Ultrasound Evaluation of Portal Hypertension - SD

Introduction

Hello, my name is Dr. Leslie Scout and I'm a professor of diagnostic radiology at Yale University School of Medicine, where I am the chief of the ultrasound service at Yale New Haven Hospital.

Today I'm going to be talking about Doppler ultrasound evaluation of portal hypertension and the hepatic vasculature.

We are going to be talking about ultrasound evaluation of portal hypertension and the hepatic vasculature.

The objectives of this presentation are to describe the normal appearance of the hepatic vasculature to describe the ultrasound findings of portal hypertension and cirrhosis to discuss the ultrasound appearance of portal and hepatic vein thrombosis.

And finally, to discuss the ultrasound evaluation of tips and the ultrasound findings in tip stenosis.

Technique

In terms of technique, it is important to try and examine these patients after they have been fasting for approximately eight to 12 hours.

And because we are looking at structures deep in the abdomen, you need to use a relatively low frequency transducer.

And typically we use a two to five megahertz curved array or phased array transducer.

And similar to all Doppler ultrasound examinations, we begin by optimizing our gray scale image and looking at the vessels on the B mode imaging.

To do this, you're going to need numerous acoustic windows.

This is probably one of the most difficult ultrasound exams that we do, and one acoustic window will not provide adequate visualization of the entire hepatic vasculature.

While we begin using the transverse anterior abdominal approach, it's important to recognize that very often we need to use oblique imaging, typically aiming towards the right shoulder to see the main portal vein, but also to use decubitus view.

You need to ask the patient to turn to the right or left to scan coronal through the flank in order to achieve complete visualization of the blood vessels to the liver.

We use color pulse as well as power doppler imaging, and you do want to maximize your settings for slow flow.

Here is an example indicating how important that is, and you can see the main portal vein in the center of this image, this color Doppler image.

And although you can see flow in the hepatic veins as well as the hepatic arteries in red here, you don't see any flow in the main portal vein and that would lead you to the conclusion that it is thrombose.

However, note the scale, the scale is relatively high and it is set too high to detect very slow flow.

When you decrease that velocity scale as you see in, we did just a few minutes later, you can see now that there is complete fill in of the main portal vein with color flow and you have the correct diagnosis, which is that there is no portal venous thrombosis.

The other things that you can do in order to maximize your ability to detect slow flow on Doppler interrogation is to decrease the size of the color box to make sure your focal focal zone is set just at the area that you're trying to image and to make sure that your wall filter is set at the lowest possible frequency before your image is degraded by noise.

It's also important that you do your doppler examination while the patient is in quiet respiration.

And the reason for this is that if you ask the patient to take a deep breath, which you often do, that they will push the liver lower into the abdomen below the rib, it is easier to look at changes will occur in the portal vein.

Once the patient takes a deep inspiration, the waveforms in the portal and hepatic veins will flatten.

The portal vein will increase in diameter and the portal vein velocity will decrease.

If you can only see these vessels during deep inspiration, you should note this or annotate this on the films.

Protocol

Our protocol includes the following.

We evaluate the aorta and obtain a waveform, doppler tracing.

There we evaluate the main right and left hepatic arteries, the main right and left portal veins, the splenic vein at the hilum and under the pancreas, the superior mesenteric vein.

Although this is the most difficult vessel to see the right middle and left hepatic veins as well as the inferior vena cava.

And I do want to emphasize particularly if you're evaluating a patient's status post liver transplantation, that it is very critical to identify and obtain a spectral doppler tracing of the waveforms in the int hepatic right and left hepatic arteries as well as portal vein.

Just seeing the main hepatic artery in portal vein is not enough.

You do need to see the int hepatic branches as well.

Normal Portal Vein Waveform and Diameter

What is the normal, what form of the, portal vein?

The normal portal vein wave form should demonstrate continuous forward hepato petal flow with relatively low velocity.

It is non pulsatile in contour, but will demonstrate mild respiratory variation.

As I mentioned already, the diameter of the portal vein will vary.

It will increase after eating with deep inspiration or if the patient performs a Valsalva maneuver or coughs or sneezes, it will decrease if you scan the patient in the upright position or after fasting or once the following, exercise.

And as we know from all doppler imaging, the velocity within a vessel varies inversely to the diameter.

If the diameter of the portal vein increases following deep inspiration, the velocity will decrease after a deep inspiration.

What is the normal diameter of the portal vein?

In quiet respiration, it should be less than 13 millimeters, and in deep inspiration it should be less than 16 millimeters.

And the normal velocity of the portal vein hovers around 20 centimeters per second, maybe between 10 and 25 centimeters per second in range.

