Diastolic Echocardiographic Examination

Transmitral Inflow

The transmitral inflow profile consists of an early diastolic (E) and atrial contraction (A) velocities ( Fig. 17.2 ). The ratio between these two peak velocities (E/A ratio) has a satisfactory correlation with LVFP in patients with established or recognized LV diastolic dysfunction. This means that when there is known diastolic dysfunction, the transmitral E/A ratio alone can often be used to accurately estimate LVFP. For example, a decreased E/A ratio (≤0.8) due to a low E velocity (≤50 cm/sec) and increased filling at atrial contraction (A velocity), along with prolongation of the deceleration time (DT), is consistent with impaired relaxation and normal LVFP; while a markedly increased E/A ratio (≥2.0), with a high E velocity and a low A velocity, along with a short DT, is consistent with advanced diastolic dysfunction and a marked elevation in LVFP.

One of the limitations of using the transmitral inflow profile alone for assessing LVFP relates to the competing effects of impaired relaxation and rising LVFP. That is, in the setting of impaired relaxation in conjunction with a moderate reduction in LV compliance and a moderate increase in LA pressure (LAP), the decreased E/A ratio and prolonged DT that are typical of impaired relaxation will normalize to create a pseudonormal transmitral inflow profile. For this transmitral inflow pattern, additional parameters are still required to accurately estimate LV filling pressures. Furthermore, the normal values for transmitral indices are age dependent. As such, age should be taken into account when evaluating the significance of transmitral Doppler filling patterns. The normal values for various age groups are listed in Table 17.1 .

Mitral Annular Velocities

The peak mitral annular e′ velocity acquired with tissue Doppler imaging (TDI) reflects LV relaxation in early diastole. In particular, the peak e′ velocity is inversely related to the time constant of isovolumic relaxation (tau [τ]); thus when there is prolongation of LV relaxation (a longer τ), there is a lower e′ velocity. Importantly, the e′ velocity is relatively load independent and since in most patients impaired relaxation is present across all grades of diastolic dysfunction, a reduction in the e′ velocity is also present across all stages of diastolic dysfunction. However, as for the transmitral inflow velocities, the normal values for the e′ velocity are age dependent. In addition, normal values for the e′ velocity is dependent on the sampled location (septal vs. lateral annulus). Therefore age as well as sampled location should be considered when evaluating these velocities. The normal e′ values for various age groups and sample locations are listed in Table 17.1 .

Another application of the e′ velocity is in the calculation of the E/e′ ratio. As stated, the transmitral inflow profile is load dependent, while the e′ velocity is relatively load independent. Hence the E/e′ ratio predominately adjusts for the influence of LV relaxation on the transmitral inflow profile. For example, in the individual with normal LAP, both the transmitral E and e′ velocities are normal; therefore the E/e′ ratio will be low. With impaired relaxation plus an elevated preload (LAP), the E velocity is normal or increased but the e′ velocity will be reduced; therefore the E/e′ ratio will be increased. Numerous studies have shown that there is a strong relationship between the E/e′ ratio, with the e′ acquired at either the septal or lateral annulus, and the LVFP. In particular, an E/e′ ratio (averaged at the septal and lateral annulus) greater than 14 is rarely seen in normal individuals of any age (see Table 17.1 ).

Tricuspid Regurgitant Velocity

In the absence of pulmonary stenosis or right ventricular outflow tract (RVOT) obstruction, the peak TR velocity combined with right atrial pressure (RAP) can be used to estimate the pulmonary artery systolic pressure (PASP). In the absence of pulmonary vascular disease, an elevated PASP is seen as a marker of elevated LVFP. In this instance, it is assumed that the increased PASP is due to left heart disease (or postcapillary pulmonary hypertension) where there is an elevation in the LAP, which leads to increased back pressure to the lungs and a subsequent increase in PASP. A peak TR velocity value of 2.8 m/sec or less is generally considered normal. Hence a peak TR velocity value greater than 2.8 m/sec, in the absence of pulmonary vascular disease, indicates elevated LVFP.

Indexed LA Volume

The LA size is considered a morphophysiologic expression of LV diastolic dysfunction and increased LVFP. For instance, a dilated LA suggests a chronic elevation in LAP. It has been found that an indexed LA volume of 34 mL/m ² distinguished the presence of elevated LVFP (defined by a septal E/e′ >15) with 86% sensitivity and 66% specificity. Thus an indexed LAV greater than 34 mL/m ² may indicate an increased LAP secondary to chronically increased LVFP.

LV Wall Thickness and Mass

Hypertensive heart disease is the most common abnormality leading to diastolic heart failure. As increased LV wall thickness and mass are common findings associated with hypertensive heart disease, the mere presence of increased LV wall thickness and mass are clues to the presence of abnormal myocardial structure and thus underlying LV diastolic dysfunction. For instance, when there is a marked increase in LV wall thickness and LV mass, an impaired relaxation pattern is expected on the transmitral inflow profile. The absence of an impaired relaxation pattern suggests a decrease in LV compliance and an increased LAP.

