DIASTOLOGY

ON-INVASIVE ASSESSMENT OF LEFT VENTRICULAR DIASTOLIC (DYS)FUNCTION AND FILLING PRESSURE

J.R.T.C. Roelandt, MD, PhD, FACC*; M. Pozzoli, MD** Thoraxcentre, Erasmus University Medical Centre Rotterdam, Rotterdam, The Netherlands **Department of Cardiology, A. Manzoni Hospital, Lecco, Italy.

the rate of relaxation of the heart ”is quite as important as the systolic contraction”… If an old man’s heart relaxes slowly, his capacity for physical exertion is thus limited.

                                                                                     (Yendell Handerson, 1923)

 Introduction

 Factors controlling diastolic LV function

 Doppler assessment of diastolic LV filling

        A. Normal transmitral Doppler flow velocity pattern
        B. Normal Pulmonary vein Doppler Flow velocity pattern
        C. Doppler tissue velocity imaging.
        D. Color M-mode transmitral flow

 The assessment of diastolic LV dysfunction

 Estimation of LV filling pressure (table 4)

 Specific conditions
       Atrial fibrillation
       Constrictive pericarditis
       Hypertrophic cardiomyopathy   
 Conclusion
 References
 Gambia Silver

Introduction

The impaired capacity of the left ventricle (LV) to fill and to maintain its stroke volume without a compensatory increase of atrial filling pressures represents diastolic dysfunction.
This results in pulmonary and/or systemic venous congestion with resultant dyspnea.
Isolated diastolic dysfunction is a relatively common problem and accounts for up to 30% of heart failure.(1).
In these patients there is a preserved systolic function (ejection fraction > 50%).
The condition often precedes the progression of systolic dysfunction and is a major determinant of the symptoms of patients with systolic heart failure.
Therefore, assessment of diastolic LV function and estimation of filling pressures is an important part of the management of patients with heart disease. It is most frequently due to coronary heart disease or LV hypertrophy (hypertension) and common in the elderly because ageing causes stiffening of the arterial system with an increase in systolic arterial pressure. This causes myocardial hypertrophy and prolonged relaxation.(2). Common conditions associated with diastolic dysfunction and heart failure are listed in table 1. Many of the problems in the assessment of diastolic function in cardiac disease have been caused by its multifactorial nature and the lack of accurate methods for its assessment.
Hi-fi pressure recordings, angiography, radionuclide techniques and combined pressure-volume relationships have been used but these methods are invasive and cannot be used for repeated examination. It is well known that physical signs provide limited information for the hemodynamic assessment of patients with chronic heart failure.(3).
Recent studies have validated Doppler flow and tissue velocity measurements for reliable and noninvasive evaluation of LV filling dynamics and rational therapeutic strategies in patients presenting with signs and symptoms of heart failure.(4,5).

                                             Table 1.

Factors controlling diastolic LV function

Diastole is divided in four phases: isovolumic relaxation phase, rapid filling phase, slow filling phase (diastasis) and atrial contraction phase.

Table 2. The different factors that influence diastolic finction of the left ventricle and their importance at different phases of diastole.

