Etodolac has been shown to have a favorable safety profile in short-term and long-term studies in both osteoarthritis (OA) and rheumatoid arthritis (RA). Two studies were conducted to further assess the gastrointestinal (GI) safety profile of this drug. These studies were designed to compare the therapeutic efficacy and upper GI effects of etodolac (600 mg/day) and naproxen (1000 mg/day) administered over 4 weeks in patients with active rheumatoid arthritis. In addition, the relative effects of the drugs on prostaglandin levels in the stomach and duodenum were assayed in one study. Fifteen patients were included in each study and received either 300 mg b.i.d. of etodolac or 500 mg b.i.d. of naproxen. In both studies, endoscopic examinations were performed on day 1 of the study and again 4 weeks later. In the second study, at the time of each endoscopy, samples of gastric and duodenal mucosa were taken for histologic study and prostaglandin assay. Endoscopy results from the first study showed significant differences in favor of etodolac between the two treatment groups. In the second study more naproxen-treated patients had abnormal endoscopy results than did etodolac-treated patients. Results from prostaglandin assays in gastric and duodenal mucosa showed no overall suppression of gastric or duodenal prostaglandin levels for etodolac-treated patients in contrast to naproxen-treated patients, who showed suppression of PGE2 and PGI2. The results of these studies show that etodolac therapy caused less gastric and duodenal injury than naproxen and also support the theory that the GI safety of etodolac may be due to selective sparing of cytoprotective prostaglandins.