Two hundred and forty-one patients with at least one ulcer at stage A1 or A2, measuring at least 5 mm in its larger diameter, were included in this Brazilian double-blind randomized study. Patients received omeprazole 20 mg in the morning (n = 120) or ranitidine 300 mg at night (n = 121) for 2 wk; unhealed ulcers were treated for an additional 2 wk. At the end of 4 wk, unhealed ulcers were treated openly with omeprazole 20 mg o.m. for 4 wk. Healing rates at 2 and 4 wk were 67.3% and 92.9% for omeprazole and 39.8% and 82.0% for ranitidine (per protocol analysis). Results were similar when analyzed as intention to treat (p significant in favor of omeprazole). Epigastric day-time pain was the most common of all symptoms (89.2%), but only heartburn at day 15 showed a significantly better response to omeprazole than to ranitidine. A multivariate analysis (logit analysis) showed that the odds in favor of healing were greater for small ulcers, nonsmokers, and omeprazole treatment.