regard to the proper technique of obtaining velocity waveforms at sites of stenosis. In areas of mild to

moderate stenosis, use of a Doppler angle of 60 degrees to the long axis of the vessel is recommended.

However, in areas of more severe stenosis and/or wall abnormalities, the Doppler angle of 60 degrees

should be defined by the long axis of the stenotic flow jet, as demonstrated by color flow.

Figure 87-4. A and B: In the presence of innominate artery very high-grade stenosis or occlusion the common carotid artery (CCA)

waveform will be blunted with a decreased amplitude and slowed upstroke (A), and there can be actually reverse flow (B) in the

ipsilateral internal carotid artery (ICA).

Figure 87-5. Internal carotid artery (ICA) waveform from the site of a >80% ICA stenosis. Both peak systolic and end-diastolic

velocities are very elevated.

The sample volume size should be kept as small as possible, usually 1.5 mm, to detect discrete

changes in flow velocity. This is important as the highest velocities may be localized to a small area in

the flow stream that emanates from the stenosis. A large sample volume that incorporates flow from

many points within the vessel in the generation of the spectral waveform may give the false impression

of disturbed flow, potentially leading to the misdiagnosis of moderate disease in an otherwise normal

vessel. In practice, the sonographer, having identified the stenosis, gently moves a small sample volume

around until the point of highest velocity is found.

Damping of spectral Doppler waveforms may be seen in the region distal to the carotid stenosis when

the lesion is severe enough to be flow reducing. The most common abnormality seen distal to a carotid

stenosis is spectral broadening caused by disturbed blood flow or turbulence. At best, poststenotic flow

disturbance is a qualitative measure of arterial stenosis; nevertheless, its detection is important. With

proper gain settings, “fill-in” of the Doppler spectral waveform generally indicates the presence of

carotid stenosis with diameter reduction of at least 50%. However, this level of disturbed flow

occasionally can be seen with nonstenotic disease. Diagnostically, the most significant poststenotic flow

2490

disturbance produces a simultaneous forward and reverse spectral Doppler signal, accompanied by poor

definition of the upper spectral border. Such disturbed flow implies the presence of severe carotid

stenosis. Severely disturbed flow distal to a highly calcified plaque may be the only substantial evidence

for the presence of clinically significant stenosis if calcification prevents direct insonation of the

stenosis.

Figure 87-6. Bony landmarks of the vertebral bodies (arrow) mark the course of the vertebral artery.

Vertebral Artery

On rare occasions, a stenosis may be found in the cervical portion of the vertebral artery associated

with cervical osteoarthritis. However, the origin of the vertebral artery from the subclavian artery is by

far the most common site of disease in the vertebral artery. The vertebral artery origin lies deep in the

base of the neck and may be difficult to access with ultrasound. Mean vertebral artery diameter is about

4 mm but the vertebral arteries are frequently asymmetric in size with one, most commonly the left,

being larger than the right.

The vertebral artery is most commonly interrogated with ultrasound further distally in the neck, from

an anteroposterior window, as it threads through the transverse processes of the cervical spine. The

vertebral bodies serve as a reference to insure the vertebral artery is actually under examination (Fig.

87-6). The artery is usually seen deeper but adjacent to the vertebral vein. Color Doppler is helpful to

locate the vessel. A spectral waveform from the vertebral artery in the mid-neck provides information

about direction of flow, waveform shape, and velocity but does not rule out disease at the origin. The

normal vertebral artery waveform is similar to that of the ICA with normal PSVs reported to be 20 to

40 cm/s.6 Velocities up to 80 cm/s are, however, frequently seen without apparent clinical importance

and may represent collateral flow through a dominant vertebral artery, or a small but disease-free,

vertebral artery. Evaluation of disturbed flow distally may help determine which elevated velocities are

associated with a vertebral artery stenosis and which are not. Velocity patterns are usually similar in the

two vertebral arteries but systolic and diastolic velocities may differ if vertebral artery diameters are

asymmetric. For this reason if there is concern for stenosis based on an elevated peak systolic velocity

(PSV) in a vertebral artery, recording of vertebral artery diameters is important.

