INTRODUCTION
The cardiac cycle is continuous. The filling of the ventricle (diastole) is followed by ventricular contraction (systole) to provide an adequate cardiac output during both rest and exercise to meet the body’s metabolic demands. Systole and diastole affect each other in an intimate manner to accomplish this goal. The normal elastic recoil after left ventricular (LV) contraction aids early filling of the ventricle, with the late diastolic atrial contraction ensuring that the myocardial sarcomeres are adequately stretched to optimize contractile force. Exercise tests the health of this integrated system by shortening the time for filling and myocardial perfusion, and a normally functioning cardiac electrical system is also needed for optimal performance.
The “new” epidemiology of LV diastolic dysfunction has been discussed in Chapter 6 of this volume. Diastolic heart failure is now recognized as a major national health problem, especially in the elderly, who have a high incidence of LV hypertrophy (LVH). These patients present with symptomatic heart failure despite a normal LV ejection fraction (LVEF) and have morbidity and mortality that is nearly equal to that of patients with reduced systolic function. Diastolic heart failure patients are also at risk for first-onset atrial fibrillation and a higher incidence of stroke. Although LVEF is normal in diastolic heart failure, ventricular contractile mechanics have been altered in a way that lengthens the isovolumic contraction and relaxation period so that the period for diastolic filling becomes shorter and may be inadequate. In both systolic and diastolic heart failure, the degree of diastolic dysfunction is a powerful predictor of prognosis.
Despite this new appreciation of the importance of both systole and diastole in maintaining normal cardiovascular physiology, the role of LV diastolic function in health and disease is incompletely understood and underappreciated by many primary care physicians and cardiologists. As discussed in Chapter 2 , diastole is a complex phenomenon with many determinants that are difficult to study individually and several phases that encompass both the relaxation and the filling of the ventricle. Physical examination, echocardiography (ECG), chest radiographs, and laboratory studies are unreliable in diagnosing diastolic heart failure in most individuals, and invasive measurements of LV diastolic properties and pressures are impractical in clinical practice. Therefore, at present, assessing the type and degree of LV diastolic dysfunction relies on evaluating the pattern of LV filling. Although this can be accomplished by radionuclide and computed tomography (CT) angiographic and magnetic resonance imaging (MRI) techniques, cardiac ultrasound is currently the method of choice because of its noninvasive and portable nature. Evaluating LV filling by two-dimensional anatomic findings, Doppler flow, and tissue Doppler imaging (TDI) have emerged as powerful clinical techniques to predict adverse cardiovascular events such as new-onset atrial fibrillation and heart failure, as well as mortality regardless of LVEF. The purpose of this chapter will be to describe the various LV filling patterns encountered in clinical practice, and what these patterns and their measurable variables reveal about LV diastolic function. In cases of possible ambiguity, ancillary variables such as left atrial (LA) volume, pulmonary venous (PV) flow velocity, and TDI that help in interpreting mitral flow velocity patterns will be discussed.
LEFT VENTRICULAR FILLING: A HISTORICAL PERSPECTIVE
In 100 bc , Galen proposed that the heart is filled by dilation of the right ventricle. Sixteen hundred years later, in 1628, William Harvey described the heart as a central pump in a circulatory system with both arteries and veins. Logically, most cases of heart failure were ascribed to damage or weakening of the heart muscle and a decrease in pumping function. Diastole was considered simply the interval in which the cardiac chambers filled passively between each pumping cycle and therefore was largely ignored.
Gradually, evidence emerged that abnormalities of ventricular filling could cause symptoms and that diastolic and systolic functions were interrelated. About 1915, Wiggers described the phases of the entire cardiac cycle, and the Frank-Starling mechanism was described, whereby LV end diastolic volume helps regulate forward stroke volume on a beat-to-beat basis. In 1930, Katz recognized the role of LV relaxation in the filling of the ventricle, suggesting that the normal heart acts like a “suction” pump, a concept proven 60 years later. A limitation in cardiac filling, not pumping function, with resultant reduced cardiac output was recognized as the cardinal feature of constrictive pericarditis, further emphasizing that disorders of filling could also cause cardiac symptoms and disease.
