Abdominal Doppler: Pearls & Pitfalls - SD

Introduction

Hello, my name's Bill Middleton.
I'm a professor of radiology at the Mallinckrodt Institute
of Radiology at Washington University in St.
Louis, Missouri, United States.
Today I'm gonna be speaking about pearls
and pitfalls of abdominal doppler.
We're gonna divide pearls
and pitfalls of abdominal doppler into two basic categories.
First we'll be specific anatomic areas
of the abdomen, and we'll talk about a little bit of anatomy
and physiology that can cause pitfalls.
And we'll go through this list of structures.
And then we will shift gears
and talk specifically about some technical parameters
and things that can cause pitfalls with various measurements and interpretations.
But we're gonna start with the portal vein.

Portal Vein

One
of the more common pitfalls in evaluating the portal vein is
slow portal venous flow.
It can cause false positive diagnoses
of portal vein thrombosis,
or at the least can cause an inconclusive
Doppler examination.
And the pearl, when dealing with that
is mesenteric augmentation.
It's based on the same principle
as we use in augmenting venous flow.
In the lower extremity,
you basically just push on the lower abdomen over the area
of bowel,
which augments flow in the mesenteric vein
and increases flow in the portal vein.
In addition to this, you can consider doing postprandial
scans in patients with slow portal venous flow
to augment portal venous flow
and make it easier to identify.
Now, here's an example of a patient
where at baseline we could see no flow within the main
portal vein, but when augmentation was performed in the
lower abdomen, enough flow
velocity increase occurred that it was possible
to see flow in that portal vein.
And this is something that can help you occasionally in
these borderline cases.

Another pitfall related to the portal vein is development
of periportal collaterals.
Now, as you know, these occur in the setting
of portal vein thrombosis,
usually chronic portal vein thrombosis,
although they can start to develop with in six
to 20 days of thrombo portal vein.
Now, generally they're seen as multiple tortuous vessels
in the portal in the port of Hetus
with detectable venous flow on Doppler analysis.
But occasionally if the portal vein is atretic,
and very difficult to visualize,
and there's a single large collateral,
that can simulate a patent portal vein.
Now the pearl in this case is
that collaterals almost always form anterior
to the portal vein and the hepatic artery.
And that location can help you when you think
that you might be dealing with
a collateral versus a patent portal vein.
And then also the collaterals are usually more tortuous,
although when there's a single straight collateral,
that's when you get into trouble.
But this diagram sort of shows the anatomy that you have
with an eretic thrombo portal vein
and a collateral anterior
to the portal vein in the hepatic artery.
And here's an example where the diagnosis was missed.
You can see in the portal heus that there's a hepatic artery and a portal vein.
This is hepatic artery,
and this is what looks like at least the patent portal vein.
Now, when you look at the waveform within this vessel,
you can see that it's a fairly monophasic venous waveform
that looks for all tens
and purposes like a patent vein portal vein.
But notice in this case that it's located anterior
to the hepatic artery.
That's your clue that this probably is not a normal portal vein,
and that what you're dealing with at least potentially,
is a single preport collateral.

Now, another artifact that occurs potentially in all veins,
but particularly as problematic in the portal vein,
is blooming artifact resolution in color doppler is lower
than it is with gray scale.
And because of this,
the doppler signal typically extends over adjacent soft
tissues and can obscure the vessel wall
and in any potential intraluminal abnormalities such
as non-obstructive thrombus.
So if you have a thrombo portal vein and you do
or partially thrombo portal vein
and you do doppler, the signal can completely obscure the
intraluminal thrombus.
So the pearl in this case is simply to be aware of this potential pitfall.
And when you are doing an examination
for suspected portal vein thrombosis, it's very critical
to use a combination of both color doppler
and gray scale to make your analysis.
So here's an example of a portal vein on color Doppler looks
entirely normal with no
intraluminal abnormalities.
But when you look at that same portal vein without the
color signal, you can see that there's fairly extensive non-obstructive thrombus.

