An ideal modality used for noninvasive imaging of congenital heart defects should be able to delineate all aspects of the cardiac anatomy, including extracardiac vessels; evaluate physiological parameters such as measurement of blood flow, pressure gradients across cardiac valves or blood vessels, and ventricular function; be cost effective, portable, and noninvasive with least risk and discomfort; and include no exposure to ionizing radiation. No single modality satisfies all of these requirements. Chest radiographic films, the original imaging tool, provide only indirect evidence of intracardiac defects that manifest primarily with volume overload. They do not provide images of the defect itself. They are less informative on pressure overload lesions that do not result in chamber enlargement. Echocardiography has become the main noninvasive imaging modality since the 1980s, providing direct images of intracardiac and some extracardiac anatomy. In recent years, noninvasive radiologic techniques such as magnetic resonance imaging (MRI) and computed tomography (CT) have emerged as supplementary modalities in areas for which echocardiography studies are deficient.
In this chapter, two-dimensional (2D) and M-mode echocardiography are presented in some detail. This will be followed by a brief discussion of the two radiologic techniques with emphasis on how to choose an optimal radiologic technique for a given patient.
Echocardiography (echo) uses ultrasound beams reflected by cardiovascular structures to produce characteristic lines or shapes caused by normal or abnormal cardiac anatomy in one, two, or three dimensions, which are called M-mode, 2D, and three-dimensional (3D) echocardiography, respectively.
An echocardiographic study currently begins with real-time 2D echo, which produces high-resolution tomographic images of cardiac structure, their movement, and vascular structures leaving and entering the heart. The Doppler and color mapping study has added the ability to easily detect valve regurgitation and cardiac shunts during the echo examination. These tests combined provide reliable quantitative information such as ventricular function, pressure gradients across cardiac valves and blood vessels, and estimation of pressures in the great arteries and ventricles. Echo examination can be applied in calculation of cardiac output and the magnitude of cardiac shunts, although this is rarely used. Reliable noninvasive hemodynamic evaluation and confident delineation of cardiovascular structures by echo have dramatically reduced the necessity for cardiac catheterization. Increasingly, patients undergo valvular or congenital heart surgery on the basis of an echo diagnosis. Transesophageal echocardiography (TEE) has markedly improved resolution of echo images. Real-time 3D echocardiography provides improved accuracy of imaging the global perspective visualization of various cardiac anomalies, but this is not presented here because it is not widely used.
Discussion of instruments and techniques is beyond the scope of this book. Normal 2D echo images and M-mode measurements and their role in the diagnosis of common cardiac problems in pediatric patients are the focus of this presentation.
Two-dimensional echo examinations are performed by directing the plane of the transducer beam along a number of cross-sectional planes through the heart and great vessels. A routine 2D echocardiogram is obtained from four transducer locations: parasternal, apical, subcostal, and suprasternal notch (SSN) positions. From each transducer position, images of the long- and short-axis views are obtained by manually rotating and angulating the transducer. The parasternal and apical views usually are obtained with the patient in the left lateral decubitus position and the subcostal and suprasternal notch views with the patient in the supine position. Figures 5-1 through 5-9 illustrate some standard images of the heart and great vessels. Modified transducer positions and different angulations make many other views possible. Measurement of important cardiac structures can be made on the freeze frame of 2D echo studies. Normal values of the dimension of the cardiac chambers, great arteries, and various valve annulus are shown in several tables in Appendix D (see Tables D-1 through D-5).
FIGURE 5-1 Diagram of important two-dimensional echo views obtained from the parasternal long-axis transducer position. Standard long-axis view (A), right ventricular (RV) inflow view (B), and RV outflow view (C). AO, aorta; CS, coronary sinus; Desc. Ao, descending aorta; LA, left atrium; LV, left ventricle; PA, pulmonary artery; RA, right atrium; RAA, right atrial appendage.
FIGURE 5-2 Diagram of a family of parasternal short-axis views. Semilunar valves and great artery level (A), coronary arteries (B), mitral valve level (C), and papillary muscle level (D). AO, aorta; LA, left atrium; LCA, left coronary artery; LPA, left pulmonary artery; LV, left ventricle; MPA, main pulmonary artery; MV, mitral valve; PM, papillary muscle; RA, right atrium; RCA, right coronary artery; RPA, right pulmonary artery; RV, right ventricle; RVOT, right ventricular outflow tract.
FIGURE 5-3 Diagram of two-dimensional echo views obtained with the transducer at the apical position. A, The posterior plane view showing the coronary sinus. B, The standard apical four-chamber view. C, The apical “five-chamber” view is obtained with further anterior angulation of the transducer. AO, aorta; CS, coronary sinus; LA, left atrium; LV, left ventricle; RA, right atrium; RV, right ventricle.
FIGURE 5-4 Apical long-axis view. A, Apical “three-chamber” view. B, Apical “two-chamber” view. AO, aorta; LA, left atrium; LAA, left atrial appendage; LV, left ventricle; RV, right ventricle.
The Parasternal Views
For parasternal views, the transducer is applied to the left parasternal border in the second, third, or fourth space with the patient in the left lateral decubitus position.
Parasternal Long-Axis Views
The plane of sound is oriented along the major axis of the heart, usually from the patient’s left hip to the right shoulder. Three major views are recorded; standard long axis, long axis of the right ventricular (RV) inflow, and long axis of the RV outflow (see Fig. 5-1).
FIGURE 5-5 Diagram of subcostal long-axis view. A, Coronary sinus view posteriorly. B, Standard subcostal four-chamber view. C, View showing the left ventricular outflow tract and the proximal aorta. D, View showing the right ventricular outflow tract (RVOT) and the proximal main pulmonary artery. AO, aorta; CS, coronary sinus; LA, left atrium; LV, left ventricle; PA, pulmonary artery; RA, right atrium; RV, right ventricle, SVS, superior vena cava.