In the 1960s, the study of ventricular biomechanics accelerated. Although most research continued to focus on LV contractile function, cardiac diseases with thickened and noncompliant ventricles with good pumping function, such as restrictive and hypertrophic cardiomyopathies, were described. Angiographic differences in LV filling patterns between normals and patients with various heart diseases were reported even when the LVEFs were similar, and these phenomena were subsequently studied by M-mode echocardiography. At the same time, the importance of LV systolic function in determining diastolic restoring forces and rate of LV relaxation was appreciated. Our current view that systole and diastole are an intertwined continuum, with each part affecting all others, gradually began to emerge.
In the mid-1980s, echocardiographic studies showed that 20% to 40% of elderly patients with heart failure had a normal LVEF and, therefore, presumably “isolated” diastolic dysfunction. However, difficulties in quantitating individual LV diastolic properties caused the clinical study of LV diastolic dysfunction to proceed slowly. LV filling patterns were analyzed by digitized M-mode, angiographic, and radionuclide methods. In 1982, pulsed wave (PW) Doppler study of mitral flow velocities to study LV filling was first reported. Doppler ultrasound, a noninvasive technique that could determine dynamic changes in LV filling after interventions and over time, revolutionized the study of LV diastolic function. Age-related changes in LV filling were quickly described, followed by the description of three basic abnormal mitral flow filling patterns that were correlated to LV diastolic variables and filling pressures. The three abnormal mitral filling patterns were found to be independent of disease type and to have clinical significance and prognostic value regardless of cardiac disease type, suggesting that different pathology altered the same basic diastolic properties. PV flow velocity helped assess the filling of the left atrium, and variables were found to aid the interpretation of LV filling patterns and pressures. Using the two flow velocities, a “natural history” of LV filling in normals and with disease patterns was described. Manipulation of preload and afterload demonstrated the dynamic nature of LV filling patterns in response to changes in loading conditions, and these changes also had prognostic significance in patients with cardiac disease. Additional Doppler methods followed, such as TDI of mitral annular motion (MAM) and the rate of color Doppler mitral inflow propagation (Vp), and model-based image processing continued to improve diagnostic accuracy and advance the new field of “diastology.” These continue to enhance our basic knowledge regarding clinical observations.
Although all echo-Doppler variables have limitations in interpreting diastolic function, the aggregate sum of the two-dimensional echo and Doppler findings provides us a powerful and practical way to noninvasively assess LV diastolic function and to objectively follow serial changes after medical intervention or with disease progression. These methods are powerful prognostic tools in asymptomatic as well as symptomatic patients with various diseases, like dilated and restrictive cardiomyopathies. As a result, the clinical syndrome of diastolic heart failure is now more readily recognized, and studies on improved diagnosis and the best treatment strategies for such patients are under way.
RELATION BETWEEN DIASTOLIC PROPERTIES AND LEFT VENTRICULAR FILLING PATTERNS
The numerous factors that affect LV diastolic properties and the filling of the left ventricle are described in Chapter 5 of this volume. Although the interaction of these factors is complex, their sum reflects the diastolic transmitral pressure gradient, which ultimately determines mitral inflow and the LV filling pattern, as shown in Figure 10-1 . Positive gradients in diastole result in flow across the mitral valve, while negative gradients decelerate or stop flow.
The different LA and LV pressures that are determined by LV diastolic properties and filling are shown in Figure 10-2 . Understanding that LV end diastolic pressure (LVEDP) can be elevated before mean LA pressure is increased is essential to understanding and interpreting diastolic function. Two key diastolic properties, the rate of LV relaxation (diastolic pressure fall) and of LV compliance (throughout all of diastole), are especially important in understanding pressures and LV filling. Normal LV contraction and relaxation are vigorous and rapid, resulting in diastolic elastic recoil that augments early diastolic filling through a suction effect (chamber volume increasing while pressure is initially still decreasing). This promotes an early mitral valve opening and helps maximize the diastolic filling period and myocardial perfusion. Rapid ejection and a predominance of early diastolic filling leaves a period of diastasis before atrial contraction as a reserve that can help maintain adequate filling when exercise shortens cardiac cycle length.
