Endoscopic submucosal dissection (ESD) for superficial gastric neoplasm is minimally invasive and achieves curative resection with an acceptable rate of adverse events. Although the safety of ESD has been demonstrated, the procedure is associated with a substantial risk of adverse events, including bleeding, perforation, and stricture.

Post-ESD bleeding is the most frequent adverse events and the incidence of post-ESD bleeding in previous studies ranges from 1.8% to 15.6%. Several studies identified that antithrombotic agents and large resection size were strong risk factors for post-ESD bleeding. Moreover, the incidence of bleeding in high-risk patients has been reported to be as high as 61.5%, depending on the definition of high-risk patients. The number of patients at high risk for post-ESD bleeding is increasing worldwide. In addition, as the indications for ESD have been expanding, more patients with large lesions undergo ESD. Therefore, it is important to prevent post-ESD bleeding in high-risk patients.

Although several previous studies have attempted to prevent post-ESD bleeding using surgical material, there is no definite prophylactic treatment to prevent re-bleeding after ESD. To date, coagulation of remnant vessels on the post-resection ulcer surface and administration of a proton pump inhibitor (PPI) after ESD are practical methods to prevent post-ESD bleeding.

Polysaccharide hemostatic powder (Endo-Clot™) is a new topical hemostatic method recently used for non-variceal upper gastrointestinal bleeding. This study aimed to identify the efficacy of hemostatic powder in preventing post-ESD bleeding in high-risk patients. This is a multicenter, prospective, randomized study.

A multicenter, prospective, randomized controlled study

• High-risk patients of post-ESD bleeding will be prospectively enrolled in the study. High-risk of post-ESD bleeding is defined as patients with taking antithrombotic agents or with large resection (specimen size ≥ 40mm).
• All ESDs were performed according to the standard procedure sequence in the hemostatic powder group and control group. The Only difference between two groups was hemostatic powder (Endo-clot™) application after standard hemostasis using hemostatic forceps in the hemostatic powder group. For patients in the study group, after hemostasis on the post-resection ulcer using conventional method and removal of specimen, Endo-clot™ (EndoClot Plus, Unc., Santa Clara, CA, USA) was applied immediately onto the post-resection surface.
• Bleeding rate of study group within 4 weeks after ESD will be compared with the control group. "

Inclusion Criteria:

• Older than 19 years old and younger than 85 years old
• Pathologically confirmed gastric adenoma and/or early gastric cancer
• Iatrogenic gastric ulcer after ESD (Endoscopic submucosal dissection) more than 40mm (at prediction)
• Patients who is taking antithrombotic drug such as aspirin and/or coumadin (and other anti-coagulation medication)
• ECOG performance status 0 or 1
• Adequate renal function (serum creatinine < 1.5 mg/dL or calculated creatinine clearance ≥ 60 ml/min)
• Adequate liver function (total bilirubin < 1.5 X the upper limits of normal (ULN), AST and ALT <3 X UNL, and alkaline phosphatases < 3 X ULN or < 5 x ULN in case of liver involvement)
• Adequate BM function (WBC ≥ 3,500/µl, absolute neutrophil cell count ≥ 1,500 /µl, platelet count ≥ 100,000/µl)
• Subjects who given written informed consent after being given a full description of the study

Exclusion Criteria:

• Previously treated by radical gastrectomy
• Adverse effect on this medication
• Pregnant or on breast feeding
• Patients who are unwilling or unable to provide informed consent, such as those with psychiatric problem, drug abuse or alcoholism

Randomization immediately after ESD