Splenic Vein

The wave form of the splenic vein is very similar to the portal vein and it will be heading in a direction away from the splenic hilum towards the liver, a oped in direction, and it also will demonstrate slight respiratory variation.

And we do evaluate the splenic vein in the renal hilum, but also as it hugs the undersurface of the pancreas heading towards the portal confluence, as you see here on the color image, it is difficult to find the superior mesenteric vein.

Superior Mesenteric Vein

But let's go back to the anatomy and that will help you know how to look for it.

Here you can see the splenic vein and where the splenic vein meets the superior mesenteric vein and forms the main portal vein.

This is called the portal confluence.

And in this, location in the body, the superior mesenteric vein is basically perpendicular to the course of the splenic vein.

That means that the easiest way to find the superior mesenteric vein is to find the splenic vein underneath the pancreas, see where it dilates at the portal confluence, and then turn 90 degrees in that location.

And you ought to see the, superior mesenteric vein elongated.

The superior mesenteric vein runs in a cradio conda dimension, basically parallel or in the same plane as the inferior vena cava or the aorta, but superior to that or more anterior to it.

And if you image in a direction like this, so that you are looking at the superior mesenteric vein, over the panc, over the, inferior vena cava, note that the uncinate process of the pancreas courses between the superior mesenteric vein and the inferior vena cava.

Similarly, if you image the superior mesenteric vein in this plane, you'll also see a little bit of that uncinate process interposed between the superior mesenteric vein as well as the aorta.

Here is an example where we have turned 90 degrees to the splenic vein in the portal confluence.

And you can see this elongates, the superior mesenteric vein where it runs over the, the, inferior vena cava.

And note, you can see that uncinate process interposed between the superior mesenteric vein and the inferior vena cava.

Now why is it so important to see the superior mesenteric vein?

Really because it's a good landmark for the coronary vein and this imaging plane that depicts the elongated superior mesenteric vein traveling in this, from the feet towards the head towards the diaphragm, this is the best plane to look for the coronary vein.

And the coronary vein will be a little venous structure that dumps into the portal confluence very near the, where the superior mesenteric vein, joins the splenic vein.

Of course, the blood in the normal coronary vein will be heading in the opposite, opposite direction towards the f.

Hepatic Artery

The hepatic artery in most patients arises from the celiac axis and is the first branch towards the right of the patient.

Once it, branches off the celiac axis, it runs in the hepato duodenal ligament, anterior and medial to the portal vein and medial or under the common bile duct in the portal heus and relative to the straight, main portal vein, the hepatic artery is a little bit tortuous in terms of its characteristic Doppler signature waveform.

It is characterized as you see in these two examples with a very sharp systolic upstroke and continuous forward flow during diastole.

In other words, it has a low resistance waveform pattern.

Hepatic Veins

The hepatic veins drain the liver and so they run in a direction away from the liver or hepatic ugal flow, and their characteristic spectral waveform, tracing has pulsatile or phasic uh, contour.

And this pulsatile waveform reflects the right atrial pressure.

This schematic demonstrates how the deflections in the waveform correspond to changes in pressure in the right atrium.

This first valley is, called the S wave, and this reflects filling of the right atrium and systole.

This indentation here, the V point is due to overfilling of the right atrium just before the tricuspid valve opens.

Then there is another surge of flow away from the liver, the D wave, which corresponds to the filling of the right atrium and diastole.

And then there is this, a wave which is called the atrial contraction or the atrial kick.

And this occurs obviously as the right atrium contracts.

And during this time period there is a little push of blood flow back towards the liver in a a pedial direction.

And here is a waveform from one of the hepatic veins.

And you can see again the S wave, the V wave, the D wave, as well as this a wave corresponding to the right atrial contraction or atrial kick.

Portal Hypertension: Definition and Causes

How do you define portal hypertension?

Portal hypertension is just indicative of an increase in the gradient between the portal system and the systemic circulation.

And this gradient can be measured relative to the hepatic veins, the inferior vena cava or even the right atrium.

But most angiographers feel that comparison.

Comparing the pressure in the portal system to the right atrium is the best way to, judge the portal systemic gradient.

Normally this gradient is less than seven millimeters of mercury.

Once the gradient rises to be between seven and 10 millimeters of mercury, this corresponds to relatively mild porter hypertension.

And if the gradient exceeds 10 millimeters of mercury, there is a significant increase in the risk of variceal bleeding.

How do people develop portal hypertension or this increased gradient?

Basically the initial thing is that they get increased resistance to portal flow in the liver, and this can be due to morphologic abnormalities, changes in the, hepatic anatomy that causes obstruction to portal flow and sometimes the body secretes vasodilators into the, vasoconstrictors into the bloodstream which cause constriction of the portal vessels and that also increases resistance to portal flow.