Its function is complex and is determined by a sequence of multiple interacting intrinsic characteristics of the myocardium and alterations in extrinsic (loading) conditions (table 2)(6).
Myocardial relaxation encompasses the end of myocardial shortening, the isovolumic relaxation period and the rapid filling phase. The time rate of relaxation is an important determinant of diastolic function and is regulated by active biochemical processes (myocyte inactivation), loading conditions, nonuniformity and elastic recoil (suction) of the myocardium.
Myocardial muscle is stiffer during contraction.
This implies that when relaxation is prolonged or incoordinate, parts of the myocardium are still in an active state during early diastole and stiffer than normal.
This situation occurs in the failing and hypertrophic myocardium and will, therefore, produce an upwards shift of the diastolic pressure-volume (DP-V) relationship during early diastole.
The DP-V relationship during the passive steady state is governed by elastic myocardial properties (compliance) and reflects the interaction between extrinsic forces resisting the stretching from within the myocardium.
Stretching represents the summation of the individual properties of muscle, extracellular matrix and coronary blood vessels. Changes in these properties occur when the relative quantity of collagen is increased, such as in hypertrophy, fibrosis, scar formation due to myocardial infarction or infiltration (e.g. amyloid) and these properties will have a greater influence on the DP-V relationship at larger LV volumes.
Since the DP-V relationship shows a concave curvilinearity, LV compliance also decreases when its volume becomes larger even the absence of these myocardial abnormalities(3). Viscoelasticity is a fundamental muscle property which manifests as an additional LV pressure increase during relatively rapid volume changes.
In the presence of cardiac hypertrophy there is an increase of perimyseal collagen weaves and the viscoelastic effects are more pronounced and predominant during atrial contraction.
Vascular volume changes (coronary engorgement) are probably not very important although their quantitative effects have not been adequately studied. Loading (or extrinsic) factors include changes in transmural pressure, vascular volume and afterload or preload.
Changes in left ventricular transmural pressure depend on changes in right ventricular pressure (ventricular interaction) and pericardial pressure, both of which may change considerably in heart failure.
An increase in right ventricular pressure shifts the septum and the LV DP-V relationship upwards over the entire volume range. Pericardial pressure shifts the LV DP-V relationship, but will change the volume range within which the ventricle is working. The questions are what the magnitude is of each of these factors and when during diastole each of them affects the DP-V relationship?

The rate of relaxation with the diastolic suction in early diastole and the LV compliance (LV fibrosis /scar, extent of relaxation and the ventricular interaction modulated by the pericardium) in mid-late diastole are the main pathophysiologic factors that affect diastolic LV function.(6).

 

Doppler assessment of diastolic LV filling

A. Normal transmitral Doppler flow velocity pattern                           B. Normal Pulmonary vein Doppler Flow velocity pattern                           C. Doppler tissue velocity imaging.                           D. Color M-mode transmitral flow

A. Normal transmitral Doppler flow velocity pattern

Doppler echocardiography provides a noninvasive means to study LV diastolic function by recording transmitral blood flow velocities versus time.
These velocity waveforms are determined by the complex interplay of hemodynamics (instantaneous pressure differences) and several other factors (table 2).
Transmitral pulsed wave (PW) Doppler flow velocities are recorded within the apical four chamber or apical long axis views and several measurements can be used to define left ventricular filling hemodynamics.
As the mitral valve is funnel-shaped, the velocities increase progressively across the mitral valve apparatus towards the outlet of the mitral funnel.
For reasons of reproducibility, all transmitral PW Doppler flow measurements should be made with the sample volume in the same position at the outlet of the mitral valve funnel. Figure 1 diagrammatically shows the normal transmitral Doppler flow velocity pattern and the parameters which can be measured. The isovolumic relaxation period (IRP), is the time interval between aortic valve closure and mitral valve opening and can be measured from the simultaneous Doppler and M-mode echocardiograms or more accurately from a simultaneously recorded phonocardiogram and transmitral Doppler curve.
IRP reflects the speed of the initial part of myocardial relaxation. Prolonged IRP is a sensitive marker of abnormal myocardial relaxation. Normal transmitral blood flow is laminar and relatively low in velocity (usually < 1 m/sec).
There is an early diastolic velocity caused by the continued myocardial relaxation resulting in a LV pressure below LA pressure which causes the mitral valve to open and rapid LV filling to occur (E wave).
E wave acceleration is directly determined by LA pressure and inversely related to myocardial relaxation. Viscoelastic properties and compliance of the myocardium then come into play, raising LV pressure and resulting in a decreased transmitral flow velocity. The rate of fall in velocity is represented by the deceleration time (DT) and is a measure of how rapidly early diastolic filling stops. DT becomes shorter when LV compliance decreases(7).
There is an inverse relationship between the mean LA pressure and

 Figure 1

Figure 1.

