This chapter is also covered by accompanying online material
Jean-Eric Develay-Morice
CHAPTER CONTENTS
The physical principles of ultrasound as applied to fetal ultrasound
What takes time?
The physical principles of Doppler
In practice: setting the controls
Application to the examination of the fetal heart
Ultrasound (US), which for a long time was considered as operator dependent, left little place for the images themselves. Recent legal developments in countries like France have changed this, requiring that the images taken from a US session be used in final reports concerning the patient.
This means that all the views obtained should be of a high standard and thus able to be interpreted by different people.
To do this we must master:
• The pure US technique.
• The framing technique.
• The adaptation of the different scales of gray as it affects the physiology of our eyes.
After reviewing the physical and biologic characteristics of US we will explain those elements that fine tune our equipment allowing an examination based on, and providing, the best possible imaging.
The physical principles of ultrasound as applied to fetal ultrasound
Tissue Elasticity
The stimulation of the US probe's crystals provokes a mechanical deformation of its surface, compressing the skin gradually, and then the underlying structures. The elasticity of these deformed structures provokes a return to their original form. This new deformation brings about, on the tissues that are passed over until the probe's return, a variation of pressure which is the inverse of the initial pressure.
This is the same principle that applies to the compression of a bicycle air pump's piston when it prevents the air from leaving: the moment there is a release the piston rises. The speed of the return varies as a function of the force of the initial compression, but this is also due to the elastic characteristics of the element that is compressed, in this case, air.
The intensity of the returning waves (and thus the quality of the image) depends on the elasticity of the tissues crossed.
Example of tissue elasticity
In this four-chamber view (Fig. 2.1), the US wave crosses tissues that present with different echogenicities in relation to their elasticity:
FIGURE 2.1 There are three different kinds of echogenicity. Anechogenic (nothing to bounce the US, which is represented in black: white arrow), hyperechogenic (most or all the US bounces off a hard structure so that it is represented in black and white as if all is reflected or absorbed: red arrow), and echogenic (intermediary hardness of the structure; the US bounces a little and is represented in gray: green arrow).
• Hard (e.g., a rib), which is hyperechogenic because all US is reflected, but also the waves are absorbed so that the structures behind cannot be visualized (red arrow).
• Liquid (e.g., a chamber), which is anechoic as there is nothing that can cause reflection so that behind this structure all the US give the impression of an augmentation of the US (white arrow).
• Intermediary (e.g., soft tissue) (green arrow).
Reflection Of Ultrasound
Ultrasound acts like light: on smooth surfaces, which are sufficiently large, the reflection of the waves occurs as with a mirror. The images will be sharp if we are directly in front of them, and they will be invisible if the angle is too great. In the latter case (Fig. 2.2) the image has been reflected away from the probe, translating into an image which is anechoic.
FIGURE 2.2 Reflection on a smooth surface can be compared to a mirror. A, B. If the US incidence is perpendicular to a smooth surface of the target, the direction of the reflected wave will be the same, thus it will return directly to the probe. This applies to the crux of the heart. C, D.If the angle is different, the entire wave is reflected away from the probe, as shown here. This also applies to the crux of the heart.
If the surface encountered is not smooth, the resulting image will be correspondingly less sharp, such as with frosted glass, because of the “scattering” of the reflected wave (Fig. 2.3).
FIGURE 2.3 Reflection as a function of the surface of the target. (A, B) If the surface is smooth, most of the wave will be reflected in the same direction. This applies to the chordae. (C, D) If the surface encountered is not smooth, the resulting image will be equally less sharp with a diminished power, such as with frosted glass “scattering” a reflected wave. This applies to a jagged surface.
We search actively for smooth surfaces that we can approach perpendicularly; this is helpful and important to remember in difficult zones.
It is important to understand that by increasing the frequency of the probe we can use the reflection, without diffusion, of a smaller surface so that intense echoes will be produced.
If by increasing the frequency we can improve the image, it is at the cost of creating a false diagnosis of hyperechogenicities (intestinal and renal—most importantly at the beginning of a pregnancy when we are most tempted to use elevated frequencies).
In the study of a smooth surface like that of the interventricular septum (IVS), the lateral approach on the four-chamber view is preferential. The interface of the myocardium and endocardium becomes more marked and the image of the chordae more distinct than the myocardial wall. The markers are clearer, giving us a very echogenic visualization because of the endocardium, and the measure of the thickness will be more precise (IVS in the case of a gestational diabetes) (Fig. 2.4).
