From Wiki Journal Club
Jump to: navigation, search
PEPTIC Investigators for the Australian and New Zealand Intensive Care Society Clinical Trials Group, Alberta Health Services Critical Care Strategic Clinical Network, and the Irish Critical Care Trials Group. "Effect of Stress Ulcer Prophylaxis With Proton Pump Inhibitors vs Histamine-2 Receptor Blockers on In-Hospital Mortality Among ICU Patients Receiving Invasive Mechanical Ventilation: The PEPTIC Randomized Clinical Trial". JAMA. 2020. 323(7):616-626.

Clinical Question

In adult patients admitted to critical care requiring mechanical ventilation, does treatment with a proton pump inhibitor compared to a histamine-2 receptor blocking agent have better in-hospital mortality.

Bottom Line

There is no difference in mortality between PPIs and H2RAs but there may be more clinically significant bleeding seen with H2RAs.

Major Points

Stress ulcer prophylaxis (SUP) has been the standard of care in critical illness for over 30 years. Routine stress ulcer prophylaxis may be associated with infectious complications including C. difficile-associated diarrhea, pneumonia, and death.[1][2] Strong evidence supporting the routine use of PPIs was lacking and following the publishing of SUP-ICU in 2018, PPI’s were associated with a non-significantly fewer bleeds, and no difference in mortality.

The PEPTIC randomized control trial, published in 2020, randomized entire units to either a PPI or H2RA used for stress ulcer prophylaxis. Once a clinician determined a patient should receive stress ulcer prophylaxis the patient would receive, in an open label fashion, which ever treatment the institution was randomized to, but the clinician could choose otherwise. Comparing patients that were randomized to receive a PPI (n=13 436) to an H2RA (n=13 392) in one of 50 ICUs in 5 countries, there was no difference in in-hospital mortality: 18.3% vs. 17.5%, ARI 0.93 (95% CI -0.01 to 1.88) P = 0.054.

The only secondary outcome to be statistically significant was Clinically Important Upper GI bleeding, 1.3% vs. 1.8%, ARR 0.51 (95% CI 0.9 to 0.12) P = 0.009, which was seen in regions outside of Canada. There were a number of issues with the trial design, first that the clinicians were allowed to cross over, leading to greater exposure to PPIs. Second Individual patient data was only collected in Canada and Ireland. Third, the use of registry data may led to recorder bias. Four, this trail did not assess the effects of feeding on stress ulcer formation.


As of August 2020, no guidelines have been published that reflect the results of this trial.


  • Multi-centre, randomized, open label, cluster cross-over, institution-level intervention, registry trial
  • N=26 828
    • PPI Approach (n=13 436)
    • H2RA Approach (n=13 392)
  • Setting: 50 ICUs in Australia, Canada, England, Ireland, and New Zealand
  • Enrollment: August 2016 to January 2019
  • Analysis:
  • Primary Outcome: in-hospital mortality, up to 90 days from index ICU admission


Inclusion Criteria

  • ≥ 18 years
  • Requiring mechanical ventilation within 24 hours of ICU admission

Exclusion Criteria

  • ICU admission diagnosis of upper GI bleed

Baseline Characteristics

‘'PPI Group displayed

  • Demographics: mean age 58(SD 17) years, 36% female,
  • APACHE II chronic comorbidities: 6.1% respiratory, 6.5% cardiovascular, 1.4% hepatic, 1.9% kidney, 5.8% immunosuppression, 2.7% metastatic cancer
  • Physiologic parameters: mean APACHE II score 18.7 (SD 8.3), mean APACHE III score 65.2 (SD 29.9)


  • Proton pump inhibitor
  • Histamine-2 receptor blocker


Comparisons are PPI vs. H2RA Groups.

Primary Outcomes

Died at the hospital by 90 days
18.3% vs. 17.5%, ARI 0.93 (95% CI -0.01 to 1.88) P = 0.054

Secondary Outcomes

Clinically Important Upper GI Bleeding
1.3% vs. 1.8%, ARR 0.51 (95% CI 0.9 to 0.12) P = 0.009
C. Diff infection
0.3% vs. 0.43%, ARR 0.11 (95% CI -0.25 to 0.03) P = 0.13
Median ICU LOS, days
3.6 (IQR 1.6-10.4) vs. 3.3(IQR 1.5-10.0), Ratio 1.0(95% CI 097-1.03)
Median Hospital LOS, days
12.2 (IQR 6.0-40.0) vs. 12.0(IQR 6.0-39.3), Ratio 1.01(0.98-1.03)

Tertiary Outcomes

Ventilator Associated Events
6.5% vs. 5.8%, ARI 1.11(95% CI -0.89 to 3.11) P = 0.28

Subgroup Analysis

Clinically Important UGIB by region
Australia / NZ: RR 0.46 [95% CI, 0.26- 0.81]
Canada: RR 0.92 [95% CI, 0.71- 1.19]
England and Ireland: RR 0.48 [95% CI, 0.33-0.69] P = 0.004

Adverse Events

Any Adverse Study Drug Associated Event
1 patient experience an allergic reaction to omeprazole and switched to ranitidine


  • Asymmetric non-adherence – physicians were allowed to cross over or give other treatment, more PPI were given overall
  • Open label intervention medications within the randomized class was used
  • Individual patient level data only collected in Canada and Ireland
  • GI Bleeding difference driven by all sites except those in Canada
  • Registry data utilized which may have recorder bias
  • Primary outcome switched after approval but before randomization
  • No washout at cross-over
  • Estimated adherence relatively low: 82% for PPI, 63% in H2RA
  • Did not assess affects of feeding


  • Health Research Council of New Zealand
  • Canadian Institutes of Health Research
  • Australian and New Zealand Intensive Care Foundation
  • Health Research Board of Ireland
  • Supported by: UK National Institute for Health Research Critical Care Clinical Research Networks and the UK Critical Care Research Group

Further Reading

  1. Marik PE et al. Stress ulcer prophylaxis in the new millennium: a systematic review and meta-analysis. Crit. Care Med. 2010. 38:2222-8.
  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.