Basics of Abdominal Doppler - HD

Introduction

I am Myron Nyk.
I'm from the University of Wisconsin and Madison.
I've been there for about 30 years.
We're gonna talk about the basics of abdominal doppler.
We will review basic wave forms.
We'll talk a little bit about hepatic veins,
portal vein, hepatic artery,
and the other major vessels.
And we'll finish with a discussion of doppler artifacts
and how they could create trouble for you
in the abdominal scan.

Why Bother Doing Abdominal Doppler?

When we do an ultrasound exam,
we have anatomic information,
but with just the push of a button,
we can get a little bit smarter.
We can capture hemodynamic information, that is
how blood enters an organ, how the pressure waves interact.
And it just decreases my diagnostic uncertainty when
I'm dictating a report.
If I see an anatomically normal organ
and I see normal flow profiles
and the arteries in the veins, that reinforces normality.
But even in the presence of a normal appearing organ,
occasionally blood flow can be distorted.

What to Look for in Doppler

Is it velocities? Is it wave forms?
A little bit of both,
but if you're just looking at
velocities, you're gonna miss things.
Waveforms are very rich in information,
and if you only hang your diagnosis on a velocity,
you can make mistakes because there are plenty of artifacts
that introduce mistakes into velocity measurements.
John Jewel, who used to be the chairman of radiology at
Wisconsin, has a great saying,
and it says A radiologist with a ruler is dangerous.
Really I like to look at waveforms more than I like
to measure velocities.

Factors Affecting Hemodynamics and Velocities

Now, there are factors.
There are conditions which will alter hemodynamics
and change velocities.
The velocities indeed will help us
to identify these conditions like stenosis
or arterial venous fistulas.
Hypervascular lesions,
tumors which pull in blood flow will increase arterial
inflow velocities, compressive masses,
be it a tumor adjacent to a vein,
or even the pressure of your transducer,
it will pressurize tissues
and it may increase velocities.
Sometimes tumors release vasoactive amines.
These are substances that can either constrict arteries
or open up the capillary bed,
and again, change flow dynamics.
And finally, vasospasm.
With certain conditions,
the arteries can just spasm down, changing
flow into an organ.
But there are other conditions that affect velocities.
There really aren't a manifestation of the disease process.
You may be trying to diagnose cardiac output.
How well is this person's heart pumping?
If it's a weak heart, then the waveform is gonna be altered.
And that's really got nothing to do with the organ
that you're looking at.
It's just a function of the waveform that's arriving at
that organ is disordered.
Blood pressure changes.
Can affect waveform medications?
Is the patient on any pressors?
The state of hydration?
Is the person dehydrated or are they over hydrated?
That will have a great impact on venous wave forms.
Whether or not the patient has been fasting
is critically important.
In looking at the splanchnic vasculature,
the superior mesenteric artery, the IMA,
the velocities of portal vein will change.
And then if you have a patient who's older,
their vessels are not as compliant.
All of these factors will affect waveforms
and velocities in an end organ,
but really are not a manifestation of a disease process in
that organ itself.

Technical Factors Affecting Flow Measurements

On top of that, if that wasn't enough,
there are technical factors
that can affect flow measurements.
You have to angle correct.
If you forget, you may critically underestimate.
A velocity insonation at angles of greater than 60 degrees,
is guaranteed
to introduce error into velocity measurements.
The closer you get to 90 degrees,
the worse off you are.
Angle correction is difficult with tortuous vessels.
When you have arteries that twist
and turn, it is difficult to identify exactly how to put
that angle correction
and just a few degrees of angle correction
as you get towards that 60, 70, 80 degree angle,
they can really significantly alter velocity estimates
where you put your sample volume in a vessel,
especially a big vessel.
That too will affect velocity measurements
because of laminar flow.
I'll show you that in a minute. And then finally,
transducer pressure can change resistance to flow.
If you push hard with your transducer,
you will change the dynamics
of flow within the underlying tissues.

Examples of Doppler Measurements

So here's a portal vein tracing.
And if you look at the one on the left, if you look closely,
realize nobody angle corrected here,
and the velocity as measured
by putting your sample volume here at
the cursor at the highest point,
it measures 0.11 roughly meters per second.
By changing the angle,
by putting a little angle correction here, then all
of a sudden we've more than doubled the velocity
that we are measuring.
Nothing's changed here.
All we've done is added the angle correction
and that then allows us to truly measure the direction of
the velocity of flow within that vessel.
'cause the computer without velocity correction automatically assumes that the flow you're measuring is directly towards
or away from the transducer.
Transducer orientation relative to the course
of the vessel is important.
Again, the closer to parallel with the course
of the vessel, you can come,
the more accurate you're going to be.
So if you're going to be measuring the left renal artery
velocity and you come from the anterior abdominal wall,
at this point, your velocities are gonna be terrible
because you are imaging perpendicular to the vessel.
You really don't wanna be coming from there.
You wanna be coming from the back.
You wanna come down the barrel of the vessel.
Now, if you have a skinny patient,
theoretically you can come from in front,
but you wanna come from the contralateral side
and aim down the barrel of the vessel.

Effects of Transducer Pressure

Alright, this is a renal transplant
and I am imaging this person with gentle pressure
with the transducer, basically no pressure.
Here's the arterial wave form.
I'm on a inter lobar artery,
and the resistive index I'm getting is 0.65.
So that's normal.
That's what we expect.
So what I then did is I took that transducer
and I pushed firmly.
I wasn't causing this patient discomfort,
but hard enough to basically blanch these tissues.
And if you look at the arterial waveform,
look at the difference.
And the only change here is the fact that I applied pressure
with the transducer.
I've basically squeezed diastolic flow way down to the point
where resistive index here at end diastole is 100%.
So just be careful if you're gonna be doing a carotid artery
and you're going to heel in
with the transducer to get a better angle.
Just don't pinch the underlying artery.
'cause it's like pinching a garden hose.
You're going to create a jet of high velocity flow.
You're going to change the dynamics within the
vessel.

