Comments on Delanghe and Joyner's editorial "testing for recombinant human erythropoietin". Preanalytical and analytical issues in indirect hematological testing.

TO THE EDITOR: In their recent editorial on the disappointing disagreement between two World Anti-Doping Agency (WADA)-accredited laboratories evaluating samples for detecting recombinant human erythropoietin (Epo) in urine, Delanghe and Joyner (4) concluded that blood-based indirect Epo tests offer an interesting alternative (2). This is ideally true, considering that the urine test is proven unreliable (4). However, indirect testing, which is designed to establish an individual hematological profile (the “hematological passport”), is a hard task to accomplish. Under ideal conditions (centralized phlebotomy, standardized and traceable procedures for handling and storage of the specimens, adherence to rigorous programs for internal and external quality assessment), laboratory tests are still intrinsically biased by a certain degree of preanalytical and analytical variability (3). In the athletic field, such biases are enormously amplified and hardly governable. Strenuous physical exercise, temperature, and humidity all have substantial influences on several hematological parameters (1). Some hematological markers are also liable; this might be an additional problem when samples are collected under various environmental and technical conditions, stored and shipped for miles to distant accredited laboratories (1, 3). Finally, longitudinal comparison of several hematological variables, especially hematocrit (Ht) and hemoglobin (Hb), lacks of clinical significance if various analytical technologies are used by different accredited laboratories (i.e., Siemens ADVIA hematological analyzers substantially overestimate Hb and Ht compared with instruments from other manufacturers, such as Sysmex and Coulter; Ref. 5). When left uncontrolled and unstandardized, all these variables would dramatically affect the reliability of indirect hematological testing to detect cheating.