A prospective, randomized trial of thrombin versus cyanoacrylate injection in the control of acute gastric variceal hemorrhage.

BACKGROUND: Acute gastric variceal hemorrhage (AGVH) is a serious complication of portal hypertension. Endoscopic cyanoacrylate glue injection is standard therapy for acute hemostasis; however, it may be associated with serious complications. The role of thrombin injection has not been confirmed. This study compared endoscopic thrombin and glue injections in the hemostasis of AGVH. METHODS: 68 eligible patients with AGVH were randomized to receive thrombin injection (33 patients) or glue injection (35 patients). The primary end point was injection-induced gastric ulcers. Secondary end points were acute hemostasis, rebleeding, and mortality within 42 days. RESULTS: Both groups had comparable baseline data. Hemostasis of active bleeding at endoscopy was 90.0 % (9/10) in the thrombin group and 90.9 % (10/11) in the glue group (P = 0.58), and 48-hour hemostasis was achieved in 93.9 % (31/33) and 97.1 % (34/35), respectively (P = 0.60). Treatment failure at 5 days occurred in two patients (6.1 %) in the thrombin group and two patients (5.7 %) in the glue group (P > 0.99). Gastric ulcers occurred in none of the thrombin group and 11/30 (36.7 %) of the glue group (P < 0.001, 95 % confidence interval [CI] 8 % - 27 %). Complications occurred in 4 (12.1 %) and 18 (51.4 %) patients in the thrombin and glue groups, respectively (P < 0.001, 95 %CI 22 % - 45 %). Two patients who received glue had post-treatment gastric ulcer bleeding. One patient in each group died. CONCLUSIONS: Endoscopic thrombin injection was similar to glue injection in achieving successful hemostasis of AGVH. However, a higher incidence of complications may be associated with glue injection.