Laminar Flow and Sample Volume Placement

In any tube in which you have flow,
the highest velocities are in the center.
The slower velocities are along the wall,
and it simply has to do with drag.
As the RBCs are bouncing along the wall, they are dragging,
they're moving slowly.
Here's A RMI phantom, which we've pulsed it,
and you can see the curve of the velocity
as it surges forward.
It forms a parabola.
And so again, the highest velocities in the center,
the slower velocities along the wall.
And in ct there's a similar phenomena you can see when you
add contrast.
So here, a patient had a contrast injection from
the superior vena cava.
The contrast came down
and kind of flowed
through the right atrium into the inferior vena cava.
And then as the flow surged back up,
it stuck the RBCs
with a contrast in 'em stuck along the wall.
And that's just 'cause they haven't been flushed yet.
That's a nice CT equivalent of
what laminar flow can do to you.
So, here's a b flow
of a jugular vein.
And you can see as this person's cardiac activity stops
and the flow surges forward during the A wave it pushes back,
but with the S and D wave, it'll surge forward.
And you can see that parabolic waveform.
There's another trick you can use,
and that's using a velocity tag on a lot of these systems.
They're pretty robust,
and you can do this little trick of
tagging a specific velocity of different color.
So here I use the green tag
and I put it at the velocities that are in the lower range
that are going towards the transducer.
And so on this image of the portal vein,
you can see the lower velocities are clinging
to the wall, and then a little lower veloc
in the branch vessel.
And so if you change it
and you put the green tag on the
higher velocities, guess what?
It's in the center. It's picking off the
higher velocity lamina in the center of the vessel.
I got really fancy with this image.
So, this is flow going away from, let's see,
yeah, it's coming towards the transducer.
And so the deep blue hues, those are the
that's the slowest flow
and you can see right up against the wall.
Then we come into a little lighter hue,
then we've got the green tag here, then the light blue,
a little more central, and now we're even aliasing
to the pink color.
And then deeper hues.
So you can see how many lamina this system
is projecting for you.
This system, you know, this is not a new scanner.
This is a pretty vintage scanner,
and yet the technology is very robust.
It can image every single pixel
and accurately detect the frequency shift and
lay it out there for you.
So when you have a vessel like this
and you can see these nice lamina of flow, there's no way
that there's any turbulence within this vessel.
That's nice to see that.
Alright, we did the same thing
with the portal vein here.
We put the green tag in kind of the mid range, then
an alias to the center.
And note you got a nice uniform flow here
with the sample volume in the middle of the vessel.
So here with the sample volume in the center
of the vessel here with the sample volume to the edge,
you have two different appearing wave forms.
This one is nice and clean
because it's picking up only the high velocities in the center of the vessel.
Whereas here I'm picking up the lower velocities
that are bumping along the wall.
In effect, if this was an artery, then you would
think consider that there was filling in
of the systolic window.
So anytime you put your sample volume in a vessel, try
to get it into the center, into the middle of the vessel.
To avoid this here, I just got rid
of the green tag on these images, so you can see better
where that sample volume is and how it affects the waveform.
Okay, center volume sample volume,
small versus sample volume wide.
Look at the difference in the tracing, how more
of the slow velocity pixels are picked up.
Because we're sampling across the spectrum of lamina across the vessels, you wouldn't want
to have your sample volume open this wide when you're trying
to characterize the waveform within any one vessel.
This is a useful tool.
When you're searching for a tiny vessel,
let's say you've got a hepatic artery
and a pediatric liver transplant, then you're trying to find
and you're having difficulty,
that's when you wanna open up your sample volume
and search, sweep a wide area of the image to try
and find that artery.
But if you're gonna try to characterize the flow within it,
sample volume small.

Arterial Waveforms and Influences

Okay, so I mentioned this earlier.
Arterial waveform is influenced by cardiac function.
The aortic valve, the integrity of the arterial system,
that's more for the systolic side.
And then downstream resistance influences the
diastolic component of that waveform.
So here are two tracings.
This one is in the celiac artery,
and this one's in the SMA, same patient, same time.
Look at the difference in these waveform.
This is a relatively low resistance wave form.
This is a high resistance waveform.
There's very little diastolic flow. Why the difference?
It's a function of what these two blood vessels are
perfusing. The celiac artery is perfusing the spleen,
the liver, a little bit of the stomach.
These are solid organs.
They basically function all the time.
They're taking all the blood that they can get.
The gut, on the other hand,
it changes its perfusion demand based on
whether or not you've eaten.
So a fasting gut really doesn't need a lot of flow.
And so the body is smart.
It will clamp down the arterials within the gut
and allow the blood to shunt to other parts of the body where it will do more good.
If you gave this patient a meal.
This will change from a high resistance
to a low resistance wave form.
Okay, here's a brachial artery.
And this arm is at rest
and we have fairly high resistance, 91% resistance,
very little fluid at end diastole,
because this artery is primarily supplying the muscles
of the forearm and these muscles aren't doing anything.
So again, the blood shunts away from 'em.
So I gave this volunteer a towel.
I rolled it up and I put it in their arm and in their hand.
And I said, here, pump this, pump this up.
It's like when you're gonna get an IV started, you
tell the patient, pump up here, pump a fist,
make a fist, you know what you're doing,
you're making them work.
And so the muscles of that arm now have a higher demand for oxygen.
They need to get waste products out.
And so the arterial bed will open up and increase flow.
And so what happens to that flow? It becomes low resistance.
The velocities pick up
and there is much more diastolic flow.
So just in the few minutes that it took for this person
to pump up, do a little work with that arm,
we went from a 91% resistance to a 62% resistance.
These are conditions you have to keep in mind
as you're doing doppler.
Not only what artery, what vessel am I looking at,
but what is the condition of the underlying organ?
Is it working hard or is it at rest?

