The aim of the present study was to evaluate whether the Epo isoforms in blood, induced by short-term and intermittent hypoxia, are different from those at normoxia at sea level and if this could be an impediment to the use of a direct Epo doping test based upon the electric charge of the Epo isoforms. Ten healthy subjects, 9 men and 1 woman, participated in the study. Median age was 22 years (range 20–32). Normobaric hypoxia was administered differently in 3 sub-groups; two groups with 12 h hypoxia and 12 h normoxia up to 10 days: IM 2000 and IM 2700 living in 16.2% and 14.9% O2, corresponding to 2000 and 2700 m above sea level, respectively, and training in normoxia. The third group, C 2700, lived in hypoxia, 14.9% O2 corresponding to 2700 m, continuously for 48 h. The mean serum Epo level increased from 10.9 IUL−1 (range 8.8–12.5) to 23.5 IUL−1 (15.6–29.1) after 2 days followed by 19.7 IUL−1 (16.1–24.1) after 10 days exposure for intermittent hypoxia. The highest values 39.5 IUL−1 (31.5–50) were obtained for the group exposed for continuous hypoxia for 48 h. The median electrophoretic mobility of the serum Epo isoforms was above the cut-off limit of 670 AMU, previously estimated for discrimination between recombinant and endogenous Epo, in all samples taken before and after exposure to hypoxia. The highest values, mean 730 mAMU (range 703–750) were obtained after 10 days of intermittent hypoxia. Conclusion: If the method had been used as a doping test, no false positive results would have been registered for the 15 serum samples from the 10 individuals exposed for hypoxia. Thus, the results indicate that the basic principle for direct detection of recombinant Epo doping, based upon the change in electric charge on Epo, can be used also on individuals having lived in a hypoxic milieu.