The first hemodynamic abnormality seen in nearly all cardiac diseases is a slower rate of LV relaxation. This is commonly associated with hypertension or LVH. Both systole and diastole are affected, even if the LVEF remains normal. In systole, the LV isovolumic contraction and ejection times become prolonged. In diastole, the slower fall in LV pressure causes the mitral valve to open later and the early diastolic transmitral pressure gradient and proportion of filling to decline. The diastasis period often disappears, and a greater proportion of filling at atrial contraction is needed to reach an optimal end diastolic volume. These changes can markedly alter the length of the diastolic filling period and resting LV filling pattern, as shown in the two individuals in Figure 10-3 . One is normal and one has hypertensive heart disease with LVH. Both have identical heart rates and LVEFs. At this stage, if LV compliance is reduced, the abnormal increase in pressure is seen initially only in late diastole at the time of atrial contraction, and mean LA pressure remains normal, so that patients are usually asymptomatic. A comparison of the durations of mitral and PV flow velocities at atrial contraction indicates this earliest abnormality of ventricular compliance and hemodynamics.
With more advanced disease, LV relaxation remains abnormal, but a decrease in LV compliance occurs throughout diastole, which increases mean LA pressure and size and begins to result in symptomatic heart failure. The increased LA pressure will oppose the effect of a slower rate of LV relaxation, causing an earlier mitral valve opening and a higher transmitral pressure gradient, so that the LV filling pattern appears more “normal.” However, in this case the increase in early diastolic LV filling is caused by increased driving pressure rather than suction created by normal ventricular elastic recoil. Patients with this “pseudonormal” LV filling begin to have heart failure symptoms and show moderate functional limitation.
With a severe decrease in LV compliance, the marked elevation in LA pressure causes early diastolic filling to predominate, while the left atrium fails and provides little additional late diastolic filling. This third and most abnormal LV filling pattern is termed “restrictive.” Patients with restrictive filling patterns also have impaired LV relaxation, but the severe decrease in LV compliance results in a marked elevation of LA pressure, which promotes a rapid initial flow of blood into the ventricle in early diastole. However, the increased proportion of early filling has an abrupt, premature termination due to a rapid increase in LV pressure with only minimal filling occurring at atrial contraction. The reduced atrial contribution indicates that LA systolic failure is present due to the chronic pressure overload. Patients with this pattern are markedly symptomatic, demonstrate a severe functional impairment, and have a poor prognosis.
Because both the rate of LV relaxation and LA pressures may have many different values that are not necessarily related, many different transmitral gradient profiles and LV filling patterns are possible. As a result, a similar-appearing LV filling pattern may occur with different combinations of these two key diastolic properties, with clinically significant differences in LV filling pressures. In these instances, the degree of abnormality of LV relaxation and compliance is indicated by anatomic abnormalities such as reduced LVEF, LVH, LA enlargement or altered PV flow, and annular TDI velocities.
EXERCISE AND LEFT VENTRICULAR DIASTOLIC FUNCTION
An increase in heart rate, as with exercise, is the heart’s diastolic “stress test.” As heart rate increases, diastolic filling time shortens. In normal individuals, several adaptations help keep early and late diastolic filling separated, a coordination of the electrical and mechanical systems that provide for increased filling and cardiac output without an elevation of diastolic pressures. Most importantly the P-R interval shortens. At the same time, faster heart rates trigger the Treppe (or “staircase”) effect, which increases LV contractility and rate of relaxation. LVEF increases and LV end systolic volume decreases. The overall effect is an earlier mitral valve opening, increased elastic recoil, and a larger early diastolic transmitral pressure gradient. LV contractility is also increased by sympathetic tone, while systemic vascular resistance falls with muscular vasodilation.