Now, another potential pitfall in the portal vein is
the presence of helical blood flow.
Now, this is not all that uncommon.
In fact, a study that was done at Kansas
showed that 20% of patients with chronic liver disease,
these are pre liver transplant patients, had an area
of helical flow in the region
of the portal bi bifurcation.
Now, not only in chronic liver disease, you see it,
but you can also see it in normal patients,
albeit at a much lower percentage.
Now, it tends to be particularly common in patients that have had tip shunts
that are status post liver transplants
or in patients with portal vein stenosis.
The problem is that on doppler analysis,
this helical blood flow can sometimes
simulate flow reversal.
And the pearl in this case is don't just rely on gray scale
and pulse doppler.
If you use colored doppler, you'll be able to see this area
of flow reversal
and be able to confirm
that it isn't throughout the portal vein
by looking at the more peripheral veins in the liver.
So here's an example of a patient
with a fairly prominent main portal vein.
And on this Doppler waveform, it looks for all intents
and purposes, like the blood flow is going in
the wrong direction.
But when you look at color doppler, it's fairly easy to see
that this is really just an area of localized helical blood flow.
And while there is reverse flow in this area, the main main portal vein flow is going in an antegrade
direction, which is easy to determine
by looking at this more peripheral segment here.

Renal Artery

Now we're gonna move on to the renal artery.
And one pitfall with evaluating the renal arteries
is identifying the location of the abnormality in patients
that have parvis tus waveform changes.
So we know that with renal artery stenosis,
which usually occurs at the origin of the renal artery,
that you can have alteration in the distal waveform
that's referred to as a parvis tardis.
Now, most of the time when you see a parvis TARDIS waveform
in the renal artery, that's due to atherosclerotic stenosis at the origin,
but it can really be due to a stenosis anywhere proximal to
where you're measuring the vessel.
So in fibromuscular dysplasia,
it may occur in the mid renal artery,
but remember, it can also occur in the proximal aorta.
So it can occur secondary to valvular stenosis
or secondary to coarctation
or other forms of aortic stenosis.
So the pearl in this case is when you see parvis TARDIS waveforms recognize
that they can occur anywhere proximal,
and that if they do occur in the aorta, they're going
to affect more than just one renal artery.
So here's an example of a patient
that has a parvis TARDIS waveform in the right renal artery.
But notice in this case,
there's also a similar pars tardis wave form in
the left renal artery.
So potentially this patient can have bilateral renal artery
stenosis, but as you look further, you can see
that the superior mesenteric artery is abnormal.
The celiac origin is abnormal,
and this tells you that there's something proximal to all
of these vessels potentially aortic valvular stenosis.
But when you look at the vessels in the upper body,
the common carotid arteries are normal bilaterally,
and the brachial arteries are normal bilaterally.
So this stenosis has to be located someplace
between the aortic valve and the renal arteries.
And in this case, it was located at the at,
at it was a coarctation located just distal
to the left subclavian.

Now, another potential pitfall
with the renal arteries is confusing
flow in the left renal vein
for flow in the right renal artery.
If you look at the diagram here, you can see
that there's an area between the IVC
and the aorta where flow in the left renal vein is very
close to flow in the right renal artery.
And since they're both going in the same direction,
it can potentially cause confusion.
Now, the pearl in this case is to realize
that venous pulsations when they occur,
are different from arterial pulsations.
It's just a completely different morphology.
And the wave form morphology in the renal artery should
be fairly similar in all segments.
And it shouldn't differ dramatically between the origin
and the mid and distal portion of the renal artery
as would occur if you're confusing the left renal vein flow
for right renal artery origin flow.
So here's an example where we can see the left renal vein
emptying into the inferior vena cava.
And you can see the left renal vein goes anterior to the aorta.
This is the right renal artery heading away from the aorta and behind the IVC,
and you can see how close together those two vessels are
and how they're both going in the same direction.
Now, waveform analysis from those vessels will show
that the left renal vein waveform,
although it is pulsatile, has a completely different
pulsatility than the pulsatility in the right renal artery.
And that should be a clue that you're dealing
with renal vein flow rather than arterial flow.
Just to show you that this truly is a potential pitfall.
This is an exam that one of my sonographers did,
and these images are very magnified,
but this is what she initially thought
was the right renal artery origin,
and it's going in the right direction when compared
to the mid right renal artery.
But look at the difference in the
morphology of these waveforms.
This is venous pulsations as opposed
to arterial pulsations.

