A recent study shows that holding off on blood transfusions can improve survival with severe acute upper gastrointestingal bleeidng (NEJM 2013; 368: 11-21). This finding is not unexpected as this has been shown in observational studies. In addition, in critical care patients without acute GI bleeding, a restrictive approach to transfusions has also been beneficial.
This study which enrolled 921 patients (>18 years) assigned 461 to a restrictive transfusion strategy (transfusion if <7 g/dL or at discretion of physician) and 460 to a “liberal” strategy (transfusion if <9 g/dL).
In additon to fewer transfusions, the restrictive group had improved survival:
- 225 in the restrictive group did not require a blood transfusion compared with 65 in the liberal group
- Survival at 6 weeks: 95% in the restrictive group compared with 91% in the liberal group. Hazard ratio 0.55 (confidence interval 0.33 to 0.92) –a 45% reduction in the relative risk of 45-day mortality.
- Recurrent bleeding occurred in 10% of the restrictive group compared with 16% of the liberal group.
- The patients with cirrhosis (and Child-Pugh class A or B) were most likely to benefit from a restrictive approach with hazard ratio of 0.30. Child-Pugh class C did not have a benefit from a restrictive approach with hazard ratio of 1.04. With the liberal approach, there was a higher portal-pressure gradient within the first five days.
The reasons for bleeding in this study included peptic ulcers in about 50%, and varices in 24%. The other causes included Mallory-Weiss tears, erosisve gastritis/esophagitis, and neoplasms.
Why does giving less blood result in better outcomes?
- Transfusion may impair hemostasis in several ways. It may result in abnormalities in coagulation properties. It may counteract splanchnic vasoconstrictive response. And, in those with cirrhosis, it can increase portal pressure (even in the presence of somatostatin). All of these mechanims may increase rebleeding.
- In addition, systemic effects from transfusion can include circulatory overload and pulmonary edema.
Also (unrelated to this posting), a thoughtful comment to a recent post on FDA regulations was posted by Ben Gold (Can the FDA prohibit free speech? | gutsandgrowth).
Related blog references: