Original paper
Standardising analysis of carbon monoxide rebreathing for application in anti-doping

https://doi.org/10.1016/j.jsams.2010.07.007Get rights and content

Abstract

Determination of total haemoglobin mass (Hbmass) via carbon monoxide (CO) depends critically on repeatable measurement of percent carboxyhaemoglobin (%HbCO) in blood with a hemoximeter. The main aim of this study was to determine, for an OSM3 hemoximeter, the number of replicate measures as well as the theoretical change in percent carboxyhaemoglobin required to yield a random error of analysis (Analyser Error) of ≤1%. Before and after inhalation of CO, nine participants provided a total of 576 blood samples that were each analysed five times for percent carboxyhaemoglobin on one of three OSM3 hemoximeters; with approximately one-third of blood samples analysed on each OSM3. The Analyser Error was calculated for the first two (duplicate), first three (triplicate) and first four (quadruplicate) measures on each OSM3, as well as for all five measures (quintuplicates). Two methods of CO-rebreathing, a 2-min and 10-min procedure, were evaluated for Analyser Error. For duplicate analyses of blood, the Analyser Error for the 2-min method was 3.7, 4.0 and 5.0% for the three OSM3s when the percent carboxyhaemoglobin increased by two above resting values. With quintuplicate analyses of blood, the corresponding errors reduced to .8, .9 and 1.0% for the 2-min method when the percent carboxyhaemoglobin increased by 5.5 above resting values. In summary, to minimise the Analyser Error to ∼≤1% on an OSM3 hemoximeter, researchers should make ≥5 replicates of percent carboxyhaemoglobin and the volume of CO administered should be sufficient increase percent carboxyhaemoglobin by ≥5.5 above baseline levels.

Introduction

The advent of multi-wavelength spectrophotometers, such as the OSM3 hemoximeter (Radiometer, Copenhagen, Denmark),1 for automated measurement of percent carboxyhaemoglobin (%HbCO) in blood, allowed expedient determination of total haemoglobin mass (Hbmass) via carbon monoxide (CO) rebreathing.2, 3, 4 Indeed, CO-rebreathing has mostly superseded other methods such as Evans Blue dye and radioactive chromium5 to quantify the erythropoietic effects of altitude training.6, 7, 8 However, determination of Hbmass depends heavily on the reliability and precision of measuring blood %HbCO,5 typically on an OSM3 hemoximeter. Although there are potentially a number of other methodological steps that could contribute to the technical error of Hbmass, such as a gross error in the dose of CO or a leak from the athlete,5 the analyser error (as a sub-component of technical error) can be most readily quantified and ameliorated.

The reproducibility of the OSM3 is the critical determinant of analyser error because this hemoximeter provides %HbCO to only .1% resolution. However, multiple measures on each blood sample will attenuate error as a function of n replicates.9 A theoretical analysis of error propagation concluded that, using an OSM3 hemoximeter, five replicates were necessary to estimate blood volume (BV) via CO-rebreathing with an error of 3.6%.10 Instead of using empirical data, these authors relied on the manufacturer's specification for repeatability of analyses of %HbCO on the OSM3. Also based on the .1% resolution of the OSM3, Burge and Skinner4 recommended that four replicates of %HbCO reduced the error of measurement of Hbmass to 1.5%. However, since these recommendations in the early to mid-1990s, some authors have used as few as one measure of %HbCO,11, 12 yet others have used the average of duplicates,13 duplicates as the average of successive single measures taken a few minutes apart2, 3, 14 or triplicates.6 Still others have measured %HbCO in quadruplicate,4, 7 four to eight replicates,15 quintuplicates,16, 17, 18 sextuplicates8 or octuplicates.19 Fewer replicates may explain why some authors have poor measurement error for Hbmass – as large as 12%.5

A meta-analysis demonstrated that the CO-rebreathing method has total measurement error of ∼2.2% for Hbmass.5 This total error likely comprises both biological and technical errors, which are independent and additive20 such that:total measurement error=(biological error2+technical error2).By substitution into Eq. (1), we can deduce that, ideally, the technical error should be ≤1% so that biological error (2%) is almost entirely the source of measurement error (of 2.2%); that is2.2=(12+22).However, the magnitude of biological error of Hbmass is uncertain.18 It is also plausible that the total error of measurement for Hbmass (2.2%) is almost completely technical in origin, with one component of technical error being Analyser Error (ideally ≤1%) and the remainder (∼2%) being residual technical error.

