Background and objectives: Strategies based on the use of upper thresholds of hemoglobin or hematocrit to detect blood doping in endurance sports have essentially failed to deter this malpractice. With the aim of establishing a more effective strategy, we analyzed the biological variations of hematologic parameters in professional athletes and investigated the possibility of defining subject-specific reference ranges that could distinguish between physiologic and abnormal variability.
Design and methods: Hemoglobin concentration, hematocrit, reticulocyte count, serum ferritin and soluble transferrin receptor levels were sequentially evaluated in 923 professional football players. Using the analysis of variance we tested the effect of age, ethnicity, exercise modalities and training phases on hematologic parameters and then estimated components of variation. The significance of the difference between two measures was obtained from the distribution of the within-subject variance (the so-called reference change). Subject-specific reference ranges were centered around the individual mean value with dispersion based on the 95th percentile of the coefficient of variation distribution.
Results: A total of 2,506 hematologic determinations were made. Exercise modalities were found to have important effects on hematologic parameters. Hemoglobin and hematocrit values were higher at the beginning of the competition season, and then declined in well-trained athletes. Aerobic exercise was clearly associated with lower values, suggesting that marginally low hemoglobin and hematocrit values should physiologically be found in endurance sports. At least five determinations were required to define subject-specific reference ranges reliably. Considering athletes showing normal indices of red cell production (i.e., reticulocyte count and soluble transferrin receptor), the 95th percentile of the coefficient of variation distribution was lower than 5% for both hemoglobin and hematocrit. Increases exceeding 10% in these latter parameters should to be considered abnormal. Score systems capable of efficiently detecting non-physiologic increases in red cell production were developed.
Interpretation and conclusions: Using proper sequential determinations of hematologic variables subject-specific reference ranges can be defined for hemoglobin and hematocrit. Thus, the hematologic passport is feasible and might be employed to exclude athletes with non-physiologic increases in hemoglobin and hematocrit from competitions. The hematologic passport should be used within a global strategy to deter blood doping.