Carotid Sonography: Doppler Evaluation and Waveform Analysis - SD

Introduction

My name is Mark Cleaver.
I'm at the University of Wisconsin at Madison,
and I'm going to talk about the basics
of carotid sonography.
I'm going to discuss the basics of carotid sonography,
and in particular the doppler
and grayscale evaluation of plaque.
I'd like to acknowledge
Leslie Scout from Yale University for additional information
that she has provided in this lecture.
I'll talk first about the technique of carotid sonography,
something about the gray scale assessment of plaque
and the vessel wall spectral doppler analysis,
color doppler, and end with some pearls and pitfalls.

Technique of Carotid Sonography

First, the technique in general,
we use a high frequency transducer in the five
to 12 megahertz range
with a patient's supine in the head on a
bolster with the chin up.
I like to turn the chin away from the side
of which I'm scanning to expose the carotid artery
to maximum advantage.
The examiner is usually to the side,
but some people prefer to examine from the head of the bed.
Here's an example of how most people do it.
The transducer is placed on the neck with the trans,
with the patient's sitting at the side of the patient.
The success of the technique depends on several factors,
but it is most important
that the patient can cooperate in some way.
If a patient has had perhaps an acute
brain injury, they may become agitated
or demented and be unable to cooperate further.
If they were to fall asleep
and to snore, then useful information is
difficult to obtain.

Gray Scale Assessment of Plaque

The gray scale assessment
hinges on several things.
One is what is the plaque characteristics,
what's the symmetry of the plaque, what are some
of the surface features of the plaque?
And then finally, present stenosis.
To remind you that the wall
of the carotid artery should be thin in many Western countries.
You'll see what is known as
intimal medial thickening, which can be up
to 1.2 millimeters and still be normal.
Exceeding that threshold is considered abnormal.
This is an example of the intimal medial thickening that
I'm describing at times.
Instead of focal plaque, you'll see a concentric thickening
of the vessel wall.
And there are several things that can give this appearance.
Radiation Takayasu's, arteritis, temporal arteritis,
and fibromuscular dysplasia.
Frequently, this is seen in the common carotid artery.
You can see in this image the wall thickening here
that surrounds the carotid artery.
In this case, the patient had Takayasu's arteritis,
which involves the major vessels from the aorta.
Here's an example of fibromuscular dysplasia.
Again, this is the carotid artery.
And you see the wall thickening.
Here is the classic appearance on angiography,
the beaded appearance of the
internal carotid artery in this case.

Plaque Characteristics

So plaque, focal plaque can have different echogenicity
that can be homogeneous or heterogeneous.
In general, heterogeneous plaque is associated
with instability because that suggests the presence
of intra plaque hemorrhage.
Here is the carotid artery.
Here you can see this plaque protruding into the lumen
vessel and the low echogenicity center indicating intra plaque hemorrhage
and plaque instability.
In general, a hard plaque is more stable than soft
plaque, and by hard plaque, I mean plaque that has a high
level of calcification.
So this would be an example of hard plaque,
and you can see the shadowing associated with
that calcification posterior to the plaque.
One thing about hard plaque is that we,
the shadowing will make segments of the vessel inaccessible
to Doppler interrogation.
If the sound waves are not reaching it
to create a gray scale image, they are neither are they
reaching it to create doppler images.
So no gray scale, no doppler, people in the field.
If you are trying to interrogate that segment of the carotid,
frequently find it helpful to interrogate the vessel from a
different angle, so, or a different perspective.

Surface Characteristics of Plaque

The surface characteristics
of the plaque has been identified
as an independent stroke risk.
The surface characteristics are
however difficult to identify with certainty,
and the characterized in some cases.
But as a rule, smooth plaque is
more stable than irregular plaque.
An irregular plaque is the kind of plaque
that can give off emboli and result in embolic strokes.
Ulceration is seen as a deep pit or a break in the plaque,
and this can be difficult to identify.
Here is an example of a plaque ulceration.
Here's the plaque. Here's the flow into the heart
of the plaque, indicating an ulceration.
Frequently in the presence of ulceration
or plaque irregularity, you'll see eddies or abnormal
or unusual patterns of blood flow.
Here's a color Doppler image
with three different sampling sites, one, two, and three.
And you can see that both the direction
and the character of flow changes radically.
As we sample different portions of this
vascular segment.

