Theresa A. Tacy
Ultrasonography is a valuable and widely available technique for evaluating the patient with suspected heart disease. Standard 2-dimensional and Doppler echocardiography play a major role in defining cardiac anatomy, assessing ventricular function, and detecting abnormal flow patterns associated with cardiac disease. In many instances, the diagnostic accuracy of echocardiography is equivalent to that of cardiac catheterization,1 and has dramatically decreased the need for diagnostic catheterization preoperatively. In addition to transthoracic echocardiography, other modalities such as fetal, transesophageal, and stress echocardiography are widely used. It is the goal of this chapter to acquaint the reader with the basic concepts, as well as the indications for each type of echocardio-graphic examination.
Ultrasound was first used during World War II, when the technique of sending sound waves through water and observing the returning echoes to identify submarines was widely used. After the war, medical investigators transferred this technology for use in medical diagnosis. The major application of diagnostic ultrasound has been imaging of tissue structures. Imaging modalities of ultrasound include M-mode echocardiography and 2-dimensional echocardiography.
M-mode echocardiography provides an “ice-pick” view of the heart by emitting a narrow ultrasound beam. Structures encountered by the beam are reflected back and displayed as a dot, and as the image scrolls through to display time, the motion of this site over the cardiac cycle is displayed (Fig. 495-1). The frame rate in M-mode echocardiography is 1000 to 4000 frames per second, yielding excellent time resolution. Thus, M-mode provides the most precise display of events that occur rapidly within the cardiac cycle, such as the opening and closing of valves, the motion of the ventricular and atrial walls, and the changes in size of the cavities during contraction and relaxation. However, owing to the narrow area of interrogation used in M-mode echocardiography, anatomic relationships are best left to other modalities.
If the dots are scrolled over time and multiple parallel beams are emitted, the beginnings of a 2-dimensional echocardiogram are noted. In today’s echocardiogram, emitted beams of sound are steered, usually over 30 times a second, over an arc of approximately 90, producing a fan-shaped or sector image. As the sector is made up of a series of lines of information, the ultrasound information represented on each one of the lines within this sector fan produces an image. The replay of this ultrasound image from the sector fan in the time that is almost instantaneous with its generation has led to the technique being called real-time imaging.
Two-dimensional echocardiography shows a tomographic slice of the heart. These slices allow the reviewer to assess the spatial and anatomic relationships in the heart. Because each view is limited to a slice through the heart, multiple planes of the heart are investigated during a complete study. On a transthoracic study, these views are obtained from multiple sites on the chest, including parasternal position, cardiac apex, subcostal position in the midline below the subxiphoid process, and suprasternal position from the suprasternal notch. Only with complete display of the cardiac images from all of these transducer positions can a reliable 3-dimensional concept of cardiac anatomy be achieved (Fig. 495-2).
FIGURE 495-1. This M-mode tracing from a normal subject is aligned properly for performing a fractional shortening, which is the percentage of change between the left ventricular end-diastolic dimension (LVEDD) and the left ventricular end-systolic dimension (LVESD). In addition, this view is useful for measuring septal and posterior wall thickness and for confirming abnormal septal wall motion.
Two-dimensional imaging is useful for anatomic information, but does not assess flow patterns or hemodynamics. To do so, the Doppler principle is applied to ultrasound technology. Johann Christian Doppler (1803–1853), an Austrian physicist, first demonstrated that the relative motion between a sound source and an observer determines the observed frequency of the sound wave. If the observer is moving relative to the sound source, the detected frequency will be different from the transmitted frequency. If the observer is moving toward the sound source, a higher frequency tone is heard. This phenomenon, induced by the relative motion of the sound source and the observer, is referred to as the Doppler effect, and the resulting change in frequency of sound is called the Doppler shift.
