Ultrasonography

Intima-Media Thickness

Thickening of the two inner layers (intima and media) of the carotid artery wall has been regarded as the first morphologic sign of atherosclerosis and also a marker for generalized atherosclerosis. Early atherosclerotic changes occur preferentially in the intima whereas medial hypertrophy is commonly seen in hypertensive patients. ^, Prospective studies have shown correlation between carotid intima-media thickness (IMT) and increased cardiovascular disease (CVD) risk. ^{, , ,} Increased common carotid IMT is a focal marker of diffuse vascular disease (atherosclerosis) and is associated with a higher risk of first-ever and recurrent stroke. ^, IMT measurements can also help refine cardiovascular event risk stratification based on the Framingham score. Nevertheless, IMT thickening may also occur in other non-atherosclerotic conditions that result in intimal hyperplasia or medial hypertrophy. Differentiating non-atherosclerotic disease using ultrasound is often limited and requires subsequent follow-up studies. In practice, IMT is defined by B-mode ultrasonography as a double-line pattern in which two parallel lines correspond to the lumen-intima and media-adventitia interfaces. Moreover, a pathology study concluded that ultrasonography accurately measures the far-wall IMT, that is, the carotid artery wall located deeper in soft tissues and further away from the transducer surface ( Fig. 46.1 ). Further studies showed good intra-observer and inter-observer reproducibility. ^, In 2013, the American College of Cardiology/American Heart Association (ACC/AHA) guidelines for cardiovascular risk assessment did not recommend carotid IMT measurement for use in clinical practice as a routine measurement of risk assessment for a first atherosclerotic CVD (class III evidence level). Accurate measurement is hampered by anatomic variability, and by focal thickening of the carotid wall, mostly due to the development of atherosclerotic plaques. There have been several attempts to standardize IMT measurement. The Mannheim Carotid Intima-Media Thickness and Plaque Consensus released the guidance on how to assess IMT: arterial wall segments should be assessed in a longitudinal view, strictly perpendicular to the ultrasound beam, and both walls should be depicted during diastole in order to obtain diameter measurements.

Early Carotid Plaque

Plaques are defined according to the Mannheim Consensus as focal structures encroaching into the arterial lumen of at least 0.5 mm or 50% of the surrounding IMT value; they should demonstrate a thickness greater than 1.5 mm as measured from intima-lumen interface to the media-adventitia interface. Similarly, a report from the American Society of Echocardiography (ASE) and the Society of Vascular Medicine and Biology specified carotid plaque as focal wall thickening that is at least 50% greater than that of the surrounding vessel wall or as a focal region with carotid IMT greater than 1.5 mm that protrudes into the lumen and is distinct from the adjacent boundary. Moreover, further studies demonstrated similar or greater predictive power for carotid plaque and CVD risk. ^, According to the ASE report, measurement of carotid IMT and detection of carotid plaques can reclassify patients at intermediate risk and predict CVD events. The groups of patients who mostly benefit from refining CVD risk assessment by measuring carotid IMT and carotid plaque are patients aged 40–70 years without a condition indicating high CVD risk; at intermediate CVD risk (i.e., 6%–20% 10-year risk of myocardial infarction or coronary heart disease death using the Framingham risk score); with family history of premature CVD in a first-degree relative; individuals older than 60 years with severe abnormalities in a single risk factor who otherwise would not be candidates for pharmacotherapy; or women younger than 60 years with ≥2 CVD risk factors. This examination is reasonable if level of aggressiveness of preventive therapy is uncertain and additional information about the burden of subclinical vascular disease or future CVD risk is needed. Imaging should not be performed if the results are not expected to alter therapy or in patients with history of CVD. ^,

Identifying High-Risk Plaques

According to the Trial of Org 10172 in Acute Stroke Treatment (TOAST) classification, in order for an anterior circulation ischemic stroke to be attributed to carotid atheromatosis, the culprit plaque has to cause a stenosis of greater than 50%. It is clear that plaques resulting in less than 50% stenosis may also cause stroke that may be classified as cryptogenic in the absence of an identifiable stroke cause. Apart from luminal narrowing, the content of a plaque may render it vulnerable, and it is known that most coronary artery plaques resulting in myocardial infarction cause luminal stenosis of less than 50%. ^, Hence, stroke risk assessment based on carotid plaque features may help stratify patients without severe carotid stenosis who have increased risk of ischemic stroke and may select those that may benefit from intensive modification of cardiovascular risk factors. ^,

Ultrasound is the most commonly used imaging method for the detection of high-risk vulnerable plaques. Pathologic changes that occur in carotid plaques were defined by Fisher as neovascularity, calcification, intraplaque hemorrhage, ulceration, and thrombosis.

Studies examining plaque features focus on echogenicity, surface, and ulceration of the lesions. Homogeneous carotid plaques mainly comprised of fibrotic tissue are imaged by ultrasound with lower intensity of echogenicity, and these plaques are rather stable with rare ulcerations. ^, In contrast, heterogeneous plaques consist of matrix deposition, cholesterol accumulation, necrosis, calcification, and intraplaque hemorrhage. Several studies showed an association between heterogeneous plaques and the occurrence of cerebrovascular events. It has been shown that B-mode imaging can determine carotid plaque features that correlate with histopathologic criteria and fibrous cap thickness, the latter being inversely associated with the risk of plaque rupture. ^, These unstable plaques may contribute to the pathophysiology of embolic stroke. Studies investigating endarterectomy specimens demonstrated an association between intraplaque hemorrhage and transient ischemic attacks and stroke. Heterogeneous plaques are associated with intraplaque hemorrhage and neurologic events; carotid plaque morphology assessment may be useful in selecting patients for carotid endarterectomy (CEA).

However, further research was unable to support these findings, and there are reports of little correlation between plaque morphology and histologic specimens. Eventually, the study of Hatsukami and colleagues showed no differences between symptomatic and asymptomatic patients undergoing endarterectomy for highly stenotic carotid lesions in volume of intraplaque hemorrhage, lipid core, necrotic core, or plaque calcification. A systematic review and meta-analysis found that factors related to symptomatic stenosis were plaque angiogenesis, complex plaque morphology, ulceration, echolucent plaques, and intraplaque motion. On the contrary, mixed echogenicity and irregular plaque surface were not correlated with plaque instability. The visualization of carotid plaque features by ultrasound may be diminished by plaque calcification depending on its extent and the localization of the plaque. In conclusion, it remains unclear whether ultrasound echogenicity can reliably distinguish between symptomatic and asymptomatic carotid plaques.

B-mode imaging can assess carotid plaque surface with a relatively good differentiation between smooth, irregular, and ulcerative plaque surfaces compared to postmortem carotid artery specimens. However, the accuracy in studies compared to CEA was poor. ^{, , ,} Furthermore, B-mode imaging showed only 47% sensitivity of yield for ulcerative plaques. Likewise, B-mode also failed to distinguish between the presence and absence of intimal ulcerations. Detection of carotid plaque ulceration with ultrasonography is influenced by the degree of the stenosis, with increasing sensitivity to 77% in plaques with 50% or less stenosis. Plaque ulcerations are prone to thrombosis and thereby cause embolization; however, it is still unclear if or to what extent carotid plaque irregularities or ulcerations increase the risk of carotid embolic event. This assertion is sustained by a study wherein medically treated symptomatic patients with angiographically confirmed ulcerations had increased risk of stroke. Interestingly, many ulcerations are smooth and thick without signs of thrombosis, , however, angiography has poor sensitivity and specificity for identification of plaque ulcerations. According to a pathologic study, asymptomatic plaques with above 60% stenosis have a higher frequency of plaque hemorrhages, ulcerations, and mural thrombi as well as of numerous healed ulcerations and organized thrombi. Similarly, a study comparing asymptomatic and symptomatic CEA plaques found complex plaque structure and complications in both groups. Consequently, there is little difference in plaque structure and surface between patients with symptomatic and asymptomatic carotid plaques. Hence, description of plaque structure and ulcerations appears to be insufficient for prediction of carotid plaque vulnerability; this concurs with guidelines from the American Society of Echocardiography and the Society of Vascular Medicine and Biology.

Another surrogate marker of plaque vulnerability appears to be intimal neovascularization within plaque. Histologically, studies demonstrated that normal vessel intima is without vasculature, and that vascularization is present in IMT and plaque formations. Moreover, histologic studies showed an association between angiogenesis and microvessels in coronary atheroma, unstable angina, and myocardial infarction. Contrast-enhanced ultrasound imaging enables the investigation of the adventitial vasa vasorum. After intravenous microbubble injection, the subsequent identification of microbubbles within the plaque indicates neoangiogenesis; this may help differentiate symptomatic from asymptomatic plaques. ^, Both visual grading and semiautomated techniques have been used to evaluate neoangiogenesis; intraplaque enhancement in contrast-enhanced ultrasound studies correlates well with histologically verified intraplaque neovascularization and hemorrhage as well as macrophage infiltration. ^, Huang et al. proposed a classification method to quantify the intensity of intraplaque enhancement: grade I for plaques with no enhancement, grade II with only arterial wall vasa vasorum enhancement, grade III with arterial wall vasa vasorum and plaque shoulder enhancement, and grade IV with extensive and internal plaque enhancement. In their study, intensity of enhancement corresponding to intraplaque vessel content, and wash-in-time reflecting the speed of flow within the plaque, were significantly associated with stroke risk.

Carotid Artery Stenosis

For the assessment of carotid stenosis by ultrasound peak systolic velocity (PSV), end-diastolic velocity (EDV) and systolic internal carotid artery/common carotid artery (ICA/CCA) velocity ratio are essential parameters ( Fig. 46.2 ). All these measurements must be assessed in the prestenotic, stenotic, and poststenotic segments of the vessel. Moreover, the degree of carotid stenosis can be grossly estimated by B-mode ultrasound. Doppler ultrasonography via the orbital approach enables detection of the retrograde collateral blood flow through the ophthalmic anastomosis in case of severe stenosis or occlusion of the ICA; however, this indirect test is insufficient in up to 20% of cases with hemodynamically significant stenosis.

The Society of Radiologists in Ultrasound Consensus Conference issued the criteria for assessment of ICA stenosis ( Table 46.1 ). The stratification of carotid stenoses is based mainly on Doppler ultrasound. When grading in deciles is required to provide counseling, St Mary’s ratio defined as the ratio of peak systolic ICA to end-diastolic CCA may also be used as proposed by the 2008 Joint Recommendations for Reporting Carotid Ultrasound Investigations in the United Kingdom. CCA occlusion can be manifested either by total occlusion of CCA, ICA and external carotid artery (ECA), or by proximal CCA occlusion with patent ICA and ECA. In the latter scenario, there could be a reversed blood flow in the ECA and antegrade flow in the ICA, or vice versa. Reversed ICA flow indicates competent circle of Willis, able to compensate for CCA occlusion and supply the ECA as well as brain vasculature.

Likewise, stenoses in the carotid siphon or in the middle cerebral artery (MCA) may alter flow in the ipsilateral carotid artery. Moreover, the increases in blood flow velocities in the proximal carotid artery may be rendered by intracranial arteriovenous malformations and shunts. Hence, these findings should lead to further examination to rule out intracranial vascular pathology. At our laboratories, all patients with symptoms of stroke or TIA undergo both carotid duplex and TCD examinations routinely.

