Chapter 2. Understanding Ultrasound System Controls

It is crucial for clinicians performing transesophageal (TEE) examinations to understand how the controls on an ultrasound machine alter the display. Without this knowledge, it is impossible to consistently optimize images, and unskilled manipulations may misrepresent diagnostic information and result in missed diagnoses. This chapter describes the controls found on most ultrasound machines, how they affect the image, and how they are used to optimize the ultrasound image. Table 2–1 presents the most commonly used controls for two-dimensional (2D) imaging.

After providing power to the machine itself, a TEE probe must be connected to the machine, register as compatible with the machine, and be selected from other possible transducer options. The basic parameters for the ultrasound examination may be defined by choosing an appropriate TEE preset. The preset provides a starting point for basic machine settings such as depth, gain, and image processing settings. The operator can adjust all the machine's variables from the initially fixed settings, as needed. Adjustments to the preset can be saved permanently under a different preset name when desired. Patient identification (name and medical record number) and any other relevant information should be entered into the machine before beginning an exam. This includes date of birth, sex, videotape number, name of person performing the examination, location, and a number of other qualifiers.

The five most common modes used during TEE examinations are 2D gray-scale imaging, color Doppler, pulsed-wave (PW) Doppler, continuous-wave (CW) Doppler, and three-dimensional (3D) imaging. The usual buttons to enable these modes are 2D, Color, PW, CW, and 3D. Other scanning modes, such as M-mode and angio, are often available but are minimally important in comparison. Figure 2–1 is an example of four common ultrasound control panels. While the number and layout of buttons and controls are different, there are many similarities. This chapter focuses on controls that affect 2D imaging, color Doppler, pulsed-wave Doppler, and continuous-wave Doppler. Three-dimensional imaging is a new and exciting addition to TEE, especially for the evaluation of the mitral valve, and will be briefly discussed at the end of this chapter.

Two-dimensional gray-scale imaging is a type of B-mode imaging (B is for brightness) in which the various amplitudes of returning ultrasound signals are displayed in multiple shades of gray. Higher amplitude signals are closer to white, whereas lower amplitude signals are displayed as closer to black. The many different shades of gray form an image or representative picture of the patient's cardiac anatomy. TEE probes generate a sector or pie-shaped display of gray-scale images, with the top portion of the sector showing the tissue closest to the transducer. Of the five modes, the 2D display mode is most commonly used and manipulated during a TEE examination. Two-dimensional imaging also provides a reference point from which to activate all three forms of Doppler (color, PW, and CW).

Overall gain or amplification is the first postprocessing function performed by the receiver and is the most important variable to adjust during a study. Overall gain controls the degree of amplification that returning signals undergo before display. By increasing gain, small voltages are changed into larger voltages by an operator-specified level of amplification. Gain is also the one control that is misused most often, with the most common mistake being the addition of too much gain to an image. Although additional gain can make the picture brighter and structures more obvious, using too much gain, or over gaining, will destroy image resolution. The appropriate amount of gain for any given image becomes apparent when reflectors and tissue interfaces can be seen, but fluid and blood appear as totally black and echo-free. Myocardium should be adequately dispersed with reflectors by setting it at a medium shade of gray, but the muscle should not approach the look of a solid white band. Only the pericardium, certain states of abnormal thickening, calcification, tissue infiltration, and surgically altered valves should be hyperechoic (very bright). Gain should be added incrementally if the picture is entirely black or if the only structures seen with clarity are normally hyperechoic. Figure 2–2 shows a TEE image with varying gain settings. Figure 2–2A is an example of overall gain setting being too low, while Figure 2–2B is proper gain setting. Figure 2–2C shows the same image with a gain setting that is too high.

The second postprocessing function of the receiver is compensation, commonly known as time gain compensation (TGC) or depth gain compensation (DGC). Compensation makes up for energy loss from the sound beam due to attenuation. Attenuation is the loss of intensity and amplitude of the ultrasound beam as it travels deeper into the body. Strong returning signals from the near field (close to the transducer) need to be suppressed, whereas signals from the far field (deeper depths) require higher amplification.

