6%) received transfusion and after 1 month, 40 patients (41.7%) still received transfusion. Average transfusion requirement per patient after 1 month was 4.7 packed RBCs (range 0–32). The median follow-up
was 13 months (range 1–102) and 76 patients (78.4%) were expired. Conclusion: Endoscopic hemostasis is not enough to control entire bleeding of gastric cancer and transfusion is needed due to delayed bleeding. To control bleeding by gastric cancer effectively, find more other hemostatic method should be considered. Key Word(s): 1. gastric cancer; 2. bleeding control; 3. endoscopy Presenting Author: TAE-HOON OH Additional Authors: TAE HOON OH, SEUNG SUK BAEK, YENA CHOI, MI JIN RYU, JI YOUNG PARK, EILEEN L. YOON, TAE JOO JEON, WON CHANG SHIN, WON CHOONG CHOI Corresponding Author: TAE-HOON OH Affiliations: Sanggye Paik Hospital, Sanggye Paik Hospital,
Sanggye Paik Hospital, Sanggye Paik Hospital, Sanggye Paik Hospital, Sanggye Paik Hospital, Sanggye Paik Hospital, Sanggye Paik Hospital, Sanggye Paik Hospital Objective: Non-variceal upper GI bleeding (NVUGIB) is a common medical problem that has significant association with morbidity and mortality. Angiographic detection and subsequent transarterial embolization (TAE) is a primary treatment option when medical and endoscopic treatments fail. We investigated clinical factors that could affect the success of the angiographic detection and prognosis after TAE in patients with NVUGIB refractory to endoscopic therapy. Methods: A retrospective analysis of EX 527 nmr the clinical data was done in patients with failed endoscopic treatment who underwent angiography for the treatment of acute NVUGIB between May 2002 and May 2013. Patients
were divided into detection or non-detection groups according to the presence of bleeding stigmata in angiographic finding. Rebleeding defined as subsequent bleeding event within 7 days and mortality within 30 days were analyzed as outcome parameters after TAE following detection in angiography. Results: A total 45 patients (37 male, mean age, 65.9 ± 14.9 oxyclozanide years) were analyzed and classified as a detection group (n = 25, 55.5%) and non-detection group (n = 20, 44.6%). Peptic ulcers were the most common cause of refractory NVUGIB. Larger transfusion amount (5.7 ± 3.9 unit vs. 3.5 ± 2.8 unit; P = 0.03), prolonged aPTT level (34.2 ± 17.3 sec vs. 21.8 ± 13.8 sec; P = 0.01) and short time interval between last endoscopy and angiography (17.5 ± 25.9 hours vs. 34.3 ± 59.5 hours; P = 0.04) were found to be significant factors for predicting angiographic detection. TAE was performed in all patients detected in angiography. Rebleeding (44%) was significantly associated with higher Rockall score (8.3 ± 1.5 vs. 6.6 ± 2.4; P = 0.046) and mortality (12%) was significantly associated with higher Rockall score (9.3 ± 0.6 vs. 7.1 ± 2.2; P = 0.002) and higher level of BUN (55.3 ± 47.4 vs. 27.6 ± 17.4; P = 0.01).