SUP-ICU: Stress Ulcer Prophylaxis in the ICU (2018)

“In this large, multicenter trial involving adult patients in the ICU who were at risk for gastrointestinal bleeding, mortality at 90 days was not lower among those who received prophylactic pantoprazole than among those who received placebo.”

— The SUP-ICU Investigators

1. Publication Details

  • Trial Title: Pantoprazole in Patients at Risk for Gastrointestinal Bleeding in the ICU.
  • Citation: 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. doi:10.1056/NEJMoa1714919.
  • Published: December 6, 2018, in The New England Jounal of Medicine.
  • Author: Mette Krag, M.D., Ph.D.
  • Funding: The Danish Council for Independent Research and others.

2. Keywords

Stress Ulcer Prophylaxis, Gastrointestinal Bleeding, Proton-Pump Inhibitor (PPI), Pantoprazole, Critical Illness, Ventilator-Associated Pneumonia (VAP).

3. The Clinical Question

In adult ICU patients at risk for clinically important gastrointestinal bleeding (Population), does prophylactic pantoprazole (Intervention) compared to placebo (Comparison) reduce 90-day all-cause mortality (Outcome)?

4. Background and Rationale

  • Existing Knowledge: Stress ulcer prophylaxis (SUP) with proton-pump inhibitors (PPIs) or H2-receptor antagonists is standard practice in many ICUs to prevent GI bleeding. However, the incidence of clinically important bleeding has decreased over time with improvements in critical care.
  • Knowledge Gap: It was unknown if routine SUP provided a net benefit. While it might reduce bleeding, there were concerns that acid suppression could increase the risk of pneumonia and Clostridium difficile infection. A large, placebo-controlled trial was needed to assess the overall effect on mortality.
  • Proposed Hypothesis: The authors hypothesized that prophylactic pantoprazole would reduce 90-day mortality in adult ICU patients at risk for GI bleeding.

5. Study Design and Methods

  • Design: A prospective, multicenter, international, randomized, double-blind, placebo-controlled, parallel-group trial.
  • Setting: 33 intensive care units (ICUs) in 6 countries.
  • Trial Period: Enrollment from January 2016 to October 2017.
  • Population:
    • Inclusion Criteria: Adult patients (≥18 years) acutely admitted to the ICU with at least one risk factor for clinically important GI bleeding (e.g., shock, mechanical ventilation, coagulopathy).
    • Exclusion Criteria: Known contraindications to PPIs, ongoing GI bleeding, or receiving therapeutic acid suppression.
  • Intervention: Pantoprazole 40 mg intravenously once daily.
  • Control: Matching placebo (saline) intravenously once daily.
  • Management Common to Both Groups: All other aspects of critical care were at the discretion of the local clinical team.
  • Power and Sample Size: The trial was powered to detect a 5% absolute difference in 90-day mortality, requiring 3350 patients.
  • Outcomes:
    • Primary Outcome: All-cause mortality at 90 days.
    • Secondary Outcomes: Included incidence of clinically important GI bleeding, pneumonia, C. difficileinfection, and new myocardial ischemia.

6. Key Results

  • Enrollment and Baseline: 3298 patients were randomized (1645 to pantoprazole, 1653 to placebo). The groups were well-matched at baseline.
  • Trial Status: The trial was completed as planned.
  • Primary Outcome: There was no significant difference in 90-day mortality between the pantoprazole and placebo groups (31.1% vs 30.4%; P=0.76).
  • Secondary Outcomes: The incidence of clinically important GI bleeding was significantly lower in the pantoprazole group (2.5% vs 4.2%; P=0.003). There were no significant differences in the rates of pneumonia or C. difficile infection.
  • Adverse Events: The rates of serious adverse events were similar between the two groups.

7. Medical Statistics

  • Analysis Principle: An intention-to-treat analysis was performed.
  • Statistical Tests Used: The primary outcome was analyzed using a logistic regression model.
  • Primary Outcome Analysis: The proportion of deaths at day 90 was compared between the two groups.
  • Key Statistic(s) Reported: Relative Risk (RR) for death at 90 days with pantoprazole: 1.02 (95% CI, 0.91 to 1.13; P=0.76).
  • Interpretation of Key Statistic(s):
    • Relative Risk (RR):
      • Formula: Conceptually, RR = (Risk in Intervention Group) / (Risk in Control Group).
      • Calculation: The paper reports the RR as 1.02.
      • Clinical Meaning: An RR of 1.02 means there was a 2% higher relative risk of death in the pantoprazole group, a difference that is not statistically significant.
    • Confidence Interval (CI):
      • Formula: Conceptually, CI = (Point Estimate) ± (Margin of Error).
      • Calculation: The reported 95% CI was 0.91 to 1.13.
      • Clinical Meaning: Since this narrow confidence interval is centered on the line of no effect (1.0), it provides a precise estimate that there is no significant difference between the two strategies. The true effect is likely somewhere between a 9% benefit and a 13% harm.
    • P-value:
      • Calculation: The reported p-value was 0.76.
      • Clinical Meaning: The p-value of 0.76 is far above the 0.05 threshold, confirming that the observed result is very likely due to chance. A result is conventionally considered statistically significant if the p-value is less than 0.05.
  • Clinical Impact Measures:
    • Absolute Risk Reduction (ARR) for GI Bleeding:
      • Formula: ARR = (Risk in Control Group) – (Risk in Intervention Group).
      • Calculation: ARR = 4.2% – 2.5% = 1.7%.
      • Clinical Meaning: For every 1000 at-risk ICU patients treated with pantoprazole, about 17 were prevented from having a clinically important GI bleed.
    • Number Needed to Treat (NNT) to prevent one GI bleed:
      • Formula: NNT = 1 / ARR.
      • Calculation: NNT = 1 / 0.017 = 58.8.
      • Clinical Meaning: Approximately 59 at-risk ICU patients need to be treated with prophylactic pantoprazole to prevent one additional clinically important GI bleed.
  • Subgroup Analyses: No significant benefit was found in any of the pre-specified subgroups.