Impaired relaxation may reduce the cardiac output achieved by reducing the diastolic filling time below that needed for optimal LV filling and myocardial perfusion. A premature fusion of early and late diastolic filling often occurs (see Fig. 10-3 ) with an inability to increase LV end diastolic volume. LV and LA filling and PV flow increase with atrial contraction of an incompletely relaxed left ventricle. Patients are affected by reduced aerobic capacity and may complain of abnormal exertional dyspnea. Those with pseudonormal and restrictive filling patterns have a significant decrease in exercise capacity. However, in these individuals, it is the increase in mean LA pressure and pulmonary congestion due to reduced LV compliance that limits exercise rather than a blunted increase in LV end diastolic volume.
DOPPLER MITRAL FLOW VELOCITY PATTERNS
Because of their noninvasive nature and ease of use, echo-Doppler techniques have become the accepted clinical standard for assessing LV diastolic function. LV filling is assessed with both continuous wave (CW) and PW Doppler techniques. Figure 10-4 shows mitral flow velocity obtained with the PW Doppler technique and the variables that are measured. These include: LV isovolumic relaxation time (IVRT); peak mitral flow velocity in early diastole (E wave) and at atrial contraction (A wave); the mitral deceleration time (DT); the E-wave velocity just before atrial contraction (E at A), also sometimes referred to as preA velocity; and the duration of mitral A-wave velocity (Adur). An E-at-A wave velocity of more than 20 cm/sec results in a peak A-wave velocity that is larger than it would have been at a slower heart rate, when mitral flow velocity has time to drop to a lower level before atrial contraction. In these cases, the E/A wave ratio may be reduced compared with values obtained at a slower heart rate, so that more reliance on other echo-Doppler variables is needed when interpreting the “fused” LV filling pattern.
Changes in Mitral Flow Velocity Patterns with Aging and Disease States
Elastic recoil and rapid LV relaxation in adolescents and young adults result in a predominance of early diastolic filling (E wave) with much less filling (10%–15%) due to atrial contraction. With normal aging, LVEF changes little, but LV relaxation slows in most individuals. The slower relaxation appears to be due largely to a gradual increase in systolic blood pressure and LV mass (hypertrophy). The result is reduced LV filling in early diastole and increased filling at atrial contraction. In most individuals, the peak E- and A-wave velocities become approximately equal during the sixth and seventh decade of life, with atrial filling contributing up to 35% to 40% of LV diastolic stroke volume. In individuals who maintain lower blood pressures and have no increase in LV mass, the age-related changes of decreasing E/A ratio in asymptomatic “normal” patients used in most reference studies are less pronounced, and normal E-wave predominance can occasionally be seen into the seventh decade of life. In these individuals, normal two-dimensional findings, LA size, and annular TDI variables confirm that diastolic function is normal. Normal age-related values for mitral variables are listed in Table 10-1 .
| AGE (yr) | n | IVRT (msec) | E (mmHg) | A (mmHg) | MDT (msec) | ADUR (msec) |
|---|---|---|---|---|---|---|
| 2–20 | 46 | 50 ± 9 | 88 ± 14 | 49 ± 12 | 142 ± 19 | 113 ± 17 |
| 21–40 | 51 | 67 ± 8 | 75 ± 13 | 51 ± 11 | 166 ± 14 | 127 ± 13 |
| 41–60 | 33 | 74 ± 7 | 71 ± 13 | 57 ± 13 | 181 ± 19 | 133 ± 13 |
| >60 | 10 | 87 ± 7 | 71 ± 11 | 75 ± 12 | 200 ± 29 | 138 ± 19 |
The three basic abnormalities of LV filling patterns were discussed previously and are shown in Figure 10-5 , where the arrows indicate that abnormal mitral filling patterns are a dynamic continuum and may worsen or become more normal with changes in loading conditions. Common usage describes the three abnormal filling patterns as “impaired,” “pseudonormal,” and “restrictive” relaxation. The diastolic property of impaired LV relaxation is present in all patterns, the difference being that with pseudonormal and restrictive filling, progressively reduced LV compliance raises mean LA pressure to levels that mask their effects on the transmitral pressure gradient and filling.