Hepatic Vessels

So lots of factors affect velocities,
but let's spend a little time now looking at wave forms.
We'll start with the hepatic veins.
There are three, the right middle and the left.
Well, that's what the books say,
but in reality, there's always accessory veins,
there's a extra right or an accessory left.
There are caudate veins.
These are unique, these are small direct branches
that go directly into the inferior vena cava.
But the main vessels converge on the IVC about a centimeter below the right atrium.
So they're very close to the right atrium
and with its associated periodicity.
So the margins
of the hepatic veins are relatively anechoic.
That's a nice way
to tell 'em apart compared to the portal vein.
And their course is relatively straight.
The portal vein is more echogenic
because it has part of Glisson's capsule that dives in along.
And there's usually a little fat in the portal triad
as it goes into the liver.
So again, the hepatic veins tend
to run a more straight course.
So here you go, three hepatic veins converging.
But like I said, most often you'll see a few accessory veins.
And then at about 5% of patients, you'll see this vein.
This is a accessory right hepatic vein.
It's typically down three
or four centimeters from the convergence of the main hepatic veins
and usually supplies the right lobe of the liver.
If you have a patient who's got a little bit
of steatosis, then you can see these little guys here,
these are caudate veins.
These veins punch directly into the inferior vena cava.
They don't drain via the main hepatic veins.
And this is why the liver behaves differently
with cirrhosis.
The main hepatic veins will get shrink wrapped
as the cirrhotic fibrotic process clamps down on them.
It resists their outflow. It chokes them.
But it's harder to do that with the tiny caudate veins.
And so the central part of the liver tends to
function a little better and it will actually hypertrophy.
And this is why. So here's the quadrate lobe.
And in a patient with severe end stage liver disease,
that part of the liver typically becomes enlarged.

Hepatic Vein Waveforms

Alright? So we're gonna talk about the waveforms.
Here is a classic hepatic vein waveform.
And to simplify it, it's two steps forward, one step back.
So what does all that mean?
Well, first of all, let me
just tell you about one thing.
I like to use the term periodicity.
When I talk about the velocity
variations in the hepatic vein.
I don't like pulsatility.
When you take somebody's pulse,
what do you put your finger on?
You put it on the artery.
So pulsatility for arteries, periodicity for velocity
variation in the veins.
And then respiratoryity, I reserve for changes in velocities due to
respiration differences.
Okay, so these little bumps on the hepatic
veins, these velocity variations have labels
to them it's the A C S V D.
Alright, what does all this mean? What's it all for?
Let's take a little time and
digest this hepatic vein tracing.
So here I've got an EKG tracing electrocardiogram.
Here's a tricuspid inflow tracing,
and here's our doppler tracing in the EKG,
the P wave occurs during it triggers right atrial contraction.
So it forces the right atrium
to push blood across the tricuspid valve.
Therefore, that valve will increase in its aperture.
It opens up a little bit
and it shoves that blood into the right ventricle
to give it a little kick
to push blood forward into the lungs.
But since there is no valve between the right atrium
and the inferior vena cava, then a small component
of blood will spit back, actually reverse direction
and go back down the inferior vena cava
and into the liver.
Okay? Same thing will happen above the heart.
It'll go backwards up into the great vessels
of the neck and arms.
Okay, so now the atrium is finishing its contraction
and the velocity starts to flow forward.
Again, it'll start filling the empty chamber, but
before it gets very far, we get
to the QRS complex on the EKG
that causes the right ventricle to contract.
Now that's a much more powerful muscle than the atrium.
And so it will slam the tricuspid valve closed
and it'll actually push back on that annulus
that's holding the valve.
And when it pushes back, it will push back against
that atrium
and against the flow, which is starting to surge in
and cause a little hiccup.
It slows it down just briefly. And that is the C wave.
And I remember it as closure
of the tricuspid valve.
Okay? So now we've got this empty right atrium
and flow just surges forward,
very briskly accelerates until you get
to a point where finally it's starting to saturate
and the velocities slow down.
So this change from accelerating to decelerating velocity of antegrade flow is known as the S wave.
And it's S because it occurs during
ventricular systole.
So the ventricle is contracted at that point.
The atrium is open and relaxed,
but its velocity starts to slow down to the point where it
actually may spit backwards a little bit if it overfills,
but in most people it'll just come
to a complete stagnant point
or just slow down significantly.
Alright, so now what happens?
Now we're at the point where the ventricle is gonna relax
and as it does so the tricuspid valve pops open,
and when it does, now there's this empty chamber in front
of this column of blood.
And so velocity start to increase again as they
surge forward into the empty right ventricle.
So this change from decelerating
to accelerating flow is known as the V wave,
and I remember it as valve opening.
Alright, so now you've got an emptying atrium
and a empty ventricle blood surges forward again
until again, it saturates and starts to slow down.
This is known as the D wave
because it's occurring during ventricular diastole.
Okay, and where are we again?
Now we've come back to the P wave
and we're starting all over.
So pretty complex waveform,
but easily understood if you understand the
electrophysiology and
the function of the tricuspid valve, that
that waveform is key to see.
Now I'll tell you, you're not gonna see the C wave very
often at all.
You've gotta be very close to the right heart to see it.
It dissipates very quickly
and you're only gonna see it in a very
small percentage of patients.
Most of the waveforms you're gonna see are like this.
So again, two steps forward, one step back,
you get a little further away from the heart
and that may not manifest as actual reversal, only
as a slowing of flow.
Get further away, it starts to get damped out.
Now how far from the heart you see this waveform is totally
a function of how well hydrated the patient is.
If the patient is dry,
then there's not much blood in the inferior vena cava.
And therefore if the vessel walls coapt, if they kiss, then
that wave stops.
It's not gonna go any farther.
It can't because there is no medium across which
to transmit that flow.
You need fluid. So if you go further away from the heart,
typically it's gonna be a flattened wave form.
But don't be surprised if you see
that wave form down in the femoral veins of a patient
who is vigorously hydrated.