TGC/DGC is seen on the machine as a column of toggles that can be manipulated along a horizontal plane. By sliding a toggle to the right, the operator increases the gain at that given depth. The TGC/DGC is normally placed at a diagonal slope of variable steepness in which the upper, or near field, toggles are often set at a lower degree of compensation, whereas the lower, or far field, toggles are often set at a higher degree of compensation. A pattern of gradual change from one toggle to the next avoids a “striped” appearance to the ultrasound picture.

Lateral gain compensation (LGC) is present on only some ultrasound machines. LGC toggles are similar to TGC/DGC toggles but act selectively on the y-axis of the picture to change the gain in vertical portions. LGCs help to bring out myocardial walls that may be hypoechoic (lacking in brightness) due to technique or positioning. However, the operator must be selective in using compensation of the gray scale because some sections of the image may not require any compensation.

Compression is another postprocessing function that is just as important as gain. Compression alters the difference between the highest and lowest echo amplitudes received by taking the received amplitudes and fitting them into a gray-scale range that the machine can display. Dynamic range (how many shades of gray appear within the image) is determined by the degree of compression applied to returning signals. Compression changes the dynamic range of the ultrasound signal with an inverse relationship. Increasing the compression produces an image with more shades of gray, whereas decreasing the compression provides a highly contrasted image with strong white and black components. Sharply decreasing the compression may be a tempting method to improve structure delineation, but it will sacrifice low-amplitude targets. Most presets will start the operator with midlevel value of compression and gain. Figure 2–3 shows the effect of compression on an image of a mitral valve. Figure 2–3A shows a high compression setting so the image has too many low amplitude targets. Figure 2–3B presents a medium compression setting. The image has the optimal gray-scale information. Figure 2–3C depicts a low compression setting, so the image has little gray-scale information.

Power can be indirectly assessed by decibel (dB) settings, mechanical index (MI), and thermal index (TI). Power, expressed in watts (W) or milliwatts (mW), describes the rate at which energy is propagated into the imaged medium. Intensity, power per unit area (mW/cm2 or some unit variation), is a concept that closely relates to power. Intensity may not be entirely uniform throughout tissues, so intensity levels predict more accurately the risk of bioeffects than do power levels. When looking at an image, changing the output power from the transducer appears to have an effect similar to that of changing the overall gain; however, alteration of power is less preferable than increasing the overall gain. Any increase in output power can raise the amount of energy that transfers into the biologic tissue, increasing risk of bioeffects. Adjustment of gain settings do not change the amount of energy transferred into the tissue. Changes in overall gain setting adjust signals that have already traveled through tissues. Thus, gain variables should always be adjusted before ever attempting to change the power.

Most machines do not describe the power level in watts or milliwatts. Instead, they indirectly describe power in terms of decibels. The 3-dB rule, intended to help clinicians understand alterations in power, states that an increase of 3 dB doubles the power or intensity from an original value, whereas a decrease of 3 dB halves the power or intensity from an original value. The exact power or intensity levels are not presented on the display, so manufacturers have placed two more variables on the display screen to help clinicians estimate power and intensity levels. The two variables shown on the display are the mechanical index and the thermal index. The MI conveys the likelihood of cavitation resulting from the ultrasonic energy during the examination. Cavitation refers to activity (oscillation or bursting) of microscopic gas bubbles within the tissues due to exposure to ultrasound energy. The TI is the ratio of output power emitted by the transducer to the power needed to raise tissue temperature by 1° Celsius.

Frequency is defined as the number of cycles per second. The frequency of most TEE probes is between 5 and 7 MHz (5,000,000 to 7,000,000 cycles/s). The frequency of the ultrasound probe can have a dramatic effect on image quality of the TEE. Lower transmitted frequencies will create very different images than higher transmitted frequencies. Resolution is the ability to detect two targets positioned closely to one another. When the wavelength is shorter (higher frequencies), the axial resolution is better. Improved resolution generates a more accurate anatomic rendering of structures displayed by ultrasound. However, the lower the sound source frequency, the more effective it is in penetrating tissue and not falling subject to attenuation. Unfortunately, low frequencies yield poorer resolution in their resulting images when compared with high-frequency images. A TEE can be performed at a higher frequency than a chest wall echocardiogram. This is largely due to the fact that the heart is closer to the probe and there is less interference from bone, lung, and other tissue.