FIGURE 5-6 Subcostal short-axis (sagittal) view. A, Entry of venae cavae with drainage of the azygous vein. B, View showing the RV, RVOT, and pulmonary artery. C, Short-axis view of the ventricles. Azg. V, azygous vein; LA, left atrium; LV, left ventricle; MV, mitral valve; PA, pulmonary artery; RA, right atrium; RPA, right pulmonary vein; RV, right ventricle; SVC, superior vena cava; TV, tricuspid valve.
FIGURE 5-7 Abdominal views. Left, abdominal short-axis view. Right, abdominal long-axis view. A, Inferior vena cava (IVC) view; B, Abdominal descending aorta (AO) view. CA, celiac axis; HV, hepatic vein; RA, right atrium; SMA, superior mesenteric artery.
FIGURE 5-8 Diagram of suprasternal notch two-dimensional echo views. A, Long-axis view. B, Short-axis view. AO, aorta; Asc. Ao, ascending aorta; Desc. Ao, descending aorta; Inn. A, innominate artery; Inn. V, innominate vein; LA, left atrium; LCA, left carotid artery; LSA, left subclavian artery; MPA, main pulmonary artery; PA, pulmonary artery; RPA, right pulmonary artery; SVC, superior vena cava.
1. The Standard Long-Axis View
is the most basic view that shows the left atrium (LA), mitral valve, and left ventricular (LV) inflow and outflow tracts (see Fig. 5-1, A). This view is important in evaluating abnormalities in or near the mitral valve, LA, LV, left ventricular outflow tract (LVOT), aortic valve, aortic root, ascending aorta, and ventricular septum. In the normal heart, the anterior leaflet of the mitral valve is continuous with the posterior wall of the aorta (i.e., aortic–mitral continuity). The trabecular septum (apical ward) and infracristal outlet septum (near the aortic valve) constitute the interventricular septum in this view. Therefore, ventricular septal defects (VSDs) of tetralogy of Fallot (TOF) and persistent truncus arteriosus are best shown in this view. Detailed discussion of localizing other types of VSDs is presented in Chapter 12. Pericardial effusion is readily imaged in this view. This is the view to evaluate mitral valve prolapse (MVP). Frequently, the coronary sinus can be seen as a small circle in the atrioventricular (AV) groove (see Fig. 5-1, A). An enlarged coronary sinus may be seen with left superior vena cava (LSVC), TAPVR to coronary sinus, coronary AV fistula, and rarely with elevated right atrial (RA) pressure.
FIGURE 5-9 Diagram of subclavicular views. A, Right subclavicular view. B, Left subclavicular view. AO, aorta; IVC, inferior vena cava; LPA, left pulmonary artery; MPA, main pulmonary artery; RA, right atrium; RPA, right pulmonary artery; RV, right ventricle; SVC, superior vena cava.
2. In the RV Inflow View
a long-axis view of the RV and RA is obtained. In this view, abnormalities in the tricuspid valve (tricuspid regurgitation [TR], prolapse) and inflow portion of the RV are evaluated (Fig. 5-1, B). The ventricular septum in this view consists of the inlet muscular septum (near the AV valve) and trabecular septum (apical ward). The right atrial appendage (RAA) can also be seen in this view. This view is good for recording the velocity of TR (to estimated RV systolic pressure).
3. In the RV Outflow View
the RV outflow tract (RVOT), pulmonary valve, and proximal main pulmonary artery (PA) are visualized (Fig. 5-1, C). The supracristal infundibular (outlet) septum is seen in this view.
Parasternal Short-Axis Views
By rotating the transducer used for the long-axis views clockwise, one obtains a family of important short-axis views (see Fig. 5-2). This projection provides cross-sectional images of the heart and the great arteries at different levels. The parasternal short-axis views are important in the evaluation of the aortic valve (i.e., bicuspid or tricuspid), pulmonary valve, PA and its branches, RVOT, coronary arteries (e.g., absence, aneurysm, stenosis), LA, LV, ventricular septum, AV valves, LV, and right side of the heart.
1. The Aortic Valve
The aortic valve is seen in the center of the image with the RVOT anterior to the aortic valve and the main PA to the right of the aorta (“circle and sausage” view) (see Fig. 5-2, A). The right, left, and noncoronary cusps of the aortic valve are best seen from this view, having the appearance of the letter “Y” during diastole. Stenosis and regurgitation of the pulmonary valve is best examined in this plane. Stenosis of the PA branches can be evaluated by Doppler interrogation and color-flow mapping and Doppler interrogation of the ductal shunt is obtained in this plane. Color-flow studies show the membranous VSDs just distal to the tricuspid valve (at the 10 o’clock direction) and both the infracristal and supracristal outlet VSDs (at the 12 to 2 o’clock direction) anterior to the aortic valve near the pulmonary valve.
2. Coronary Arteries
With a slight manipulation of the transducer from the above plane, the ostia and the proximal portions of the coronary arteries are visualized. The right coronary artery (RCA) arises from the anterior coronary cusp near the tricuspid valve, which should be confirmed to connect to the aorta; there are some venous structures that run in front of the aorta (cardiac vein) but do not connect to the aorta. The left main coronary artery arises in the left coronary cusp near the main PA. Its bifurcation into the left anterior descending and circumflex coronary artery can usually be seen clearly. The proximal coronary arteries can also be seen in other long-axis views.
3. Mitral Valve
The mitral valve is seen as a “fish mouth” during diastole. This view is good for measuring the mitral valve area in patients with mitral stenosis and it is the best view to identify a cleft mitral valve (see Fig. 5-2, C).
4. Papillary Muscles
Two papillary muscles are normally seen at the 4 o’clock (anterolateral) and 8 o’clock (posteromedial) directions. The trabecular septum is seen at this level. This view is good in the assessment of apical portion of the LV such as hypertrophic cardiomyopathy, noncompaction of the apex and apical mass (see Fig. 5-2, D).