Blood flow patterns within the heart and great vessels are evaluated by measuring blood velocity and displaying this information in various formats. Ultrasonic Doppler blood velocity measurement is performed by both spectral and color flow Doppler. Spectral Doppler displays blood velocity vertically on a scale as a function of time (horizontal axis) (Fig. 495-3). If the direction of flow is toward the transducer, the flow velocity information is displayed above the baseline (as positive velocity values), and if moving away from the transducer, is displayed below the baseline. Spectral Doppler assessments of flow can be performed using 1 of 2 modes of ultrasound, either pulsed wave (PW) or continuous wave (CW) mode. Each of these has specific benefits and limitations that are complementary in application. In the PW mode, measurements can be performed within a small range and at a precise site determined by the operator. However, in the PW mode the maximum velocity that can be measured is limited. In pediatric cardiology, PW Doppler velocimetry is used to define the origin of flow disturbances within the cardiovascular system, such as flow events through valve orifices and the direction of blood flow. The CW mode has no limit to the maximum measurable velocity; however, the site of velocity measurement cannot be controlled, and blood flow at all depths along the axis of the Doppler beam is measured and displayed. Thus in practice, the operator usually uses PW Doppler for precise determination of the site, direction, and pattern of flow, and CW to measure higher velocities accurately.
Color Doppler flow-mapping techniques have become an important aspect of the echocardiographic examination as its use improves the diagnostic accuracy of the study. In color Doppler echocardiography, information on blood flow velocity is computed, color encoded, and simultaneously superimposed on the 2-dimensional echocardiographic image. Flow traveling toward the transducer has red hues, ranging from a deep red at low velocities to a yellow hue at higher velocities, whereas blood traveling away from the transducer is assigned a deep blue hue at lower velocities and a light blue at higher velocities. Disturbed flow is displayed as a mosaic color that includes green. The information is updated several times a second, producing a real-time map of flow during the cardiac cycle. In this manner, color Doppler provides information regarding direction, velocity, and flow disturbances so that normal and abnormal flow patterns can be easily identified (Fig. 495-4).2 Color flow mapping techniques are qualitative at present and are not used for precise velocity measurements.3 Their main utility is in recognizing abnormal flow patterns and as an aid in aligning spectral Doppler beam for accurate flow-velocity measurement. Since M-mode echocardiography, 2-dimensional echocardiography, and spectral and color Doppler velocity measurements each assess a different aspect of heart anatomy, function, or flow information, the complete echocardiogram comprises all of them.
Echocardiography not only describes anatomy and flow aberrations but also is used to measure cardiac dimensions and function and to estimate pressure drop and shunt magnitude. Chamber and vessel dimensions as measured by either M-mode or 2-dimensional images are compared with normal data. This information aids in the assessment of the effects of a shunt lesion, such as assessing left ventricular dilation in a patient with aortic insufficiency. The dimension in question is related to normal data, often by calculating a z score. A z score is the number of standard deviations above or below the average indexed size of the structure in question on the basis of normative data.4 If zscore data are not readily available, normal plots of ventricular dimensions; valve sizes; or great vessels size related to patient body surface area, patient weight, or age and height exist for the pediatric population and are useful for comparison.5-8
Left ventricular function is commonly assessed during a routine examination by M-mode measurement of the left ventricular shortening fraction, which is the percentage of change of the left ventricular chamber diameter during a cardiac cycle. Normal values for left ventricular shortening fraction remain unchanged throughout childhood, and range from 28% to 44%.9-11 Left ventricular shortening fraction is easily measured, but it does not take into account the loading conditions of the left ventricle. Thus, it estimates global left ventricular function but not its contractility. For example, a patient with severe systemic hypertension may have a depressed left ventricular shortening fraction because of the increased afterload to which it is exposed, but the contractility of the ventricular muscle is normal. Shortening fraction is limited in its ability to assess global ventricular function as well. It assumes that left ventricular geometry is normal. If this is not true, as occurs when septal wall motion is flattened or paradoxical, shortening fraction will not reliably reflect left ventricular systolic function. In such instances, a measure of change in volume rather than change in dimension should be used. Ejection fraction is the most common measure used and is usually determined using Simpson’s rule. Normal values for ejection fraction range from 56% to 78%.14
