Ultrasound Evaluation of the Carotid Arteries - SD

Introduction

Hello, my name is Leslie Scout
and I'm a professor of radiology at Yale University School
of Medicine, where I am chief
of ultrasound at Yale New Haven Hospital.
And I'm excited to be talking
to you today about ultrasound evaluation
of the carotid arteries.
It's a subject that's very dear to my heart
and something that I'm very excited about
because it involves not only imaging
but also pathophysiology.
It's extremely interesting.
In addition, it really offers you the chance
to make a significant impact on a patient's health
and future because by diagnosing carotid artery stenosis,
you can help prevent stroke,
which is a significant worldwide problem.
I'm here to share my experiences
and the things I've learned over the years
and I hope that these will help you in your practice
and help improve patient care in your area.
Thank you.

Ultrasound Evaluation of the Carotid Arteries

This next segment will be a presentation
of ultrasound evaluation of the carotid arteries.
We're going to talk a little bit about the background
technique and then focus on characterization of plaque
grading of stenosis of the internal carotid artery
with peak systolic velocity criteria.
And then finally end up with evaluation
of the carotid artery waveform.

Background

In terms of background,
and to put this topic in perspective,
stroke is a major cause of morbidity
and mortality in the United States.
And it's estimated
that there are over half a million new cases
of stroke per year in the United States in this decade.
And it's estimated that approximately 20 to 30%
of all strokes are due
to disease at the carotid bifurcation.
Now, stroke is more commonly caused
by emboli rather than ischemia occurring distal
to a pressure or flow reducing stenosis.
And the question is therefore, why do we grade a carotid stenosis if it's an embolic phenomenon rather than being directly related to reduced pressure gradient distal to a stenosis?
And the reason is that when you have a carotid stenosis, turbulent blood flow develops, characterized by high velocity jets, and these probably contribute to plaque friability.
Hence, the risk of embolization is increased
and that risk is directly related to the degree of stenosis.
In other words, the higher the stenosis,
the more likely there is to be a high velocity jet
and the higher the risk of embolization.
In addition, the more plaque there is at a particular location, the more likely it is that it will be irregular or hemorrhagic.
And that also increases its friability
and its likelihood of embolization.
To put it in clinical context,
many studies have demonstrated
that there is a clear benefit in patients
who undergo carotid endarterectomy
for hemodynamically significant stenosis in comparison
to patients who are treated
with optimal medical therapy.
And the three studies
that are quoted the most commonly are the North American
symptomatic carotid Endarterectomy trial
and the asymptomatic carotid atherosclerosis study as well
as the European carotid surgery trial.
And these three studies were all multicenter prospective
studies and they clearly demonstrated
that patients had a significant decrease in stroke
and morbidity if they underwent endarterectomy for
stenosis in the 60 to 70% range.
One of the things though, I do wanna point out
that when carotid endarterectomy was compared
to optimal medical therapy,
this optimal medical therapy at the time
that these studies were done in the 1990s really only
included anti-platelet therapy as well as diet modifications and exercise modifications.
And there are new trials currently underway at many places which are comparing carotid endarterectomy with current optimal medical therapy, which includes oral statins to lower cholesterol levels.
So what we all need to pay attention to when these trials are published to determine if this decrease in morbidity remains constant in the patient population that is treated with oral statin medication.

Indications for Carotid Ultrasound Examination

With that as background, what are the indications
for a carotid ultrasound examination?
Most of the time we do this to evaluate symptomatic patients
with either neurologic symptoms such as TIA or stroke
or patients who have an audible cervical brewery.
In addition, we're often asked to screen patients prior
to undergoing cardiac surgery
and increasingly to screen high risk patients
because of many comorbid factors such as obesity,
hypertension, and known elevated cholesterol level,
we frequently do carotid ultrasound examination
to follow up a known stenosis
or to follow up a patient
who has undergone carotid endarterectomy
or carotid stent placement.

Technique of the Examination

What is the technique of the examination?
There are three basic components of this exam.
First of all, we want to characterize the plaque,
then we want to do the doppler part of the exam
to grade the stenosis in the internal carotid artery using
peak systolic velocity criteria.
And finally, we want to evaluate the waveform.
We perform these examinations with a patient's supine
with a head slightly hyperextended
and turned slightly to the contralateral side.
You can interrogate the carotid arteries using either an
anterior or a posterior approach,
but whatever approach you use, it is very important
that you use the highest frequency transducer possible
and if at all possible to use a linear array transducer
because both increased transducer frequency as well
as the linear array configuration increases the spatial
resolution of the examination.
Now, in a patient who has a short, thick neck
or deep vessels, we sometimes have
to use a lower resolution,
lower frequency curved array transducer in order
to visualize the vessels.
So as I said, the first thing that we do is
to evaluate the plaque, and we do this on both sagittal
and transverse images using both gray scale
and color doppler.
And we focus our examination on the proximal
and distal common carotid artery as well as the origin
and the bulb of the internal carotid artery.
But we look also at the mid internal carotid
and distal if we can see it, as well as the origin
of the external carotid artery.
And when we evaluate plaque, we are looking
to characterize three different things.
First of all, the echo texture
and you want to differentiate hypo coic
from echogenic plaque.
We then wanna look at the surface characteristics
of the plaque and we differentiate ulcers
or plaque within an irregular surface from
smooth surface plaque.
And finally, we want
to estimate the percent diameter reduction,
just using the gray scale and the color flow imaging.
And the reason that we look at all things is that several prospective studies have shown that when there is enough plaque to cause a stenosis, if the plaque is hypo coic, if it has it irregular surface or if there are ulcers, this type of plaque is associated with an increased risk of neurologic events and increased rate of plaque deposition.
So these are important things to recognize.

