Background & Aims: Helicobacter pylori-negative idiopathic ulcers are increasingly recognized. The secular trend and long-term outcome of this condition are unknown. Methods: We prospectively studied consecutive patients with bleeding gastroduodenal ulcers from January to December 2000. The incidence and etiology of ulcers during this period were compared with that between September 1997 and August 1998. H pylori-negative idiopathic ulcers were defined as negative tests for H pylori, no exposure to analgesics within 4 weeks, and absence of other risk factors for ulcers. After the ulcers had healed, patients with H pylori-negative idiopathic ulcers and patients with H pylori ulcers who received eradication therapy were followed up for 12 months without anti-ulcer drugs. Results: Six hundred thirty-eight patients had bleeding ulcers: 213 (33.4%) were H pylori ulcers, and 120 (18.8%) were H pylori-negative idiopathic ulcers (vs 480 [50.3%] H pylori ulcers and 40 [4.2%] H pylori-negative idiopathic ulcers in 1997-1998; P <. 001). H pylori-negative idiopathic ulcers accounted for 16.1% of patients who were admitted for bleeding and 42.4% of patients who bled while in the hospital (P <. 0001); 28.3% of patients with H pylori-negative idiopathic ulcers had histologic evidence of past H pylori infection. The probability of recurrent ulcer complications in 12 months was 13.4% (95% CI: 7.3%-19.5%) in patients with H pylori-negative idiopathic ulcers and 2.5% (95% CI: 0.4%-4.6%) in patients with H pylori ulcers who received eradication therapy (P =. 0002). Conclusions: The incidence of H pylori-negative idiopathic bleeding ulcers is rising. These ulcers are prone to recurrent complications.