Now, another arterial pitfall in the kidneys
that can occur, particularly in transplants, is due
to probe pressure.
And remember that renal transplants are in a very
superficial location, so it's easy
to compress the parenchyma with the transducer.
And when you do that, you depress arterial flow
to a greater extent than you depress systolic flow.
So this can result in an abnormally
elevated resistive index.
Now, the pearl on this case is to suspect
that this is the case when you have RIS
that vary in a transplant from one location to the other
or from one time to another,
and just alert the people that are doing the exams that they need to have a light hand
and not use excessive pressure when doing renal transplants.
So this is an example of a renal transplant,
very thin individual, very superficial transplant.
And at one point in the examination you can see
that the arterial waveform from the transplant
looks entirely normal.
But at another point in the examination, you can see
that the waveform looks extremely abnormal
with complete absence of end diastolic flow.
And this was entirely due
to variations in transducer pressure.

Another pitfall that can occur
with the renal artery is situations where your doppler signal has simultaneous venous and arterial flow,
and we call this venous arterial contamination.
And when that happens,
the venous flow can potentially simulate
diastolic flow reversal.
Now, if you're dealing with a renal transplant,
diastolic flow reversal can be a sign
of renal vein thrombosis.
The pearl in this case is to look for simultaneous flow
above and below the baseline that occurs in systole
and simultaneously occurs in diastole.
As we see here, there's flow
below the baseline in systole.
There's also flow below the baseline
and above the baseline in diastole.
That would not happen with the two
and fro pattern that you get with renal vein thrombosis.
And then in addition, if you compare the venous,
the isolated venous wave form with the suspected diastolic flow reversal from an arterial waveform,
you can see that they're identical.
Now here's a couple of examples.
These are actually in native kidneys,
but you can see this waveform on the left looks very much
like a two andro waveform with antegrade flow in systole
and retrograde flow in diastole.
But notice that there's also some antegrade flow in diastole
here that overlaps with this apparent retrograde flow
that tells you that this is not really a true example of a to and fro wave form.
The same thing is present here where we have a little bit
of reversal from,
or a little bit of v venous flow that occurs
during systole and overlaps the antegrade systolic flow.
And this venous flow overlaps the antegrade diastolic flow.
Now, this truly is a pitfall.
This is an example above of two
and fro flow in a patient with renal transplant venous thrombosis.
And you notice that there's systolic flow above the baseline
and there's diastolic flow below the baseline,
but they never occur simultaneously.
And this is an example from the literature
where this was reported as diastolic flow reversal.
But I think if you look real carefully,
this in fact is probably an example of contaminated arterial signal with venous flow.

Hepatic Artery

Now we're gonna move on to the hepatic artery.
And you can get the same pitfall
with hepatic artery wave forms
where the portal venous signal overlaps with the
hepatic arterial signal.
And when that happens, it can lead to a misinterpretation of the arterial waveform.
And in particular, it can cause problems
with measuring hepatic artery resistive indices.
So you can see here when the portal venous waveform is added
to the arterial, the portal venous wave form can obscure
diastolic flow,
and make it difficult to measure an accurate RRI.
So the pearl in this case is decrease your sample volume
size as much as possible.
So you're either sampling the artery or the vein,
but not both simultaneously.
Move your sample volume around.
And if you still can't eliminate this overlap,
then just select a different hepatic artery to sample from.
So here's an example. We can see a portal triad here
with both portal venous and hepatic arterial flow.
This is what the portal venous signal looks like.
This is what the initial hepatic artery signal looked like.
It looked like there was a fair amount of diastolic flow,
but notice that this diastolic flow is exactly the same
as this portal venous flow.
And this is actually due to the phenomenon
that we were just discussing now when we were very careful and tried to really isolate
that hepatic artery.
You can see there truly is some diastolic flow here,
but it's a little less than the portal venous flow.
So the measurement of RI here would be correct.
The measurement of RI here would be incorrect.

