We compared the efficacy of epinephrine injection and epinephrine injection followed by sodium tetradecyl sulfate in controlling active ulcer bleeding. Out of 2814 patients who underwent endoscopy for gastrointestinal bleeding, 200 patients with actively bleeding ulcers seen at the time of endoscopy were randomized to receive epinephrine injection alone (99 patients) or epinephrine injection followed by 3% sodium tetradecyl sulfate (101 patients). After the procedure the patients were transferred to the surgical gastroenterology ward and were treated by surgeons who were unaware of the mode of treatment. The patients underwent routine endoscopy 24 hours later, and epinephrine injection was repeated if active bleeding was seen again. Emergency surgery was performed for the following: (1) arterial spurting not controlled endoscopically, (2) failure of the blood pressure or pulse to stabilize after 4 units of blood, (3) total transfusion of more than 8 units of blood, or (4) rebleeding as defined by hematemesis with pulse greater than 100 beats/min or blood pressure less than 100 mm Hg after stabilization. The two groups were comparable in age, sex, site of ulcer, and severity of bleeding. Initial hemostasis was obtained at the time of endoscopy in 94% of the epinephrine group and 97% of the epinephrine plus sodium tetradecyl sulfate group. No difference in outcome was seen in the two groups as measured by emergency surgery requirement, blood transfusion, hospital stay, and hospital mortality. Endoscopic epinephrine injection is effective in controlling active ulcer bleeding. The additional injection of sodium tetradecyl sulfate confers no additional advantage.