To assess the comparative efficacy of omeprazole 20 mg, a proton pump inhibitor, versus ranitidine 150 mg twice a day, an H2-receptor antagonist, in healing duodenal ulcers we performed a randomized, double-blind, multicenter trial in 309 patients with endoscopically diagnosed ulcers. Patients were treated for up to four weeks and were seen at week 2 and at week 4, if unhealed at week 2, for determination of ulcer status by endoscopy, review of daily self-assessment symptom diaries, and clinical laboratory including fasting serum gastrin. Gastrin levels were repeated two weeks after cessation of study medication. Evaluation of baseline demographic and laboratory parameters demonstrated no significant differences between the two groups at entry. At week 2, 42% of the omeprazole and 34% of the ranitidine-treated patients were healed (P = NS). At week 4, there was a 19% advantage in ulcer healing for the omeprazole-treated patients in comparison to those treated with ranitidine (82% vs 63%, respectively, P less than 0.05). Healing of ulcers greater than or equal to 1.0 cm occurred in 83% of those treated with omeprazole versus 37% treated with ranitidine (P less than 0.01). There were no significant differences in rate of pain relief or incidence of clinical laboratory abnormalities. Mean fasting serum gastrin value during treatment increased over the baseline in both groups, (P less than 0.05). The percent change was significantly greater with omeprazole but few patients had elevations above the upper limit of normal for the assay. Both drugs were well tolerated. Omeprazole 20 mg demonstrated superiority in healing duodenal ulcers at four weeks in comparison to ranitidine 150 mg twice daily and was more effective in healing ulcers greater than or equal to 1.0 cm.