Now, another potential pitfall
that can affect the hepatic artery and
can affect other arteries for that matter is patients that have left ventricular assist devices.
These can produce very abnormal waveforms
that simulate a parvis TARDIS waveform.
And this is important when you're evaluating liver
transplants or hepatic artery potential hepatic artery injuries
because parvis TARDIS waveforms would be a positive finding
in both of those situations.
Now, LVA DS
that are non pulsatile will cause a marked blunting
of the arterial signal.
And this can not only simulate arterial stenosis,
but it can even simulate venous flow
'cause it looks so monophasic.
So the pearl in this case is
recognize when the waveform looks extremely abnormal,
that you may be dealing with a patient that has A-L-V-A-D,
look at other arteries
because all of the arteries will be affected similarly.
And then simply pull up a chest x-ray on the patient
and it'll be obvious
that the patient has a left ventricular assist device.
And here's an example. This was a woman
with congestive heart failure, coagulopathy
and abnormal liver function test, who needed a exam
to evaluate for hepatic artery thrombosis?
And here on the magnified view of the port of hetus,
you can see the portal vein and hepatic artery.
Separately, this is the signal
that came from the main portal vein,
and this is the signal that came from the
main hepatic artery.
And notice how venous this signal looks.
This is because this patient had a non pulsatile LVAD
and you just couldn't get a pulsatile waveform within the
hepatic artery or any of the other arteries.

Gallbladder

Now we'll move on to the gallbladder.
Twinkling artifact is a very useful artifact,
but it can simulate true doppler signal,
and can potentially be confused with real blood flow.
And this is an example in the gallbladder of a patient
that came in with an outside diagnosis of HEPA
or a gallbladder mass.
And you can see on the gray scale image
that the gallbladder is completely filled with what appears
to be soft tissue.
Now, most of the time we would assume
that this is just sludge,
but when you get a doppler signal from it, you say, well,
gosh, maybe that isn't just sludge.
Maybe that is a vascularized soft tissue mass.
Well, the fact of the matter is this isn't true flow.
This is twinkling artifact that's simulating flow.
And the pearl in this case is that when you're dealing
with twinkling artifact, the signal is very alias.
So you're gonna have very highly saturated colored doppler signals.
And also if you try to take wave form,
it will be completely non physiologic.
You won't get an arterial
or a venous signal from a twinkling artifact.
And that's opposed to true masses, which will have
detectable flow on color doppler.
And the flow will be linear
like vascular structures as opposed to little tiny dots
of color, which is what you get with twinkling artifact.
And then finally, when you do waveform analysis,
you'll get a detectable signal
that has a physiologic appearance.

Scrotum

Now dealing with the scrotum,
there's a pitfall called acoustic streaming.
Now, acoustic streaming is a mechanical
non cavitation bio effect that occurs
because when waves travel through fluid,
they transfer momentum to the particles in the fluid and cause those particles.
And the fluid itself to move
this motion is always in the direction
of the sound propagation.
It occurs most prominently
with the higher frequency transducers
and particularly above seven megahertz.
And the motion velocity tends to be higher
with colored doppler
and pulse doppler than it is in the gray scale imaging mode.
So the pearl in this case is to note that
with acoustic streaming, the velocity is always constant
and it's always away from the probe regardless of
what the direction of the probe is.
So here's an example of a spermatic seal on the color Doppler image.
Here you can see that there is a signal
generated within the hepatic
or within the spermatic seal.
It's always a signal that's negative
because it's going away from the
direction of the sound pulse.
Now you can see on gray scale
that there is some motion within the little particles within
the spermatic seal, but notice on color Doppler
where we've turned color doppler on,
but we've suppressed the color Doppler signal.
You can see when color doppler is on
that the particles move much faster
in the Doppler mode than they move over here
in the gray scale mode.
Now, this doesn't just happen with strata seals
and it doesn't just happen in the scrotum for that matter.
It can occur in any sort of fluid collection
or cystic structure that has low level internal echoes.
And this just happens to be a case of hydros seal
that has some particulate matter in it,
and you can see the signal both on color doppler
and this monophasic negative signal on pulse doppler.

