727

Abstract

Introduction: Acid suppressing agents like histamine-2 receptor antagonists (H2RAs) and proton pump inhibitors (PPIs) are commonly used to prevent gastrointestinal hemorrhage in critically ill patients. The stronger acid suppressing effects of PPIs may reduce the rate of bleeding but enhance the occurrence of infectious complications, specifically pneumonia and Clostridium difficile infection (CDI). This study evaluated the occurrence of gastrointestinal bleeding, pneumonia, and CDI in critically ill patients and assessed risk factors for these outcomes. Methods: A pharmacoepidemiological cohort study was conducted evaluating adult patients requiring mechanical ventilation for ≥ 24 hours and administered either an H2RA or PPI for ≥ 48 hours while intubated across 71 hospitals between January 1, 2003 and December 31, 2008. The primary outcomes were secondary diagnoses of gastrointestinal bleeding, pneumonia, and CDI occurring ≥ 48 hours after initiating invasive ventilation. Odds ratios (OR) and 95% confidence intervals (CI) were determined using nonparsimonious multivariate regression models. Secondary outcomes included intensive care unit (ICU) and hospital lengths of stay, mortality, and total ICU and hospital costs. Results: Of the 35,312 included patients, 13,439 (38.1%) received H2RAs and 21,873 (61.9%) received PPIs. The rates of gastrointestinal bleeding (2.1% vs. 5.9%, p<0.0001), pneumonia (27% vs. 38.6%, p<0.0001), and CDI (2.2% vs. 3.8%, p<0.0001) were lower in the H2RA group. After adjusting for covariates, odds ratios of gastrointestinal hemorrhage (OR=2.15; 95% CI 1.77-2.61, p<0.0001), pneumonia (OR=1.2; 95% CI 1.02-1.42, p=0.032), and CDI (OR=1.33; 95% CI 1.09-1.63, p=0.0052) were greater in the PPI group. Other variables independently associated with gastrointestinal hemorrhage were age, acute respiratory failure, sepsis, shock, kidney injury, hepatic injury, myocardial infarction, coagulopathy, and parenteral nutrition. Male sex, hepatic injury, kidney injury, neurologic injury, transplant, sepsis, shock, acute respiratory failure, duration of mechanical ventilation, congestive heart failure, duration of acid suppressing therapy, and parenteral nutrition were associated with pneumonia. Chronic kidney injury, transplant, inflammatory bowel disease, sepsis, shock, atrial fibrillation, congestive heart failure, duration of mechanical ventilation, and the use of carbapenems, piperacillin, or parenteral nutrition were associated with CDI. Median ICU stay (4 days vs. 5 days, p<0.00001), hospital stay (6 days vs. 6 days, p<0.00001), ICU mortality rate (10.8% vs. 17.8%), non-ICU hospital mortality rate (5.1% vs. 7.8%, p<0.00001), median ICU costs ($17,076 vs. $18,946, p<0.001), and median hospital costs ($5952 vs. $6282, p<0.01) all favored H2RA therapy. Conclusions: PPI therapy poses higher risk of gastrointestinal hemorrhage, pneumonia, and CDI compared to H2RA therapy in mechanically ventilated patients. Numerous other risk factors are evident. Until these findings are validated in a prospective comparative trial, these data suggest H2RAs are preferred to PPIs for the prevention of gastrointestinal hemorrhage in adult critically ill patients.