As you likely remember, flow to the liver, 75% of it comes from the portal vein and about 25% of it only comes from the hepatic artery.

And when you get increased resistance to portal venous flow, this decreases the flow to the liver and your body compensates trying to increase the flow.

And the way the body does that is by increasing the splank nick flow.

First thing that happens is that the body will retain salt so that it retains fluid and so the patient becomes relatively volume overloaded and it increases the blood volume.

In addition, the body will secrete vasodilators to counteract the vasoconstrictors that I've mentioned previously.

And this will also cause dilatation of the blood vessels in the abdomen in an effort to increase flow to the liver.

In addition, sympathetic tone is increased and this also causes relative vasso dilatation.

Basically you have a summation of these two factors, increased resistance to portal flow, as well as increased splank neck flow, which actually volume overloads flow to the liver and that actually makes the, increased peripheral vascular resistance in the liver.

Worse, these two things combine to increase the portal systemic gradient, and the end result of this increase in port systemic gradient is that the body develops portosystemic collaterals to shunt blood away from the liver.

And flow will actually ultimately reverse in the portal vein again to shunt blood away from the liver.

Ultrasound Findings of Portal Hypertension

What are the ultrasound findings of portal hypertension?

First, one of the first things that you can see is a change in diameter in the portal vein.

A diameter that is greater than 13 millimeters in quiet respiration or greater than 16 millimeters in deep inspiration.

And while this is a relatively insensitive finding, it has a very high specificity of nearly a hundred percent.

The next thing will happen is that the wave form will become relatively flat and there will be a loss of the respiratory variation.

Subsequent to that, the velocity will decrease decreasing from 20 centimeters to second till 10 to five till eventually you'll get, bidirectional flow kind of wishy-washy to and fro flow within the portal vein, and then frank reversal will occur.

On these uh, images, you can see that there is uh, increase in the diameter of the main portal vein, and you can see this very flat wave form on these images.

You can see that the flow in the portal vein is opposite in color to flow in the hepatic artery.

It is reversed on the color image and it is below the baseline on the spectral tracing heading in a direction away from the liver capsule.

This is the appearance of reverse flow in the portal vein on the spectral doppler tracing here as well as the color doppler image.

These are relatively late findings and you cannot use these findings to actually try and grade the degree of portal hypertension or to estimate the portal systemic gradient.

Those gradients still have to be obtained an geo, an geographically, but again, these are late findings of portal hypertension.

What they indicate though, of course, is reduced flow in the portal vein.

And the body does compensate by increasing flow in the hepatic artery, so the hepatic artery will dilate and develop this corkscrew appearance.

And here on this color Doppler image, you can see that that the red hepatic artery sort of spirals around in a corkscrew appearance spiraling around the portal vein.

And it's easy on a gray scale image to mistake this for dilatation of the bile duct.

If you see a dilated tubular structure in the Porto Heus in a patient with portal hypertension, again, check with color because it may just be a dilated corkscrewed hepatic artery.

Other things that occur are increase in the size of the spleen, the de development of ascites and the development of varice.

Varices

Varice can occur from uh, two different, uh, pathways, either reversal of flow within existing veins or recanalization of embryonic channels.

And your sensitivity of finding these varis on ultrasound really just depends on how hard you look for them.

And if you look hard, you should probably be able to find them in about 90% of all cases.

The ones that are the most important to identify are the gastroesophageal varices.

And the key to finding them is knowing where to look and knowing the the vein that feeds them, which is namely the coronary vein.

Here is a schematic, demonstrating that varice can occur really practically almost anywhere throughout the abdomen.

But in order to see again that coronary vein, note that here you have the splenic vein.

Perpendicular to that is a superior mesenteric vein.

And just above that in the same plane will be the coronary vein and you follow that coronary vein to the diaphragm.

This is where you're going to see the gastroesophageal varis.

Here is an example of a dilated and tortuous coronary vein, and you can see the coronary vein in a normal patient, but it is abnormal if it's greater than four millimeters in diameter, or if you see flow heading towards the liver here, blue away from the transducer, towards the diaphragm.

Flow heading towards the head is abnormal, and if the diameter of that coronary vein exceeds seven millimeters, there is substantial increased risk of variceal bleeding.

In terms of looking for the varices themselves, you can try and find the aorta and follow it towards the head where it pierces the diaphragm and look for varices, that way near the uh, hiatus of the aorta.

Or you can look between the left lobe of the liver and the spleen just above the stomach.

You'll see these dilated, uh, serpiginous channels on the gray scale imaging.

And when you put the color doppler on, you can see that these fill in with color flow.

Another very common varix to see is the paraumbilical vein.

And the paraumbilical vein is an exten, uh, extension of the left portal vein.