The Apical Views
For apical views, the transducer is positioned over the cardiac apex with the patient in the left lateral decubitus position.
Apical Four-Chamber View
The plane of sound is oriented in a nearly coronal body plane and is tilted from posterior to anterior to obtain a family of apical four-chamber views (see Fig. 5-3). This view displays all four chambers of the heart, the ventricular and atrial septa, and the crux of the heart. This is the best view to image the LV apex, where an apical VSD is commonly seen.
1. Coronary Sinus
In the most posterior plane, the coronary sinus is seen to drain into the right atrium (see Fig. 5-3, A). The ventricular septum seen in this view is the posterior trabecular septum.
2. The Apical Four-Chamber View
(see Fig. 5-3, B) evaluates the atrial and ventricular septa and size and contractility of atrial and ventricular chambers, AV valves, and some pulmonary veins and identifies the anatomic RV and LV and detecting pericardial effusion. Normally, the tricuspid valve insertion to the septum is more apicalward than the mitral valve (5–10 mm in older children and adults), with a small portion of the septum (called the AV septum) separating the two AV valves. A defect in this portion of the septum may result in an LV–RA shunt. In Ebstein’s anomaly, the insertion of the septal tricuspid valve is displaced more apically. The inlet ventricular septum (where an endocardial cushion defect occurs) is well imaged in this view just under the AV valves. VSDs in the entire length of the trabecular septum are well imaged, including apical VSD. The membranous septum is not imaged in this view. The anatomic characteristics of each ventricle are also noted, with the heavily trabeculated RV showing the moderator band. Abnormal chordal attachment of the AV valve (straddling) and overriding of the valve are also noted in this view. The relative size of the ventricles is examined in this view.
3. Apical Five-Chamber View
Further anterior angulation of the transducer demonstrates the so-called apical five-chamber view. This view shows the LVOT, aortic valve, subaortic area, and proximal ascending aorta. In this view, color-flow imaging allows qualitative assessment of aortic regurgitation. The membranous VSD is visualized just under the aortic valve, and the infracristal outlet VSD is also imaged in this plane.
Apical Long-Axis Views
The apical long-axis view (or apical three-chamber view) shows structures similar to those shown in the parasternal long-axis view (see Fig. 5-4, A). In the apical two-chamber view, the LA, mitral valve, and LV are imaged. The left atrial appendage can also be imaged (see Fig. 5-4, B). The view of the LV apex provides diagnostic clues for cardiomyopathy, apical thrombus, and aneurysm.
The Subcostal Views
Subcostal long axis (coronal) and short-axis (sagittal) views are obtained from the subxyphoid transducer position, with the patient in the supine position.
Subcostal Long-Axis (Coronal) Views
are obtained by tilting the coronal plane of sound from posterior to anterior (see Fig. 5-5). The coronary sinus is seen posteriorly draining into the RA , similar to that shown for the apical view (see Fig. 5-5, A). Anterior angulation of the transducer shows the atrial and ventricular septa. This is the best view for evaluating the atrial septum, including atrial septal defect (see Fig. 5-5, B). Further anterior angulation of the transducer shows the LVOT, aortic valve, and ascending aorta (see Fig. 5-5, C). The parts of the ventricular septum visualized in this view (apical ward) are membranous, subaortic outlet, and trabecular septa. The junction of the superior vena cava (SVC) and the RA is seen to the right of the ascending aorta (see Fig. 5-5, C). Further anterior angulation shows the entire RV including the inlet, trabecular and infundibular portions, pulmonary valve, and main PA (see Fig. 5-5, D). The ventricular septum seen in this view includes the (apical ward) supracristal outlet, infracristal outlet, and anterior trabecular and posterior trabecular septa.
Subcostal Short-Axis (or Sagittal) Views
(see Fig. 5-6) are obtained by rotating the long-axis transducer 90 degrees to the sagittal plane.
To the right of the patient, both the superior and inferior venae cavae are seen to enter the right atrium (see Fig. 5-6, A). A small azygous vein can be seen to enter the SVC, and the right PA can also be seen on end beneath this vein (see Fig. 5-6, A).
A leftward angulation of the transducer shows the RVOT, pulmonary valve, PA, and the tricuspid valve on end (see Fig. 5-6, B). This view is orthogonal to the standard subcostal four-chamber view, and both views combined are good for evaluation of the size of a VSD.
Additional leftward angulation of the transducer will show the mitral valve (not shown) and papillary muscle (see Fig. 5-6, C), similar to those seen in the parasternal short-axis views.
Subcostal Views of the Abdomen
Abdominal short- and long-axis views (Fig. 5-7) are obtained from the subxyphoid transducer position, with the patients in supine position.
1. Abdominal Short-Axis View
is obtained by placing the transducer in a transverse body plane (see Fig. 5-7, left). It demonstrates the descending aorta on the left and the inferior vena cava (IVC) on the right of the spine as two round structures. The aorta should pulsate. Both hemidiaphragms, which move symmetrically with respiration, are imaged. Asymmetric or paradoxical movement of the diaphragm is seen with paralysis of the hemidiaphrgam.
2. Abdominal Long-Axis View
is obtained by placing the transducer in a sagittal body plane. The IVC is imaged to the right (see Fig. 5-7, right, A), and the descending aorta is imaged to the left of the spine (Fig. 5-7, right, B). The IVC collects the hepatic vein before draining into the right atrium. The eustachian valve may be seen at the junction of the IVC and the RA. The failure of the IVC to join the RA indicates interruption of the IVC (with azygous continuation, which is frequently seen with polysplenia syndrome). Major branches of the descending aorta, celiac artery, and superior mesenteric artery are easily visualized. A pulsed-wave Doppler examination of the abdominal aorta in this view is important to identify coarctation by demonstrating persistent diastolic flow and delayed upstroke of systolic flow.