Duplex ultrasonography including B-mode and PW Doppler imaging improves the accuracy of carotid artery sampling with spectral Doppler. Doppler frequency spectrum measures blood flow velocities used to estimate the NASCET degree of carotid stenosis. These velocity measurements may be technically difficult if an extensive (>2 cm) plaque calcification is present causing shadowing artifact, thus obscuring the view of the residual lumen. When examining plaques with extensive calcification, indirect findings in the intracranial circulation through transcranial Doppler may aid in the detection of a severe stenosis, but in most of these cases other imagine modalities should be considered.

For the assessment of residual vessel lumen, CDFI and PDI are more accurate than B-mode imaging. Transverse lumen reduction on CDFI correlates with diameter reduction on carotid stenosis angiography. ^, Owing to better visualization of the residual vessel lumen by PDI, assessment of local diameter and area reduction of carotid stenosis is carried out more accurately than by CDFI. ^, 3D ultrasound angiography is a precise imaging method for the quantification of carotid artery atherosclerosis. ^, Color Doppler flow patterns may add information in the assessment of the stenosis in the carotid artery. However, despite these advantages, diameter or area reduction calculations derived from B-mode and flow imaging techniques are not recommended, and velocity criteria and ratios remain the mainstay of grading carotid stenosis with ultrasound. Imaging information is mainly used to determine:

• 1.

  plaque or thrombus presence,

• 2.

  if the residual lumen is less than 50% or greater than 50% narrowed by visual inspection,

• 3.

  complete occlusion seems to be present with no detectable flow within the lumen.

Carotid Artery Dissection

Ultrasonography may contribute to the diagnosis of carotid artery dissection that can manifest with diverse ultrasound findings. Most ICA dissections begin at the skull base and dissect down toward the carotid bifurcation, often being impossible to illustrate through ultrasound imaging. CDFI may record reversed systolic blood flow at the origin of the ICA and absent or minimal diastolic blood flow that concurs with high-resistance bidirectional Doppler signal. In cases where the dissection involves the origin of the ICA, a tapered lumen with a characteristic dural-tail appearance ( Fig. 46.3 ) as well as a floating intimal flap may be detected by B-mode imaging. The true lumen can be oppressed by the false lumen thrombus, and subsequently, a low-velocity Doppler waveform can be imaged. The flow direction in a patent false lumen may fluctuate from forward to reversed or bidirectional. The overall flow dynamics hinges on the thrombus presence within the false lumen, the entry and exit flaps in patent false lumen, the flap wall motion, and the extent of the dissection. In some cases, a retromandibular high velocity waveform may signal a distal cervical carotid artery stenosis. Within a few weeks to months, more than two thirds of patients have successive normalization of ultrasound findings. Moreover, carotid aneurysms may supervene the course of ICA dissection. Dissection of the CCA, though rare, is much easier to illustrate in B-mode, may present with an acute ischemic stroke, and should never be missed as it may arise from the extension of an aortic dissection to the cervical vessels, an absolute contraindication of IV thrombolysis ( Fig. 46.4 ).

Fibromuscular Dysplasia

Fibromuscular dysplasia (FMD) is an idiopathic non-atheromatous, non-inflammatory arteriopathy that most commonly affects the wall of renal arteries, and carotid and vertebral arteries, but it can also occur in almost any artery. The most frequently affected segments in both the carotid and vertebral arteries are their middle and distal portions, which are commonly inaccessible to ultrasound. Indirect Doppler signs of upstream stenosis/occlusion or concomitant anomalies (vascular loops, ectasia) may point to the diagnosis, but an angiography is always required ( Fig. 46.5 ).

Inflammatory Disease of the Carotid Arteries

Takayasu arteritis is associated with concentric intima-media thickening (IMT) on B-mode imaging in proximal cervical vessels (common carotid, vertebral, innominate and subclavian arteries), which is typically diffusely distributed. Of note, 79% patients have bilateral diffuse increase of IMT. Typically, patients with Takayasu arteritis have affected CCA with sparing of ICA and ECA. Active lesions are significantly thicker than inactive ones; however, hyperechogenicity is present in active as well as inactive stages of the course of Takayasu arteritis. Takayasu arteritis may also lead to subclavian artery stenosis manifesting with subclavian artery syndrome ( Fig. 46.6 ).

