Carotid Artery Stenosis Imaging 

Stroke (brain attack) represents one of the most serious causes of mortality and morbidity in the United States and throughout the world. Each year, 150,000 patients die as a direct result of a cerebrovascular accident (CVA), while 600,000 patients experience the morbidity of aphasia, blindness, or paralysis.

Among patients with stroke, extracranial carotid disease represents the cause of approximately one half of cases. The goals of carotid imaging can be described as early detection, clinical staging, surgical road mapping, and postoperative therapeutic surveillance (see the images below).

Imaging helps in detecting associated conditions early. Coronary artery disease, peripheral arterial stenosis, and hypertension are associated with an increased incidence of carotid arterial stenosis. Carotid arterial screening remains controversial. Good evidence supports evaluation for carotid stenosis prior to coronary arterial bypass surgery.^{[1, 2, 3]}

Detection of a carotid bruit is a common physical examination finding that may lead to a referral for carotid duplex ultrasonography. The correlation between a carotid bruit and a hemodynamically important carotid stenosis is reported to be between 10-20%. A cardiac murmur may be transmitted to the neck. Stiff, calcified, or torturous vessels may generate a bruit in the absence of stenosis.

Preferred examination

A complete medical history should be taken first. On the basis of risk factors and the family history and current symptoms, selecting a screening examination such as carotid duplex ultrasonography, computed tomography angiography (CTA), or magnetic resonance angiography (MRA) of the carotid artery may be appropriate (see the images below).^[4]

Carotid duplex imaging is performed most commonly in patients with moderate risk factors. In the author's practice, duplex ultrasonography is the initial triage examination for patients with an asymptomatic bruit and known asymptomatic carotid vascular disease, and in patients with a complete stroke without prior history of carotid stenosis.

In symptomatic patients and in most who present with abnormal carotid ultrasonographic findings, another imaging test is performed. MRA or CTA offer full depiction of the cervical and cerebral portions of the common carotid artery (CCA) and internal carotid artery (ICA). In most cases, a diagnostic evaluation for cerebral vascular disease can be performed by using either MRA or CTA. The immediate availability of CTA in many critical hospitals on a 24-hour basis makes CTA attractive in the care of patients who present after hours in the emergency department.

In the selection of patients for acute treatment for cerebral ischemia, axial CT images should be reviewed by using the criteria of less than one-third involvement of the middle cerebral distribution or by using the Alberta Stroke Program Early CT Score (ASPECTS). The role of cervical-cerebral angiography is evolving as less invasive alternative tests have become available. Many experienced vascular surgeons are more comfortable with cervical-cerebral catheter angiography than with other studies.

In the author's practice, clinicians perform carotid angiography routinely, and cerebral angiography or digital subtraction angiography (DSA) is reserved for patients who are likely candidates for surgery or for patients who may benefit from angiographically based carotid intervention.

Limitations of techniques

Although carotid duplex imaging offers an excellent means of initial evaluation of the extracranial cerebral vessels, the presence of dense calcifications in the carotid plaque tends to make the study less accurate.

Because carotid duplex imaging does not help in assessing the intracranial portion of the carotid artery, tandem lesions of the ICA may be missed. In a similar manner, proximal stenosis of the innominate artery and the left carotid artery cannot be evaluated near the origins from the aortic arch.

MRA is contraindicated in patients who have cardiac pacemakers or cerebral aneurysm clips or in those who have undergone certain other medical procedures. In addition, MRA is highly motion sensitive. Many patients require sedation. Because of artifacts related to the MRA image, the degree of stenosis may be overestimated.

CTA requires that iodinated contrast agents be injected at a relatively high flow rate. Patients with renal disease may not tolerate intravenous contrast agents. Motion artifacts remain a problem if the examination is performed by using older CT scanning equipment.

Cerebral angiography also involves the injection of iodinated contrast agents. The overall contrast dose is similar to that required for CTA. The performance of catheter-based cervical-cerebral angiography depends on the skill and experience of the angiographer. Overall major morbidity rates are 0.1-1%. Injury may occur in the form of iatrogenic stroke or bleeding around the catheter introduction site.

Angiograms do not provide much information concerning the nature of the plaque lesion. Cerebral angiography is the most costly means of carotid stenosis evaluation. If cases are selected carefully, the overall risk of diagnostic angiography, together with the morbidity related to carotid surgery, is less than the risk of stroke for the untreated patient (see the image below).