The Suprasternal Views
The transducer is positioned in the suprasternal notch to obtain suprasternal long-axis (see Fig. 5-8, A) and short-axis (see Fig. 5-8, B) views, which are important in the evaluation of anomalies in the ascending and descending aortas (e.g., coarctation of the aorta), aortic arch (e.g., interruption), size of the PAs, and anomalies of systemic veins and pulmonary veins. In infants, the transducer can be sometimes placed in a high right subclavicular position.
The Suprasternal Long-Axis View
(see Fig. 5-8, A) is obtained by 45-degree clockwise rotation from the sagittal plane in the suprasternal notch to visualize the entire (left) aortic arch. Failure to visualize the aortic arch in this manner may suggest the presence of a right aortic arch. Three arteries arising from the aortic arch (the innominate, left carotid, and left subclavian arteries in that order) are seen. The innominate vein is seen in cross-section in front of the ascending aorta and the right PA behind the ascending aorta. Manipulation of the transducer farther posteriorly and leftward will image the isthmus and upper descending aorta, a very important segment to study for the coarctation of the aorta.
The Suprasternal Short-Axis View
(Fig. 5-8, B) is obtained by rotating the ultrasound plane parallel to the sternum. Superior to the circular transverse aorta, the innominate vein is seen to connect to the (right) superior vena cava, which runs vertically to the right of the aorta. The right PA is visualized in its length under the circular aorta. Beneath the RPA is the LA. With a slight posterior angulation of the transducer, four pulmonary veins are seen to enter the LA.
The Subclavicular Views
The Right Subclavicular View
(see Fig. 5-9, A) is obtained in the right second intercostal space in a sagittal projection. This view is useful in the assessment of the SVC and right atrial junction as well as the ascending aorta. The right upper pulmonary vein and the azygous vein can also be examined in this view.
The Left Subclavicular View
(see Fig. 5-9, B) is useful for examination of the branch PAs. The transducer is positioned in a transverse plane in the second left intercostal space and a little tilted inferiorly. The main PA is seen left of the ascending aorta (circle), and it bifurcates into the right and left PA branches.
Quantitative Values Derived from Two-Dimensional Echocardiography
1. Dimensions of Cardiovascular Structures
Several tables of normal values of cardiovascular structures that were measured from still frames of 2D echocardiography are shown in Appendix D. These tables are frequently useful in the practice of pediatric cardiology. They include M-mode measurements of the LV (Table D-1); stand-alone M-mode measurements of the RV, aorta, and LA (Table D-2); aortic root and aorta (Table D-3); pulmonary valve and PAs (Table D-4); and AV valves (Table D-5). The normal dimension of the coronary arteries is shown in Table D-6.
2. Left Ventricular Mass
LV mass is an indication of LVH. Although the thickness of LV walls (interventricular septum and LV posterior wall) identifies people with increased LV mass, LV mass has become a valuable marker of end-organ damage in patients with systemic hypertension. LV mass can be estimated from M-mode echocardiography or 2D echocardiography measurements.
a. M-mode echo method. LV mass is usually derived from 2D-guided M-mode echo measurement, with assumptions that the LV is spherical in shape and the thickness of the wall measured at the basal area of the LV is representative of the entire LV. It is assumed that a smaller sphere formed by the endocardium is inside a larger sphere formed by the pericardium, and therefore, the difference in the volume of the two spheres would be the volume of LV muscle. Although less accurate than the 2D echo method, the M-mode method is simpler to obtain and thus more popular than the 2D method. There is controversy as to indexing the LV mass for body size, having been variably indexed by body weight, height, body surface area, or 2.7 power of height. M-mode–derived LV mass indexed to 2.7 power of height is popular and normal data by that method are presented in Appendix D (Table D-7).
b. 2D echo method. In the 2D-measured echo technique for calculating LV mass, the LV is assumed to be a bullet shape rather than a sphere, and this technique has been shown to be more accurate than the M-mode method. The LV volume can be estimated from the short- and long-axis views in systole or diastole. The LV area in the short-axis view is calculated by the biplane Simpson’s method. The formula for such a volume would be 5/6 the LV area of the short axis multiplied by the length of the ventricle (obtained from the long axis):
The LV volume so calculated is then converted to mass by multiplying it by the specific gravity of muscle (usually taken as 1.05). This cumbersome method is less popular and not routinely performed in most laboratories. Normal values of 2D echo-derived LV mass indexed to the body surface area are presented in Appendix D (Table D-8).
M-mode echocardiography, which graphically displays a one-dimensional slice of cardiac structure varying over time, was one of the earliest tools of the echocardiography. Currently, an M-mode echo is obtained as part of 2D tomographic images. M-mode echo is used primarily for the measurement of the dimension (wall thickness and chamber size) and LV function (fractional shortening, wall thickening). It is also useful for assessing the motion of cardiac valves (mitral valve prolapse, mitral stenosis, pulmonary hypertension) and movement of cardiac wall and septa (in RV volume overload).
Figure 5-10 shows examples of M-mode measurements of the dimension of the RV, LV, LA, and aorta and LV wall thickness during systole and diastole. Line 1 passes through the aorta and LA, where the dimensions of these structures are measured. Line 2 traverses the mitral valve. Line 3 goes through the main body of the RV and LV. Along line 3, the dimensions of the RV and LV and the thickness of the interventricular septum and posterior LV wall are measured during systole and diastole. Pericardial effusion is best detected at this level.
FIGURE 5-10 Examples of M-mode measurement of cardiac dimensions. The dimension of the aorta (AO) and left atrium (LA) is measured along line 1. Line 2 passes through the mitral valve. Measurement of chamber dimensions and wall thickness of right ventricle (RV) and left ventricle (LV) is made along line 3. The following measurements are shown in this figure: (a), RV dimension; (b), LV diastolic dimension; (c), interventricular septal thickness; (d), LV posterior wall thickness; (e), LA dimension; (f), aortic dimension; (g), LV systolic dimension. AMV, anterior mitral valve; ECG, electrocardiogram; PMV, posterior mitral valve; T, transducer.