This diagram shows intracardiac pressure tracings from the left ventricle and left atrium with the corresponding Doppler mitral (MVF) and pulmonary vein flow (PVF) velocity patterns.
The isovolumic relaxation period (IRP) is the interval between aortic valve closure (Ac) and the onset of LV filling at mitral valve opening (Mo) on the PW Doppler trace.
Aortic valve closure is best defined on a simultaneously recorded phonocardiogram but a more practical way is by positioning the sample volume of the PW Doppler halfway between the anterior mitral leaflet and the outflow tract.
The time interval between the end of the outflow tract velocity and the onset of the mitral inflow velocity waveform represents IRP.
During relaxation there is a pressure cross-over between left atrial and left ventricular pressure, which causes the mitral valve to open and rapid filling to occur (E wave).
In this part of the cardiac cycle LV relaxation is still ongoing causing a continuing drop in LV pressure.
The area under the E wave is the time velocity integral (TVI) and reflects the contribution of the rapid filling phase in the LV diastolic filling.
The deceleration time (DT) is a measure of how rapidly early diastolic filling stops. It is represented by the time-interval between the E peak and a point on the baseline where the descending limb crosses the baseline. The A wave is associated with atrial contraction and is an important index of diastolic function. The area under the A wave is the time velocity integral (TVI) and reflects the contribution of atrial contraction to LV diastolic filling. The normal pulmonary venous flow usually has a biphasic (occasionally triphasic) flow with a slightly greater systolic (S wave) than diastolic wave (D wave) and a small retrograde flow wave during atrial contraction (AR) The AR wave may become larger with increasing age.

DT. Inertia effects may cause continued forward low velocity flow during mid-diastole. The higher LA pressure during its contraction causes an increase in velocity (A wave) and is an important parameter of diastolic function. Commonly, the E/A ratio is used to assess LV diastolic function and is > 1.

Table 3. Assessment of diastolic (dys)function transmitral Doppler velocities

The normal reference values in normal healthy subjects
 (± 1 SD) are presented in Table 3.

B. Normal Pulmonary vein Doppler Flow velocity pattern

The pulmonary veins are readily imaged from the transesophageal approach.
However, in many patients the flow velocity profile of the right upper pulmonary vein can be interrogated from a foreshortened apical cross section with the Doppler sample volume placed just inside the vein.
The normal pulmonary vein flow pattern is diagrammatically in figure 1.
It is usually biphasic with a predominant systolic forward flow (S wave) and a less prominent diastolic forward flow wave (D wave).
Occasionally, there may be a triphasic flow pattern with two distinct systolic flow waves of which the initial flow into the left atrium results from atrial relaxation followed by a further inflow due to the increase in pulmonary venous pressure.
The D-wave occurs when there is an open conduit between the pulmonary vein, LA and LV and reflects the transmitral E wave.
A retrograde flow wave into the pulmonary vein (AR wave) occurs during atrial contraction and its amplitude and duration are related to LV diastolic pressure, LA compliance and heart rate.
8 In normal subjects, the amplitude of the AR wave is generally less than 25 cm/sec and its duration is shorter than the A wave of the transmitral A wave.

C. Doppler tissue velocity imaging.

Doppler tissue velocity imaging (DTI) provides the velocity of motion within tissue.
In the absence of gross shape abnormalities or severe regional wall motion abnormalities, mitral annulus motion reflects LV volume changes rather than pressure differences between the LA and the LV and is less load dependent than the transmitral inflow velocity pattern.
Mitral annulus velocity can be measured by DTI.
It should be noted that mitral annulus velocities may differ around its circumference and that an average value of septal, anterior, lateral and posterior velocities should be used especially in patients with regional wall motion abnormalities.
In the systolic phase the annulus shows anterior motion which starts after the isovolumic contraction.
The diastolic phase includes the IRP, early rapid diastolic filling (E’ wave) followed by diastasis and atrial contraction causes the late diastolic A’ wave (figure 3).
The E’ wave is larger than the A’ wave and mirrors the transmitral flow velocity pattern.
Peak systolic mitral annulus velocity correlates with the peak rate of LV pressure rise (dP/dt).
9 TDI annulus velocity measurements can be used for differentiating pseudo-normal from normal transmitral inflow patterns since early diastolic velocity of the mitral annulus (E’ wave) is relatively pre-load independent and correlates with LV relaxation(10).
The peak transmitral E wave velocity/TDI E’ wave velocity ratio has been found to correlate with main pulmonary capillary wedge pressure. The normal mitral annulus TDI velocity pattern is shown in figures 3 and 4