Characterization of Plaque

Here are some examples of different echo texture in plaque.
Now on the slide on your left,
you can see this almost completely anti coic
or black plaque at the origin
of the internal carotid artery.
And this would be described as hypo coic plaque.
Whereas on the slide on the right,
you can see this very bright echogenic plaque at the origin
of the left internal carotid artery.
This plaque in this patient is very irregular.
It's relatively heterogeneous
and it has this focal arrow area centrally,
which is hypo colic, which most likely
represents intra plaque hemorrhage.
And again, this is associated with a worse prognosis.
This is another patient
who has very echogenic plaque both on the anterior as well
as the posterior wall.
And you can see that that echogenic plaque along the
posterior wall casts a very, very dense distal dirty shadow
as indicated by the yellow arrows.
And you could imagine that if
that plaque were on the anterior wall
and you had that shadowing over the lumen of the vessel,
it would be very hard to evaluate to the lumen,
the residual lumen of the vessel.
Another couple of examples on the left, again,
hypo coic plaque.
And on the right you have some very irregular,
more echogenic plaque.
It is important to recognize
that you actually may only be able to see
that hypo coic plaque on color doppler imaging.
In other words, it may be so hypo coic
that you can't see it on the gray scale imaging
and it will appear as a signal void on the
color Doppler image.
And here is an example on the gray scale image.
You see a little bit of echogenic plaque,
maybe some little internal echoes here in the what looks
to be the lum of the vessel.
But when you put the color on, you can see
that it outlines the plaque
and in fact, there's a lot more plaque there, much of it,
very hypo coic
that you would not have suspected from the gray scale image.
Another example that I've shown you previously, this is
that patient with that irregular echogenic plaque at the
origin of the left internal carotid.
But it doesn't look like it's significantly decreasing the
arterial diameter.
However, when you put the color on, you can see
that in fact there's some color blooming that overrides
and hides the plaque
that I showed you on the gray scale image.
But if you look up above it
where you could really not see much on the gray scale image,
you can fact see there's a great deal of hypo coic plaque,
which is causing a string sign.
And there is at least a 95% stenosis
of the mid internal carotid artery,
which frankly you wouldn't have suspected on the
gray scale image alone.
Here is another example showing some plaque at the origin
of the right internal carotid artery.
Some post stenotic dilatation perhaps,
but again, no significant plaque on the gray scale image.
But on the color Doppler image, you can see
that indeed there's about an 80% stenosis there
and hypo coic plaque that you didn't appreciate
with a gray scale imaging alone.
The color doppler can also be very helpful
for outlining the surface of the plaque
and will help you differentiate smooth,
which is relatively benign plaque from the more irregular
undermined or fissure plaque that is associated
with a worse prognosis.
Here on this patient, you can actually see that the surface
of the plaque is echogenic and indeed it looks very smooth.
And this is confirmed with the color doppler.
And again, you can estimate
that there is about a 50% stenosis there.
And on this patient, however you can see
that the power doppler demonstrates a very irregular
surface to that plaque.
Probably with some ulcers.
And again, you can estimate easily the stenosis
of at least about 70 to 80%.
Ulcers are a very interesting thing to try
and diagnose on imaging, whether by ultrasound
or with an angiogram.
And even our angiographers know
that we're really not very good at differentiating an ulcer
from just a very irregular
but endothelial plaque service.
You must remember that an ulcerated plaque is actually
histologic diagnosis,
and it doesn't mean just a divot in the plaque
because that can be endothelial and not be an active ulcer.
But nonetheless, on ultrasound
and angiography we describe a divot in a plaque
that's deeper than two millimeters from the residual lumen
as being suggestive of an ulcer.
If you actually see on an angiogram evidence of
that color button edema around it, that indeed suggests
that there is active ulceration there.
But again, we can't be absolutely specific.
Nonetheless, if we see a dividend,
the plaque deeper than two millimeters,
and here you can see one on the sagittal image confirmed on the transverse image,
we do raise the question of ulcerated plaque.
Most of the images I've shown you have been sagittal
or longitudinal images,
and they're really the best way
for looking at the percent stenosis as I've tried
to show you the surface of the plaque.
And it gives you a good view of the extent
of the plaque to look at the echo texture.
But don't forget to do the transverse imaging.
These are very good for evaluation of plaque burden
and we use them to confirm the findings on the sagittal
image and to make sure that we're not off center and
therefore overestimating the degree of the plaque.
However, we do not categorize the percent stenosis on a transverse image because it tends to overestimate the degree of stenosis.
In other words, as you see on this example,
you see a great deal of plaque and the vessel is expanded,
but the lumen is still more than three millimeters in diameter.
So although there's a lot of plaque there,
the percent stenosis on the gray scale longitudinal image
or color image will be more likely to be in the 50
to 70% range, not in the 80 to 90% range,
as you might suspect on the transverse image.
So we don't grade on the transverse image
because again, it tends
to overestimate the percent stenosis.