Giant cell arteritis can present with stroke symptoms, typically of the vertebrobasilar territory. In these cases, the vertebral artery may rarely show hypoechoic wall thickening on cervical duplex ultrasonography. An examination of the superficial temporal artery with high-frequency 12-MHz to 15-MHz B-mode transducers can detect hypoechoic circumferential thickening (the halo sign). The halo sign is moderately sensitive (68%) but highly specific (91%) when present at the superficial temporal artery and can also be used to guide biopsy as well as monitor treatment ( Fig. 46.7 ).

Vertebral Artery Stenosis

Extracranial vertebral artery examination by ultrasound is confined to its origin from the subclavian artery, inter-transverse segments between the third and sixth vertebrae, and the atlas loop. Diagnosis and classification of vertebral artery stenosis is more demanding than of the carotid arteries. However, several studies defined PW Doppler criteria to assess vertebral artery stenosis that are comparable to those in the diagnosis of carotid artery stenosis. Detecting a focal and significant PSV increase is a reliable sign of vertebral stenosis, since tortuosity of the proximal vertebral artery segment, ICA lesions, and vertebral artery asymmetry may result in a relative increase in flow. The velocity increase should be found over a relatively short segment of the vertebral artery with normal or decreased prestenotic and poststenotic velocities. The optimal cutoffs of PSV to identify a proximal extracranial vertebral artery stenosis ≥50% range between 90 and 110 cm/s.

Elongated and multiple stenoses in the vertebral artery may not produce focal velocity elevations, which could be a source of false-negative cervical duplex ultrasonography studies. For assessment of vertebral arteries, it is important to consider the variability of arterial caliber and the presence of numerous collateral pathways that allow supply to the basilar artery even if there is a vertebral occlusion. Flow in the vertebral arteries in over 95% of patients may be quantified by CDFI. This technique also allows recognition of the origin, proximal segment, location of extracranial vertebral stenosis, and the atlas loop. In addition, normal values of flow velocities in the origin, the proximal segment, and the inter-transverse segment have been identified by this imaging method. In order to assess velocities in a vertebral artery it is requisite to know the Doppler values from the contralateral vertebral artery as well as from the carotid arteries since many abnormalities in the contralateral vertebral artery (e.g., aplasia, hypoplasia, stenosis, and occlusion) or severe stenosis in the carotid arteries may be accompanied by altered blood flow in the vertebral artery. Vertebral artery stenoses are most commonly located in the origin from the subclavian artery, while the atlas loop and the intracranial segment are affected less frequently. Finally, stenoses in the inter-transverse segments are less common.

Vertebral Artery Dissection

Diagnosis of the vertebral artery dissection in the V2 through V4 segments is challenging since there are no defined salient ultrasonographic findings. Dissection of the atlas loop may by demonstrated by absent blood flow, low bidirectional flow, or by low post-stenotic blood flow. The stenotic segment within dissection of the V1 segment may be detected by ultrasound. Absent blood flow in the inter-transverse segments may be evidence of vertebral dissection. Moreover, at the same level the sonographer may detect a localized broadening of a vessel diameter with hemodynamically present stenosis or occlusion. ^, Rarely, ultrasound imaging may provide a direct finding of intramural hematoma, but these are usually missed, especially if they cause no hemodynamically significant stenosis or if they are located outside the arterial segment. ^, For all patients with stenosis or occlusion, ultrasound imaging is a convenient method to follow up vertebral artery dissection; it does not, however, identify vertebral artery pseudoaneurysms. In order to assess the length of dissection, TCD plays a useful role. In conclusion, if there is a suspicion of dissection, further imaging work-up should be performed even if ultrasound results are negative.