topeeng.gif (8416 bytes)

[ Scientific Activities - Actividades Científicas ]

Assessment of Diastolic Function Using Tissue Doppler Echocardiography

Leonardo Rodriguez, M.D, FACC

Cardiovascular Imaging Center
Department of Cardiology
Cleveland Clinic Foundation


Diastolic dysfunction is an important cause of morbidity in patients with heart disease. It may exertional symptoms in elderly patients with hypertensive heart disease. Diastolic dysfunction can be defined as the inability of the heart of maintaining normal diastolic pressures during left ventricular filling. Diastolic dysfunction may be caused by abnormalities in relaxation, increased stiffness or a combination of both.Constrictive pericarditis is another cause of severe impairment of ventricular filling.

The gold standard for evaluation of diastolic dysfunction has been the direct measurement of left ventricular pressure with simultaneous left ventricular volume estimation. This allows us to construct pressure-volume curves and determine the exact relationship between these parameters. Unfortunately, this requires invasive placement of intraventricular catheters with sophisticated micromanometers, which make this method unsuitable for routine use or for serial studies.

Echocardiography is now the method more commonly used for assessment of diastolic function. In addition to high-resolution 2-D images, Doppler flow velocity offers important information about the dynamics of ventricular filling. Pulsed wave Doppler of the mitral and pulmonary veins are used for routine assessment of left ventricular diastolic function. Similarly, Doppler flow of the tricuspid and hepatic veins is used to evaluate right ventricular diastolic function.

An important limitation of the spectral Doppler assessment of diastolic function is its dependence of loading conditions. With worsening left ventricular diastolic function there is a compensatory increase in left atrial pressure, increase in the velocity of the E wave of the mitral inflow and pseudonormalization of the filling pattern (normal E/A ratio and deceleration time).

In an individual case it can be difficult to know whether the patient is in the descending or ascending limb of the curve.

Recently, 2 new technologies, color M-mode and Doppler Tissue echocardiography, have emerged that are very promising in complementing the information provided by Doppler echocardiography and may allows us a more complete evaluation of diastolic function.

Doppler Tissue Echocardiography

Doppler tissue echocardiography (DTE) is a recent modality developed for the assessing of left ventricular function. The physical principles of this method are similar to those of the flow Doppler velocities. The fundamental differences between these 2 modalities are related to machine instrumentation. The Doppler signals coming from blood flow have low amplitude and high velocity. In contrast, DTE signals have high amplitude and low velocities. In order to display DTE signals the low pass filter is eliminated and Doppler gain lowered. DTE shares the same limitations that flow Doppler, the most important of which is angle dependence.

DTE Modes:

Pulsed Doppler: this is the easiest way to measure myocardial velocities and has been used for interrogation of myocardial or mitral annular velocities. Using this modality a sample volume is placed on the myocardium and a spectral display is obtained. An example of pulsed Doppler spectral velocities is shown. Typically there is one systolic wave and 2 diastolic waves

Image6.gif (10627 bytes)

Color Doppler mapping: color-coded velocities are displayed over the 2D image. Different color maps can be used more often showing velocities toward the transducer coded in red and away from the transducer coded in blue.


DTE color M-mode: displays velocities along a single scan line.

More recently some machines can display velocities along a curvilinear cursor (curved anatomical M-Mode) that can follow the ventricular contour and measure simultaneously velocities in different myocardial segments.

Power Doppler imaging: This modality displays not velocities but the signal amplitude and has the advantage of being angle independent. Intense investigation is under way to evaluate the role of Power Doppler in echocardiographic myocardial perfusion using intravenous contrast agents.

Myocardial velocity gradients (MVG): this is more complex method that measures the difference in velocity between the endocardium and the epicardium. MVG appears to be independent of translation.

Myocardial strain: measuring the rate of change in velocity between 2 points its possible to calculate myocardial strain rate. It appears to be promising in the evaluation of regional systolic function during dobutamine echo. It may be applicable for regional diastolic function assessment.