Pitfalls in Characterization of Plaque

There are a couple of pitfalls in characterization of plaque
that you should be wary of.
And the most common and most
therefore significant is color blooming.
Because color blooming may overwrite the plaque and
therefore obscure it
and underestimate the stenosis, particularly
if you freeze your image during systole.
And so here's an example
of plaque seen very well on the gray scale image,
and you can see it decreases the arterial diameter
by about 50%.
But on the color Doppler images,
it looks like there's no stenosis at all.
So remember when you evaluate plaque to use gray scale
and color as well as longitudinal
and confirm with the transverse images,
incorrect gain is also a problem.
And if the gain is too low,
it will make the plaque artificially look hypo coic
and you can overcall this vulnerable
or plaque that you think has a worse prognosis.
So make sure your gain settings are right,
remembering if they're too low, you're going
to think all the plaque looks hypo coic, and
therefore more worrisome.
And the last thing I wanna mention about
regarding plaque is fibro intimal hyperplasia.
This has been described in the recent literature
as a risk factor for cardiovascular disease.
And on this patient on the left, you can see
that there is diffuse echogenic fibro intimal thickening,
best seen always along the posterior wall
of the carotid artery.
And you always make the measurement along the posterior wall
and the common carotid artery.
And this patient on the right has more hypo coic
and thicker fibro intimal thickening.
And again, the thicker the carotid wall seemed is this seems to be related with increased risk for cardiovascular disease.

Pulse Doppler Interrogation

Having evaluated the plaque,
the next thing we do is move on
to pulse doppler interrogation in order to quantitate the degree of stenosis more accurately.
And you obtain a pulse doppler tracing in the proximal com
carotid artery and in the distal com carotid artery.
And it's very important that
that distal measurement be made two to three centimeters
below the bulb so that you don't make that measurement where the carotid artery widens.
As it joins the bulb and forms the internal
and external carotid artery,
we make at least three measurements in the internal carotid
artery, again at the origin, mid
and distal, one measurement at the origin
of the external carotid artery.
And then we measure a velocity
or get a tracing in the mid vertebral artery.
In addition to these basic tracings, we also sample and obtain a doppler tracing at any area that demonstrates focal color aliasing narrowing or any area where we see significant plaque deposition.
When you make the doppler tracing, it is absolutely critical
that the angle be kept at less than 60 degrees
and that angle ought to be kept constant if possible,
on follow-up examinations.
In other words, you don't wanna make measurements at 60
degrees in a patient in January of 2007
and then repeat it at 55 degrees in September of 2007.
You make at least three measurements
and record the highest peak systolic velocity
and diastolic velocity.
And then you calculate a peak systolic velocity ratio,
which is defined as the peak systolic velocity at the site
of the stenosis divided by the peak systolic velocity
in the distal common carotid artery.
A couple of important technical pointers.
It's critical that you carefully choose that doppler angle.
It ought to be between 30 and 60 degrees.
And as I mentioned, you should try
and use the same angle for serial studies.
If you make the measurements
with a doppler angle greater than 60 degrees, it tends
to artificially increase the peak systolic velocity.
And this is an example demonstrating that
this is the same patient on the left.
The angle is at 60 degrees, which is correct,
the velocity is 143,
and here when that angle is incorrect
and at 68 degrees you can see in the same patient,
the velocity is increased to nearly 200.
So if your angle is two high, you're going
to get an artificially elevated peak systolic velocity and
therefore you will overestimate the degree of stenosis.
Another thing to be careful about is
how you actually measure and calculate that doppler angle.
And there has been some controversy in the literature should
the measurement be made parallel to the wall, as you see
with this red line here
or along these yellow lines parallel to the jet of blood.
And quite clearly the consensus of opinion now is
that when you calculate that doppler angle,
you should be calculating it parallel to the jet
of blood within the lumen rather than to the wall
of the carotid artery.
Just like gain can be a problem in evaluating plaque,
if the gain is incorrect,
it can affect the doppler peak systolic velocity as well.
And this is an example lent from me from Dr. Tomi by Dr.
Christopher Merrit from Thomas Jefferson University,
where again, he made a continuous tracing
of peak systolic velocity in a volunteer's
internal carotid artery.
And all he did here was at the beginning he had the gain up
too high, so you can see background speckles,
and then he gradually turned the gain down till it was too
low and you didn't see a very sharp white tracing.
And you can see that there was a fairly significant range in
velocity from 70 with a gain up too high
to 55 centimeters per second when the gain was too low.
So again, at the very least, don't change your doppler gain
during the examination.

Normal Values and Waveforms

What are normal values that you're looking for?
These will be variable patient to patient,
but the normal peak systolic velocity is most typically in
the United States between 70
and a hundred centimeters per second.
It decreases as you travel distally down the common carotid
and internal carotid artery.
But the absolute number clearly depends upon pulse rate,
cardiac output, blood pressure, age,
and also body habitus.
You ought to characterize the waveform as well.
And the normal waveform of the internal carotid artery is,
as you see here, and is characterized
by a sharp systolic upstroke, a thin spectral envelope flow
above the baseline
and with continuous forward diastolic flow.
Now in the external carotid artery,
the waveform looks quite different.
There is substantially decreased diastolic flow little
sometimes no end diastolic flow.
And often you see this early diastolic notch.
The common carotid artery, which feeds both
of these vessels, has an intermediate waveform
and an intermediate amount of diastolic flow.
When you're evaluating the vessels above the bifurcation,
it's absolutely critical to know whether
or not you're looking at the external carotid
or the internal carotid artery when you measure the
peak systolic velocity.
And the absolute best way to be sure
that you're looking at the external carotid artery is
to identify branches in that vessel.
There are no branches in the neck in the
external carotid artery.
Another thing you can do is temporal tapping
and you can tap over the ophthalmic artery just lateral to the eye.
And when you do that, you will see a deflection in the
spectral tracing best seen during diastole.
Unfortunately, you can sometimes see
that in the internal carotid artery as well,
though it's usually not as sharply defined as you see here.
And if you compare one vessel to the other, it won't be quite as high velocity in the internal carotid artery in comparison to the external, as I mentioned, the external carotid artery will have a higher resistance waveform with less diastolic flow.
You'll have an early diastolic notch.
It tends to be more medial and anterior and also smaller.
To reiterate regarding the internal crowded artery,
there will be no branches
that you can visualize in the neck.
It is larger, has a lower resistance
with continuous diastolic flow
and tends to be more lateral and posterior.