It runs in the echogenic ligament terrace towards the capsule of the liver out beyond the margin of the liver, down the anterior abdominal wall as you see here towards the inferior epigastrics where it dumps into the semi systemic circulation at the level of the external iliac vein.

And it is abnormal if it is greater than three millimeters in diameter or if you can see flow extending in a hepato fetal direction, but beyond the liver capsule.

And one of the interesting things about the paraumbilical vein of course, is that if it is patent, it can prevent reversal of flow in the main portal vein.

If you have a patient who has portal hypertension and you know that and you examine the main portal vein and you don't see any reversal flow, you ought to be thinking that perhaps there is a patent paraumbilical vein because it basically is a shunt through the liver parenchyma.

And it will allow blood to keep going towards the liver in the main portal vein as it shunts through the liver in the uh, patent paraumbilical vein.

And this is a example of a patent paraumbilical vein on the mr.

Again, you can see it in the ligamentum te running between the lateral and medial segments of the left lobe of the liver.

And here is another example on an ultrasound on the gray scale image.

You can see this uh, tubular structure within the echogenic ligamentum terrace.

This fills in with color doppler with flow heading well beyond the liver capsule in this patent.

Paraumbilical vein.

Varice though can really occur just about any place.

And you want to survey the abdomen quickly in its entirety.

And anytime you see serpiginous, uh, dilated structures again check them with color doppler to see if their blood have uh, blood flow.

Here you see uh, varice below the inferior margin of the left lobe of the liver, uh, somewhat separate from the diaphragm, separate from the stomach, and these are varice of the short gastric or left gastric veins.

This patient had dilated tubular structures around the lower pole of the left kidney and you can see this huge varix coming from the outer surface of the spleen down around the left kidney.

And these are just large left retroperitoneal varice.

These two patients had spontaneous splinter shunts and you can see in the patient on the left that they dilated venous structure comes out of the splenic hilum, kind of makes a U-turn as it goes towards this dilated left renal vein.

And this is a spontaneous ple renal shunt.

And in this patient you can see again a different kind of a U-turn where the dilated venous structure comes from the splenic hilum to the renal vein and then crosses over the midline.

In this dilated renal vein, this patient had a development of a spontaneous porta cable shunt and you can see that there is a, uh, direct connection between the portal vein and the inferior vena cava on this color Doppler image.

As I said, though varice really can occur any place, and here are some unusual examples.

Here is the inferior vena cava and notice that there are these massively dilated varices in the left flank.

This patient had a spontaneous splenic cable shunt.

Here's this big spleen and you can see this large varix crossing over the abdomen to the right side of the patient anterior to the abdominal aorta where it dumps in directly to the inferior vena cava.

That's a spinal cable shunt, and this is a transverse image of the mid abdomen just demonstrating huge varis to the left of the aorta within the retroperitoneum.

Cirrhosis

Cirrhosis is a different phenomenon than portal hypertension, though it often occurs with portal hypertension.

However, you can have cirrhosis without portal hypertension and you can't have portal hypertension without cirrhosis.

I just want to point out that the ultrasound findings of cirrhosis are different than the ultrasound findings in portal hypertension.

They are not the same thing and these are really not so much vascular findings as changes in the contour and the echo texture of the liver.

The first thing that you'll see is a coarse echo texture of the liver.

And what do I mean by coarsening of the echo texture?

Generally the ultrasound image is made up of lots of little dashes and typically these are very fine and hard to distinguish as separate paint strokes.

But when you see these dashes that are much thicker and the liver looks very heterogeneous and you can really almost imagine very thick brush strokes here, creating this image of the liver parenchyma, this is what is typically described as coring of the echo texture.

The other thing that happens is that the liver decreases in size and the surface becomes very nodular.

Here you see three different patients with um, ascites and with the patient with ascites, it is very easy to see this nodular surface contour, uh, of the liver.

It's harder to see this in a patient without ascites, but if you look carefully in the near field, the most, uh, anterior surface of the liver with a linear ray transducer, remember the linear ray transducer does have better spatial resolution than the curved array.

Sometimes you can see the nodular contour even if it is not outlined by the presence of ascites.

As I said, the total volume of the liver, uh, shrinks often quite dramatically, but also volume redistribution occurs and in particular the right lobe of the liver decreases in size.

And there is compensatory increase in the size of the lateral segment of the left lobe as well as the caudate lobe.

And here on this transverse image, you can see the caudate lobe extending in front of the inferior venia cava and towards the midline.

And on the sagal image, you can see again a very big caudate lobe Underneath its echogenic ligamentum veno and patients with cirrhosis have increased risk for hepatocellular carcinoma.

In fact, patients with cirrhosis develop numerous uh, focal nodules.

