Background: The main purpose of this study was to compare omeprazole (ome) plus two antibiotics (OMC) with omeprazole plus placebo (OP) with regard to gastric ulcer relapse for a period of 2 years in patients who were Helicobacter pylori-positive at inclusion. Methods: Using double-blind randomization 125 patients with gastric ulcer were treated with either OMC (ome 20 mg b.i.d., metronidazole 400 mg b.i.d., clarithromycin 250 mg b.i.d.) (n = 64) or OP (ome 20 mg and placebo) (n = 61) for 1 week, followed by ome 20-40 mg o.d. until healing was confirmed endoscopically after 4, 8 or 12 weeks. Endoscopy and H. pylori diagnostics using culture, histology and serology were performed 6, 12 and 24 months after treatment or at symptomatic relapse. At inclusion, 35%of the OMC group and 38%of the OP group were taking non-steroidal anti-inflammatory drugs (NSAIDs). Nine percent (11/125) of the ulcers were malignant. Results: The prevalence of H. pylori was 82%and the eradication rate 88%in the OMC group and 3%in the OP group. More than 90%of the ulcers were healed after 12 weeks. After 2 years, 76%of patients in the OMC group were in remission compared with 28%in the OP group (ITT) (P < 0.001). Sixty percent of patients in the OMC group that continued to take NSAIDs were in remission after 2 years compared with none in the OP group. Atrophy but not intestinal metaplasia decreased after treatment. Conclusions: Gastric ulcers are mainly caused by H. pylori, and relapse is effectively prevented by H. pylori eradication, even in patients on NSAIDs. Background: The main purpose of this study was to compare omeprazole (ome) plus two antibiotics (OMC) with omeprazole plus placebo (OP) with regard to gastric ulcer relapse for a period of 2 years in patients who were Helicobacter pylori-positive at inclusion (N = 64) or OP (ome 20 mg and placebo) (n = 61) for 1 week, followed by ome 20-40 mg od until healing was confirmed endoscopically after 4, 8 or 12 weeks. Endoscopy and H. pylori diagnostics using culture, histology and serology were performed 6, 12 and 24 months after treatment At%, 35% of the OMC group and 38% of the OP group were taking non-steroidal anti-inflammatory drugs (NSAIDs). Nine percent (11/125) of the ulcers were malignant. Results: The prevalence of H. pylori was 82% and the eradication rate 88% in the O More than 90% of the ulcers were healed after 12 weeks. After 2 years, 76% of patients in the OMC group were remission compared with 28% in the OP group (ITT) ( P 0.001). Sixty percent of patients in the OMC group that continued to take NSAIDs were in remission after 2 years compared with none in the OP group. Atrophy but not intestinal metaplasia decreased after treatment. Conclusions: Gastric ulcers are mainly caused by H . pylori, and relapse is still prevented by H. pylori eradication, even in patients on NSAIDs.