Echocardiography has revolutionized the diagnostic approach to patients with congenital heart disease. A comprehensive cardiovascular ultrasound imaging and hemodynamic evaluation is the initial diagnostic test used in the assessment of any congenital cardiac malformation. Since echocardiography became a part of clinical practice in the 1970s, the technology used for cardiac imaging has been in a nearly constant state of change. New techniques have been introduced at an increasingly rapid pace, especially since the 1990s. In this chapter, we will review the basic physical properties of ultrasound and the primary modalities used in clinical imaging. These discussions will provide an important foundation that will allow us to understand the more advanced methods of imaging and functional assessment, which will be covered in more detail later.
WHAT IS ULTRASOUND?
Diagnostic ultrasound generates images of internal organs by reflecting sound energy off the anatomic structures being studied. An ultrasound imaging system is designed to project sound waves into a patient and detect the reflected energy, then converting that energy into an image on a video screen. The types of sound waves used are given the name “ultrasound” because the frequencies involved are greater than the frequencies of sound that can be detected by the human ear. The average human ear can respond to frequencies between 20 and 20,000 Hz. Therefore, it stands to reason that ultrasound waves have frequencies greater than 20,000 Hz. In clinical practice, most imaging applications actually require frequencies in excess of 1 MHz. Current cardiac imaging systems have the ability to produce ultrasound beams varying between 2 and 12 MHz. A typical diagnostic ultrasound system consists of a central processing unit (CPU), the video image display screen, a hard drive for storage of the digital images, and a selection of transducers. The transducers both transmit and receive ultrasound energy.
■Sound wave: Series of cyclical compressions and rarefactions of the medium through which the wave travels (Fig. 1.1).
■Cycle: One alternation from peak compression through rarefaction and back to peak compression in a sound wave.
■Wavelength (λ): Distance from one peak (or trough) of the wave to the next (one complete cycle of sound).
■Velocity (v): Speed with which sound travels through a medium. The ultrasound propagation velocity in human tissue is 1540 m/s.
■Period (p): Time duration required to complete one cycle of sound.
■Amplitude (A): Magnitude of the sound wave, representing the maximum change from baseline to the peak of compression or a rarefaction in a cycle.
■Frequency (f): Number of sound cycles occurring in 1 second.
■Power: Rate at which energy is transferred from the sound wave to the medium. This is related to the square of the wave amplitude.
■Fundamental or carrier frequency (f0): Frequency of the transmitted sound wave.
■Harmonic frequency (fx): Sound waves that are exact multiples of the carrier frequency. The first harmonic frequency is twice the frequency of the carrier wave.
■Bandwidth: Range of frequencies that a piezoelectric crystal can produce and/or respond to.
Figure 1.1. Graphic depiction of a sound wave. The portions of the wave above the dashed baseline represent compression of the medium by the energy in the wave. Conversely, the portions of the wave shown below the baseline represent rarefaction. The portion of the wave that lies between one peak and the next, or one valley and the next, is referred to as the period. Wavelength is the distance covered by one period. Amplitude (A) refers to the maximum change from baseline caused by the wave (by either compression or rarefaction).
Diagnostic ultrasound imaging relies on the ability of high-frequency sound waves to propagate (travel) through the body and be partially reflected back toward the sound source by target tissues within the patient (Fig. 1.2). The imaging system generates the imaging beam by electrically exciting a number of piezoelectric crystals contained within a transducer. The imaging beam is then focused and projected into the patient. As the ultrasound beam travels through the patient, some of the energy will be scattered into the surrounding tissue (attenuation), and some will be reflected back toward the source by the structures in the beam’s path. These reflected waves will provide the information used to create images of the internal organs. This is the same imaging strategy used in sonar technology to detect objects below the water’s surface. The intensity (amplitude) of the reflected energy wave is proportional to the density of the reflecting tissue (see later). The reflected ultrasound energy induces vibrations in the transducer crystals and an electric current is created. This current is sensed by the CPU and converted into a video image.
The ultrasound image generation is based primarily on the amount of energy contained in the reflected wave and the time between transmission of the ultrasound pulse and detection of the reflected waves by the transducer crystals. The interval between transmission and detection of the reflected waves is referred to as “time of flight.” The depth at which the ultrasound image is displayed is determined by this time interval. Reflections from structures in the far field take longer to return to the transducer than do reflections from objects close to the sound source. This time interval is sensed by the CPU and directly converted into distance from the sound source based on the speed of ultrasound propagation within tissue.
