Journal of Applied Physiology

Testing for recombinant human erythropoietin in urine: problems associated with current anti-doping testing

Carsten Lundby, Niels J. Achman-Andersen, Jonas J. Thomsen, Anne M. Norgaard, Paul Robach


Background: the main action of recombinant human erythropoietin (rHuEpo) is to increase the oxygen carrying capacity of the blood. To prevent a possible misuse of rHuEpo, this is tested in urine samples collected from athletes by World Anti-Doping Agency (WADA)-accredited laboratories. Recently the test has met serious critiques, and the aims of the present study were to investigate the detection power of the test as well as the variability in the test power comparing the results of two WADA-accredited laboratories. Methods: eight human subjects were studied for 7 wk and treated with rHuEpo for 4 wk with 2 wk of “boosting” followed by 2 wk of “maintenance” and a post period of 3 wk. Urine samples were obtained during all periods. Results: laboratory A determined rHuEpo misuse in all subjects during the boosting period, whereas laboratory B found no misuse, with one sample to be negative, and the remaining seven to be suspicious. The detection rates decreased throughout the maintenance and post period when total hemoglobin mass and exercise performance were elevated. During this period, laboratory A found only two of 24 samples to be positive and three to be suspicious, and laboratory B found no positive or suspicious samples. Conclusion: this study demonstrates a poor agreement in test results comparing two WADA-accredited laboratories. Moreover, after the initial rHuEpo boosting period the power to detect rHuEpo misuse during the maintenance and post periods appears minimal.

  • doping
  • detection
  • exercise

erythropoietin (Epo) is a glycoprotein hormone that is mainly produced in the kidney (7) and enhances the oxygen carrying capacity of the blood by increasing hemoglobin mass and concomitantly decreasing plasma volume (14). Injections with recombinant human Epo (rHuEpo) greatly increase aerobic performance in humans (6, 17) and although the majority of the increase in performance is associated with oxygen transport, non-Hb mediated functions of rHuEpo may also be speculated to increase performance (see Ref. 13 for references). rHuEpo is banned for performance enhancement by the World Anti-Doping Code. Accredited World Anti-Doping Agency (WADA) laboratories test for potential rHuEpo doping using a method developed by the research group of Lasne (10). The test is based on the assumption that endogenous and rHuEpo separate according to their charge heterogeneity (18, 19), and Epo isoelectric focusing is the official method used on a routine basis in all WADA-accredited anti-doping laboratories. The method has met serious critiques, with cases of false-positive testing in an experimental setting (2) [albeit the findings hereof has been disputed by others (4, 11)] and very low detection power (9). In addition, the half-life of rHuEpo is rather short, making detection difficult 3 days after injection (15), and a case report has even implied that detection may be impossible after as little as 12–18 h after injection (1). Also the so-called ON-model has been criticized (12), and other potential pitfalls of urine rHuEpo testing have recently been reviewed by Delanghe and coworkers (5).

In the present study, rHuEpo was injected in eight human subjects for 4 wk to study hematological effects of rHuEpo. At the same time, we took the opportunity to assess performance and the detection power of the WADA test, as well as to test the variability of rHuEpo urine analysis in two different accredited laboratories.


Eight healthy male (23 ± 3 yr, 181 ± 7 cm, 77 ± 5 kg) volunteers (university students, non athletes) participated in the study. 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.

rHuEpo (5,000 IU: NeoRecormon, Roche, Mannheim, Germany) was injected as follows. For the first 14 days (“boosting” period), subjects were injected with rHuEpo every second day and, for the following 2 wk, a single injection was given weekly (“maintenance” period). Prior to the injection, baseline measurements were obtained (repeated twice). Furthermore, blood samples from an arm vein were obtained prior to rHuEpo injections and on days 14, 16, 21, 23, 28, 35, 42, and 49. Urine samples were collected on days 0, 16, 23, 30, 35, 42, and 49 before any blood sampling; total hemoglobin mass was also determined at these times. Exercise tests were conducted prior to rHuEpo treatment and on days 35, 42, and 49.

The venous blood samples were analyzed for hematocrit and total plasma Epo concentration (Quantikine, R&D Systems, Minneapolis, MN). Total hemoglobin mass was determined by the carbon monoxide rebreathing method (3). Urine samples were obtained under sterile conditions with visual inspection by an experimenter. Each urine sample was handled according to WADA rules and immediately divided into four or five containers and stored at −80°C. The samples were subsequently shipped to laboratories for later analysis packed in dry ice (78.5°C or colder) and were reported by the laboratories to have arrived packed in dry ice and still to be in the solid frozen state. Total transport time was less than 24 h for both laboratories. Identical urine samples were analyzed in duplicates according to WADA regulations ( by two independent WADA laboratories that were blinded toward the treatment. The results were categorized by these WADA-accredited labs as “negative,” “suspicious,” or “positive.” Only samples determined “positive” can be used to potentially ban an athlete from competition, whereas samples found “suspicious” and “negative” do not imply restrictions. In “real life,” WADA requires labs that confirm an adverse analytical finding for rHuEPO to ask another lab to repeat the analysis. This was not done in the present study.