D. Color M-mode transmitral flow

With color M-mode both the temporal and spatial propagation of flow is visualised along a single scan line over the entire length of the LV throughout diastole rather than in a single sample volume. The rate of flow propagation from the LA to the LV is determined by the rate of relaxation and the diastolic suction of the LV and is helpful in

Figure 2

Figure 2. These are diagrams showing typical mitral valve flow (MVF), pulmonary vein flow (PVF), tissue Doppler mitral annulus velocity (TDI), and color M-mode patterns of the various stages of diastolic dysfunction (modified from Garcia and Thomas, Echocardiography 1999;16:501-8). Figure 3

Figure 3. These recordings are obtained from a normal individual and patients with diastolic dysfunction. A) normal individual — 70 years old (6.581.360); B) abnormal relaxation in a 55-year-old man with LVH (on hemodialysis- 1.677.171); C) a 44-year old man with LV hypertrophy (8.090.101) and D) a 41-year-old man with end-stage heart failure (5.534.419) (see also diagrams of figure 3). differentiating between normal and pseudo-normal transmitral flow velocity patterns(11,12,13). In practice, a M-mode cursor is placed parallel to the mitral inflow and adjusted so that the longest column of color flow from mitral valve to apex is recorded. The propagation velocity (Pv) is measured by the slope along a distinct isovelocity (aliasing) line during early filling from the mitral valve plane up to 4 cm into the LV cavity. The normal value is > 40 cm/sec. The normal propagation pattern is shown in figures 3 and 4. The color M-mode method offers a practical means to rapidly screen for a diastolic inflow abnormality during the examination procedure. The assessment of diastolic LV dysfunction The basic patterns of abnormal transmitral and pulmonary vein Doppler flow were first described by Kitabatake et al. (14). and are diagrammatically shown in figure 2. Different grades of LV diastolic dysfunction result in different transmitral and pulmonary venous flow velocity patterns. Pattern 1 is seen in patients with impaired myocardial relaxation, normal diastolic LA and LV pressures. The pattern is characterised by a prolonged IRP, a decrease in the E wave amplitude, prolonged DT and an increase in A wave amplitude, the latter being a reflection of compensatory increase in atrial contribution to diastolic filling. The E/A ratio is < 1. The pulmonary venous flow velocity pattern may show a diminished D wave as a result of a reduced early diastolic filling and an increased AR wave when the LV end-diastolic pressure is elevated. The E’ wave on the TDI recording is decreased and the rate of flow propagation in the LV on the color M-mode is decreased. The abnormal relaxation pattern is a common pattern in elderly but represents the earliest manifestation of diastolic dysfunction in younger individuals. Figure 4 Figure 4. Diagrams of the basic abnormal transmitral Doppler flow velocity patterns corresponding to different grades of LV dysfunction. Pattern 1 is seen in patients with abnormal relaxation. IRP is prolonged, the E wave decreased, DT prolonged, the A wave increased resulting from increased atrial contribution to left ventricular filling. The E/A ratio is < 1 and the D wave on the PVF becomes smaller proportional to the reduction of the E wave. Pattern 2 shows pseudo-normalisation when the LV compliance decreases causing an increased diastolic filling pressure (DFP). In these patients, there is increased flow reversal into the pulmonary veins and the amplitude of the AR wave increases as well as its duration. Pattern 3 is the typical waveform of a restrictive LV filling pattern when the LV compliance further deteriorates causing an increased E wave velocity, shortened deceleration time (DT) and low-A wave velocity. Most of the atrial contraction results in a flow reversal into the pulmonary veins and the AR duration is much longer than the transmitral A duration.