FIGURE 2.4 Cardiac application. (A) In the study of a smooth surface like that of the IVS, the lateral approach on the four-chamber view is preferential. (B) The interface of the myocardium–endocardium becomes more marked (green arrows). In this approach an important pitfall has to be avoided. The image of the chordae is more distinct than the myocardial wall allowing it to be confused with the endocardium, giving the impression of an increased thickness of the septum (red arrow). (C) On the contrary, an apical view will limit our view of the septum (red arrow).
It is always advantageous to use perpendicular incidences in looking at interesting structures. This allows us to “highlight” the small smooth surfaces that would be invisible in the case of an oblique incidence.
The Principle Of The Shortest Path
The attenuation that limits exploration is in function of the quality or nature of the tissue, but also the quantity of the tissue that is passed through (Fig. 2.5).
The path we choose is a function of the obstacles that we confront.
FIGURE 2.5 The shortest path. As an exercise we can search for the approaches that limit the distance to the fetus, positioning the left thorax in front in order to facilitate our study of the heart. (A) This difficult path provides poor visualization of the heart, the tissue associated with the ribs causing an important attenuation (red arrow). (B) Often after studying the spine, pressure on the back and kidneys will cause the fetus to move, allowing a better pathway with less attenuation for the examination of the heart. There will be less tissue thickness (green arrow) and fewer obstacles will greatly decrease the attenuation as well.
Going Around Obstacles
The more the US beam meets obstacles, the more the wave is attenuated, weakening equally the incident and reflected beam and thus finally reducing the information that can be furnished by the reflected waves. The attenuation depends on the structure of the obstacle, going from a discrete hypoechogenicity to an US wall, blocking the US and, generating impassable shadow zones, as with calcification.
Conversely, we can use these anechoic zones to facilitate the propagation of US.
In this spirit, we look for “acoustic windows” to obtain the best image, for instance using the form of a liquid near our target area to better frame the target itself (Fig. 2.6).
FIGURE 2.6 Going around obstacles. (A) A cone of shadows masking the right inferior venous return (red arrow). (B) Going around the rib to finally arrive at this vein (green arrow).
We must always look for the best acoustic windows, keeping absorbent structures as far as possible outside of the trajectory of the incident wave. The very quality of our images depends directly on our ability to achieve this goal.
US can always be described as a “fight against time.” There is always a compromise between the time and the quality that we are trying to obtain. (It does not matter if we are on vacation or in a rush, one hour always equals 60 minutes.)
We must also remember that the speed of the US propagation is independent of the constants that we may have chosen.
This speed is constant at 1560 m/s; one wave of US always has the following characteristics:
• 103 µs for a round trip to a depth of up to 8 cm
• 51.5 µ for the same trajectory up to 4 cm
• emitting time is so short that it is negligible in relation to the reception
• these data are absolute and invariable.
There is a limited time to emit and receive. We know that the crystal has two phases—emission and reception—which can never occur simultaneously.
At the same time we must realize that the sharpness of the final image depends on the number of points taken to form it.
Each point corresponds to a round trip of the US wave between the probe and the structure it represents.
The image always results from a compromise between quality and frame rate (i.e. the number of images per second), which depends on our specific objectives.
To improve the frame rate, we choose those settings, from a series of parameters, which reduce useless time loss in relation to what we are looking for. These parameters are as follows:
• the surface to be explored
• the number of crystals stimulated
• the distance traveled
• the number of focus zones
• the use of color Doppler.
What takes time?
The Surface To Be Explored
The larger the surface explored and the greater the number of beam lines will vastly increase the number of pulsations to far more than what is needed. This will increase the time taken to form the image to the point that the frame rate will be greatly diminished.
It is necessary for us to limit the fields explored in our zone of interest in order to increase frame rate.
The Number Of Crystals Stimulated
This number depends on the size of the field. The greater the number of crystals stimulated, the more time we need to obtain the image. If we choose a narrow field, fewer crystals will be stimulated and this allows us to:
• Increase the number of lines per image for better definition without jeopardizing the frame rate.
• Keep the same number of lines per image in order to increase the frame rate.
Most machines allow for a lateral reduction of the field programs concerning the heart.
In selecting a 2D zoom—and without question when we work in color—it is always preferable to choose a “box” that is higher rather than wider to increase the frame rate.
