMUCOSAL DISSECTION IN PATIENTS WITH INFLAMMATORY BOWEL DISEASE Siddhant Yadav, Edward V. Loftus, Louis M. Wong Kee Song, Nayantara Coelho-Prabhu Background/Aim: Patients with chronic colitis due to inflammatory bowel disease (IBD) have an increased risk of colorectal cancer. There is increasing evidence to suggest that endoscopic removal of polyps by endoscopic mucosal resection (EMR) and endoscopic submucosal dissection (ESD) can prevent or lower the cancer risk, but data are scarce about removal of large polyps. We sought to describe our experience with EMR and ESD in IBD patients with polyps >20 mm in size. Method: We performed a retrospective chart review of patients with IBD and colonic polyps >20 mm in size who underwent EMR and/or ESD at our institution between January 1st, 2012 and June 2nd, 2016. Patient demographics, IBD characteristics, polyp morphologic features and histology, resection techniques, adverse events and lesion recurrence were noted. Result: Fifty-four patients (30 men) with IBD were included in the study. Thirty-one (57%) patients had ulcerative colitis, 16 (30%) had Crohn's disease and 7 (13%) had indeterminate colitis. Colonic involvement was found in 89% of cases. Polyps were located in the ascending colon (39%), transverse colon (33%) and sigmoid colon (28%). Twelve patients (22%) had scarring noted during colonoscopy. According to the Paris classification, 54% of patients had polypoid sessile (Is) polyps, while 39% of patients had non-polypoid (IIa, IIb, IIc) polyps. Forty six patients underwent an
AGA Abstracts
S-786