Standardization of reticulocyte counts in the athlete biological passport: A practical update. (28th October 2021)

• Record Type: Journal Article
• Authors: Equey, Tristan; Sletten, Camilla; Dehnes, Yvette; D'Onofrio, Giuseppe; Brugnara, Carlo; Baume, Norbert; Aikin, Reid
• Is Part Of: International journal of laboratory hematology. Volume 44:Number 1(2022)
• Page Start: 112
• Page End: 117
• Publication Date: 2021-10-28
• Subjects: anti‐doping -- athlete biological passport -- blood doping -- laboratory hematology -- reticulocytes; Hematology -- Periodicals; Blood -- Diseases -- Periodicals; Hematology -- Periodicals; 616.15005
• DOI: 10.1111/ijlh.13732
• Languages: English
• ISSNs: 1751-5521

Abstract

Abstract: Introduction: The athlete biological passport monitors blood variables over time to uncover blood doping. With the phasing in of a new series of blood analyzers, the Sysmex XN series, it was necessary to examine the comparability of results with the previously employed XT/XE series. A previous comparison between XN and XT/XE series suggested a small but significant bias between the two instruments in the measurements of RET%. Here, we examined the comparability of RET% on the XN and XT/XE platform using data collected over the first year since the transition. Methods: The comparability of results obtained from XN and XT/XE instruments was assessed using three datasets: (i) 767 blood samples measured on both instrument series in 22 WADA‐accredited laboratories, (ii) 27 323 samples measured on either instrument across 31 laboratories, and (iii) 119 clinical samples and 110 anti‐doping samples measured on both instruments in a single laboratory. Results: Analysis of the three datasets confirms the previous observation of a bias toward higher RET% values for samples measured on Sysmex XN instruments compared with the XT/XE series. Using data across a larger number of XN instruments and a larger athlete population, the current work suggests that the bias is proportional and slightly higher than previously observed across most of the range RET% values. Conclusion: A model is proposed for the comparison of data across XN and XT/XE technologies whereby the instrument bias increases proportionally with RET% measured on Sysmex XN Series, but where the rate of increase is negatively related to IRF%.