Department of Sports Medicine and the Center for Transfusion Medicine, University Hospital of Freiburg, Freiburg, Germany.
Transfusion. 2011 Aug;51(8):1707-15. doi: 10.1111/j.1537-2995.2011.03076.x. Epub 2011 Mar 7.
Since no direct detection method for autologous blood transfusions exists, the most promising attempt is the Athlete Biological Passport (ABP) and its adaptive model that enables a longitudinal monitoring of hematologic measures to identify patterns of blood manipulations. The purpose therefore was to evaluate the performance of this adaptive model for the detection of autologous blood transfusions in a longitudinal blinded setting.
Twenty-one subjects were divided into a doped group (multiple transfusions of 1-2 units of red blood cells, n = 11) and a control group (n = 10). The time course of a cycling season (42 weeks) was simulated including three major competitions (Classics, Grand Tour, World Championships). Up to 10 venous blood samples were ordered per subject by a blinded investigator mimicking the intelligent approach in obtaining hematologic data for the adaptive model (hemoglobin [Hb] concentration, reticulocyte percentage, OFF-score).
Retrospective analysis allowed identification of four (probability >99%) or three (probability >99.9%) abnormal samples for Hb and eight (probability >99%) or five (probability >99.9%) abnormal samples for OFF-hr in doped subjects. Four doped subjects (36%) presented an abnormal OFF-hr sequence and three doped subjects (27%) an abnormal Hb sequence; there were no false-positive sequence results. The best possible sensitivity was 82% when a combination of all tests was used.
This investigation provides evidence that the adaptive model allows detection of autologous blood transfusions with a good sensitivity. An intelligent testing approach and the adherence to World Anti-Doping Agency's ABP operating guidelines are nevertheless determinant in the success.
由于不存在直接检测自体输血的方法，最有希望的尝试是运动员生物护照（ABP）及其自适应模型，它能够对血液学指标进行纵向监测，以识别血液操纵的模式。因此，本研究旨在评估该自适应模型在纵向盲法设置下检测自体输血的性能。
将 21 名受试者分为实验组（多次输注 1-2 单位红细胞，n=11）和对照组（n=10）。模拟了一个周期（42 周）的时间过程，包括三个主要比赛（古典赛、大环赛、世界锦标赛）。由一位盲法调查员为每位受试者安排多达 10 次静脉血样，模仿自适应模型获得血液学数据的智能方法（血红蛋白[Hb]浓度、网织红细胞百分比、OFF 评分）。
回顾性分析可识别出 4 个（概率>99%）或 3 个（概率>99.9%）Hb 异常样本和 8 个（概率>99%）或 5 个（概率>99.9%）OFF-hr 异常样本，在实验组中。有 4 名实验组（36%）出现异常 OFF-hr 序列，3 名实验组（27%）出现异常 Hb 序列；没有假阳性序列结果。当联合使用所有检测时，最佳可能的灵敏度为 82%。
本研究提供了证据表明，自适应模型可以检测到自体输血，具有良好的灵敏度。然而，智能检测方法和遵守世界反兴奋剂机构的 ABP 操作指南是成功的决定因素。
