BACKGROUND: Our aim was to study the efficacy of oral human recombinant epidermal growth factor (EGF) in the treatment of duodenal ulcers, on the basis of its repairing actions in the gastrointestinal tract. METHODS: A placebo-controlled, multicenter, randomized, and double-blind study was conducted. Treatment groups were A) placebo solution, B) 10 microg/ml of human recombinant (hr)-EGF, and C) 50 microg/ml of hr-EGF, three times daily during 6 weeks. Patients, 15-65 years old, with a duodenal ulcer >4 mm, who gave their written informed consent to participate were eligible. Exclusion criteria were gastric ulcer and more than one duodenal ulcer, ulcer-related complications, and previous treatment with oral EGF or other specific anti-ulcer drugs in the previous 2 weeks. The main outcome variable was ulcer healing, evaluated by endoscopy after the 2nd, 4th, and 6th week. RESULTS: One hundred and three patients were included. The groups were comparable with regard to age, sex, toxic habits, antecedents of ulcerous disease, initial size and depth or the ulcer, initial symptoms, and positivity for Helicobacter pylori. The ulcers were healed in a larger proportion of patients treated with hr-EGF at the highest dose (70.6% in group C versus 40.0% and 35.3% in placebo and low-dose groups, respectively (P = 0.007)). The difference was significant from week 4 on. Groups A and B did not differ. Eighty-eight percent of group C patients were cured or improved versus 57% and 56% in groups A and B, respectively. No adverse reactions were reported. CONCLUSIONS: Oral hr-EGF was effective in the treatment of duodenal ulcer at a 50-microg/ml dose every 8 h but not at 10 microg/ml.