European Journal of Applied Physiology

, Volume 101, Issue 4, pp 481–486

Prolonged administration of recombinant human erythropoietin increases submaximal performance more than maximal aerobic capacity


  • J. J. Thomsen
    • Copenhagen Muscle Research CentreRigshospitalet
  • R. L. Rentsch
    • Department of Molecular BiologyUniversity of Copenhagen
  • P. Robach
    • ENSA
  • J. A. L. Calbet
    • Department of Physical EducationUniversity of Las Palmas de Gran Canaria
  • R. Boushel
    • Department of Exercise ScienceConcordia University
  • P. Rasmussen
    • Copenhagen Muscle Research Centre, Department of AnaesthesiaRigshospitalet
  • C. Juel
    • Department of Molecular BiologyUniversity of Copenhagen
    • Department of Sport ScienceUniversity of Århus
    • Copenhagen Muscle Research CentreRigshospitalet
Original Article

DOI: 10.1007/s00421-007-0522-8

Cite this article as:
Thomsen, J.J., Rentsch, R.L., Robach, P. et al. Eur J Appl Physiol (2007) 101: 481. doi:10.1007/s00421-007-0522-8


The effects of recombinant human erythropoietin (rHuEpo) treatment on aerobic power (VO2max) are well documented, but little is known about the effects of rHuEpo on submaximal exercise performance. The present study investigated the effect on performance (ergometer cycling, 20–30 min at 80% of maximal attainable workload), and for this purpose eight subjects received either 5,000 IU rHuEpo or placebo every second day for 14 days, and subsequently a single dose of 5,000 IU/placebo weekly/10 weeks. Exercise performance was evaluated before treatment and after 4 and 11 weeks of treatment. With rHuEpo treatment VO2max increased (P < 0.05) by 12.6 and 11.6% in week 4 and 11, respectively, and time-to-exhaustion (80% VO2max) was increased by 54.0 and 54.3% (P < 0.05) after 4 and 11 weeks of treatment, respectively. However, when normalizing the workload to the same relative intensity (only done at time point week 11), TTE was decreased by 26.8% as compared to pre rHuEpo administration. In conclusion, in healthy non-athlete subjects rHuEpo administration prolongs submaximal exercise performance by about 54% independently of the approximately 12% increase in VO2max.




In humans maximal aerobic power (VO2max) is limited by oxygen delivery rather than metabolic capacity (di Prampero and Ferretti 1990;Turner et al. 1993;Wagner 2000), and when arterial oxygen content (CaO2) is increased VO2max is also increased (Calbet et al. 2006). At sea level CaO2 is mainly determined by the haemoglobin concentration, which may be altered by several means. Berglund and Ekblom (1991) performed the first study investigating the effects of rHuEpo administration on [Hb] and VO2max. They injected 20–40 IU kg−1 body mass for a period of up to 6 weeks and demonstrated an increase in [Hb] and concomitant also in VO2max. Subsequently several other groups have confirmed that injecting between 50–200 IU kg−1 at a frequency of about three times a week for 10–30 days increases VO2max (Audran et al. 1999; Birkeland et al. 2000; Parisotto et al. 2001), and also lower dosages (20–50 IU kg−1) have been shown to induce similar results (Russell et al. 2002). Although VO2max is a commonly used measure for aerobic performance, it was recently shown in top professional cyclists (all with several remarkable international victories) that VO2max varied from 65.5 to 82.5 ml kg−1 min−1 (Lucia et al. 2002). The lower value corresponds to levels also observed in amateur athletes whereas the higher value is associated with world-class endurance athletes. Although a relatively high VO2max is needed for outstanding endurance performance, also other factors such as exercise economy, the ability to exercise at a given fractional percentage of VO2max, and lactate metabolism have been shown to play a capital role (Bassett and Howley 2000). Although the positive effect of rHuEpo treatment on VO2max is clearly established, it remains unknown as to what its impact is on endurance performance.