Patient education

For patient education information, see the Stroke Center and Dementia Center, as well as Stroke, Transient Ischemic Attack (Mini-stroke), and Stroke-Related Dementia.

Standard radiographs may demonstrate calcification in the carotid vessels in the neck; however, only large calcified plaques are demonstrated on radiographs. In general, the information provided by radiographs of the neck or skull is not clinically helpful except to alert the clinical physician that the patient may be at risk for carotid stenosis.

Degree of confidence

In general, calcification in a plaque indicates a chronic disease pattern. However, the absence of calcification on radiographs of the neck or lateral aspect of the skull does not exclude significant extracranial carotid disease.

Calcification in the carotid bulb is not fully correlated with significant carotid arterial stenosis. High-grade stenosis of the proximal carotid artery may occur in the absence of calcifications, and dense carotid calcification may be seen in the absence of a high-grade carotid stenosis.

Axial computed tomography (CT) scanning of the cerebral circulation provides an accurate means of assessing stenosis and carotid plaque (see the images below). Although early attempts to apply CT scanning in the evaluation of the carotid artery were limited by movement artifacts and thick scanning sections, current multisection CT scanners allow for the acquisition of thin (eg, 1.0-2.0 mm) axial images within a brief time (a single breath hold).

Intravenous contrast material must be injected rapidly enough (3-4 mL/s for a total volume of 120-150 mL of 300-320 mg/mL nonionic contrast agent) to achieve a contrast density of at least 150 HU or in the innominate and carotid inflow to continuing distally into the intracranial carotid artery. Imaging begins just before the contrast density peaks in the carotid artery.^{[5, 6, 7]}

Initially, all images should be reviewed in the axial plane. Multiplanar and curved multiplanar reformatted images are often helpful. The intraluminal diameter should be measured by using an electronic workstation with electronic calipers. If the image of the carotid artery is enlarged before measurement, error is reduced. Measurements are made across the lumen through the narrowest portion of the proximal ICA and across the area of the ICA that is above the stenosis and is believed to be normal.

The degree of stenosis is calculated according to criteria developed by the North American Symptomatic Carotid Endarterectomy Trial (NASCET) and is reported as a percentage of stenosis.

Carotid ulceration

The carotid arteries may be a source of cerebral emboli via release of the contents of the plaque, via turbulence or clot formation on the ulcerated surface. Carotid ulcers are best demonstrated by using the higher resolution of catheter angiography; however, the use of multisection CTA increases the likelihood of carotid ulcer identification. Larger deeper ulcers are successfully depicted with most CTA techniques (see the image below).

Carotid plaque

Dense calcifications in the carotid artery limit the accuracy of measurements of the degree of stenosis across the plaque if maximum intensity projection (MIP) or shaded-surface display imaging is used. After first attempting to use axial images to measure the degree of stenosis, curved, multiplanar, reformatted images obtained through the lumen help to show the stenosis.

Certain workstation techniques can be applied that can image reverse the pixels above 300 HU. Image inversion subtracts calcified plaque from the image, which tends to result in a clearer image of the contrast agent–filled carotid lumen. Occasionally, contrast may be seen entering the subintimal area of the plaque. Complex and dissected carotid plaques may be detected in this manner.

Carotid thrombosis

Care must be exercised in cases of carotid thrombosis. Very slow flow rates may be missed if the timing of the intravenous contrast agent bolus or the peak density of the contrast material is less than optimal. It is always necessary to compare 3-dimensional (3D) volume images to axial images (see the images below).

The intracranial collateralization pattern helps make the diagnosis of thrombosis and offers important clinical information.

Tumors of the cervical region may surround the CCA or the ICAs. One of the advantages of cross-sectional imaging, such as CTA, is the identification of the tissues that surround the carotid and vertebral arteries.

Degree of confidence

Carotid CTA represents a reliable means of estimating the degree of stenosis in extracranial and intracranial vessels. Limitations in the degree of confidence are related to technical factors. Current multisection CT scanners allow for the acquisition of up to 16 sections for each gantry rotation. Each rotation may require as little as 0.4 seconds. Axial collimation for cervical-cerebral CTA is performed by using a collimation of 0.75 mm with a reconstruction of 1.5 mm for each axial image.

Axial images, multiplanar reformatted images, and 3D volume MIP and volume-model images contribute to the sensitivity and accuracy of multisection CTA. By using available multisection CT scanners, extracranial carotid stenosis can be diagnosed to a degree of accuracy equal to or exceeding that of catheter-based angiography.

