Point-of-care ultrasound (PoCUS) refers to ultrasound imaging performed by the treating clinician to augment the physical exam for diagnosis and procedure safety when consultative imaging is unavailable or impractical. Like other clinical skills, the scope of PoCUS should directly reflect individual skill level and training. Still, PoCUS is not a perfect tool, and clinical context or pretest probability will affect false-positive and false-negative rates. Therefore, clinicians should understand how to integrate PoCUS findings with other available data, including other POCUS findings, to ultimately improve diagnostic certainty and optimize patient management.
Ultrasound systems consist of a display, a set of controls, a computer, and a transducer or ultrasound probe usually attached with a cable. Laptop-sized systems are portable but maintain a large screen and set of controls. Transducers and cables can be fragile and expensive. Users should take care to avoid twisting or rolling over the cord, dropping the probe, or damaging it with sharp objects like needles or alcohol-based antiseptics.
There are now numerous handheld systems available, some of which integrate with a smartphone or tablet. Transducers with integrated batteries can be bulky and awkward to manipulate. In addition, using them with a larger tablet can be challenging to hold and control with the same hand. Still, compared to larger systems, handheld devices are more portable, cheaper and incorporate newer technologies that make the transducers and cables more robust. In addition, multiple manufacturers now compete in this growing market, so the price, usability, and functionality of handheld devices should continue to improve.
Image 1.1 Built to meet US military specifications for austere settings, the laptop-sized Sonosite M-Turbo ultrasound system remains a popular choice for hospital wards, emergency departments, and global aid organizations. All but three of the images in this textbook were obtained with this system.
Ultrasound transducers use an array of piezoelectric crystals that expand and compress in response to an alternating electric current. This oscillation generates a pulse of high-frequency sound that propagates into underlying tissue. Immediately, echoes of the pulse start to reflect from various points in the tissue beneath the transducer. Echoes occur whenever the pulse gets to a change in tissue density and stiffness, also known as acoustic impedance. A more significant change reflects a larger portion of the original pulse. These echoes return to the transducer causing the same piezoelectric crystals, which are in listening mode between pulses, to oscillate again and generate an alternating current, which the computer interprets into an image. Any energy in the ultrasound wave that does not reflect back is either scattered, absorbed as heat, or transmitted deeper. Air has a much different density and stiffness than the ultrasound probe and most tissues. Therefore, a transition to air will reflect almost all of the pulse. This is why it is necessary to use a gel or liquid between the transducer and the tissue surface to displace any air and allow the high-frequency sound waves to pass through. Wray et al. published a blinded comparison of alternatives to ultrasound gel and concluded that water was non-inferior to commercial ultrasound gel.1 It is often adequate to use a damp paper towel or gauze when ultrasound gel is unavailable, contaminated, or expired. Do not neglect to clean the transducer and use a sterile gel/liquid and barrier whenever scanning to guide an invasive procedure.
In B-mode or brightness mode, the computer makes a 2D grayscale image by plotting brightness or echogenicity based on how strong the echo was for each pixel on a grid. The x-axis represents which crystal or beamline created and heard the echo, and the y-axis represents how much time passed after the pulse went out such that time equals depth. Higher frequency sound with shorter wavelengths will improve axial resolution (y-axis), albeit at the expense of more attenuation since higher frequency sound energy is more easily absorbed into heat. Incorporating more crystals or beamlines can improve the lateral or horizontal resolution, but at the expense of temporal resolution since it will take longer to generate a single frame as each beamline pulse and listening cycle occurs sequentially.
Image 1.2 Using one beamline (orange arrow) after another, B-mode imaging plots consecutive columns of pixels to create a 2D grayscale image. The brightness of each pixel in a beamline is determined by how strong the echo is over time, such that later echoes determine the brightness of lower/deeper pixels. This image reveals bright pixels corresponding to a strong echo from a bone surface 2-3 cm from the tissue surface (white arrow).
Almost no echoes occur as sound travels through homogeneous tissue, like fluid, because density and stiffness are uniform throughout. As a result, fluid appears black or anechoic in B-mode. In contrast, more heterogeneous tissue that varies in density and stiffness reflects and scatters more of the ultrasound pulse and produces stronger echoes that are bright or hyperechoic.