Normal M-Mode Echo Values
Two frequent applications of M-mode echo in clinical practice are measurements of dimension of cardiac structures and LV systolic function.
Cardiac Chamber Dimensions
Most dimensions are measured during diastole, coincident with the onset of the QRS complex; the LA dimension and LV systolic dimension are exceptions (see Fig. 5-10). The dimensions of the cardiac chambers and the aorta increase with increasing age and thus normal values are expressed as function of growth (see Appendix D).
Left Ventricular Systolic Function
LV systolic function is evaluated by the fractional shortening (or shortening fraction) or ejection fraction. The ejection fraction is a derivative of the fractional shortening and offers no advantages over the fractional shortening. Serial determinations of these measurements are important in the management of conditions in which LV function may change (e.g., in patients with chronic or acute myocardial disease).
Fractional shortening (or shortening fraction) is derived by the following:
where FS is fractional shortening, Dd is end-diastolic dimension of the LV, and Ds is end-systolic dimension of the LV. This is a reliable and reproducible index of LV function, provided there is no regional wall-motion abnormality and there is concentric contractility of the LV. If the interventricular septal motion is flat or paradoxical, the shortening fraction will not accurately reflect ventricular ejection.
Mean normal value of FS is 36%, with 95% prediction limits of 28% to 44%. FS is decreased in a poorly compensated LV regardless of cause (e.g., pressure overload, volume overload, primary myocardial disorders, doxorubicin cardiotoxicity). It is increased in volume-overloaded ventricle (e.g., VSD, patent ductus arteriosus, aortic regurgitation, mitral regurgitation [MR]) and pressure overload lesions (e.g., moderately severe aortic valve stenosis, hypertrophic obstructive cardiomyopathy).
Ejection fraction relates to the change in volume of the LV with cardiac contraction. It is obtained by the following formula:
where EF is ejection fraction and Dd and Ds are the end-diastolic and end-systolic dimensions, respectively, of the LV. The volume of the LV is derived from a single measurement of the dimension of the minor axis of the LV. In the preceding formula, the minor axis is assumed to be half of the major axis of the LV; this assumption is incorrect in children. Normal mean ejection fraction is 66% with ranges of 56% to 78%.
Doppler echocardiography combines the study of cardiac structure and blood flow profiles. The Doppler effect is a change in the observed frequency of sound that results from motion of the source or target. When the moving object or column of blood moves toward the ultrasonic transducer, the frequency of the reflected sound wave increases (i.e., a positive Doppler shift). Conversely, when blood moves away from the transducer, the frequency decreases (i.e., a negative Doppler shift). Doppler ultrasound equipment detects a frequency shift and determines the direction and velocity of red blood cell flow with respect to the ultrasound beam. By convention, velocities of red blood cells moving toward the transducer are displayed above a zero baseline; those moving away from the transducer are displayed below the baseline.
The two commonly used Doppler techniques are continuous wave and pulsed wave. The pulsed wave emits a short burst of ultrasound, and the Doppler echo receiver “listens” for returning information. The continuous wave emits a constant ultrasound beam with one crystal, and another crystal continuously receives returning information. Both techniques have their advantages and disadvantages. Pulsed-wave Doppler can control the site at which the Doppler signals are sampled, but the maximal detectable velocity is limited, making it unusable for quantification of severe obstruction. In contrast, continuous-wave Doppler can measure extremely high velocities (e.g., for the estimation of severe stenosis), but it cannot localize the site of the sampling; rather, it picks up the signal anywhere along the Doppler beam. When these two techniques are used in combination, clinical application expands.
NORMAL DOPPLER VELOCITIES IN CHILDREN AND ADULTS
Children, Mean (Ranges) (m/sec)
Adults, Mean (Ranges) (m/sec)
From Hatle L, Angelsen B: Doppler Ultrasound in Cardiology, 2nd ed. Philadelphia, Lea & Febiger, 1985.
The Doppler echo technique is useful in the study of the following: (1) detecting the presence and direction of cardiac shunts; (2) studying stenosis or regurgitation of cardiac valves; (3) assessing stenosis of blood vessels; (4) assessing the hemodynamic severity of a lesion, including pressures in various compartments of the cardiovascular system; (5) estimating the cardiac output or blood flow; and (6) assessing diastolic function of the ventricle (see later discussion). The Doppler echo is usually used with color flow mapping (see below) to enhance the technique’s usefulness.
Normal Doppler velocities in children and adults are shown in Table 5-1. Normal Doppler velocity is less than 1 m/sec for the PA; it may be up to 1.8 m/sec for the ascending and descending aortas. Mitral and tricuspid inflow velocities are usually less than 1 m/sec.
Measurement of Pressure Gradients
The simplified Bernoulli equation can be used to estimate the pressure gradient across a stenotic lesion, regurgitant lesion, or shunt lesion. One may use one of the following equations.
where (P1 – P2) is the pressure difference across an obstruction, V1 is the velocity (m/sec) proximal to the obstruction, and V2 is the velocity (m/sec) distal to the obstruction in the first equation. When V1 is less than 1 m/sec, it can be ignored, as in the second equation. However, when V1 is more than 1.5 m/sec, it should be incorporated in to the equation to obtain a more accurate estimation of pressure gradients. This is important in the study of the ascending and descending aortas, where flow velocities are often more than 1.5 m/sec. Ignoring V1 may significantly overestimate the pressure gradient in patients with aortic stenosis or coarctation of the aorta.
To obtain the most accurate prediction of the peak pressure gradient, the Doppler beam should be aligned parallel to the jet flow, the peak velocity of the jet should be recorded from several different transducer positions, and the highest velocity should be taken. An example of a Doppler study in a patient with moderate pulmonary stenosis is shown in Figure 5-11. The peak instantaneous pressure gradient calculated from the Bernoulli equation is not the same as the peak-to-peak pressure gradient measured during cardiac catheterization. The peak instantaneous pressure gradient is usually larger than the peak-to-peak pressure gradient. The difference between the two is more noticeable in patients with mild to moderate obstruction and less apparent in patients with severe obstruction.
