4.2.1 The principles of echocardiography
2DE and M-mode echocardiography
A understanding of ultrasound physics and the control settings of ultrasound machines is required if good quality images are to be obtained. Ultrasound is high-frequency sound which is produced when a piezoelectric crystal, mounted in a transducer, is stimulated by an electrical current. The sound waves are too high in frequency to be audible. They are thought to be harmless to tissue at the intensities used in diagnostic imaging. The passage of sound waves depends on the acoustic impedance of body tissues. Sound is reflected by interfaces between materials of different acoustic impedance.
The majority of the ultrasound waves pass through structures on to other structures lying further from the surface, but reflected sound returns to strike the crystal, deforming it and producing electric signals which correspond to the degree of deformation. This electrical information is transformed by electronics in the ultrasound machine so that it can be displayed on a cathode-ray tube as pulses of light. Because the speed of sound within the body is relatively constant, the depth of the tissue interface can be known and reflected echoes are displayed on the screen on a depth scale.
In echocardiography, sound is directed into the body and is reflected by interfaces between tissues of different acoustic impedance such as myocardium, valves and blood. Blood reflects little sound so it appears relatively black (hypoechoic, or anechoic) compared with the myocardium which reflects more of the ultrasound and therefore appears relatively white (hyperechoic or echoThe endocardium and valves are the most echogenic structures. Ultradoes not pass through air or bone. Because the heart is surrounded by lung over the majority of its surface and is contained within the bony cage of the thoracic cavity, the ultrasound beam must be aimed through gaps (which are known as acoustic windows), in order to produce images of the heart.
Frequencies of 2-10 MHz are used in diagnostic ultrasound. The lower end of this range is used in transducers designed for equine echocardiography because low frequency sound penetrates a greater depth of tissue. A trade-off in the use of low frequency transducers is that there is a loss of image detail because the increased wavelength results in reduced resolution.
M-mode echocardiography and two-dimensional echocardiography (2DE) are the two techniques which are usually used to produce images of the heart in real-time. In M-mode echocardiography (motion mode, or time-motion mode), the crystal is stationary and the beam produced is a pencil-beam of sound. The signal is produced almost continuously. Echoes are displayed on the screen on the Y axis, with time displayed on the X axis (Figure 4.4). This produces an almost continuous image of the position of the cardiac structures which are in the line of the beam.
Echocardiography has become more popular since the advent of 2DE, which produces an image of cardiac structure. 2DE images are also referred to as B-mode images (brightness mode). Originally, static B-mode images were probut improvements in technology allowed the image to be rapidly updated resulting in a real-time image. 2DE images are more easily understood than Mtraces because of the greater spatial detail. A comparison has been made between the two techniques and different ways of illuminating a room. M-mode is the equivalent of looking round a dark room with a narrow-beam torch. 2DE is the equivalent of using a floodlight.
A 2DE image can be produced in one of three ways. The first and simplest is to have multiple crystals mounted in line to produce a 'curtain' of sound. This method is known as linear array and is used in many rectal probes in current use in veterinary medicine. The problem with the use of this technology for echo-cardiography is that it is difficult to aim the curtain of sound through the acoustic windows, so some of the heart is likely to be obscured from view and only a limited number of views can be obtained. This prevents a complete echoinvestigation, although it may allow detection of gross abnormsuch as a pericardial effusion, very poor myocardial contractility, or large vegetations on the heart valves.
To be able to examine the heart in a large number of image planes, a point source is required with a sector or 'fan' of sound produced by sweeping the sound in an arc. Transducers which produce this arc of sound are called sector scanThe sector can be produced by rotating or oscillating the crystal mechanically; this type of transducer is called a mechanical sector scanner. Alternatively the sector can be produced by using an array of crystals which are electronically stimulated in sequence to produce a fan-shaped beam. These are known as phased-array transducers. Because the beam has to be swept through an arc, a finite time is taken to produce each sector image. The arc is then repeated and the image is updated. The quality of the image therefore depends on how many lines of data are displayed per arc and how often the image is updated. The rate at which the image is updated is known as the frame rate.