Aorta

Now moving on to the aorta.
There's a artifact called mirror image artifact,
which you're all very familiar with.
It's very common with colored doppler analysis,
and it can result in duplicated doppler signals.
When that happens in a aorta
that's had a stent graft, it can create a false positive impression of an endo leak.
Now, the pearl in this case is that you need
to view from different approaches.
So for instance, if this is aortic aneurysm
with this stent graft within it,
if you view it from an anterior approach,
then the mirror image artifact will be immediately deep
to the true flow within the lumen.
Now, on the other hand, if you view it from a lateral
approach, then the abnormal signal will continue
to be located here on the opposite side of the stent.
And on your image, it will look like it's right there
as opposed to a true endoleak, which would appear
to be right here if you came from a lateral approach.
Also, if you compare the waveform from the stent
and the waveform from the artifact, you'll see
that they're similar in size and shape,
although they may differ in intensity.
So here's an example of endoluminal stent graft.
And from an anterior approach,
you can see flow within the stent,
and then this mirror image artifact deep to the stent.
When we come from a lateral approach, we would expect if
that was, if this was truly an endoleak,
that it would occur over here on this side of the stent,
but it's occurring deep to the stent again,
and that's typical of a mirror image artifact.

General Technical Pitfalls

Now we're gonna move on to some pitfalls that
can occur virtually any place.
And one of those is baseline ambiguity.
When the baseline is moved to either the top
or the bottom of the doppler scale erroneously, a signal
can be alias into the opposite side of the scale.
And when this happens, it can lead to confusion
and velocity measurements.
So for instance, here's a arterial waveform
where everything is set up properly.
The baseline is down at the bottom,
the arterial signal is a positive signal,
and when you measure this velocity, you get a velocity
of 20 centimeters per second.
Now, if the baseline was in the middle of the scale,
then you would see that this is an alias signal down here,
and you would recognize that this is not something
that you can measure from.
If however, you move the baseline all the way to the top,
the entire signal is alias.
And if you don't recognize that
and you measure this as the peak velocity,
what you'll actually be measuring is this distance from the
cursor to the baseline rather than the distance
that you want, which is from the cursor down here.
And in addition, you'll get a negative result
rather than a positive result.
So in order to avoid this, the pearl is
to just don't move your baseline all the way to the top
or the bottom of the scale unless you absolutely have to
and then pay attention to the sign of the velocity.
If you're expecting a positive velocity,
you should have a positive signal.
If you expecting a negative signal,
then you should have a negative velocity.
And here's an example. This is a tip shunt where
as you can see in the colored upper image,
the flow in the shunt is going away from the transducer.
This is a negative signal.
And when we measured this here, we got a velocity
that was 48 centimeters per second,
but it's a positive velocity, and that's
because the baseline is actually up here.
The baseline is actually right up here.
So what we're measuring in this case is this
velocity rather than this velocity.
Let's just show that a little bit more dramatically.
So here's the abnormal waveform.
The baseline is actually right down here.
And so the distance
that we're measuring is this distance, not this distance.
If we move the baseline up to where it should be,
then we're actually measuring the proper velocity when we put our cursor down here
and notice that it's a negative result rather than a
positive result, as you would expect.

Now, probe drift is another potential problem.
When you are doing a scan, it's not uncommon for your probe
to change its orientation slightly
or for the vessel to move slightly, particularly
with the patients holding their breath
but not holding their breath terribly well.
And this can produce non-representative wave forms
and lead to inaccuracies in ri calculations.
So as you can see with this waveform, as we start to drift,
the flow starts to decrease gradually.
So in this case, ri should be measured here
rather than being measured here
because this is not representative of true diastolic flow.
So to minimize this, just make sure
that your diastolic closer cursor
and your systolic cursor are close together
and measure the in diastole
and early systolic
velocity rather than in diastole
and systole from earlier in the cardiac cycle.
And here's just an example.
This is a waveform
where we've got some slow drift in the transducer
towards the end of the sweep.
And notice that when we measure the ri this way,
we get a measurement of 0.75.
When we measure it here, it's a more accurate RI of 0.68.
So this is really the way we should be doing our IRI
measurements to avoid any chance of probe drift.
This is not the correct way.