Flow in the vertebral artery is normally antegrade with a rapid upstroke and continuous diastolic

flow. In cases of anatomic subclavian steal, color flow provides an important initial clue to the diagnosis

as flow in the artery will be retrograde in the same direction as the vein. Spectral Doppler must still be

used to verify arterial flow and will demonstrate reverse or bidirectional flow in cases of subclavian

steal. Depending on the degree of subclavian stenosis the vertebral artery flow may be reversed or

biphasic, with flow above and below the baseline (Fig. 87-7A,B). End-diastolic flow will be zero or near

zero reflecting the high resistance of the peripheral arm circulation rather than low-resistance cerebral

circulation. High-resistance vertebral artery flow may also be seen with antegrade flow and like the

situation for the ICA may represent distal stenosis or occlusion of the vertebral artery or an atretic or

hypoplastic distal vessel.

2491

Figure 87-7. A and B: Both reverse vertebral artery flow (A), and vertebral flow both above and below the baseline (B) indicate a

proximal innominate or subclavian artery stenosis.

External Carotid Artery

The ECA is smaller in diameter than the ICA at the level of the carotid bulb but similar in diameter

beyond the bulb. It has little clinical significance in most cases but can serve as an important source of

collateral flow to the brain in cases of ICA very high-grade stenosis, or occlusion. The ECA may also

serve in such cases as a conduit for emboli to the brain; the so-called carotid stump syndrome.7

The ECA waveform has a sharp upstroke, a prominent dichroitic wave in late systole or early diastole,

and velocity that is near or at the zero base line in end-diastole. The PSV of the ECA is normally higher

than the ICA. The ECA may adopt the characteristics of the ICA in end-diastole as the resistance in the

face and scalp decreases with temperature change and/or in the presence of disease. In cases of common

carotid occlusion with preservation of patency of the carotid bulb, reverse flow in the ECA can provide

antegrade flow in the ICA maintaining patency of the ICA even with CCA occlusion.

Classifying Carotid Stenosis

Duplex criteria for quantifying carotid artery stenosis were developed by comparing duplex-derived

spectral waveforms and contrast arteriograms. Duplex-derived categories of stenosis are relatively broad.

Sensitivities and specificities for spectral analysis of duplex-derived waveforms for detecting an ICA

stenosis of >50% to 99% are between 90% and 95%. There are numerous spectral criteria for

classifying ICA stenosis. Some focus on categories of stenosis, whereas others focus on threshold levels of

stenosis.

One of the most widely accepted classification schemes for categories for ICA stenosis was developed

at the University of Washington under the direction of Dr. Eugene Strandness. These criteria were useful

in the study of the natural history of carotid atherosclerosis and in clinical practice. In the University of

Washington system, velocity waveform analysis and spectral criteria were used to classify ICA

angiographic stenosis as normal, 1% to 15%, 16% to 49%, 50% to 79%, and 80% to 99% stenosis and

occlusion. Prospective validation of these criteria has demonstrated an overall agreement of 82% with

contrast angiography. The ability of the criteria to detect carotid disease is 99% sensitive and the ability

of the criteria to recognize normal arteries is 84% specific.3

Criteria for detecting carotid artery stenosis scanning have undergone reevaluation to remain relevant

to current clinical practice. This reevaluation was stimulated by the randomized trials testing the

efficacy of carotid endarterectomy (CEA) that took place over the last two decades (Table 87-1).8–12 The

trials had a profound impact on validating the indications for CEA in patients with carotid bifurcation

atherosclerosis. The trials identified significant benefit, in terms of stroke reduction, for CEA in patients

with specific levels of ICA stenosis. In particular, patients with symptomatic ICA stenosis >70% to 99%

2492

had dramatic benefit from CEA, whereas patients with symptomatic ICA stenosis between 50% and 69%

and patients with asymptomatic ICA stenosis between 60% and 99% also benefited, albeit to a lesser

extent, from CEA.