Portal Vein and Hepatic Artery

Alright, so let's go to the other side of the liver.
The portal hepatis gets about a liter
and a half of blood per minute, about a quarter
of cardiac output.
Only one other organ system gets more,
and that's the kidneys that get about 30%.
Two thirds of this blood comes in the portal vein
and a third in the hepatic artery.
But the hepatic artery brings all that oxygen.
Normally you get a nice brisk upstroke in systole and it's
because the arterial system wide open,
that arterial surge in systole should just scream right into
that artery and
the velocity should accelerate very quickly at end.
Diastole velocity should be about 0.2 meters per second.
Systolic velocities may vary,
usually 0.6 meters per second at end.
Diastole about 0.2, this isn't absolute,
but it's just a rough approximation resistive index, usually
around 60 to 70%.
Okay, so that's how we calculate the resistive index.
It's the distance from peak
to the trough versus peak to the baseline.
And then again, the end diastolic velocities roughly 0.2
meters per second.
The portal vein should have a relatively flat flow
profile.
A little bit of velocity variation, a little bit
of periodicity is okay, but not too much.
Okay, why? Well,
because the portal venous system is isolated from cardiac
activity by the capillary plexus of the gut
and the spleen on one side, on the inflow side,
and the liver sinusoids on the other side, the outflow side.
So this vascular system is isolated.
Now a little bit of periodicity is okay,
and this is usually introduced by
pressure variations within the liver
as the hepatic arterial systolic pressure surges into the
liver or maybe the A wave
from the hepatic veins.
So they will influence the portal vein flow velocities and
cause subtle undulations.
The velocity itself in the portal vein should be roughly
around 0.2 meters per second in a fasting patient.
So relatively close to diastolic flow.
And this relationship of portal vein velocity
to hepatic artery velocity is known as the liver
vascular index.
Okay? It's easy to get
because these vessels are right next to each other.
So you can do the wide sample volume and capture both in the same tracing.
Or you can put your sample volume in one vessel
and then move it to the other one in the same tracing portal.
Venous flow, it basically percolates through the liver.
The pressure gradient between the portal vein
and the right atrium is not very much.
So normally flow pressures in the portal venous system are about zero
to plus six millimeters of mercury.
Whereas in the right atrium minus two
to plus five millimeters of mercury,
that's just a few millimeters of mercury.
So really the portal venous blood just gently percolates
through this liver as the liver processes it.
Okay, so here's that tracing with the side by side
artery and vein.
And you can see end diastolic very close
to the portal vein velocity.
This is a normal liver vascular index.
Now what about this tracing hepatic arterial velocities are
up portal venous velocities are down,
and now there's a discrepancy between velocities
and end diastole
and the portal vein, this is an altered
liver vascular index.
What does this mean? It can mean anything.
Now, early literature back in the eighties started talking
about this being particularly good at diagnosing
hepatocellular carcinoma.
Unfortunately, that's not true.
This effect occurs with all sorts of liver disease,
be it hepatitis from any condition, any cause,
tumor in the liver, be it lymphoma or metastatic disease.
Anything can do this is really not very specific.
And the other thing I don't like
to talk about is it's not compensatory the hepatic artery.
It's bringing oxygen to the liver.
It's fueling the engine of the liver.
The liver's like a big factory.
It's doing a lot of work for the body.
The portal vein is bringing blood into the liver
to be processed.
But if the liver is dysfunctional
because of any of these conditions, then a portal vein
flow simply cannot enter the liver.
Whereas the hepatic artery,
however, it keeps pushing flow into the liver
because it is fueling the disease process.
So this concept is useful.
You know, you may be doing an ultrasound if somebody has
elevated liver function testing,
everything looks pretty good,
but you see this kind of a liver
and this what this is the starry sky liver.
You see the fatty portal triad standing out brightly against
the background of the liver parenchyma.
So does this mean that this is a hepatitis,
that this is an inflamed liver?
You know, that's what it might be.
But you know, you and I, we see these occasionally and
there's no liver enzyme elevation.
There's no, you know, we might be doing it
for a completely unrelated reason.
And so what do we do with this?
Do we talk about it or do we ignore it?
I would encourage you in a patient like this to get
that liver vascular index
because if it is altered, if the portal vein flow is down,
hepatic arterial flow is up, there's something going on
and it deserves to be mentioned
and you know, maybe this patient needs
to see a hepatologist, maybe they need a better history,
maybe we need to find out what's going on.
As the liver disease worsens the degree of flow slowing
or reversal increases except in the presence
of a paraumbilical vein.
We'll get back to that later.
So the liver disease will correlate with this.