And if you looked at a cirrhotic liver pathologically, you could see in fact that it is full of nodules.

However, most of these will be regenerative nodules and typically you don't see regenerative nodules on ultrasound examination.

If however, in a patient with cirrhosis, you see a focal nodule.

The burden of proof is upon you to determine that it does not represent hepatocellular carcinoma.

Any nodule that you see in a cirrhotic liver, no matter what the echo texture, whether it is echogenic as in this case or heterogeneous and hypo coic with or without a hypo coic halo as you see here, but any discreetly visualized liver nodule, no matter what the echogenicity, you need to make sure that this is not hepatocellular carcinoma.

Precipitating Causes in Portal Hypertension

In a patient with known portal hypertension who presents to you because of deteriorating liver function tests or changes in the, uh, signs and symptoms consistent with worsening of their disease such as ascites or lower extremity edema, it's important to look for precipitating causes that could have made that portal hypertension suddenly worse.

And the things that you wanna look for are particle vein thrombosis, but Chiari syndrome or thrombosis of the hepatic veins as well as hepatocellular carcinoma.

Portal Vein Thrombosis

What are the ultrasound findings of portal vein thrombosis?

First of all, you'll see distension or dilatation of the portal vein with internal echoes and absence of flow on color doppler imaging.

However, you can have false positives due to slow flow to the slow flow phenomenon.

And because of this, you want to again, always ensure that you have maximized your machine settings for detection of slow flow on color doppler.

In addition, because pulse doppler and power doppler are more sensitive for detection of flow than color doppler, you always want to confirm your impression on the color doppler with a spectral tracing or color power angiography.

And if further questions, remain, you should get a confirmatory study using MR or ct.

Here are some examples showing dilatation of the main portal vein with internal echoes consistent with thrombus.

On this patient on the right, you suspect that the, uh, color doppler settings are maximized for slow flow because you can see flow in the inferior vena cava deep to this.

But again, you would want to confirm with power or a pulse doppler tracing.

When the thrombus is not occlusive, you can see that it is outlined by the color flow and you can see a signal void and that can help you make the diagnosis of portal vein thrombosis.

And here is another patient with complete thrombosis of the portal vein.

And you can see this linear streaks within the liver, their hypo coic and right underneath the uh, hepatic artery, which you can see on this color, color power Doppler image, but also on the color Doppler image On your left here is another example where the portal, main portal vein is descended there contains intraluminal thrombus.

You can see that there's no flow on the color Doppler image.

And when you see this, you do want to confirm on, uh, other imaging planes, but you also want to look for the extent of the thrombus.

Here you can see that this patient with thrombus in the main portal vein, it extends all the way down to the portal confluence, and in fact, the splenic vein underneath the pancreas is thrombosis as well.

Again, a reminder that technique is important.

Here is an example where initially the sonographer thought that there was portal vein thrombosis, but note, just as I showed you at the beginning of this lecture, the velocity scale is quite high.

And when you reduce the velocity scale, you can see that there is fill in with color of the main portal vein and you come to the correct diagnosis, namely that it is patent and there is no thrombus.

When you think though you've maximized everything, you still don't see any flow.

Confirm that there's no flow on the spectral doppler tracing as well as the power doppler because both of these are more sensitive to slow flow than the color even when it is maximized.

And the last thing regarding portal venous thrombosis that I wanna mention is that it is important to try and differentiate bland thrombus from tumor thrombus associated with hepatocellular carcinoma.

And the way to do that is to look for vascularity within the thrombus.

And if you see arterial pulsations within the, uh, internal echoes of that distended portal vein, you know there's no way that there should be arterial signal within that portal vein.

That means there has to be tumor thrombus, uh, causing, uh, the arterial signal due to, uh, vascularization of the thrombus.

And you know, there must be underlying tumor.

How good is ultrasound for the detection of portal venous thrombosis?

The sensitivity is pretty high in the negative predictive value about 98%.

And what does that mean to your practice?

That means that if you have a negative doppler examination, in other words you see flow in the Doppler examination, you don't need to look farther to see if there's non occlusive thrombus or if there's thrombus that you just, uh, uh, couldn't detect.

The real problem with the Doppler ultrasound examination is that the specificity isn't that high and the positive predictive value not quite so high because you can have false positive exams due to a low flow state in patients with severe portal hypertension or cirrhosis.

Again, you need to maximize your settings to detect slow flow.

Remember to use power and pulse doppler and to confirm with either MR or ct.

Cavernous Transformation

In about 50% of patients with portal venous thrombosis, they develop cavernous transformation of the portal vein.

And this can occur within two to four weeks following, uh, acute portal vein thrombosis.

And it is due to the presence of, uh, or the development of periportal and peri gallbladder collaterals.