Thus, the aim of the present study was to investigate if prolonged rHuEpo administration increases aerobic endurance performance. For this purpose eight subjects received either 5,000 IU rHuEpo or saline every second day for 14 days, and subsequently a single dose of 5,000 IU/saline weekly for 10 weeks. The exercise tests were performed prior to any injection and after 4 and 11 weeks of rHuEpo/saline treatment.


A total of 16 healthy male volunteers participated in the study. They were divided into two groups, one receiving rHuEpo (Epo, age: 27 ± 7 (mean ± SD) years; height: 180 ± 4 cm; weight: 83 ± 7 kg) and one receiving saline injection (Con, age: 26 ± 5 years; height: 178 ± 5 cm; weight: 82 ± 6 kg). The study was approved by the local ethical committee of the communities of Copenhagen and Frederiksberg and conformed to the Declaration of Helsinki. All subjects gave written informed consent to participate. This study was part of a larger experimental protocol.

Experimental design

The treatment was blinded to the Con group (unaware if treated with rHuEpo or placebo, single-blinded), but not to the Epo group (aware that they were administered rHuEpo). This was done due to ethical considerations associated with the very invasive experimental setup used in other parts of this study (Epo group only, to be published elsewhere). The rHuEpo treatment aimed to increase the haematocrit rather rapidly to around 50% and keep it at this level for the remaining study period. Two weeks prior to rHuEpo/saline treatment all subjects received 100 mg iron day−1 orally, and this was maintained throughout the entire study period. Following baseline measurements a dose of 5,000 IU (∼60 ± 4 IU/kg body mass) of rHuEpo (NeoRecormon, Roche, Mannheim, Germany) dissolved in 0.3 ml saline was injected as follows—first 2 weeks, one injection every second day; third week, three injections on three consecutive days; week 4–13, one injection every week. All injections occurred between 0800 and 1000 hours, and were preceded by 30 min of supine rest and a venous blood sample (6 ml) for analyses. See Lundby et al. (2007) for further details.

Exercise tests

The exercise test consisted of (1) incremental bike exercise until exhaustion (VO2max) and (2) time to exhaustion (TTE) at a given percent of the maximum attained workload during the incremental exercise test. Prior to the experiments all subjects performed both tests to familiarise themselves with this type of testing. During these tests the subjects chose their cadence themselves (allowed between 70 and 120 rpm), and the chosen cadence was kept for the remaining study. Following these lead-in practice trials the experiments started and both exercise trials were conducted before rHuEpo/saline treatment (Pre), and again after 4 (4 W) and 11 (11 W) weeks of treatment. The incremental exercise test was initiated with 15 min of warm-up at 100 W on a bike ergometer (Monark E839, Varberg, Sweden). Then the workload was increased by 40 W every 90 s until exhaustion. Pulmonary VO2 and CO2 production (VCO2) were measured continuously (Quark b2, Cosmed, Rome, Italy). Before each test ambient conditions were measured, then the gas analyser and the flowmeter were calibrated with high precision gases and a 3-L calibration syringe, respectively. During submaximal and maximal exercise the VO2 was recorded as averages of 30 s intervals. TTE was also performed on the bike ergometer and was initiated with a 15 min warm-up at 150 W after which the workload was increased to 80% of the maximal attained workload in the pre-incremental exercise test. All TTE tests were completed at this level (same absolute workload) and were not changed during the duration of the experimental period. At time point 11 W, however, an additional TTE test was performed at 80% of the actual VO2max (same relative workload). The subjects were encouraged to exercise for as long as possible through strong verbal motivation and music. During the test blood samples were obtained from a forearm vein every 5 min and at termination of exercise.

Blood analysis

Resting venous blood samples were analysed for haematocrit and haemoglobin concentration by a Radiometer ABL-700 (Radiometer, Copenhagen, Denmark), and also exercise lactate was quantified using this machine. Arterial oxygen saturation (SaO2) was measured by pulse oximetry (Datex, Helsinki, Finland).

DEXA scanning

All Epo subjects were scanned for body composition in a dual-energy X-ray absorptiometer (DEXA; Lunar, GE Medical Systems, Madison, WI, USA) before and after the study period. Results are shown in Table 1.
Table 1

Body mass index (BMI, kg/m2), body fat % (Fat, %), weight (Weight, kg), bone mineral content (g/kg), fat mass (Fat mass, kg), and muscle mass (Muscle mass, kg) in all subjects before (Pre Epo) and after (Post Epo) 14 weeks of rHuEpo treatment. Values are mean ± SD


Pre Epo

Post Epo

BMI (kg/m2)

25.2 ± 1.7

25.6 ± 2.0

Body Fat (%)

17.5 ± 7.2

18.2 ± 5.8

Weight (kg)

82.0 ± 7.1

83.2 ± 6.6

BMC (g kg−1)

3.3 ± 0.2

3.3 ± 0.2

Fat mass (kg)

13.9 ± 6.3

14.6 ± 5.1

Muscle mass (kg)

64.8 ± 6.1

65.3 ± 6.0

Statistical analysis and calculations

Arterial oxygen content was determined as (Hb(g/l) × 1.34 × SaO2) + (PaO2 × 0.03). PaO2 was estimated from the SaO2. Differences between groups were determined by two-way ANOVA followed by Tukey’s post hoc test. Statistical difference was set to P < 0.05. Values reported are mean ± SD.


Blood and exercise data

The rHuEpo treatment increased CaO2 by 11.0 and 9.4% (P < 0.05) in week 4 and 11, respectively, and as expected no differences were found in the placebo group (Table 2).
Table 2

Haemoglobin (Hb, g l1), haematocrit (Htc, %), arterial oxygen content (CaO2, ml l1), and maximal oxygen uptake (VO2max, l min1) before (Pre), and after 4 (4 W) and 11 (11 W) of treatment with saline (Con) or rHuEpo (Epo)












144.8 ± 10.2

145.7 ± 9.8

143.3 ± 8.9

161.8 ± 9.4*§

145.6 ± 3.2

159.4 ± 13.4*§


44.4 ± 2.8

44.7 ± 3.0

43.9 ± 3.1

49.5 ± 2.8*§

44.8 ± 2.3

48.8 ± 4.0*§


190.2 ± 10.9

191.4 ± 12.9

189.1 ± 10.8

212.5 ± 12.3*§

191.3 ± 12.9

209.3 ± 17.6*§


319 ± 18

330 ± 19

323 ± 17

374 ± 25*§

318 ± 21

371 ± 11*§


3.76 ± 0.4

3.95 ± 0.4a

3.81 ± 0.3

4.31 ± 0.5a

3.75 ± 0.4

4.27 ± 0.2a

Values are mean ± SD

aN = 7

* P < 0.05 compared to Pre, §P < 0.05 compared to Con

The test–retest coefficient of variation for the incremental exercise test was 3.2%, and 8.8% in the time-to-exhaustion test. Both values are well below the found differences in VO2max and TTE of rHuEpo treatment, as reported below.

With rHuEpo treatment VO2max increased (P < 0.05) by 9.1 and 8.1% in week 4 and 11 respectively, whereas no changes were found in the placebo group (Table 2). During the VO2max tests gross efficiency remained unchanged for both groups throughout the study period, and when the whole body economy was calculated as the slope of the regression relating exercise work load to pulmonary oxygen uptake, economy was: 11.6 ± 0.1 and 11.2 ± 0.1 ml O2 min−1 W−1 in the Con and Epo group before treatment and 11.9 ± 0.1, 11.6 ± 0.1 ml O2 min−1 W−1 in week 4 and 11.4 ± 0.1 and 11.2 ± 0.1 ml O2 min−1 W−1 in week 11, respectively.

