Measurement of mixed venous carbon dioxide pressure by rebreathing during exercise.

This study compared rebreathing methods currently used for the measurement of mixed venous PCO2 (PVCO2) during exercise. Four mathematical procedures were used to derive the asymptote of the exponential rise in end-tidal PCO2 during 15 sec rebreathing of low (2-6%) CO2 mixtures ('exponential' methods); the derived PVCO2 was compared to that obtained with an 'equilibrium' method, in which high CO2 mixtures were rebreathed to obtain an equilibrium of CO2 between the lungs and rebreathing bag. Precision was established by analysing duplicate rebreathings at each power output. The most precise of the four exponential procedures (coefficient of determination 0.98) used iterative statistical analysis to obtain an equation for the best fit curve of end-tidal PCO2 with time (t), solved for t = 20 sec. The PVCO2 derived by this procedure was similar to that obtained by the equilibrium method (r = 0.96), and both yielded estimates of cardiac output that were within the previously published expected range. Variations in the initial rebreathing bag volume and CO2 concentration, and in the breathing frequency during rebreathing had little effect on the derived PVCO2.