Measurement of Stenosis

The grayscale assessment is usually done both in transverse
and longitudinal planes.
It's important to identify certain landmarks,
notably the carotid bulb in the carotid bifurcation,
and to identify the individual components,
the common carotid, internal carotid,
and external carotid arteries.
When we do see plaque, we often measure a percent diameter
and assess for area stenosis.
To remind you a bit about
how these things are typically reported, usually
stenosis are reported in terms of a percent diameter stenosis,
which is not the same
as cross-sectional area reduction
because a 50% diameter stenosis corresponds
to about a 70% area reduction.
And similarly, a 70% diameter reduction
corresponds to critical stenosis
or severely flow limiting lesion.
This type
of stenosis is typically measured in the transverse plane,
but beware of asymmetric or soft
and echolucent plaque, which can be missed
in general power Doppler if available.
Can define the margins of the plaque with greater
precision than color doppler?
When you do measure the luminal diameter narrowing,
typically there's two ways of doing it.
Most programs do it using what's called the NASCET criteria.
That's an acronym of N-A-S-C-E-T,
and the NASCET criteria is based on a
endarterectomy trial, the North American
stroke treatment and endarterectomy trial.
And it's measured by measuring the luminal diameter
of the vessel at the stenosis
and dividing that by the diameter
of the normal downstream ICA.
So the by criteria, the luminal stenosis would be measured
as a ratio of the diameter A over the diameter C.
Some people have embraced the concept
of measuring what's known
as a residual lumen at the stenosis.
In general, the diameter of the ICA is five
to six millimeters,
and if there is a residual luminal diameter of two millimeters
or less than the lesion is in the surgical range,
that is the greater than 70% diameter reduction.

Spectral Doppler Analysis

Spectral doppler in a normal
and unaffected vessel.
The flow within the carotid artery is known as laminar.
This is a even parabolic flow.
You can see the diagram here
with a relatively narrow range of velocities.
And this narrow frequency spectrum is illustrated in the Doppler wave form
as a narrow envelope of velocities
that enclosed what's sometimes called the acoustic window or a spectral window here,
an area
of relatively little blood flow here in the lower frequency ranges.
So it's an envelope of higher frequencies.
However, you don't see parabolic
or laminar flow a condition known
as spectral broadening.
This indicates disturbed flow with a larger range
of velocities, and again, a filling in of this spectral
or acoustic window, because now we have representative blood cells moving at different velocities
from baseline all
the way to the peak envelope of velocities.
Graphically, it's illustrated as a filling in
of the spectral waveform here,
and also as multiple color flurries on color doppler imaging.
If you do see spectral broadening,
the most common cause would be focal atherosclerosis.
But there are other things that can give you the same result,
namely the disturbed flow
that results from an arterial venous fistula,
disturbed flow at an endarterectomy site.
The presence of tortuous vessels
and also spectral broadening is common
and normally seen in small vessels.
Here's an example of an arterial venous fistula
and the spectral broadening at the sampling of the fistula,
where you can see this multiple levels of velocity being represented here
on the spectral waveform velocities,
and the waveform character can change radically
at the endarterectomy site.
Oftentimes, this has to do
with decreased vascular compliance,
but the presence of both elevated velocities in,
and high peak systolic velocities,
and a spectral broadening phenomenon is common at the site
of prior surgery.
Tortuous vessels are notorious
for causing spectral broadening.
Oftentimes, particularly in older patients,
you'll see vessels that make abrupt turns.
And as you could see on this color image at the abrupt
turns, the flow dynamics are changing rapidly.
Here's an example of sampling at prior to a turn,
and then after the turn, and you can see that the acoustic
or spectral window is filled in after the turn
because we have a broadening of the velocity spectrum or this spectral broadening that
I'm describing.
Not all spectral broadening is easy to identify.
There are pitfalls in the identification
of spectral broadening, specifically, if the gain setting
of the doppler is not set correctly, you can fill in
that acoustic or spectral window artifactually.
Secondly, if you don't center
your sampling gate within the lumen of the vessel,
but rather have it eccentric in the lumen, say close
to the wall, you can get spectral broadening on the basis
of detected wall motion of the vessel.
And finally, this can be normal when the vessel abruptly changes caliber
as it does as it enters the carotid bulb.
Here's an example of wall motion.
You can see that my range gate is opened up widely
to encompass both the center stream
and the periphery of the vessel, as well as the wall itself.
And you can see a lot of spectral broadening here.
If you sample outside the center stream,
here's center stream sampling,
you do have an acoustic or spectral window.
If you sample closer to the wall,
you'll get the spectral broadening phenomenon.