The changes in LV filling with normal aging and with cardiac disease states can be combined into a “natural history of LV filling,” which is shown together with their corresponding PV flow velocities in Figure 10-6 . Although theoretical when proposed in 1992, the progression of abnormalities in LV filling patterns with disease states (from impaired relaxation to pseudonormal to restrictive), together with changes in LV relaxation and compliance, has been documented in experimental models of congestive heart failure and clinically observed in patients with restrictive cardiomyopathies. Many variations of LV filling patterns that do not exactly match the three “classical” abnormal patterns are common because of the multiple combinations of the rate of LV relaxation and compliance. However, the abnormal LV filling patterns remain specific to the alterations in diastolic properties rather than to the type of cardiac disease, with all three patterns, depending on disease stage, being seen in disorders as diverse as restrictive and dilated cardiomyopathies.
This “natural history” of LV filling explains how both young normal individuals and patients with severe disease and a restrictive filling pattern can have a high proportion of filling in early diastole and an audible S 3 gallop. It also shows that PV flow velocity has its own changes that occur with normal aging and in cardiac disease states (discussed below) and that these associated PV filling patterns are more distinctive than some similar-appearing normal and abnormal mitral flow velocity patterns.
Left Ventricular Filling and Changing Cardiac Loading Conditions
Simple maneuvers in the echo laboratory to reduce (Valsalva) or increase (leg raising) preload demonstrate that mitral flow velocity patterns are a dynamic continuum. Plotting the changes in E-wave velocity and mitral DT in individuals after altering loading conditions has even been proposed as a load-independent index of normal and abnormal diastolic filling.
During the strain phase of a Valsalva maneuver, preload (mean LA pressure) is reduced, and in normals peak mitral E-wave velocity decreases by at least 20% during maximum strain with a smaller decrease in peak A-wave velocity ( Fig. 10-7 ). Similarly, in patients with an impaired relaxation filling pattern, mean LA pressure is normal, so that with preload reduction, the whole diastolic transmitral pressure gradient decreases and both E- and A-wave velocities decrease. In patients with reduced systolic function and an impaired relaxation pattern, increasing preload by leg raising may result in no change in the filling pattern. However, if a pseudonormal mitral flow velocity pattern results, it indicates a higher cardiovascular morbidity than if their E/A ratio remains less than 1.
With pseudonormal mitral flow patterns, the Valsalva strain lowers the elevated LA pressure and “unmasks” the underlying impaired LV relaxation. A notable feature of this change is the increase in mitral A-wave velocity and duration as the left atrium ejects into a ventricle that has a lower pressure. Preload-sensitive patients with restrictive filling patterns will revert to a pseudonormal filling pattern. In individuals who perform an adequate Valsalva and yet remain restrictive, LV stiffness is markedly increased, even at the more normal filling pressures. These patients have a poor prognosis.
Grading the Degree of Left Ventricular Diastolic Dysfunction by Mitral Flow Velocity Alone
A simplified grading system for diastolic function based on the three abnormal mitral flow velocity patterns alone is shown in Figure 10-8 and has been found to be useful in many patients. Grades Ia and Ib represent an impaired relaxation filling pattern with normal mean LA pressure and LA size. The difference is whether all LV filling pressures are normal (Ia) or whether LV pressure increase with atrial contraction is abnormal and increases LVEDP (Ib). This is important because an increase in LVEDP is the first hemodynamic abnormality of diastolic dysfunction (see Fig. 10-2 ). Grade II indicates pseudonormal filling with increased mean LA pressure, and grade III is a restrictive LV filling pattern. Grade IV is restrictive filling that does not revert to pseudonormal with preload reduction, indicating the most advanced LV diastolic dysfunction and the worst prognosis.
