Proton pump inhibitor doesn't reduce ICU mortality

Reuters Health Information: Proton pump inhibitor doesn't reduce ICU mortality

Proton pump inhibitor doesn't reduce ICU mortality

Last Updated: 2018-10-24

By Gene Emery

(Reuters Health) - Giving intensive care unit (ICU) patients a proton-pump inhibitor (PPI) to prevent gastrointestinal bleeding does not seem to affect mortality, according to a European trial of nearly 3,300 adults.

The 90-day death rate was 31.1% with 40 mg of the PPI pantoprazole given intravenously versus 30.4% with placebo (P=0.76). Both were given once a day.

And although the drug cut the rate of clinically important gastrointestinal (GI) bleeding (2.5% vs. 4.2%), the rates of a composite of GI bleeding, pneumonia, myocardial ischemia and Clostridium difficile infection were similar, at 21.9% with the drug and 22.6% with placebo.

"The good news is that severe gastrointestinal bleeding is rare in ICU patients and that prophylactic pantoprazole may be avoided," coauthor Dr. Anders Perner of Rigshospitalet in Copenhagen told Reuters Health by email.

The findings of the trial, known as SUP-ICU, were reported Wednesday at the European Society of Intensive Care Medicine's 31st Annual Congress in Paris and online by the New England Journal of Medicine.

"In our view, the take-home message from this trial is that, given the lower incidence of clinically important upper gastrointestinal bleeding in the ICU, prophylaxis with a PPI, if initiated, should be reserved for seriously ill patients who are at risk for this complication," said Drs. Alan Barkum of McGill University in Montreal and Marc Bardou of the University of Burgundy Franche Comte in Dijon, France, in an accompanying editorial.

They also acknowledged that the definition of a high risk of bleed has not been standardized.

Dr. Perner said he has changed the way he practices based on the findings.

"So have many of the investigators in the SUP-ICU trial team. Time will show what others do," he told Reuters Health. "The results support the ongoing theme that less is more in the ICU. Simplifying intensive care is rational as this will reduce the risk of drug-drug interaction, side-effects and medication errors in general."

Mechanical ventilation, hepatic failure, kidney failure and coagulopathy are known to be risk factors for gastrointestinal bleeding in the ICU.

The U.S. Food and Drug Administration has not approved the use of PPIs to prevent stress ulcers, but they are frequently used for that purpose.

The patients remained in the ICU for a median of six days after randomization and their treatment with the PPI or placebo lasted a median of four days.

When the researchers looked just at clinically important gastrointestinal bleeding events, a secondary endpoint, there was a difference of 1.7 percentage points. But the researchers didn't calculate a P-value because of a lack of adjustment for multiple comparisons.

There was no difference between the two groups when the researchers looked just at a combined endpoint of pneumonia and infection with C. difficile.

The same was true when it came to the percentages of days alive without life support.

The study did not require diagnostic endoscopy to differentiate between stress ulcers and bleeding from other causes, not did the researchers determine if a patient was being treated for acid reflux disease before admission.

Asked if it should be surprising that the treatment hasn't been tested earlier, Dr. Perner said, "A lot of what we do in intensive care has not been tested. The surprising part is that health care systems and societies have not established processes that ensure the testing of what we do to the sickest patients in the hospital."

The patients were treated in ICUs in Denmark, Finland, Switzerland, Norway, the Netherlands and the United Kingdom.


N Engl J Med 2018.

© Copyright 2013-2024 GI Health Foundation. All rights reserved.
This site is maintained as an educational resource for US healthcare providers only. Use of this website is governed by the GIHF terms of use and privacy statement.