Spectral Doppler Criteria for Carotid Stenosis

Now, we'll talk a little bit about spectral Doppler criteria
and the diagnosis of carotid stenosis.
There are different parameters that are used,
the most common being peak systolic velocity
and diastolic velocity,
the ICA-CCA ratio both in the peak systole
and in diastole.
And then finally, the presence
or absence of spectral broadening.
There is a useful reference that where this
criteria is tabulated.
I prefer two publications of this journal,
of this article, both in radiology
and in ultrasound quarterly.
It is a consensus conference that was produced
by the Society of Radiologists and Ultrasound.
The criteria that was agreed upon at
that consensus conference is like this,
a diameter stenosis in the ICA of less than 50%, which is
to say a non-significant stenosis
Is identified when the velocity,
the peak systolic velocity is less than
125 centimeters per second.
The end diastolic velocity is less than 40
centimeters per second.
And the ICA-CCA ratio at peak systole is less than two
for diameter stenosis in the 50 to 69% range,
which is a borderline range that exists
between non-significant stenosis
and stenosis that is non-surgical, is identified
with a peak systolic velocity in the 125
to 230 centimeters per second range
and end diastolic velocity in the 40
to 100 centimeters per second range.
And an ICA-CCA peak systolic ratio between two
and four, then diameter
stenosis that would be considered in the surgical range
or critical stenosis that is greater than 70% is identified
by peak systolic velocities exceeding 230 centimeters per
second and end diastolic velocity exceeding 100
centimeters per second.
An ICA-CCA peak systolic ratio that exceeds four.
It is interesting to note that the,
though you may not have the criteria immediately in front
of you, but you can make a general estimate
of degree of stenosis, remembering that the peak systolic velocity
less than 125 is insignificant.
But if you double that factor to, let's say,
around a 250, 230, then you're in the area
of critical stenosis.
Similarly, you don't get into the area
of significant stenosis until about 40 centimeters per second.
And then if you double that, that's about 80
or a hundred centimeters per second.
And then you're in the range of critical stenosis.
And similarly, the ICA-CCA ratio starts
to become significant at two
and becomes critically significant at four.
So if you remember a hundred and twenty five, forty
and two, you can remember really the essence of this table.
Now, remember though that you have to look at everything on the image.
It's not enough simply to look at numbers.
It's very important to look both at the color
and the gray scale, because there will be a high grade
stenosis above which the flow is limited,
and the velocity will decrease.
And this chart shows it
as the diameter stenosis gets more severe velocity increases
because of that jet effect at the stenosis,
but then we'll drop off precipitously when
that stenosis starts to become occlusive.
So don't simply look at the peak systolic velocity
and make your determination from that alone
that would be a mistake.

Color Doppler Factors in Carotid Stenosis Assessment

All right, now to talk about some of the color doppler factors to consider in the assessment of carotid stenosis.
First of all, we'll talk about detection
of luminal abnormalities, the presence
or absence of helical flow and boundary layer separation.
And finally, some limitations of the carotid doppler analysis.