CT also offers an excellent means of detecting a tumor that might surround the carotid artery in the neck.

When the carotid artery is imaged by using CTA, artifacts include motion, poor cardiac function, and dense carotid calcification. Even so, overall accuracy for carotid CTA exceeds 95%. Intracranial CTA with a multisection scanner reportedly helps in identifying intracranial carotid stenosis with a sensitivity of greater than 98%, a specificity of 99%, and an accuracy of greater than 98%.

Multisection CT scanners have been introduced that acquire 8 scans with each gantry rotation. As a result, CTA now represents the best overall means of investigating carotid stenosis in a less invasive and more cost-effective manner.

Motion artifacts, however, can reduce the diagnostic efficacy of carotid CTA. Sudden movement, breathing, or swallowing by the patient during scanning may result in a misregistration of the axial images on 3D or multiplanar reformatted images. In such cases, only measurements taken from the axial images should be considered.

Diagnostic problems may also occur if the vessels of the carotid bulb are very dilated and torturous. Superimposed jugular veins and arteries may hide stenosis. A careful review of axial images, together with carefully performed curved, multiplanar, reformatted images through the carotid lumen, demonstrates the stenotic area in most patients.

MRA usually is performed by using 2 primary methods, ie, time-of-flight and ultrashort, T1-weighted imaging (see the images below).

A 3D time-of-flight image of the carotid artery or a contrast-enhanced, short-TE (echo time), short-TR (recovery time) image is interpreted in much the same manner as CTAs or carotid angiograms of the same area. The stenotic area of the ICA should be evaluated according to the North American Symptomatic Carotid Endarterectomy Trial (NASCET) criteria.

Carotid artery stenosis is seen in the images below.

Time-of-flight imaging

Time-of-flight imaging is performed without an intravenous contrast agent by using a spoiled gradient-echo sequence. The images are displayed with an MIP protocol in multiple projections.

Because of the effects of turbulence, 3D time-of-flight imaging tends to cause overestimation of high-grade stenoses. In some cases, an area of discontinuity may be generated in the area of the stenosis. This results from turbulent blood flow patterns at the point of a high-grade stenosis and within very stenotic longer stenoses.

Contrast-enhanced MRA

Contrast-enhanced MRA is performed by using a timed and rapid injection of a gadolinium-based contrast agent, such as gadolinium dimeglumine.^{[8, 9]}

Because the volume of contrast agent is limited to 15-20 mL in most cases, timing of the contrast agent bolus and good venous access are essential. The images are obtained by using a short TR, short TE, and T1-weighted technique (TR/TE/flip angle, 4.9/2.4/35°). The images are displayed in multiple projections by using an MIP technique.^[10]

MRA results obtained by using a very short-TR, short-TE, gadolinium-enhanced, timed bolus technique can be interpreted in much the same manner as those of carotid angiography and CTA.

The results of contrast-enhanced MRA are closely correlated with angiographic and operative findings. A gadolinium-based contrast agent has the effect of shortening the T1 qualities of the blood flowing through the stenosis. Residual lumen is displayed with high signal intensity. High-grade stenosis tends to remain an attenuated, but visible, lumen.^[11] Results of contrast-enhanced MRA are usually better than those of 3D time-of-flight imaging.

Advances in contrast-enhanced MRA have allowed for improved imaging speed without the need for temporal interpolation. Time-resolved, contrast-enhanced carotid MRA with a sensitivity-encoding (SENSE) reconstruction technique enables visualization of the carotid artery without superimposed venous structures.

The early identification of stroke in the vascular distribution of a related carotid stenosis helps in focusing on an ipsilateral carotid lesion. T1-weighted, diffusion-based imaging effectively depicts focal cerebral ischemia (see the images below), and the finding is often correlated with a proximal carotid stenosis. In other cases, intracranial MRA demonstrates no flow pattern, confirming carotid thrombosis.

Gadolinium-based contrast agents (gadopentetate dimeglumine [Magnevist], gadobenate dimeglumine [MultiHance], gadodiamide [Omniscan], gadoversetamide [OptiMARK], gadoteridol [ProHance]) have been linked to the development of nephrogenic systemic fibrosis (NSF) or nephrogenic fibrosing dermopathy (NFD). The disease has occurred in patients with moderate to end-stage renal disease after being given a gadolinium-based contrast agent to enhance MRI or MRA scans.