Denser tissue like bone absorbs more sound energy into heat. Therefore, using a lower frequency transducer and scanning through an acoustic window of less dense, more homogenous tissue, such as the intercostal space, the bladder, or the liver, will help image deeper structures. In addition, if scanning from the flanks, you can turn the patient into a lateral decubitus position to displace overlying fatty tissue.
The relative angle of the reflective surface also influences how bright or echogenic the tissue appears. Perpendicular angles maximize echogenicity, especially with smooth flat surfaces like tendons, muscle fibers, bone, nerve sheaths, or a metal needle. These smooth surfaces may have few discernible echoes but instantly become hyperechoic when perpendicular to the beamline. This phenomenon is called anisotropy.
Image 1.3 Counterintuitive movements may be necessary to get beamlines through an intercostal space and optimize the acoustic window.
After finding and optimizing the acoustic window, which may involve repositioning the patient or having them adjust their lung volumes, the next step is to fine-tune the image with specific transducer movements. It is helpful to consciously separate these movements such that only one viewing plane is adjusted at a time.
Image 1.4 Subcostal cardiac view showing distinct transducer movements: sliding, rotating, rocking, and fanning. Sliding can be further distinguished as 'sliding' when moving along the long axis (shown here) and 'sweeping' when moving along the short axis.
The ultrasound system’s power or volume setting regulates the total energy output per ultrasound pulse. This soundwave energy generates heat when tissues absorb it. Higher frequencies are absorbed more easily, and denser tissues like bone absorb more of the pulse’s energy. In addition, ultrasound waves can expand and contract microbubbles that may reside within tissues. This mechanical effect can generate even more heat and tissue injury. However, clinical ultrasound is relatively safe because most machines cap the power setting to avoid significantly heating tissues or disturbing microbubbles during regular use. To ensure the settings are safe, always scan a specific area using the appropriate mode or preset on the machine. Be especially careful when examining sensitive structures like the eye or developing fetus/embryo. As a rule, avoid power or pulse-wave Doppler when scanning a fetus/embryo or a patient’s eye. Do not press on an injured eye; avoid scanning altogether if a globe rupture is suspected. Never hold the transducer in a fixed position for any longer than necessary. Instead, take the transducer off the patient when real-time imaging is not required. In doing so, your practice will be consistent with the established "As Low As Reasonably Achievable" or ALARA principle.2
Image 1.5 Transducer types that are commonly used in PoCUS.
The linear array transducer uses a flat line of crystals to form a characteristic rectangular image. It uses higher frequencies (5-15 MHz) to create detailed images of superficial structures such as bones, joint spaces, tendons, blood vessels, and other subcutaneous tissue. The curvilinear, convex array or abdominal transducer has crystals arranged in a curved line. It uses relatively low frequencies (2-5 MHz) to image deep structures such as the liver, gallbladder, aorta, kidney, uterus, and ovaries. The phased array transducer or cardiac probe (1-5 MHz) is used to image the heart through intercostal spaces by cycling through phases that focus and steer the beamlines in different directions to create a fan-shaped image.
Image 1.6 The transducer or probe marker (A) corresponds to the orientation marker (B) on the B-mode image. An additional narrative label or graphic pictogram (C) is also essential to orient those reviewing an image after the exam.
The palpable probe marker is used to orient the transducer with the corresponding marker on the image. By convention, unless a standard view of an oblique organ like the heart or indicated otherwise as out-of-plane, all ultrasound images should be oriented with the marker towards the patient's head in longitudinal views (coronal plane from side or sagittal plane from front). Thus, the probe marker on the left side of the image would be towards the patient's head. In transverse views (axial plane), images should be oriented with the probe marker towards the patient's right. Thus, the probe marker on the left side of the image would be towards the patient's right. However, when using ultrasound for real-time procedure guidance, keep the marker towards the operator's left to correspond to the left side of the screen.
