Institute of Digestive Disease, The Chinese University of Hong Kong, Hong Kong (J.Y.L., R.S.T., H.C., B.S., F.K.C.). Division of Gastroenterology, Department of Medicine, Faculty of Medicine, Chulalongkorn University and King Chulalongkorn Memorial Hospital, Thai Red Cross Society, Bangkok, Thailand (R.P., R.R.). Changi General Hospital, Singapore (A.K., J.L., T.L.A.). Department of Surgery, Prince of Wales Hospital, Shatin, Hong Kong (Y.Y.). Nanyang Technological University, Singapore (J.J.S.).
BACKGROUND The effectiveness of the hemostatic powder TC-325 as a single endoscopic treatment for acute nonvariceal upper gastrointestinal bleeding is uncertain. OBJECTIVE To compare TC-325 with standard endoscopic hemostatic treatments in the control of active bleeding from nonvariceal upper gastrointestinal causes. DESIGN One-sided, noninferiority, randomized, controlled trial. (ClinicalTrials.gov: NCT02534571). SETTING University teaching hospitals in the Asia-Pacific region. PATIENTS 224 adult
patients with acute bleeding from a nonvariceal cause on upper gastrointestinal endoscopy. INTERVENTION TC-325 (n = 111) or standard hemostatic treatment (n = 113). MEASUREMENTS The primary outcome was control of bleeding within 30 days. Other outcomes included failure to control bleeding during index endoscopy, recurrent bleeding after initial hemostasis, further interventions, blood transfusion, hospitalization, and death. RESULTS 224 patients were enrolled (136 with gastroduodenal ulcers [60.7%], 33 with tumors [14.7%], and 55 with other causes of bleeding [24.6%]). Bleeding was controlled within 30 days in 100 of 111 patients (90.1%) in the TC-325 group and 92 of 113 (81.4%) in the standard treatment group (risk difference, 8.7 percentage points [1-sided 95% CI, 0.95 percentage point]). There were fewer failures of hemostasis during index endoscopy with TC-325 (3 [2.7%] vs. 11 [9.7%]; odds ratio, 0.26 [CI, 0.07 to 0.95]). Recurrent bleeding within 30 days did not differ between groups (9 [8.1%] vs. 10 [8.8%]). The need for further interventions also did not differ between groups (further endoscopic treatment: 8 [7.2%] vs. 10 [8.8%]; angiography: 2 [1.8%] vs. 4 [3.5%]; surgery: 1 [0.9%] vs. 0). There were 14 deaths in each group (12.6% vs. 12.4%). LIMITATION Clinicians were not blinded to treatment. CONCLUSION TC-325 is not inferior to standard treatment in the endoscopic control of bleeding from nonvariceal upper gastrointestinal causes. PRIMARY FUNDING SOURCE General Research Fund to the University Grants Committee, Hong Kong SAR Government.