Estimating Arterial PCO₂ from Flow-Weighted and Time-Average Alveolar PCO₂ During Exercise

Abstract

The appropriateness of the ventilatory response to muscular exercise is best considered with respect to the precision of arterial PCO₂ (P_aCO₂) regulation for moderate exercise and by the degree of the compensatory hyperventilation at work rates which engender a metabolic acidemia. But in order to avoid the necessity for intra-arterial sampling, which for sufficient data-density usually requires an indwelling catheter, several investigators have proposed non-invasive techniques for P_aCO₂ estimation. Techniques which are based upon assumptions of dead space volume (V_D) or of a constant relationship between P_aCO₂ and end-tidal PCO₂ (P_ETCO₂) are not useful owing to the large inter-subject variability in the former case¹ and the invalidity of the assumption in the latter.^2,3 DuBois et al.⁴ proposed that the mean alveolar PCO₂ could be derived from a “reconstruction” of the intra-breath time profile of P_ACO₂ (i.e., time-average: P_ĀCO ^T₂ ), which in normal subjects would closely approximate P_aCO₂. Gumming,⁵ has shown that when the cumulative CO₂ output (VCO₂) is expressed as a function of the cumulative expired volume (V), the resulting relationship is — to a very close approximation — linear (following a small lag phase) with a slope which represents P_ĀCO₂ (i.e., flow-weighted average: P_ĀCO ^F₂ ) and an intercept on the volume axis which is a measure of the “series” dead space (V_D)). We were interested, therefore, in developing on-line, breath-by-breath techniques for determining these estimators and to define the acccuracy with which they reflect directly-measured P_aCO₂ during exercise.