Quality control technique to reduce the variability of longitudinal measurement of hemoglobin mass

The sensitivity of the athlete blood passport to detect blood doping may be improved by the inclusion of total hemoglobin mass (Hb(mass)), but the comparability of Hb(mass) from different laboratories is unknown. To optimize detection sensitivity, the analytical variability associated with Hb(mass) measurement must be minimized. The aim of this study was to investigate the efficacy of using quality controls to minimize the variation in Hb(mass) between laboratories. Three simulated laboratories were set up in one location. Nine participants completed three carbon monoxide (CO) re-breathing tests in each laboratory. One participant completed two CO re-breathing tests in each laboratory. Simultaneously, quality controls containing Low (1-3%) and High (8-11%) concentrations of percent carboxyhemoglobin (%HbCO) were measured to compare hemoximeters in each laboratory. Linear mixed modeling was used to estimate the within-subject variation in Hb(mass), expressed as the coefficient of variation, and to estimate the effect of different laboratories. The analytic variation of Hb(mass) was 2.4% when tests were conducted in different laboratories, which reduced to 1.6% when the model accounted for between-laboratory differences. Adjustment of Hb(mass) values using quality controls achieved a comparable analytic variation of 1.7%. The majority of between-laboratory variation in Hb(mass) originated from the difference between hemoximeters, which could be eliminated using appropriate quality controls.