Does Epo save lives?

Efficacy and Safety of Epoetin Alfa in Critically Ill Patients
Corwin et al. NEJM Volume 357:965-976. September 6, 2007.

A multicenter study was conducted whose primary objective was to demonstrate that the administration of epoetin alfa to critically ill subjects reduces the proportion of subjects requiring red blood cell (RBC) transfusion as compared with placebo. (ClinicalTrials reference)

Secondary outcomes included:
Cumulative number of units of red blood cell transfusions received from Study Day 1 through Study Day 42.
Change in hemoglobin from Study Day 1 through Study Day 29.
Mortality through Study Day 29.
Cumulative mortality through Study Day 140.

The study was a prospective, randomized, placebo-controlled trial, which enrolled 1460 medical, surgical, or trauma patients between 48 and 96 hours after admission to the intensive care unit. Epoetin alfa (40,000 U) or placebo was administered weekly, for a maximum of 3 weeks; patients were followed for 140 days.

What did they find?

1) No difference in red cells transfused.
2) Higher hemoglobin in the epoetin alfa group. (Hemoglobin increase = 1.6±2.0 g per deciliter vs. 1.2±1.8 g per deciliter, P<0.001)
3) A tendency towards decreased mortality in the epoetin alfa group, especially in the trauma subgroup. (adjusted hazard ratio for death at 140 days in the trauma subgroup, 0.40; 95% CI, 0.23 to 0.69)
4) Increased thrombotic events in the epoetin alfa group. (hazard ratio, 1.41; 95% CI, 1.06 to 1.86)

Are the findings important?

If epoetin alfa is associated with decreased mortality (independent of blood transfusion!), this is an important and new finding. The finding appeared to be most robust in the trauma patients. With any unexpected finding, it is necessary to ask whether this is merely a chance finding. Statistical significance alone is insufficient to justify a change in clinical practice. There has to be biological plausibility for a finding. The authors speculate that it may be the action of epoeitin alfa as a cytokine with antiapoptotic activity that leads to the reduction in mortality.(JAMA. 2005;293:90-95. Kidney International (2006) 70, 246–250. Critical Care 2004, 8:337-341.) {Link to a diagram showing possible cytoprotection with epoeitin alfa}

Are there safety concerns?

Absolutely. This study found an increased risk of thrombosis in the epoeitin alfa group. Previous studies in patients with renal failure and with cancer have shown that epoeitin alfa is associated with increased risk of thrombosis and even of increased mortality. (Lancet 2003; 362(9392): 1255-1260. J Clin Oncol 2005;23:5960-5972. J Natl Cancer Inst 2006;98:708-714. N Engl J Med 2006;355:2085-2098.)


Erythropoietin (Epo) binds its receptor (Epo R) on the surface of red blood cell progenitors in the bone marrow causing proliferation, maturation, and differentiation, thereby preventing or correcting anemia. Epo may also bind Epo R expressed on the surface of cancer cells and may elicit tumor growth via cell proliferation, protection from apoptosis, and/or angiogenesis. (See - Brower, V. (2003) Nat. Med. 9:1439.)

So what should the intensivist do?

1. Administer epoetin alfa to all anemic critically ill patients unless it is specifically contra-indicated.
2. Exercise clinical judgment (e.g. consider epoetin alfa in certain settings such as acute hemorrhage where patients are otherwise healthy and there is no known risk for thrombosis or cancer).
3. Await the results of further studies.
4. Conduct further studies yourself.
5. Be highly skeptical and take the view that this was an industry funded study of an expensive drug that showed questionable benefit and heightened known safety concerns.

Further Reading on Erythropoietin (Review articles with free full text):

• Arch Intern Med. 2004 Feb 9;164(3):262-76.
• Blood. 2000 Aug 1;96(3):823-33.
• Cancer. 2006 Jan 1;106(1):223-33.
• Chest. 2005 Nov;128(5 Suppl 2):576S-582S.
• Chest. 2005 Nov;128(5 Suppl 2):568S-575S.