Transfusion Strategy in Upper GI Bleeding

Posted by Daniela Lamas • January 2nd, 2013

A 55-year old man presents to the Emergency Department, weak and dizzy following three episodes of dark, blood-streaked emesis.

He’s a long-time drinker with hepatitis C and peptic ulcer disease. Knowing the danger of an upper gastrointestinal bleed, you quickly place two large-bore IV’s and begin volume resuscitation with fluids.

Labs return with a hemoglobin level of 8 g/d. You reflexively move to call the blood bank to order a transfusion, but then pause.

Does this patient actually need a transfusion?

You know that restrictive transfusion strategies have been proven beneficial in the critically ill, but these studies generally excluded patients with acute bleeding from a gastrointestinal source.  When it comes to your patient and the question at hand, the jury is still out.

Until now. In this week’s issue of NEJM, Candide Villanueva and colleagues report their results: even in GI bleeding, less is more.

To address the controversial question, the authors enrolled nearly 1,000 patients with severe acute upper GI bleeds. Half received transfusions with a target hemoglobin of 7 g/dL and the other half were transfused to a hemoglobin of 9 g/dL.

All included in the study had clear evidence of upper GI bleeding, either from vomiting blood or from melena.  The mean hemoglobin at admission was similar in both groups; 9.6 g/dL in the group with the restrictive strategy and 9.4 g/dL in the other group.

All patients underwent emergent upper endoscopy, as is standard practice to investigate bleeding source. Peptic ulcers were identified as the source of bleed in nearly half those enrolled, and esophageal varices in 20 percent.

Those with “massive exsanguinating bleeding” – for whom transfusions could be life-saving – were excluded, as were those with concomitant heart attacks. Patients both with and without liver disease were included, and the groups were stratified according to the presence or absence of cirrhosis.

The results were striking. Mortality at 45 days was significantly lower in the group with fewer transfusions.  Furthermore, the patients with a lower transfusion goal were less likely to re-bleed and were discharged from the hospital sooner.

What might explain these findings?

In their discussion, the authors specifically address why aggressive transfusion might be harmful for patients with cirrhosis and portal hypertension, by increasing portal pressure and leading to more bleeding.  Indeed, those with cirrhosis who were randomly assigned to liberal transfusion parameters had higher portal pressures – leading, perhaps, to the higher rates of further bleeding observed in this group.

But restrictive transfusions also benefited those with bleeds from non-variceal sources – such as ulcers. To address this point, the authors note that transfused blood itself might be harmful, by altering the coagulation cascade and interfering with the body’s own response to blood loss.

How applicable are these results?

In an accompanying editorial, Loren Laine, M.D. writes that two groups – specifically, patients in hypovolemic shock, and those with cardiovascular disease – might still benefit from higher transfusion thresholds. But for the majority of patients, the study’s results are clinically directive: “Most patients with upper gastrointestinal bleeding, with or without portal hypertension, should have blood transfusions withheld until the hemoglobin level drops below 7 grams per deciliter,” Laine writes.

For our patient, then?

Rethink that reflex and put down the phone.

Leave a Reply

Please note: Comment moderation is enabled and may delay your comment. There is no need to resubmit your comment.

Submitting a comment indicates you have read and agreed to the Terms and Conditions.