BACKGROUND:Cardiac output monitoring is most important where cardiovascular stability is potentially threatened, such as during major surgery and in critically ill patients. However, continuous monitoring of cardiac output is still not performed routinely during anesthesia and critical care, because of invasiveness, expense, and inaccuracy of available technologies. METHODS:A technique termed the capnodynamic method was tested for breath-to-breath measurement of pulmonary blood flow from lung carbon dioxide mass balance, using measured carbon dioxide elimination and end-tidal concentration. A prototype measurement system was constructed for a feasibility study in six anesthetized sheep. Large and rapid fluctuations in cardiac output were generated by repeated dobutamine and esmolol challenge. Measurements were compared with an indwelling ultrasonic flow probe placed on the ascending aorta or pulmonary artery. RESULTS:Cardiac output measured by the flow probe varied between zero and 8.67 l/min, with a mean of 3.50 l/min. Overall mean bias [SD of the difference] between the methods (capnodynamic - flow probe) was -0.25 [0.94] l/min, r = 0.79 (P < 0.001). During periods of stability in cardiac output of 5 min or more, mean bias was -0.20 [0.55] l/min. The method successfully indicated two cardiac arrest events, which were induced in one of the animals. CONCLUSIONS:The method satisfactorily tracked wide fluctuations in cardiac output in real time. The capnodynamic method may have potential for continuous noninvasive cardiac output monitoring in patients undergoing anesthesia for major surgery, and in critical care, on a routine basis.