By Betty Tran, MD, MSc

Assistant Professor of Medicine, Pulmonary and Critical Care Medicine, Rush University Medical Center, Chicago

Dr. Tran reports no financial relationships relevant to this field of study.

SYNOPSIS: The authors of this multicenter, blinded, randomized trial found that among critically ill adults at risk for gastrointestinal (GI) bleeding, fewer patients in the pantoprazole group exhibited clinically important GI bleeding compared to placebo, although mortality at 90 days was similar in both groups.

SOURCE: Krag M, et al. Pantoprazole in patients at risk for gastrointestinal bleeding in the ICU. N Engl J Med 2018;379:2199-2208.

Prophylactic use of proton pump inhibitors (PPIs) in critically ill patients at risk for gastrointestinal bleeding (GIB), such as those receiving mechanical ventilation, is standard protocol in the ICU. In fact, it is a component of the daily ICU checklist as an important reminder to ensure compliance. However, data supporting their use are limited in terms of quality, and there are concerns surrounding higher risks of pneumonia and Clostridium difficile that may limit any benefits.1,2 The Stress Ulcer Prophylaxis in the Intensive Care Unit (SUP-ICU) trial was conducted in 33 ICUs in Denmark, Finland, the Netherlands, Norway, Switzerland, and the United Kingdom between 2016 and 2017. Investigators screened critically ill adults (age > 18 years) who were admitted to the ICU for an acute condition with at least one risk for GIB (shock; use of anticoagulants, nonsteroidal anti-inflammatory drugs, acetylsalicylic acid, and IV thrombolysis; acute renal replacement therapy; mechanical ventilation > 24 hours; chronic liver disease; and coagulopathy).

Ultimately, 3,298 patients were enrolled in the trial and were randomized either to IV pantoprazole 40 mg daily until ICU discharge or death (maximum 90 days) or placebo. If a patient was readmitted to the ICU within 90 days, the original assigned regimen was resumed. The primary outcome was death by 90 days after randomization. Secondary outcomes included clinically important GIB, new infection with either pneumonia or C. difficile, percentage of days alive without life support, serious adverse reactions (e.g., anaphylaxis, pancytopenia, acute hepatic failure), and a composite outcome of clinically important ICU events (GIB, pneumonia, C. difficile infection, or myocardial ischemia).

In addition to analyzing the per-protocol population, investigators assessed the primary outcome in predefined subgroups, including presence of liver disease, history of or ongoing coagulopathy, medical vs. surgical ICU, presence of shock, use of mechanical ventilation, and a Simplified Acute Physiology Score (SAPS) II above 53 at baseline (a score of 53 was chosen as predictive of a 50% mortality rate). At 90 days, similar mortality rates were seen in both the pantoprazole and placebo groups (31.1% vs. 30.4%, respectively; relative risk [RR], 1.02; 95% confidence interval [CI], 0.91-1.13; P = 0.76). There was no heterogeneity in the effect of pantoprazole on 90-day mortality in any predefined subgroup except for the group of patients with SAPS II scores > 53, where the mortality appeared higher in the pantoprazole group (RR, 1.13; 95% CI, 0.99-1.30; P = 0.05). In terms of secondary outcomes, the rate of clinically important ICU events was similar in both groups: 21.9% in the PPI group vs. 22.5% in the placebo group (RR, 0.96; 95% CI, 0.83-1.11). More patients in the placebo group demonstrated clinically important GIB compared to the PPI group (4.2% vs. 2.5%, respectively; RR, 0.58; 95% CI, 0.40-0.86). Rates of one or more new infections with pneumonia or C. difficile, serious adverse reactions, and days alive without life support were similar between the two groups. The authors did not provide any P values for the secondary outcomes, as no adjustments were made for multiple comparisons.


Although SUP-ICU was a large, multicenter, randomized, placebo-controlled, blinded trial, its results are unlikely to change current practice at this time. The trial did not reveal a significant difference in 90-day mortality between patients at risk for GIB who received a PPI vs. those who did not.

First, the trial was powered to detect an absolute between-group difference of 5%, which may be a large margin to detect. Second, there are no data regarding whether patients were receiving enteral nutrition at baseline, which can modify the effect of PPI prophylaxis on the development of stress ulcer-related GIB as well as risk of pneumonia.3 Third, the finding of higher 90-day mortality among patients receiving a PPI with a SAPS II score > 53 is hypothesis-generating at best. The finding was of borderline significance, and the trial was not powered to address the primary outcome in this subgroup. Finally, although it is difficult to interpret the secondary outcomes in this trial given no adjustment was made for multiple comparisons, the investigators did report an increase in the number of patients with clinically important GIB in the placebo group compared to those receiving a PPI without a difference in the rates of infection.

The composite secondary outcome that combined these measures (referred to as “clinically important events in the ICU”) was similar in both groups, but inferences are difficult to draw from this finding as this composite endpoint combines outcomes that are affected by PPI prophylaxis in opposite directions (e.g., PPI prophylaxis theoretically increases the risk of infection but reduces the risk of GIB). Currently, the standard of care in the ICU is stress ulcer prophylaxis in patients at risk for GIB. Considering that there was an observed higher rate of GIB in patients receiving only placebo without clear evidence of increased risk of infection or mortality in patients receiving PPI prophylaxis in this trial, this standard practice will continue until further evidence emerges to dictate otherwise.


  1. Krag M, et al. Stress ulcer prophylaxis versus placebo or no prophylaxis in critically ill patients. A systematic review of randomised clinical trials with meta-analysis and trial sequential analysis. Intensive Care Med 2014;40:11-22.
  2. Alhazzani W, et al. Efficacy and safety of stress ulcer prophylaxis in critically ill patients: A network meta-analysis of randomized trials. Intensive Care Med 2018;44:1-11.
  3. Huang HB, et al. Stress ulcer prophylaxis in intensive care unit patients receiving enteral nutrition: A systematic review and meta-analysis. Crit Care 2018;22:20.