Validation and application of single breath cardiac output determinations in man

Cardiac outputs by single breath (Qsb) and Fick (Qf) procedures were compared in five healthy males during supine rest and exercise with Qf ranging from 6-19 L X min-1. The prolonged exhalation (SB) was not controlled. The Qsb calculations incorporated an equation of the CO2 dissociation curve and a "moving spline" sequential curve-fitting technique to calculate the instantaneous R from points on the original expirogram. The resulting linear regression equation for all 38 comparisons obtained (r = +0.76, p less than 0.001, mean difference +/- S.D. = 2.93 +/- 2.72 L X min-1) indicated a 24% underestimation of Qf. A substantial portion of the variability during exercise (n = 28) was due to a difference in alveolar ventilation between the time of the mixed expired (E) gas collection and the SB maneuver. When Qsb was corrected (Qsb) by a linear regression based on the difference between Re and Rsb during exercise and by adding 2.44 L X min-1 at rest (the mean difference), the relationship was greatly improved (Qsb = 0.14 + 0.99 Qf, r = +0.93, mean difference +/- S.D. = 0 +/- 1.47 L X min-1). A subsequent study during upright rest and exercise to 80% of VO2max in 6 subjects indicated a close linear relationship between Q'sb and VO2 for all 95 values obtained (r = +0.94), with slope and intercept close to published studies utilizing invasive cardiac output measurements. Considerations of measured blood gases in relation to estimated values suggested that underestimates of Qf arose, at least in part, from arterial desaturation during the SB maneuver. Detailed computational procedures are provided for implementing this improved Qsb procedure.