When diastolic LV function deteriorates, LV compliance progressively decreases and there is an increase of LA pressure and the diastolic filling pressure.
The transmitral flow pattern and more particularly the E wave normalise.
This pseudo-normal pattern (pattern 2) is a transition pattern from impaired relaxation to restrictive filling.
In certain clinical conditions this pattern may be confusing and must be distinguished from the normal pattern.
In patients that are regularly followed, progression of disease and pseudo-normalisation is usually recognised.
The problem is the patient with minimal symptoms who presents with a normal transmitral velocity pattern.
A psuedo-normal pattern can be identified as follows:

    The LA is (moderately) enlarged (a normal  
    LA size makes diastolic dysfunction very
      unlikely).15

    The AR wave on the pulmonary venous flow 
  trace has an amplitude > 25 cm/sec and  is longer than the transmitral A wave.15,16

    The E’ wave on the TDI recording is
  diminished and the transmitral E/TDI E’ ratio is > 15.10 Color M-mode shows a
      reduced flow

     Color M-mode shows a reduced flow
  propagation rate (< 40 cm/sec).

     Altering loading conditions e.g. by
  nitroglycerine, nitroprusside, or Valsalva maneuver brings out an impaired relaxation
      pattern.

     The rate of fall of mitral and/or aortic
  regurgitation velocity (negative dP/dt) measured from continuous wave Doppler
      velocity recordings provides a direct measurement of the LV relaxation rate and is
      decreased.17,18

As LA pressure increases with further decrease of LV compliance and deterioration of diastolic LV

Table 4

Table 4. Identification of patients with elevated LV filling pressure

function a restrictive filling pattern develops (pattern 3).
The higher LA pressure causes in an earlier opening of the mitral valve and a shortened IRP.
The low compliance of the LV causes a rapid increase in the early LV pressure and a shortened inflow and DT.
The E/A ratio is > 2. Forward diastolic pulmonary vein flow stops in mid-late diastole and during atrial contraction there is a significant flow reversal resulting in a prolonged AR.
DTI shows a diminished E’ wave and color M-mode a low flow propagation rate.

Patients with restrictive diastolic filling can be further classified in those with a reversible and those with irreversible restrictive pattern by altering the loading conditions. Decreasing preload by nitroglycerin or nitroprusside will change the reversible restriction pattern to an impaired relaxation pattern while the pattern of irreversible restrictive remains unchanged(19). The distinction is important for prognostication and management of patients with advanced heart failure. Those with a reversible pattern can still benefit from different therapeutic strategies while those with an irreversible restrictive physiology are candidates for cardiac transplantation. Recent studies have shown that patients with an A’ wave > 5 cm/sec on the TDI mitral annulus recording have a reversible restrictive filling.

Estimation of LV filling pressure (table 4)

It appears that echo/Doppler techniques provide important information for assessment LV diastolic dysfunction which is the cause of an increased LV filling pressure. The question is whether LV filling pressures can be estimated non-invasively by these methods.
Several parameters either alone or in combination can be used to identify patients with an elevated LV filling pressure (table 4). However, these parameters should always be interpreted in the clinical context considering age, symptoms and functional status.
The initial E wave velocity is mainly determined by the pressure in the LA at mitral valve opening and its peak acceleration rate correlates with LV filling pressure.20 In general, a high E wave amplitude indicates a high and a low E wave amplitude, a low mean LA pressure.
In the presence of systolic dysfunction, an E/A ratio >> 2 indicates a high LA pressure while a low E/A ratio indicates a low LA pressure.
When the mean LA pressure is high as a result of a diseased non-compliant LV DT is be shortened.
A DT of less than 150 ms represents a mean LA pressure of 20 mmHg.
It should be noted that in young subjects with a fast relaxation, rapid suction and a highly compliant LV may have a high E wave and a short DT.
The TDI mitral annulus velocity in these subjects is elevated while it is decreased in patients with diastolic dysfunction. On the other hand, patients with a dilated LA and LV will always have an increased filling pressure. The problem for assessment of LV filling pressure is the patient with mild functional impairment and LV systolic dysfunction who has a pseudo-normal transmitral flow velocity pattern.
Indices derived from the pulmonary vein velocity pattern may be helpful to identify a pseudo-normal pattern. The S/D ratio or the systolic forward flow fraction (in percent) decreases and a systolic forward fraction of less than 40% indicates a pulmonary artery wedge pressure higher than 18 mmHg.(21,22).