Collaterals within the gallbladder wall, and these occur primarily within the hepato duodenal ligament as well as the lesser mentum.

These do not work particularly well in terms of decompressing flow around the liver.

And it is very common to see esophageal varis in patients with cavernous transformation of the portal vein.

The findings on doppler ultrasound are characteristic and you see this, uh, spaghetti type of appearance of red and blue flow within these varis, a thick vascular gallbladder wall with uh, tortuous vessels within it.

And here you can see a confirmation of this finding these, uh, dilated vessels in the port of Hetus on the CT scan of this patient with cavernous transformation of the portal vein.

Budd-Chiari Syndrome (Hepatic Vein Thrombosis)

The ultrasound findings of Bud Chiari syndrome or hepatic vein thrombosis are very similar to the ultrasound findings for portal vein thrombosis.

You look for non visualization of flow in the hepatic veins or the inferior vena cava and the presence of echogenic intraluminal thrombus within the hepatic vein.

Here you see the intraluminal thrombus, again, these sort of hypo coic streaks through the liver heading away from the inferior vena cava, which do not fill in on the color Doppler image.

Another characteristic finding is been described as portalization of the hepatic venous waveform.

At the beginning of this lecture, I pointed out that the normal waveform in the hepatic veins is very pulsatile.

And when the waveform is flat, as you see in this example, and you do not see transmission of cardiac pulsatility from the heart, this suggests that there is something obstructing the transmission of the right atrial pressure from the heart to the area that you're interrogating.

And typically that is a thrombus, uh, somewhere in the vessels between the two locations.

Sometimes it can be due to external compression.

Here is another example of a patient, uh, that demonstrates that waveform note that the inferior vena cava is distended and contains thrombus.

Here you can see the thrombus on the transverse image in the inferior ven cava.

It's not occlusive, there's a little flow around it, but when you look at the waveform in the hepatic veins, uh, before this, you can see this flat, uh, because the transmission of the right atrial, uh, pressure is blunted by the presence of this thrombus within the inferior vena cava.

And the last thing that you want to look for to help make the diagnosis of bud Qri syndrome is the presence of collateral vessels.

And these can be intraparenchymal as you see here in the porta between the portal vein and the inferior vena cava.

Or they can be capsular between the hepatic and portal veins.

Normally you shouldn't see blood flow this close to the liver because the vessels are typically so small you don't see them with doppler.

But when you see these large vessels going right to the capsule a little bit beyond and then flowing back into the liver, these are cap capsular collaterals consistent with, uh, bud Chiari syndrome, you can occasionally have a false positive flat waveform, uh, indicating Bud Chiari syndrome.

That is only due to the fact that the patient has increased their intraabdominal pressure by taking a deep breath or by external compression.

Here you see an example of a patient that has a very flat wave form in the inferior vena cava, but a few minutes later you can see that it's normally pulsatile, uh, as I showed you at the beginning of the lecture.

And the only difference between these two images was that the patient had taken a deep breath and held it when we obtained this initial tracing of the vera vena cava.

Remember with a patient who takes a deep breath that can flatten the waveform in the inferior vena cava, uh, so can external compression.

In the obese patient that has a lot of weight in the anterior abdominal wall or a patient with ascites or a big mass in the liver, sometimes those things can compress the inferior vena cava and cause flattening in the waveform in the hepatic veins.

Gastroesophageal Hemorrhage and TIPS Procedure

The last thing I want to talk about is a gastroesophageal hemorrhage.

And this is due of course to bleeding from, uh, gastroesophageal varis and it's estimated that about a third of patients with uh, GE varices will ultimately bleed.

And if a patient bleeds once they have about a 70% chance of bleeding.

Again, risk factors for bleeding include a portosystemic gradient greater than 12 millimeters of mercury, a diameter of the coronary vein that exceeds seven millimeters, ascites, acute alcohol ingestion, and actually heavy exercise.

These patients are generally treated medically, but if medical therapy fails, uh, a tips procedure is currently the preferred secondary line of therapy and is now preferred over placement of a surgical shunt.

And the reason for this is that placement of a tips avoids the risks of general anesthesia and major abdominal third surgery, and it doesn't alter the extrap pad vascular anatomy.

The patient's position on the transplant list will not change.

There are very few contraindications to placement of atip, but the two biggest ones are inadequate liver reserve.

If you shunt too much blood around, uh, the liver and you don't have enough adequate, uh, liver parenchyma to detoxify the blood, the patient will end up with severe hepatic encephalopathy.

And if you shunt too much blood to the systemic circulation, it can overload the right heart and precipitate severe right heart failure.

In patients who have poor, um, cardiac status or inadequate liver reserve, sometimes the tips, uh, placement can make the clinical situation worse.

