Objective—To compare cardiac output measured by the transoesophageal Doppler and thermodilution techniques.
Design—Prospective direct comparison of paired measurements by both techniques in each patient.
Setting—Intensive care unit in a cardiovascular centre.
Patients—65 patients after open heart surgery (mean (SD) age 53 (12) years).
Interventions—Cardiac output was measured simultaneously by the transoesophageal Doppler and thermodilution techniques. Cardiac output was measured again after a mechanical intervention or volume loading.
Results—The limits of agreement were −2·53 to +0·83 1·min−1 for cardiac output measured by the Doppler and thermodilution techniques. This suggests that the Doppler method alone would not be suitable for clinical use. The second measurement of cardiac output by thermodilution was compared with cardiac output estimated from the first and second Doppler measurements and the first thermodilution measurement. The limits of agreement (−0·55 to +0·51 1·min−1) were good enough for clinical use.
Conclusions—After cardiac output had been measured simultaneously by both the Doppler and thermodilution techniques, subsequent transoesophageal Doppler alone gave a clinically useful measurement of cardiac output.