Now, another pitfall in measuring a resistive indices is
when your doppler scale is too high,
when you have a very high doppler scale
and a resulting small waveform minor changes in cursor
placement can lead to large differences in RI measurement.
So you should never try to measure a resistive index
or velocity for that matter when the waveform is this small,
adjust your doppler scale so
that you've got a nice big waveform to measure from.
And then small changes in your cursor placement will have
very minimal change in velocity measurements
and RI determinations.
So here's an example. Here's a waveform
that's been obtained at a relatively low doppler scale.
And what I've done is place two different cursors in diastole and systole
and notice that there's very little difference in the height
of these cursors, and it's the height
that's important in measuring resistive indices.
So they're very close together
and with a waveform that's this big,
the resulting difference in your RRI measurement is very
small 0.73 and 0.72.
But when your waveform is really small,
because of a high doppler scale,
small differences in cursor placement have a more dramatic
effect on your RRI measurements.
And in this case, the RRI changes from 0.75
to 0.67, so from an abnormal result
to a normal result.

Now, another pitfall that can cause problems primarily in determining flow direction
is aliasing.
So on color doppler aliasing will produce a wraparound in
your color assignments.
If you look at the doppler scale here, when you alias,
the wraparound is gonna go from the high frequency shifts to the low frequency shifts,
but true flow reversal will occur.
And that will produce a transition between the very yellowish reds
and the greenish blues depending on what your doppler scale
or your doppler map is like.
But it will cause color changes from here
to here when there's true flow reversal.
It occurs from the lower frequency shifts on either side
of the baseline, and your transition will be from the dark
reds to the dark blues,
and there will be a little black line in between
because of the low frequency filter.
So here on a tip shunt, this is aliasing,
which is very obvious because of this color transition.
And this is true change in flow direction as you can see
with the little black line between the two
and the change from dark red to dark blue.
Now, aliasing is actually a useful artifact.
You can see in this case, this highlighted an area of stenosis.
But if you're trying to eliminate
or diminish aliasing, you can increase your doppler scale.
And then also in determining what's flow reversal and what's aliasing, just pay close attention
to your color transitions.
And here's an example of a splenic vein.
On a high Doppler scale plus
or minus 20, you can see
that it looks like it's going in the normal direction antegrade towards the portal splenic confluence.
But when we decrease the doppler scale,
there's enough aliasing
that this can really simulate flow reversal.
So pay attention to that when you're using color doppler to determine flow direction.
Here's an example of a tip shunt
where the doppler scale is so low that we've got
so much aliasing artifact that it's wrapping
around multiple times and you just get this mosaic
appearance on color doppler.
Now when you readjust the doppler scale
to a more appropriate value, you can see
that the flow looks much more normal in this dip shunt.

Now with pulse doppler, aliasing induced wraparound can cause confusion in analyzing the signal.
And here's an example
of a arterial wave form from a dialysis shunt.
Now, with a relatively high doppler scale,
everything looks fine as the scale is dropped.
We start to see some aliasing here below the baseline.
But this is typical aliasing and is easy to recognize.
As we continue to drop the doppler scale, we're starting
to get more overlap and wraparound of arterial signal,
and this is becoming a little bit more confusing.
And then finally, if we really drop the doppler scale,
there's complete overlap of arterial systolic
and diastolic flow.
And this looks completely non physiologic.
So this type of a waveform would be difficult
to analyze unless you increased your doppler scale.
So the pearls are increase your doppler scale since the doppler frequency shift is proportional
to the transmitted frequency,
decreasing the transmitted frequency will also minimize aliasing.
So that's something to attempt as well.

Now when you're measuring acceleration
or when you're trying to
recognize parvis tardis changes,
the sweep speed is important.
With respect to acceleration measurements,
minor changes in cursor placement can lead
to very large differences in systolic acceleration when the
sweep speed is slow.
So for instance, here we have a sweep speed
where we've got multiple cardiac cycles kind
of all crowded into one sweep speed.
When you're measuring acceleration which is the systolic upstroke here, the slope
of the systolic upstroke this is gonna give you an inaccurate acceleration.
What you really want to do is to speed up your sweep speed
so that the signals are stretched out.
And at this point, minor changes in these cursor placements
won't affect your acceleration measurement as much.
So here's an example. Here's a arterial signal
that's been obtained with a slow sweep speed.
When we measured the acceleration twice,
and you can see there's two cursors here
in systole diastole.
The difference range from 200 centimeters per second
to 500 centimeters per second.
And that's for just minor changes in these
cursor placements.
Now with a much faster sweep speed,
these minor changes in cursor placement result in much less
in inaccuracy in acceleration measurement.
I also realize that just subjectively when you're analyzing for parvis tardis waveform
that the sweep speed will make a difference.
A fast sweep speed like is illustrated here,
stretches out the arterial signal
and can give you apparent harvest tardis effects even when
they're not there.
So this certainly looks like a blunted signal when it's at a very fast sweep speed.
But if we slow the sweep speed down
and look at the exact same signal,
it looks much less abnormal.
So the pearl in this case is when you're subjectively
comparing waveform, for instance, the upper
and the lower pole of the kidney
or the right kidney in the left kidney, you should do it
with similar sweep speeds.
And here's just an example.
This is a waveform that you might think is abnormally blunted has a parvis type of appearance when viewed at a very fast sweep speed,
but looks much less abnormal when viewed at
a slow sweep speed.