Grading Carotid Stenosis

With that as a background as to
how you do the examination, let's now focus on
how we actually grade the carotid stenosis.
And this is based on a simple principle which really forms
the basis of all vascular ultrasound,
which is the simple equation, namely that flow is equal
to velocity times the area.
And we extrapolate that to the evaluation
of the carotid arteries by saying blood flow is equal
to peak systolic velocity times the diameter.
And basically what that means is that in order
to maintain flow volume, if the diameter
of the vessel decreases, the velocity must increase.
And this is a beautiful graft of one of the most important experimental studies done in vascular ultrasound.
And it demonstrates just what I showed you along the baseline is decreasing the diameter.
This is a velocity along the vertical axis.
And the yellow line is what happens to that velocity
as you decrease the diameter.
And the pink line is what happens to flow volume.
And you see that as the diameter decreases it,
the velocity stays fairly constant till you reach a stenosis
of about 50%.
And at that point, the velocity tends
to increase exponentially,
though it maintains flow volume till you reach about a 70%
stenosis at that time flow volume does
decrease a little bit, but velocity continues
to increase exponentially
and this is the basis
for doing the carotid ultrasound examination.
So armed with this information,
numerous authors have done studies comparing peak systolic
velocity measurements in the carotid arteries
with angiography.
And this is an example of one such study trying
to identify the appropriate threshold to diagnose a greater than 70% carotid stenosis.
And in their hands, the peak systolic velocity
that was most accurate for identifying these hemodynamically significant stenosis was 210 centimeters per second, a peak systolic velocity ratio of three.
Now, this was another group did the exact same study comparing carotid ultrasound with angiography, and they found in their group of patients that the appropriate threshold value was much higher.
It was 325 centimeters per second with a ratio of four.
So how do you deal with the fact that you have many reputable scientific investigators who find such widely disparate threshold values for identifying a greater than 70% stenosis?
Dr. Ed grant from California did a meta-analysis looking at multiple studies, and he found the same phenomenon for peak systolic velocity, which I'm showing here as he did for peak systolic velocity ratio, which is namely that the accuracy stays fairly constant for a wide range of peak systolic velocities.
So you can see that of peak systolic velocity of 1 75 centimeters per second is really as accurate as a peak systolic velocity of 300 centimeters per second for identifying a greater than 70% stenosis.
What changes though is the sensitivity and specificity.
So at these lower numbers, you have a higher sensitivity,
but lower specificity.
And at the higher numbers, you have lower sensitivity,
but you have higher specificity.
And indeed he found the same thing
for the peak systolic velocity ratio.
So what can you conclude from this?
You can conclude that there is a range of absolute numbers
and ratios for any given percent stenosis,
and these will be both laboratory as well as machine
or equipment dependent.
You really cannot accurately differentiate carotid stenosis
at 10% increments.
In other words, I can't tell you accurately
that one patient has a 65% stenosis
and another patient has a 72% stenosis.
In general with ultrasound, all of these studies have shown
that we tend to overestimate the degree
or percentage of carotid stenosis,
and they have all have shown
that we are actually more accurate
for detecting high grade stenosis, which is a good thing
because these are the ones
that are clinically important.
We are much less accurate
for detecting lower grade stenosis, stenosis,
specifically less than about 50% diameter reduction.
But this really isn't a big thing
because these are not stenosis
that are clinically important.
In fact, you would not recommend operation for these patients.
So how do you make sense of this?
The way I like to think about it is the following.
The criteria that you should use depends upon how you plan
to use the Doppler ultrasound examination.
If you plan to use it as a screening test,
and you are going to get correlative imaging in these
patients with C-T-A-M-R-A or angiography
before you refer them for surgery.
Your criteria should emphasize high sensitivity
and you should use those lower thresholds
of peak systolic velocity and systolic velocity ratio.
However, if you're going to use ultrasound
as a diagnostic test, in other words, instead
of an angiogram, the criteria should emphasize high
specificity and you should use the higher thresholds.
And if you can't make up your mind,
this actually helps you in how
to interpret the specific numbers that you get.
In other words, if the number is very high,
the peak systolic velocity is say,
400 centimeters per second, then you know you're likely
to be very specific and you can use it as a diagnostic test
and refer that patient
to surgery without a correlative imaging.
On the other hand, if the number is low, if it's two 20
or two 30, maybe then this is a screening test.
And if there are other questionable features
of the exam, for example, very little plaque,
or if there are comorbid problems such
that the patient isn't a great surgical candidate
and you really only want to send the patient to surgery,
if you're really sure that there is a stenosis greater than
70% than these patients
with these lower velocities are the ones that you ought to triage first to CTA or MRA.
Another thing to remember about these doppler criteria is
that they're only validated for the internal carotid artery
and not for other vessels in the neck.
So in order to make some sense
and to reach a consensus about these wide variation in reported thresholds for numbers of peak systolic velocity and ratio, about five years ago, the Society of Radiologists and ultrasound convened a consensus conference to try and analyze this data and their recommendations, which most of us now use as at least a starting point for evaluating carotid stenosis are as follows.
They use a peak systolic velocity greater than two 30,
a ratio greater than four,
and an end diastolic velocity greater than a hundred
centimeters per second as suspicious for a greater than
or equal to 70% internal carotid artery stenosis.
If the peak systolic velocity is less than 125 centimeters
per second, if the ratio is less than two
and diastolic velocity less than 40,
then there's likely less than a 50%
internal carotid artery stenosis.

