Objective: To compare the incidence of nosocomial pneumonia in critically injured patients randomized to one of three stress ulcer prophylaxis regimens.
Design: Prospective, randomized clinical trial.
Methods: Mechanically ventilated patients admitted to the trauma intensive care unit of a Level I trauma center received sucralfate, antacid, or ranitidine.
Measurements and main results: Two hundred forty-two patients were randomized: sucralfate, n = 80; antacid, n = 82; and ranitidine, n = 80. There was no statistically significant difference in pneumonia rates among the treatment groups (p = 0.875). Pneumonia occurred more frequently in patients with gram-negative retrograde colonization from stomach to trachea (p = 0.02), but this accounted for only 13% of all pneumonias in the study population. The death rate in patients with pneumonia was not statistically different among the three groups. Although 20% developed overt gastrointestinal bleeding, no episode was clinically significant. Mean gastric pH was > 4 in 95% of the study population, including 88% of patients receiving sucralfate. The death rate in the antacid group was significantly higher (p = 0.046) but not because of increased gastrointestinal bleeding or pneumonia.
Conclusions: Our results show no difference in the incidence of nosocomial pneumonia in mechanically ventilated trauma patients during the first 4 days of stress ulcer prophylaxis with sucralfate, antacid, or ranitidine. There is a trend toward decreased pneumonia in the sucralfate group after study day 4. Even after controlling for injury severity, the mortality rate in the antacid group was significantly higher; the reasons for this are unknown.