The energy contained within the reflected waves is related to the amplitude of those waves. The amplitude of the reflected waves can be measured based on the amount of electric current produced by the receiving crystals. The brightness of the image created by the ultrasound system is determined by the amplitude of the reflected waves. Bodily fluids, such as blood, effusions, and ascites, will transmit nearly all of the energy contained within the imaging beam. Because there is little reflected energy, these areas are displayed as black (or nearly black) on the imaging screen. Air is not dense enough to allow transmission of sound frequencies in the ultrasound range. Therefore, all of the imaging energy present in the beam will be reflected at an air–tissue interface, such as at the edge of a pneumothorax or of the normal lung. This nearly 100% reflection is translated into a very bright (usually white) representation on the imaging screen. Other very dense tissues, like bone, will also reflect virtually all of the energy and be displayed as very bright echo returns. Structures beyond these very bright “echoes” cannot be displayed, because no ultrasound energy reaches them. These areas are often referred to as acoustic shadows. Fat, muscle, and other tissues will transmit some of the imaging beam and reflect a fraction of the sound wave. The amount reflected is related to the density of the tissue, and the amount of returning energy sensed by the crystals in the transducer will determine how brightly an image will be displayed on the video screen.
Figure 1.2. Ultrasound transducer transmitting a plane of ultrasound through a heart in a parasternal, sagittal, or long-axis plane (left). The myocardial and valvar structures reflect the ultrasound energy back to the transducer. The crystals within the transducer detect the returning energy, and the processors within the ultrasound system quantitate the intensity of the reflected waves and the time required for the ultrasound energy to travel from the transducer to the reflector and back. The intensity of the returning signal determines the brightness of the display (right), and the time defines the depth at which the signal is displayed. The central processing unit filters and then converts this information into a video display (right), which corresponds to the anatomy encountered by the plane of sound as it traversed the chest. AV, aortic valve; LA, left atrium; LV, left ventricle; RV, right ventricle.
During a clinical examination, echocardiographers are usually less conscious of amplitude than they are of the frequency of the transmitted sound beam. The frequency of the ultrasound waves has a tremendous impact on the ability to produce images of anatomic structures. The greater the frequency, the greater is the resolution of the resulting image. However, high-frequency sound beams lose more of their energy to surrounding tissues (attenuation) and, therefore, do not penetrate human tissue as well as low-frequency beams. Thus, the echocardiographers must always balance penetration (lower frequencies) and resolution (higher frequencies). For example, the heart of an adult patient will have structures that are positioned farther from the transducer than they are in the heart of a child. Therefore, lower imaging frequencies are usually required to produce adequate images in older patients.
The advent of harmonic imaging has significantly enhanced the ability to examine these older patients with surface echocardiography. Human tissue is not homogeneous in character. As a result, when an imaging beam is reflected by the target, the reflected sound energy exists not only in an unaltered state but also in multiples of the carrier frequencies (harmonic waves). Modern ultrasound transducers now have sufficiently broad bandwidths to vibrate not only at the carrier or fundamental frequency but also at the first harmonic frequency of the transmitted wave. The first harmonic frequency has twice the number of alternating cycles of compression/rarefaction relative to the transmitted frequency. For example, the first harmonic frequency of a 4-MHz ultrasound beam will be 8 MHz. This allows the ultrasound system to transmit at a relatively low frequency but to detect (to image with) reflected waves of a much higher frequency than the original wave. Thus, harmonic imaging combines the advantages of low-frequency transducers (penetration) with the improved resolution associated with higher-frequency imaging.
COMMON IMAGING FORMATS AND IMAGING ARTIFACTS
The earliest echocardiograms displayed either the amplitude or the brightness of the reflected waves on an oscilloscope. These were referred to as either A-mode (amplitude) or B-mode (brightness) echocardiograms. When video screens were linked to echocardiographic systems, it became possible to display the information in “real-time.” These “echocardiograms in motion” were referred to as motion mode studies, or M-mode. M-mode scans display the brightness of the ultrasound reflections, as well as the distance from the transducer and the time at which the reflection occurs. This allows the examiner to visualize cardiac activity as it occurs. M-mode scanning interrogates targets along a single line within the patient. Advances in transducer construction, image processing, and video display allowed multiple M-lines to be fused into a sector of scanning, including usually 80 to 90 degrees of arc. Two-dimensional sector scanning produced a “flat” tomographic display of the areas being interrogated. Two-dimensional imaging remains the primary imaging modality in modern anatomic echocardiography (see Fig. 1.2). However, additional advances in transducer and image processing capabilities now allow real-time volumetric three-dimensional interrogation of the cardiovascular system (Fig. 1.3). As this technology continues to progress, it is likely to, once again, revolutionize the way in which echocardiographic data are acquired.
Figure 1.3. Improved transducer and processing capabilities allow ultrasound systems to scan in “volumes” of interest rather than just in planes. This image was obtained during real-time three-dimensional imaging. The patient had significant Ebstein malformation. The transducer was positioned at the right ventricular apex, and the resulting volume of sound was cropped to display the ventricular cavity at the level of the functional tricuspid orifice (FTO). The abnormalities of the tricuspid valve leaflets are clearly seen. The large anterior leaflet and remnant of the septal leaflets (A/STL) are highlighted (long arrows), and the smaller inferior tricuspid leaflet (ITL) can be seen parallel to the diaphragm (short connected arrows). Ao, aorta; LV, left ventricle; RA, right atrium; S, ventricular septum.