Examples of Stenosis Grading

Here are some examples as to
how you might use this in practice.
And these are relatively old examples
because in fact, nowadays these are really the only time
that I get angiographic correlation.
So in this patient, the velocity in the common carotid is
about 60 centimeters per second,
and it reaches about
two hundred and fifty, two hundred forty centimeters per
second in the proximal right internal carotid artery.
And when you look on the angiogram, in fact,
you see there's just about a 70% stenosis
as you would expect using those threshold values proposed by the SRU consensus conference.
Another example where you see
that the peak systolic velocity in the internal is nearly
350 centimeters per second,
much higher than that threshold value.
So you're anticipating
that this will be a higher stenosis may mean 90 to 95%,
which is indeed confirmed by the angiogram.
And in this patient, you can do your own angiogram
with the color of power doppler,
and it looks like there's about a 90% stenosis.
And indeed, peak systolic velocity is nearly 400
centimeters per second.
And another example, this one is interesting, you see
that echogenic shadowing plaque in the anterior wall,
you can't see the lumen well,
and this is when transverse imaging really helps you,
but the vessel is clearly dilated there.
There's a lot of eccentric plaque.
And the question is though, really is the vessel just big,
or is the diameter actually really decreased?
And when you do your doppler, you see that in the common,
the velocity is about 50 centimeters per second,
but it's 460 centimeters per second
in the origin, the internal.
So again, you know, this is a high grade stenosis,
and so this patient we would refer directly to surgery
and not do an MR or a CTA.

Discordances Between Gray Scale and Doppler Findings

That's what I'm going to talk about in terms
of the Doppler criteria.
But what I want to warn you
and what I want to use is an introduction to the next part
of the lecture, is that whatever doppler criteria you choose
to adapt in your laboratory, it is absolutely critical
that you always correlate the gray scale appearance
and the appearance of the waveform
with these doppler criteria
because there are exceptions to the rule
that I just outlined.
So what do you do if there's a discordance between
what you see on gray scale and the Doppler findings?
And this is really when it becomes,
I think, the most interesting.
What do you do if the peak systolic velocity is
elevated but there's no plaque?
So you don't see a stenosis?
You should consider the possibility
of a tortuous vessel
or a contralateral occlusion or stenosis.
We all know that the velocity increases around a curve.
And when the vessel is curving, it's very difficult
to assess the correct doppler angle
because the direction of blood flow changes rapidly.
And so this means that we very often overestimate the degree
of stenosis or the peak systolic velocity in these very
tortuous vessels.
And one of the things you can do to help you is a, you see
that there, the vessel is very tortuous.
You see on gray scale there's absolutely no plaque.
You confirm that on color
and on the pulse oler trace, you remember
that any true stenosis ought to cause
abnormalities in the waveform post stenotic turbulence fill into the spectral envelope.
And brewery, another thing
to think about if you have an increased peak systolic
velocity, but not really a lot of plaque, is
that there might be a problem on the other side of the neck
because a contralateral high grade stenosis
or inclusion will increase the velocity opposite to it.
Unfortunately, the amount
that it increases the velocity is really quite variable
and unpredictable.
And so therefore the use of the,
that velocity ratio doesn't in fact always compensate.
And here's an example.
So in the right internal carotid artery,
you can see it's occluded.
The flow stops here right at the bulb.
You have a peak systolic velocity
of 260 centimeters per second in
the internal carotid artery.
This is clearly above that threshold at the SRU proposed
of 230 centimeters per second.
But when you look at that ICA carefully on gray scale
and color, you can see that
although there's a little bit of plaque here, the lumen of
that vessel really is only minimally narrowed,
certainly less than 50%.
And really all that has happened is
that you have increased flow on the left in comparison because of the occlusion on the right.
And there really isn't a stenosis here at all.
So when you have a high grade occlusion
or high grade stenosis
or occlusion on one side, you really have trouble using those threshold velocity criteria on the other side.
What about the opposite?
If you have lots of plaque
but the velocity isn't as elevated as you would expect,
think about tandem lesions, long segment stenosis
or a very, very tight stenosis.
So here's an example.
You have a peak systolic velocity
of only about 200 centimeters per second.
By those SRU criteria, you would expect somewhere
between a 50 and 70% stenosis.
But on the angiogram you can see in fact
that these are very tight stenosis.
So why was our peak systolic velocity so much lower?
The reason was is there was a tight stenosis in the
common carotid artery proximal
to the internal carotid artery.
And so when you have tandem lesions,
the velocity will not be as elevated in the distal stenosis.
In addition, if there's a long segment stenosis,
the peak systolic velocity also won't be as elevated
as you would expect for the degree of stenosis,
although the diastolic velocity also usually remains high.
And this is likely due to increased inflow resistance.
And this is proportional in fact
to the length of the stenosis.
So here is an example
and on the gray scale images, I saw a lot of plaque.
The color doppler confirms in fact
that there's really a long segment stenosis
that's quite narrowed in at least two places.
But even though I sampled very carefully up
and down the length of that entire stenosis,
the highest velocity I could get was 160, which is clearly
below the SRU criteria.
But again, looking at this really you shouldn't have a problem on the color flow imaging and identifying the fact that there's probably a 90% stenosis there.
This is an older image of another long segment stenosis.
Just to point out that these aren't very common.
Typically carotid stenosis are only about a centimeter
or less in length.
And if you see a long segment stenosis,
you could should consider other etiologies such
as prior radiation therapy, carotid dissection arteritis,
including fibromuscular dysplasia.
But one of the most important things to remember is
that if you have a very tight stenosis greater than 95%
stenosis, especially if it's a long segment,
the peak systolic velocity actually will decrease.
And if you go back to that experiment
that I showed you previously,
and you put a line at say 110 centimeters per second
across that curve of peak systolic velocity, you see
that it bisects that curve in two places at about
50% stenosis.
In other words, a non hemodynamically significant stenosis,
you could see that with a hundred centimeters per second
peak systolic velocity,
but it also bisects the left end
of the curve at a very, very tight stenosis.
So what happens is velocity
increases exponentially from about 50%
to about 98%, 97%.
And at that point, it decreases rapidly
until it reaches zero when it's occluded.
And the only way that you're going to be able
to differentiate between the high grade stenosis on the left
hand side of the graft
and the not so severe stenosis on the right hand side
of the graft is by looking at the gray scale image
and the waveform.
So what are the clues to a tight stenosis?
Decreased diameter of the lumen on the gray scale
or the color image is absolutely key,
but also changes in the waveform.
You might see something called a knocking waveform
proximally or a TTUs parvus waveform distally.
The knocking wave form, which you see proximal
to an occlusion or a high grade stenosis, is characterized
by a very low peak systolic velocity.
Here you see this is only about 25 centimeters per second
and little reversed
or absolutely no diastolic flow,
which is I presented earlier, is not normal for the internal
or common carotid artery only in the external carotid
artery, the TARDIS parvis wave form is characterized
by a sort of old hill
or old mountain, a kind of appearance,
delayed systolic upstroke.
You don't see that nice sharp systolic upstroke decreased
peak systolic velocity in this rounded systolic peak.
And this occurs distal to a high grade stenosis.
And so you ought to look at these features
before you grade a stenosis
and just remember that you can't a hundred percent rely
just on the single number.
And so here is an example where somebody
who was inexperienced with this fell into this trap.
This was a portable exam done in a patient in the intensive
care unit who had an acute stroke.
And this was read by someone with an inexperienced
as being normal because they looked at the peak systolic
velocity and saw there was only 37 centimeters per second.
So they figured there couldn't be a stenosis.
What you need to do is to compare the left side
with a right side.
And when you do that, you can see that
although they didn't see any flow here, they just figured
'cause it was portable and it was a poor exam.
But if you look at the right, in fact,
you can see perfectly good flow in a widely patent perfectly normal sized internal carotid artery.
So if they could have seen it on the right,
they should have been able to see it on the left.
And so in fact, there's no flow here on the left.
In addition, although that flow is normal, it's nearly half
the velocity of the flow on the right
and the waveform looks totally different.
There's rounding of this peak,
the systolic upstroke is not right is not sharp,
and there's much decreased diastolic flow.
So there's asymmetry in the waveform as well
as asymmetry in the gray scale and color flow image.
And in fact, this patient had a very
tight stenosis proximally.
They had an MRA, which shows delayed filling
of the intercerebral carotid circulation on the left.
And unfortunately the next day the patient had a large
stroke and completely occluded the left
internal carotid artery.
And the angiogram showed that the only filling
to the left side from the brain came
through collateral flow through the circle
of Willis on this MRA.
And in fact, there's no flow in the distal
internal carotid artery.
So it is important to recognize that changes in the shape
of the arterial waveform do provide
clues to other pathology.
And that's really what's fun about the carotid ultrasound
examination because if you look at these little nuances,
it gives you hints to proximal
or distal cardiovascular disease
that you're not even looking for or looking at.
And it can also help you identify some uncommon iatrogenic
conditions and you look, can look
for changes in systole diastole
and changes throughout the cardiac cycle.

