top of page




September 6, 2007

Efficacy and Safety of Epoetin Alfa in Critically Ill Patients.

Mazen Kherallah

Summarized by: 

What was the research question?

  • Does the treatment with recombinant human erythropoietin (epoetin alfa) reduce the need for red-cell transfusions in critically ill patients?

How did they do it?

  • A prospective, randomized, placebo-controlled trial at 115 centers in the USA

  • 1460 critically ill medical, surgical, or trauma patients between 48 and 96 hours and hemoglobin <12 g/dL, were randomized to receive weakly 40,000 U of Epoetin alfa for three weeks (733 patients) or placebo (727 patients).

  • Patients with history of acute ischemic heart disease, PE, DVT, or stroke were excluded.

  • RBC transfusion was left for the discretion of the treating team, but a target of 7-9 g/dL was recommended. EPO was held if Hg >12 g/dL.

  • The primary end point was the percentage of patients who received a red-cell transfusion.

  • Secondary end points included the number of red-cell units transfused, mortality, and the change in hemoglobin concentration from baseline.

What did they find?

  • Percentage of patients who received any RBC transfusion was not significantly different in the EPO group compared to placebo (46% vs. 48.3%, P=0.34).

  • Hg concentration was higher in the EPO group compared to the placebo group (1.6 g/dL vs. 1.2 g/dL, P<0.001).

  • 29-day mortality tended to be lower in the EPO group but not significantly different (8.5% vs. 11.4%, HR 0.79; 95% CI, 0.56 to 1.10)

  • In the pre-specified trauma group, 29-day mortality was significantly better in the EPO group compared to the placebo group (3.5% vs. 6.6%, HR 0.37; 95% CI, 0.19 to 0.72, NNT 32). Same for the 180-day mortality (6% vs. 9.2%, HR 0.86; 95% CI, 0.65 to 1.13).

  • EPO was associated with a higher risk of thrombotic vascular events (16.5% vs 11.5%, HR 1.41; 95% CI, 1.06 to 1.86, P=0.008, NNH 21).

Are there any limitations?

  • Mortality was a secondary endpoint, and the study is not powered to confirm a beneficial effect.

  • Compared to EPO-1 and EPO-2 which showed a decrease RBC requirement, this study did not, but likely secondary to the change in the transfusion threshold in the ICU.

What does it mean?

  • Epoetin alfa did not reduce the number of RBC transfusion or mortality in general ICU patient and increased the incidence of thrombotic events. However, in a predefined trauma group, mortality was decreased.

  • No routine use of epoetin alfa is critically ill patients. More data is needed in traum patients.


bottom of page