Ultrasound, like any imaging technique, will occasionally produce an erroneous image. These are referred to as artifacts. The echocardiographer must be aware of these imaging anomalies to avoid misinterpretation of the images. Image formation depends on the reflection of ultrasound energy. As a result, the most common artifact encountered is due to the fact that structures that lie parallel to the beam of sound produce no return. The structures do not appear on the video image. This is referred to as parallel dropout. Imaging areas from multiple angles of interrogation is a way to avoid this problem. Very dense, and therefore bright, structures produce echocardiographic shadows that lie beyond the intense return and parallel to the plane of sound. This shadowing reduces or eliminates the information obtainable in these areas. Imaging from multiple windows and angles of interrogation is the most effective strategy when faced with such shadows. In extreme cases, such as shadowing caused by a prosthetic valve, the transducer may need to be placed posterior to the heart to avoid the shadow. These are situations in which transesophageal echocardiography can be extremely useful. Echo-dense structures can also distort the image lateral to the bright reflector. This occurs due to scattering of the ultrasound energy in nonparallel directions to the original beam. This distortion has been referred to as side-lobing. This effect can artificially broaden the appearance of a bright structure, such as a calcified valve or thickened pericardium. Enhanced focusing and filtering capabilities have significantly reduced this issue in modern equipment. Other unusual echo returns can be encountered. These returns often have an arc-like appearance on the video screen. Alterations in frequency, depth of image, or video frame rate will frequently eliminate these returns from the image, confirming their artifactual nature.
THE DOPPLER EFFECT AND CARDIOVASCULAR HEMODYNAMICS
Structures of interest to the echocardiographer are generally not stationary. It is well known that objects in motion reflect sound energy differently than do objects at rest. When an energy wave reflects off or is produced by a moving target, the frequency of the resulting wave is altered based on the direction and speed of the target. This phenomenon was first described by Austrian scientist Christian Doppler in 1843 while he was studying distant stars. Doppler found that the change (or shift) in the frequency of the wave produced or reflected by an object in motion is directly proportional to both its speed and the direction of the motion relative to the observer. This shift has become known as the Doppler effect in honor of its discoverer. A classic example of the Doppler effect is the change in the perceived pitch of a train horn as it approaches and then passes by a stationary observer. The train’s horn produces a sound of a single, constant pitch, defined by the frequency of the sound waves. However, when the train is in motion, the frequency/pitch that will be “heard” by the observer will be greater or less than the transmitted frequency, depending on the direction of motion. If the train is traveling toward the observer, the perceived frequency is greater than the transmitted frequency (more cycles per second). Conversely, the pitch is lower than the transmitted frequency if the train is moving away from the observer.
These shifts in frequency occur because targets in motion toward an observer will physically encounter and reflect the wave more often than will a stationary target. This increase in “encounter rate” compresses the reflected wave and thereby increases the number of cycles per second in the “reflection,” increasing the frequency (Fig. 1.4). If the target is moving away from the observer, the energy wave will encounter the target, and be reflected, less often (Fig. 1.5). The frequency of the reflected wave will therefore be reduced. In the case of cardiac ultrasound, the transducer is the stationary observer and the moving reflector is either the red blood cells within the vascular system or myocardial tissue in motion. The changes in frequency that occur due to target motion are referred to as the Doppler shift, or Δf.
Figure 1.4. Doppler shift: impact of a reflector moving toward the transducer. The motion of the targets (red blood cells) toward the ultrasound source compresses the reflected wave, reducing its period (–ΔP) and wavelength. The frequency of the reflected wave (fr) is thereby increased relative to the transmitted wave (ft) of ultrasound.
Δf = ft – fr
The shift from the original frequency (Δf) can be determined by measuring the frequency of the reflected wave (fr) and determining the difference between that value and the transmitted frequency (ft). This difference is routinely calculated by ultrasound systems.
Figure 1.5. Doppler shift: impact of a reflector moving away from the transducer. The motion of the targets (red blood cells) away from the ultrasound source decreases their interaction with the transmitted wave. As a result, the reflected wave has an increase in its period (+ΔP) and wavelength. The frequency of the reflected wave (fr) is thereby reduced relative to the transmitted wave (ft) of ultrasound.