Waveform Analysis for Various Conditions

What happens if you see increased peak systolic
velocity if it's bilateral?
Think of aortic regurgitation, hypertension, thyroid,
toxicosis or bradycardia.
If it's unilateral and there's no no significant plaque
or stenosis, think about the tortuous vessel
or contralateral stenosis that I told you about already.
What about decreased peak systolic velocity?
If it's bilateral
and the waveform is normal, think about things
that decrease the cardiac output.
So here is an example,
very low peak systolic velocity in the common on the right,
in the common on the left.
And in the internal, on the right,
in the internal on the left, the velocity is only 20
to 30 centimeters per second, and that is really not normal.
And this is due to ischemic cardiomyopathy.
And the ejection fraction is only 20%.
And you can see that that left ventricle really doesn't hardly contract at all.
And I've mentioned the tardis parvis waveform again,
to repeat, you have delayed systolic upstroke,
decreased peak systolic velocity,
and the classic appearance of this rounded systolic peak as opposed to the nice sharp pointy systolic peak that I've shown you before.
And this occurs distal to a high grade stenosis.
It becomes more pronounced the more distal you
sample to the stenosis.
And what's fun about this is that the pattern
of distribution in the vessels in the neck can help localize
where the stenosis is
because you know it has to occur proximal to the vessel
with the tardis parvis waveform.
So this patient has bilateral tardis parvis waveforms.
It affects the right common, the right internal,
the right vertebral artery, as well
as the left internal carotid artery so that you know
that the stenotic lesion has to be proximal to the left
carotid, the right carotid and the right vertebral.
So it's in guess where the aortic valve
and this patient has a severe aortic stenosis.
This valve barely contracts at all.
The lumin is almost obliterated. How about this patient?
There's tardis parvis in the right common,
the right internal, you have a sharp upstroke on the left
and you have reverse flow in the right vertebral.
So this lesion has to be proximal to the right vert,
the right common, but it doesn't affect the left side.
So you can tell this is gonna be a stenosis at the origin
of the right innominate artery.
How about this patient? The right side has a sharp upstroke,
but the left side has tardis parvus.
It affects the left common and the internal,
but the right common and internal have a sharp upstroke.
So sure enough, this has
to be on the left, can't affect the right.
And so this is at the origin
of the left common carotid artery.
I mentioned the knocking wave form again just to refresh your memory.
Low peak systolic velocity, little to reversed
or no diastolic flow.
This occurs proximal to an occlusion or high grade stenosis.
And think about atherosclerosis, dissection vasospasm
or increased intracranial pressure.
Here is a patient where you have reverse flow in early
diastole, no end diastolic flow in the common.
As you go up the common, the peak systolic velocity
decreases, the reverse flow flow becomes more pronounced.
You know, there needs to be an occlusion above that.
And sure enough, there's a high grade stenosis
with a string sign in the right internal carotid artery.
And as I said, this will always worsen as you get closer
to the obstructing lesion.
So another patient, no diastolic flow in the common
lower velocity in the internal.
And sure enough, a nearly complete occlusion in the distal ICA.
Another example,
this looks completely normal on the left normal diastolic
waveform in the left column comment.
But you can see no diastolic flow in the right
and you can see that this gets worse
as you go towards the head.
Velocity decreases in the internal no diastolic flow,
in fact markedly reversed early diastolic flow.
But if you look on the color image,
that vessels widely open.
So that obstruction has to be higher than that.
And sure enough, if you look in the head, you can see
that there is this dense middle cerebral artery sign on the ct
and there has been acute thrombus
of the middle cerebral artery.
And you can see the stroke on the CT
as well as on the mr.
What about if you see that wave form bilaterally?
I hear you see it on the right and on the left,
and again, worse in both internals.
And this is due to increased intracranial pressure.
And you can see here there's been herniation across the
midline in this patient with a large stroke
cerebral vasospasm
or diffuse cerebral arteritis can cause the same thing.
What about this case?
And this will help you avoid a making a mistake.
I think if you pay attention to these waveforms.
Is this normal or abnormal?
This is the left side
and I will tell you that this is a normal waveform
and here's the right common in comparison
and there's no diastolic flow.
The sonographer brought me this tracing
of the internal carotid artery
and it looked basically normal with low flow
and a lot of diastolic flow.
It looks like an internal carotid artery tracing.
But the problem is
that you have this abnormal flow in front of it.
And so unless there's a stenosis between this point
and this point in the distal common carotid artery,
which I could see well,
and there wasn't, this just can't be right
because this abnormal waveform
with no diastolic flow means there has to be a stenosis up
above it in the internal carotid artery.
So in fact, this isn't the internal carotid
artery that's a branch.
They thought that was the external,
but in fact, that's not, that's a branch off of this vessel,
which in fact is the external carotid artery.
So this was an occluded right internal carotid artery.
What the text thought was the internal was the external,
which had an internalized waveform