Distance Traveled From The Point Of View Of Time
The deeper the field of exploration, the greater the wave trajectory. The user has no choice but to wait.
It is therefore necessary to limit depth of field as much as possible using only what is necessary.
Some US machines limit the time of reception of echoes to those that are coming from the field being zoomed.
This gives a time advantage which can be used for increasing the frame rate and the quality of the image (by using those settings that reduce time consumption, such as the density of beam lines or the number of focal zones).
The Number Of Focal Zones
This control is often forgotten but has to be constantly adapted to the situation. This setting can optimize US since the loss of quality can be very important if the zone of interest is far from the focal zone, notably with the very high frequencies that are used in the endocavity examination.
Something that is often overlooked—but which should be attended to when the position of the focus has been correctly adjusted—is that each sweep will be more and more precise in the small horizontal band corresponding to the focal zone. It is possible to increase this band by making several sweeps, each with a different focal band. In proceeding like this, we keep only the different focal bands from each sweep, which we will then combine to create the final images. When an image is constructed in this fashion it obviously takes more time to acquire (Fig. 2.7).
The frame rate is divided by the number of focal zones used, which can be as high as eight or nine.
FIGURE 2.7 Focus. (A) The image remains sharp only within the narrow band. (B) The multiplication of narrow bands provides us with a better definition for a greater depth, but this is at the cost of lowering the frame rate. Number of focus: green arrow; frame rate: red arrow.
If the zone of interest is reduced and limited to the proper extent this constant can be overlooked. In doing this the US equipment will automatically adjust its maximal focal zone to the zoom which will then be entirely within this focal zone.
The Use Of Color Doppler
The use of the Doppler effect allows us to detect the presence of movement. These then allow us to quantify direction and speed by studying the variation of the frequency undergone by the incident wave on a moving target.
Color Doppler is the greatest “time eater”. Each pixel (the unit measure of each point of color) represents a pulse. Each pulse analyzes the variation in the frequency of the reflected wave provoked by the movement of each small target (whose size is only one pixel). To do this we must add the time necessary to take the 2D image.
This explains the interest in reducing, to the greatest extent possible, the surface explored in color Doppler; the time that is gained can then be used to increase image cadence or quality.
By the same mechanism, the reduction of the 2D field—even if it is less efficient than color Doppler to keep an image cadence that is satisfactory—should not be overlooked when we are studying the heart.
The time taken by Doppler in pulsed or color modes has led manufacturers to propose an independent freeze for each of these modes (in general 2D) leaving more time for what is left (in general either color and/or pulsed Doppler). Certain manufacturers refuse to use the triplex mode (2D, pulsed Doppler, and color) in order to preserve the best image quality.
The physical principles of doppler
Doppler Color And Time
Movement can be studied with four Doppler modes, as follows.
Continuous Doppler
This is created by variations in the frequency of reflected waves from all the tissues that are passed through by a line that is perpendicular to the probe's crystal. All these frequency variations are represented on the same curve, as if the two targets are moving in opposite directions. However, this mode is not used in obstetrics.
Pulsed Doppler
This mode allows for a limited zone of sampling along the beam line. Its correct reduction will avoid superimpositions of different flows. Ultrasound will arrive in a focused fashion to a very reduced quantity of tissues situated between the calipers, causing a stimulation which is very close over time and which can provoke excessive heating. In practice, this mode must be used prudently, especially during the first trimester.
Color Doppler
Pulses are distributed over a large volume and are thus much less concentrated by unit of tissue. The risk of heating is diminished, but as we have seen before, this procedure results in a distinct diminution of frame rate. In this mode information from reflected waves is transformed into colour as a function of the direction and speed of flow. Here the studied zone is a plan with many more samples than in pulsed Doppler. Each sample is seen as an individual colored point, the totality forming a real vascular cartography.
Power Doppler
Using the same principle as color Doppler, imaging here rests on the analysis of the variation in amplitude, and not frequency. The results give us the capacity to represent flow, no matter what the direction.
These techniques follow different rules:
Incident Angle
For pulsed and color Doppler, the greater part of the principles for 2D apply, except for the incident angle because these techniques are very angle dependent.
Power Doppler also uses the principles of 2D, but since it is only concerned with the variation in the amplitude of the reflected wave on a moving target (and not on its frequency), it is not angle dependent. On the other hand, it does not give information concerning direction or speed of movement.