THE DOPPLER EQUATION
Additional studies of these frequency shifts have shown that the speed (velocity) of the moving target can be determined mathematically. This mathematical relationship is known as the Doppler equation. The components of the Doppler equation are the frequency shift (Δf), the speed of the wave within the medium (c), the original transmitted frequency (ft), and the cosine of the angle between the original wave and the direction of the moving reflector. This angle is referred to as the angle of interrogation, or is designated by the Greek letter theta, θ. If Δf is known, then the Doppler equation can be easily solved for velocity as shown in the following equation:
Δf = 2ft × [(V × cos θ/c)]
Velocity = (Δf × c)/[(2ft) × cos θ]
The influence of the angle of reflection (also referred to as the angle of interrogation) can be minimized by aligning the transducer’s beam nearly parallel to the flow being investigated (angle = 0 degrees, cos θ = 1; Fig. 1.6). In clinical practice, one strives to maintain the angle of interrogation at less than 20 degrees, because the cosine values of all angles less than 20 degrees are essentially equal to 1 (see Fig. 1.6). If the Doppler beam can be aligned with the flow in this way, the angle of reflection will not influence the calculation of velocity by the Doppler equation and it can be ignored. Because the speed of sound in tissue is constant, the only remaining variable in the Doppler equation is the frequency shift, which is measured by the ultrasound system. This relationship allows a nearly direct calculation of the velocity of any moving reflector in the path of the sound wave.
RELATIONSHIP BETWEEN VELOCITY AND PRESSURE DIFFERENCES
Bernoulli was an Italian physicist who had an interest in fluid dynamics. He found that flow velocities were directly related to the pressure difference (P1 – P2, or ΔP) across a flow restrictor (analogous to a stenosis; Fig. 1.7). It was Bernoulli’s work that allowed the Doppler effect to be used in assessing cardiovascular hemodynamics. Bernoulli’s equation states that the pressure difference between two points on opposite sides of a restrictor within a flow stream is related to the difference between the squares of the flow velocities at those two points. His mathematical description of this relationship included terms related to convective acceleration, flow acceleration, and viscous friction (see Fig. 1.7). For clinical purposes, we can simplify this relationship by making a few assumptions. First, one assumes that the influence of friction between the flowing column of blood and the vessel/chamber wall is negligible. This is a reasonable assumption because we are usually interrogating flows in the center of relatively large vessels or chambers. Similarly, the flow acceleration term can usually be ignored because the flow in the area of interest is not accelerating significantly. These two assumptions simplify the Bernoulli relationship to the following equation:
ΔP = ½ρ(V22 – V21)
V2 represents the flow velocity beyond the flow restrictor (valve or other stenosis), V1 represents the flow velocity just proximal to the restrictor, and ρ represents the mass density of the fluid—in this case, blood, which is constant. In the human, with a relatively normal hemoglobin concentration, ½ρ is equal to 4. Therefore, the Bernoulli equation as it relates to Doppler echocardiography is usually expressed as ΔP = 4 (V22 – V21). If the proximal velocity (V1) is relatively low (<1 m/s), then it, too, can be ignored without altering the results of the calculation significantly. In clinical practice, this “very” simplified Bernoulli equation, ΔP = 4(V22), is actually the equation that is most commonly used. Neglecting the proximal velocity is almost always appropriate when evaluating a velocity profile produced by regurgitation. However, in valvar or vascular stenosis, the accuracy of ΔP is improved by including V1 in the relationship. Whenever such assumptions are made, one must make a conscious effort to beware of clinical situations in which the assumptions are being violated. The most common clinical situation that violates these assumptions is encountered with prosthetic aortic–to–pulmonary artery shunts. These prosthetic tubes are small. Therefore, their walls do exert friction on the column of moving blood and the acceleration within the area of interest can also be important in “vessels” of such a small size. Cyanotic patients with significant polycythemia also create a problem for the Bernoulli equation because their blood viscosity is greater than normal, altering the value of ρ.
Figure 1.6. Influence of the angle of interrogation (θ) on calculated Doppler velocity. The Doppler equation relates the velocity to the frequency shift of the reflected wave but also includes the cosine of the angle that the ultrasound beam makes with the flow direction. The angle term (cosine θ) in the Doppler equation can be neglected when the angle is less than 20 degrees. This is because the cosine values of angles less than 20 degrees are approximately 1 (right). Therefore, when using Doppler echocardiography, the examiner should use two-dimensional and color flow guidance to achieve directions that are parallel to the blood flow being evaluated or at least minimize the angle of interrogation. RA, right atrium; RV, right ventricle; V, velocity.
Figure 1.7. Complete Bernoulli equation. This diagram and the expanded Bernoulli equation describe how the flow velocities of a fluid moving through a restriction are related to the pressure difference across that restriction. For most situations in clinical echocardiography, the flow of acceleration and viscous friction terms can be ignored (see text). In most human subjects, the term “½ρ” is approximately equal to 4. The resulting “expanded” Bernoulli equation predicts that the pressure “drop” (P1 – P2) across the restriction can be estimated by the difference between the squares of the flow velocities proximal and distal to the restriction or P1 – P2 = 4 (V22 – V21).