NSF/NFD is a debilitating and sometimes fatal disease. Characteristics include red or dark patches on the skin; burning, itching, swelling, hardening, and tightening of the skin; yellow spots on the whites of the eyes; joint stiffness with trouble moving or straightening the arms, hands, legs, or feet; pain deep in the hip bones or ribs; and muscle weakness.

Degree of confidence

Intracranial MRA has a sensitivity of 92-95%, a specificity of 91%, and an overall accuracy of 91% in the detection of intracranial carotid stenosis.

Systematic and random errors affect the results of cerebrovascular MRA. Systematic errors are primarily the result of artifacts and are related to 3D time-of-flight imaging. Areas within the carotid artery that generate turbulence may develop a recirculating pattern of blood flow. In the returning blood flow, the lumen is recorded as artifactually narrowed. Short-segment occlusions may be suggested in patients with very high-grade stenosis (>85% stenosis).

Time-of-flight MRA results should be reported in a manner similar to that for carotid duplex ultrasonography results. Although measurements should be attempted using the NASCET criteria, the significance of the recorded stenosis should be reported with caution.

Certain variations of the CCA may be difficult to interpret by using MRA findings. Narrow bands of intimal thickening may result in a high degree of stenosis; however, the central luminal narrowing is difficult to demonstrate on MRA examination. The proximal ICA may become folded or kinked. Such a kink generates turbulence just beyond the fold. A short-segment pseudo-occlusion may be seen on a time-of-flight study.

The ectatic appearance of fibromuscular hyperplasia may be difficult to resolve by using time-of-flight imaging alone.

Random errors may occur when measurements of the internal lumen are made mechanically. Measuring from enlarged images reduces the relative degree of random error.

The spasm of migraine can simulate carotid dissection on MRA studies performed during an acute migraine event. Assessment of the proximal carotid artery near its origin from the aortic arch requires a good contrast bolus, breath-holding and, in general, good cooperation on the part of the patient.

False-negative carotid MRA results are primarily the result of limitations of resolution. Small surface ulcers may not be resolved on MRA images. At other times, deeper ulcers may contain blood clots, which reduce the apparent size of some ulcers.

B-mode ultrasonography

The ultrasonographic characteristics of symptomatic and asymptomatic carotid plaques are different. Symptomatic plaques are more likely to be hypoechoic and highly stenotic, while asymptomatic plaques are hyperechoic and moderately stenotic. Evaluation of the surface of the plaque has not been demonstrated to be a satisfactory index of plaque instability.

After carotid endarterectomy, intimal thickness varies in the surgical site. The thickness of the neointima has been correlated with carotid wall stiffness and restenosis after carotid surgery. Comparison of current carotid ultrasonographic velocities to those obtained before carotid endarterectomy is important. In general, the velocity of flow should become reduced with successful surgery. However, velocities may remain elevated after surgery due to scar formation, which results in carotid wall stiffness. Restenosis may occur as a result of surgery or generalized atherosclerotic vascular disease.

Doppler ultrasonography

Doppler ultrasonography is the primary noninvasive test for evaluating carotid stenosis.^{[12, 13, 14]}

Primary examination of the carotid plaque is somewhat subjective, because terms such as soft plaque or irregular surface are often used to describe the primary ultrasonographic images. The degree of stenosis is better measured on the basis of the waveform and spectral analysis of the CCA and its major branches, especially the ICA.

As an example, consider the following case history. A 70-year-old man presented with a suspected stroke. Symptoms included right-sided hemiparesis and dysphasia. Duplex ultrasonographic findings revealed significant stenosis of the proximal left ICA.

The image showed mostly soft, echogenic plaque protruding from the posterior aspect of the bulb and origin of the ICA. A fairly narrow residual lumen remained. Spectral Doppler imaging confirmed the high-grade stenosis, with peak systolic velocity of 325 cm/s and diastolic velocity of 105 cm/s. In the author's laboratory, a diastolic cutoff of 90cm/s is used to distinguish 60-79% from 80-99% lesions in patients such as this man.

The blunted flow in the distal ICA was compatible with the high-grade, proximal obstruction. The CCA showed narrowing of approximately 40%. The opposite ICA was only mildly narrowed and demonstrated normal findings on spectral Doppler examination. The patient underwent carotid angiography. Angiographic findings helped to confirm an 80% ICA stenosis.

The patient was eventually treated with carotid endarterectomy. Follow-up studies in the author's laboratory (not shown) demonstrated a fully patent ICA. Some residual hemiparesis remained on the right side, but the patient was much improved from his status after the initial stroke.

