Abstract

The variability of Doppler echocardiographic estimation of cardiac output at the aortic orifice was investigated in eight healthy subjects. Cross sectional echocardiograms of the aortic orifice and aortic Doppler velocities were recorded and measured by four echocardiographers. Between subject variability was significantly larger than within subject variability for all variables. Variability owing to different echocardiographers and different measurement times was small compared with total variability. Coefficients of variation for aortic annular diameter, aortic velocity integral, and heart rate were 4.1%, 6.4%, and 5.0% respectively. The coefficient of variation for cardiac output was 8.8% and the 95% confidence interval for measurement of cardiac output by the Doppler method was 4.45 to 6.35 l/min. One echocardiographer reanalysed all the recordings and the results showed that recording the echocardiograms introduces a significantly larger source of error than measuring them. Thus serial measurements of cardiac output by the Doppler method can be performed with acceptable reproducibility; this indicates that the technique can be accurately applied in clinical practice.