On completion of this chapter, you should be able to:
Distinguish normal from abnormal intracranial arterial anatomy
Understand the physiology and hemodynamics within the circle of Willis
Outline the technical aspects and proper instrumentation control settings used to perform a high-quality transcranial Doppler imaging examination
Describe the characteristics of the spectral Doppler waveforms obtained from arteries evaluated during a transcranial Doppler examination
List the clinical applications of transcranial Doppler imaging and describe expected findings and diagnostic criteria associated with each
Discuss common errors associated with performing and interpreting a transcranial Doppler imaging examination
As discussed in the previous chapter, noninvasive Doppler evaluation of the extracranial vasculature has become a reliable and effective method for detecting arterial disease. However, the development of a noninvasive method to interrogate the intracranial arterial system has suffered in comparison because of the attention focused on surgically correctable lesions of the carotid bifurcation and the difficulty in penetrating the skull using ultrasound technology.
Fortunately, technical sophistication has progressed and experience has been gained in the area of transcranial Doppler (TCD) imaging, since its inception into clinical use in 1982.
TCD imaging was first used as a method to detect cerebral arterial vasospasm after subarachnoid hemorrhage, but has since been used in a wide variety of clinical applications ( Box 38-1 ). With its continued use and growing popularity as a noninvasive diagnostic tool, a better understanding of intracranial arterial hemodynamics will be gained by using TCD in many different clinical settings. Although TCD imaging can provide valuable information regarding intracranial circulation, it is a technically difficult imaging modality to perform; therefore thorough knowledge of anatomy, physiology, and pathologies of the intracranial arterial system are required. Furthermore, accurate interpretation of a patient’s TCD examination is not possible without knowledge of the amount and location of atherosclerotic disease in the extracranial vasculature. Carotid and vertebral duplex imaging should be performed before the TCD examination, as extensive extracranial disease may cause changes in the velocity profile or direction of blood flow in the intracranial arterial system.
Assessment of intracranial collateral pathways
Detection of cerebral emboli
Detection of feeders of arteriovenous malformations
Diagnosis and follow-up of intracranial vascular disease
Documentation of the subclavian steal syndrome
Evaluation of the hemodynamic effects of extracranial occlusive disease on intracranial blood flow
Evaluation of migraine headaches
Evaluation of the vertebrobasilar system
Monitoring vasospasm in subarachnoid hemorrhage
Monitoring evolution of cerebral circulatory arrest
Monitoring during anticoagulative or fibrinolytic therapy
Monitoring after head trauma
Screening of children with sickle cell disease
Intracranial arterial anatomy
Blood supply to the brain is provided by the internal carotid (anteriorly) and vertebral (posteriorly) arteries, both of which originate extracranially and terminate intracranially. Familiarity with the anatomy of the large intracranial arteries and the arteries that compose the circle of Willis are prerequisites when it comes to performing accurate TCD imaging studies. Therefore this chapter will focus on the intracranial portions of the carotid and vertebral arteries and the arteries that comprise the circle of Willis ( Figure 38-1 ). The extracranial portions of the internal carotid and vertebral arteries are discussed in Chapter 37 .
Internal carotid artery
The internal carotid artery (ICA) is divided into four main segments: (1) the cervical ICA originates at the common carotid bifurcation (carotid bulb) and ends as it enters the carotid canal of the temporal bone at the base of the skull; (2) the petrous section of the ICA begins at the entrance of the carotid canal within the petrous portion of the bone and continues until it traverses the cranial portion of the foramen lacerum and passes into the cavernous sinus; (3) the cavernous segment extends from the foramen lacerum and cavernous sinus entrance to just medial of the anterior clinoid process; and (4) the supraclinoid portion of the ICA enters the intracranial space at the anterior clinoid and continues to its termination where it bifurcates into the middle cerebral and anterior cerebral arteries. During a TCD examination, the terminal portion of the ICA, just proximal to its bifurcation, and the more proximal carotid siphon are evaluated. The carotid siphon is an S -shaped curve in the ICA formed by a posterior and then anterior bend. This begins in cavernous segment and continues to the ICA bifurcation. The internal carotid siphon is a common site of atherosclerotic disease in adults.
The ophthalmic artery (OA) originates as the first branch of the ICA just distal to the cavernous sinus. The OA travels anterolaterally and slightly deep through the optic foramen to perfuse the globe, orbit, and adjacent structures. This artery has three major groups of branches: (1) the ocular branches, (2) the orbital branches, and (3) the extraorbital branches. The branches of the OA often play an important role in collateral pathways that form as a result of disease of the internal or external carotid arteries. The OA is evaluated during a TCD examination .
