The protective effect of rebamipide against 0.15 N HCl solution containing 40% ethanol (HCl-ethanol)-induced mucosal lesion in the stomach was evaluated in rats and humans. In rat experiments: (1) rebamipide prevented development of mucosal damage induced by HCl-ethanol in a dose-dependent manner when tested at 10, 30, 100 and 300 mg/kg, i.p., and the prevention at 30-300 mg/kg was significant (P less than 0.05); (2) rebamipide significantly (P less than 0.05) increased mucosal blood flow by continuous i.v. dosing at 10 mg/kg/hr by the hydrogen gas clearance method; (3) rebamipide at 10 mg/kg, i.v., also significantly (P less than 0.05) inhibited reduction in gastric mucosal blood volume and tended to suppress oxygen saturation of hemoglobin following hemorrhagic shock by organ reflectance spectrophotometry. A double-blind crossover study was conducted to compare the gastric protective effect of rebamipide with an inactive placebo using 6 healthy male volunteers. Rebamipide was administered to them at a clinical dose of 300 mg/day orally for 7 days, and then HCl-ethanol was given once to induce mucosal damage. Rebamipide significantly (P less than 0.05) inhibited formation of mucosal lesion, decrease in the number of mucous granules and dilatation of intercellular space. These clinical findings indicate that rebamipide was protective against HCl-ethanol-induced gastric lesion at a clinical dose, and the mechanism of this gastric effect was considered to involve, in part, the improvement of mucosal microcirculation.