Estimation of Mixed Venous Pco2 for Determination of Cardiac Output in Children (Discussion)We found in healthy children that PvC02 obtained using the exponential method was similar to the downstream corrected equilibrium value, but significantly different from the uncorrected equilibrium value. This is similar to what has been described in adults.’ This suggests that in children, as in adults, either the equilibrium method with downstream correction or the exponential method can be used when calculating PvC02.

Using the PvC02 obtained by the equilibrium method with the downstream correction, we found a very high correlation, r2=0.95, between Q and Vo2 in healthy adults and children of both sexes (equation 3, Fig 2). Our results agree closely with those of dye dilution studies’ (equation 4, Fig 2). We found that there was a slight additional effect when a grouping factor, indicating whether a subject was a child or an adult, was included, but this effect disappeared when weight was included as one of the independent variables in the stepwise linear regression equation.

We therefore suggest, for ease of use and applicability to both children and adults, that the following equation (equation 6) be used in the prediction of Q: Q (L/min) = 1.42+5.80 • Vo2 (U min) + 0.06 • wt (kg), r2=0.97, standard error of the estimate (SEE)=0.67. Using a forced intercept of zero, our equation (equation 8) was not significantly different from that of Jones (equation 7), which was extrapolated from values in the literature. Since the exponential method is reliable in determination of Q in mild to moderate lung disease as well as in severe lung disease in adults, we would suggest that the indirect Fick method may be applicable to a wide patient population, although arterial Pco2 may need to be measured directly in patients with severe airflow limitation. In addition, because the exponential method is well tolerated and can be used during progressive exercise, the indirect Fick method can be used for the determination of Q in a wide variety of clinical and research studies.

Our results are in keeping with studies in adults’ and children, that found good agreement between Q calculated using downstream corrected equilibrium values for PvC02 and that measured by dye dilution techniques. Furthermore, Gadhoke and Jones found that the relationship between Q and Vo2 did not differ between boys and men. While Godfrey et al’ acknowledged the existence of an alveolar to blood gradient in children that was no different from that found in adults, they suggested that the uncorrected value for PvC02 should be used in children to avoid overestimation of cardiac output. It is unclear why our results differ from those of Godfrey et al. While the use of an infrared analyzer to measure Pco2 can affect C02 readings due to collision broadening, this would not explain the difference as we used a mass spectrometer, which is not subject to collision broadening, and Godfrey et al controlled for the effects of collision broadening in their study (personal communication). Paterson and Cunningham suggested that the C02 dissociation derived by Godfrey et al may have raised the values of Q in children, as compared to the curve of McHardy, which we used in our study, or that of Comroe. In the present study, both children and adults were evaluated using the same exercise apparatus and protocol, and both groups had similar values to those obtained by dye dilution.

When Q in mL/min/kg, calculated using the corrected PvC02, was plotted against Vo2 in mL/min/ kg, most of our data points for adults fell within the 95th percentile confidence interval described by Faulkner et al in their study of 50 normal adult male subjects at various steady-state work loads using the exponential method (Fig 3), although our values tended to be somewhat higher. Differences between our values and those of Faulkner et al may result from slight differences in the equations used to convert partial pressures to contents as the equations used by Faulkner et al were not specified. Most of our data points for children using the downstream correction were higher than those predicted by the 95th percentile confidence interval of Faulkner et al (Fig 2). In our stepwise regression analysis for prediction of Q, we found a difference between children and adults. In their study, Faulkner et al found a lower intercept for the normalized relationship in men older than 40 years of age who had a Vo2 of <30 mL/min/kg, and suggested that the difference in intercepts may be due to body fat in excess of 18%. We did not have data on body fat on all the adults and children who we tested, but we speculate that as children tend to have a lower body fat percentage than adults; this is a possible explanation for their relatively higher values for Q per kilogram of body weight.

It should be noted that our prediction equations apply to adults and children performing submaxi-mal exercise, with a range of V02 from 0.5 to 2.5 L/min. We cannot comment on whether they are applicable to the determination of Q at higher levels of Vo2.

We conclude that C02-rebreathing techniques, equilibrium and exponential, may be used in children to estimate Q noninvasively. In children, as in adults, the downstream correction factor should be applied to PvC02 obtained by the equilibrium method. With this, a prediction equation for Q is obtained, based on Vo2, which is very similar to that obtained from dye dilution studies. When weight is included in the equation, the relationship is further strengthened, and one equation can be used to describe Q in children and adults.