All cardiac images should be oriented to achieve the standard short-axis and long-axis views regardless of the exact orientation of the transducer. By convention, the marker will switch to the right side of the image in cardiac modes. In addition, saved images should be adequately labeled with a pictogram and text to clarify the location, orientation, exam indication, examiner, and at least some identifier so you can later tie the images to a specific patient. It may be helpful also to include contextual information such as the patient's blood pressure at the time, especially when examining the heart.
Parameters that affect ultrasound image quality and interpretation include the following.
Machines have various presets or scanning modes, such as cardiac or abdominal for each transducer type. Each mode changes adjustments like frequency, focus, frame rate, image orientation, and specific post-processing effects, which may not be individually adjustable. Be aware that scanning modes not designed for ocular or fetal scanning may expose these sensitive tissues to a higher mechanical index or thermal index than is considered safe. Therefore, always use the appropriate scanning mode.
Depth will simultaneously adjust the frequency (axial resolution & attenuation) and the interval between each 'send and listen cycle' or pulse repetition frequency (PRF), impacting the frame rate. A higher frame rate helps distinguish changes in tissues over time, such as valve leaflet movement. To maximize frame rate and resolution, reduce depth until the area of interest encompasses 80-90% of the image.
Image 1.7 This parasternal long-axis view of the heart improves significantly when excess depth (A) is minimized while still keeping the whole heart in the image (B).
Gain is akin to adjusting the traditional film speed or the ISO setting on a digital camera. Increasing gain can improve detail in a relatively hypoechoic region of the image. However, more gain can also decrease detail in the hyperechoic areas. The goal is to find the sweet spot for the structure of interest in the image. Avoid the mistake of increasing gain to brighten the image when scanning in a bright room or viewing a dim screen. Always optimize the machine's screen brightness and ambient light in the room.
Image 1.8 The left image (A) was obtained with inadequate gain, and no dissection is visible despite increasing brightness after the clip was saved. When the gain setting is increased, a dissection of the same abdominal aorta becomes visible (B).
Some machines let you adjust the frequency. For example, the Sonosite M-Turbo has three settings, “RES” for a higher frequency, “GEN” for a medium frequency, and “PEN” for a lower frequency. Lower frequency settings may improve the view of tissues that are deeper than usual, such as the heart in a subcostal view in an obese patient.
Image 1.9 Changing from GEN (A) to the PEN (B) mode on the Sonosite M-Turbo system uses a lower frequency to decrease attenuation and improves this subcostal four-chamber view.
M-Mode or 'motion' mode displays echoes over time for a single beamline. It allows you to measure movement on a single line with high temporal resolution to measure rapid motion. It is also helpful for displaying motion or lack of motion in a single image instead of a video clip. Remember to minimize the duration of scanning a fetus. This method is quick and safe. It should be the only mode used to assess fetal heart rate.
Image 1.10 M-Mode is used to measure fetal heart motion with a high temporal resolution for accurate HR measurement.
Color Doppler depicts motion and velocity on a 2D image within an overlaid 'color box' that displays pixels of various colors based on detected motion. The computer interprets the frequency shift between the send and the receive signal to determine the direction and velocity of blood flow or tissue motion. The magnitude of the Doppler shift decreases as the velocity becomes more perpendicular to the beamlines. Therefore, try to align the beamlines so they are parallel to the blood flow or tissue motion of interest. By convention, blue signals flow away and red towards (BART), and lighter colors signal faster movement. Color Doppler is essential for detecting abnormal heart valve function and identifying blood vessels.
It is critical to optimize image depth and B-mode gain before entering color Doppler mode. Once in color Doppler mode, the gain adjustment will adjust the color gain. Inappropriately high color gain can create false color signals in adjacent stationary tissues. Color flash artifacts also worsen with increasing color gain. These are spurious Doppler signals arising from transducer or whole-tissue movement. Anechoic structures are more prone to producing color flash artifacts. Therefore, avoid mistaking whole-tissue movement caused by respiration or transducer movement from actual flow inside a hypoechoic structure such as a gallbladder or blood vessel.
Image 1.11 Color Doppler mode detects mitral regurgitation with rapid abnormal backflow across the valve. Blue signals flow away and red towards (BART).