In the North American trials of CEA, ICA stenosis was calculated from arteriograms by comparing the

diameter of the minimal residual lumen to the diameter of the distal cervical ICA.13 The University of

Washington duplex criteria for categories of stenosis pre-existed the endarterectomy trials and were

developed comparing the diameter of the residual ICA lumen at its narrowest point with an estimate of

the diameter of the ICA bulb if it were free of atherosclerosis. Because the bulb has a greater diameter

than the distal ICA, the two methods of measurement do not give the same calculated percentage of

angiographic stenosis for the same lesion. Calculations of angiographic stenosis using the distal ICA as

the reference vessel result in lower calculated stenosis percentages than calculations using the bulb as

the reference site. This effect is particularly striking for modest lesions.

Table 87-1 Major Randomized Trials Assessing Efficacy of Carotid

Endarterectomy

In a review of 1,001 internal carotid angiograms, 34% of the ICAs were classified as stenosis of 70%

to 99% using the ICA bulb as the reference vessel. In contrast, when the distal cervical ICA was used as

the reference site, only 16% of the ICAs were classified as 70% to 99% stenosis.14 More than 99% of the

distal ICA-based calculations of stenosis were less than bulb-based calculations. Thus, the duplex stenosis

criteria using the bulb as the reference vessel are not and were not directly applicable to the results of

the clinical trials.

Current Criteria for ICA Stenosis

Since the randomized CEA trials were completed, additional duplex criteria were developed comparing

duplex scans to angiographic ICA stenosis using the distal ICA as the reference vessel in calculating

angiographic stenosis. Such criteria are considered more useful by many in selection of patients for

carotid intervention because they are directly applicable to the threshold levels of carotid stenosis

addressed in the CEA trials.

The initial studies addressing the issue of duplex criteria relevant to the CEA trials were performed at

the Oregon Health & Science University (OHSU)5,15 with subsequent publications from many different

institutions, with many different proposed criteria for identification of clinically relevant threshold

levels of ICA stenosis.16–21

Recognizing that duplex criteria from different centers differed for the threshold levels of

angiographic stenosis determined by the CEA trials, a panel of authorities from a variety of medical

specialties assembled to review the carotid ultrasound literature. The panel developed a consensus

regarding the key components of the carotid ultrasound examination and reasonable criteria for

stratification of ICA stenosis.22

The consensus committee recommended that all carotid examinations be performed with grayscale

imaging, color Doppler, and spectral Doppler. Examinations should be performed by a credentialed

vascular technologist in accordance with the standards of an accrediting body. Doppler waveforms

should be measured with an insonation angle as close to 60 degrees as possible but not exceeding 60

degrees, and the sample volumes should be placed within the area of maximal stenosis. The panelists

recommended the consistent use of relatively broad diagnostic strata to estimate the degree of ICA

stenosis. The panel also concluded that Doppler is relatively inaccurate for subcategorizing ICA stenosis

<50%, and recommended that these lesions be reported under a single category as <50% stenosis, and

that subcategories for minor degrees of stenosis not be used.

4 The consensus panel noted that PSV is easy to obtain. However, data suggest reproducibility of PSV,

2493

even among experienced vascular technologists, has sufficient problems that PSVs should not be used as

a continuous variable in clinical carotid duplex scanning. Even so, the degree of stenosis estimated by

ICA PSV and the degree of narrowing of the ICA lumen seen on grayscale and color Doppler should

correlate with PSV as the primary parameters for determining ICA stenosis. Additional parameters such

as ICA/CCA PSV ratio and ICA EDV are secondary parameters and should be employed as internal

checks. They are especially useful when ICA PSV may not be representative of the extent of disease.

After their discussions the consensus panel recommended criteria, stratifying ICA stenosis into specific

categories relevant to the CEA trials (Table 87-2). These criteria have not been subjected to widespread

retrospective or prospective evaluation and do not represent the results of any one laboratory or study.

They are not meant to serve as a substitute for continuous quality assurance in individual laboratories.

BILATERAL HIGH-GRADE ICA STENOSIS

Doppler-derived flow velocities from the ICA opposite an ICA occlusion or high-grade stenosis may

suggest a higher degree of narrowing than is observed angiographically.2 This likely is because of

compensatory flow. Duplex scan overestimation of stenosis is more common in less-severe categories of

stenosis than in higher-severity categories.23

Table 87-2 Consensus Panel Recommendations for Classification of Internal

Carotid Artery Stenosis22

2494

Figure 87-8. Color flow image and velocity waveforms from a stented internal carotid artery. A peak systolic velocity of 120 cm/s

indicates the stent is widely patent. The stent itself is easily recognized as a hyperechoic stripe (arrow) parallel to the vessel wall.

