Determination of cardiac output at rest and during exercise by carbon dioxide rebreathing method in obstructive airway disease.

We evaluated the accuracy of the CO2 rebreathing method (CO2rb) for measuring cardiac output at rest and during steady-state exercise in 15 patients (mean +/- SD age, 59.7 +/- 7.5 yr) with obstructive airway disease. At rest, there was a significant correlation between direct Fick and CO2rb methods using measured arterial PCO2 (r = 0.70; p = 0.002), but not with using end-tidal PCO2 (r = 0.38; p = NS). During exercise, there was greater correlation with CO2rb using arterial PCO2 (r = 0.79; p = 0.001) than using end-tidal PCO2 (r = 0.63; p = 0.007) compared with the direct Fick determination. Correlation between the CO2rb and direct Fick methods was greater with moderate air-flow obstruction (n = 6) than with severe airway disease (n = 9), and the CO2rb method was more accurate during exercise than at rest. The CO2rb method using either end-tidal or arterial PCO2 underestimated the direct Fick measurement in 13 of 15 patients at rest, which may reflect inadequate equilibration between alveolar and oxygenated mixed venous PCO2. However, no consistent error was observed during exercise when higher CO2 production and an increased venoarterial PCO2 difference would diminish potential inaccuracies. We concluded that the CO2rb technique is an acceptable method for measuring cardiac output during exercise in patients with moderate and severe obstructive airway disease as long as arterial PCO2 is directly measured rather than estimated from end-tidal PCO2.