Assessment of diastolic function using DTE

Myocardial velocities during left ventricular filling have been used in the assessment of diastolic function. Clinical studies have found that the early diastolic velocity (Em) is inversely correlated with the time constant of isovolumic relaxation (t). Mitral annular velocities appear to be less dependent on loading conditions than mitral inflow velocities. Diastolic velocities of mitral annular motion have used by us and others for assessing LV diastolic function in a variety of clinical conditions. Motion of the mitral annulus reflects the dynamics of the longitudinal axis of the heart.

Effects of age and LVH on DTE velocities:

Diastolic annular velocities related to age as the early diastolic velocity decreased and the atrial systolic velocity increased in older subjects. Consequently, the early/late annular velocity ratio also decreased with age. In normal subjects these changes closely follow changes in E/A ratio of mitral inflow velocities. However, in pathological conditions this concordance is altered. The relative independence between mitral inflow velocities and annular velocities is evident in the patients with left ventricular hypertrophy where the E/A ratio of the mitral inflow may be higher than the Eann/Aann by DTE. It is likely that some of the patients with higher E/A ratios have increased left atrial pressures with pseudonormalization of the mitral inflow pattern but with persistent slow annular motion.

This can be explained by the fact that mitral annulus velocities reflects structural mechanics while mitral Doppler inflow velocities reflects fluid dynamics which are influenced not only by the rate of filling but also by the left atrial pressure. DTE velocities have used to differentiate the physiologic hypertrophy of athletes from hypertrophic cardiomyopathy.

Differentiation of normal from pseudonormal left ventricular filling pattern.

Mitral Doppler inflow velocities are the most commonly used method for assessing left ventricular filling. One of the major limitations of this method is its dependence on loading conditions. In a patient with delayed relaxation (E/A<1) further impairment of diastolic function may lead to an increase in left atrial pressure. Elevated filling pressures will cause an increase in E wave velocity and therefore pseudonormalization of the E/A ratio. DTE has been used for differentiation between normal and pseudonormal pattern. In the latter group early diastolic velocities are significantly reduced. Farias et. al. from our laboratory studied patients with delayed relaxation, pseudonormal and restrictive mitral filling pattern and compared them with normal control group. He found that the traditional Doppler indices have a "U" shape as they progress from normal to restrictive pattern. In contrast, DTE showed a uniphasic decline in velocities.

Constrictive pericarditis vs. restrictive cardiomyopathy

Differentiation between constrictive pericarditis and restrictive cardiomyopathy is difficult using Doppler echocardiography. Both conditions show a tall E wave and short deceleration time (restrictive filling pattern). Garcia et al. measured annular velocities in patients with constrictive pericarditis and restrictive cardiomyopathy. Myocardial Ew velocity was significantly higher in normals (14.5 ± 4.7 cm/sec) and patients with constriction (14.8 ± 4.8 cm/sec) compared to those with restriction (5.1 ± 1.4 cm/sec). By multivariate analysis, Ew was the best parameter for differentiating constriction from restriction. This study indicated that longitudinal axis expansion velocities are markedly reduced in patients with restrictive cardiomyopathy. These velocities had no correlation with peak early transmitral flow velocities suggesting again that they are relatively preload independent. Thus, the measurement of longitudinal axis expansion velocities provides a clinically useful separation between constrictive pericarditis and restrictive cardiomyopathy and may prove to be valuable in the study of other patients with diastolic function.

Left atrial function and estimation of pulmonary wedge pressure.

Late diastolic annular velocities correspond to the motion of the mitral annulus during atrial contraction. These velocities correlate closely with left atrial systolic function. We compared the left atrial fractional area change (a surrogate for left atrial ejection fraction) with the A wave annular velocities. The correlation between these two parameters was good with the advantage that annular velocities were very easy to record and measure.

