Purpose: A new method to estimate hemoglobin mass (Hbmass) requires capillary blood and rebreathing a carbon-monoxide (CO) bolus for 2 min. We hypothesized that incomplete circulatory mixing of CO could confound this method, so we compared capillary with venous blood to determine whether sampling site altered the percentage of carboxyhemoglobin (%HbCO) and the reliability and accuracy of the "2-min Hbmass." The conventional 20-min CO-rebreathing procedure was used as the Hbmass criterion.
Methods: In the first experiment (N = 12), both fingertip capillary and antecubital venous blood were sampled 4 and 6 min after commencing 2 min of CO-rebreathing. Within 8 d, these subjects completed two 2-min and one 20-min CO-rebreathing periods. For the latter, capillary and venous blood were collected simultaneously after two 10-min periods of rebreathing. In a second experiment (N = 6), both capillary and venous blood were sampled 4, 6, 8, 10, and 12 min after commencing 2 min of CO-rebreathing. A third experiment (N = 6) evaluated the reliability of a modified 2-min CO-rebreathing test with capillary blood sampled at minutes 8 and 10.
Results: Typical error (TE) for the first two 2-min tests was 1.1% (90% confidence limits 0.9-1.8%), but the average Hbmass from 2-min capillary blood was 4.8% lower than from venous blood for the 20-min procedure. In the second experiment, peak venous %HbCO occurred at minute 6, and the difference between capillary and venous values was minimal (mean ± SD; 0.08 ± 0.07, 0.01 ± 0.09) at minutes 8 and 10. TE for the third experiment was 1.2% (0.8-2.5%).
Conclusion: A modified 2-min CO-rebreathing procedure using capillary or venous blood sampled 8 and 10 min after starting CO-rebreathing allows complete circulatory mixing and provides an accurate and reliable estimate of Hbmass.