Stented Carotid Arteries

Carotid artery stenosis may be selectively treated with an intraluminal stent. However, ultrasound

criteria developed for native ICAs are likely not applicable to stented carotid arteries, especially to

more modest lesions in the stented arteries. With rare exceptions the number of patients in studies of

stented carotid arteries that have actually had severe in-stent restenosis is small. No study has correlated

the degree of stenosis or increase in ICA velocities within a stented carotid artery with clinical

symptoms or outcomes. Studies have also not evaluated the effect of the stented carotid artery on the

opposite nonstented artery. It is generally agreed that PSVs >125 cm/s, will be needed to identify a

>50% stenosis in a stented ICA. Criteria to detect high-grade stenosis in native ICAs still work

reasonably well to identify high-grade stenosis in stented ICAs (Fig. 87-8).24,25

CCA and ECA Stenosis

Criteria used for classifying disease in the ICA have not been prospectively tested for application to the

ECA or the CCA. However, as with the ICA, relative degrees of stenosis may be determined by the

presence of plaque with B-mode imaging, aberration in color flow on duplex examination, spectral

broadening, and increases in PSV. Greater than 50% stenosis can be inferred by the presence of a focally

increased PSV followed by poststenotic turbulence. As noted above, the CCA waveform normally has

attributes of the ICA and ECA. The CCA will take on the quality of the “normal” vessel (ICA or ECA)

when the other is occluded. If there is a proximal CCA (or innominate artery) high-grade stenosis or

occlusion, the ipsilateral CCA Doppler flow quality will be dampened with low PSVs and delayed

upstroke of the waveform compared with the nonstenotic contralateral side. Poststenotic turbulence

also may be seen. There are no widely employed validated criteria to give a diameter reduction for

stenosis in the CCA or ECA. Greater than 50% stenosis in the ECA is often inferred by PSV >125 cm/s

associated with poststenotic turbulence whereas a PSV >200 cm/s may be an appropriate threshold for

>50% stenosis in the CCA.26

LOWER EXTREMITY ARTERIAL DISEASE

The noninvasive vascular laboratory, in combination with the history and physical examination, has a

critical role in providing objective diagnosis of lower extremity arterial occlusive disease in

asymptomatic patients, those with intermittent claudication, and in those with critical arterial

insufficiency. Both plethysmographic and ultrasound techniques are used in the evaluation of peripheral

arterial disease (PAD).

In many patients the clinical history and physical examination are all that is necessary to establish a

diagnosis of intermittent claudication (IC). Almost all patients with IC have diminished or absent lower

extremity pulses. However, palpation of pulses is subjective and poorly reproducible. Therefore, the

2495

pulse examination in a patient suspected of having arterial disease should be supplemented with

objective testing in the noninvasive vascular laboratory. On occasion a patient will give a history

suggesting IC, yet have what appear to be normal resting lower extremity pulses. In such cases, exercise

testing with postexercise Doppler-measured ankle pressures and ankle–brachial systolic blood pressure

ratios is crucial to confirm the diagnosis of IC secondary to arterial stenosis or occlusion (see below).

Many patients with PAD have atypical leg symptoms or are asymptomatic. Such patients have a

similar increased risk of cardiovascular death as patients with typical IC symptoms, leading some to

advocate for screening for PAD in patients with atypical leg symptoms or atherosclerotic risk factors.

Patients with exercise-induced lower extremity and/or buttock pain should be asked about the

location of their pain, its relationship to walking, severity and duration of symptoms, and symptom

progression over time. Only exercise-induced muscular pain of the calf, thigh, or buttock, relieved

within a few minutes of rest and reproduced by additional walking, can be predictably improved by

lower extremity arterial revascularization. There are no data on the response of atypical leg symptoms

to revascularization in patients with evidence of PAD.