Nagueh et al studied 100 patients with simultaneous Doppler and invasive hemodynamics. They found a good correlation between PCWP and a ration of the E wave of the mitral inflow and E wave of the mitral annular velocities (E/Ea), PCWP=1.55+1.47(E/Ea), r=0.86. We studied a group of patients with severe LV dysfunction and congestive heart failure and found a good correlation between wedge pressure and Sa and Aa waves of the mitral annular velocities.

Color M-mode flow propagation

Color M-mode Doppler is a pulsed Doppler technique that allows to obtain a spatio-temporal velocity map with a temporal resolution of 5 ms, a spatial resolution of 300 µm, and a velocity resolution of about 3 cm/sec. A major difference exists between Color M-mode Doppler and normal pulsed Doppler as the latter yields information on the temporal course of velocity at a fixed spatial point while color M-mode Doppler allows acquisition of information on velocity, time and space. By obtaining information on these three parameters, a more complete picture of the filling pattern can be acquired.

Distance (cm)

Time (sec)

Time, velocity and space are, along with pressure, variables of all hydrodynamic systems. Brun et al, using the color M-mode of early left ventricular filling found that the slope of the transition color-noncolor was strongly correlated with the time constant of isovolumetric relaxation, tau (t). He also showed that this slope tended to be slower in patients with conditions associated with abnormalities of the relaxation process, such as dilated and ischemic cardiomyopathy and in patients with hypertrophied ventricles. In these groups of patients the slope of flow velocity propagation was not related with the peak E wave of pulsed Doppler. Stugaard et al. using the time difference between the maximal velocity at the mitral level and the maximal velocity at the apex, also found a strong correlation with tau. She studied patients with coronary disease during coronary occlusion and observed significant changes in flow propagation during ischemia. One the most important aspects of the color M-mode flow propagation is that appear to be independent of loading conditions. Clinical studies and more recently animal studies all have found that color M-mode flow propagation is not affected by changing pre-load conditions. This load independence of flow propagation is a unique and very important feature among all the Doppler-base method of diastolic function assessment. Garcia et al. recently published the application of color M-mode for calculation of capillary wedge pressure. We found that a combination of pulsed wave peak E velocity and the slope of flow propagation allowed us to estimate capillary wedge pressure (CPWP=5.27*(E/Vp) + 4.6; r=0.80).

The main problem with color M-mode flow propagation is its measurement. Several methods have been proposed. The earliest description by Brun used the slope of the transition between no color and color. This has the limitation of being affected by isovolumic flow and has significant interobserver variability. Stutgaard used the time delay of peak velocities from base to apex. Although it has been shown to have better inter and intraobserver agreement, it requires decoding of digital velocities a feature not available in current ultrasound machines. Takatsuji et al. modified the aliasing velocity to determine the slope of a line connecting the maximal velocity near the mitral orifice and the point in the mid-ventricle at which the velocity decreased to 70% of its initial value. They found that the flow propagation was significantly slower in patients with abnormal diastolic function compared to normals.

Measuring the slope of propagation or the time delay don’t take advantage of the vast amount of information that is present in a color M-mode tracing. Using the velocity, temporal and spatial data contained in the color M-mode Doppler, it is possible to obtain intracavitary gradients. This requires intensive computational resources but it may be the ultimate gold standard. Currently in our laboratory, we measure the slope of the first isovelocity line that produces an uninterrupted contour. The Nyquist velocity is typically set around 45 cm/sec and the sweep speed at 100 cm/s. As the value of the slope can differ depending on the ultrasound system used, it is important for each laboratory of obtain its own set of normal values.


Doppler tissue echocardiography is a new promising technique, which offers useful information about left ventricular diastolic function. Myocardial or annular velocities are easy to obtain and offer a rapid way to differentiate normal from pseudonormal pattern and constrictive from restrictive physiology. Color M-mode with its relative independence from preload makes it a unique marker of diastolic dysfunction. Preliminary data is promising that with a combination of Doppler mitral inflow and DTE or color M-mode parameters it will be possible to estimate non-invasively left ventricular filling pressures.