You are here

Erythropoietin In-Depth: An Experiment of Nature Highlights the Non-Hematopoietic Activity of Erythropoietin

When a paper entitled “Erythropoietin doping in cycling: lack of evidence for efficacy and a negative risk–benefit”1 was published in 2012, I felt a little triumph on several levels.  First, it always feels good to have more evidence that cheaters never prosper.  On a scientific level, the paper supported my bias that evolution has created an organism so finely tuned ,that when normal, it cannot be improved pharmacologically, at least not without a costly downside.   As far back as 1998, approximately two dozen deaths of elite cyclists were anecdotally linked to erythropoietin doping.2

Finally, the 2012 paper seemed to confirm what I’d been teaching medical students for years:  that oxygen delivery is maximal at a hematocrit of about 40%, which is why that value is considered “normal.”  As JH Jandl wrote in his classic textbook Blood: Textbook of Hematology, “That O2 flow at high rates of shear is maximal at physiologic levels of hematocrit reaffirms the splendor of natural selection.”3 The value of 40% for the optimal hematocrit was reported at least as early as 1967 4 and continues to be reported in 2014.5  

Because viscosity is inversely proportional to flow, higher hematocrits with higher viscosities should decrease oxygen delivery and impair athletic performance.  Therefore, I was surprised to learn of Eero Mäntyranta, who had a hematocrit of 60% and was the greatest Finnish cross-country skier of all time, with seven Olympic medals, including three gold, and two World Championships.  He died in 2013 at age 76.  His erythrocytosis was due to a mutation of the gene encoding the erythropoietin receptor which made him hyper-responsive to the hormone erythropoietin, which upregulates erythrocyte production.6  

The author David Epstein suggested that Mother Nature gifted Mäntyranta with what Lance Armstrong and company achieved through technology.7 Further, Epstein wrote “I thought he must have been the strongest seventy-three year old I had ever met.”  

The mutation is transmitted as an autosomal dominant trait.  At least 6 kindreds with mutations of the erythropoietin receptor are known.  These mutations do not appear to affect longevity.  Indeed, the condition is sometimes called “benign familial erythrocytosis.” Many affected individuals are unaware of any abnormality, although hypertension, cardiovascular disease, and headaches, which responded to therapeutic phlebotomy have been reported.   

Contrast with Polycythemia Vera

Polycythemia vera patients can also have hematocrits of around 60%.  The life expectancy of untreated polycythemia vera is 1 ½ to 3 years,8  which demonstrates the profound negative impact of hyperviscosity on health -- much greater than hypercholesterolemia or primary hypertension.  Life expectancy is now 10 to 20 years with therapy, the mainstay of which is therapeutic phlebotomy.7  In a study of polycythemia vera patients with a mean hematocrit of 60%, the incidence of thrombosis was 14% in ten years.9 Why are the prognoses of benign familial erythrocytosis and polycythemia vera so different? 

The answer is the activity of erythropoietin.  Erythropoietin upregulates endothelial expression of nitric oxide, which is a vasodilator and inhibits platelet activation.  Of course, these are useful properties of a hormone which increases red cell mass and blood viscosity.  In subjects with benign familial erythrocytosis, the former action can normalize total peripheral resistance despite the extremely high viscosity caused by a hematocrit of 60%, in which case normalized peripheral resistance would limit thrombosis. Erythropoietin levels are lower in polycythemia vera than in any other disease,10 which undoubtedly influences the negative prognosis.

Erythropoietin activity can also explain the lack of atherosclerosis in adult patients with cyanotic congenital heart disease, in whom hematocrits can exceed 70%.    

Extramural coronary arteries in [cyanotic congenital heart disease] dilate in response to endothelial vasodilator substances supplemented by mural attenuation caused by medial abnormalities. Basal coronary flow was appreciably increased, but hyperemic flow was normal. Remodeling of the microcirculation was responsible for preservation of flow reserve. The coronaries were atheroma-free because of the salutary effects of hypocholesterolemia, hypoxemia, upregulated nitric oxide, low platelet counts, and hyperbilirubinemia.11

In particular, increased blood flow, hypocholesterolemia, increased nitric oxide levels, and thrombocytopenia would limit development of mural thrombi, the precursor lesion to the atherosclerotic plaque.  I suspect that the vascular remodeling of the microvasculature and especially the epicardial coronaries, which involves loss of smooth muscle and fibrosis, may be due to long-term effects of erythropoietin among other mediators and would not occur to a sufficient degree when erythropoietin is administered alone for athletic training.