because the right internal crowded artery was occluded.
So when we went back
and we followed the distal common, we followed it right up
to this obstructing plaque
and saw that there was no flow in the internal carotid artery.
So again, it can help you not mistake an internalized waveform of the external carotid artery.
Some very unusual waveforms that you see quite uncommonly.
One is this puls biser,
which is two prominent systolic peaks with an interpose,
mid systolic retraction indicated by the arrow.
Usually the second peak is higher.
This tends to occur with aortic valve disease
or hypertrophic obstructive cardiomyopathy.
And here's a patient where you see that biser pulse.
And on the CT you can see that calcified valve,
which was both stenotic as well as regurgitant.
In this particular patient,
I've only seen pulses alternates a couple of times.
This is characterized by first a high peak, then a low peak
with irregular cardiac rhythm.
This tends to be due to intrinsic myocardial disease,
metabolic disease,
and sometimes impairment of venous return.
What about a patient
who has reversed diastolic flow bilaterally
but doesn't have decreased peak systolic velocity?
So it's a high resistance waveform,
but a normal to elevated peak systolic velocity.
This is due to aortic regurgitation.
And you can see the regurgitant blood flow here.
This is called the water hammer pulse.
You tend to only see it if the aortic
regurgitation is severe.
The p systolic velocity is normal to increase,
but you see reverse diastolic flow in the
common carotid arteries.
It's always bilateral,
but the waveform improves as you go distally.
And here is another example
where you see this abnormal flow in the common carotid
arteries due to regurgitation and often distally.
You now see there's the a normal amount
of flow in the internal carotid artery,
but you see that biser waveform
as I described before.
This is a very abnormal waveform, but it's very regular.
This is due to the intraaortic balloon pump.
And inflation of the balloon causes that second peak
of forward flow during early diastole.
And you see a flow reversal at the end of diastole,
which corresponds to deflation of the balloon.
Occasionally we get called to the unit in a hurry
because there's a acute swelling in the neck when they've tried
to stick in a central line
because instead of hitting the vein,
they've hi the carotid artery
and they've caused a pseudo aneurysm.
And here you see the rent in the carotid artery
and this large pseudo aneurysm on the transverse image
filling with color.
And here's the angiogram.
Another example demonstrating the waveform in the neck,
which has the classic two and fro pattern.
But you only see that if the neck is actually thin,
as in a case I just showed you.
If the neck is very wide, the waveform can be even normal, but often quite bizarre.
This is a patient who developed a pseudo aneurysm following
a carotid endarterectomy.
And I've included it
because it shows that nice typical yin yang appearance of the color flow in the pseudo aneurysm itself, but it doesn't have the typical to and fro pattern in the neck.
And that's because that neck is so wide.
Ultrasound can be used to diagnose carotid dissection.
This is something you should think about in a young patient
who has a stroke or a patient who has neurologic symptoms with no obvious plaque.
The wave form can be quite variable, very bizarre,
decreased peak systolic velocity.
And it depends upon the length of patency
and the size of the true or false lumen.
And whether you're sampling the true
or the false lumen, usually you can recognize it on the gray
scale images 'cause you can see the flap.
Often it's not as thick as this.
More often it's nice and thin as you see here,
and you can see it both transversely as well as Sally.
And this patient shows that in fact again has a couple of abnormalities.
You saw that thin dissection.
You see velocities though are elevated
and there is reverse flow.
So they're not elevated probably
because the diameter is decreased
because of the dissection.
First of all, you can see color fills the whole thing,
but with, if it were elevated velocity
'cause of a stenosis, you'd have increased diastolic flow.
Instead you have this high resistance waveform pattern.
This looks exactly like that water hammer pulse
that I described to you before.
And it was present actually
that high resistance waveform pattern
and the high velocity in the right side,
which didn't have the dissection.
And so that should make you think of aortic regurgitation.
And sure enough, this pa was a patient with Marfan syndrome
who had an aortic dissection and aortic regurgitation
and the dissection extended into the left
common carotid artery.
Here's another example demonstrating a nice thin echogenic dissection flap.
The both the false and the true lumen were patent.
And just to show you how variable that wave form can be
when you see a long segment stenosis with increased velocity
and no significant echogenic plaque,
look carefully at the wall for an intramural hematoma.
And if you see that,
particularly if you see it on both the anterior
and a posterior wall, as you see here,
think about carotid dissection
because sometimes the false lumen is not patent
or you just have intramural hematoma
and it will dissect for a long way up the lumen
and can cause a long segment stenosis without
any obvious plaque.
If you see an internal carotid artery that is narrowed
but has a very beaded irregular appearance
and tends to involve the mid portion of the carotid artery
and distal more than it involves the proximal,
think about fibromuscular dysplasia,
particularly in your young female patient.
And note that because of the multiple tandem stenosis up higher,
this patient had a tardis parvis.
So waveform.

