Clinically significant gastrointestinal bleeding in critically ill patients with and without stress-ulcer prophylaxis

Objective: To compare the rates of clinically significant gastrointestinal bleeding and the number of blood units and endoscopies required for gastrointestinal hemorrhage between patients receiving or not receiving stress-ulcer prophylaxis. Design: Historical observational study comparing two consecutive periods: with (phase 1) and without stress-ulcer prophylaxis (phase 2). Design and setting: A 17-bed intensive care unit in a university teaching hospital. Patients: In phase 1 there were 736 patients and in phase 2 737. Those in the two phases were comparable in age and reason for admission; clinically significant gastrointestinal bleeding rates did not differ between the two phases, but patients in phase 2 were more severely ill. Measurements and results: Comparable numbers of blood units were transfused per bleeding patient in the two phases, especially for patients with significant gastrointestinal bleeding. During each phase 19 fibroscopies were performed for significant bleeding, and two patients required surgery. The clinically significant gastrointestinal bleeding rate and outcome did not differ in patients with at least one risk factor. Total expenditures directly related to gastrointestinal bleeding were similar during the two phases; the total cost incurred by stress-ulcer prophylaxis was estimated at epsilon6700. Conclusions: Our results suggest that stress-ulcer prophylaxis does not influence the clinically significant gastrointestinal bleeding rate in intensive care unit patients or the cost of its management.