Although grading LV diastolic dysfunction by mitral flow velocity pattern alone can be helpful, many patients (especially the elderly) may be misclassified. The rate of LV relaxation and LV compliance and filling pressures are a continuum, and similar LV filling patterns are possible with different combinations of these diastolic properties. A major area of misinterpretation concerns patients with markedly impaired LV relaxation where LA pressures are elevated with an increase in the E-wave velocity, yet the filling pattern appears to be impaired (E/A ratio <1) because of partial fusion of early and late diastole (see Fig. 10-3 ) or because LV relaxation is so abnormally slow that only marked increases in mean LA pressure will result in pseudonormal filling. Figure 10-9 shows an example of this latter phenomenon. In this case, three patients with hypertrophic cardiomyopathy have different combinations of the speed of LV relaxation and LA pressure, giving similar mitral E/A wave ratios of approximately 2, yet the mean LA pressure varies threefold because of the markedly different rates of LV relaxation. In these cases, ancillary data such as two-dimensional anatomic abnormalities, LVEF reduction, LVH, LA enlargement, or altered PV flow or annular TDI velocities are indicators that abnormal diastolic function and pressures are present. Also some mitral filling patterns are “atypical,” meaning that a biphasic mitral DT, a mid-diastolic filling “hump,” or some other unusual feature is present. These less common LV filling patterns do not fit well into a grading scheme for diastolic dysfunction that uses only the mitral flow velocity pattern and are best understood by evaluating the abnormal diastolic properties and physiology that the altered LV filling reflects.
Grading the Degree of Left Ventricular Diastolic Dysfunction in Epidemiologic Studies
Recent landmark studies show that many asymptomatic individuals with a normal LVEF have abnormal diastolic function that is a risk factor for future development of adverse cardiovascular events such as new-onset heart failure, a first episode of atrial fibrillation or stroke, and death. In an effort to improve predictive value beyond that of analyzing mitral flow velocity alone, the classification of the degree of LV diastolic abnormality in these studies included mitral inflow velocity variables (especially the E/A wave ratio), their response to the Valsalva maneuver, PV flow velocity, and TDI of the mitral annulus. Figure 10-10 shows the multiple echo-Doppler criteria for grading diastolic dysfunction in these epidemiologic studies and provides a preview of the ancillary variables that aid in this interpretation (discussed later in this chapter).
PULMONARY VENOUS FLOW VELOCITY
PV velocity, usually obtained with PW Doppler from the right upper pulmonary vein during transthoracic apical imaging, reflects the filling dynamics of the left atrium. With experience, high-quality PW Doppler transthoracic recordings can be obtained in approximately 85% to 90% of patients. Within a short time after mitral flow velocity patterns were correlated with hemodynamics, it became apparent that PV flow velocity could be of additional help in assessing LV filling patterns, especially impaired relaxation filling with increased LVEDP and pseudonormal LV filling. While there are now several additional echo-Doppler variables to help identify pseudonormal filling (LA size, color M-mode [CMM] inflow propagation velocity, mitral annular TDI), PV flow velocity analysis remains unique and indispensable to identifying an abnormal increase in LV pressure at atrial contraction.
The various components of PV flow velocity change with age, mitral flow velocity, and disease states and can be matched to their corresponding mitral Doppler patterns (see Fig. 10-6 ). The hemodynamic determinants of PV flow velocity have been studied in vivo and clinically and related to individual variables ( Fig. 10-11 ). These include peak forward flow velocity in early systole (PVs1), late systole (PVs2), and early diastole (PVd) and peak reverse flow velocity at atrial contraction (PVa) and its duration (PVa dur). A dip or “notch” between PVs1 and PVs2 in early LV systole represents the C wave seen in LA pressure recordings at the time of mitral valve closure. PVs1 blends with PVs2 velocity, and this “notch” is not distinctly seen in most (about 70%) transthoracic flow velocity recordings but is observed more often with a first-degree atrioventricular (AV) block and in virtually all transesophageal-echo (TEE) recordings in patients in sinus rhythm.