The vertebral arteries are large branches of the subclavian arteries. The left vertebral artery is dominant in approximately 50% of individuals, the right in approximately 25%, and codominant in the remaining 25%; that is to say, vertebral artery size asymmetry is common. The vertebral artery is divided into four segments: (1) extravertebral, (2) intervertebral, (3) horizontal, and (4) intracranial. The intracranial portion begins as it pierces the dura and arachnoid immediately below the base of the skull at the foramen magnum. It continues anterior and medial to the anterior surface of the medulla and unites with the contralateral vertebral artery to form the basilar artery. Several major branches arise from this segment of the vertebral artery, with the posterior inferior cerebral artery (PICA) being the largest, and commonly arising approximately 1 to 2 cm proximal to the confluence of the two vertebral arteries to form the basilar artery. During TCD examinations, it is the intracranial segment of the vertebral artery that is evaluated. The PICA along with another branch of the vertebral artery, the anterior spinal artery, can sometimes be visualized during a TCD examination.
The basilar artery (BA) is evaluated during TCD imaging and is formed by the union of the vertebral arteries at the lower border of the pons (pontomedullary junction). The basilar, when combined with the vertebral arteries, is often referred to as the vertebrobasilar system, which perfuses the posterior portion of the circle of Willis. From its origin, the BA extends anteriorly and superiorly and bifurcates into the paired posterior cerebral arteries. There are several branches of the BA, including the anterior inferior cerebellar arteries, the internal auditory (labyrinthine) arteries, the pontine branches, and the superior cerebellar arteries, just proximal to the posterior cerebral arteries. The BA is often variable in its pathway, size, and length; is typically tortuous; and may be duplicated or fenestrated.
Circle of willis
The circle of Willis was first described in 1664 by Thomas Willis, characterized by the arterial anastomoses at the base of the brain. The circle is composed of the A1 segments of the two anterior cerebral arteries, anterior communicating arteries, posterior communicating arteries, the terminal portions of the ICAs, and the P1 segments of the two posterior cerebral arteries. These intracranial arteries form a polygon vascular ring at the base of the brain that permits communication between the right and left cerebral hemispheres (via the anterior communicating artery) and the anterior and posterior systems (via the posterior communicating artery). These communications are important when there is significant disease or occlusion within one of the major cervical arteries, as they serve as compensatory perfusion mechanisms. Variations in the circle of Willis are common, as it is estimated that an anatomically complete (“classic”) circle of Willis is present in only one third of the population; however, physiologically adequate circle of Willis exists in approximately two thirds. Significant hypoplasia and absence of the posterior communicating artery, the anterior communicating artery, the A1 segment of the anterior cerebral artery, and the P1 segment of the posterior cerebral artery are the most common variations.
Middle cerebral artery
The middle cerebral artery (MCA) is the larger terminal branch of the ICA. From its origin, the MCA extends laterally and horizontally in the lateral cerebral fissure. The horizontal segment may course superficial or deep. The MCA either bifurcates or trifurcates before the limen insulae (a small gyrus), where the branches turn upward into the Sylvian fissure, forming its genu (“knee”). The vessels travel around the island of Reil, which is a triangular mound of cortex, and run posterosuperiorly within the Sylvian fissure. The terminal branches of the MCA anastomose with the terminal branches of the anterior cerebral and posterior cerebral arteries.
The MCA can be divided into four segments (M1 to M4). The main horizontal section of the MCA, from its origin to the limen insulae, is the M1 segment. The M1 segment gives rise to numerous small lenticulostriate branches. The M2 segment is composed of the branches overlying the insular surface in the deep Sylvian fissure. The M1 segment and the origin of the M2 segment are evaluated during TCD imaging. The initial MCA bifurcation is a common site for intracranial aneurysmal formation. The MCA may also be the site of arterial stenosis and/or occlusion.
Anterior cerebral artery
The anterior cerebral artery (ACA) is the smaller of the two terminal branches of the ICA. From its origin, the ACA courses anteromedially over the optic chiasm and the optic nerve to the interhemispheric fissure (longitudinal cerebral fissure). The proximal horizontal portion of the ACA is known as the A1 segment and is connected to the contralateral A1 segment via the anterior communicating artery (ACoA; see following section). The A1 segment is evaluated during a TCD imaging examination and serves as a midline marker. The contour of the A1 segment may take a horizontal course, ascend, or slightly descend. The complete absence of the A1 segment is unusual. An anomalous origin of the ACA is rare, and asymmetry between the bilateral A1 segments is uncommon. A direct inverse correlation exists between size of the A1 segment and the size of the ACoA. A small or hypoplastic A1 is typically found in conjunction with a large ACoA, as the contralateral A1 segment supplies most of the blood flow to both distal ACA territories. It has been well documented that individuals with anomalies of the A1 segment have a higher incidence of ACoA aneurysms. Stenosis or occlusion of the ACA may be found, but this has been found be less common than in other intracranial vessels.
Distal to the ACoA, the ACA angles superiorly and travels in the interhemispheric fissure. The ACA curves anterosuperiorly around the genu of the corpus callosum. The segment of the ACA extending from the ACoA to the distal ACA bifurcation (callosomarginal artery and pericallosal artery) is termed the A2 segment. The proximal portion of the bilateral A2 segments may be visualized in some patients during TCD imaging and should be documented if possible. The distal A2 segments anastomose with branches of the posterior cerebral arteries. A large medial striate artery (recurrent artery of Heubner) is a major branch of the proximal A2 segment in approximately 20% of individuals, but this artery can also originate from the distal A1 segment (approximately 15%), or from the ACA/ACoA junction (approximately 60%).