Many published lists exist of carotid flow velocities, which are associated with a graduated degree of stenosis. A listing of flow velocity/carotid stenosis criteria used in the author's department is shown in the table below.

Table. Carotid Stenosis Criteria (Open Table in a new window)

Doppler ultrasonographic findings associated with stenosis

Ipsilateral CCA-to-ICA flow ratios may not be valid in the setting of contralateral ICA occlusion.

CCA waveforms may have a high-resistance configuration in ipsilateral ICA lesions, while ICA waveforms may have a high-resistance configuration in ipsilateral, distal ICA lesions. In addition, ICA waveforms may be dampened in ipsilateral CCA lesions.

Long-segment ICA stenosis may not have high end-diastolic velocity.

Velocities supersede imaging in grading stenosis.

Imaging can be used to downgrade stenosis in the setting of turbulence caused by kinking.

In addition to the typical evaluation of the flow rate in the proximal ICA, it has been shown that the flow directly within the ophthalmic artery is highly specific for severe carotid stenosis.

Transcranial Doppler ultrasonography

Transcranial Doppler imaging is sensitive and specific in the detection of hemodynamically significant intracranial ICA stenosis. By using a mean flow velocity of 100 cm/s, transcranial Doppler images help in identifying most intracranial stenotic lesions, with good specificity.

Degree of confidence

The success of duplex ultrasonography (see the images below) depends on careful technique. Results vary somewhat among laboratories. The author's practice has chosen to base its results on a composite mix of the published results comparing the degree of proven stenosis versus flow velocity and ICA-to-CCA velocity ratios. Results should be reported in the context of established criteria as a percentage of stenosis. The table above represents the criteria used in the author's practice. The results should be confirmed with carotid CTA or carotid MRA before surgery.

Discovery of a reversed direction of flow in the ophthalmic artery is closely associated with high-grade ipsilateral ICA stenosis, with a sensitivity of 55%, a specificity of 100%, a negative predictive value of 82%, and a positive predictive value of 100%.

Transcranial duplex ultrasonography with a threshold of 100 cm/s helps in identifying hemodynamically significant lesions, with a sensitivity of 93.9% and a specificity of 91.2%.

The use of duplex ultrasonography is not valid in the setting of contralateral ICA occlusion. CCA waveforms may have a high-resistance configuration with ipsilateral ICA lesions. ICA waveforms may have a high-resistance configuration in ipsilateral distal ICA lesions. ICA waveforms may be dampened with ipsilateral CCA lesions. Long-segment ICA stenosis may not have high end-diastolic velocity. Velocities supersede imaging in grading stenosis. Imaging can be used to downgrade stenosis in the setting of turbulence caused by kinking.

After carotid surgery, recorded flow velocities may be elevated more than those recorded during preoperative examinations. Comparison of the results of the current examination to those of a prior study is essential.

Single photon emission CT (SPECT) scanning and positron emission tomography (PET) scanning of the brain may demonstrate areas of cerebral ischemia versus areas of complete cerebral infarction.^{[15, 16, 17]} However, these studies provide only indirect imaging information related to carotid stenosis.

By using a dynamic imaging method, an assessment of the inflow of technetium-99m (^99m Tc) hexamethylpropylamine oxime (HMPAO) can be used to estimate the hemispheric cerebral blood flow. If such an assay is combined with a provocative challenge with acetazolamide, it is possible to differentiate patients with migraine from patients with an irreversible cerebral ischemia.

There may, however, be confusion in the findings between an infarct of the brain and a low-grade brain neoplasm or abscess. Moreover, in the evaluation of stroke, false-negative findings may occur in small areas of infarction, which may be below the resolution of a nuclear examination. Comparison with other studies (such as CT scanning or MRI), which have higher resolution, is recommended.

Carotid angiography via a catheter injection of contrast agent is considered the standard of diagnostic imaging of the cervical and intracranial carotid arteries against which other techniques are often judged (see the images below).

The diameters of stenosis can be measured directly, and luminal diameter ratios can be expressed. All contrast-enhanced angiographic studies depend on the radiographic density of iodinated contrast agent compared with normal blood and the density of the wall of the carotid artery (see the images below).

Lesions may be smooth, irregular, or focal, or they may involve a long segment (see the image below).

The NASCET study was based on images created during catheter angiography (see the image below).

Angiographic technique

In the author's institution, all cervical-cerebral angiography is performed by using the Seldinger technique. The procedure initially involves entering the lumen of the artery of access with a needle. The femoral artery is used as access in most cases, while the axillary arteries may be entered as an alternative.

