Effect of scheduled second-look endoscopy on peptic ulcer bleeding: a prospective randomized multicenter trial

This study aimed to investigate the effectiveness of scheduled second-look endoscopy (EGD) with endoscopic hemostasis on peptic ulcer rebleeding and to identify the risk factors related to the need for second-look EGD.

We prospectively randomized patients who had endoscopically confirmed bleeding peptic ulcer with stigmata of active bleeding, visible vessel, or adherent clot into 2 groups between August 2010 and January 2013. Hemoclip application or thermal coagulation and/or epinephrine injection were allowed for initial endoscopic therapy. The same dosage of proton pump inhibitor was injected intravenously. The study group received scheduled second-look EGD 24 to 36 hours after the initial hemostasis, and further therapy was applied if endoscopic stigmata persisted, as above. Those patients who developed rebleeding underwent operation or radiologic intervention despite the additional endoscopic therapy. Outcome measures included rebleeding, amount of transfusion, duration of hospitalization, and mortality.

After initial endoscopic hemostasis, 319 eligible patients were randomized into 2 groups. Sixteen (10.1%) and 9 (5.6%) patients developed rebleeding (P = .132), respectively. There was also no difference in surgical intervention (0, 0% vs 1, .6%, P >.999) or radiologic intervention (3, 1.9% vs 2, 1.2%, P = .683), median duration of hospitalization (6.0 vs 5.0 days, P = .151), amount of transfusion (2.4 ± 1.7 vs 2.2 ± 1.6 units, P = .276), and mortality (2, 1.3% vs 2, 1.2%, P > .999) between the 2 groups. Multivariate analysis showed that grades 3 to 4 of endoscopists' estimation to success of initial hemostasis, history of nonsteroidal anti-inflammatory drug (NSAID) use, and larger amounts of blood transfusions (≥4 units of red blood cells) were the independent risk factors of rebleeding.

A single EGD with endoscopic hemostasis is not inferior to scheduled second-look endoscopy in terms of reduction in rebleeding rate of peptic ulcer bleeding. Repeat endoscopy would be helpful in the patients with unsatisfactory initial endoscopic hemostasis, use of NSAIDs, and larger amounts of transfused blood. (Clinical trial registration number: KCT0000565; 4-2010-0348.).