Anterior communicating artery
The anterior communicating artery (ACoA) is a short vessel that connects the A1 segments of the ACAs at the interhemispheric fissure. The ACoA is typically a single vessel, but it may be duplicated, absent, or a multichanneled system. The ACoA is typically short in length; however, this has been found to be variable. Longer ACoAs are found to be curved, tortuous, and/or kinked. The ACoA is often the location for congenital anomalies and is the most common site for intracranial aneurysm formation (25%), as well as the most common site for aneurysms associated with subarachnoid hemorrhage. A Doppler waveform of the ACoA may be captured at midline, but it cannot be visualized during TCD imaging.
Posterior cerebral arteries
At the approximate level of the ponto-mesencephalic junction, the basilar artery terminates by bifurcating into the bilateral posterior cerebral arteries (PCAs). From their origin, each PCA travels anterolaterally to perfuse the posterior portion of the occipital lobe. The portion of the PCA extending from its origin to its junction with the posterior communicating artery (PCoA) can be referred to as the precommunicating portion, but is more commonly termed the P1 segment. The P1 segment is evaluated during TCD imaging. Throughout this segment is the origin of many perforating branches that serve to perfuse the brainstem and thalamus. The portion of the vessel extending posteriorly from the PCoA to the posterior aspect of the midbrain is the P2 segment of the PCA, which may be visualized during TCD imaging. The proximal portion of the PCA is typically asymmetric. In cases of “fetal” origin, the P1 segment is hypoplastic or smaller than the PCoA. It is uncommon for occlusive disease to be limited to the PCA, but if it does occur, the P2 segment is most commonly affected. The PCA is further divided into the P3 and P4 segments, but these cannot be visualized during TCD imaging.
Posterior communicating artery
The posterior communicating artery (PCoA) is a paired artery that travels posteriorly and medially to provide a connection between the ICA and PCA. The PCoA is highly variable in size and may angle upward or downward. Hypoplasia is a common PCoA anomaly, which occurs in an estimated 25% of individuals. However, the PCoA can be enlarged in circumstances of a hypoplastic posterior cerebral artery, which occurs in 10% to 20% of cases. This is termed a fetal origin of the posterior cerebral artery. The PCoA does not generally function as an important collateral pathway except in the presence of extensive extracranial occlusive disease bilaterally or absence of a patent ACA. The PCoA is typically evaluated during TCD imaging.
Intracranial arterial physiology
The circle of Willis provides a communication between the internal carotid, external carotid, and vertebrobasilar systems. On its discovery, Thomas Willis stated its function to be a compensatory mechanism in the case of carotid or vertebral stenosis, a belief that is still largely accepted today. , In essence, the circle of Willis provides multiple communications between intracranial arteries and collateral pathways to ensure that the cerebral perfusion remains intact and the brain is supplied with an adequate amount of oxygenated blood, even in the presence of a flow-limiting lesion. Hemodynamic principles teach us that distal to a significant stenosis exists turbulent and low-velocity blood flow. In cases of stenosis proximal to the intracranial arterial system, an inadequate amount of oxygenated blood may be reaching the area of the brain that is perfused by the stenosed artery. It is in this situation that the arteries composing the circle of Willis may shunt blood and be used as a collateral pathway to provide arterial blood to the area that is in need of it. The same idea remains true in situations of artery occlusion, in that collateral pathways are used to divert blood to preserve adequate and entire brain perfusion.
An alternative and somewhat overlapping theory explains that the circle of Willis serves to dissipate areas of elevated pressure within the intracranial arterial system caused by distal significant stenoses. As mentioned, a significant stenosis causes turbulent and low-velocity blood flow distal to the narrowed segment. This sudden deceleration of blood flow within a closed system causes the kinetic energy previously possessed by the blood to be transferred to the arterial walls, resulting in a propagating shock wave toward the brain. Therefore the circle of Willis serves to absorb this shock wave, by transferring the pressure it creates to another lower-pressure compartment in the intracranial arterial system.
Technical aspects of transcranial doppler imaging
Transcranial Doppler imaging is used to investigate the intracranial arterial circulation and has been referred to as one of the most complex and in-depth physiologic tests in vascular medicine, as it requires a thorough understanding of intracranial vascular anatomy, physiology, and pathology. To obtain consistently reliable studies with TCD imaging, the operator must appreciate the importance of proper patient positioning, use the available anatomic landmarks that are important for accurate identification of the intracranial arteries, and be knowledgeable about proper use of the instrument’s controls. The accuracy of the examination will be raised by using the gray-scale image and the color display to guide the TCD evaluation. The outcome of a TCD examination are spectral Doppler waveforms representing the flow characteristics of the intracranial vasculature. Analysis of these waveforms provides information regarding their hemodynamic properties and any fluctuations from normal physiologic flow.
