For patients with severe acute GI bleeding, outcomes are significantly better when a restrictive transfusion strategy is used – limiting the hemoglobin threshold to 7 g/dL – rather than a liberal transfusion strategy allowing a 9 g/dL threshold, according to a report published online Jan. 2 in the New England Journal of Medicine.
In a single-center randomized controlled trial involving 889 patients, the restrictive transfusion strategy resulted in significantly lower mortality, lower rates of rebleeding, less frequent need for rescue therapy, fewer complications, and shorter hospitalizations than did the liberal transfusion strategy. "Our results suggest that in patients with acute GI bleeding, a strategy of not performing transfusion until the hemoglobin concentration falls below 7 g/dL is a safe and effective approach," said Dr. Càndid Villanueva of Hospital de Sant Pau, Barcelona, and his associates.
"Current international guidelines recommend decreasing the hemoglobin threshold level for transfusion ... from 10g/dL to 7 g/dL" in such patients, but these recommendations are based on trials involving critically ill patients with normovolemic anemia that did not include subjects with acute bleeding. "Transfusion requirements may be different for patients with acute hemorrhage due to factors such as hemodynamic instability or rapid onset of anemia" resulting from extremely low hemoglobin levels.
In particular, results of animal studies suggest that transfusion can be especially harmful in patients with bleeding from portal hypertension sources, "since restitution of blood volume after hemorrhage can lead to a rebound increase in portal pressure, which is associated with a risk of rebleeding," the investigators noted.
To examine the effects of different transfusion strategies in this setting, Dr. Villanueva and his colleagues enrolled adults who presented with hematemesis, melena, or both, randomly assigning 444 to receive restrictive transfusion (with a target range for the posttransfusion hemoglobin level of 7-9 g/dL) and 445 to receive liberal transfusion (with a target range of 9-11 g/dL).
The study protocol permitted transfusions to be administered at the discretion of the attending physician any time symptoms or signs of anemia developed, massive bleeding occurred, or surgical intervention was needed, as well as when hemoglobin levels dipped below the assigned threshold.
All the study subjects underwent emergency gastroscopy within 6 hours of presentation, with appropriate treatment when the source of the bleeding was identified. Diagnoses included peptic ulcer, esophageal varices, cirrhosis, portal hypertension, and nonvariceal lesions.
The primary outcome measure, mortality from any cause at 45 days, was significantly lower in the restrictive-strategy group (5%) than the liberal-strategy group (9%). Death resulted from uncontrolled bleeding in 0.7% vs 3.1% of the 2 groups, respectively, the researchers said (N. Engl. J. Med. 2013 Jan. 2 [doi:10.1056/NEJMoa1211801]).
The rate of rebleeding also was significantly lower with the restrictive strategy (10% vs. 16%), and length of hospital stay was significantly shorter. In addition, rescue therapy for esophageal varices with balloon tamponade or a transjugular intrahepatic portosystemic shunt was required less often in the restrictive-strategy group than in the liberal-strategy group, as was emergency surgery to control further bleeding from peptic ulcer.
The rate of overall complications was significantly lower with the restrictive strategy (40%) than with the liberal strategy (48%), as was the rate of serious adverse events. In addition, transfusion reactions and cardiac events such as pulmonary edema were more frequent with the liberal strategy.
"Our results are consistent with those from previous observational studies and randomized trials performed in other settings, which have shown that a restrictive transfusion strategy did not increase, and even decreased, the mortality observed with a liberal transfusion strategy," Dr. Villanueva and his associates said.