Low-dose antacid treatment without and with additional ranitidine in the prevention of acute gastroduodenal stress ulceration in high-risk patients was compared in a randomized, double-blind clinical trial with endoscopic examination of the upper gastrointestinal tract before entry and after the 7-day study period. Of 67 patients who entered the study, 56 could be evaluated. Only one patient, allocated to antacid and placebo, had massive acute upper gastrointestinal bleeding, arising from a gastric ulcer. Although gastric intraluminal pH was better controlled with additional ranitidine treatment, the occurrence of mucosal lesions did not depend on the assigned treatment. We conclude that there is no significant difference between the two regimens in terms of endoscopically visible lesions and clinical outcome.