Other Major Vessels

Alright, let's look at some other vessels, the aorta.
Okay, ultrasound indications.
An elderly smoker, we want
to rule out aneurysm somebody with pain.
We might wanna rule out
dissection somebody with a lot of pain.
We wanna rule out rupture or leak.
Typically that gets done with ct.
So how do we image the aorta?
Well, whatever window works, a lot of patients are gassy.
So we may want to come in from the side.
We wanna push the bowel gas out of the way.
If you're measuring diameters,
you wanna make sure you watch out for tortuosity
and maybe do a little angle correction with your probe when you're measuring diameters.
But in a nice skinny patient, we'll see the aorta up just below the diaphragm.
We'll see the origin of the celiac and the SMA.
We'll come down a little bit from there to the level
of the renal arteries.
We'll see their takeoff.
It's that nice banana peel look, come down a little lower.
You get to the mid to distal aorta.
Again, a little doppler to show you
where the branch vessels are.
And then finally you get to the iliac arteries
and you see flow within those.
So that's nice.
You do your transverse measurements, you measure
for aneurysm, you get down
to the iliacs, you measure those two.
Again, just be careful
because a tortuous vessel like this,
if you're holding your transducer perpendicular the torso
of the patient and you're trying
to measure this iliac artery,
you're gonna way over measure the diameter.
Now the doppler flow dynamics in the aorta,
very high resistance now where I'm talking about
below the level of the renal arteries.
So why is it such a high resistance?
Well, what is this blood vessel supplying?
It's primarily supplying the musculature
of the lower extremities and those muscles are at rest
'cause the patient is just laying
there in front of you.
So again, that plexus does not need a lot of flow,
therefore, the arterials will clamp down
and you end up with a high resistant system.
The celiac artery lives right here.
It's about a centimeter inferior to the diaphragm.
It gives rise to a splenic artery.
Hepatic artery, the left gastric.
And it typically will have a fairly
low resistance flow.
Here's the splenic celiac artery right there.
Here's the SMA, another centimeter lower to it.
It supplies the mesentery of the gut
and the right colon.
Sometimes you can see a hepatic artery
coming off the SMA.
That's a replaced hepatic artery that occurs in a small percentage of patients.
And again, two very different arterial waveforms.
The celiac artery will have a low resistance wave form
and the SMA will have a high resistance wave form kidneys, there's the main renal artery.
It divides into the segmental,
subsequently the interlobar between the lobes.
And then it the arcuate artery arcs over the top
of the medullary pyramids
and then the interlobular arteries.
These are tiny little guys in between the glomeruli
with power doppler.
We can see them, we can see these little interlobular arteries out there in the periphery.
The normal main renal artery flow should have about 1.6
to 1.4 meter per second velocity.
You start getting over that.
If you're getting towards two meters,
you might worry about stenosis.
The resistive index should be relatively low, 56 to 70%.
And the systolic acceleration time should be brisk
less than 0.07 seconds from the onset to the peak
of systole.
In the venous side,
we should have relatively continuous flow.
We may see a little respiratoryity
and you actually, very often you'll see the
cardiac periodicity.
That A S V D wave form. Now you can see that effect here.
A S V D, it's subtle, but it's there.
And the reason I'm seeing it,
because there's a full column of fluid
between this renal vein and the right atrium.
Why? Well, there's 30% of cardiac output coming
out of that renal vein.
It's likely going to maintain that cava
well distended with fluid.

Doppler Artifacts

Alright, let's talk a little bit about doppler artifacts
and how they can give us trouble in the belly.
We can divide 'em into color and spectral
and then we can divide 'em into technical and physical tech.
Physical being just a function
of the limitations of physics.
Technical being that we set the knobs incorrectly,
and I'll let you guess which one happens more often.

Velocity Range Errors

So let's talk about velocity range errors.
And typically with that we mean that the scale,
the pulse repetition frequency is either set too high
and you lose information or it's set too low and you alias.
So these settings are all about
how fast the flow is in the vessel
which you are investigating.
So here's the portal vein.
Very simply seen when I took the sample,
the pulse repetition frequency
and set it to 60, it's too high.
I don't see any flow in here on color. Why?
Because the flow is being suppressed.
That image is suppressed, the color's being suppressed
by the wall filter.
So it's see that little black area
between the two directions?
It blocks out the painting
of color in an area.
And so no flow is seen.
If we take the PRF
and change it to a lower scale
now we appropriately show the flow coming up into the liver
towards the transducer, take the scale even lower,
this is inappropriately low for the velocity in this vessel.
And now we're starting to see aliasing.
So we're picking off the high color going away from the transducer,
but this is simply oversaturating the system
and it's picking off the color going the other way.
Now you might say to yourself, I know
that I'm not gonna make this mistake,
but this is one
that I made a mistake on.
This is, you can see this is a nodular cirrhotic liver.
This person was ready to go for their liver transplant.
Now, this was many years ago.
You can see it's an older system,
but I'm imaging this portal vein
and I don't see flow within it.
Lots of hepatic arterial flow.
And so I would hesitate sending this patient
to surgery if the portal vein was thrombosed.
So they ended up sending this patient over
to the angiographers who injected the portal vein
and they showed it to be patent.
So what my mistake was, I had the scale set too high.
Here are two carotid artery tracings on this side.
What you see is aliasing. The scale is low.
This blue in the center of the vessel does not mean
that the blood flow is going in the opposite direction.
No, this is aliasing and I know it's aliasing
because when I go from the different
between the different colors,
I'm not going across this black wall filter.
I'm wrapping around from the high color saturation
of the higher velocities to the bright yellow.
So we're going from turquoise to yellow.
So that's when you know it's aliasing.
Take your pulse repetition frequency up
and we get rid of the aliasing,
but there's a little trade off here.
Take a look at the wall of these vessels.
Look at this one versus this one.
So when I have the pulse repetition frequency set low,
I get a better appearance of the interface
between the flowing blood and the vessel wall.
Whereas if I set it higher, I'm losing some
of these very low velocity pixels.
And this looks more ragged.
It's the same vessel, it looks more jagged
because the very low flow
that's bumping along the wall is being suppressed
by the wall filter.
And so the vessel wall itself looks more coarse.
Now the take home point is if you are doing a carotid
artery ultrasound and you are trying
to characterize the vessel integrity, you know,
is there atherosclerotic disease?
Let's live with the aliasing
and we will see a nicer interface
between the flow in the vessel and the vessel wall.
Leave it to remove the aliasing,
but you're going to add a more coarse appearance
to the vessel wall in spectral Doppler aliasing, same thing.
It's like you take some scissors, cut off the top
of the tracing and paste it from below,
change your baseline, change the scale and
you'll get rid of that artifact.