8. Strengths of the Study

  • Study Design and Conduct: This was a very large, international, multicenter, randomized, double-blind, placebo-controlled trial, representing the highest level of evidence.
  • Generalizability: The pragmatic design and inclusion of a large number of diverse ICUs make the findings highly generalizable.
  • Patient-Centered Outcomes: The primary outcome of 90-day mortality is a strong, patient-centered endpoint.

9. Limitations and Weaknesses

  • Internal Validity (Bias): No major limitations to internal validity.
  • External Validity (Generalizability): The results are specific to pantoprazole and may not apply to other forms of acid suppression.
  • Other: The overall rate of bleeding in the placebo group was low (4.2%), which may have made it difficult for a reduction in bleeding to translate into a mortality benefit.

10. Conclusion of the Authors

“Among adult patients in the ICU who were at risk for gastrointestinal bleeding, mortality at 90 days and the number of clinically important events were not significantly lower with pantoprazole than with placebo. However, the number of episodes of clinically important gastrointestinal bleeding was lower with pantoprazole.”

11. To Summarize

  • Impact on Current Practice: The SUP-ICU trial was a landmark study that has forced a re-evaluation of the routine practice of stress ulcer prophylaxis. It definitively showed that while PPIs do reduce GI bleeding, this does not translate into a survival benefit in a general at-risk ICU population. This has led to a more nuanced approach, suggesting that SUP should be targeted to the highest-risk patients, rather than being a universal practice.
  • Specific Recommendations:
    • Patient Selection: For the general population of adult ICU patients with risk factors for GI bleeding.
    • Actionable Intervention: Prophylactic pantoprazole reduces the risk of GI bleeding (NNT ~59) but does not improve survival and does not appear to increase the risk of pneumonia. The decision to use it should be based on an individual patient’s bleeding risk.
    • Expected Benefit: A reduction in clinically important GI bleeding, but no effect on mortality.
  • What This Trial Does NOT Mean: This trial does not mean that SUP should be abandoned entirely. It is still effective at preventing bleeding, and may be very important for patients at extremely high risk (e.g., severe coagulopathy, history of peptic ulcer disease).
  • Implementation Caveats: The findings support a more thoughtful, de-escalating approach to SUP, questioning its necessity in every patient with a risk factor.

12. Context and Related Studies

  • Building on Previous Evidence: This trial was the definitive placebo-controlled study that was needed to assess the overall risk-benefit balance of modern SUP with PPIs, a question left open by older trials like the SIC trial (1998) which compared two active agents.
  • Influence on Subsequent Research: This trial, along with the subsequent REVISE trial (2024), has provided a massive amount of data that is reshaping international guidelines on stress ulcer prophylaxis, moving toward a more targeted rather than routine approach.

13. Unresolved Questions & Future Directions

  • Unresolved Questions: Which specific subgroups of patients have a high enough risk of bleeding that the NNT to prevent a bleed is low enough to justify routine prophylaxis?
  • Future Directions: Research is focused on developing better risk-stratification tools to identify the highest-risk patients who are most likely to benefit from SUP.

14. External Links

15. Framework for Critical Appraisal

  • Clinical Question: The question was of fundamental importance, challenging a deeply ingrained and nearly universal practice in critical care.
  • Methods: The large, international, multicenter, double-blind, placebo-controlled design was methodologically superb.
  • Results: The trial had a clear and robustly neutral result for its primary outcome of mortality. The positive finding for the secondary outcome of GI bleeding was also clear and statistically significant.
  • Conclusions and Applicability: The authors’ conclusion is a direct and fair interpretation of the data. The results are highly applicable to general ICU practice worldwide and provide a definitive, evidence-based rationale to reconsider the routine, reflexive use of stress ulcer prophylaxis in all at-risk patients.

16. Disclaimer and Contact

This summary is provided by the Academic Committee of ESBICM (ACE) to facilitate the understanding of this study; readers are advised to refer to the original trial document for a deeper understanding. If you find any information incorrect, or missing, or it needs an update or have a request for a specific critical care trial summary, kindly write to us at academics[at]esbicm.org.

Scroll to Top