PVs1 occurs in early ventricular systole as a result of LA relaxation and a decrease in downstream LA pressure at a time when PV pressure is relatively constant. PVs2 peaks later in systole and is due to the increase in PV flow and pressure that results from systolic right ventricular (RV) stroke volume. Maximal PVs2 velocity, its time-velocity integral (TVI), and deceleration of flow reflect the pressure difference between PV and downstream LA pressure as blood fills both vascular systems. The descent of the AV rings with ventricular systole increases LA size and compliance while helping decrease the downstream LA pressure. PV diastolic flow velocity (PVd) initially follows early diastolic mitral flow velocity; but in mid-diastole, LV filling slows while PVd flow continues with ongoing LA enlargement and appendage filling. PV flow reversal due to atrial contraction (PVa) is determined by LA contractility and LV compliance, with more and longer reverse flow seen as LV compliance decreases.
Changes in PV flow velocity and their relation to LV compliance and LA pressures are shown in Figure 10-12 . In patients with impaired relaxation filling, decreased LV compliance, and normal mean LA pressure, early diastolic filling (mitral E wave) is reduced. In these cases, LA hypertrophy (LAH) provides a sufficient atrial “kick” to fill the left ventricle to its optimal end diastolic volume. An important benefit of LAH is that it confines the abnormal pressure rise in the atrium, ventricle, and pulmonary veins to the short period associated with atrial contraction so that mean LA pressure remains normal (see Fig. 10-2 ). PVs1 is increased so that the PV systolic-to-diastolic flow velocity ratio remains above 50%. If LV compliance decreases further, the atrium may begin to decompensate, enlarge, and have contractile dysfunction. A reduced systolic fraction of PV antegrade flow (<40%) indicates LA systolic dysfunction (decline in PVs1) and has a high specificity for a pseudonormal LV filling pattern and increased mean LA pressure (see Fig. 10-12 ). The PV A wave and its relation to mitral A-wave duration are of special importance and are discussed in detail below.
Relation of Mitral to Pulmonary Venous A-Wave Duration
As shown in Figure 10-13 , when both A-wave flow velocity durations are accurately recorded, this relation is an important indication of LV A-wave pressure increase and LV end diastolic pressure, even in the pediatric age group. Laboratories that do not routinely record PV flow velocity will miss the unique aspects of this derived variable. Unlike other Doppler variables, the relation is independent of age, meaning that mitral A-wave duration remains equal to or longer than PV A-wave duration throughout life in normal individuals. Secondly, the relation of the two A-wave durations is our only echo-Doppler variable that directly relates to the rise in LV pressure at atrial contraction, thereby separating patients with impaired relaxation who have normal filling pressures from those who have an increased LVEDP (>12 mmHg), the first hemodynamic abnormality seen with diastolic dysfunction.
Under normal circumstances when the left atrium contracts, the net volume and duration of flow should be greater forward into the left ventricle than backward into the pulmonary veins. If the PV A wave is increased in either velocity (>35 cm/sec) or duration (>30 msec longer than mitral A wave), LV A-wave pressure is increased and LVEDP is elevated (see Fig. 10-13 ). Even in cases where the atrium is enlarged, markedly hypokinetic, and failing, the PV A-wave duration of reverse flow continues to be more than 30 msec longer than the abbreviated mitral A-wave inflow duration. If the PV A-wave duration is difficult to measure due to a suboptimal recording, referencing the end of both A-wave flows to the QRS complex is helpful, as PV flow duration is abnormal if it exceeds that of the mitral A wave. A detailed guide to obtaining high-quality PW Doppler PV flow velocity recordings has been published.
The interpretation of mitral versus PV A-wave duration may not be reliable if the mitral velocity at the start of atrial contraction is greater than 20 cm/sec, or if atrial contraction occurs before PVd has reached the zero velocity baseline. In the first case, the peak mitral A-wave velocity, TVI, and A-wave duration are longer than normal to accommodate the increased atrial stroke volume that is present; and in the second case, the PV Adur is shorter because it starts above the conventional measuring point of the zero velocity baseline.