The exercises were performed using a bicycle ergometer with concomitant determination of pulmonary gas exchange as described previously (17).

Statistical differences over time were assessed by the nonparametric Friedman test, and the nonparametric Wilcoxon test was used as post hoc test. Statistical difference was set to P < 0.05. All values reported are means ± SD.


The rHuEpo treatment was effective in increasing total Hb mass at all measuring points (Table 1). Similarly, the hematocrit was augmented except on day 49, where it had returned to base values. Aerobic exercise power was increased (5.4–7.9%; P = 0.034) during the performance tests on days 35, 42, and 49. The urine data are shown in Table 1.

View this table:
Table 1.

Total hemoglobin mass, hematocrit, maximal pulmonary oxygen uptake, and number of samples determined “positive” from laboratories A and B, respectively, obtained before any rHuEpo injection, during, and after rHuEpo injections


Laboratory A determined rHuEpo misuse in all subjects during the “boosting” period, whereas laboratory B found no tests to be judged as positive. In the “maintenance” period laboratory A found six positive results indicating misuse and two suspicious results in a total of 16 samples. In the same period, laboratory B found five samples to be suspicious. During the posttreatment period laboratory A found two of 24 samples to be positive and three to be suspicious. Laboratory B found all these samples to be negative. There was little consistency in results between laboratories A and B, with a suspicious sample from laboratory A not being confirmed as a suspicious or positive sample from laboratory B and vice versa.

The results can be interpreted as laboratory A having a higher positive determination rate compared with laboratory B. In the boosting period, this can not be attributed to false positives since the subjects were injected with rHuEpo. Although this would seem satisfying for laboratory A, it may be of some concern that this lab found a sample to be positive on day 35 when the same person was negative on day 30. This is of concern because no injection had occurred between the two measurement points. In addition, the fact that two samples were determined positive on day 35 from laboratory A suggests that either the detection window is wider than the previous reported three days (15) in some subjects or they are false positives. Regardless, it is of main concern that the correlation between test results from laboratories A and B are poor and that laboratory B was not able to determine a single sample as being positive. The origin of the differences in analytical outcome between the laboratories is unknown (and beyond the scope of the present study), but could include interlaboratory standardization difficulties because of the use of irreproducible carrier ampholyte gels (16). Also, the highly concentrated urine sample is difficult to solubilize and apply on the gel (8). Although highly concentrated, some 20% of analyzed urine samples do not show detectable Epo (16), suggesting low test sensitivity. The American WADA laboratory has analyzed over 2,600 urine samples for rHuEpo, and here nine have been found positive. This low detection rate could be interpreted as either athletes are not misusing rHuEpo or that the testing for short-acting peptide hormones in urine is difficult.

During the maintenance period and 3 wk after the last rHuEpo injection, when total hemoglobin mass and aerobic exercise capacity were still elevated, only two of 48 samples were found positive. Moreover, it is worth noting that only two of 16 samples were found positive on day 30 (i.e., 48 h following previous injection), whereas this time point lies within the detection time window of 3 days (15). This implies that athletes may take advantage of rHuEpo doping without great risk of being tested Epo positive. It is noteworthy that a maintenance dose of rHuEpo, which in this study was limited to 2 wk, can by the athlete be elongated throughout a whole season (14). The practical implication is then that urine testing during competition in the season is of little or no value. The only strategy that provides a possible chance for obtaining a positive urine Epo test is to find the athlete for out of competition testing during a “boosting” period.

In conclusion, it would seem that for urine determinations of rHuEpo to be as efficient as can be with present laboratory techniques, elite athletes should be tested out of competition at strategic time points in their preparation for a season or for major events. The obtained samples would have to be analyzed by more than one laboratory. On the other hand, it should be considered whether it is at all advisable to spend more energy and money on the present procedures or develop new rHuEpo tests considering the availability of pharmacological agents such as continuous erythropoietin receptor activators (CERA) and Hydrol Proxylases having essentially the same effect as rHuEpo without the necessity of (mis)using rHuEpo. The potential implementation of the blood passport where longitudinal monitoring will be used to identify changes indicative of doping practices seems to surpass many of the above problems.


  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


View Abstract