Prediction of Intracardiac or Intravascular Pressures
The Doppler echo allows estimation of pressures in the RV, PA, and LV using the flow velocity of certain valvular or shunt jets. Estimation of PA pressure is particularly important in pediatric patients.
FIGURE 5-11 Doppler echocardiographic study in a child with a moderate pulmonary valve stenosis. The Doppler cursor is placed in the main pulmonary artery near the pulmonary valve in the parasternal short-axis view. The maximum forward flow velocity (negative flow) is 3.91 m/sec (with an estimated pressure gradient of 61 mm Hg). There is a small regurgitant (positive) flow seen during diastole.
The following are some examples of such applications:
1. RV (or PA) systolic pressure (SP) can be estimated from the velocity of the tricuspid regurgitation (TR) jet, if present, by the following equation:
where V is the TR jet velocity.
For example, if the TR velocity is 2.5 m/sec, the instantaneous pressure gradient is 4 × (2.5)2 = 4 × 6.25 = 25 mm Hg. Using an assumed RA pressure of 10 mm Hg, the RV systolic pressure (or PA systolic pressure in the absence of PS) is 35 mm Hg. (Assuming that the RA pressure of 10 mm Hg is too high for patients who do not have severe TR or RV failure. An RA mean pressure of 5 mm Hg is more reasonable in most children and adolescents).
2. RV (or PA) systolic pressure can also be estimated from the velocity of the VSD jet by the following equation:
where V is the VSD jet.
For example, if the VSD jet flow velocity is 3 m/sec, the instantaneous pressure drop between the LV and RV is 4 × 32 = 36 mm Hg. That is, the RV systolic pressure is 36 mm Hg lower than the LV systolic pressure. If LV systolic pressure is assumed to be 90 mm Hg, the estimated RV systolic pressure is 54 mm Hg (90 – 36 = 54). In the absence of PS, the PA systolic pressure will be about 54 mm Hg. Note that the systolic pressure obtained in the arm cannot be assumed to be the same as LV systolic pressure; the arm pressure is usually 5 to 10 mm Hg higher than the LV systolic pressure (see peripheral amplification in Chapter 2).
3. LVSP can be estimated from the velocity of flow through the aortic valve (V) by the following equation:
where V is the aortic flow velocity. The same precaution applies as above in that the arm systolic pressure is slightly higher than the LV systolic pressure.
Measurement of Cardiac Output or Blood Flow
Both systemic blood flow and pulmonary blood flow can be calculated by multiplying the mean velocity of flow and the cross-sectional area as shown in the following equation:
where V is the mean velocity (cm/sec) obtained either by using a computer program or by manually integrating the area under the curve. CSA is the cross-sectional area of flow (cm2) measured or computed from the two-dimensional echo. Usually, the PA flow velocity and diameter are used to calculate pulmonary blood flow; the mean velocity and diameter of the ascending aorta are used to calculate systemic blood flow, or cardiac output.
Signs of diastolic dysfunction may precede those of systolic dysfunction. Mitral inflow velocities obtained in the apical four-chamber view can evaluate LV diastolic function. There are two flow waves in the AV valves; the E wave and the A wave (Fig. 5-12). The E wave occurs during the early diastolic filling phase, and the A wave occurs during atrial contraction. The E wave is taller than the A wave in normal children and adults. However, the A wave may be normally taller than the E wave in the first 3 weeks of life.
The following simple measurements are useful in evaluating diastolic function of the ventricle (see Fig. 5-12).
1. The E and A wave peak velocities and the ratio of the two (E/A ratio)
2. Deceleration time (DT): The interval from the early peak velocity to the zero intercept of the extrapolated deceleration slope
3. Atrial filling fraction: The integral of the A velocity divided by the integral of the total mitral inflow velocities
4. Isovolumic relaxation time (IVRT): The interval between the end of the LV outflow velocity and the onset of mitral inflow; this is easily obtained by pulsed-wave Doppler with the cursor placed in the LV outflow near the anterior leaflet of the mitral valve and is measured from the end of the LV ejection to the onset of the mitral inflow
In normal children, the mitral Doppler indexes are as follows. The average peak E velocity is 0.6 m/sec, the average peak A velocity is 0.3 m/sec, and the average E:A velocity ratio is 2.0. Detailed normal values for other measurements are shown in Figure 18-6.
Abnormalities in the diastolic function are easy to find, but they are usually nonspecific and do not provide independent diagnostic information. In addition, they can be affected by loading conditions (i.e., increase or decrease in preload), heart rate, and the presence of atrial arrhythmias. Two well-known patterns of abnormal diastolic function are a decreased relaxation pattern and a “restrictive” pattern (see Fig. 18-6). The decreased relaxation pattern is seen with hypertrophic and dilated cardiomyopathies, LVH of various causes, ischemic heart disease, other forms of myocardial disease, reduced preload (e.g., dehydration), and increased afterload (e.g., during infusion of arterial vasoconstrictors). The “restrictive” pattern is usually seen in restrictive cardiomyopathy but is also seen with increased preload (e.g., seen with MR) and a variety of heart diseases with heart failure.
FIGURE 5-12 Selected parameters of diastolic function. (See text for discussion). A, second peak velocity; DT, deceleration time; E, early peak velocity; IVRT, isovolumic relaxation time; LV, left ventricle.
A color-coded Doppler study provides images of the direction and disturbances of blood flow superimposed on the echo structural image. Although systematic Doppler interrogation can obtain similar information, this technique is more accurate and time saving. In general, red is used to indicate flow toward the transducer, and blue is used to indicate flow away from the transducer. Color may not appear when the direction of flow is perpendicular to the ultrasound beam. The turbulent flow is color coded as either green or yellow.