Gain-Related Artifacts

Alright? What happens when I take this gain
and I turn it too high?
What I get is spectral bandwidth broadening
mirror imaging in the tracing
or an indistinct velocity envelope.
So here's a tracing and it's a liver or kidney transplant.
And what I've done here is take the gain
and crank it up as I go across this tracing.
And so when you start with a gain set appropriately,
we have a nice systolic window.
We have a very clean velocity envelope,
and we have sort of a darker
tracing turn the gain up a little bit
and the tracing becomes a little brighter.
But look what's happening to the systolic window.
It starts filling in.
And then it gets a little bit more whiskery here.
It gets a little dirtier.
So when you image you want to see the tracing well,
but just be careful, don't overdo it.
'cause you can get into trouble. Alright, here,
this is a phantom, it's a block of gel,
and all I'm doing is I'm moving my transducer back
and forth across the block of gel.
So everything here is moving at the same velocity.
So here I hit one wall, I'm going to one side,
and then I'm gonna slow down.
I hit the other wall, start going in the other direction.
And so I'm creating this
artifactual tracing.
This is just me moving the transducer across the gel,
and I'm just doing it steadily gain, relatively low.
I'm turning the gain up a little bit.
The tracing is brighter, yes.
But look at the systolic window.
Well, let's call it the systolic window
for the purpose of this discussion.
It's filling in instead of being nice
and open, now it's starting to fill in,
turn the gain up a little higher,
filled in even more really bright tracing.
But what's this, what are you starting to see here?
Let's turn that gain up even more.
Systolic window's almost completely wiped out.
And look at this.
There's something that's going in the
other direction.
Wrong. There's nothing going in the other direction.
This is just a block of gel and I'm moving in this way.
What's going on here?
This is that mirror imaging in the tracing.
Turn it up even a little higher,
and you start seeing these spikes of noise.
What is this? This is crosstalk.
Crosstalk.
What does that mean? When you have these tracings projecting either towards
or away from the transducer,
these are like two different channels on your stereo system.
If you went and bought an expensive stereo, one
of the things the person selling it
to you we'll talk about is channel separation.
How good the stereo is.
It's separating channel A from channel B,
from the right, from the left.
And with good expensive systems,
they're very good at it.
They keep this from happening.
But in cheaper systems you'll get this crosstalk.
And so at higher volumes, the stereo doesn't sound so good.
While the same thing happens with doppler,
if you have the gain cranked up too high, the signal
talks across the channel
and will bleed over to the other side
and get projected artifact.
And then when you make it really powered up,
you start getting these crackles.
This is artifact. It is like on your stereo.
When it's really loud, it starts sounding very painful
to the ear because it's crackling.
It's actually the system is overwhelmed
and it's sending out noise instead of sound.
Now there's one scenario in Doppler where that artifact,
where that phenomenon is very useful,
and that is in identifying gas bubbles in the portal vein
when they are floating by.
So if you have a patient with pneumatosis intestinalis
and there's gas bubbles in the portal venous system, this is
what you see on the tracing.
So as a bubble goes by, it's not going
by at a higher velocity than the rest of the RBCs.
It's going along at the same velocity.
However, your system is set to correctly
paint the reflectors of the red blood cells.
But when a air bubble comes by, it's a much,
much more intense reflector.
The sound coming back is much louder and
therefore a noise spike is painted on that tracing.
And that's how you identify air bubbles.