Vertebral Artery Evaluation

The last thing I wanna talk about quickly
is the vertebral artery.
This typically has a low resistance waveform
with a sharp systolic upstroke.
Looks a lot like the internal carotid artery.
If there's a stenosis,
you'll see an increase in peak systolic
velocity as you do here.
But remember, the vertebral arteries can be asymmetric in
size and the increase in peak systolic velocity
can occur just because one side is hypoplastic.
So look higher and see if there's tardis parvis waveform.
If you see that, that will confirm
that the increase in velocity is due to proximal stenosis.
When you have a subclavian steel,
you'll see reversed flow again, it will be blue
or below the baseline or a negative.
Again, we've just switched the baseline,
so the negative is on top instead of below.
And if you see that, always check the subclavian artery.
You'll see high velocity centrally here,
450 centimeters per second.
And if you look out in the lateral subclavian
or axillary artery, you'll see that tardis parvis waveform.
This is a very unusual way form it's characterized
by this mid systolic retraction.
You see this early, very skinny systolic narrow peak mid systolic retraction.
Then velocity goes up again.
It can be above the late baseline or below the baseline.
And it has been described by somebody, I must say,
with a good friend of mine
who has a lot more imagination than I do
as the bunny waveform.
So this is called the vertebral bunny waveform.
And this is a presti waveform.
And the spectrum of this waveform changes
with varying degrees of subclavian steel
and the deeper the mid systolic retraction, as you see here,
the titer that subclavian stenosis is going to be.
Now, if you do provocative maneuvers such as
to exercise the hand
or inflate a blood pressure cuff till the hand is numb
and then deflate it, you can actually accentuate
that mid systolic retraction.
So this is an example from Dr.
Cleaver's article at baseline, that systolic retractions
above the baseline, but
after provocative maneuvers, it deepens
and is below the baseline.

Conclusion

In conclusion, I hope I've been able to demonstrate
to you that doing accurate carotid ultrasound can help you refer patients for carotid endarterectomy.
And the referral of appropriate patients will result in a
significant decrease in the incidence of stroke.
Ultrasound evaluation
of the carotid arteries is highly accurate in diagnosing
these high grade clinically significant stenosis in the 70
to 99% range.
In addition, you can use the doppler parameters in your own
laboratory either to maximize sensitivity,
which you wanna do, if you want to use this examination
as a screening examination,
or to maximize specificity if you want
to use the Doppler ultrasound examination
as a diagnostic study.
So if you wanna maximize your sensitivity, you want
to use value lower numbers as your threshold values.
If you wanna maximize your specificity,
of course you wanna use higher thresholds,
but whatever velocity parameters you decide to use,
it's absolutely critical
that you correlate these velocity measurements
with gray scale or color images as well.
But whatever velocity parameters that you decide
to use in your laboratory, it is absolutely critical
that you correlate these velocity measurements
with a gray scale and color images as well as
with waveform analysis.
Remember that the waveform in the common
and internal carotid arties should be
symmetric the right to the left.
And if you see an abnormal pattern that's high resistance
and it's bilateral, consider aortic regurgitation.
If it's unilateral, consider distal occlusion
or high grade stenosis.
If you see that TARDIS parvis waveform with
that rounded systolic peak, consider a proximal stenosis
and the distribution of that
within the great vessels in the neck will help you localize
the cy of stenosis
because you know it has to be proximal
to all vessels that are affected.
If you see a global decrease in peak systolic velocity
in all the vessels in the neck, consider things
that decrease cardiac output, hypotension,
severe aortic stenosis, and left ventricular
or thoracic aneurysm.
And if you see a global decrease in peak systolic velocity,
think about things that decrease cardiac output, such
as hypotension, a decreased ejection fraction,
severe aortic stenosis, and left ventricular
or thoracic aneurysm.
I hope I've convinced you
that carotid ultrasound is exciting
and that it is useful clinically,
and I hope that the things
that I have learned over the years will help you
and your patients in your practice.
Thank you.