Intraluminal Bands

First of all, the intraluminal band are sometimes seen in
the performance of carotid doppler.
And a band that crosses the lumen
of the carotid artery can result from
basically three sources.
The most common being a free edge of atherosclerotic plaque.
It can also indicate dissection
or the presence of fibromuscular dysplasia and luminal webs.
This is an example of a free edge of atherosclerotic plaque
that is waving within the lumen.
And we've done an m mode showing the motion of
that free edge here.
Dissections, carotid and arterial dissections are usually seen as thin septations
or bands as shown here in this carotid artery
that shows different flow profiles on either side.
In other words, if you put a spectral doppler cursor on this
side of the band
or septation, you'll get a different appearance to
that waveform than you will see on the other side.
In general, acute dissections can be very difficult to see
because it is simply a very thin layer of intima
that's raised in the lumen.
In general, these become more visible
as fibrin deposits on the dissection
and the dissection becomes chronic.
Here is an example of another dissection, quite difficult
to see on gray scale, but on color doppler, the presence of
that septation becomes more visible.
And if you were to sample on either side,
you would get different flow profiles.
Another example. Now this is a chronic plaque.
You can see that this is more echogenic here,
And here it is on longitudinal than that.
And you can see one of the flow profiles there,
and if you sample it, you often get this,
what some people call the stump thump phenomenon.
Anyway, short percussive thumping waveform here, which is both muted
and low velocity
fibromuscular dysplasia can, one of the manifestations
of fibromuscular dysplasia is a web
that is seen protruding into the lumen.
And this is an example of a young woman who had a stroke with no history of trauma or disease.
And so if you see this in a young patient, you have
to think about the presence of fibromuscular dysplasia.

Helical Flow

Helical flow is a phenomenon that's commonly seen
and normally seen in many patients.
This is flow that is helical in the sense
that it has two components of motion.
It is moving forward,
and it is moving in a circular pattern within the vessel
lumen that's demonstrated here on these transverse images.
And this case flow is going in a clockwise direction.
You can see that in tr
but it is also moving, of course, cephalad towards the head
if you angle your transducer to demonstrate it.
So it is going in two different directions.
It's going in a circular motion,
and it is going forward both in that helical pattern.
If you sample the helical flow
and longitudinal on either side of the vessel, you'll get,
you can pick up that component
of circular flow.
It will look like it's going away from you at one point, one portion of the vessel.
And on the opposite side,
it'll look like it's coming towards you,
which in fact it is.

Boundary Layer Separation

It is not uncommon to see flow reversal,
particularly around the bulb.
This is a phenomenon that some people refer to
as boundary layer separation.
And it is a consequence of the fact that as a laminar flow enters a capacious area
or a suddenly widened area such as the capacious
or the enlarged bulb, you'll get flow reversal at the periphery of the vessel.
As that flow at the periphery, it flows
to the point of flow reversal.
Some people have so much that suggested
that the absence of this boundary layer separation
phenomenon at the carotid bulb itself indicates the presence
of disease at this location.
Here's an example of that. Here's the color doppler.
You can see this is true flow reversal
and depicted here as a flow
below the baseline in this spectral doppler tracing.

