Does gastrointestinal stress ulcer prophylaxis in the intensive care unit improve outcomes?

In critically ill patients at high risk of gastrointestinal (GI) bleeding, the use of proton pump inhibitors (PPIs) for stress ulcer prophylaxis leads to less bleeding without increasing the risk of infectious complications associated with PPIs. Although there was no overall effect on mortality with PPI use in this patient population, the study was powered to detect at least a 5% decrease in mortality. Smaller decreases in mortality, if they exist, would not have been detected. (LOE = 1b)

Krag M, Marker S, Perner A, et al, for the SUP-ICU trial group. Pantoprazole in patients at risk for gastrointestinal bleeding in the ICU. N Engl J Med 2018;379(23):2199-2208.  [PMID:30354950]

Randomized controlled trial (double-blinded)

Government

Concealed

Inpatient (ICU only)

For critically ill patients, prophylaxis for stress ulcers with PPIs is common but there is concern about potential adverse effects associated with these drugs, specifically infectious complications. In this multicenter European study, investigators randomized adult patients admitted to the intensive care unit (ICU) who had at least one risk factor for GI bleeding (shock, use of anticoagulants, renal replacement therapy, mechanical ventilation, liver disease, or coagulopathy) to receive either intravenous pantoprazole 40 mg daily (n = 1645) or matching placebo (n = 1653) until discharge from the ICU or death. The 2 groups were similar at baseline, except for slightly more patients with chronic lung disease and coagulopathy in the pantoprazole group and more patients who underwent emergency surgery in the placebo group. The average age was 67 years, and the majority of patients had medical admissions to the ICU and were receiving mechanical ventilation and vasopressors. For the primary outcome of death by 90 days, no difference was detected between the 2 groups (31.1% in pantoprazole group vs 30.4% in placebo group; P = .76), but the pantoprazole group did have a decreased rate of GI bleeding (2.5% vs. 4.2%; relative risk 0.58; 95% CI 0.40 - 0.86). Additionally, there was no difference in the number of patients in each group who had at least 1 of 4 clinically important events (GI bleeding, pneumonia, Clostridium difficile infection, and myocardial ischemia) and no serious adverse reactions were reported in either group.