Instrument controls and control settings vary depending on the manufacturer of the color Doppler imaging system. Therefore it is important for the operator to be familiar with the particular imaging system being used. Spectral Doppler waveforms obtained are typically saved digitally and archived, but it may also prove beneficial to document the audio signal of the waveforms being evaluated. Audio recordings are also typically stored digitally.
TCD imaging is performed using a low-frequency (4 MHz) phased-array imaging transducer that, after shifting the frequency within the bandwidth, is operating between 2 and 3.5 MHz. High-quality TCD imaging depends on proper adjustment of several instrument controls. This includes gray-scale (GS), color Doppler (CD), and spectral Doppler (SD) imaging parameters. A deep knowledge of ultrasound physics is required in order to understand how adjusting different ultrasound controls will independently affect the image and how they will affect each other. Improper settings may result in a false finding of the examination.
Instrument controls to consider when adjusting the GS image during TCD imaging are frequency, sector width, image depth, overall gain, time gain compensation (TGC), focal zone number and placement, frame rate, and dynamic range. Proper GS optimization is key as B-mode imaging plays an important role in performing an optimal TCD examination.
CD imaging is also very important while performing TCD, and its instrument controls should likewise be carefully optimized. Instrument controls to consider when adjusting the CD display during TCD include CD gain, scale (pulse repetition frequency [PRF]), wall filter, sensitivity (ensemble length), and persistence. As previously mentioned, CD imaging is integral to the TCD examination, as it can be used to visualize anatomic landmarks within the brain, as well as identify the location of certain intracranial vessels. Conventional CD scale orientation for TCD examinations uses the red portion of the scale to represent blood flow toward the transducer and the blue portion to indicate flow moving away from the transducer. By keeping this color assignment constant, intracranial blood flow direction in the arteries can be readily recognized. While imaging, the CD box, also known as the region of interest (ROI), is moved to the area of being examined on the GS image. During TCD imaging, the entire circle of Willis can often be captured within a small color ROI. It is important to consider the size of the color ROI while scanning, because if it becomes too large, the imaging frame rate will decrease and hinder the performance of the examination. Therefore it is recommended that instead of a large color ROI, a small color ROI should be used, and the anterior and posterior circulations evaluated separately by moving it to the appropriate location for what is being evaluated.
The third component of TCD imaging requires the use of SD imaging to achieve real-time display of Doppler shift frequencies, represented by the SD waveform. Time is recorded along the horizontal axis, and velocity (frequency shift) is recorded on the vertical axis. This allows for velocity measurements of the SD waveforms to be recorded in centimeters per second (cm/sec). TCD imaging, like all ultrasound imaging, is based on the Doppler effect. The ultrasound transducer emits a frequency of a known value, and evaluates the value of the returned frequency, to determine what is known as the Doppler shift. A positive Doppler shift, indicating the movement of blood flow toward the transducer, is displayed above the SD baseline, and a negative Doppler shift, indicating the movement of blood flow away from the transducer, is displayed below the SD baseline. This means that the information gained from SD waveforms depends on the direction of blood flow relative to the transducer; therefore it is imperative that proper transducer placement and angle of insonation are used. Accurate recording of the intracranial SD waveform is critical, as this is the basis for interpretation of the TCD imaging examination. To this end, along with transducer placement, instrument controls to consider when obtaining SD waveforms are the SD gate (sample volume) size and position, SD gain, velocity (frequency shift) scale (PRF), baseline, wall filter, sweep speed, and output power. During TCD imaging, Doppler output power should be increased to ensure adequate penetration and enhance the quality of the SD waveforms. It is important to note, however, that this increase in output power should be done at the lowest level necessary and applied for the shortest duration possible, while still obtaining accurate clinical information. This is in accordance with the “as low as reasonably achievable” (ALARA) principle, which should be applied during all TCD imaging examinations.
TCD evaluation of the intracranial arteries is performed with a large sample volume (5 to 10 mm) to obtain a good signal-to-noise ratio. Furthermore, intracranial arterial SD waveforms are acquired without angle correction, as an assumption of a zero-degree angle is usually recommended. , This is due to a Doppler shift measurement error that occurs when the angle of insonation exceeds 30 degrees. Several investigators have evaluated the potential use of angle-adjusted (corrected) velocities during TCD imaging and have caused somewhat of a controversy on the topic, stating that angle correction reduces the inaccuracy of velocity measurements and allows for measures to be compared with diagnostic criteria for detection of vessel stenosis, all while maintaining intrarater and interrater levels of reproducibility. This caused the release of a practice recommendation consensus statement in 2008, which states that angle correction should only be performed in circumstances in which the SD sample volume can be located in a sufficiently long vessel segment, and should be omitted in curved or tortuous vessels. It is important to note that velocities that have undergone angle correction may tend to be elevated compared with those that have not. To this end, if the TCD examination being performed is for follow-up purposes and is to be compared with a previous TCD study, angle correction should not be performed if it was not done in the prior examination to allow for a proper comparison to be made. ,
Bone tissue of the skull very heavily attenuates ultrasound, making the insonation of the intracranial vasculature very difficult. Therefore thinner areas of the skull, or naturally occurring foramen or fissures, are used instead for insonation. These are known as acoustic windows. Four windows are commonly used during TCD imaging: the transtemporal, transorbital, suboccipital, and submandibular ( Figure 38-2 ). During every TCD examination, the operator should do the following:
Examine the blood flow throughout the entire course of the circle of Willis and its major branches.