Injection of indocyanine green, dextrose in water, saline, or the patient’s blood into a peripheral or central vein produces microcavitations and creates a cloud of echoes on the echocardiogram. Structures of interest are visualized or recorded by 2D echo at the time of the injection. This technique has successfully detected an intracardiac shunt, validated structures, and identified flow patterns within the heart. For example, an injection of any liquid into an intravenous line may confirm the presence of a right-to-left shunt at the atrial or ventricular level. To a large extent, this technique has been replaced by color-flow mapping and Doppler studies.
Other Echocardiographic Techniques
Improvement in image resolution makes visualization of the fetal cardiovascular structure possible, thereby permitting in utero diagnosis of cardiovascular anomalies. To obtain a complete examination, the transducer is placed at various positions on the maternal abdominal wall. Doppler examination and color mapping are performed at the same time. The optimal timing for performance of a comprehensive transabdominal fetal echo is 18 to 22 weeks of gestation. Images can be more difficult to obtain after 30 weeks of gestation because the ratio of fetal body mass to amniotic fluid increases (Rychik et al, 2004).
Accurate diagnosis of congenital heart defect (CHD) via fetal echo provides many benefits, including improved surgical outcome for infants with CHDs. In addition, fetal echo continues to teach physicians more about human fetal cardiac physiology. It also enables physicians to study the effects of cardiovascular abnormalities and abnormal cardiac rhythms in utero and then assess the need for therapeutic intervention.
Indications for fetal echo are expanding, and examples are shown in Box 5-1. Increased nuchal translucency present on obstetric ultrasonography at 10 to 13 weeks of gestation has been associated with an increased risk of CHD even in the absence of a chromosomal anomaly. Infants conceived via intracytoplasmic sperm injection and in vitro fertilization have up to a threefold increased prevalence of CHDs.
BOX 5-1 Examples of Indications for Fetal Echocardiography
Family history of CHD
Metabolic disorders (e.g., diabetes, PKU)
Exposure to teratogens
Exposure to prostaglandin synthetase inhibitors (e.g., ibuprofen, salicylates, indomethacin)
Autoimmune disease (e.g., SLE, Sjögren’s syndrome)
Familial inherited disorders (Ellis van Creveld, Marfan, Noonan’s disease)
In vitro fertilization
Abnormal obstetric ultrasound screen
Increased first trimester nuchal translucency
Multiple gestation and suspicion of twin–twin transfusion syndrome
CHD, congenital heart disease; PKU, phenylketonuria; SLE, systemic lupus erythematosus
From Rychik J, Ayers N, Cuneo B, et al: American Society of Echocardiography Guidelines and standards for performance of the fetal echocardiography. J Am Soc Echocardiogr 17:803-810, 2004.
By placing a 2D or multiplane transducer at the end of a flexible endoscope, it is possible to obtain high-quality 2D images by way of the esophagus. Color-flow mapping and Doppler examination are usually incorporated in this approach.
If satisfactory images of the heart or blood vessels are not possible from the usual transducer position on the surface of the patient’s chest (e.g., in patients with obesity or chronic obstructive pulmonary disease), physicians may use TEE. This approach is especially helpful in assessing thrombus in native or prosthetic valves, endocarditis vegetations, thrombi in the left atrial chamber and appendage, and aortic dissections. TEE is often used for patients who are undergoing cardiac surgery. TEE can monitor LV function throughout the surgical procedure, as well as assess cardiac morphology and function before and after surgical repair of valvular or congenital heart defects. TEE requires general anesthesia or sedation and the presence of an anesthesiologist because a lack of patient cooperation can result in serious complications. Use of this technique in pediatric patients is somewhat limited to intraoperative use and in some obese adolescents and adolescents with complicated heart defects for whom the risk of general anesthesia or sedation is worth taking for the expected benefit. Schematic drawings of biplane images of TEE are shown in Appendix D (Fig. D-1).
To provide an intravascular echo, the ultrasonic transducer is placed in a small catheter so that vessels can be imaged by means of the lumen. These devices can evaluate atherosclerotic arteries in adults and coronary artery stenosis or aneurysm in children with Kawasaki’s disease.
Radiologic Techniques: Magnetic Resonance Imaging and Computed Tomography
Although the conventional echo study remains the mainstay of noninvasive evaluation of cardiac patients, it has limitations in complete delineation of cardiac anomalies. In addition to being operator dependent, echocardiography may not provide optimal quality of images of cardiovascular structure because of postoperative scars, chest wall deformities, overlying lung tissue, large body size in adolescents, and obesity. In particular, extracardiac structures such as the PAs, pulmonary veins, and aortic arch cannot always be adequately imaged by echo study because of acoustic window limitations. As to the coronary arteries, only the proximal portion can be adequately imaged by echo studies. Although TEE may improve the quality of images, it is not only a more invasive technique requiring deep sedation or anesthesia but also it does not always provide the needed information for patient care. Cardiac catheterization with a higher complication rate may become necessary to make complete anatomic and physiologic evaluation of the patient with complex cardiac pathology. Fortunately, however, some noninvasive radiologic techniques, such as MRI and cardiac CT, are now available that can obviate the need for cardiac catheterization.
Both MRI and CT can provide images of cardiovascular structures and other intrathoracic structures that are not usually imaged by echo studies. However, one of the radiologic techniques may be better than the other in its capability and its practicality. For example, in young pediatric patients, CT may have a broader appeal because of its short duration of study, but exposure to ionizing radiation and contrast administration requirements are important drawbacks. On the other hand, MRI has capability of offering quantitative ventricular function, myocardial viability, and tissue characterization, which CT lacks. However, MRI requires a longer scanning time and cannot be used in patients with implanted metallic objects, such as pacemakers or intracardiac cardioverter-defibrillators (ICDs).