Based up the assumption of independence of errors, the aim of this study was to determine empirically the number of replicates required to yield an analyser error of ≤1% for an OSM3 hemoximeter. Quantifying and minimising analyser error becomes particularly relevant if authorities seek to use Hbmass as part of the biological passport21 in the fight against blood-doping.11, 14

Section snippets

Methods

Nine recreationally active non-smoking volunteers, seven men and two women, participated in this study. The mean (and standard deviation (SD)) age, height and body mass of the men was 38.4 (9.4) years, 183.3 (6.7) cm and 80.4 (7.1) kg, and the corresponding values for the women were 25.1 (.5), 166.7 (4.2) and 53.4 (5.1). All experimental procedures were approved by the Australian Institute of Sport (AIS) Ethics Committee and written consent was obtained prior to testing. The funding body of the

Results

Table 1 presents the SD of the %HbCO for each of the three OSM3 hemoximeters at both Low and High %HbCO, and for the pooled SD. The SD and associated confidence limits are similar for both Low and High %HbCO; for instance the SD (and confidence limits) for five replicates for OSM3-1 are .039 (.029–.059) %HbCO for Low and are .038 (.034–.042) %HbCO for High. The raw values in Table 1 are more readily understood when expressed as Theoretical Analyser Error (Fig. 1); for example the pooled SD

Discussion

Our results suggest that Hbmass via CO-rebreathing might be a suitable candidate for inclusion as part of the biological passport21 in the fight against blood-doping11, 14 because Analyser Error of the method can be expediently reduced to ≤1%, which is much less than the 10% magnitude that an athlete might seek to illegally boost their Hbmass.11, 14 In order to reduce the Analyser Error of measuring %HbCO with an OSM3 hemoximeter to ≤1%, our results indicate that five replicates are necessary

Practical implications

  • Total body haemoglobin measured carefully with carbon monoxide rebreathing is sufficiently accurate to be a useful adjunct in the fight against blood-doping.

  • Carbon monoxide rebreathing can readily be conducted in the field and is thus a practical anti-doping technique, not merely a laboratory measure.

  • In order to make accurate measures an athlete would require a dose of carbon monoxide to bind to ∼6% of their total haemoglobin, at least 8 h before competition to prevent any decrease in their

Acknowledgements

We are grateful for funding support of the Australian Institute of Sport Applied Research Centre.

References (28)

  • N. Fogh-Andersen et al.

    Diode-array spectrophotometry for simultaneous measurement of hemoglobin pigments

    Clin Chim Acta

    (1987)
  • J.A. Raub et al.

    Carbon monoxide poisoning—a public health perspective

    Toxicology

    (2000)
  • N. Prommer et al.

    Loss of CO from the intravascular bed and its impact on the optimised CO-rebreathing method

    Eur J Appl Physiol

    (2007)
  • W. Schmidt et al.

    The optimised CO-rebreathing method: a new tool to determine total haemoglobin mass routinely

    Eur J Appl Physiol

    (2005)
  • C.M. Burge et al.

    Determination of hemoglobin mass and blood volume with CO: evaluation and application of a method

    J Appl Physiol

    (1995)
  • C.J. Gore et al.

    Errors of measurement for blood volume parameters: a meta-analysis

    J Appl Physiol

    (2005)
  • P. Robach et al.

    Living high–training low: effect on erythropoiesis and maximal aerobic performance in elite Nordic skiers

    Eur J Appl Physiol

    (2006)
  • J.P. Wehrlin et al.

    Live high–train low for 24 days increases hemoglobin mass and red cell volume in elite endurance athletes

    J Appl Physiol

    (2006)
  • C.J. Gore et al.

    Increased serum erythropoietin but not red cell production after 4 wk of intermittent hypobaric hypoxia

    J Appl Physiol

    (2006)
  • W.G. Hopkins

    Measures of reliability in sports medicine and science

    Sports Med

    (2000)
  • P. Christensen et al.

    Precision of a new bedside method for estimation of the circulating blood volume

    Acta Anaesthesiol Scand

    (1993)
  • T. Pottgiesser et al.

    Hb mass measurement suitable to screen for illicit autologous blood transfusions

    Med Sci Sports Exerc

    (2007)
  • Y.O. Schumacher et al.

    Haemoglobin mass in cyclists during stage racing

    Int J Sports Med

    (2008)
  • P. Robach et al.

    Recovery of plasma volume after 1 week of exposure at 4,350 m

    Pflugers Arch

    (2002)
  • Cited by (33)

    View all citing articles on Scopus
    View full text