Restrictive transfusion improves outcomes in acute upper GI bleeding

Reuters Health Information: Restrictive transfusion improves outcomes in acute upper GI bleeding

Restrictive transfusion improves outcomes in acute upper GI bleeding

Last Updated: 2017-04-03

By Anne Harding

NEW YORK (Reuters Health) - Following a restrictive blood-transfusion policy in patients with acute upper gastrointestinal bleeding reduces mortality and rebleeding, a new systematic review and meta-analysis suggests.

“These results support more widespread implementation of restrictive transfusion policies for adults with acute upper gastrointestinal bleeding,” Dr. Vipul Jairath of Western University in London, Canada, and colleagues write in their report, online March 23 in The Lancet Gastroenterology and Hepatology.

Restrictive transfusion is increasingly being used by medical and surgical specialties, the researchers note. This trend, as well as the emergent nature of acute upper GI bleeding, makes it difficult to conduct clinical trials comparing restrictive and liberal transfusion policies in these patients, they add.

In the new study, Dr. Jairath and his team reviewed four published randomized controlled trials and one unpublished trial, including a total of 1,965 patients. Patients treated with restrictive policies received 1.73 fewer RBC units, on average, than those treated with liberal transfusion policies.

Restrictive transfusion reduced the risks of all-cause mortality by 35% (relative risk, 0.65; p=0.03).

There was no significant difference in the mortality reduction seen in patients with cirrhosis and nonvariceal GI bleeding, respectively. However, mortality changes differed in patients with and without ischemic heart disease at baseline (RR, 4.38 vs. 0.58; p for interaction=0.03).

The overall absolute risk reduction was 2.22% for all-cause mortality with restrictive transfusion, and the number needed to treat (NNT) to prevent one death was 45.

The relative risk for rebleeding was 0.58 for patients managed with restrictive transfusion (p=0.004), with no differences among patients with cirrhosis, those with non-variceal bleeding, and those with ischemic heart disease.

The absolute risk reduction for rebleeding was 4.21% with the restrictive strategy, and the NNT to prevent rebleeding was 24.

There were no differences in acute myocardial infarction, ischemic stroke or acute kidney injury between the restrictive and liberal transfusion groups.

Dr. Ernst Kuipers of Erasmus MC University Medical Center in Rotterdam, the Netherlands, who wrote an accompanying editorial, told Reuters Health by email, “The key message is that restrictive blood transfusion should indeed be used for patients with upper GI bleeding, striving for a threshold hemoglobin value of 70 g/l.”

“So far, we have insufficient data to determine whether this is also preferable in patients with ischemic heart disease,” Dr. Kuipers added. “We know that patients with ischemic heart disease are on one hand more sensitive to impaired oxygen supply as a result of hypovolemia (less circulating fluid due to blood loss) and anemia. On the other hand, they are also more sensitive to fluid overload. With this balance, we assume that restrictive transfusion also benefits patients with ischemic heart disease, but we need further study to definitely answer this.”

Breaking with routine management of patients can be difficult, Dr. Kuipers noted. “In a patient with blood loss, in this case due to upper GI bleeding, it may feel as a counterintuitive challenge to be restrictive with blood transfusion, yet the evidence is that this approach benefits patients, apart from also being cost-saving.”

Dr. Jairath was not available for an interview by press time.


Lancet Gastroenterol Hepatol 2017.

© Copyright 2013-2019 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.