The quantitative assessment of flow events in the heart and great vessels is performed using spectral Doppler velocity measurement of flow velocities. In many situations the blood velocity can be used to predict a pressure drop across the area of flow. Using the modified Bernoulli equation, the pressure drop is 4v2, where v is velocity in meters per second. Thus, a peak jet velocity of 3 m/s predicts a peak pressure drop of 4 × 32, or 36 mm Hg. High-velocity jets can be measured arising from a ventricular septal defect, patent ductus arteriosus, valvar stenosis, valve regurgitation, or from a pulmonary artery band or aortic coarctation (Fig. 495-3). This gradient can be used to predict a chamber or vessel pressure or may be used to assess stenosis severity.15-18
FIGURE 495-2. A: This frame, taken from a normal subject, is a parasternal long-axis view taken slicing the heart from a parasternal location at the fourth left interspace, subtending a sector with the right ventricle (RV) anteriorly. The sound beams then pass through the ventricular septum and the aorta (AO), the left ventricular cavity (LV), and the left ventricular posterior wall. Behind the ascending aortic root is the left atrium (LA). B: From the same normal subject in the parasternal short-axis view (P S Ax) from the third intercostal space. The sector subtends the right side of the heart as it winds around the aortic root (AO) in the center of image. The right atrium (RA) is separated from the right ventricle (RV) by the tricuspid valve. The pulmonary artery (PA) is separated from the right ventricle by the pulmonary valve. The cusps of the aortic valve can also be identified in this figure. C: A very high right parasternal view (in the right infraclavicular region) permits imaging of the main (MPA) and branch pulmonary arteries (RPA, LPA). The relationship of the RPA to the ascending aorta (AAO) is easily seen. The thymus is evident in this image. D: The apical 4 chamber views from a normal subject, demonstrating the scan from the anterior to the posterior aspect of the heart (from the apex to base). The right atrium (RA) is separated from the left atrium (LA) by the faint echo of the interatrial septum, and the left and right ventricles (LV and RV) can be seen separated from their respective atria by the tricuspid and mitral valves in the open position and from each other by the ventricular septum. E: When the transducer is angulated anteriorly from this position, the aortic root (AO), can be seen arising from the left ventricle. F: In the subcostal coronal plane with posterior angulation, the right atrium (RA), interatrial septum (IS), and pulmonary veins (PV) entering the left atrium (LA) are well seen. G: When the transducer is swept anteriorly, the plane of the beam goes through the left ventricular outflow tract. In this image, the tricuspid valve is seen between the right atrium (RA) and right ventricle (RV), as well as the aortic valve (AV). The right pulmonary artery (RPA), which courses underneath the ascending aorta, is also visible. H: With the ultrasound beam oriented in the sagittal plane of the body and angulated leftward, the left ventricle (LV) is seen in cross-section; the right ventricle (RV) and pulmonary artery (PA) can be seen wrapping around the left ventricle. I: With the ultrasound beam oriented in the sagittal plane of the body and angulated rightward, the superior vena cava is seen entering the right atrium (RA), the left atrium (LA) is seen, as is the right pulmonary artery (RPA) in cross-section passing behind the superior vena cava. J: The aortic arch from the suprasternal notch sagittal view in a normal infant. The scan comes from the suprasternal notch area and the sector subtends the innominate vein (IV) superiorly, as it crosses in front of the ascending aorta (AAO). The whole arch is seen from the ascending aorta to the descending aorta (DAO). The left carotid artery (LCA) can be seen arising from the aortic arch. The circular right pulmonary artery can be seen running under the arch (RPA).