One of the main problems in equine echocardiography is the great depth which is required for the whole heart to be displayed on the screen. Because the transducer has to wait longer for the sound to be reflected back from tissue interfaces at greater depth, the whole arc takes much longer to produce than those with less depth, and the frame rate is relatively low. Fortunately, the sign of this is offset to some extent by the slow resting heart rate in the horse, which means that the number of frames per cardiac cycle is much the same in equine echocardiography as it is in human echocardiography.
The principles of Doppler echocardiography (DE) are somewhat more complex than those of ultrasound imaging. DE has been used in equine medicine since the late 1980s and has proved particularly valuable in evaluation of congenital heart disease and acquired valvular heart disease. It is used to provide information about blood flow, following evaluation of the structure and the size of the heart using 2DE and M-mode studies. DE is also very useful in a research setting for assessment of cardiac function. A thorough DE examination requires rigorous technique and is extremely time-consuming. Equipment for DE is still relatively costly and the technique is likely to remain limited to referral centres for the foreseeable future. Recently, colour-coded DE has become available to a few institutions. This technique reduces the time required for Doppler examinations and makes them easier to perform, but has the same physical limitations.
The Doppler principle is evident to anyone standing by a railway line or a busy road as traffic moves past: the pitch of the sound of an engine drops as the vehicles pass. This is because sound waves are compressed as a source of sound moves towards an observer and are therefore higher in frequency than those emitted when the source is travelling away from the observer (Figure 4.5). The principle also applies to sound which is reflected off a moving target. The change in frequency of the sound is proportional to the velocity of the target in relation to the source of the sound.
DE involves the emission of ultrasound waves of known velocity which are reflected from interfaces and return to the transducer, in the same way as for echocardiographic imaging. The equipment detects changes in the frequency of the reflected sound in comparison to the emitted sound. In DE, the most important interfaces which reflect the sound are red blood cells (RBCs). Thus, if the emitted ultrasound is reflected off RBCs moving towards the transducer the reflected sound will be of a higher frequency (and shorter wavelength) than the emitted sound. The change in frequency (frequency shift) is proportional to the velocity of the cells towards the transducer. A computer inside the Doppler unit calculates the velocity of the moving blood from the Doppler shift equation (Table 4.4). The calculated velocity is displayed on a velocity/time graph with blood flow towards the transducer displayed above a baseline and flow away displayed below it. This form of display is known as spectral Doppler (Figure 4.6). The standard form of DE, in which the transducer emits and receives sound waves simultaneously, is known as continuous-wave (CW) Doppler.
A critical feature of the Doppler shift equation is the effect on the angle of the beam to blood flow (Figure 4.7). If the cells are moving in a direction oblique to the line of the ultrasound beam, the velocity calculated will be an underestimate of their true velocity unless this angle is known and can be included in the calUnfortunately, the exact angle is seldom known. Estimating it and using the angle correction software which is available on many machines usually only adds to the potential inaccuracy of the method. This means that, at all times, every effort should be made to keep the line of the ultrasound beam parallel to blood flow. In practice, angles under 15o either side of parallel are acceptable because the error will be less than 4%.
Initially, DE was performed independently of echocardiographic imaging. Later, it became possible to display Doppler information at the same time as an image, a technique known as duplex imaging. This allows the echocardioto guide the position of the Doppler beam within the heart. Another advance was the development of pulsed-wave Doppler (PWD). With standard CW Doppler, the frequency shift in the reflected ultrasound can come from anywhere along the course of the beam. However, it is often helpful to obtain information about blood flow in specific parts of the cardiac chambers and great vessels. PWD sends a pulse of known, short duration, and then records returning echoes for a limited period some time later. By limiting the period during which returned echoes are detected, the distance from the transducer of the targets which return the echoes is known. The 'gated' period can be displayed on the screen as a small box known as the sample volume. This corresponds to the area of the heart from which the Doppler signals are received. The time taken for the PWD sound-wave to reach the near end of the sample volume can be termed T1. The time taken for it to reach the far end of the sample volume can be termed T2. The machine starts to listen to returning echoes after T1x 2 and stops after T2 x 2 (Figure 4.8). The sample volume can be guided into specific; areas of interest such as the atrial side of atrioventricular (AV) valves to detect regurgitant blood flow, or the right ventricular side of a ventricular septal defect (VSD) to detect blood flowing through the defect. Thus DE can be used to detect blood flow, to identify the direction of flow, and to calculate the velocity of flow.
