subsequently managed with bowel rest, intravenous fluids and antibiotics without improvement. Effectiveness was a composite of length of hospitalization plus addi- tional hospital care due to adverse events or readmissions (measured in hospital days). Costs of care were obtained from the National Inpatient Sample and pub- lished literature. Outcomes estimates (e.g. procedure complications and readmis- sions) were obtained from published literature. Primary analysis was performed from the health care perspective. Cost-effectiveness was measured in incremental cost- effectiveness ratios (ICERs) and compared to the national average cost of one hos- pital bed per diem as reference. One-way sensitivity analyses and tornado diagrams were generated to evaluate effect modifying variables. Results: Analysis of a hypo- thetical cohort of patients with acute cholecystitis unsuitable for cholecystectomy showed that EGBD dominates PGBD. Compared to PGBD, ERC-TP was a cost-saving strategy and EUS-GB was cost-effective requiring $1,312 per hospitalization day averted (Figure 1). Both interventions were acceptable under the cost of one hos- pital bed per diem ($2,338 average across the country). Compared to ERC-TP, EUS- GBD required expending $8,950 more per hospitalization day averted. Our model was significantly affected by luminal apposing metal stent (LAMS) price, the length of stay in patients managed conservatively and the length of stay of patients requiring surgery (i.e. those with technical or clinical failure after drainage attempt). Cost- effectiveness was minimally affected by variations in technical success rate of each endoscopic procedure. (Figure 2) Conclusions: EGBD is cost-effective compared to PGBD, favoring ERC-TP over EUS-GBD. Further efforts are needed to make EGBD available in more medical centers, reduce equipment costs and shorten inpatient care. Figure 1. Cost-effectiveness analysis comparing three treatment strategies Figure 2. Tornado diagram comparing ICERs of EUS-GBD vs. ERC-TP Su1321 EFFECTIVITY AND SAFETY OF EMERGENT DOUBLE BALLOON ENTEROSCOPY IN PATIENTS WITH OVERT ONGOING OBSCURE GASTROINTESTINAL BLEEDING Goktug Sirin*, Ali Erkan Duman, Hasan Yilmaz, Zeki Islamoglu, Altay Celebi, Sadettin Hülagü Gastroenterology, Kocaeli University Medical Faculty Hospital, Kocaeli, Turkey Background and Aims: Double Balloon Enteroscopy (DBE) is a modality that has proven its efficacy in the diagnosis and treatment of small bowel diseases, espe- cially mid GI bleeding, in recent years. Currently, a consensus regarding the role of emergent DBE in overt ongoing obscure gastrointestinal bleeding is not exist (OO- OGIB). In this study, we wanted to compare the diagnostic and therapeutic yields of emergent and non-emergent DBE in patients with OO-OGIB and evaluate whether this condition created a security problem or not. Methods: Prospectively collected data of the DBE procedures performed between November 2007 and December 2018 in Kocaeli University Gastroenterology Clinic were evaluated. Emergent DBE was defined as an enteroscopy performed within the first 24 hours of presentation, after the case confirmed as OO-OGIB by conventional upper and lower endoscopy both of them failed to show the location of bleeding. Non-emergent group cate- gorized as Urgent (24-72 hours) and non-urgent (later 72 hours) subgroups, also. Demographic, clinical, laboratory, and technical data were reviewed. The diagnostic and therapeutic yields evaluated and compared between emergent and non-emer- gent DBE groups. All procedures has been made using therapeutic DBE systems manufactured by Fuji Film Co, Japan. Results: We performed 1371 DBE examinations in a total of 989 patients for probing the small intestine between september 2007 and December 2018. Among them, 487 patients who has suspicious of mid GI bleeding (OGIB) were enrolled in this study (253 men and 234 women, with an age range of 19 – 99 years, and