Conclusions: Restarting NOACs after endoscopic resection did not increase the risk of delayed bleeding. NOACs present a low risk for delayed bleeding and reduce the length of hospital stay compared with heparin bridging therapy; hence NOACs may be safe and useful for periendoscopic management of patients requiring anticoag- ulants. Sa1006 Prophylactic Endotracheal Intubation in Critically Ill Patients with Upper Gastrointestinal Bleeding is Associated with Higher Cardiopulmonary Unplanned Events Peter Junwoo Lee* 1 , Umar Hayat 2 , Hamid Ullah 2 , Rocio Lopez 3 , John J. Vargo 1 1 Digestive Disease Institute, Cleveland Clinic, Cleveland, OH; 2 Medicine Institute, Cleveland Clinic, Cleveland, OH; 3 Department of Quantitative Health Sciences, Cleveland Clinic, Cleveland, OH Aim: To compare the incidence of cardiopulmonary complication rates between pa- tients who underwent endotracheal intubation versus those who didn’t, in critically ill patients with brisk upper gastrointestinal bleeding (UGIB). Methods: A prospec- tively collected and maintained endoscopy database utilizing ProVationÒMD was used to screen for all patients who underwent an upper endoscopy (EGD) in an intensive care unit (ICU) at the Cleveland Clinic between 2004 and 2012. Patients aged 18 years or older who had presented with hematemesis and/or any patient with melena with consequential hypovolemic shock with/without cirrhosis were included. We collected relevant laboratory values, APACHE scores, Glasgow Coma Scale (GCS), Charlson Comorbidity Index (CCI) and several validated prognostic scores relevant to UGIB. The prognostic scores we collected include Glasgow Blatchford Score, AIMS65 score and Rockall score. Primary outcome was a com- posite of several cardiopulmonary complications occurring within 48 hours of the endoscopic procedure. These included pneumonia, pulmonary edema, ARDS, persistent shock/hypotension after the procedure, arrhythmia, myocardial infarction, and cardiac arrest. Patients who underwent an EGD in a non-ICU setting (i.e. endoscopy suite) and patients who were intubated for a reason other than UGIB were excluded. Propensity score matching was used to match each patient 1:1 in variables that could influence the outcome of interest. These included Glasgow Blatchford Score, CCI, APACHE score and location of UGIB event (i.e. community or in-hospital). Results: A total of 156 patients were included. The baseline character- istics were similar between the two groups. Comorbidity scores, prognostic scores, proportion of patients with underlying cirrhosis and distribution of different etiol- ogies of GIB were similar between the intubated group and the non-intubated group. The overall cardiopulmonary complication rate was significantly higher in the intubated group compared to the non-intubated group (11.5 vs 26.9% pZ0.015). This remained significant (pZ0.044) after adjusting for Glasgow Blatchford score and propensity score. This absolute risk difference yielded a number needed to harm (NNH) of 6. The mortality rate was higher in the non-intubated group but this did not reach statistical significance (16.7 vs 9% pZ0.15). Conclusion: Prophylactic endotracheal intubation prior to an EGD for a brisk UGIB in a critically ill patient may lead to an increased risk of a cardiopulmonary unplanned event. The benefits and risks of intubation should be carefully weighed when considering airway pro- tection before an EGD in this group of patients. EGD Findings Overall (N[156) No Pre-EGD Intubation (N[78) Pre-EGD Intubation (N[78) p value N Summary N Summary N Summary Factor EGD findings (Non- exclusive) Esophageal Ulcer 152 5 (3.3) 78 3 (3.8) 74 2 (2.7) NS Mallory Weiss Tear 152 6 (3.9) 78 2 (2.6) 74 4 (5.4) NS Stomach Ulcer 152 27 (17.8) 78 14 (17.9) 74 13 (17.6) NS Duodenal Ulcer 152 29 (19.1) 78 19 (24.4) 74 10 (13.5) NS Dieulafoy Lesion 152 4 (2.6) 78 2 (2.6) 74 2 (2.6) NS Esophageal Varices 152 50 (32.9) 78 21 (26.9) 74 29 (39.2) NS Gastric Varices 152 6 (3.9) 78 4 (5.1) 74 2 (2.7) NS Bleeding Esophageal Varices 61 10 (16.4) 27 4 (14.8) 34 6 (17.6) NS Forrest Classification 1a/b 80 17 (21.3) 43 8 (18.6) 37 9 (24.3) NS 2a 80 17 (21.3) 43 8 (18.6) 37 9 (24.3) NS 2b 80 14 (17.5) 43 7 (16.3) 37 7 (18.9) NS 2c 80 3 (3.8) 43 2 (4.7) 37 1 (2.7) NS 3 80 18 (22.5) 43 12 (27.9) 37 6 (16.2) NS Sa1007 Complications of ERCP in Post-Liver Transplant Patients Daniel Pievsky* 1 , Richard J. Arrigo 1 , Nneoma Okoronkwo 1 , Michelle A. Pievsky 2 , Andrew H. Zabolotsky 1 , Raj Sarkar 1 , Nikolaos Pyrsopoulos 1 , Sushil Ahlawat 1 1 Rutgers New Jersey Medical School, JERSEY CITY, NJ; 2 Fairleigh Dickinson University, Teaneck, NJ Endoscopic retrograde cholangiopancreatography (ERCP) is the standard of care for the diagnosis and treatment of biliary complications in patients who have undergone orthotopic liver transplant (OLT). However, there is a paucity of data on predictors of complications associated with ERCP in the post-OLT population. We examine the outcomes and risk factors associated with post-ERCP complications in OLT recipi- ents in our tertiary medical center over a 15 year period. A retrospective analysis was conducted on all ERCP procedures performed between 1997 – 2012 on patients that had undergone an OLT. Data regarding demographics, laboratory results, immu- nosuppresion protocol, diagnostic findings, therapeutic interventions during ERCP and post ERCP complications within 7 days of the procedures, including pancreatitis, cholangitis, post-sphincterotomy bleeding, subcapsular hematoma, perforations and cardiopulmonary events were recorded. Predictive factors were determined using univariate analyses and entered into a stepwise forward binary logistic regression model. A total of 219 ERCP procedures were performed on 120 patients. The median age of the patients was 53 years (range 4 – 77 years). A full list of patient baseline characteristics and results of univariate analyses is shown in Table 1. There were 28 total complications (12.7%) noted with 15 cases of mild to moderate pancreatitis, 6 cases of cholangitis, 2 cases of post-sphincterotomy bleeding and 5 cases of car- diopulmonary complications (see Table 2). No deaths related to the procedure were noted. Multivariate analysis revealed that, a history of pancreatitis (OR 4.3; 95% CI 1.6 – 11.6; p Z 0.004) and placement of biliary stent during ERCP (OR 3.1; 95% CI 1.2 – 7.8; pZ0.02) independently predicted post-ERCP complications in OLT pa- tients. Our overall complication rate of 12.7% is slightly higher than the rate for the general population (4-12%) but is well within the range of 2-18% which was reported in other small studies of post-ERCP complications in OLT patients. Our rate of post- ERCP pancreatitis was also similar to that of general population and there were no cases of severe pancreatitis. An increased risk of complications was found in those patients who had a history of pancreatitis and in those that had a biliary stent placed during the procedure. None of the patients in our study received indomethacin for prophylaxis of post-ERCP pancreatitis. The role of indomethacin is still uncertain in this population and warrants further study. Our data support the growing body of evidence that OLT patients are not at increased risk compared to those without liver transplants when undergoing ERCP procedures. Table 1. Baseline Patient Characteristics prior to ERCP Variables Procedures without Complications (N [ 191) Procedures with Complications (N [ 28) P value Age, years, mean (SD) 52 (10) 53 (9) 0.55 Gender, female, n (%) 44 (23) 7(25) 0.81 Time between OLT and ERCP, months (SD) 28 (47) 28 (52) 0.98 Previous sphicterotomy (%) 53 (28) 4 (14) 0.17 Total bilirubin (SD) 6.1 (8.4) 6.3 (8.2) 0.91 Alkaline phosphatase (SD) 479 (382) 478 (485) 0.99 GGT (SD) 743 (799) 488 (439) 0.02 AST (SD) 143 (173) 133 (148) 0.77 ALT (SD) 158 (162) 106 (78) 0.01 WBC (SD) 7.2 (3.7) 6.1 (2.7) 0.12 Hemoglobin (SD) 11.5 (1.9) 11.1 (1.7) 0.30 Platelets (SD) 168 (79) 177 (89) 0.57 INR (SD) 1.2 (0.85) 1.2 (0.97) 0.87 Creatinine (SD) 1.1 (0.4) 1.3 (0.5) 0.09 History of acute pancreatitis (%) 19 (10) 8 (29) 0.01 History of chronic pancreatitis (%) 11 (6) 4 (14) 0.11 History of post-ERCP pancreatitis (%) 17 (9) 5 (18) 0.17 Abstracts www.giejournal.org Volume 83, No. 5S : 2016 GASTROINTESTINAL ENDOSCOPY AB201