Identify, optimize, and acquire SD waveforms at at least two locations in each artery.
Identify, optimize, and acquire any abnormal SD waveforms or CD signals.
Measure the highest velocity at each location in which an SD waveform is obtained.
Using the transtemporal window, the terminal ICA (t-ICA) or C1 segment of the ICA, MCA, ACA, PCA, and PCoA can be interrogated ( Figure 38-3 ). , , , , This approach is performed with the patient in the supine position, with the patient’s head aligned straight with the body. The transducer is placed on the temporal bone, cephalad to the zygomatic arch and anterior to the ear, and angled slightly upward and anterior to the contralateral ear. This transducer orientation produces an imaging plane that is a transverse oblique view. This view has the advantage of providing simultaneous visualization of the anterior and posterior intracranial circulation in many patients. The ipsilateral hemisphere is at the top of the image and the contralateral hemisphere at the bottom. Accordingly, the left side of the image is anterior and the right side of the monitor is posterior. Although the contralateral hemisphere is visualized in many patients, each hemisphere should be separately studied through the ipsilateral window to obtain the best artery-to-transducer angle. However, in patients with only a unilateral transtemporal window, evaluating the contralateral hemisphere is acceptable.
A generous amount of acoustic gel is necessary to ensure good transducer-to-skin contact. The acoustic gel will also aid in angling the transducer when it requires that the footprint be lifted from the skin’s surface in order to obtain optimal SD signals. Locating this window can be difficult and at times frustrating, as the transtemporal window is known to vary in size and location among individuals, as well as from the ipsilateral to contralateral side. Furthermore, transtemporal TCD imaging depends on the penetration of the ultrasound beam through the temporal bone. Other windows used during the TCD examination are usually less difficult to find because they consist of the much less dense natural ostia, which allows easy intracranial penetration of the ultrasound beam. However, in situations where a dense and thick temporal bone exists, the magnitude of Doppler signal attenuation will be greater, resulting in less penetration of the ultrasound beam. It has been estimated that up to only 6% of the acoustic beam intensity reaches brain tissue while insonating through the cranial bones. The ability for the ultrasound beam to penetrate the temporal bone is also influenced by the age, sex, and race of the individual. Hypertosis is the greatest obstacle for transtemporal TCD imaging, which is more prevalently found with increasing age, in females, and in individuals of African descent. , It has been reported in the literature that an estimated variation between 3% and 34% of individuals have inadequate or absent transtemporal acoustic windows. , One strategy to decrease the amount of time it takes to find the transtemporal window is to begin the TCD examination with maximum (100%) output power and, once it is located, decrease the output power until an appropriate balance between image quality and patient safety is achieved.
Once the transtemporal window is located, identifying bony landmarks ensures position at the correct level within the skull to locate the circle of Willis ( Figure 38-4 ). In the transtemporal approach, these anatomic landmarks are the petrous ridge of the temporal bone, sphenoid bone, cerebral falx, suprasellar cistern, and cerebral peduncles. In this view, the highly reflective echoes of the lesser wing of the sphenoid bone can be seen extending anteriorly, and the petrous ridge of the temporal bone can be visualized extending posteriorly ( Figure 38-5 ). The ipsilateral temporal lobe can also typically be seen at the top of the image in this view.
After the bony landmarks have been identified, CD is activated to evaluate the intracranial vasculature via the transtemporal window ( Figure 38-6 ). First, the t-ICA is visualized by slightly angling the transducer inferiorly. Typically t-ICA blood flow direction is toward the transducer, but this depends on the artery’s anatomic configuration. The t-ICA mean velocity is normally 39 ± 9 cm/sec. While imaging at this location, a mirror imaging artifact presents due to the adjacent bone.
Next the transducer is angled slightly anterosuperiorly to image and evaluate the MCA and ACA. Using the aforementioned bony landmarks, the MCA can be visualized traveling adjacent to the sphenoid wing. From this approach, blood flow in the main trunk of the MCA (M1 segment) should be traveling toward the transducer ( Figure 38-7 ). Normal mean velocity of the MCA is 62 ± 12 cm/sec. The M2 branches are also often visualized, typically traveling toward the transducer, but may appear to be traveling away due to the curvature in their shape.
The A1 segment of the ACA is displayed just posterior to the MCA on the ultrasound image, as it travels away from the transducer toward midline. It may be necessary to angle the transducer slightly anteriorly and superiorly to optimally visualize the ACA. A normal ACA mean velocity is 50 ± 11 cm/sec. Here it may be necessary to decrease the color Doppler PRF to enhance the visualization of this artery due to its lower relative velocity and increased depth from the transtemporal window. The initial portion of the A2 segment often can be visualized traveling away from the transducer in an anterior direction at midline.