BOX 5-2 Advantages and Disadvantages of Magnetic Resonance Imaging and Computed Tomography
CT, computed tomography; ICD, implantable cardioverter-defibrillator; IV, intravenous; LV, left ventricular; MRI, magnetic resonance imaging; RV, right ventricular.
Physicians and cardiologists often face the situation to make a decision as to which noninvasive technique best serves a patient. This section provides some insights into the advantages and disadvantages of noninvasive radiologic imaging techniques in pediatric patient care. The decision as to which test to order depends on the physician’s knowledge, the availability of the technique, and the expertise of cardiovascular radiology consultants in the area. Advantages and disadvantages of MRI and CT are summarized in Box 5-2. Additional discussions on the topic follow.
Magnetic Resonance Imaging
Selected clinical applications and advantages of cardiac MRI in pediatric cardiology are listed below.
1. MRI offers excellent images of intracardiac morphology that is better than CT images.
2. Ventricular function. Cardiac MRI has become the gold standard for the quantification of ventricular function for both the LV and RV because it does not rely on geometric assumptions as used in echo study (and cardiac catheterization). It provides accurate information on stroke volume, systolic function, ventricular mass, and regurgitation fraction. A frequent application of RV function study relates to older children who had surgical repair of TOF with resulting pulmonary regurgitation of significant magnitude. MRI studies play a major role in determining the timing of pulmonary valve placement in these patients.
3. Equal quality images of extracardiac structures such as the PAs and their distal branches, the aorta, the pulmonary veins, and the systemic veins as CT does
4. Almost equal quality imaging of trachea and mainstem bronchus
5. Almost equal quality imaging of small vessels such as the coronary artery (in patients who had Kawasaki’s disease or the arterial switch operation for TGA)
6. Capability of tissue characterization and myocardial viability, which CT lacks
7. Regional wall motion abnormalities of the LV and RV
The following are considered disadvantages of cardiac MRI:
1. Patients must remain still in the scanner bore for 45 to 60 minutes to minimize motion artifact during image acquisition. Accordingly, many infants and small children require sedation or anesthesia. Most children age 8 years and older can cooperate sufficiently for a good-quality MRI.
2. Implanted metallic objects are of particular concern in the MRI environment because they could potentially undergo undesirable movement if the magnetic fields are sufficiently strong. MRI can be used to image patients with implanted intravascular coils, stents, and occluding devices when the implants are believed to be immobile. However, the wires and clips may cause localized image artifacts. Surgical clips and sternotomy wires are typically only weakly ferromagnetic.
3. The presence of cardiac pacemaker or ICD is considered a contraindication to MRI.
4. The presence of an intracranial, intraocular, or intracochlear metallic object is also considered a contraindication to MRI.
Computed tomography provides excellent quality imaging of extracardiac vasculature and may provide this information much more quickly (5–10 minutes) than MRI. Preferential applications of CT may include the following:
1. Almost equal quality definition of intracardiac anatomy as MRI has.
2. Better quality imaging and faster results of coronary artery anomalies than MRI images, a major application of CT angiography
a) Anomalous origin of left coronary artery
b) Possible RV-dependent coronary circulation in patients with pulmonary atresia with intact ventricular septum
c) Postoperative arterial switch operation for D-TGA
d) Coronary involvement after Kawasaki’s disease.
3. Comprehensive evaluation of the pulmonary arteries and veins, preoperatively and postoperatively.
4. Evaluation of the patency of vascular shunts and conduits, including patients after such operations as B-T shunt, bidirectional Glenn shunt, or Fontan procedure.
5. Evaluating status after balloon dilatation with or without stent placement; for example, in the aorta (for coarctation) and pulmonary arteries. CT has a clear advantage over MRI in metallic stent evaluation.
6. Imaging vascular rings and pulmonary artery sling (as an alternative to MRI or cardiac catheterization).
7. Evaluating aortic collaterals prevalent in certain cyanotic CHD, such as TOF.
8. Clear imaging of the airways is a unique advantage of CT over other imaging modalities, including vascular or nonvascular narrowing, anomalies, and dynamic obstruction (airway malacia, air trapping).
9. Capability of studying patients with pacemakers and ICD in whom cardiac MRI is prohibited and in patients with metallic implants that create unmanageable artifacts on cardiac MRI (e.g., steel coils).
The following are considered disadvantages of cardiac CT:
1. There is significant radiation exposure associated with CT. Radiation exposure during cardiac CT is estimated to be slightly greater than exposure during a pediatric cardiac catheterization. In children with complex CHDs, the risk may be compounded by their previous or subsequent exposure to ionizing radiation during cardiac catheterization or with serial CT studies.
2. Iodinated contrast materials are more toxic to the kidneys. Adverse reactions with nonionic, iodine-based contrast agents (e.g., hypotension, bradycardia, tachycardia, and even angina) occur at three times the rate of reactions to MRI gadolinium-based compounds.
Choice of Imaging Modalities
The age of the patient may be an important factor in choosing an imaging modality in studying pediatric cardiac patients. Prakash and colleagues (2010) have made the following suggestions.
For Infants and Children Younger than 8 Years, echo studies will provide accurate diagnosis of even complex CHDs in most cases. Sedation may be required in some infants and children. Therefore, the need for using MRI or CT study arises only rarely. MRI can be used to answer most of the questions regarding ventricular size and function and extracardiac vasculature. However, when the question is primarily the extracardiac vasculature, CT can also be used. Its use should be balanced against the risk of ionizing radiation exposure.
For Adolescents and Adults, echo remains the primary diagnostic modality. However, MRI plays an increasing role, especially for the evaluation of the extracardiac thoracic vasculature, ventricular volume and function, and flow measurement. MRI is usually preferred over CT or cardiac catheterization in this age group because it avoids exposure to ionizing radiation and can provide a wealth of functional information. CT is used in patients with contraindications to MRI, such as those with cardiac pacemaker or ICD, and those in whom concomitant evaluation for coronary disease is necessary