In pulsed and color Doppler modes the angle needs to be calculated to be as close as possible within the axis of movement. It thus generates the largest variations of frequency at a constant speed.
Let us think of this another way. Imagine seeing a child throwing a ball in the direction of a car. The ball returns more quickly if the car is moving towards the child and less quickly if the car is going away. This variation of speed will be even more important if the angle of the throw is close to the angle of the axis along which the car is moving.
To see this simply in pulsed or color Doppler we consider that the principle concerning the variation of speed applies as follows:
• The wave returns more quickly if it hits a target coming towards the probe.
• The wave returns more slowly if it hits a target going away from the probe.
• These variations become optimal if the axis of the pulse is the same as that of the movement.
The physical principle corresponds to a variation in the periodicity of the reflected waves by an obstacle. This variation of periodicity allows the machine to calculate the direction and speed of movement.
Example
To search for pulmonary veins, we choose an approach to the fetus that allows us to be as close within the axis as possible in order to be able to detect the slowest flow, but at the same time allowing the best color filling (Fig. 2.8).
FIGURE 2.8 Incident angle. (A) To search for pulmonary veins we choose a perpendicular allowing for better visualization of their walls (red arrow) but like this flow cannot be visualized in color Doppler. (B) If we alter the path to be as close to the axis of flow as possible, the wall will not be seen, but flow will be optimized. This allows for the detection of the slowest flow, but will also allow the best color filling (green arrow).
Pulse Repetition Frequency And Aliasing
Pulse repetition frequency (PRF) is the frequency of pulses generated by the probe (not to be confused with the frequency of the probe, which is the characteristic of the wave that is emitted, i.e. its periodicity). A given level of PRF signifies that the examination is only accurate between a minimum and maximum flow speed.
If the speed is below the minimum limit, the flow will be ignored by the device, and thus not seen.
If it is above the maximum limit, the flow will be represented by a chaotic juxtaposition of pixels going in opposite directions. It can even represent a direction that is totally the opposite of the real movement; this is the phenomena of aliasing.
This specific aspect of aliasing is well known: contiguous pixels of opposing colors. These criteria should lead us to increase the PRF until we have a homogeneous color.
On the other hand, the absence of color should lead us to lower the PRF to look for slower speeds.
We must constantly adapt the PRF to the speed of the flow we are studying:
• pulmonary veins: slow speed; low PRF
• aorta: rapid flow; high PRF.
A good setting gives a homogeneous color with few pixels out of the vessel itself, thus giving a precise outline of the vessel (Fig. 2.9).
FIGURE 2.9 Pulse repetition frequency. (A) PRF too high: the flow is unknown. (B) PRF too low: the flow is detected, but is too imprecise. (C) PRF is correct: the flow is detected and indications are precise (speed, delimitation).
The direction of the trajectory influences the setting of the PRF concerning incident waves, which is an essential element in Doppler adjustments.
The consequence of bad settings can translate into erroneous interpretations concerning the flow direction and speed; this is also aliasing.
There is no aliasing in power Doppler because this technique gives us neither speed nor direction of flow.
The phenomenon of aliasing corresponds to the false information that is produced when reflected waves arrive outside of the period that was reserved for them. This happens most often after the departure of the next series of waves.
While the speed of an US wave is constant, the frequency of the reflected wave undergoes variations. These are provoked on the US beam by the movement of the target.
The US wave generates a sinusoid displacement of the tissues around an equilibrium point that can be represented as a curve. The frequency of the wave is represented as a function of the distance between two equilibrium points. This curve can be compared to a spring at rest attached between a wall and a car. If the car moves in the direction of the wall, the spring will compress and the coils will come closer together. The opposite will happen if the car moves away from the wall: the spring will decompress and the coils will move apart. The spring will not change if the car remains stationary. This is, more or less, the principle behind Doppler.
The curve demonstrates that at the same point an ordinate can be found which is both on the ascending and descending slope.
The US machine receives information only from points located on the curve. For every given point, its ordinate value is the same on the ascending slope as it is on the descending slope. The machine determines the slope by finding these points as a function of the time when each one of them is expected to arrive, a determination based on the supposed speed of the target. The adjustment of PRF defines this “time slot”. If the real speed of the target is different from the estimation, the ascending point could arrive at the same time as the descending point, thus leading us to believe that the variation in the frequency is far more important than it really is (Fig. 2.10).