CLINICAL APPLICATIONS: INTERPRETATION OF INDIVIDUAL MITRAL FLOW VELOCITY VARIABLES
Studying the factors that influence individual mitral flow velocity variables is often useful in interpreting LV filling patterns in difficult or uncommon cases. It is also a powerful tool for deciding which diastolic property is most abnormal, which helps in the planning of possible clinical interventions in individual patients.
Left Ventricular Isovolumic Relaxation Time
LV IVRT is the interval from aortic valve closure to mitral valve opening and the start of mitral inflow (see Fig. 10-4 ). This interval can be a powerful tool for physicians in helping to evaluate diastolic dysfunction and filling pressures, especially when the mitral E wave is increased or atrial fibrillation is present. We suggest it be measured on all echo-Doppler studies.
In normal patients, IVRT varies with age, being shorter in the young, who have rapid LV relaxation that results in an earlier mitral valve opening, and then becoming lengthened as relaxation slows with age. Figure 10-14 compares the effect of changes in the rate of LV relaxation and LA pressure on the IVRT interval, timing of the mitral valve opening, and LV filling pattern using a normal subject and a patient with impaired LV relaxation.
The IVRT interval is most helpful when it is at its extremes, meaning farthest from expected norms, being either short (<60 msec) or long (>110 msec). It is less useful in-between these values. A normal IVRT for a middle-aged adult is approximately 80 msec. A short IVRT (<60 msec) indicates an early mitral valve opening; a long IVRT (>100 msec), a delayed LV relaxation and a late valve opening. In patients with impaired relaxation filling and normal pressures, a prolonged IVRT is an early indicator of LV diastolic dysfunction. If mean LA pressures remain normal, extremely slow LV relaxation can result in IVRT values that approach 200 msec. The higher filling pressure in pseudonormal filling patterns causes the mitral valve to open earlier, so the IVRT shortens. More normal values of 60 to 100 msec are usually seen, and the IVRT value is less useful. A short IVRT of 40 to 60 msec can be seen in young, healthy, normal individuals or in patients with very high mean LA pressure and restrictive filling. This clinical distinction is easily made by normal versus abnormal two-dimensional anatomic findings, especially of LA size and contractile function.
A common instance when IVRT duration can be very helpful is when heavy mitral annular calcification is present and there is a question of mild calcific mitral stenosis versus LV diastolic dysfunction. The mitral annular narrowing may increase E-wave velocity, the E/A wave velocity ratio, and LA size, making the assessment of LV diastolic dysfunction difficult. If the pressure half-time is normal or borderline but the IVRT is short (20 msec less than expected for age), LA pressure is likely elevated due to a noncompliant left ventricle. Conversely, if E-wave velocity is increased but the IVRT is normal, mild annular narrowing is likely the cause for the increased velocity. A markedly prolonged pressure half-time indicates calcific mitral stenosis, in which case a short IVRT would also indicate increased mean LA pressure, similar to the auscultatory findings of a short opening snap interval.
Left Ventricular Intracavitary Flow During Isovolumic Relaxation Time
LV IVRT flow is usually apically directed, from one part of the ventricle to another, occurring when both aortic and mitral valves are closed during IVRT. The importance of recognizing IVRT flow is twofold; it indicates that abnormal, dyssynchronous LV relaxation is present between apex and base and that its peak velocity should not be confused with the mitral E-wave velocity that immediately follows it. In the normal situation, no significant LV flow is detected during the IVRT period. When present, IVRT flow is most easily detected from an apical transducer position with CW or CMM Doppler because both techniques scan the long axis of the ventricle. PW Doppler can then be used for better velocity definition. When present, IVRT flow is usually 20 to 60 cm/sec but can occasionally be as high as 1 to 2 m/sec, as seen in Figure 10-15 .