The posterior circulation is imaged by angling the transducer slightly posteriorly and inferiorly, using the cerebral peduncles as an anatomic landmark. Typically, the two cerebral peduncles are identical in size and shape and are of intermediate echogenicity. Using this landmark, the P1 segment of the PCA can be visualized wrapping around the cerebral peduncle, traveling toward the transducer. For adequate interrogation of this artery, the color Doppler PRF may need to be slightly decreased, as the PCA is characterized with a slower relative blood flow velocity (39 ± 10 cm/sec under normal circumstances). The ipsilateral and contralateral P1 segments can often be visualized at their origin at the bifurcation of the basilar artery. In this view, the ipsilateral P1 segment is traveling toward the transducer, and the contralateral P1 segment is traveling away. The ipsilateral P2 segment may be displayed in red just distal to the origin of the PCoA, but will be displayed in blue distally as it wraps around the cerebral peduncle and changes direction. The color display of the P2 segment is variable because of the vessel’s anatomic route and relative orientation to the transducer.
The ACoA is not visualized during TCD imaging due to its short length. Contrarily, the longer PCoA can very often be visualized connecting the anterior and posterior circulations (the ACA and PCA, respectively) ( Figure 38-8 ). The mean peak velocity in the PCoA is 36 ± 15 cm/sec, and the direction of blood flow may be toward or away from the transducer. Again, the color Doppler PRF may need to be decreased to appropriately image the PCoA. Additionally, the use of power Doppler (PD) imaging may prove helpful in locating the PCoA, as PD imaging is not susceptible to a loss of information in the area of the vessel being insonated at 90 degrees. This makes it ideal for imaging the PCoA, as it often courses parallel to the skin line.
It is important to note that although the anterior and posterior portions of the circle of Willis can often be simultaneously visualized on a single image, variations in anatomy among the population often require minor changes in transducer position on the surface of the skin or its angle of insonation, to permit individual evaluation of either portion.
The transorbital window allows for the imaging and evaluation of the ophthalmic artery and the carotid siphon. , , , , The U.S. Food and Drug Administration (FDA) has approved certain imaging transducers on various manufacturers’ equipment for evaluation of the orbit. It is important for all operators to contact the appropriate ultrasound company to determine which transducer is approved for orbital imaging on their system. Furthermore, care should be taken to minimize the amount of acoustic energy exposure to the eye, as a traumatic subluxation of the crystalline lens can occur. Therefore the ultrasound output power should be significantly decreased, and should remain under 10% for the entirety of the time that insonation through the transorbital window is occurring. , In addition to the concentration of acoustic energy, the total time of insonation also needs to be minimized to prevent further soft tissue and ocular damage. The current FDA maximum acoustic output allowable levels (derated) for ophthalmic imaging include a spatial peak temporal average intensity of 17 mW/cm 2 and a mechanical index of 0.28.
The transorbital window imaging portion of the TCD examination begins with the patient in a supine position, with the transducer gently placed on the closed eyelid. A liberal amount of acoustic gel is important to aid in the reduction of transducer pressure needed. The ultrasound transducer is aligned in a transverse orientation, with the orientation marker pointing medially (while scanning on both the right and left sides). Therefore evaluation of either eye produces an image with medial (nasal) on the left, lateral (temporal) on the right, and the globe at the top.
Once transorbital imaging has commenced, the transducer is pointed toward the optic canal to image the carotid siphon. The carotid siphon is located in the medial portion of the transorbital window and is located at a depth of approximately 58 to 65 mm. The direction of blood flow can be used to determine which portion of the carotid siphon (parasellar, genu, or supraclinoid) is being insonated ( Figure 38-9 ). Normal flow is toward the transducer in the parasellar portion, bidirectional in the genu, and away from the transducer in the supraclinoid portion. The mean velocity in the carotid siphon is 47 ± 14 cm/sec.
The ophthalmic artery is generally identified adjacent to the optic nerve; therefore the ultrasound beam should be directed slightly medially along the anteroposterior plane for its insonation. The color Doppler PRF should be decreased to adequately visualize this vessel. The blood flow within the ophthalmic artery is toward the transducer under normal circumstances, with a mean velocity of 21 ± 5 cm/sec. Spectral Doppler waveforms obtained in the ophthalmic artery demonstrate higher resistivity than those from the carotid siphon, as this artery supplies blood to the globe and its structures ( Figure 38-10 ).
Performing TCD imaging via the suboccipital window allows for the evaluation of the vertebral and basilar arteries of the vertebrobasilar system ( Figure 38-11 ). , , , , Before imaging, the patient should be lying on his or her side with the head bowed slightly toward the chest. This position increases the gap between the cranium and the atlas, and has been found to create the best opportunity for TCD imaging. The orientation marker on the transducer is pointed to the patient’s right side while the transducer is placed on the posterior aspect of the neck, inferior to the nuchal crest. The best images from this approach are acquired with transducer placement slightly off midline and angled toward the bridge of the patient’s nose.