It is advisable to adjust the PRF constantly to the speed of the examined flow. The clearest example of this “tuning” is that used to study intracardiac flow and that used for the study of the PVs.
FIGURE 2.10 Ultrasound wave. The curve corresponds to the movement of one point of a target around its equilibrium point as a function of time (the green point when it returns to its equilibrium point, the red point when it is moving up). If the movement of the target changes the frequency of the US, one sinusoid apparently becomes longer. This is a function of direction and speed. PRF consists of the calculation of two time zones: one for the ascending curve (green box) and one for the descending curve (red box) when the target moves. (A) The curve of the vibration of a point that moves on an immovable target. PRF data define the “time box” (green and red box) when the descending and ascending points (respectively green and red points) will arrive. The speed of the movement will change the period of the curve (the time to return to the equilibrium point). (B) When the movement of the target occurs in the direction of the probe the frequency will be higher and the ascending point will come before the position it was in when the target did not move, and it will appear to “fall” more quickly, in the red box. This speed will be represented as positive. (C) If the movement occurs in the opposite direction to the probe, the frequency will be lower so that the descending point (green point) will arrive a little later than if the target were motionless. This point will be in the green box but a little later. (D)Aliasing. Here, the movement also occurs in the direction of the probe but its speed is faster when compared to what was predicted. The variation in the frequency will be such that the ascending point (red) will be found within the zone reserved for the descending point (green box and point). The represented color of the point cannot be differentiated by the machine so it will be calculated as a green point (a point returning to its equilibrium point) and not a red one in this box (situations shown in C and D cannot be differentiated). The subsequent interpretation of the speed will be wrongly represented as negative.
No matter what the frequency, or the depth that is used, the emitting time is negligible in relation to the time necessary for reception.
The 3D Technique
Recent and rapid technological progress has opened up new possibilities of investigation. The 3D technique is in the process of revolutionizing fetal cardiac imaging. It consists of acquiring a volume which is constituted by the juxtaposition of multiple US frames, acquired through a lateral mechanical sweep. This had been limited for many years to stationary targets because of the size of the acquisition which had been detrimental to image quality.
!!! Attention !!!
While the use of 3D makes it easier for us to acquire the right views, it can never compensate for the mediocrity of the initial images.
Recent advances in computer science have allowed for a considerable reduction in the time needed for acquisition, thus giving us access to volume in real time. Image quality has progressed as well, and cardiac structures can now be studied with great precision.
One of the new applications, the spatiotemporal image correlation or STIC method allows us to reconstruct the acquired views in function of the cardiac cycle, in classic color Doppler or power modes (Fig. 2.11).
FIGURE 2.11 Appearance of the four-chamber view with the valves and their situation with 3D.
Another possibility of image treatment, which is even more recent, is a mode called inversion. In this mode, hypoechogenic structures are isolated and colored, giving us a cast or mold of the cardiac chambers and vessels. This technique has the advantage of being independent of the incident angle. Contrary to Doppler, it allows the visualization of vessels undergoing widely varied speeds of flow, permitting us to have all these vessels viewed together. Moreover, as it does not use the Doppler mode, we have much more time to concentrate on finessing the image or the frame rate.
Finally one the biggest hopes for improving image quality while reducing acquisition time is the practical use of matrix volume acquisition where the mechanical sweep is replaced by an electronic one. In comparing the evolution of this technique to that of 2D, from a manual sweep of a single crystal to an electronic sweep we should expect a clear improvement in the technical quality of our examinations.
For the moment, limited to use in postnatal examinations, the exploration of the fetal heart by MRI will probably also be of great use in pathology, as soon as its resolution and acquisition speed allow it to be used efficiently.
In Practice: The Settings
The present day quality of US is generally sufficient to allow us to determine the normality of the heart during our first examination while keeping the same settings throughout the entire examination. However, it is useful to understand what affects the image in order to know how to vary the constants to improve it.
In practice: setting the controls
The 2D Settings
Zoom, focus, and gain can be modified continually, while other parameters are preset and remain constant.
Zoom
At the price of being destabilized by the new images when we first use the zoom, this function allows us to better visualize certain objects while sometimes leading us to explore structures that were not yet suspect (Fig. 2.12).