Pearls and Pitfalls

I'll conclude the lecture with a series of pitfalls
and pearls.
First of all, there can be some variability
of doppler measurement based on technique
instrument variability.
In other words, one machine to another
physiologic differences between patients.
And finally, vessel and plaque morphology.
There is quite a lot of controversy about the need
for a uniform angle from one study to another.
Another, suffice it to say that many people insist
that the color doppler angle should be the
same for serial studies.
And in general, those people advocate a 60 degree
angle of insonation
that the angle matters is indisputable.
If you measure here
and you change your angle, you can get a change from 140
to 195 centimeters per second.
So that is quite a large change just
by changing the angle of insonation.
So you want to keep that as uniform as you can.
Remember that the angle of insonation radically affects the reliability of the measurement.
Measurements become much more variable
when the doppler angle exceeds 60 degrees.
So all measurements of carotid velocity elevation
or velocity should be obtained at 60 degrees angle
of insonation or less.
There has been some controversy about whether the
doppler angle should be parallel to the wall
or parallel to the jet, remembering that the plaque,
of course, is frequently asymmetric
and creates jets that are not parallel to the wall,
but rather directed towards the wall.
And the consensus is
that the doppler angle should always be
measured parallel to the wall.
Remember that gain can cause problems.
Here's an example of two plaques.
This one looks like it has intra plaque hemorrhage,
but if you turn up the gain even a little bit, you can see
that there is quite a lot echogenicity within the plaque,
and that this is not the intra plaque hemorrhage
that it appears sometimes the gain can
be adjusted too high.
This is an example of that.
This is gain that is adjusted way too high,
but when shown in color doppler what appears to be plaque
is clearly not present.
Finally, spectral doppler gain can affect the measured velocity.
If the gain is set too low, you can not detect some
of the faster velocities within the vessels.
And here, as I adjust the gain down lower, you can see
that some of the fast velocities are screened out,
and you get a difference of about 20 centimeters per second.
There is some instrument variability.
This is perhaps less than has been touted,
but suffice it to say that if you do serial exams on patients,
you should try to do it as much
as possible on the same machine.
If you use different machines, you can get some variability
that's based simply on the variability between instruments.
Then there are physiological differences between patients
that can result in different measurements.
Patients who are tachycardic,
or have a fast heart rate will tend
to have a lower peak systolic velocity
and a higher end diastolic velocity
because of the lower filling
of the cardiac chambers
and the shortened time for the diastolic flow to decline.
Patients with an exceptionally low heart rate
or bradycardia will tend
to have lower end diastolic velocities
because the decline within the waveform is greater
and the peak systolic velocities tend to be greater.
Hypertension tends
to result in higher peak systolic velocities.
And again, if there are the presence
of cardiomyopathies such that
there is decreased cardiac output in general,
that results in a decline.
In peak measured peak systolic velocities,
this sort of decreased cardiac output is seen
with cardiomyopathies and hypotension.
Here's an example of this.
This is a decreased peak systolic velocity
measured in this case, in the range
of about 40 centimeters per second.
In this patient with a cardiomyopathy, you can see that those effects are seen throughout the carotid
system, both in the right system
and the left, as well as the common carotid
and the internal carotid.
So the central abnormalities such as cardiac abnormalities
or cardiomyopathies are seen globally
through the carotid system.
Cardiac abnormalities most often
directly affect the absolute measure
of peak systolic velocity.
Using the peak systolic velocity ratio will usually
compensate and give you some handle on degree stenosis.
So if you have a patient with a low peak systolic velocity,
but with obvious plaque, I would suggest
that you rely more specifically on the
ICA-CCA peak systolic velocity ratio
and correlate of course, with the appearance on color
and gray scale.
Arrhythmias cause all sorts of problems for
velocity precision and measurement.
It is important that if you can identify
a strip of several identical waveforms
that you do your measurement there, either that
or choose the highest peak systolic
velocity seen on the strip.
Some people's arrhythmias are so irregular
that you won't be able to find a pattern of three
or more consistent waveforms.
And in that case, your last resort is
to use the highest measured peak systolic velocity
that you obtain.
Furthermore,
because the doppler, the vessel is changing direction very rapidly,
it's difficult to identify a correct doppler angle.
In general, peak systolic velocity tends
to be over measured in tortuous vessels.
And so I strongly suggest that you correlate
with gray scale imaging and look for focal narrowing
because a true stenosis should cause post
stenotic turbulence.
Here is an example of a true stenosis
here, both, I should say
that long segment stenosis can be another source of velocity variation.
As you might imagine,
a long segment stenosis in this plaque will be artifactually low
because of the flow limiting quality
of a long segment stenosis.
Tandem lesions can frequently cause aberrant waveforms,
and by tandem lesions, I mean lesions
that are seen both proximal and distally.
And measurement in the intervening segment will give you
very bizarre waveforms as the flow pattern will swirl
and eddy in that isolated segment.
In general, the peak systolic velocity is less than
expected in the segment, margined on either side by flow
limiting lesions, tandem lesions.
Then, some cases if you have a high grade stenosis on
one side, that will cause velocity elevation
on the other side.
In some studies, the use of ratios
that is the ICA-CCA ratio does not compensate and is,
and the effects from one side to the
to the other can be highly variable.
So in those cases, I recommend you use gray scale measurement
and color doppler to give you a better sense
of the degree of stenosis on the contralateral side,
the side away from the high grade stenosis.

Conclusion

So this is an introduction to basic carotid sonography.
I talked a little bit about how the study is performed,
how to assess plaque
and gray scale, spectral doppler analysis,
color doppler analysis, and then some pearls
and pitfalls, and a separate lecture.
I will talk about waveform analysis. Thank you.