Erroneously Displayed Flow Reversal

Alright?
So erroneously displayed flow reversal.
I showed you two cases already.
Aliasing or the gain being too high.
You can have just a poor system
with lousy balance in the circuitry,
or you can have it caused by grating lobes or side lobes.
All right? Most of us, when we think of sound coming out
of a transducer, we think of it sort
of like a laser, but it's not.
It's there are lobes of sound.
The sound is focused
and there are additional lobes that come off to the side,
the side lobe and the grating lobe.
Now, these are very weak,
but nevertheless these are still focused beams of sound
and they can interrogate things to the sides and make you
and maybe fool you.
All right? So the side lobe right along the main beam
and then the grating lobes,
these guys are way off to the side.
These grating lobes,
artifacts are the worse on those very tightly curved
intracavitary phased array probes.
All right? There's a block of gel
and I put a needle through the gel.
Like if you were doing a biopsy, when you guide a biopsy,
you ever sometimes you see this thing
that you think it's a needle,
but it really doesn't make sense where it is.
And then all of a sudden you pick up like, oh,
whoa, here's the needle.
Well, there's the needle. What's this?
Alright, this is an artifact, this is a grating lobe,
a focus of ultrasound that comes out way off to the side.
It interacts with the needle
and then comes back to the transducer.
Now the transducer, when the ultrasound system,
when it paints a reflector,
it does not take into account these
grating lobes or side lobes.
It always projects on the image according
to if it was coming back from the main beam.
So because this sound took longer
to travel to the target
and come back the ultrasound system will paint a dot further down from
where your actual interaction with the needle is.
So you have this artifactual needle appearing deeper
than the true needle is.
So, okay, that's it in gray scale.
Does it happen in doppler? And it does. Very subtle.
But here, this is a phantom
and flows only going in one direction.
So I should only be seeing it coming this way
towards a transducer.
But instead I got this over here. What's this?
This is a grating lobe that's coming across
and interrogating the flow in a way to the other side.
Now you notice it's very weak on this tracing.
I love this one.
The sample volume is here, it's right in the middle of gel.
There's nothing moving here. This is where the movement is.
And so I had to, it was a little tricky for me to do,
but I was able to put the grating lobe right into
that stream of flow.
And here you get this tracing.
So, does it happen clinically? It can, it's pretty rare.
This is a renal transplant that had a very high resistance.
And as I'm scanning it, all of a sudden here's a sample
volume right in the middle of the kidney.
And yet I'm getting this very low resistance waveform.
I have no clue where this is coming from.
The grating lobe is coming out
and interrogating some vessel somewhere.
I don't know where, but you know, if you see this
and it makes no sense, here's my advice,
don't take the picture
and then you don't have to explain to some surgeon
what a grating lobe artifact is.
Just don't take the picture.

Bidirectional Flow Artifact

Alright?
Bidirectional flow artifact.
It happens when you image perpendicular to the vessel.
So if you're just perpendicular to the vessel,
you may get flow showing up on both sides.
And I'll tell you, with these new ultrasound system, that's
hard to do, you really gotta work at it
because just a degree off from perpendicular.
And these systems will accurately
paint the direction of flow.

Mirror Image Artifacts

All right? If you do ultrasound along,
and if you may see mirror image artifacts.
So if you're near the diaphragm
and you have a cyst in the liver,
you can paint a cyst in the lung.
Okay, here's the TIPS. What's this thing?
That's the diaphragm. This is not a catheter in the lung.
This is simply mirror image artifact ultrasound hit here,
reflected in this direction, came back.
And so this projects deeper.
Are these accessory hepatic veins in the lung?
No, it's simply mirror image artifact off the diaphragm.
It's taking this and projecting it deeper.
Tracheal cartilage, same thing.
There's nothing back here.
The airway, that's where the air is.
So everything deeper to that line is mirror image.
And so you think there may be two sets
of tracheal cartilages?
No, that's not, that's artifact.
So if you have a blood vessel and the sound comes towards
and comes back, that's where it should be.
But if the sound hits the diaphragm goes to that structure,
interfaces with it,
what the computer thinks is it came from over here and
therefore it'll paint a structure there for you.
Okay? So what blood vessel is this?
This isn't a, it's just not a real blood vessel.
Here's the hepatic vein. This is the inferior vena cava.
The diaphragm is over here.
And what's happening is sound came hit that part
of the diaphragm, bounced off of it,
and then went right down the barrel
of the inferior vena cava got shifted and shifted red
because the flow's coming towards the transducer.
The sound bounced back to the diaphragm, back
to the transducer.
This is artifact.
Anytime you're imaging anywhere near the lung,
you can get into trouble.
Okay? Here's a interesting mirror image artifact.
We're going a biopsy
snap and out.
Did you see the artifact? Look down here.
Okay, there's the diaphragm.
And just pay attention over here.
When I was watching the resident do this biopsy,
I got scared for one second.
That's not the biopsy needle tip.
It was fortunately there in our lesion.
This is all just mirror imaging artifact.

Color on Nonvascular Structures

Color and nonvascular structures was written
up a long time ago.
And this is an important artifact.
It is a function
of the motion discriminators in the systems.
For every dot on the image, these ultrasound systems have
to decide whether they're going to paint color
or gray scale.
And the discriminators basically decide
for the intensity of reflection.
If it's bright and the frequency shift is small,
it's gonna assume it's an artifact.
It's gonna assume either motion
of the transducer in your hand
or maybe the patient's breathing.
But if it's a dark reflector, it will assume
that this motion is real, such
as a red blood cell moving in a vessel.
And so what do you think of this is a gallbladder.
Is there a cholecystoarterial fistula here?
You know, it better not be, it's my gallbladder.
The only thing I was doing here was taking the transducer
and moving it in and out, pushing against everything.
And so at a point in this image, everything was moving.
And look where the system painted the color.
It painted only where the low intensity reflectors are.
So if you're taking an image of a cyst
and you're ready to hit the freeze button
and the patient blows their breath out,
and just as you freeze it, the color paints into that cyst.
Don't take the picture. Again, you don't have
to explain this artifact.

Spectral Bandwidth Broadening

Spectral bandwidth broadening is an artifact.
It has to do with angle dependence
and again, a decreased angle.
So if this is your direction of flow
and you're coming in with your insonating beam in
that direction, that's horrible.
You're not gonna have accurate portrayal of the tracing.
You're gonna have bandwidth broadening
if you come in at that angle.
That's so good. Now we're getting better.
Now we're getting even better yet. And that is the best.
This is where you're gonna get your best
doppler velocity measurements.
The best tracings. So here's a phantom.
And the only thing I'm doing differently between this image
and that one is we are electronically steering the beam.
This is steeper, this is closer to 90 degrees.
This is a little more shallow. So this one is better.
And look at the differences in the tracing.
I didn't change anything other than that.
Look at the systolic window here.
It's much better seen at a shallower angle
to the actual direction of flow.
So imaging wise, it's better,
but also it's better for velocity measurements.
Okay, so here we go.
We've got a velocity measurement of one meter per second
and 0.85 meters per second.
The only thing different between the two
is electronic steering.
Now this is a phantom.
And in that phantom I set the velocity
at 0.8 meters per second.
Which one measured more accurately?
The one that is less perpendicular
to the direction of flow, the further towards perpendicular,
the more inaccurate your velocity
measurement is going to be.