Other parameters are the amplitude and duration of the AR wave.
When the LV becomes more diseased and non-compliant, transmitral flow after atrial contraction rapidly stops and the flow reversal in the pulmonary veins will increase.
An AR wave velocity > 25 cm/sec and an AR duration longer than 30 ms of that of the transmitral A wave are indicative of a LV end-diastolic pressure of > 15 mmHg.
However, pulmonary vein flow parameters are often difficult to measure especially at faster heart rates.(16).
Two indices which aim at correcting the E wave velocities for the confounding relaxation rate are the early diastolic TDI mitral annulus velocity (E’ wave) and the color M-mode LV propagation velocity (PV).
23,24 The E/E’ ratio is a practical index which can be calculated in most patients. A ratio < 8 indicates a normal filling pressure while a ratio > 15 corresponds to a filling pressure > 15 mmHg.25 The ratio can also be used in a patient with sinus tachycardia (fused mitral E and A waves) and atrial fibrillation.(20,26,27). An E/PV ratio < 2 suggest an elevated LV filling pressure.
Several formulas which allow the estimation of LV filling pressures have been proposed and are presented in table 5.

Table 5. Formulas for the estimation of LV filling pressure

Specific conditions

Atrial fibrillation                           Constrictive pericarditis                           Hypertrophic cardiomyopathy

Atrial fibrillation

In atrial fibrillation there is no atrial contraction and consequently no A wave on the transmitral flow velocity pattern, no AR wave on the pulmonary vein flow and no A’ wave on the TDI mitral annulus velocity recordings. Nonetheless, some parameters can be used to identify patients with an elevated LV filling pressure (table 6).

Table 6. Diastolic dysfunction and atrial fibrillation

Peak E wave velocity and DT vary with the length of the cardiac cycle.
It has been shown that peak E wave acceleration rate correlates with LV filling pressure but this parameter is often difficult to measure.
20 DT is shortened when LV filling pressure is elevated but should not be measured from short RR interval cycles because filling stops very early and the DT may become artificially short.
Therefore it is practical to select cardiac cycles corresponding to a heart rate of 60-80 beats/min.
The S wave on the pulmonary vein flow is low in amplitude since forward flow in to the LA is predominantly diastolic.
The E/E’ ratio,26 and the initial deceleration rate of the D wave,28 flow propagation velocity can be used to estimate LV filling pressure (table 5).

Constrictive pericarditis

This condition can mimic the manifestations of restrictive cardiomypathy and clinically difficult to distinguish.
In patients with constrictive pericarditis an increase in respiratory variation of the early mitral inflow velocity is seen while this is not present in restrictive cardiomyopathy.
These phasic changes result from exaggerated ventricular interdependence.(29).
TDI shows a decreased mitral annulus velocity in restrictive cardiomyopathy while the velocity is normal in constrictive pericarditis.(9).

Hypertrophic cardiomyopathy

Prolonged relaxation is the predominant diastolic abnormality in hypertrophic cardiomyopathy (prolonged IRP, low E wave, slow DT and increased A wave).
Relaxation can be markedly delayed in some myocardial areas resulting in a triphasic mitral inflow pattern.
When the disease progresses and LA pressure increases the early E wave velocity increases and the DT shortens as a result of the diminishing LV compliance.
In some patients an intracavitary reversed gradient can be produced by apical relaxation during the IRP producing flow from base to apex.
A formula to estimate the pre A-wave diastolic pressure has been proposed by Nagueh et al.(30).

Conclusion

Doppler echocardiography provides important information on left ventricular filling dynamics and allows assessment of diastolic LV function.
The left ventricular filling pressure, which is a major determinant of the symptoms of patients with heart failure, can be estimated.
Because it is a non-invasive method, it has great advantages for the management of many patients with heart failure by guiding therapeutic effects and predicing prognosis.

References

1.    Vasan RS, Benjamin EJ, Levy D. Prevalence, clinical features and prognosis of
       diastolic heart failure: an epidemiologic perspective.
       J Am Coll Cardiol 1995;26:1565-4.