Ischemic pain at rest is a clinical diagnosis. It is suspected when a patient complains of pain and/or

numbness in the forefoot, toes, or instep at rest. In such patients the vascular laboratory can objectively

confirm and quantify the magnitude of arterial insufficiency. Ischemic pain is generally associated with

an ankle pressure <50 mm Hg and an ABI ≤0.4.

PLETHYSMOGRAPHIC TECHNIQUES

Volume Flow

Calf or foot blood flow can be recorded with a mercury-in-silastic strain gauge. Electrical resistance of

the mercury column depends on the length of the column. Changes in electrical resistance, reflect

changes in the volume of the extremity, and are based on detection of minute changes in the length of

the column. However, neither calf nor foot blood flow at rest differ between normal subjects and

patients with even rather severe IC.27 Hyperemic volume flow is often lower in patients with occlusive

disease but such testing can be quite painful for the patient. Measurements of volume flow therefore are

not very useful in the evaluation of lower extremity ischemia.

Pulse Volume Recordings (PVR)

5 Air plethysmography can be used to display pulse volume waveforms. PVRs are obtained with

partially inflated blood pressure cuffs that detect volume changes sequentially down a limb. Volume

changes beneath the cuffs resulting from the pulse wave result in small pressure changes within the

cuffs. These changes are displayed as arterial waveforms with the use of appropriate transducers. A

normal pulse volume waveform has a sharp systolic upstroke and peak, as well as a prominent dichroic

notch on the downward portion of the curve. With increasing proximal arterial occlusion, the dichroitic

notch is lost and the waveform bows out on the down portion of the systolic portion of the curve. The

peak of the pulse wave also becomes rounded and there is loss of amplitude of the wave as disease is

more severe. With even more severe disease there are nearly equal upstroke and downstroke times and

in very severe proximal disease the pulse wave is absent (Fig. 87-9).28,29 An absence of quantitative data

limits the utility of PVRs. More quantitative data for the evaluation of lower extremity arterial disease

are available using ultrasound-based techniques.

2496

Figure 87-9. Pulse volume recordings. The plethysmographic waveforms are normal on the right and severely abnormal in the left

leg with loss of amplitude of the waveform, slowing of the upstroke of the waveform and outward bowing of the downstroke of

the waveform.

ULTRASOUND TECHNIQUES

Ankle–Brachial Index

6 The ankle–brachial index (ABI) is a very useful measure of overall lower extremity perfusion. The test

is performed with the patient supine and having rested for a few minutes. A pneumatic pressure cuff is

placed just above the ankle and inflated to supra systolic levels. A continuous wave Doppler probe is

positioned over the posterior tibial or dorsalis pedal artery distal to the cuff. The cuff is deflated and the

pressure reading when the Doppler signal returns is recorded. The sequence is repeated for the other

ankle artery and the values compared with the highest brachial artery systolic pressure also obtained

with the Doppler. For clinical purposes, the higher ipsilateral dorsal pedal or posterior tibial pressure is

divided by the higher brachial artery systolic pressure, yielding an ABI for that lower extremity. The use

of a ratio makes the test relatively independent of day-to-day variations in arterial blood pressure.

A normal ABI is 1.0 to 1.2, with progressively lower values corresponding to worsening arterial

disease (Table 87-3). ABIs below normal indicate relative lower extremity arterial insufficiency and

increased risk of cardiovascular death with the risk increasing with the magnitude of depression of the

ABI. This test has several limitations. Significant bilateral subclavian or axillary artery occlusive disease

results in a falsely elevated ABI. Calcified tibial arteries that can frequently occur with diabetes or renal

failure may be inadequately compressed by the pressure cuff. This also results in a falsely elevated

(supra systolic) ankle pressure. An ABI >1.4 should therefore also be considered abnormal. Such

patients often have severe arterial disease and are also at increased risk of cardiovascular death. When

the ABI may be falsely elevated, qualitative analysis of Doppler-derived analog or plethysmographic

waveforms or measurement of digital systolic pressures is a more appropriate indicator of the arterial

status of the lower extremity. Other problems with ABI include confusion with venous signals when the

ankle arterial pressure is low or unmeasurable and relative insensitivity to certain patterns of

progression of arterial disease. A tibial artery can occlude without a change in ABI if the remaining

2497