Workloads for the TTE test were 264 ± 17 W for the rHuEpo group, and 255 ± 16 W for the placebo group during all TTE tests. In week 11 TTE tests were also conducted at the same relative exercise intensity as before saline/rHuEpo treatment, i.e. at 80% of the week 11 VO2max, and the workload was 297 ± 13 and 254 ± 15 W for the rHuEpo and saline group, respectively (Table 2). With rHuEpo treatment TTE was increased by 54.0 and 54.3% (P < 0.05) after 4 and 11 weeks of treatment, respectively (Fig. 1). When workload was normalised in week 11 to elicit the same relative exercise intensity, the TTE was decreased by 26.8% (P < 0.05) compared to pre-values.
Fig. 1

Time to exhaustion (min) at 80% of Pre treatment VO2max before (Pre) and after 4 (4W) and 11 (11W) of treatment with saline (Con) or rHuEpo (Epo). Hatched bars indicate that the exercise intensity was normalized to the same relative exercise intensity as before rHuEpo treatment. Values are mean ± SD. *P < 0.05 compared to Pre, §P < 0.05 compared to Con

Saline treatment had no effects on venous lactate concentration during TTE exercise (Fig. 2). In the rHuEpo treatment group, however, venous lactate was decreased (P < 0.05) during the warm-up period, and also in the sample obtained 10 min into TTE exercise in week 4 and 11. When workload was normalised to the same relative exercise intensity, 10 min lactate concentration was increased (P < 0.05) as compared to the pre-treatment value (Fig. 2).
Fig. 2

Venous lactate (mmol l−1) before (Pre) placebo (Con) or rHuEpo (Epo) treatment during warm-up (Warm-up), 10 min of TTE exercise (10 min), and at termination of exercise (Max), and after 4 (4W) and 11 weeks (11W) of treatment with placebo/rHuEpo. Hatched bars in week 11 indicate TTE exercise intensity normalised to 80% of present VO2max. Values are mean ± SD. *P < 0.05 from pre treatment value


The main findings of this study were that (1) following an aggressive two-week rHuEpo injection period, a weekly low dose rHuEpo injection is sufficient to increase VO2max after 4 and 11 weeks of administration, and (2) rHuEpo administration significantly prolonged TTT at 80% of maximal workload.

VO2max and submaximal cycling performance

The presented increase in VO2max of 9.1 and 8.1% after 4 and 11 weeks, respectively, falls in line with previously published data ranging from 6 to 9% (Audran et al. 1999; Birkeland et al. 2000; Parisotto et al. 2001; Russell et al. 2002). In these studies, however, rHuEpo was injected at a much higher frequency than in the present study, and the treatment period was considerably shorter, i.e. three times a week with 50 IU kg−1/25 days (Parisotto et al. 2001), three times a week with 70 IU kg−1/4 weeks (Birkeland et al. 2000), or even daily dosing with 50 IU kg−1 for 26 days (Audran et al. 1999). In a recent study by Russel and co-workers (2002) a dose of 50 IU kg−1 was injected three times a week for three weeks followed by 20 IU kg−1 three times a week for 4 weeks, and for the first time it was demonstrated that low-dosage administration of rHuEpo may also increase VO2max. Considering the short half-time of rHuEpo (McMahon et al. 1990), there seems to be a fair chance for national and international anti-doping agencies to catch athletes employing injection regimes with a frequency as described in the previous studies. However, with one weekly injection of 5,000 IU (∼60 IU/kg body mass) as described in the present investigation, the difficulty for doping agencies to strike down on rHuEpo misuse seems to increase considerably. Interestingly, Russel et al. (2002) demonstrated that VO2max might remain improved for up to 4 weeks after terminating rHuEpo administration, making detection even more difficult. Although the lifespan of an erythrocyte is on average 120 days, it remains to be seen that haematocrit and haemoglobin concentration will remain elevated for such long periods after termination of rHuEpo injections. In fact, increases in [Hb] due to acclimatisation to altitude falls rather rapidly on return to sea level, with a destruction of the newest blood cells occurring first (Rice et al. 2001).