Transesophageal echocardiography (TEE) has been used in children as small as 2.3 kg by inserting a phased-array ultrasound transducer mounted on the tip of a fiber-optic gastroscope as small as 6 mm in diameter. This device allows imaging of the heart from the esophagus and stomach when transthoracic images are of poor quality, when the chest is open, or when a transthoracic study is impractical because of extensive chest bandaging.19-22 The most common indication for TEE in children is during cardiac surgery. Intraoperative TEE has had a great impact by (1) providing additional information preoperatively that alters the planned surgical approach, (2) identifying postoperative structural abnormalities after the initial surgical procedure that require additional surgical intervention while the patient is still in the operating room, and (3) identifying postoperative abnormalities that alter medical management immediately after separation from cardiopulmonary bypass support.23 Transesophageal echocardiography also can play an integral role during cardiac catheterization during placement of devices and stents24,25 for monitoring of catheter placement during transeptal or ablation procedures.26,27
FIGURE 495-3. Continuous wave Doppler signal across an aortic coarctation taken from the suprasternal position, aiming inferiorly down the descending aorta. The velocity scale is in meters per second and identifies the scale of 0 to 6 m/s. The velocity of flow across the coarctation increases in systole, and because the direction of flow is away from the transducer, the signal is represented below the baseline. The signal increases to a velocity (V) of 2.6 m/s, which predicts a peak pressure gradient (PG) of 27 mm Hg.
FIGURE 495-4. This parasternal long-axis view with color Doppler demonstrates a small ventricular septal defect that permits shunting from the left ventricle (LV) to the right ventricle (RV) in systole. This flow is in practice color encoded with red hues, since in this view the direction of flow is toward the transducer.
Stress echocardiography is more common in the adult population, yet indications for stress echocardiography are increasing in children. Whereas standard echocardiographic studies are performed in a resting state, the purpose of stress echocardiography is that the patient can be evaluated during exercise or during an enhanced inotropic state (accomplished by dobutamine infusion). Stress echocardiography may demonstrate abnormal ventricular function in asymptomatic patients with normal resting state assessments.28 Stress echocardiography is useful in assessing exercise intolerance after surgical intervention,29and has been shown to be helpful in evaluating the patient with Kawasaki disease.30
Contrast echocardiography, using cross-sectional or M-mode modalities, is useful for detecting right-to-left intracardiac shunts as well as for validating cardiac structures. Microcavitations from agitated saline injected into a peripheral vessel or via catheter can be produced by injecting 1 to 2 mL of saline, dextrose, or other intravenous solution mixed with a small quantity of blood. As the capillary bed traps the microbubbles (less than 200 µm in diameter), when injected on the right side, they do not enter the systemic circulation, unless an intracardiac communication is present. The microbubbles will, however, follow the blood flow patterns so that when right-to-left shunts occur, even in degrees not detectable by other techniques such as angiography or oximetry, these shunts can be detected by the microbubbles in the left heart. The technique is particularly valuable for defining intra-atrial right-to-left shunting.
Transtelephonic echocardiography is useful in areas where a pediatric echocardiographer is geographically distant, yet immediate review of studies are needed to determine patient management. This method uses one of a variety of means to transmit echocardiographic images using either video transmission services or computer lines. The echocardiographic images are transferred to a regional tertiary care center, where a pediatric cardiologist can review the images while they are being performed or as soon as the study is completed. Often, real-time telephone connections between the sonographer performing the study and the pediatric cardiologist are established when the study begins, and the cardiologist assists the sonographer with verbal instructions. Thus far, transtelephonic echocardiography has been a reliable and cost-effective means for assessing patients in remote locations and has been especially effective in avoiding unnecessary neonatal transports.31
The costs of and indications for echocardiography remain misunderstood by many primary care physicians. In today’s environment, cost-effective means for diagnosis are a major concern to the primary care physician, yet patients are often referred for echocardiography instead of a cardiology consultation to evaluate murmurs. Bensky et al studied the causes of this practice and determined that the majority of referring physicians underestimated the cost of echocardiography drastically, believing it was cost-equivalent to cardiac consultation.32 Primary care physicians also believe a cardiologist will obtain an echocardiogram as part of their evaluation of a child with a murmur,32 despite ample published literature documenting cardiology consultation alone is often sufficiently diagnostic33-35 and is a much more cost-effective means36 for assessing a heart murmur. Despite these facts, the request for echocardiography amongst primary care physicians as a screening tool for heart disease is rising.37