Pulmonary Venous Inflow

The normal pulmonary venous profile is characterized by three distinct waveforms: (1) systolic forward flow (S velocity), (2) diastolic forward flow (D velocity), and (3) atrial flow reversal during atrial systole (Ar velocity) (see Fig. 17.8 later in the chapter). In approximately one-third of normal individuals, the S waveform may be biphasic consisting of an S1 velocity and an S2 velocity.

Pulmonary venous waveforms are influenced by LA contraction, relaxation, and pressure as well as by LV compliance and filling pressures. The pulmonary venous S velocity is closely related to the LAP. The pulmonary venous D velocity is influenced by changes in LV filling and compliance, and essentially parallels the transmitral E velocity. The magnitude and duration of the Ar waveform is influenced by LV late diastolic pressures, atrial preload, and LA contractility. In particular, increased LAP decreases the pulmonary venous-to-LA pressure gradient in systole and increases the pulmonary vein–LA pressure gradient during diastole; this results in a decreased peak S velocity, an increased peak D velocity, and a subsequent decrease in the S/D ratio. Generally a decreased S/D ratio less than 1 in adult patients with a depressed LVEF is indicative of increased LAP.

The pulmonary venous flow profile can also be used to identify patients with an elevated LV end-diastolic pressure (LVEDP). When the LVEDP is increased, there is an earlier than normal crossover of pressures between the LA and the LV, which effectively shortens the transmitral A wave duration. As a result, the pulmonary venous Ar duration is longer than the transmitral A duration since the transmitral A duration is shortened by the elevated LVEDP. Difference between the pulmonary vein Ar and the transmitral A durations has shown a strong correlation with the LVEDP with a difference of 30 msec or more providing a good indicator for an elevated LVEDP above 15 mmHg.

Isovolumic Relaxation Time

The isovolumic relaxation time (IVRT) is the time interval between aortic valve closure and mitral valve opening. When LV relaxation is impaired, the fall in LV pressure is slow so that the normal crossover between the LA and LV pressures is delayed, which also delays the opening of the mitral valve and prolongs the IVRT. However, as for the transmitral inflow profile, the IVRT is both load dependent and age dependent ( Table 17.2 ). Importantly, despite the presence of impaired relaxation, increased LAP normalizes the IVRT; with a further elevation in the LAP, the IVRT shortens. Hence the use of the IVRT in isolation has limited accuracy in the assessment of LV diastolic function. However, when combined with other key variables, the IVRT may be useful in assessing LV diastolic function in patients with heart failure and reduced LVEF, coexistent mitral valve disease, mitral annular calcification, or atrial fibrillation.

Color M-mode Propagation Velocity

Similar to transmitral inflow, LV intracavitary filling is comprised of two waveforms: an early diastolic wave and an atrial contraction wave (see Fig. 17.10 later in the chapter). Using color M-mode (CMM), these waveforms can be mapped as blood flows from the mitral annulus toward the LV apex. The flow propagation velocity (Vp) is a measurement of the velocity at which flow propagates within the LV during early diastole. The Vp is indirectly related to the time constant of relaxation (τ) such that the longer it takes for the ventricle to relax, the slower the Vp.

The E/Vp ratio has been found to correlate with the LVFP. Like the E/e′ ratio, the E/Vp unmasks the influence of LV relaxation on the transmitral inflow profile. For example, in the individual with normal LAP, both the transmitral E velocity and Vp are normal; therefore the E/Vp ratio will be low. With impaired relaxation plus an elevated preload (LAP), the E velocity is normal or increased but the Vp is slow; therefore the E/Vp ratio will be increased. However, the Vp can fall into the normal range in patients with normal LV cavity size and ejection fraction, despite the presence of impaired relaxation; hence the E/Vp ratio may not accurately predict LVFP in these patients.

A more detailed description of the role of CMM propagation velocity in the assessment of LV diastolic function is discussed in Chapter 11 .

LV and LA Strain

LV global longitudinal strain (GLS) is a measure of longitudinal deformation of the LV during systole, and abnormally low values indicate impaired global longitudinal function. A reduction in GLS can precede the development of overt systolic dysfunction as measured by traditional methods such as LVEF. Thus abnormal LV GLS identifies patients with abnormal myocardial mechanics and therefore suggests the presence of diastolic dysfunction, but as a single parameter it has not been found to be an accurate predictor of LVFP.

LV global diastolic strain rate (GDSR) in the isovolumic relaxation period and in early diastole has demonstrated good correlation with the time constant of relaxation (τ). Reduced values suggest the presence of impaired relaxation. When combined with the transmitral E velocity, GDSR has been shown to correlate with LVFP.

LA strain is an emerging parameter used for the assessment of LA function. Peak LA strain has been shown to be abnormal in patients with systolic and diastolic heart failure. In patients with preserved LVEF, peak LA strain values have demonstrated a stepwise decline with increasing grades of diastolic dysfunction and thus may allow more accurate categorization of diastolic dysfunction severity.

LV and LA strain, although not part of the current diagnostic algorithms for diastolic function assessment, are promising parameters and await ongoing validation in larger clinical studies. A more detailed description of the application of strain imaging in the assessment of LV diastolic function is discussed in Chapter 10 and 12 .