The principles of fluid dynamics allow Doppler echocardiography to do more than evaluate pressure differences. It is also possible to calculate the volume of blood that flows past a specific point in the cardiovascular system using a combination of imaging and range-gated Doppler techniques. The hydraulic formula states that the rate of flow within a tube is equal to the product of the tube’s cross-sectional area (CSA) and the flow velocity of the fluid (Fig. 1.8). Flow volume can be calculated as the product of the CSA and the “stroke distance” traveled by the flowing fluid. Stroke distance is the integral of the fluid flow velocity over time. Flow volume is equal to CSA multiplied by the time-velocity integral (TVI) of the fluid’s velocity (see Fig. 1.8). This concept is easily applied to the determination of flow volume within blood vessels or across cardiac valves (Figs. 1.9 and 1.10). Most vessels and valves are relatively circular, and CSA is used in the geometric equation for the area of a circle [CSA = π × (radius)2]. The examiner measures the diameter (D, in cm) from a two-dimensional scan. The radius (R) is one-half the diameter and π = 3.14. Some prefer to simplify the relationship by combining the constant numerical terms into one value and simply using the measured diameter in the equation rather than the radius. This results in the following relationship: CSA = 0.785 × D2. The other component of the hydraulic relationship, stroke distance, is determined by tracing the pulsed-wave (PW) Doppler signal to “integrate” the velocity profile (determine the average flow velocity during the period being examined, usually one cardiac cycle). The product of this TVI and the CSA is equal to the volume of blood flow. This technique can be used to determine chamber stroke volumes, cardiac outputs, and regurgitant or shunt volumes. Specific applications of these relationships will be covered in more detail in later chapters of this text.
Figure 1.8. Hydraulic formulas. These equations show the relationships of flow rate and velocity to the cross-sectional area of a column of moving fluid, the velocity of the fluid, and the integral (time-weighted average) of that velocity. A, cross-sectional area; TVI, time-velocity integral; V, velocity; ∫ velocity, integral of velocity.
Figure 1.9. Use of Doppler echocardiography in calculating flow volume in a blood vessel. TVI is obtained by tracing the Doppler signal from a PW Doppler sample volume positioned within the “area” of interest. PW, pulsed-wave; TVI, time-velocity integral.
Figure 1.10. Use of the hydraulic formula to calculate the stroke volume crossing the aortic (left) and mitral (right) valves. Cross-sectional area is determined by measuring the two-dimensional diameter of the valve at the annular hinge points. Radius (R) is determined by dividing the diameter by 2. Cross-sectional area is calculated using the formula for the area of a circle (πR2). The pulsed-wave Doppler sample volume is placed between the annular hinge points of the valve being examined. In this way, one can be certain that the Doppler flow time-velocity integral (TVI) corresponds directly to the “area” of interest. R, radius; SD, stroke distance.
DOPPLER ECHOCARDIOGRAPHIC MODALITIES
The most common Doppler modalities used during an echocardiographic examination are color flow, PW, and continuous-wave (CW) Doppler. These techniques are focused on the description of blood flow. Other Doppler techniques, such as tissue Doppler imaging, color kinesis, and Doppler-derived myocardial deformation imaging (strain), are used to describe myocardial activity. The remainder of this chapter will focus on the mechanics of the Doppler techniques used to describe blood flow. The topics of myocardial Doppler examinations will be addressed in later chapters.
Use of the Doppler equation allows determination of blood flow velocities within the heart and central blood vessels. Spectral Doppler techniques (CW and PW) are used to describe relatively discrete flowing streams of blood. By convention, blood flows that are directed toward the transducer have their velocities displayed as positive (upward) deflections above a zero-velocity baseline (red signal, Fig. 1.11). Conversely, flows that are directed away from the transducer’s position are displayed as negative (downward) deflections below the baseline (blue signal, Fig. 1.11). The oldest Doppler echocardiographic technique is CW Doppler. This type of Doppler interrogation is performed using two independent ultrasound crystals within the transducer. One crystal continuously transmits a beam of ultrasound (Fig. 1.12), usually at a relatively low frequency (2 MHz). The other crystal acts as a continuous receiver for the reflected ultrasound waves. Because the sound beam is continuously produced and detected, this type of Doppler will detect velocity profiles from all moving targets in the beam’s path. Because the sound beam is generated continuously, the pulse repetition frequency is essentially infinite (see later discussion of Nyquist limit). The continuous nature of beam generation and sensing allows detection of large frequency shifts and therefore very high-velocity flows. CW Doppler is therefore the primary modality used to define the rapid blood flows encountered in valvular regurgitation or stenoses of any etiology. The continuous nature of CW Doppler also means that there is no spatial information contained in the signals. In other words, this method cannot be used to localize the flow being interrogated. As a result, postprocessing and signal filtering are usually optimized to display the maximum velocity profile encountered. As a general rule, CW Doppler is best recorded with both high gain and filter settings.
