In recent studies of cardiopulmonary resuscitation, an increase in end-tidal carbon dioxide tension (PETCO2) signifies an increase in cardiac output (QT) as spontaneous circulation resumes. We hypothesized that changes in QT might generally be measured by changes in PETCO2. In five pentobarbital-anesthetized dogs, we inflated percutaneously inserted vena cava balloons to impede venous return and to decrease QT (measured by pulmonary thermodulation). The PECTCO2 was measured at the airway opening by sidestream infrared capnometry. In 32 vena cava balloon inflation sequences during constant ventilation in five dogs, the percent decrease in PETCO2 directly correlated with the percent decrease in QT (slope = 0.73, R2 = 0.89, P less than 0.001). During decreased QT, reduced CO2 delivery to the lungs decreased alveolar PCO2 to cause part of the decrease in PETCO2. The remaining reduction in PETCO2 resulted from the increase in alveolar dead space (in turn due to lower pulmonary perfusion pressures during reduced QT), which diluted the CO2 from perfused alveolar spaces to further reduce PETCO2. During a sustained reduction in QT, increasing CO2 accumulation in the peripheral tissues and in venous blood began to restore CO2 delivery to the lung and PETCO2 toward baseline levels. Reciprocal changes occurred during increases in QT when the vena cava balloons were deflated. The linear relationship between changes in PETCO2 and QT in animals supports a decision to perform clinical studies necessary to determine whether a change in PETCO2 will be useful as a noninvasive, continuous monitor of a change in QT during anesthesia or intensive care.