Polycythemia in Neonates

Neonatal polycythemia is an interesting disease because of its mixed etiologies.  When caused by fetal hypoxia, the disease is driven by erythropoietin.  When caused by delayed clamping of the umbilical cord or placing the newborn below the level of the placenta, the polycythemia is not driven by erythropoietin.  These two groups should be studied separately, because the former group is protected by erythropoietin, while complications of hyperviscosity disproportionately affect the latter.  Instead, every review I have read lumps the two conditions together, and consequently, data on prognosis and recommendations for management are murky.12  

Erythropoietin Responsiveness and Pathophysiology

It is possible that the detrimental hyperviscosity of erythrocytosis can be mitigated by erythropoietin.  Then, why do 20% to 30% of anemic renal failure patients on supplemental erythropoietin develop hypertension as a consequence of the hyperviscosity?  The answer could be the presence of truncated (as opposed to mutated) erythropoietin receptors on endothelial cell membranes.  The truncated receptors lack the intracytoplasmic domains, which activate the downstream intracellular pathways responsible for nitric oxide production.  The ratio of truncated to full erythropoietin receptor mRNA in circulating endothelial progenitor cells correlated with the change in blood pressure following initiation of erythropoietin therapy.13  

Soluble erythropoietin receptor, a truncated protein which contains only the extracellular domain of the receptor, has the beneficial activity of antagonizing erythropoietin and decreasing red cell mass in order to normalize blood viscosity. The presence of dysfunctional receptors on endothelial cells could be due to incorrect post transcriptional modification of erythropoietin receptor mRNA.      

Physically, is the Mäntyranta kindred a breed of new, improved humans?  Mäntyranta’s nephew Pertti won an Olympic skiing gold medal in 1976 and a bronze in 1980, and his niece Elli was twice world champion in the 3x5K relay in 1970 and 1971.14  The mutation is energetically costly, no doubt, but in the 21st century culture of overabundant consumption, that is not a major drawback.   

We may be experiencing the rare privilege of watching the rise of a beneficial mutation, the evolutionary equivalent of finding a needle in a haystack.  At the least, the Mäntyranta story demonstrates that evolution has insured that it is better to come by your erythrocytosis honestly. 


1. Erythropoietin doping in cycling:  lack of evidence for efficacy and a negative risk-benefit.  Heuberger JAAAC, Tervaert JMD, Schepers FML, Vliegenthart ADB, Rotmans JI, Daniels JMA, Burggraaf J, Cohen AF.  British Journal of Clinical Pharmacology 2012; 75(6):  1406-1421.

2. Backtalk; Lifesaving Drug Can Be Deadly When Misused.  Longman J.  NY Times, July 26, 1998. 

3. Blood: Textbook of Hematology.  Jandl JH.  Little, Brown, and Company. Boston, 1996, p.159.

4. Determinant of the optimal hematocrit.  Crowell JW, Smith EE.  J Appl Physiol 1967; 22:  501-504.

5. What can we learn from Einstein and Arrhenius about the optimal flow of our blood?  Schuster S, Stark H.  Biochimica et Biophysica Acta 2014;1840(1):271-6.

6. Truncated erythropoietin receptor causes dominantly inherited benign human erythrocytosis.  de la Chapelle A, Traskelin A-L, Juvonen E.  Proc Natl Acad Sci USA 1993; 4495-4499.

7. Eero Mäntyranta.  Epstein D. [accessed 4/14/2014].

8. Polycythemiavera.BesaEC.KoyamangalathK,ed.

9. Polycythemia vera in young patients: a study on the long-term risk of thrombosis, myelofibrosis and leukemia. Passamonti F, Malabarba L,Orlandi E, Barate C, Canevari A, Brusamolino E, Bonfichi M, et al.  Haematologica 2003;88:13-18.

10. Polycythemia vera: myths, mechanisms, and management.  Spivak JL. Blood 2002; 100(13):  4272-4292.

11. Cyanotic Congenital Heart Disease: The Coronary Arterial Circulation.  Perloff JK. Curr Cardiol Rev. 2012; 8(1): 1–5.

12. Polycythemia of the newborn. Karen J Lessaris KJ.  Rosenkrantz T, ed.

13. Hypertension induced by erythropoietin has a correlation with truncated erythropoietin receptor mRNA in endothelial progenitor cells of hemodialysis patients.  Ioka T, Tsuruoka S, Ito C, Iwaguro H, Asahara T, Fujimura A, Kusano E.  Clinical Pharmacology & Therapeutics 2009; 86(2):  154-159.

14. The Sports Gene.  Epstein D.  CURRENT, New York, 2013, p. 281.


LAST UPDATED:  2014-12-30


Stay Connected