a mean of 59.4 years). Patients with overt ongoing OGIB accounted for 13.9% of the total number of patients with OGIB (68/487). Eighty-five DBEs were performed in 68 patients with overt OGIB (42 (61.7%) males, mean age: 58.0Æ7.3 years]. The diagnostic yield of emergent DBE was significantly higher than non-emergent DBE (100 vs. 70.0%, P<0.05). Difference between emergent group and non-urgent group was most significantly, also (P<0.01). The most common diagnosis were angiodysplasias (27.9%), erosions/ulcers (22.1%), dieulafoy lesions (19.1%), diverticula (13.2%), tumors (11.7%), and polyps (5.9%), respectively. Endoscopic therapy was more frequently performed in emergent DBE patients versus non-urgent group (88 vs. 64%, PZ0.03). The most commonly used endo- scopic therapies were hemostatic clips (40%), hemostatic clips plus argon plasma coagulation (20.0%), argon plasma coagulation (30.0%), epinephrine injection (5.7%) and polypectomy (4.3%). None of the patients enrolled in this study devel- oped severe complications due to endoscopic examination. Conclusions: DBE ex- amination when performed in terms of emergent is a safe and effective option for patients with overt ongoing OGIB. The diagnostic and therapeutic impact of DBE was higher in this setting. Su1322 OBESITY IS CORRELATED WITH PROLONGED DOUBLE BALLOON ENTEROSCOPY PROCEDURE DURATION Victor Ciofoaia*, Andrea Koralewski, Matthew J. Crawford, Alex M. Kesler, Paul T. Kroner, Andrea C. Rodriguez, Fernando Stancampiano, Frank Lukens, Mark Stark, Bhaumik Brahmbhatt Gastroenterology, Mayo Clinic Florida, Jacksonville, FL Background: Double balloon enteroscopy (DBE) allows the diagnosis and therapy of small bowel disorders with a low complication rate, however it remains a resource- intensive procedure that is not available in many endoscopy units, with equipment, expertise and the duration of the procedure remaining a frequent concern. In this study we attempt to identify procedure, operator and patient variables impacting the duration of the procedure. Method: We reviewed a 20% sample of all DBE per- formed by three endoscopists during 1/2012 - 11/ 2017 at Mayo Clinic Florida rep- resenting 506 patients. Basic demographic, clinical data and procedure details were obtained by chart review after obtaining adequate IRB approval from our institution. In order to determine the variables that influence procedure duration we employed logistic regression using Stata 14. Results: The mean age for our population was 64.4 years; 51.49 patients were female and 88% Caucasian. 10% of procedures were inpatient procedures. 260 were anterograde enteroscopies, 187 retrograde enter- oscopies and 59 combined upper/lower enteroscopies. Complete enteroscopy was achieved in 276 procedures (54.54%). Average procedure time was 65.28 minutes (SD 40.7) for the anterograde DBE, 56.27 minutes (SD 31.64) for retrograde DBE and 103.64 minutes ( sd 48.1) for combined anterograde/retrograde DBE. Average fluoroscopy time was 118.17 seconds (SD 78.5). The most common indications and findings are presented in table 1 and table 2. Obese patients (BMI > 30) had prolonged procedure time when compared to normal weight and overweight patients (Odds RatioZ3.74, pZ0.05), maintained even after adjusting for age, gender, race, American Society of Anaesthesiologists score, prep quality, anterograde versus retrograde approach, procedure indication and total fluoroscopy time used. Conclusion: DBE remains a time consuming pro- cedure when compared with other endoscopies. DBEs required significant time and do not always allow for a total enteroscopy. Obesity is the most important factor predicting procedure duration. Top Indications Patients (N) Percentage (%) GI bleeding 262 51.78 Abnormal CT imaging 63 12.45 Small bowel obstruction 57 11.26 Abdominal pain 42 8.30 Suspected tumor/mass 37 7.31 www.giejournal.org Volume 89, No. 6S : 2019 GASTROINTESTINAL ENDOSCOPY AB331 Abstracts