skip to Main Content

GI BleedingGoal-Directed Management of Gastrointestinal Bleeding

Gastrointestinal bleeding accounts for more than 450,000 hospitalizations annually in the United States and is a major consumer of blood products.1 Blood transfusions are given to 21- 43% of these patients, and acute upper gastrointestinal bleeding has a mortality rate of 10- 14%.2,3 In our large database, GI bleeding consistently ranks in the top 5 list of patient groups receiving blood products, along with cardiac surgery, orthopedics, oncology, and critical care/ trauma.

In a number of hospitals, GI bleeding is actually the #1 consumer of blood products, which often comes as a surprise to hospital leadership.  For reasons discussed below, this is a high risk/ high volume patient population that needs to have a more proactive and evidence-based approach.

Recognizing the prevalence and consequences of GI bleeding, an international panel created consensus recommendations on the management of nonvariceal upper gastrointestinal bleeding first in 20034 with an update published in 2010.2 These consensus recommendations cover a variety of patient care variables, including resuscitation, risk assessment, endoscopy management, pharmacologic management, post-endoscopy in-hospital care and post discharge management.  In spite of the existence of these guidelines, it is highly likely that they have not been universally implemented.  With regards to transfusion management, the 2010 update2 specifically endorsed a transfusion threshold of 7 gm/dL based largely on data from the Transfusion Requirements in Critical Care study5, although they noted that GI bleed patients were excluded from that study.  Also published in 2010 was an observational study of 4,441 acute upper gastrointestinal bleed patients showing that early red blood cell transfusion, which they defined as a transfusion threshold > 8 gm/dL, was associated with a two-fold increase in re-bleeding and a 28% increase in mortality.6

Published this month in the New England Journal of Medicine was a prospective, randomized trial of restrictive vs. liberal transfusion practice in 889 acute upper gastrointestinal bleeding patients7 that seems to definitely answer the question of transfusion strategies.  The study included patients with both variceal and non-variceal upper GI bleeding, excluding only those with “massive exsanguination” or those at very low risk of bleeding.  The results of this large study significantly favored a restrictive approach for both morbidity and mortality, and are in line with the growing body of evidence supporting “less is more.”  Compared to the liberal strategy group, patients managed with a restrictive approach had lower mortality at 45 days (5% vs. 9%; odds ratio of 0.55, p= 0.02), lower rebleeding rates (10% vs. 16%; OR 0.62, p= 0.01) and rates of emergency surgery (2% vs. 6%; p= 0.04), shorter length of stay (9.6 vs. 11.5 days; p= 0.01) and an overall lower rate of serious adverse events (40% vs. 48%; OR 0.73, p= 0.02).  The authors concluded: “in patients with acute gastrointestinal bleeding, a strategy of not performing transfusion until the hemoglobin concentration falls below 7 g/dL is a safe and effective approach.”7

An interesting side note within this study was their use of urgent endoscopy, which was performed in all patients within the first six hours and is a more aggressive timeline than the current standard of care.  While it seems intuitive that a shorter door-to-procedure time might improve outcomes, the consensus guidelines recommend endoscopy within 24 hours, and specifically state that there is no additional benefit of urgent endoscopy (<12 hours) over early endoscopy (>12 hours).2 Does this study also suggest that urgent endoscopy is superior to early endoscopy, and did this strategy contribute to the outcome of this study in some way?

In spite of the statistics that support GI bleeding as a major healthcare issue, my experience is that this patient population has not attracted as much attention as other bleeding patients within hospitals, such as cardiac surgery, orthopedics and trauma.  It is also clear that this problem will continue to grow as the population ages and with the explosion of new anticoagulant drugs, many of which don’t have specific reversal agents.  I see some analogies in the current management of GI bleed patients to the suboptimal management of septic patients prior to the advent of evidence-based, “goal directed” guidelines.8 Implementation of these “Surviving Sepsis” guidelines has resulted in a standardized and coordinated approach to these high risk patients, and has substantially lowered mortality rates.  There is a similar need to standardize practices for transfusion, pharmacologic therapies and endoscopy for GI bleed patients, as well as a significant need to coordinate their care among a wide group of stakeholders.  In the past, these patients were often managed by a single gastroenterologist for their entire hospitalization.  Today, the care of these complex patients typically involves a variety of specialties, including emergency room, critical care, and hospitalist physicians, and gastroenterologists may be viewed more as “proceduralists” rather than the primary attending physician.

What would be the impact of implementing evidence-based protocols for the management of GI bleed patients?  Based upon the NEJM study and putting the impact in terms of “number needed to treat,” 25 patients would have to be treated using a restrictive transfusion strategy (rather than a liberal transfusion strategy) to prevent one additional death at 45 days.1 To be more specific, for the average hospital that treats about 200 GI bleed patients annually, implementing evidence-based protocols would significantly reduce blood use and healthcare costs and would save eight lives each year.  Implemented nationally, these protocols could save millions of healthcare dollars and over 15,000 lives each year.  It is clearly time for the Goal Directed Management of Gastrointestinal Bleeding.

Selected References

  1. Laine L. Blood Transfusion for Gastrointestinal Bleeding. New England Journal of Medicine. 2013;368(1):75–76. Available at:
  2. Barkun AN, Bardou M, Kuipers EJ, et al. International consensus recommendations on the management of patients with nonvariceal upper gastrointestinal bleeding. Annals of internal medicine. 2010;152(2):101–13. Available at:
  3. Crooks C, Card T, West J. Reductions in 28-day mortality following hospital admission for upper gastrointestinal hemorrhage. Gastroenterology. 2011;141(1):62–70. Available at:
  4. Barkun A, Bardou M, Marshall JK. Consensus recommendations for managing patients with nonvariceal upper gastrointestinal bleeding. Annals of internal medicine. 2003;139(10):843–57. Available at:
  5. Hébert PC, Wells G, Blajchman M a, et al. A multicenter, randomized, controlled clinical trial of transfusion requirements in critical care. Transfusion Requirements in Critical Care Investigators, Canadian Critical Care Trials Group. The New England journal of medicine. 1999;340(6):409–17. Available at:
  6. Hearnshaw SA, Logan RFA, Palmer KR, Card TR, Travis SPL, Murphy MF. Outcomes following early red blood cell transfusion in acute upper gastrointestinal bleeding. Alimentary pharmacology & therapeutics. 2010;32(2):215–24. Available at:
  7. Villanueva C, Colomo A, Bosch A, et al. Transfusion strategies for acute upper gastrointestinal bleeding. The New England journal of medicine. 2013;368(1):11–21. Available at:
  8. Dellinger RP, Levy MM, Rhodes A, et al. Surviving Sepsis Campaign. Critical Care Medicine. 2013;41(2):580–637. Available at:
Back to top