1033 Underestimation of Small Bowel Bleeding in Users of Nonselective Nonsteroidal Anti-Inflammatory Drugs (NSAIDs) and Low-Dose Aspirin (ASA): A Prospective Study

Abstract

G A A b st ra ct s the recurrence of LDA-related peptic ulcers has not been reported. We conducted a randomized, double-blind, double-dummy, active-controlled trial to assess the efficacy, doseresponse relationship and safety of rabeprazole for peptic ulcer recurrence in Japanese patients in long-term LDA therapy. Methods: Patients were eligible for the trial if they had an endoscopically confirmed gastric or duodenal ulcer scars and were receiving long-term LDA (81 mg/day or 100 mg/day) therapy for cardiovascular and cerebrovascular protection. Patients on long-term LDA therapy were also eligible if they had a previous history of endoscopically confirmed active peptic ulcers, but did not present them at baseline endoscopy. Patients were randomly segregated into three groups receiving rabeprazole 10 mg once daily (standard dose in Japan), rabeprazole 5 mg once daily or teprenone (mucosal protective agent commercially available in Japan) 50 mg three times per day as an active control. The primary endpoint was recurrence of gastric and duodenal ulcers over 24 weeks. Secondary endpoints included an occurrence of bleeding ulcers with stigmata of recent hemorrhage or drop in hemoglobin level of more than 2 g/dl, changes from baseline in the modified Lanza score (endoscopic scoring system for gastroduodenal damage) or upper gastrointestinal symptom score, and newly-developed erosive esophagitis. Results: Among 472 randomized patients, 452 patients (rabeprazole 10 mg, n=151; rabeprazole 5 mg, n= 150; teprenone 150 mg, n=151) constituted the full analysis set. The cumulative recurrence rates of gastric and duodenal ulcers over 24 weeks in the 5and 10-mg rabeprazole groups were 2.8% and 1.4%, respectively, and both were significantly lower than that in the teprenone group (21.7%). The cumulative occurrence rate of bleeding ulcers over 24 weeks in the teprenone group was 6.1%, while bleeding ulcers were not observed in the 5or 10-mg rabeprazole groups. The rates of worsening in the modified Lanza and upper gastrointestinal symptom scores in the 5and 10-mg rabeprazole groups were significantly lower than those in the teprenone group. Significantly fewer patients developed erosive esophagitis in the 5 mg rabeprazole group (2%) and 10 mg rabeprazole group (0%), compared with the teprenone group (8.6%). Rabeprazole was well-tolerated at both doses. Conclusions: Rabeprazole prevents the recurrence of gastroduodenal ulcers with no evidence of a major dose response effect in patients taking LDA, and is well-tolerated.