A randomized, controlled trial of sucralfate versus antacid as prophylaxis against upper gastrointestinal bleeding was carried out in 100 critically ill intensive care unit patients. Both groups were comparable with regard to sex, age, duration of prophylaxis, and the number of risk factors per patient. There was no gross bleeding in any of the patients. Three of the 48 sucralfate patients and 2 of the 52 antacid patients showed strongly positive results of Gastroccult tests of three successive hourly samples of gastric aspirate and were considered treatment failures. We conclude that sucralfate therapy is as effective as an antacid regimen in the prevention of gastrointestinal bleeding in critically ill patients. In view of a considerable savings in nursing time, it offers the opportunity for considerable reduction in the cost of intensive care while providing effective protection.