Next let's talk about tissue vibration.
This is a artifact
that occurs when there's turbulent blood flow
turbulence cause pressure fluctuations in the lumen
of the vessel, which can cause the vessel wall to vibrate,
and this vibration can be transmitted into the
perivascular soft tissues.
Now this is actually a useful artifact on color doppler,
but it can cause confusing appearances on pulse doppler.
So here's the effect.
Turbulent blood flow in this case due to a stenosis,
causes vessel wall vibrations
and then vibrations in the soft tissue,
which then give you this type of a doppler signal in the soft tissues.
Now on a pulse doppler,
the tissue vibration will result in a very high intensity,
but a low amplitude signal.
And that's shown here on this image
where we're actually sampling the tissues next
to a dialysis graft anastomosis.
And notice that the signal is very bright
because it's coming from soft tissue interfaces.
But it's not very high
because these vibrations aren't occurring at a high velocity.
Now, as you can see here, this causes symmetric signal
above and below the baseline
because the reflectors in the soft tissues are moving back
and forth rather than just going in one direction.
Now, this may be isolated if you put your sample volume over
the soft tissues and as in this anastomosis area here.
But here's an example of a renal artery stenosis
where we're sampling from the artery itself.
And you see the arterial signal here
and the very high velocities due to the stenosis.
But also notice that buried within this arterial signal
is a signal from vibrating soft tissues.
So it's not nearly as high, but it's quite strong
because it's coming from soft tissues
rather than red blood cells.
So the pearl in this case is
to correlate your doppler signal
with your color doppler image so that you don't confuse
tissue vibration for true vascular flow.
And then when this happens, increase your scale until the arterial signal is clearly identified
as discreet from the tissue vibration.
So here's an example. This is a femoral artery AV fistula, and you can see what looks like
a vascular signal here,
but this is actually the tissue vibration.
Notice you can see vibrating tissues
on the color Doppler image.
And the true arterial signal is all of this signal up here,
which is hard to recognize
because it's been alias at the limit
of the positive doppler scale.
And it's wrapping around now into the negative scale.
So you can see how this can produce a problem.

Now finally,
one particular pitfall that we all encounter all the time
is with respect to doppler sensitivity.
You're examining a vessel, you're trying to show
that there's blood flow present and
because of sensitivity issues, slow flow, et cetera,
you just can't quite do it.
So remember that you need to adjust your doppler parameters
to increase sensitivity.
So with respect to the doppler scale,
it should be decreased.
The doppler gain should be increased.
Doppler angle should be decreased.
Transmit frequency in most cases in the abdomen
should be decreased.
Wall filter decreased.
The packet length that's available on your unit should be increased
and the sample volume should be increased.
The one that I really want
to emphasize is your transmit frequency.
I can't tell you how many times I've seen cases
where we haven't been able to detect flow in a vessel
because we're transmitting at too high
of a transmit frequency.
And here's a good example.
We're looking at a transmit frequency of 3.3 megahertz,
and we can't see any flow in the main portal vein in this
patient that's had a tip shunt.
When we drop the transmit frequency down to two megahertz
flow is readily detectable despite the fact
that no other changes have occurred in the
technical parameters.

Conclusion

So that concludes this lecture on pearls
and pitfalls of abdominal doppler.
I hope this will help you in your day-to-day practice
as you go forward
and do your abdominal doppler examinations.
Thank you very much.