Reproducibility of Doppler indices of left ventricular systolic and diastolic function in patients with severe chronic heart failure

Doppler ultrasound has been utilized to evaluate cardiac output and left ventricular filling pressure in patients with various cardiac diseases. This method would be particularly useful for non-invasive haemodynamic monitoring in patients with severe chronic heart failure. However, few data exist on the reproducibility of Doppler indices in this population. To determine the magnitude of technical and biological variability of Doppler indices, serial Doppler echocardiographic studies were performed in 26 patients with severe heart failure in the usual clinical setting. Short-term intra- and inter-observer, mid-term and day-to-day variabilities of stroke volume, cardiac output, maximal early and late diastolic velocities of mitral flow, rate of decrease and deceleration time of flow velocity in early diastole and the colour Doppler area of the mitral regurgitant jet were evaluated by two cardiologists. For each source of variability, correlation coefficients with standard errors and mean differences between paired determinations with their standard deviations were calculated. Short-term (intra- and inter-observer) variability was small for each considered variable: the mean differences of measurements were within 10% of the mean value for most parameters and no systematic error was found. Stroke volume and cardiac output showed a significant increase in the afternoon. The standard deviations of differences between day-to-day measurements were 41 ml (9% of the mean value) and 39 ml (11% of the mean value), respectively. Doppler indices assessing diastolic filling and colour Doppler area of mitral regurgitant jet showed greater variations on a mid-term and day-to-day basis. ‘Spontaneous’ shifts from a predominant early diastolic velocity to a predominant late diastolic velocity mitral flow pattern were observed in four patients with coronary artery disease and less compromised left ventricular ejection fraction. We conclude that, in patients with severe heart failure, Doppler measurements have a good short-term intra- and inter-observer reproducibility and, therefore, may be suitable for assessing acute interventions. For systolic indices, day-to-day variability was also fairly small, provided that a rigorous—but practicably obtainable—methodology is used. The relatively large mid-term and day-to-day variability of diastolic variables, observed in less compromised patients, should be considered when serial Doppler studies are performed.