Tu1556 Transpancreatic Sphincterotomy (Goff Septotomy) Is Safe and Effective in Patients With Failed Wire/Contrast Guided Biliary Cannulation Siddharth B. Javia*, Priyanka Priyanka, Nathaniel Avila, Meera Avila, Dishita S. Pandya, Erik F. Rahimi, Shashideep Singhal, Nirav Thosani Gastroenterology, Hepatology & Nutrition, University of Texas Health Science Center at Houston, Houston, TX Background: Deep common bile duct (CBD) cannulation using standard technique is possible in majority of cases. Remaining cases may require advanced cannulation techniques such as transpancreatic sphincterotomy (Goff septotomy), needle knife sphincterotomy or EUS guided CBD cannulation. We study success rate and complication rates of Goff septotomy. Methods: This is a retrospective cohort study of patients (age 18-99), who underwent ERCP with CBD cannulation from July 1, 2014 to June 12, 2015 (nZ140). Patients with evidence of prior sphincterotomy or previously placed CBD stent, patients undergoing endoscopic retrograde pancrea- tography, patients whose ampulla was not reached due to difficult anatomy were not included. Data was collected for age, gender, indication, type and success of biliary cannulation and complications like acute pancreatitis, gastrointestinal bleeding, perforation. Fisher Exact test was used to test significance level between groups. Results: Total of 140 patients with native ampulla underwent biliary ERCP. Mean age of patients was 52.9 and 55(39.3%) patients were males. 112 out of 140 (nZ80%) patients had successful CBD cannulation using standard technique. In remaining 28 patients, 25(18%) had advanced cannulation technique performed. 20 out of these 25 patients had pancreatic duct (PD) cannulation while attempting CBD cannulation and subsequently underwent Goff septotomy to access biliary system. Goff septot- omy was successful in 15(75%) out of 20 patients. 5 out of 20 patients did not have successful CBD cannulation with Goff septotomy. Out of these 5 patients, one had rendezvous biliary access through t- tube, two patients had biliary access by needle knife sphincterotomy, one patient has EUS guided biliary access, one patient had unsuccessful needle knife sphincterotomy but subsequently patient had negative intraoperative cholangiogram and subsequent EUS guided ERCP was not warranted. In standard CBD cannulation group, 5 (4.46%) patients developed mild acute pancreatitis compared to 1(5%) patient in Goff septotomy group (pZ0.964). Pro- phylactic PD stents were placed in 14(12.5%) of patients in standard biliary cannu- lation group compared to 5(25%) of patients in Goff septotomy group (pZ0.586). In standard CBD cannulation group, 4 (3.5%) developed gastrointestinal bleeding out of which 2 (1.79%) had severe post sphincterotomy bleeding (drop of Hb>2 gm/dl). None of the patients who underwent Goff septotomy had gastrointestinal bleeding. There was no perforation in either group. There was no difference in overall complication rate between the standard cannulation (8%) group versus the Goff septotomy (1%) group (pZ0.936). Conclusion: Goff septotomy is effective in achieving biliary access when standard cannulation techniques fail and it is not associated with higher complication rates compared to the standard technique. Standard biliary access (N[112) Goff septotomy (N[20) p value Age* 53.15 51.9 Male 45(40%) 6(30%) Female 67(60%) 14(70%) Indication Cholangitis 9(8.0%) 2(10%) Choledocholithiasis 61(54.5%) 8(40%) Abnormal Imaging 22(19.6%) 6(30%) Complication Overall complication 9 (8.03%) 1(5%) 0.936 Acute pancreatitis 5(4.5%) 1(5%) 0.964 Gastrointestinal bleeding 4(3.6%) 0 Severe gastrointestinal bleeding 2(1.8%) 0 Perforation 0 0 * mean. Tu1557 Refractory Anastomotic Bile Leaks After Orthotopic Liver Transplantation Are Associated With Hepatic Artery Compromise and Carry High Morbidity Tomas DaVee* 1 , Sunil K. Geevarghese 2 , James C. Slaughter 3 , Patrick S. Yachimski 1 1 Medicine-Gastroenterology, Vanderbilt University, Nashville, TN; 2 Transplant Surgery, Vanderbilt University, Nashville, TN; 3 Biostatistics, Vanderbilt University, Nashville, TN Introduction: Anastomotic bile leaks are common following orthotopic liver trans- plant (OLT), and standard treatment consists of placement of a biliary endopros- thesis. The objectives of this study were to identify risk factors for refractory anastomotic bile leaks and to determine the morbidity associated with refractory bile leaks after OLT. Methods: Consecutive adult patients who underwent ERCP for treatment of an anastomotic bile leak following OLT between 2009 and 2014 were retrospectively identified from an institutional database. A refractory leak was defined as a bile leak that persisted following placement of a plastic biliary endo- prosthesis, and required repeat endoscopic or surgical intervention. Results: Forty- three subjects met study inclusion criteria. Median age was 57 years and 84% (36/ 43) of subjects were male. Initial endoscopic treatment consisted of placement of a transpapillary plastic biliary stent in 98% (42/43), of which 95% (41/43) traversed the anastomosis. Forty percent (17/43) of patients experienced a refractory leak following initial stent placement. Hepatic artery thrombosis was identified in 18% (3/17) of patients with refractory leaks vs 0% (0/26) with resolution of bile leak following initial stent placement (PZ0.06). Hepatic artery stenosis was identified in 35% (6/17) of patients with refractory leaks vs 4% (1/26) of patients with res- olution of bile leak following stent placement (PZ0.01). Overall, hepatic artery complications (HAC) were found in 53% (9/17) of patients with refractory bile leaks vs 8% (1/26) without refractory leaks (P<0.001). In patients with refractory bile leaks vs those with resolved leaks, 24% (4/17) underwent surgical biliary revision vs 8% (2/26; PZ0.2); and 12% (2/17) repeat OLT vs 8% (2/26; PZ0.9). Among patients with refractory leaks, salvage endoscopic therapy resulted in leak resolution in 53% (9/17) of cases; leak resolution was achieved with placement of a plastic stent in 45% (5/11) of cases and with a metal stent in 66% (4/6) of cases (PZ0.6). Overall, surgical biliary revision was required in 14% (6/43) and repeat OLT in 9% (4/43) of patients. Mortality among patients with refractory bile leak was 24% (4/17) vs 12% (3/26) with resolution of bile leak (PZ0.40). Time-to-event analysis revealed an association between refractory bile leak and the combined outcome of death, repeat OLT, or surgical biliary revision (PZ0.01). Conclusion: Refractory bile leaks following OLT are associated with compromised hepatic ar- tery flow, and result in a high degree of morbidity and mortality. Given the severity of complications found in this study, large-scale prospective investigation is warranted to define optimal endoscopic management strategies for patients who may be at risk for refractory bile leaks. AB606 GASTROINTESTINAL ENDOSCOPY Volume 83, No. 5S : 2016 www.giejournal.org Abstracts