FIGURE 2.12 Zoom. (A) The heart is too small to see the details with a poor frame rate of 20 frames per second (fps). (B) The appropriate zoom to study the heart with a better frame rate of 40 fps. (C) A greater zoom emphasizes structures and the different echogenicities at a frame rate of 100 fps. (D) This zoom can lead us to unsuspected structures such as septal attachments of the tricuspid valve (green arrow) but only if we have a good trajectory. The high frame rate allows us to acquire intermediary positions of the valve during its movements.
Zoom especially allows us to make adjustments targeted at the region we are studying, without letting the operator be perturbed by modifications to that image by neighboring structures.
In some US equipment the use of this parameter will only treat waves coming directly from this zone, allowing us to increase, in a substantial fashion, the image cadence, or use settings that take more time for an equivalent cadence.
Example of using the zoom
This is essential to the study of the heart. When we take the appropriate approach, the zoom can allow us to study previously unsuspected structures such as the crux of the heart at the septal attachments of the tricuspid valve, which would not be visible, even with the same approach if we were not using the zoom. Here we use a very specific setting, which gives preference to the interfaces and image cadence (structures whose echogenicity is very different, the heart pumping at around 145 beats per minute), but which is unusable with other organs.
Focus
Each sweep will be more precise on the small horizontal band called the focal zone, which we can move about the interest zone. Though the settings involving focus are often forgotten, we should always be sure that we have constant optimization of the placement frame (Fig. 2.13).
FIGURE 2.13 Focus. (A) The focus is too high (red arrows) and the interest zone is less precise. (B) Better definition with an appropriate position of the focus (green arrows). (C) Limiting the zoom to the heart will automatically place the focus in a good position (green arrows).
This problem is quickly resolved when we use a significant zoom. If the interest zone is sufficiently reduced, allowing for an important zoom, the focus settings can be forgotten because the machine will automatically place its focal zone within the zoom.
It is possible to enlarge this band by conducting several sweeps, each with a different focal zone. We then keep the focal bands of each sweep, which will eventually be juxtaposed to create the definitive image. The price we pay here is in time. The frame rate is divided by the number of focal zones used (the number of focal zones can go as high as eight or nine).
• For zones without movement, and when studying the entire range of a depth of field, multiply the focal zones (aortic arch and descending aorta).
• For zones of fast movements and small size use one focal zone (crux of the heart).
Gain
Expressed in decibels (dB), as the logarithmic unit, it is too often elevated with the accompanying risk of “burning” the image (making it too white) and blurring the differences of echogenicity essential to the diagnosis. It should be adjusted to obtain the maximum amount of information about the studied structures, which often occurs to the detriment of the rest of the image. The characteristics of the heart are very different from other tissues, and setting the gain after using the zoom allows us to optimize the image (Fig. 2.14). Certain manufacturers have made this function automatic.
FIGURE 2.14 Gain. (A) The gain is too high: gray nuances are crushed. (B) The gain is too weak: structures such as valves with little echogenicity are ignored. (C) Correct gain setting: an adapted gray scale is given for the zone studied.
In color mode, the adjustment of gain is essential:
• Too weak: there would be an absence of flow or a feeble flow diameter, which is not the case.
• Too elevated: the amount of parasite signals will be multiplied, giving an image that would be difficult to interpret and showing vessel contours whose diameters are greater than that in reality.
Preset elements
Predefined on machines in the form of specific programs (for instance “general”, “cardiac”, or “superficial”), the precise knowledge of these settings is not essential, but can help in optimizing images or in researching precise elements such as very rapid movements like the opening of the cardiac valves.
Dynamic range
This is expressed in dB. Lowering of the dynamic hardens the image by increasing the differences between the gray levels, but it equally brings about a reduction in the quantity of grays that are available, thus the image becomes more black and white (Fig. 2.15).
FIGURE 2.15 Dynamics. (A) A high dynamic blocks contrast and leaves contours blurred. (B) Lowering the dynamic makes contours more precise. (C) Correct gain setting: an adapted gray scale is given for the zone studied so that the valve (green arrow) is particularly well seen.
This low dynamic range is not recommended when trying to differentiate a myocardial structure from pericardiac ascites because it makes the myocardium very hypoechogenic (like liquid). In case of doubt, increasing the dynamic will increase the gray scale augmenting only the myocardial echogenicity and making it visible, which will now make differentiation from ascites possible.
To study the fetal heart, a relatively weak dynamic is chosen in order to obtain clearer interfaces by diminishing the echogenicity of the blood.