On the GS image obtained through this acoustic window, the foramen magnum and the occipital bone can be used as the anatomic landmarks to aid in locating the vertebrobasilar system. The foramen magnum will appear as a large, circular, and anechoic area, and the occipital bone can be identified by its bright and echogenic appearance. Under normal circumstances, the blood flow within the arteries comprising the vertebrobasilar system will be traveling away from the transducer and can be visualized on CD imaging as a blue “ Y ,” which represents the confluence of the vertebral arteries to form the basilar artery ( Figure 38-12 ). Because of the transducer orientation, which is located on the posterior portion of the body, the right vertebral artery is displayed on the left side of the image, and the left vertebral is on the right. The basilar artery is located deep to the vertebral arteries. The mean velocity is 38 ± 10 cm/sec in the vertebral arteries and 41 ± 10 cm/sec in the basilar artery. Because the posterior circulation has lower velocities than the anterior circulation, the operator may need to decrease the color Doppler PRF to visualize the arteries of the vertebrobasilar system.
Attention to the CD gain setting is critical when trying to measure the exact depth of the vertebral artery confluence, as an inappropriately high setting could result in a falsely elevated measure. Appropriate CD settings will also aid in the visualization of vertebral artery branches that can often be imaged in this approach. The most commonly visualized branch of the vertebral arteries is the PICA, which is usually seen curving as it carries blood away from the suboccipital window transducer location. Moving the transducer slightly inferior on the neck and angling superiorly provides optimal visualization of the distal portion of the basilar artery. The terminal portion of the basilar artery bifurcating into the PCAs cannot be visualized in this acoustic window.
Evaluation of the intracranial vasculature through the submandibular window is an approach that is a continuation of the TCD imaging examination, used to interrogate the distal portion of the ICA. , , , , Before imaging, the patient is placed in a supine position, with the head tilted up slightly. The transducer is placed at the angle of the mandible and is angled slightly medially and cephalad toward the carotid canal. The orientation marker on the transducer should be pointed superiorly. The distal ICA can be typically found at a depth of 35 to 80 mm. Normal blood flow in the distal ICA travels away from the transducer. Careful Doppler evaluation is important to distinguish the ICA’s low resistance signal from the higher resistance signal of the external carotid artery (ECA). The mean velocity in the retromandibular portion of the ICA is normally 37 ± 9 cm/sec.
The quality of the intracranial image depends on proper ultrasound instrumentation adjustment and optimization. Initially in a TCD examination, anatomic landmarks are identified using GS imaging to aid in locating various intracranial vessels. Therefore GS parameters such as frequency, sector width, image depth, overall gain, TGC, focal zone number and placement, frame rate, and dynamic range must be appropriately optimized to facilitate this. To obtain quality CD images, increase the color gain to the appropriate level, maintain a small sector width and color ROI width to keep the highest possible frame rates, change the color PRF depending on the vessel being evaluated and the patient’s individual hemodynamics, and remain aware of color sensitivity and persistence settings. Just as the GS image aids in locating intracranial arteries, the CD image aids in determining the proper placement of the SD gate for the acquisition of an SD waveform. The TCD examination is interpreted based on the information obtained from the SD waveform. Therefore SD signals are obtained along the path of the artery using the color display as a road map. At each depth setting, it is important to adjust the SD gate and angle the transducer to optimize the SD signal. Additionally, it is important to remember that the CD image is displayed in only two dimensions; consequently, tortuous intracranial arteries often cannot be displayed along their entire length as a continuous pathway.
Interpretation of transcranial doppler imaging
Proper identification of the intracranial vasculature is, of course, the first step in evaluating the TCD examination. GS imaging should first be used to identify anatomic or bony landmarks (when appropriate), followed by CD imaging to visualize the blood flow within the intracranial vasculature. Each TCD window provides access to specific arteries. The identification of those arteries are based on the following:
Depth of the sample volume. The sample volume depth is measured in millimeters (mm) and is the distance from the face of the ultrasound transducer to the middle of the SD gate. The SD gate should be kept large (5 to 10 mm) to achieve the best signal-to-noise ratio. The depth ranges at which the intracranial arteries can be located from each TCD window are listed in Table 38-1 . The depths of the arteries may vary with each individual, but in most adults the artery in question will usually fall within these ranges. Midline is in the range of 70 to 80 mm in most adults.
Mean Velocity (cm/sec)
62 ± 12
50 ± 11
39 ± 9
39 ± 10
36 ± 15
21 ± 5
47 ± 14
Bi, away, toward
38 ± 10
41 ± 10
37 ± 9
Angle of the transducer. The angle of the transducer is important, as different arteries can be insonated at the same depth from the same acoustic window. The operator’s position at patient’s head will enable the best perception of the transducer-artery angle because it permits orientation of the ultrasound transducer relative to the body’s axes and planes.
