Since the introduction of blood analysis performed before major cycling events in 1997, there have been discussions concerning the quality of the results. The aim of our study therefore was to measure blood samples and compare the results obtained on the field with those obtained in the laboratory. For this it was necessary to have blood samples analysed with different instruments to determine the exactness of the results and evaluate the performances of these instruments in order to validate the haematological testing used to reveal athletes abusing recombinant erythropoietin. We report on the haematological analysis of 177 professional cyclists who took part in the Tour de France 2001. All the blood samples were withdrawn in the morning between 7 and 9 am in Dunkerque (France) and were analysed immediately with a transportable analyser. Then the samples were flown to Lausanne (Switzerland) and were reanalysed in two independent ISO 17 025 accredited laboratories with three different analysers. The results confirmed that the most effective haematological follow-up should be performed under standardized pre-analytical conditions and with identical analysers of the same manufacturer to avoid too many variations notably on the haematocrit level and the reticulocyte count. Furthermore, this study suggests that analyses performed on the site are good and could enable the federations to perform a urinary test to detect rhEPO abuse right after the blood analysis. This time saving is essential to fight efficiently recombinant erythropoietin doping, because the half life of the hormone is very short.