Figure 1.11. Continuous-wave (CW) Doppler interrogation of the right ventricular outflow tract (RVOT) (left) and representation of the resulting Doppler tracing. The tracing is shown with a simultaneous, single-lead electrocardiogram to allow accurate timing of Doppler events to the phases of the cardiac cycle. The baseline of the Doppler tracing represents a velocity of zero, or no flow. Blood flow directed toward the transducer is displayed above the baseline, as a positive velocity profile (red shaded area). In this case, the positive signal represents pulmonary valve (PV) regurgitation. Blood flow directed away from the transducer is displayed below the baseline as a negative velocity profile (blue shaded area). In this case, the negative flow signal represents right ventricular (RV) ejection across the pulmonary valve (PV). LA, left atrium; PA, pulmonary artery; RA, right atrium.
Unlike CW Doppler, PW Doppler only transmits the interrogating sound beam intermittently. Once the pulse of ultrasound is generated, the crystal then “listens” for the reflected wave. This characteristic allows the examiner to specifically interrogate flows within an area of interest, often referred to as the sample volume (see Fig. 1.12). For this reason, PW Doppler is referred to as “range gated.” This means that PW Doppler will detect frequency shifts/velocity profiles that occur only within a defined region of interest, allowing one to localize the origin of the flow pattern. The advantage that range gating provides is offset by the limited ability of PW Doppler to assess high-velocity flows without distorting the flow envelope. This limitation stems from the intermittent or “pulsed” nature of the Doppler beam. The highest velocities that can be displayed by PW Doppler are determined by the pulse repetition frequency (PRF) of the interrogating sound beam. If the frequency shift is greater than 50% of the PRF, the resulting signal will “alias.” These aliased signals are displayed as if the reflector was moving in a direction opposite to its actual motion, or on the incorrect side of the tracing’s baseline (Fig. 1.13). The maximum velocity that can be displayed without aliasing is referred to as the Nyquist limit (NL = 0.5 × PRF). The continuous nature of the ultrasound beam used for CW Doppler results in an essentially infinite PRF, and therefore there is no theoretical limit to the maximum velocity that could be recorded by CW Doppler techniques.
Figure 1.12. Physical difference between pulsed-wave (PW) Doppler (left) and continuous-wave (CW) Doppler (right). CW Doppler uses a constant interrogating beam of ultrasound (large arrows) and continuously senses the reflected ultrasound energy (small arrows). In contrast, PW Doppler intermittently transmits the interrogating beam in pulses and detects the reflected energy between those pulses. This allows the ultrasound system to focus on the reflections occurring within a single area of interest, at a specified distance from the transducer. This characteristic, unique to PW Doppler, is referred to as range gating and allows definition of Doppler flow profiles at specific points within the cardiovascular system.
Figure 1.13. Aliasing. These pulsed-wave (top, PW) and continuous-wave (bottom, CW) Doppler recordings were taken from an examination of a patient with both aortic stenosis (AS) and regurgitation (AR). Both tracings were obtained with the transducer at the cardiac apex. The PW sample volume was placed in the left ventricular outflow tract (LVOT), just proximal to the aortic annulus. The PW tracing shows both the relatively laminar, systolic LVOT flow profile, as well as the turbulent, diastolic, high-velocity flow associated with AR. Since the AR velocity (4 to 5 m/s) is greater than the Nyquist limit (NL), much of the PW signal is displayed below the baseline. This occurs even though the AR flow is actually directed toward the transducer at the apex. This phenomenon where a velocity profile “wraps around” to the inappropriate side of the zero-velocity baseline is referred to as aliasing. The CW tracing displays both the high-velocity systolic (AS) and diastolic (AR) signals correctly and unambiguously. The AS flow signal, traveling away from the apex, is all displayed below the baseline. Conversely, the AR flow signal is seen entirely above the zero-velocity baseline, as expected for a flow stream moving away from the transducer. The continuous transmission/detection of the ultrasound beam creates an essentially infinite NL, allowing accurate display of very high-velocity profiles but eliminating the ability limit flow detection to a single region of interest.
Color flow Doppler is based on PW Doppler technology. Color Doppler creates a “map” of velocity data, coded in shades of red and blue. This map is then displayed over an image of the cardiovascular structures within a defined region of interest (Fig. 1.14). Color Doppler data can be displayed with images obtained using any imaging format—M-mode, two-dimensional, or even three-dimensional volumetric scans. The choice of background imaging is made based on the imaging task being performed. Color mapping allows the echocardiographer to “see” both normal and abnormal blood flow patterns and directly relate them to the structures involved (Fig. 1.15). In essence, color flow Doppler is the echocardiographic equivalent to angiography, but with added value. Color Doppler maps have both direction and velocity data encoded in the signals. By convention, flows that are directed toward the interrogating transducer will be shown in shades of red. Flows that are directed away from the transducer are seen in blue (Fig. 1.16). Students often use a simple mnemonic (BART) to assist in remembering the directionality of color coding. BART stands for “blue away—red toward.”
