A prospective randomized study of the injurious effects of aspirin and naproxen on the gastroduodenal mucosa in patients with rheumatoid arthritis.

A prospective double-blind randomized placebo-controlled study was conducted to compare the injurious effects of aspirin and naproxen on the gastroduodenal mucosa. Patients with rheumatoid arthritis attending the immunology clinic of a tertiary care center were included in the study. Ninety consecutive patients were randomized to receive aspirin, naproxen, or placebo. Clinical evaluation was done every 2 wk. Upper gastrointestinal endoscopy, biopsies from the gastric antrum and duodenum for histological examination, and Helicobacter pylori studies were done at entry and after 2 months. The number of patients with dyspeptic symptoms was similar in the aspirin (12) and naproxen (9) groups. In both cases, however, it was higher (p less than 0.01) than in the placebo group (none). Naproxen was associated with a higher score of endoscopic abnormalities in the gastric body (0.6 +/- 1.3) than aspirin (0.2 +/- 0.6) or placebo (0.03 +/- 0.18). H. pylori was present in a similar proportion of patients in the three groups (33-43%). A greater fall in hemoglobin was observed with aspirin (0.35 +/- 0.48 g%) and naproxen (0.44 +/- 0.53 g%) than with placebo (0.09 +/- 0.4 g%). The present study suggests that aspirin and naproxen produce a similar degree of injury to the gastroduodenal mucosa.