Does bronchial thermodilution allow estimation of cardiac output?

Objective: Transcapillary heat transfer after injections of cold saline into the right atrium generates bronchial thermodilution curves resembling those observed in the aorta. Under the assumption that no indicator is lost or gained within the pulmonary capillary bed and changes in blood temperature are instantaneously recorded in the bronchial system, we tested the hypothesis that flow rates calculated from bronchial temperature-time curves are similar to those from aortic curves. Design: Comparative study of two cardiac output estimates in five dogs. Setting: Research laboratory for Experimental Anaesthesiology. Interventions: Cardiac output was decreased (repeated withdrawal of blood) and increased (infusion of colloids or dobutamine) in order to study a wide range of cardiac outputs. Measurements and results: Thermistors were placed in a bronchiole (wedge position) and in the ascending aorta of anaesthetized dogs. Bronchial and aortic thermodilution curves were recorded after injection of 5 ml ice-cold saline into the right atrium. We found that bronchial thermodilution yields flow estimates similar to those from aortic curves. Correlation between the two flow estimates was acceptable (r = 0.84) with a mean difference between the two of less than 2 %. Conclusion: We conclude that the Stewart-Hamilton equation may be extended to bronchial temperature-time curves for estimations of cardiac output. At this time, however, we do not advocate bronchial thermistors as suitable and less invasive alternatives to pulmonary arterial catheters for routine cardiac output measurements in patients.