The imaging frequency is dependent on the transducer used; however, each probe has a frequency bandwidth. Frequency bandwidth is the range of frequencies any selected probe is able to transmit. Broadband transducers are commonly used because of their ability to transmit over a wide range of frequencies. Some ultrasound systems offer a wide bandwidth and allow an operator to select a part of the bandwidth spectrum that is most appropriate to make the images. During a TEE, the highest possible frequency should be used to distinguish targets on the display. Such a strategy takes advantage of the close proximity of the heart to the TEE probe and will produce more highly resolved images.

Standard transducers transmit and receive frequencies that are the same or within a very close range. The frequency from such a transducer is known as the fundamental frequency. As the beam propagates through tissues, it distorts and creates additional frequencies that are multiples of the fundamental frequency. These additional frequencies are the harmonic frequencies and are created by the tissues themselves. Whereas the fundamental frequency may undergo a large amount of distortion, the harmonic frequencies do not. In patients with poor sound transmission due to obesity or dense muscle tissue, the harmonic frequencies may produce a better image than the fundamental frequency. In general, it should be unnecessary to use the harmonics during a TEE because of the proximity of the transducer to the heart. In fact, when using harmonics, the image likely will have poorer resolution than the image created with the fundamental frequency. During contrast studies, harmonics will improve image quality whether using agitated saline solution or one of the transpulmonary contrast agents. Figure 2–4 demonstrates the differences between a fundamental low frequency (Figure 2–4A), a fundamental high frequency (Figure 2–4B), and a harmonic image (Figure 2–4C). Notice how harmonic imaging makes the aortic valve appear thicker in the harmonic image compared to the fundamental images.

Because the ultrasound beam does not necessarily maintain the same width as it travels into the depths of tissue, many systems use a technique known as focusing to better evaluate a specific depth. Focusing the ultrasound beam is a process accomplished by mechanical or electronic means and is available as a system control. As the beam proceeds deeper into the tissue, the beam gradually tapers to a narrow region known as the focal zone. The central point of the focal zone is the focal point, an area that will have the highest intensity (mW/cm2) of the transmitted ultrasound energy from the transducer. Of interest to the echocardiographer is the fact that manipulation of focal zones produces a higher quality image because it produces better returning signals for the machine to process and display.

All machines have a control that increases or decreases the overall image depth. Decreasing the depth increases the frame rate by reducing the amount of information that the machine has to process and shortening the time required for the beam to travel to and return from a target of interest. Regardless of frame rate, some cardiac views require a deeper field of view for adequate display. Each machine has a limit as to how deep it is able to image. Although this can become a limitation in transthoracic echocardiograms of extremely obese patients, depth is seldom a limiting factor in acquiring a TEE picture. Figure 2–5 shows an example of the use of deep (Figure 2–5A) and shallow (Figure 2–5B) depth settings.

Reducing the width of the sector is another excellent way to increase frame rate and isolate an area of interest. Just as depth will reduce the amount of information that needs to be processed, so will narrowing the sector. See Figure 2–6 for a display of wide (Figure 2–6A) and narrow (Figure 2–6B) sector images.

All forms of Doppler analysis on the ultrasound machine, including color Doppler, are based on the Doppler effect, the perceived change in frequency that occurs between a sound source and a sound receiver. When a reflector is stationary and the transducer is stationary, no Doppler shift occurs. In the human body, the constantly moving red blood cells serve as the reflector creating the Doppler shift. It should be remembered that color Doppler provides data only on moving reflectors and that stationary reflectors detected within each color packet are eliminated from processing. Color Doppler is commonly layered over a 2D image to provide information on blood flow within the context needed to adequately interpret the color data.

The direction of flow, relative to the transducer, is always shown on the machine's display with a color map. Manufacturers frequently provide a red-and-blue color map, with red designating flow toward the probe and blue representing flow away from the probe. The machine operator can change many color display properties. A color invert option will flip the color map; this would present blue as flow toward the probe, and red as flow away from the probe. Figure 2–7 depicts this color-invert option. The color map can also be changed to display flows in an entirely new set of hues or intensities. One set of color maps are velocity maps, these display Doppler velocities with two preselected colors (typically red and blue). Some color maps add an element of green or yellow to differentiate between laminar and turbulent flows; these are variance maps. While using a variance map, color Doppler analyzes velocities, direction of flow, and areas of turbulent versus laminar flow. Figure 2–8 shows examples of mitral regurgitation by TEE using two different color maps. Figure 2–8A shows a velocity map, a map of direction and velocity. Figure 2–8B is an example of a variance map, showing direction, velocity, and turbulence.