A heparin-coated guide wire is passed through the hub of the needle into the lumen of the artery. Often, a J -shaped guide wire is used initially to avoid intimal trauma. A 4F pigtail catheter is generally introduced over the guidewire into the ascending aortic arch. Nonionic contrast (320 mg/mL of organically bound iodine) is injected at a rate of 20-25 mL for a total volume of 40-50 mL. The left anterior oblique projection is most common. If imaging is performed using a digitally subtracted technique, less contrast material is needed.

After the shape, smoothness, and patency of the proximal right CCA, the right subclavian artery, the left CCA, and the left subclavian artery are inspected, a catheter is selected to assist in the selective catheterization of the right CCA, the left CCA, and either vertebral artery or both of them. A 0.035-inch guidewire with a soft, straight tip is used to exchange the pigtail catheter for either a simple angle-tip catheter (eg, one with an HN1 shape) or one with a more complex hook or short-radius, curved shape.

The guidewire chosen for the exchange may have a variable degree of flexibility in the distal several centimeters near the tip. With the guidewire leading into the proximal right and left CCA origins, the cerebral-shaped catheter is positioned in the CCAs below the carotid bulb. Vertebral injections are performed with the catheter in the vertebral artery near the origin of the vertebral artery to avoid spasm.

After a small test injection is disinterred to verify the location and security of the catheter tip position, each of the carotid arterial circulations is studied. Images in a 30° ipsilateral left anterior oblique or right anterior oblique projection are obtained to clearly outline the carotid bifurcations. In some cases, the lateral and the anterior projections are needed.

DSA performed by using a C-arm imager permits an optimal degree of rotation based on the patient's anatomic form. Imaging should include the intracranial carotid circulation in most cases. Selective catheterization of 1 or both vertebral arteries should be performed if clinically indicated in patients with vertebral basilar symptoms.

Each injection of contrast agent into the CCA is given at a rate of 6-7 mL/s for a total volume of 10-12 mL, depending on the estimated flow rate in the CCA. Selective vertebral artery angiography is generally performed using an injection rate of 4-5 mL/s for a total volume of 6-8 mL. Somewhat less contrast agent is needed if DSA imaging techniques are applied.

The degree to which intracranial arteries beyond the ICA-MCA trifurcation must be studied should be based on an understanding of the patient's clinical presentation and symptoms. Collateral circulation is best understood by having full visualization of the intracranial circulation, as well as the aortic arch and cervical carotid circulations.

Measurements and other findings

The most common method used to express measurements of the carotid artery is the NASCET criteria, which state that a percentage of stenosis is expressed as a factor of 100, ie, the diameter of the normal ICA located above the carotid bulb divided by the diameter of the proximal ICA at the narrowest point.

Special consideration must be given when there is a critical degree of stenosis, often termed the string sign. The presence of contrast agent in a markedly restricted lumen may be an indication of a critical proximal focal stenosis or longer segmental narrowing. Depiction of residual intracranial arterial flow helps to establish patency (see the image below).

Other findings noted on carotid angiography include calcification, ulceration (see the first image below), fibromuscular hyperplasia, carotid kinking or folding (see the second and third images below), focal thrombus formation (see the fourth image below), and intimal dissection.

The plaque may become complicated by marked intimal thickening or subintimal hemorrhage.

After a carotid thrombosis forms, extensive collateralization patterns may be demonstrated, or partial recanalization may occur, resulting in a complex pattern of alternating narrowing and dilatation (see the image below).

Degree of confidence

Findings on catheter-contrast angiography of the carotid artery are accurate if measurements and calculation of the degree of stenosis are made carefully using an enlarged image. Systematic and random errors may occur during stenosis measurement. The calculation of the degree of stenosis should be based on the NASCET criteria if possible.

False-positive examinations may occur if injection of the contrast agent is not timed properly, resulting in a washed-out appearance on the image.

If subtracted images are developed, overlying metal or dense calcifications result in shadowing of some of the plaque-artery lumen, making an accurate measurement of the residual lumen difficult. Movement, coughing, and the presence of metal artifacts may prevent an accurate examination. Rapid injection may cause a standing wave to form, which may appear similar to fibromuscular hyperplasia.

False-negative results may occur if a nonselective angiographic technique is used. Superimposition of other arterial and venous structures may prevent adequate depiction or measurement of the stenosis.

Despite technical limitations, catheter-based angiographic studies remain the most accurate means of assessing the degree of carotid stenosis.