And those are really the only two significant, uh, absolute contraindications to placing a tips in terms of the following the patient.

Once the tip shunt has been placed, it's important to recognize that stenosis typically occur at the hepatic end.

In addition, it's also important to note that about 73% of all tip stenosis or malfunctions are detected while the patient is asymptomatic.

If you only examine a patient, uh, only examine the patient's tips once they're symptomatic because of bleeding, uh, redevelopment of ascites or abnormal liver function test, you usually have waited too late to identify the stenosis.

It is recommended that patients be placed on a monitoring schedule, that they get a baseline exam within 48 hours of having the tips shunt placed then at three months, six months, and 12 months and annually following that because you want to identify the stenosis when it's asymptomatic and relatively minimal and before occlusion occurs because it is much easier to fix the stenosis than it is to fix an occlusion.

TIPS Imaging Protocol

In terms of the imaging protocol, just like all doppler imaging, you start with the basics and you start with the gray scale imaging.

The tips has been placed either for control of variceal bleeding to deal with refractory ascites or pleural effusion.

And since that is the outcome that we are trying to affect with the tips, mainly reduction of the varice ascites and pleural effusion, the first thing to look for is whether or not those things have recurred.

Because if you see once again that the patient has varice, a large amount of ascites are big pleural effusion, you know that the tips cannot be working.

In addition, we measure the velocity in the main portal vein, and this velocity in the main portal vein should be measured in the midline about two to three centimeters before the proximal end of the steps.

Tips in the proximal mid and distal shunt in the inferior vena cava or the draining vein, you want to assess the direction of flow in the main and intrahepatic portal veins, the splenic vein and hepatic veins.

And lastly, you want to look at the hepatic artery and just like the examination of the liver parenchyma, excuse me, of the uh, liver vascularity, you will need to use numerous windows as well as velocity scales to do a complete examination.

The velocity scales will have to be relatively high for the tips because velocity is typically in the a hundred to 200 centimeter per second range, but you're going to need to decrease those velocity scales to look at the intrahepatic parenchymal portal and hepatic veins.

In addition, just like looking at the normal liver vasculature, you want to do this examination at the end of normal expiration.

If you can only examine the patient and see the tips after they've taken a deep breath, you need to recognize that the portal vein velocity will decrease and it can look therefore, like there has been a change in, uh, velocity in the tips when really all that has happened is that the patient has taken a deep breath.

Normal Findings of TIPS

What are the normal findings of a tip shunt on ultrasound examination?

Color should fill the lumen of the tips.

You should see increased flow in the main portal vein.

It increases from about 20 centimeters per second to 40 centimeters per second.

There will be increased flow in the hepatic artery, and typically you see high velocity, relatively pulsatile flow in the tips between 102 hundred centimeters per second.

In addition, the tips has been placed such that the gradient along the tips should be less than the gradient in the liver.

Parenchyma and blood will always flow according to the lowest gradient, and therefore flow should be towards the tips.

Towards the liver in the main portal vein, but also towards the tips from the intrahepatic vein.

You should see reverse flow in the left and right portal veins.

Here is an example of a normal tips and you can see that it has a very echogenic wall.

It is completely filled in with color, and you see pulsatile flow, high velocity 142 centimeters per second along the length of the tips.

And this arrow in the main portal vein indicates where you want to make the velocity measurement in the main portal vein, again, in the midline, about two to three centimeters towards the midline or towards the left from the proximal end of the tips.

Diagnosing TIPS Stenosis

How do you diagnose a tip stenosis with doppler ultrasound?

These criteria are very, very controversial.

And I'm just going to provide you with some general, uh, guidelines because it is really hard to say exactly how to precisely diagnose a stenosis.

First of all, look for the varice reac accumulation of fluid in the peritoneal cavity or the pleural space because if this has occurred, it is really proof that the tips is not working.

The gradient has increased and therefore a stenosis must be present on your color Doppler image of the tips.

Look for a filling, filling defect, luminal narrowing or color aliasing demonstrating an increase in velocity.

Look for a decrease in velocity in the main portal vein.

If you see that it is less than 30 centimeters per second in the vessel feeding the tips, this is a bad sign and, uh, often indicative of a distal narrowing in the tips.

Low velocity in the tips itself is also worrisome.

And although you'll see many different numbers quoted in the literature, most people now recommend using a velocity threshold of 50 to 60 centimeters per second.

If you see velocity below that, you should be worried about a distal stenosis.

Look for an increase in velocity at the site of the stenosis, anything in excess of 200 centimeters per second, and look for a change in velocity, either a decrease or an increase of more than 50 centimeters per second temporally from one exam to the next or spatially along the length of the tips.

And lastly, look for change in direction of flow in the left or right portal veins.