2.    Klein AL, Burstow DJ, Tajik AJ, Zachariah PK, Bailey KR, eward JB.
       Effects of age on  left ventricular dimensions and filling dynamics
       in 117 normal persons.  Mayo Clin Proc 1994;69:212-24.

3.   Stevenson LW, Perloff JK. The limited realibility of physical signs for estimating
       hemodynamics in chronic heart failure. JAMA 1989;261:884-8.

4.   Bonow RO, Udelson JE. Left ventricular diastolic dysfunction as a cause of
      congestive heart failure. Ann Intern Med 1992;117:502-10.

5.   Nishimura RA, Tajik AJ. Evaluation of diastolic filling of l eft ventricle in health and
      disease: Doppler echocardiography is the clinician’s Rosetta stone.
       J Am Coll Cardiol 1997;30:8-18.

6.    Gilbert JC, Glantz SA. Determinants of left ventricular filling and of the diastolic
       pressure-volume relation. Circulation Res 1989;64:827-852.

7.    Ohno M, Cheng CP, Little WC. Mechanism of altered patterns of left ventricular filling
       during the development of congestive heart failure. Circulation 1994;89:2241-50.

8.    Klein AL, Tajik AJ. Doppler assessment of pulmonary venous flow in 
       healthy subjects and in patients with heart disease. 
       J Am Soc Echocardiogr 1991;4:379-92.

9.    Garcia MJ, Rodriguez L, Ares M, Griffin BP, Thomas JD, Klein Al. 
       Differentiation of constrictive pericarditis from restrictive cardiomyopathy:
       assessment of left ventricular diastolic velocities in longitudinal axis by
       Doppler tissue  imaging. J Am Coll Cardiol 1996;27:108-114.

10.  Sohn DW, Chai IH, Lee DJ, Kim HC, Kim HS, Oh BH, Lee MM, Park YB, 
       Choi YS,  Seo JD, Lee YW.
       Assessment of mitral annulus velocity by Doppler tissue imaging in
       the evaluation of left ventricular diastolic function.
       J Am Coll Cardiol 1997;30:474-80.

11.  Brun P, Tribouilloy C, Duval AM, Iserin L, Meguira A, Pelle G, Dubois-Rande JL.
       Left ventricular flow propagation during early filling is related to wall relaxation:
       a color M- mode Doppler analysis. J Am Coll Cardiol 1992;20:420- 32.

12.  Takatsuji H, Mikami T, Urasawa K, Teranishi JI, Onozuka H, Takagi C, Makita Y,
       Matsuo H, Kusuoka H, Kitabatake A. A new approach for evaluation of left ventricular
       diastolic function: spatial and temporal analysis of left ventricular filling flow
       propagation by color M-mode Doppler echocardiography.
       J Am Coll Cardiol 1996;27:365-71.


13.  Garcia MJ, Smedira NG, Greenberg NL, Main M, Firstenberg MS, 
       Odabashian J,  Thomas JD. Color M- mode Doppler flow propagation velocity is
       a preload insensitive  index of left ventricular relaxation: animal and human validation.
       J Am Coll Cardiol. 2000;35:201-8


14.  Kitabatake A, Inoue M, Asao M, Tanouchi J, Masuyama T, Abe H, Morita H, Senda S,
       Matsuo H. Transmitral blood flow reflecting diastolic behavior of the left ventricle in
       health and disease: a study by pulsed Doppler technique.
       Jpn Circ J 1982;40:92-102.

15.  Appleton CP, Galloway JM, Gonzalez MS, Gaballa M, Basnight MA. Estimation of left
       ventricular filling pressures using two-dimensional and Doppler echocardiography in
       adult patients with cardiac disease: additional value of analyzing left
       atrial size, left atrial ejection and the difference in duration of pulmonary venous and
       mitral flow velocity at atrial contraction. J Am Coll Cardiol 1993;22:1972-82.

16.  Rossvoll O, Hatle LK. Pulmonary venous flow velocities recorded by transthoracic
       Doppler ultrasound: relation to left ventricular diastolic pressures.
      J Am Coll Cardiol 1993;21:1687-96.