Blood flow direction. The normal direction of blood flow in the intracranial arteries relative to the ultrasound transducer is listed in Table 38-1 . From the different acoustic windows, SD waveforms may demonstrate blood flow direction away from, toward, or in both directions from (bidirectional) the transducer. If the direction of blood flow is reversed from the established norm, it can be assumed that the artery is functioning as a collateral channel, or that it may be due to an anatomic variant.
Spatial relationship. Understanding the spatial relationship of one artery to another is necessary to properly identify the intracranial arteries. Using the t-ICA bifurcation as a guide, the location of the intracranial arteries is determined with greater technical ease.
Traceability of an artery. The operator should be able to “trace” the anatomic route of the artery in stepwise fashion by increasing or decreasing the depth setting of the SD gate. The SD gate should be moved slowly along the path of the vessel to thoroughly assess its entire length, using the CD display as a guide.
Adjacent anatomic structures. Using the anatomic structures visualized on GS imaging as a guide to correctly identify the intracranial arteries is especially important. It is the major advantage of using the imaging versus the nonimaging TCD technique.
Standard interpretation criteria for TCD examinations have been reported in the literature. , , These criteria were previously developed using the nonimaging TCD technique, although they have been applied to TCD imaging. The same measurements are being obtained during both techniques, but TCD imaging permits identification of structural landmarks that assist in accurately locating and obtaining SD waveforms of the intracranial arteries.
The evolution of the TCD imaging examination has made it apparent that important underlying physiologic variables, in addition to pathology, may affect blood flow velocities within the intracranial arterial system. , Therefore it is important that the operator be aware of the factors that may affect intracranial hemodynamics, allowing them interpret each case on an individual basis.
The most important factor influencing intracranial arterial velocity measures obtained in a TCD examination is patient age. Several investigators have found lower levels of velocities in individuals with increasing age. , , The downward trend in intracranial arterial velocities is most likely multifactorial, possibly due to changes in cardiac output, age-related decrease in cerebral perfusion, decreases in metabolic demands, and/or elevated hematocrit levels.
Women between 20 and 60 years of age have demonstrated higher flow velocities than men, by an estimated 10% to 15%. , Possible explanations for these differences are lower levels of hematocrit being found in women, the intracranial arteries of women may be smaller in diameter, and/or women having higher hemispheric cerebral blood flow than men. , It is important to note that there are thought to be no detectable velocity differences between genders after age 70.
Hematocrit (Hct) is the percentage of red blood cells by volume in whole blood and is a major determinant of blood viscosity. Blood viscosity, or the “thickness of blood,” is an important factor that influences intracranial arterial velocities, as blood hematocrit and viscosity levels are inversely related to intracranial arterial blood flow velocity. , , It has been found that a decrease in Hct levels from 40% to 30% can cause velocity increases by up to 20%. A clinical example of this are the expected findings of elevated intracranial arterial velocities in the presence of anemia (Hct less than 30%). If anemia is the cause of elevated velocities, these changes should be detected from all of the intracranial arteries. Focal or localized velocity increases suggest a different cause.
Carbon dioxide reactivity.
Changes in arterial carbon dioxide (CO 2 ) partial pressure (Pco 2 ) have a direct effect relationship on cerebral blood flow and intracranial arterial velocities. , , Alteration of cerebral vascular resistance and changes in TCD signals result from changes at the level of the arteriolar channels. Clinical examples include hyperventilation (a deficiency of CO 2 in the blood known as hypocapnia ), which causes a decrease in the MCA mean velocity and an increase in the PI. Conversely, hypoventilation (an excess of CO 2 in the blood referred to as hypercapnia ) causes an increase in MCA mean velocity and a decrease in PI. This information suggests that general changes in the intracranial arterial velocities caused by CO 2 reactivity must be taken into account when interpreting TCD data.
Heart rate and cardiac output.
Intracranial arterial velocities are a reflection of an individual’s heart rate. Therefore most experienced TCD examiners caution against obtaining SD waveforms during times when fluctuations in the cardiac rhythm can occur. This can occur when the patient is yawning, agitated, experiencing pain, or in any other situation that could cause a change in heart rate. To this end, any cardiac arrhythmia will be reflected in the TCD SD waveform. If there is a question concerning a change in the patient’s heart rate, several SD waveform tracings should be obtained before relying on the calculations. To compensate for extreme cases of bradycardia or tachycardia, the ultrasound instrumentation, such as SD tracing sweep time or PRF, may need to be adjusted. Changes resulting from cardiac output not associated with hemodilution have little effect on the cerebral blood flow if autoregulation is intact. This suggests that TCD velocities should be theoretically independent of small changes in cardiac output. Data on the relationship between cardiac output and intracranial arterial velocities are limited, and further investigation in this area is necessary.
Many parameters are involved in the accurate interpretation of TCD data. Each patient must be considered individually because of the variety of physiologic factors that affect intracranial arterial hemodynamics ( Box 38-2 ). The parameters that affect the intracranial velocities can be categorized into those factors that are (1) proximal to the circle of Willis, (2) at the level of the circle of Willis, and (3) distal to the circle of Willis.