And if you see flow going, uh, towards the liver capsule or away from the tips in the intrahepatic veins, that's a bad sign.

Or if you see flow going away from the tips in the main portal vein away from the liver hepato ugal, that's a very bad sign.

It means that the gradient in the tips must be high.

It is easy to make the diagnosis of thrombus and complete occlusion.

And here you can see the echogenic wall of the tips, the color doppler flow in the vessels around it, but absolutely no flow within the tips itself.

And in this other patient, you can see in addition that the thrombus is extended now into the main portal vein, which is occluded, although there's a little bit of, uh, flow noted in the portal confluence on color and power doppler.

You can see narrowing of the, uh, tips here at the hepatic end where it ends in the right hepatic vein.

And on this color Doppler image, you not only see narrowing at the hepatic end of the tips here is the diaphragm, this echogenic line here, but also this, uh, change in the color, uh, signal indicative of ene and increased velocity.

Where you see this color aliasing, that's where you should look for the increased velocity and you should place your uh, uh, sample, uh, gate there to assess velocity.

And you note that in the proximal portion of the tips velocity is low under 60 centimeters per second, and where the color aliasing occurs, it's very high, 140.

There's a substantial increase in velocity indicating a stenosis of the tips, which you can see on the color Doppler image as well.

Low velocity approximately in the main portal vein.

You sample here in the main portal vein, and you can see that the velocity is only 25 centimeters per second.

That's a bad sign indicative of a distal stenosis.

And this patient has low velocity in the tips itself.

The velocity is only 15 to 20 centimeters per second.

That's very, very low, well under that 50 to 60 centimeter per second threshold.

And when you see velocity this low, the tips almost certainly has a tight stenosis, in fact is likely almost ready to occlude.

And lastly, look for change in direction of flow in the intrahepatic veins.

And here's an example of how the direction of blood flow changed within the left portal vein.

Initially, when the tips was well functioning, you can see flow in this direction heading away from the transducer below the baseline.

It's heading towards the tips where the gradient is lower and that's normal.

However, a year later, the came patient came back for surveillance and now blood is heading towards the transducer towards the liver capsule.

It's above the baseline.

And what that means is that the gradient, the portosystemic gradient in the liver, parenchyma in the left lobe has to be lower than the gradient in the tips.

And since you know that the gradient won't improve in the cirrhotic patient, what it means is that the gradient in the tips had to increase and it does that because it has become stenosis.

Conclusion

In conclusion, doppler interrogation has excellent accuracy for diagnosing complete occlusion of a tip shunt.

The criteria for diagnosing a stenosis of a tip shunt remain controversial, and the controversy is likely due to differences in patient population from center to center, differences in the type of stent that is deployed.

Differences in angiographic criteria at what Porto systemic gradient does one angiography department consider that the uh, uh, procedure has been successful and also likely differences in ultrasound technique.

But as general guidelines, you should be worried about a stenosis.

If the velocity in the main portal vein two to three centimeters before the proximal, uh, tips is less than 30 centimeters per second, a velocity less than six 50 to 60 centimeters per second within the tips, a decrease in velocity of more than 50 centimeters per second, uh, from one exam to the next or along the length of the tips and any area of significant increase in velocity at the site of the stenosis.

In addition, you should be very concerned about change in the direction of the, uh, flow of blood in the rest of the intrahepatic, uh, portal veins.

When they initially were flowing towards the tips or in a hepato ugal direction, and then they reverse, so the blood flow is now flowing towards the liver capsule or in a hepato petal direction, that is a bad sign.

But you should remember that ultrasound is a screening test and the sensitivity and specificity are inversely related.

And if every patient that you refer with a possible, uh, tip stenosis are demonstrated to have a tip stenosis on angiography, you are likely missing some cases.

You are going to, and you should in fact have some false positive examinations.

And how you titrate exactly when you refer a patient to a follow-up angiogram really depends on the tolerance of your patients and the clinicians for these false positive studies.

It is very important to do some clinical correlation to know when to intervene in terms of the ultrasound evaluation of portal hypertension.

Ultrasound is not an accurate means of assessing, uh, specific, uh, level or number for the port of systemic gradient.

It's not a good way to actually grade cirrhosis that still needs to be done with biopsy or by angiographic procedure.

But you can see findings of portal hypertension, namely increase in diameter of the portal vein and reversal of flow direction in the portal vein.

But these are late findings of portal hypertension when you evaluate your patient with portal hypertension or suspicion of portal hypertension.

In addition to the Doppler exam, remember to look for varice and to look for causes of worsening liver function, namely portal vein thrombosis, bud Chiari syndrome, or hepatic vein thrombosis, as well as tumor, uh, most commonly hepatocellular carcinoma.

Thank you very much.