To study the myocardium, we raise the dynamic to increase the sensibility of the gray scale allowing a better visualization for the study of the structures of the muscle and not simply its contours.
Frequency
Corresponding to wavelength, the higher the frequency the more precise the image will be by multiplying “mirrors” (the higher the frequency, the more that structures of smaller size act as “mirrors”, generating intense echoes). At the same time attenuation increases in an exponential fashion with the thickness of the tissues that are passed through, so that the depth of field that is studied rapidly limits the zone of exploration.
The speed of conduction is independent of the frequency and has no influence on frame rate.
To simplify we begin with the highest frequency possible, whether in 2D, or color or pulsed Doppler. Then we lower it if the attenuation is too great.
The density of pulse lines per image
This adjustment is also called resolution or space-time, depending on US manufacturers.
An image is constructed based on a certain number of lines per pulse. As with frequency, the more this density is elevated, the more precise the image. Increasing frequency and density of lines improves the resolution.
We seek to increase beam lines on a target that moves little.
An efficient zoom allows an identical frame rate to increase the density of these beam lines.
Persistence
(The scale is variable depending on the specific machine.) Persistence is the superimposition of images in time which gives the “smooth” and more regular character to an image (Fig. 2.16). The heart beats at 145 times per minute and is so rapid that the images that are superimposed often have slight differences which generate a blur, such as in the visualization of the cardiac valves in their intermediary positions. This is similar to what happens when you move when taking a photograph.
FIGURE 2.16 Persistence. (A) Maximum persistence: the image is less hard, but produces a blur, particularly on the valves in their intermediary positions. Also, the valves will appear thickened. (B) No persistence: more aggressive, the image stops every blur so that each frame become “exploitable.” This setting is very important in the study of the valves and septum. Its measurement can be overestimated with the chordae being very hyperechogenic because they are parallel to the septum. This uncertainty will be cleared up by the precision of each frame, demonstrating the movements of the chordae by considering two frames. With still lower values, the cavities will appear even less echogenic with increased contrast.
In the study of the fetal heart, persistence eliminates blur.
Contours
(The scale is variable depending on the specific machine.) This setting allows us to favor those linear elements perpendicular to the US beam (Fig. 2.17). It is especially useful in organs such as the heart with numerous interfaces (creating very intense reflective echoes) particularly with the cardiac valves. The value of this adjustment is only seen when the incidence is perpendicular.
FIGURE 2.17 Contours. (A) Few “contours” give a blur leading to problems in the study of the heart. (B) The maximum use of contours gives a better visualization of the septum and valves.
This method is extremely useful in measuring the septum for gestational diabetes. The endocardium is smooth, the opposite of the myocardium. The echoes illustrate this, allowing us a clear limit to the septum we are measuring, but only if we have a perpendicular approach to benefit from the “mirror” effect of the smooth interfaces, i.e., by a transverse approach.
These settings will allow a good visualization of the offsetting of the atrioventricular valves at the level of the crux of the heart, with a highlighting of the oblique direction of the fibrous bridge that crosses the septum linking the mitral and tricuspid valves.
Doppler Settings
The same physical principles for 2D mode apply here, with special attention to the incidence of the US beam and the time factors in color Doppler where each pixel (point) corresponds to an individual roundtrip of the wave.
We must remember to optimize the size of the sampling “box” of the zone studied, to the narrowest possible, to limit the number of stimulated crystals (Fig. 2.18). This holds true whether we are in pulsed or color Doppler to gain time.
FIGURE 2.18 Zoom. (A) A wide 2D window at 17 fps. (B) The same color “box” with a limited 2D field at 26 fps. (C) A small 2D field with a small “box” used for the study of the pulmonary veins at 50 fps. This has the advantage of avoiding coloring the four chambers (red arrow), which would “drown” the image of the pulmonary vein.
The direction of the incident wave
In order to emphasize the variations of frequency which allow good Doppler sensitivity, as in the example presented in “Physical principles” where the child was playing with a ball, we have to be as close as possible within the axis of movement (Fig. 2.19).
FIGURE 2.19 US direction. (A) The perpendicular direction of US in the pulmonary veins optimizes the visualization of their walls in 2D (red arrow) but leads us to believe that there is an absence of flow. (B) By simply modifying the incident angle it is possible to make the vein's flow appear (green arrow), as the “walls” of the veins are made to disappear.