Acoustic shadowing occurs when superficial tissues attenuate the sound more than usual, such that minimal ultrasound energy can get to deeper tissue and back. For example, ribs, stones, prosthetic materials, and air in the bowel or lung commonly cause acoustic shadowing.
Image 1.12 Sliding the transducer so the beamlines run between the ribs (B) removes the acoustic shadowing that darkens much of the image (A).
Acoustic enhancement is the opposite of shadowing. More superficial hypoechoic structures, especially anechoic fluid like a full bladder, create less attenuation than usual, making underlying tissues appear brighter than expected.
Image 1.13 Scanning the pelvis with a full bladder pushes the bowel out of the way and enhances deeper pelvic structures of interest. Sometimes, reducing gain can optimize the view of acoustically enhanced deep structures of interest.
Reverberation artifact occurs when some of the ultrasound pulse can bounce back and forth between highly reflective surfaces. Some energy from each reverberation gets back to the transducer after each cycle, which appears as equidistant lines deeper than the reflective surface. These artifacts fade and taper at increasing depths. They are seen below the pleura-air interface in the chest (A-lines) and below lung surface irregularities, metallic needles, catheters, hard debris, or calculi (commit tail artifacts).
Image 1.14 A-lines below the pleura-lung interface are normal reverberation artifacts that suggest the lung air-to-tissue/fluid ratio is at or above normal.
Edge artifact occurs deep to the outer edges of hypoechoic structures like large blood vessels. Ultrasound waves refract or bounce off the edge, leaving a dark, blurred line as relatively few echoes return from directly deep to the edge. The gallbladder can have these edge artifacts, which should not be confused with acoustic shadowing from dense gallstones.
Image 1.15 Example of edge artifacts with shadows coming off the edge of the IVC and hepatic blood vessels (arrows).
Mirror image artifact usually occurs around the heart or diaphragm where ultrasound reflects off the hyperechoic border, hits an object like a cardiac valve or hyperechoic liver lesion, bounces back to that hyperechoic surface, and then reflects up to the transducer. The delay plots a faint mirror image of the object distal to the hyperechoic surface.
Image 1.16 This parasternal long-axis view of the heart is an example of a mirror artifact where ultrasound reflects off the lung surface and back again after hitting echoic parts of the heart. These echos eventually get back to the transducer and show up as a mirror reflection of the heart below the pericardium-lung interface.
Ringed-down artifacts occur when ultrasound gets inside small bits of fluid or tissue surrounded by air. Pulmonary B-lines are ringed-down artifacts, which are a sensitive marker of increased lung tissue/fluid-to-air ratio. In this example, ultrasound energy gets into lung tissue, and resonant frequencies bounce around for a continuously variable duration before returning to the transducer. The continually returning echoes produce a continuous hyperechoic line (light-ray like), which goes from the pleural line to the bottom of the image without fading and moves back and forth with lung sliding. Similar ringed-down artifacts are seen in gaseous stomach content and subcutaneous air that might accumulate with a traumatic pneumothorax or severe soft-tissue infection.
Image 1.17 Lung ultrasound views with a transducer (A) and a curvilinear transducer (B). Both views reveal multiple B-lines, which start at the pleural line and emanate to the bottom of the image. B-lines are examples of ringed-down artifacts.
Interference artifacts can occur when nearby equipment interferes with the ultrasound system. This interference usually creates repetitive geometric patterns and makes image interpretation impossible. If this occurs, try unplugging the machine, using a different power outlet, or turning off nearby electrical equipment. A second transducer, used simultaneously on the same patient, can also generate this interference artifact.
Lastly, ultrasound systems assume the sound propagation speed is 1540 meters per second. However, different tissues have slightly different sound propagation speeds, as sound travels faster through stiffer tissue. For example, the speed of sound in air is 330 m/s, fat is 1450 m/s, water is 1480 m/s, and bone is 4080 m/s. This discrepancy leads to speed displacement or propagation velocity artifact, where the depth of echoes traveling through a significant amount of fluid or fat will appear slightly deeper than their actual depth.