We tested the hypothesis that therapeutic endoscopy using the Nd:YAG (neodymium:yttrium-aluminum-garnet) laser would benefit patients with acute peptic-ulcer bleeding. Over 43 months, 174 patients with active bleeding (n = 32) or stigmata of recent bleeding (n = 142) due to peptic ulcers were randomly assigned during endoscopy to either standard treatment with laser photocoagulation or therapy without photocoagulation. There were no significant differences in a number of outcomes between the group treated with laser photocoagulation and the control group. Continued bleeding or rebleeding was observed in 22 percent of the laser-treated group and in 20 percent of the control group. Urgent surgery was necessary in 16 percent of the laser-treated patients and in 17 percent of the controls. Laser-treated patients spent a mean of 41 hours in the intensive care unit, and controls spent a mean of 32 hours. The mean hospital stay was 12 days in the laser-treated group and 11 days in the control group. One death occurred in each group. When patients with active bleeding were analyzed separately, there was no significant difference in outcome, even though laser photocoagulation stopped active bleeding in 88 percent of cases. Among patients with visible vessels, rebleeding occurred in 5 of 14 (36 percent) who received laser treatment and 2 of 15 (13 percent) who did not. Laser treatment precipitated bleeding in four patients and duodenal perforation in one. We conclude that Nd:YAG-laser photocoagulation does not benefit patients with acute upper gastrointestinal bleeding from peptic ulcers.