Pulse repetition frequency
Pulse repetition frequency should be adapted continually to the speed of the target and the US angle so that the reception of the wave arrives in the predicted time and location reserved for it (Fig. 2.20):
FIGURE 2.20 Pulse repetition frequency. (A) PRF is too high for the flow so it is ignored (red arrow). (B) PRF is too low for the flow, causing aliasing. Most of the color is red, representing a false direction for these veins, with some blue side by side, which represents an opposite direction (red arrow). We can only conclude that there is a flow. (C) PRF is well set giving a homogeneous flow in blue representing the correct direction (green arrow), with a clear indication of the speed.
• Too high: The positive speed will be considered as negative and the smallest movement will be encoded into a color which gives an image too saturated to be interpretable.
• Too low: The greater part of the vessels will be “drowning” the vessel you are looking for.
If it is too weak, flow will be ignored.
For studying the heart, the PRF should be set at the maximum to search only the most rapid flow. On the other hand, it should be much lower to reduce the “box” when studying the pulmonary veins.
Color gain
The setting for color gain is often forgotten. It is preferable to make this adjustment last after having selected the zone of exploration and the PRF (Fig. 2.21).
FIGURE 2.21 Gain. (A) The gain is too high, the image artifact “burns” the true flow. (B) The gain is too weak and the flow is overlooked. (C) The gain is correct, visualizing the flow with a good filling of the four chambers.
If the gain is:
• Too weak, we will have the impression that there is no flow and we could underestimate the flow diameter, even the size of the vessel.
• Too high, we will have a number of parasites that will mask the flow. Filling will appear blurred, giving the impression of a flow diameter, and thus a vessel size, greater than in reality.
Application to the examination of the fetal heart
The Echo-Structure
The echogenic gradient of cardiac tissue is particularly important, linked to the differences in cardiac tissue blood, but also because of the structural differences of the various cardiac tissues themselves.
• The myocardium is a tissue which is not very dense, made by fibers going off in every direction, and thus, for US, it is not angle dependent.
• Valvular tissue and chordae are, on the contrary, very dense tissues, of a fibrous nature having a single direction generating an important mirroring effect, leaving them very angle dependent for US examination.
• The endocardium is an intermediary tissue for US reflection. A perpendicular approach can differentiate it from the myocardium, a characteristic which is particularly interesting in the estimation of the septum's thickness in a diabetic pathology.
The Position Of The Fetal Heart
The fetal heart is found lying flat on a horizontally viewed diaphragm, surrounded by the ribs. The pathways to approach it are thus narrow in a fetus whose position is variable (head, breech, transverse, etc.) and in almost constant movement. All of these elements taken together require that we have an excellent understanding of the position and anatomy of the heart in order to determine, in advance, the approach path to take which will favor each of our reference views.
The movements
The target, i.e., the heart, moves quickly and instinctively we must follow these movements. Not only does our target move quickly, but the position of the fetus itself constantly varies.
• The heart is beating at an average of 145 beats per minute, necessitating that we favor frame rate, especially for the kinetics of the valves.
• The respiratory movements sometimes force us to place ourselves in the correct position in relation to the fetus, and then wait patiently until the structures we want to study pass before the US window to be captured.
• There are certain movements of the fetus that we can use by knowing when to wait, or even by provoking them ourselves, causing subsequent changes in the position of the fetus which will be more favorable for the view we are looking for.
The aim of the US specialist should be to see the elements directly from the front taking great care to avoid “wave eaters” that precede the target, always placing ourselves in the axis of movement of the Doppler.
Further reading
Arbeille, P. Mise au point 2003 sur les risques d'effets biologiques par échographie, Doppler pulsé et couleur in SFAUMB 2003. www. sfaumb. org.
Chaoui, R, et al. Three-dimensional (3D) and 4D color Doppler fetal echocardiography using spatio-temporal image correlation (STIC). Ultrasound Obstet Gynecol. 2004; 23(6):535–545.
Kremkau, FW. Diagnostic ultrasound: principles and instruments, 5th ed. Philadelphia: W. B. Saunders Co. Ltd; 1998.
Kremkau, FW, Taylor, KJ. Artifacts in ultrasound imaging. J Ultrasound Med. 1986; 5(4):227–237.
Lee, W, Goncalves, LF, Espinoza, J, et al. Inversion mode: a new volume analysis tool for 3-dimensional ultrasonography. J Ultrasound Med. 2005; 24(2):201–207.