Figure 1.14. Production of a two-dimensional ultrasound image (left) and a color flow Doppler map obtained in the same parasternal long-axis projection (right). The color flow Doppler encodes flows that are directed, even partially, toward the transducer in shades of red. Conversely, flows away from the transducer are coded in blue.
Figure 1.15. Two systolic, parasternal long-axis echocardiographic images displaying the color flow principles described in Figure 1.14. The regurgitant flow crossing the mitral valve is moving at an angle to the plane of sound (thin arrow) that is directed away from the transducer. Therefore, the mitral regurgitation flow is coded in blue. The flow generated by systolic ejection of the left ventricle (LV) is moving toward (thick arrow) the position of the transducer and is coded in red. AV, aortic valve; LA, left atrium; RV, right ventricle.
Because color Doppler is based on PW technology, it shares the ability of PW Doppler to localize flow velocities in space, but it also has an upper velocity limit beyond which the flow pattern becomes aliased. When a flow velocity exceeds the Nyquist limit during color flow imaging, the color map will take on characteristics of flows that are opposite the actual direction of the flow stream. This phenomenon is also referred to as “aliasing.” Disordered or turbulent blood flows are displayed by the inclusion of variance in the color map. This variance is displayed by adding shades of green to the color coding for the direction of flow and is often perceived as “speckling” in the flow stream (Figs. 1.16 and 1.17).
Color flow Doppler is particularly useful for detecting the sources of abnormal blood flows (see Fig. 1.17). Most normal cardiovascular flow patterns are relatively laminar, low velocity, and undisturbed. Flows generated by septal defects, valvar regurgitation, and stenoses of any kind generally will display greater velocity and variance. These features make these flow streams stand out from the more laminar background, allowing them to be quickly and easily recognized. When interpreting color flow Doppler signals, the echocardiographer must remember that, unlike spectral Doppler techniques, color flow does not demonstrate peak or maximal velocities. Rather, the color flow Doppler velocity is a representation of the mean flow velocity in the area being interrogated. While this makes prediction of pressure differences difficult, it is actually an advantage when determining flow volumes. The fact that color Doppler velocities represent the mean flow velocity is one of the major reasons that proximal isovelocity surface area (PISA) analysis of regurgitant and stenotic flows is possible.
Figure 1.16. Color palette used during color flow Doppler mapping. The color coding includes information regarding not only direction but also velocity and turbulence. Velocity is indicated by decreasing intensity of color shade. Turbulence is displayed by adding shades of green to the dominant red or blue directionally determined color.
Figure 1.17. Parasternal long-axis images acquired from a patient after patch closure of a ventricular septal defect (VSD). The VSD patch is shown (left, arrow). There is no obvious gap detected on two-dimensional imaging. The addition of color flow Doppler interrogation (right) allowed rapid recognition of an abnormal flow arising from the junction of the patch and the muscular ventricular septum. This flow has high velocity (note the speckled pattern and variance coding) and a narrow origin from the left ventricular cavity. These features suggest that the defect is actually small, despite the broad color flow disturbance seen within the right ventricular cavity. LA, left atrium; LV, left ventricle.
Although many of the concepts outlined in this chapter are old ones, they provide an essential foundation for anyone investigating the cardiovascular system with ultrasound technology. The chapters that follow will build on these foundations, but when one encounters confusing or inconsistent information during an echocardiographic examination, it is often useful to reflect on these basic principles of ultrasound imaging to determine what portion of the acquired data may be based on an inaccurate assumption or influenced by a technical limitation of ultrasound imaging.
The authors gratefully recognize the generous contributions of Dr. Jae Oh to this chapter. He has generously contributed many of the figures used in this chapter from his teaching library. His zeal for echocardiographic education and investigation is an example to us all.
Currie PJ, Hagler DJ, Seward JB, et al. Instantaneous pressure gradient: a simultaneous Doppler and dual catheter correlative study. J Am Coll Cardiol. 1986;7:800–806.
Currie PJ, Seward JB, Chan KL, et al. Continuous wave Doppler determination of right ventricular pressure: a simultaneous Doppler-catheterization study in 127 patients. J Am Coll Cardiol. 1985;6:750–756.
Edler I, Hertz CH. The use of ultrasonic reflectoscope for the continuous recording of the movements of heart walls. 1954. Clin Physiol Funct Imaging. 2004;24:118–136.
Hatle L, Angelsen B. Doppler Ultrasound in Cardiology. 2nd ed. Philadelphia, Pa: Lea & Febiger; 1985.
Hatle L, Brubakk A, Tromsdal A, et al. Noninvasive assessment of pressure drop in mitral stenosis by Doppler ultrasound. Br Heart J. 1978;40:131–140.
Houch RC, Cooke J, Gill EA. Three-dimensional echo: transition from theory to real-time, a technology now ready for prime time. Curr Probl Diagn Radiol. 2005;34:85–105.
Kremkau FW, ed. Diagnostic Ultrasound: Principles and Instruments. 6th ed. Philadelphia, Pa: WB Saunders; 2002.