17.  Nishimura RA, Schwartz RS, Tajik AJ, Holmes DR, Jr. Noninvasive measurement of
       rate of left ventricular relaxation by Doppler echocardiography. Validation with
       simultaneous cardiac catheterization. Ciculation 1993;88:146-55.

18.  Yamamoto K, Masuyama T, Doi Y, Naito J, Mano T, Kondo H, Nagano R, 
       Tanouchi J,  Hori M, Kmada T. Noninvasive assessment of left ventricular
       relaxation using  continuous- wave Doppler aortic regurgitation velocity curve. 
       Its comparative value to  the mitral regurgitation method. 
       Circulation 1995;91:192-200.

19.  Pozzoli M, Traversi E, Cioffi G, Stenner R, Sanarico M, Tavazzi L. Loading
       manipulations improve the prognostic value of Doppler evaluation of mitral flow in
       patients with chronic heart failure. Circulation 1997;95:1222-30.

20. Nagueh SF, Kopelen HA, Quinones MA. Assessment of left ventricular filling
      pressures by Doppler in the presence of atrial fibrillation.
      Circulation 1996;94:2138-45.

21.  Kuecherer HF, Kusumoto F, Muhiudeen IA, Cahalan MK, Schiller NB. Pulmonary
       venous flow patterns by transesophageal pulsed Doppler echocardiography: realtion
      to parameters of left ventricular systolic and diastolic function.
      Am Heart J 1991;122:1683-93.

22.  Pozzoli M, Capomolla S, Pinna G, Cobelli F, Tavazzi L. Doppler echocardiography
       reliably predicts pulmonary artery wedge pressure in patients
       with chronic heart failure with and without mitral regurgitation.
       J Am Coll Cardiol 1996;27:883-93.

23.  Nagueh SF, Middleton KJ, Kopelen HA, Zoghbi WA, Quinones MA. Doppler
       tissue imaging: a noninvasive technique for evalutaion of left ventricular 
       relaxation and  estimation of filling pressures. J Am Coll Cardiol 1997;30:1527-33.

24.  Garcia MJ, Ares MA, Asher C, Rodriguez L, VanderVoort P, Thomas J. An index of
       early left ventricular filling that combined with pulsed Doppler peak E velocity may
       estimate capillary wedge pressure. J Am Coll Cardiol 1997;29:448-54.

25. Ommen SR, Nishimura RA, Appleton CP. Clinical utility of Doppler echocardiography
       and tissue Doppler imaging in the estimation of left ventricular filling pressures. A
       comparative simultaneous Doppler catheterization study.
       Circulation 2000;102:1788-94.

26.  Sohn DW, Song JM, Zo JH, Chai IH, Kim HS, Chun HG, Kim HC. Mitral annulus
       velocity in the evaluation of left ventricular diastolic function in atrial fibrillation.
       J Am Soc Echocardiogr 1999;12:927-31.

27.  Nagueh SF, Mikati I, Kopelen HA, Middleton KJ, Quinones MA, Zoghbi WA. Doppler
       estimation of left ventricular filling pressure in sinus tachycardia.
       A new application of tissue doppler imaging. Circulation 1998;98:1644-50.

28.  Chirillo F, Brunazzi MC, Barbiero M, Giavarina D, Pasqualini M, Francheschini-
       Grisolia E, Cotogni A, Cavarzerani A, Rigatelli G, Stritoni P, Longhini C. Estimating
       mean pulmonary wedge pressure in patients with chronic mitral fibrillation from
       transthorcic Doppler indexes of mitral and pulmonary venous flow velocity.
       J Am Coll Cardiol 1997;30:19-26.

29.   Diastology

Gambia Silver

Gambia Silver Professor of Cardiology, Head, Department of Cardiology, Thoraxcentre University Hospital Rotterdam, The Netherlands. Correspondence to: Professor J.R.T.C. Roelandt, Thoraxcentre University Hospital Rotterdam, Dijkzigt P.O. Box 2040/Dr. Molewaterplein 40 3000 CA/3015 GD ROTTERDAM The Netherlands Fax: +31.10.436.2995 Email: roelandt@card.azr.nl