Although numerous studies have determined the effects of rHuEpo injections on VO2max, no study has directly assessed the submaximal performance benefits associated with rHuEpo injections in human subjects. Since only very few athletic competitions are performed at VO2max, this may have created the false assumption that performance is enhanced to the same extend as VO2max. Time-to-exhaustion tests performed at the same absolute exercise intensity was increased by 54.0 and 54.3% after 4 and 11 weeks of treatment, respectively. Of note, however, is that time-to-exhaustion tests are likely to increase more with a given stimulus than would be the case for time-trial type of exercise (Burgomaster et al. 2005; Leddy et al. 2007). To date only one study has evaluated the effects of blood doping on performance at sea level. In a double-blind, placebo, crossover, experimental design, subjects significantly improved their 10 km running time after injection with 400 mL of autologous, previously frozen deglycerolized red blood cells (Brien and Simon 1987). This study and the present results indicate the powerful submaximal stimulus, most important for athletic performance, of autologous blood doping or rHuEpo administration. It should be noticed that the main effects of rHuEpo administration are on cardiovascular parameters, and that for example the skeletal muscle is likely not to have undergone similar changes, or at least not to similar extent. Thus, the increments in performance observed with rHuEpo administration or autologous blood doping are different from those obtained through training where the most critical steps in the oxygen transport system are optimised.

At first glance it would seem obvious that the main reason for VO2max and TTE performance to be increased following rHuEpo treatment is caused by the augmented oxygen carrying capacity of the blood. However, also other factors such as a potential increase in total blood volume could have increased exercise performance. In some studies, but not in all, it has been shown that plasma volume expansion increases exercise performance (Warburton et al. 2000). However, as we have recently reported, no significant changes were observed in the whole blood volume in response to rHuEpo treatment in these subjects (Lundby et al. 2007).

In the TTE tests performed at the same absolute exercise intensity, the increased performance could in part have been a consequence of the reduction in relative exercise intensity, i.e. 80% (pre) versus 68% (4W) and 69% (11W). At 80% of VO2max (same relative intensity) TTE was 26.8% less at week 11 than before rHuEpo treatment. The increase in TTE performance can therefore, at least to some extend, be related to the reduction in relative exercise intensity. The fact that TTE was reduced at 80% of VO2max after rHuEpo treatment is likely related with the curvilinear relationship between endurance time and exercise intensity, but also indicates that the ergogenic effect of rHuEpo treatment cannot be explained by the improvement in VO2max alone.

Interestingly, it has recently been demonstrated that rHuEpo may also have effects on brain function (for review see (Jelkmann 2005), and healthy subjects treated with rHuEpo may feel improvements in mood (Miskowiak et al. 2007) and perceived physical conditioning (Ninot et al. 2006). Both these findings could also have affected physical performance somewhat in the present study.


This study was conducted with normal healthy subjects and it remains unknown if the performance enhancements observed in this study may also be valid for elite athletes. Although VO2max and time-to-exhaustion tests do not reflect the exercise task that for example participants in the Tour de France have to overcome, the data presented in the present study could indicate the difficulties in competing at the top level when other cyclists are using rHuEpo. Also, in comparison to altitude training camps such as “live high, train low” where the hypoxic stimulus of the altitude is used to increase Hb mass in order to increase performance, potential doping with rHuEpo seems more effective. Since rHuEpo treatment may cause activation of the Epo receptor and for example induce angiogenesis, rHuEpo abuse could also be more effective than with autologous reinfusion of blood.

It is generally accepted that TTE tests are associated with a larger coefficient of variation as compared to time-trial testing covering a fixed distance (Jeukendrup et al. 1996). However, in the present experiment the coefficient of variation between the pre-experiment practice trial and the experimental trial prior to any treatment was 8.8%, and thus well below the reported improvement of approximately 55% with rHuEpo treatment.

In conclusion, in healthy non-athlete subjects rHuEpo administration prolongs submaximal exercise performance by about 54% independently of the approximately 12% increase in VO2max.


This study was supported financially by Anti Doping Denmark (CL), Kunststyrelsen (CL), and by the Bundesamt für Sport (BASPO) in Magglingen, Switzerland.

Copyright information

© Springer-Verlag 2007