One of the problems with the use of PWD is that the rate at which the pulses of sound are produced is limited by the fact that a new pulse cannot be sent until the preceding one has been received. The number of pulses which can be sent per second is known as the pulse repetition frequency (PRF). The greater the depth of the sample volume, the lower the PRF. The PRF becomes a limiting factor when measuring the frequency shift associated with high velocity jets. This is because of a physical factor known as the Nyquist principle. This states that frequency shift (and therefore velocity) can only be measured accurately when the sampling rate is twice or more than twice that of the frequency shift being measured (Table 4.4). When the Nyquist limit is exceeded, the flow of blood will be displayed in the opposite direction to its true direction (aliased) which can be confusing. Aliasing is a particular problem in horses because the depth of many areas of interest means that PRF is relatively low, while the velocity of jets associated with valvular regurgitation or a VSD are usually quite high. For this reason, quantitative data about high-velocity jets is best derived from CW Doppler, which is not affected by aliasing.
When a PWD sample volume is placed in an area of laminar flow, for example the flow which is usually found in the pulmonary artery, a clear line will be dison the velocity/time graph indicating that the majority of the blood cells are moving at the same speed. Where this clear line is seen it is described as an 'envelope' (Figure 4.6). A line can be drawn around the envelope, and the area under the line is the velocity/time integral (VTI). The VTI is directly proportional to the stroke volume ejected through the valve and, if the area of the vessel is accurately measured and the VTI is the maximum which can be recorded from the site (i.e. the ultrasound beam is parallel to the line of blood flow), the stroke volume can be calculated. Cardiac output is then simply calculated by multiplying the stroke volume by the heart rate (Table 4.4).
The clarity of the envelope of a PWD signal depends on the range of blood flow velocities within a sample volume. When the sample volume is placed in an area of turbulent flow associated with a regurgitant jet which is causing a cardiac murmur, a range of RBC velocities will be seen because blood is flowing in difdirections relative to the transducer. It may even be displayed either side of the baseline.
The principles of physics can be put to good use by clinicians. An example is the use of DE to estimate pressure gradients, a task which previously required invasive catheterisation techniques. Once accurate measurements of the frequency shift associated with a jet of blood flowing between two chambers have been made, it is possible to estimate the pressure gradient between these chambers using a derivative of the Bernouilli equation. A simplified version of this equation is suitable for use in most instances (Table 4.4). An example of the use of this principle is estimation of the pressure gradient between the left ventricle (LV) and right ventricle (RV) in an animal with a VSD. This is estimated from CW Doppler measurement of the velocity of blood shunting through the defect. If the RV pressure is raised then the defect is likely to be of clinical significance and may affect the horse's athletic ability (see section 5.5.5).
Colour-flow Doppler mapping
Colour-flow DE (colour-coded DE) is a form of PWD which is subject to exactly the same physical limitations as standard PWD. The direction of flow in a large number of individual sample volumes is measured and colour-coded. Usually, flow away from the transducer is coded blue, flow towards the transducer is coded red. The colour scale is usually displayed on the side of the screen. The tone of the colour depends on the velocity of flow and its brightness on the intensity of the signal (i.e. on the number of RBCs reflecting sound). The colour-coded pixels are displayed in a sector superimposed on a 2DE image so that the location of the flow within the heart is known. It is important to restrict the angle of the 2DE image and colour sectors as much as possible because the inforrequires time to be collected and may produce this result in a very slow frame rate, particularly if the depth display is large. As with PWD, aliasing occurs; with colour-flow Doppler the blood will be encoded with the opposite colour from that expected. Some machines code areas of turbulent or disturbed flow with a 'variance' pattern in green.