The color mode also has an adjustable scale. The numbers above and below the color map indicate the range of mean velocities that can be displayed, typically in centimeters per second, without aliasing. Usually, the machine calculates an optimal scale based largely on depth. Lowering the scale number lowers the pulse repetition frequency and therefore the Nyquist limit (pulse repetition frequency/2 = Nyquist limit). Although the lower scale is more sensitive to flow, it is more susceptible to aliasing. Raising the scale will reduce sensitivity to flow, raise the pulse repetition frequency and Nyquist limit, and make the color display less likely to alias. Figure 2–9A shows normal diastolic flow through the mitral valve in diastole. In Figure 2–9B, the Doppler scale has been lowered to 15.4 cm/s resulting in distortion of the normal diastolic flow. Altering the scale may be beneficial in select instances, but the scale should be left alone most of the time, and particularly when grading valvular regurgitation.

Quality of color imaging depends on the number of pulses per color packet, or packet size, and on the frame rate of the overall picture with color. Each tiny color packet represents the mean velocity within that particular color packet. The more pulses in the color packet, the more accurate the received data and the color representation, but these variables must be balanced carefully because a larger number of pulses in the color packets also decreases frame rates. The frame rate of the image with color Doppler can be highly dependent on the machine operator. For optimal imaging, the depth should be kept as shallow as possible to have the highest frame rate. The color box also should be as narrow as possible and fully cover the area of interest. Wider color boxes will lower frame rates because more time is required to interrogate a flow in a larger area. In contrast, the length of the color box has a minimal effect on the frame rate. Keyboard controls, such as trackball, size, and position controls, can be used to change the length, width, and placement, respectively, of the color box.

Smoothing determines the degree to which the color packets progressively transition into adjacent color packets. A low smoothing setting will make the individual color packets highly independent of one another and create a speckled impression of color flow. A high degree of smoothing will make the color packets appear as though they were blended together. The appearance of color filling improves with higher levels of smoothing. If the color appears bright and flashy within the color box, it may help to decrease the color gain. Table 2–2 lists the most commonly used controls for adjusting the color Doppler display.

PW Doppler samples velocities at a specific point along the beam axis. A gate designates the sampling point and its position within a 2D image, and the trackball moves the pulse sample gate within the ultrasound image. The disadvantage of PW lies in its susceptibility to aliasing. This type of Doppler is best for low-velocity flows or selecting a specific area to interrogate blood flow.

CW Doppler possesses a clear advantage over PW Doppler in its lack of a Nyquist limit. This means that the high-velocity flows will not alias. However, if the scale on the Doppler waveform display is set too low, it may appear to wrap around in the same manner as PW Doppler. If this occurs, the scale (cm/s) can be increased to the desired velocity range. CW Doppler is not depth specific. While using CW, the ultrasound machine samples all along the beam axis, constantly sending and receiving Doppler signals. Figure 2–10 compares the Doppler spectral traces of a high-velocity jet in a patient with mitral regurgitation. Figure 2–10A shows a PW Doppler trace with the sample site in the mitral valve orifice. The flow is aliased and an accurate peak velocity cannot be measured. In Figure 2–10B, the CW Doppler spectral trace shows no aliasing and the peak velocity of the mitral regurgitation is easy to accurately measure.

PW and CW Doppler placements are straightforward and take very little time to learn. It is of primary importance to have the beam axis at an angle as close as possible to 0° (parallel) with blood flow. On the sector display, the dotted line designating Doppler beam placement is positioned within the area of interest using the position button and the trackball. When using PW Doppler mode, the sample gate can move freely vertically along the beam path. Only flow at the point designated by the sample gate will be displayed. Very small, incremental changes in beam axis or sample gate placement can produce significant changes in pressures, velocities, and patterns of flow. Hypertrophic obstructive cardiomyopathy (HOCM) investigation consistently proves the importance of good sample gate and beam axis position. In CW Doppler mode, an operator can adjust only the focal point of the beam, without eliminating any specific depths from the overall Doppler display.