Twinkle Artifact

All right, let's talk about twinkle artifact a little bit.
This is due to clock jitter from a irregular interface.
Typically stones
and crusted sutures, catheters, anything in the body that
has an irregular interface.
Now it's very useful to confirm calculi.
Okay, you saw that bright structure.
You turn on the color and you get noise
and it's due to just a irregular surface.
And so as the sound come back, it's distorted.
There's a pregnant female, here's her left or her right ureter.
See that little bright thing? What could that be?
Turn on the color
and boom, there is twinkle artifact.
It's very impressive. It just nails that calculus.
It's very useful. I was this was a neonate
with biliary atresia.
We were trying to find the gallbladder.
I wasn't sure where the gallbladder was
until I turned on my color
and saw the twinkling of the crystalline material
in within the gallbladder.
There it was, okay, renal calculi.
It's excellent at identifying renal
calculi, but just be careful.
This artifact, although it's it will
just make these stones shine at you.
It overestimates their size.
Do not measure stone size from here to here.
You'll be way overestimating the stone size.
Okay? You should measure 'em off the image.
So very often we'll see, you know, these are paraplegics,
they come to us 'cause they're immobile
and they mobilizing their calcium.
We rule out kidney stones
and you know, we see these bright things.
They're have twinkling on color.
The also the spectral tracing is abnormal.
Sometimes maybe it's tiny
and you think it's a blood vessel turn on
your spectral doppler.
It won't sound anything like a normal
arterial venous waveform.
It actually sounds like you're tuning a short wave radio.
Here are two examples
and you can see how these tracings have been altered.
And when you listen to them, really,
it sounds like you're tuning in a radio.
Alright, this woman had a bladder suspension procedure.
You had UTIs often.
And so the technologist is doing the scan
and she sees this thing in the middle of the bladder.
She turns on her color and there's flow here.
And she says, Myron,
there's some unusual blood vessel in the bladder.
And I looked at that and I go, oh man, that can't be
because this spectral tracing just makes no sense.
This is artifact. And what this was,
it was not a blood vessel,
it was a suture material from the bladder suspension.
The suture had poked a suture through the center
of the bladder and the strands became frayed.
They got a little crystalline deposition.
And so that's where that twinkle artifact was coming from.
When you're looking at a gallbladder
and you see this mass like cluster?
It could be tumor, but it could also be tumor effect
sludge.
Most of the time it's sludge.
When you turn on your color, just be careful.
This is not perfusion. This is twinkle artifact.
Make sure you throw your sample volume on there
to convince yourself this is just noise.
This is not an arterial or a venous wave form.
Okay, here's another patient's gallbladder image.
Lots of color in here.
This is not blood flow to a tumor, it's twinkle artifact.

Conclusion

So in conclusion, Doppler, use it, use it.
Don't wait for somebody to order it.
It you have to keep your skills up.
If somebody asks you to do doppler,
because it's a very complicated case,
it's not gonna be easy to do.
You want to learn
and refine your doppler skills on normal patients.
If you have a few minutes between patients, play with it.
Make sure you know it.
And then apply these skills on the abnormal patients.
Some disease conditions really can only be
diagnosed by doppler.
Don't hesitate to turn it on. It can make you smarter.
Here, here's a case. There's a very young college student was
came to us
because she had abdominal pain.
And we did this imaging study.
Now, this was a long time ago, but this was
before I wrote the first Doppler book.
And the tech came by
and showed me beautiful images of the liver,
the gallbladder, the pancreas.
Everything looked absolutely normal.
We were gonna call this a normal study,
but since I had a little time between the patients
and I needed to capture some tracings for the book,
I came in there and I started imaging her vasculature.
And I started taking some tracings, normal aorta,
normal portal vein, normal hepatic vein,
hepatic artery.
What do you think of that tracing? Is that a normal tracing?
It's actually not. This is a tardus parvus waveform.
Where's the systolic peak?
It's flat and it's elongated.
So from having a perfectly normal study, all
of a sudden we've got something going on here.
And then when we looked at the celiac artery,
almost over two and a half meter per second velocity.
What's going on here?
This young woman had an arcuate ligament syndrome
where the diaphragm was coming across the celiac artery
and pinching it
and causing decreased perfusion to the liver,
making it relatively ischemic and causing the pain.
Here it is on the CT angio. You can see this defect.
There's the celiac artery right here,
and you can see how it's compressed at that point.
So compression on the celiac artery arcuate
ligament syndrome, that's her diagnosis,
and we would've completely walked past it.
Had I not stumbled into it with the ultrasound here.
Here's another patient. Same thing.
You can see the jet and it's kind of fun.
You can see the difference between inspiration
and expiration is that diaphragm slides across the celiac,
it can make it worse.
So use your doppler.
You'll be surprised what you could see with it.
Thank you very much.
And if you liked what you saw in this lecture
and you wanna see some more material,
if you wanna learn more about Doppler,
we've just published our third edition
of the clinical Doppler ultrasound textbook.
You can find it on Amazon,
you can find it in bookstores, you can go to conferences.
The vendors are selling it. Hopefully you
might learn a little more about doppler.
Thank you.