Omoto R, Kasai C. Physics and instrumentation of Doppler color flow mapping. Echocardiography. 1987;4:467–483.
Tajik AJ, Seward JB, Hagler DJ, et al. Two-dimensional real-time ultrasonic imaging of the heart and great vessels: technique, image orientation, structure identification, and validation. Mayo Clin Proc. 1978;53:271–303.
Zamorano J, Cordeiro P, Sugeng L, et al. Real-time three-dimensional echocardiography for rheumatic mitral valve stenosis evaluation: an accurate and novel approach. J Am Coll Cardiol. 2004;43:2091–2096.
1.The frequency of a sound wave is determined by:
A.the distance between consecutive peaks of the sound wave.
B.the energy contained in the sound wave.
C.the number of sound waves occurring in one second.
D.the speed of the sound wave Through the medium.
2.An ultrasound image is created by:
A.reflection of the transmitted sound wave by the target.
B.scattering of the transmitted sound wave by the target.
C.frequency shifts in the transmitted sound wave induced by target motion.
D.rarefaction of the transmitted sound wave by the target.
3.Harmonic imaging creates an image using:
A.the fundamental frequency of the transmitted ultrasound wave.
B.the first harmonic frequency of the transmitted ultrasound wave.
C.the second harmonic frequency of the transmitted ultrasound wave.
D.the Doppler shift induced in the transmitted ultrasound wave.
4.The frequency shift induced in the reflected sound wave by a moving target is:
A.determined by the velocity of the target.
B.related to the tangent of the angle of interrogation between the sound wave and the target.
C.an example of the Bernoulli effect.
D.seen only with harmonic waves.
5.The Nyquist limit of a pulsed wave Doppler signal is:
B.one-half of the transmitted frequency of the ultrasound beam.
C.one-half of the first harmonic frequency of the ultrasound beam.
D.one-half the pulse repetition frequency of the transmitted ultrasound beam.
6.The velocity reading produced by a continuous wave Doppler signal can be used to predict a difference in pressures by using:
A.the Doppler equation.
B.the Bernoulli equation.
7.Color-flow Doppler is an example of:
B.pulsed Doppler imaging.
C.continuous wave Doppler imaging.
D.tissue Doppler imaging.
1.Answer: C. All of the potential responses describe features of a sound wave, but frequency refers to the number of sound waves occurring within one second. The distance between two consecutive waves is the wavelength. The energy contained in a wave is related to the amplitude (height of the peak and trough). The speed of the sound wave in a medium is constant and is not related to the frequency.
2.Answer: A. Reflected waves are sensed as they return to the transducer and the CPU then transforms the reflected energy into an image using both the intensity of the reflected signal (to determine brightness) and the time between transmission and reflection detection (to determine depth). Sound energy that is scattered does not return to the transducer and cannot participate in image formation. Frequency shifts are caused by moving targets and are used in Doppler echocardiography but not in image formation. Rarefaction describes the reduced density of the medium (tissue or air) observed at the troughs of the sound wave, and is a characteristic of any sound wave but not directly responsible for image formation.
3.Answer: B. Harmonic images use the first harmonic frequency in the reflected wave to create an image, instead of the primary (fundamental) frequency of the transmitted wave. The first harmonic frequency is twice the frequency of the transmitted (carrier) wave and is produced by both stationary and moving targets and is therefore useful for imaging. The second harmonic frequency is usually not detectable because available transducers do not have an adequate bandwidth.
4.Answer: A. Frequency shifts associated with targets in motion are related to the velocity of the target. The cosine of the angle between the interrogating ultrasound beam and the target is a part of the Doppler equation, which describes that relationship (not the tangent of that angle). The Bernoulli equation is used to determine the velocity of the target causing the shift in frequency and both fundamental and harmonic waves can develop frequency shifts.
5.Answer: D. The Nyquist limit is related to the rate at which the sound wave is emitted by the source (pulse repetition frequency and is equal to ½ of that frequency). Continuous wave Doppler has an infinite Nyquist limit by virtue of the continuous transmission of the interrogating sound beam. The frequency of the beam will impact the Nyquist limit but is not the primary determinant of its value.
6.Answer: B. The Bernoulli equation (in its simplest form - 4V2) describes the relationship between velocity and pressure difference. The Doppler equation describes the relationship between frequency shift and target velocity. LaPlace’s law describes, and Ohm’s law is related to, resistance in electrical circuits.
7.Answer: B. Color-Doppler imaging shows flow patterns within a specified region of interest and therefore is “range-gated.” Continuous-wave Doppler techniques do not allow depth specificity due to the continuous nature of the transmitted beam. Harmonic imaging is not a Doppler technique and tissue Doppler imaging is a different example of a pulsed wave Doppler application, but is not related to color-flow Doppler, as it is used to detect velocity of the myocardium (not the blood pool).