PW and CW Doppler displays may need to be “cleaned up” before they are acceptable for acquisition. Relevant information should be extracted and refined from the Doppler display. Once Doppler sampling placement is optimal, the resulting Doppler signal can be easily manipulated. Common controls to manipulate the signal include Doppler sweep speed, gain, scale, baseline, compression, and reject. In case Doppler uniformity is preferred, a Doppler invert button is also available. The Doppler invert feature will flip the image vertically along the baseline to display flows toward the transducer below the baseline and flows away from the transducer above the baseline. Even though Doppler invert provides no additional Doppler information, it can make a more aesthetically pleasing comparison of Doppler frames during interpretation.

Sweep speed changes the number and width of Doppler waveforms that can be displayed in a single picture. Increasing the sweep speed effectively zooms in on one or a few Doppler waveforms at a time. High sweep speed is optimal for patients in normal sinus rhythm with multiple uniform Doppler readings over time. Conversely, sweep speed should be reduced for a patient with significant Doppler variations between cardiac cycles or for arrhythmic patients. Other common indications for very slow sweep speed include assessment of tamponade or constriction, for which the presence or absence of respiratory variation needs to be demonstrated. Figure 2–11 is an example of three spectral traces with various sweep speeds.

Gain amplifies or de-amplifies the returning signal from the moving red blood cells. The gain is often preset, but changing gain can compensate for low- or high-density red blood cell movement. The Doppler spectral signal itself should be a midlevel gray, and the Doppler background should be black. If the signal is bright white, Doppler gain should be reduced. If it is very faint or totally black but present, Doppler gain should be increased. Increasing the audio volume to hear the Doppler shift may be helpful in optimizing the spectral trace. Audio volume provides critical information when using Pedoff-style probes, which produce Doppler signals with no 2D image for guidance. Figure 2–12 compares a Doppler spectral trace with too low a gain setting (see Figure 2–12A) with optimal gain setting (see Figure 2–12B). Figure 2–12C is an example of a very high Doppler gain setting.

The scale setting controls the highest and lowest velocity (cm/s) levels that can be presented on the Doppler display. If the baseline is in the center of the Doppler display y-axis, then the highest velocities that can be detected above and below the baseline (toward and away from the probe) should be the same. If a flow velocity is very high or very low, it may be necessary to alter the Doppler scale.

The baseline is a horizontal line showing the zero velocity point over time. Velocities can be shown symmetrically above and below the baseline. The alternative is to eliminate unwanted information by raising or lowering the zero velocity point along the y-axis of the Doppler display to focus attention on flow above or below the baseline. Figure 2–13 depicts the effect of shifting the Doppler spectral trace baseline.

Doppler compression changes the number of gray shades assigned to the spectral display. Increasing the numerical level of Doppler compression creates a softer, smoother, gray Doppler display. A lower numerical level of Doppler compression creates a harsh display with more black-and-white contrast and fewer shades of gray. Figure 2–14 shows an example of changing the compression on the Doppler spectral trace. Figure 2–14A is an example of high Doppler compression setting showing too many low amplitude signals. In Figure 2–14B, the trace displays optimal shades of gray. In Figure 2–14C, the spectral trace is at a lower compression setting.

Reject is a function that can eliminate low-velocity signals, which usually are seen near the baseline. Reject brings the Doppler background to black or close to it. Reject also helps to define the borders of the Doppler signal. When tracing or measuring, the reject should be set at an acceptable level to eliminate baseline noise. Figure 2–15 shows the use of reject on the Doppler spectral trace. Figure 2–15A shows a spectral trace with reject set higher, thus eliminating the background noise. Figure 2–15B shows an example of low reject with more background noise seen on the display.

Doppler measurements are often necessary for a complete cardiac assessment. Relevant controls to accomplish this task include freeze, caliper, trace, enter, erase, and the trackball. Caliper measurements of PW and CW Doppler can be made with the caliper, trackball, and enter controls, and typically are used for machine calculations of instantaneous velocity and pressure. Traced Doppler measurements, made by using the trace, trackball, and enter buttons, provide information on peak velocity, and peak and mean pressure gradients. The erase button may be pressed one or more times to remove one or all Doppler measurements from the image waveforms. Figure 2–16 shows examples of typical Doppler measurements. Figure 2–16A displays a simple caliper measurement of an aortic stenosis Doppler velocity and peak gradient. Figure 2–16B shows the same Doppler spectral trace with more advanced measurements performed under the system's analysis package.

Doppler measurements can be recorded inside or outside the machine analysis package. The analysis package on any machine is intended to label, identify, and present measurements in an organized manner. Calculations can be simplified: once certain Doppler measurements are accumulated, the machine may be able to compute additional helpful values by using previously programmed ultrasound equations. Table 2–3 lists the most commonly used controls for PW and CW Doppler variables.

Zooming an area of interest is another good way to isolate a part of cardiac anatomy or to remove unwanted information from the display sector. Zooming trims from the x- and y-axes of the sector, increases frame rate, and increases spatial resolution. Zoom can be selected on the machine and an initial zoom box will appear. Zoom parameters for the box can be changed by using the size and position buttons, and the trackball until the area of interest is adequately covered. Figure 2–17A shows a normal, full-sector-size image. Figure 2–17B shows a zoomed view of the same image, focusing on the left atrial appendage.

Although the freeze button is fairly simple and straightforward, it should not be underestimated. When measuring 2D, PW, or CW images, an operator must first freeze the image and sometimes scroll backward or forward in time to obtain the correct frame. Individual frames can be compared and scrolled as needed. In TEE, the freeze button should be used whenever the probe is not actively needed (eg, when introducing the probe into the esophagus, after completion of the examination, or during cardiopulmonary bypass). Freezing the image stops probe sound transmission, thus decreasing the chance of the probe overheating.

All echocardiographic images can be measured and quantified. Buttons used for gray-scale assessments may include freeze, caliper, trace, enter, and erase. Measurements of 2D images can be made outside or inside the analysis package. If measurements are made within the analysis package, the operator will need to select the analysis or measurement package button and then the proper name for the subsequent measurement. A pair of calipers is provided for exact linear measurements of gray-scale images. The first caliper has to be entered onto a point to have the second caliper appear. Once the second caliper is set, then the machine calculates the linear distance between the two points. Multiple linear caliper measurements can be made. Calipers are removed by using the erase button. To obtain the area or circumference of an object, the trace, trackball, and enter buttons are commonly used. On some ultrasound machines, once the trace function is selected, the first given caliper is set to a starting point and the trackball is used to follow the borders of the desired region. When the border is outlined and the dotted tracing line is returned to the starting point, the outline must be entered before the machine will calculate any area. The operator has the ability to backtrack a traced line or completely remove any given trace by using the erase function.

Annotations are labels that are in the form of text or pictures; they can be used to accompany an image for communication. Nonstandard views, unusual anatomic variations, or highly edited standard views may need some further explanation. It is also helpful to annotate when a standard view is not available or is of very poor quality. Naming anatomic landmarks, body positioning of the patient, relationships to other anatomic points of reference, or timing of events is often helpful for the interpreting physician. For example, it may be useful to label an image as a pre- or postsurgical event. Applied comments should be removed after the image is recorded so that later pictures are not confusing as a result of incorrect labeling.

Three-dimensional TEE is a new diagnostic tool that has already proven to be helpful in the preoperative assessment of mitral valve disease. This is especially true for mitral valve prolapse patients being considered for repair rather than replacement. Once a 3D TEE image is obtained, it can be rotated and cropped in order to display specific anatomy, and then these images can be saved as still pictures or moving loops. Figure 2–18 is an example of a flail mitral valve imaged with a 3D TEE probe pre- and post-repair. The 3D has been cropped and rotated into the “surgeon's” view so that the mitral valve is viewed from the left atrium down towards the ventricle.

The 3D TEE image can also be saved as a dataset from which additional images can be derived after the study is completed. Emerging software is available for on-line or off-line analysis of these 3D datasets, and can provide additional information on valve size and structure. Figure 2–19 is an example of this type of derived information obtained from a 3D TEE dataset. Further research is required to verify the clinical utility of these data.

Image quality and diagnostic accuracy depend heavily on the knowledge and skill of the clinician operating the ultrasound machine. Awareness of the basic knobs associated with 2D, color Doppler, PW Doppler, and CW Doppler is absolutely essential for performing a good TEE examination. In addition to the controls for adjusting a 2D TEE image, 3D TEE adds a variety of tools for manipulating 3D data sets. Although simple display manipulation may seem difficult at first, image optimization becomes second nature with time and experience.