Doppler Imaging of Liver Transplants: Diagnoses and Dilemmas - HD

Introduction

Hi, my name is Dr. Deborah Rubins. I'm from the University of Rochester, and I'm gonna be talking to you today about Doppler imaging of liver transplants diagnoses and dilemmas.

Background on Liver Transplants

As a background, the indications for liver transplant are quite diverse. They range from acute irreversible liver failure to chronic end stage liver disease, which can be from hepatitis B, c, or other causes, for hepatocellular carcinoma. For pediatric, causes that are congenital, including biliary atresia cirrhosis, or even congenital hepatic fibrosis.

Types of Liver Transplants

And there are two types of liver transplants we'll be discussing. There's an orthotopic liver transplant, which is basically from a nearly deceased person and the whole liver is transplanted. There's also a live donation, in which case one part of a liver is resected from a living donor and usually the right lobe is used for adults and the left lobe for pediatric cases.

Statistics on Liver Transplants

Just to give you some background of how many people need liver transplants in 2009, these are the latest statistics from the transplant network. And, there were over 16,000 patients on the transplant waiting lists, newly registered patients over 11,000. Unfortunately the total number of liver transplants is only six and a half thousand. You can see that there were a lot of people waiting for livers.

Again, just to remind you, there are two types of liver transplants. Most of the ones that we're gonna be seeing and talking about are the deceased donor liver transplants. Those have been increasing slowly since basically the year 2000, but have sort of plateaued off, and that's just because of the lack of available livers. Living donors certainly would have a potential for rising, but there's some complications associated with living donor transplants and so they have not taken off, as we would hope they could.

Vascular Connections in Liver Transplants

So what are the key vascular connections and the vessels we're gonna be looking at in liver transplants. There are basically four anastomosis, one for the hepatic artery, two you might need for the inferior vena cava, one superior and one inferior, and one for the portal vein. And this takes some time. The surgeons actually would like to eliminate one. And so now we are mostly seeing this kind of anastomosis, which is called a piggyback in which you can eliminate one of the IVC anastomosis. You eliminate the inferior one and you just basically hook the superior portion of the IVC from the donor onto the recipient in this sort of side to side or end side fashion. Creating a piggyback. And you can see that here when you look, Sally, you can see the native IVC behind and the donor IVC on top. And when you look in transverse, you can actually see the, again, donor here on the top and the native posteriorly before they've connected.

Imaging Protocol for Liver Transplants

So what's our imaging protocol when we're thinking about the vascular anatomy for the liver transplants? We wanna do a gray scale of the liver and the spleen. Also the biliary tree and the perio pad spaces. We're gonna use color and spectral doppler. We wanna be sure to see all these anastomosis. So we wanna look at the hepatic arteries as well as the int pad vessels main right and left the portal vein main right and left all three hepatic veins, right middle and left, the splenic vein and the IVC.

Hepatic Artery Complications

And the major complication we're gonna be looking for initially in the liver transplant is the hepatic artery. And it can be either completely thrombosis or narrowed, which means stenosis. The less common complications are actually on the venous side, including the portal vein, hepatic vein, or IVC.

Why is hepatic artery so important? Thrombosis or stenosis is probably the most common complication, 13%, but it is the leading cause of graft failure from biliary duct necrosis. The bile ducts depend on the hepatic artery for their integrity. And when you lose the bile ducts, you end up in infarction, stenosis, abscess formation, et cetera. And the thing that's great about ultrasound is even in asymptomatic patients, we can diagnose this complication and up to 10% of the time, if we screen these patients aggressively postoperatively. And the way you take care of the hepatic artery is basically arterial revision or retrans plantation.

So here you can see actually a normal hepatic artery with a rapid up stroke and a good diastolic flow. And you can see it easily here on color as well without the hepatic artery. As I said, the bile ducts don't do well and it is a disastrous complication. So here we can see these echogenic foci along the portal region, which is basically actually in the bile ducts. This is areas of necrotic debris. You can see here on ct, again, this low attenuation lesion is all debris and dilatation here of the bile ducts. And again, PTC shows you these very, very irregular and destroyed bile ducts. And this patient will go on unfortunately to have many abscesses and live not a very unhappy life trying to take care of his biliary problems.

Diagnosis of Hepatic Artery Complications

So how do we make the diagnosis and with the hope of maybe avoiding these complications? The best criteria that we have in the literature actually is to use something called the resistive index and or the acceleration time. And the resistive index, as you may remember, is the peak systolic velocity minus the end diastolic velocity, all divided by the systolic velocity. And the idea is that you want to have your diastolic flow normally should be about less than a little less than half of your systolic peak. Okay, so this is obviously abnormal. There's a lot of diastolic flow here. This is a low resistive index. And you can see here again, the echogenic regions along the porta in this patient who also already has sym biliary necrosis from hepatic artery stenosis.

Problems with this diagnosis, if you have very small hepatic arteries after surgery, you may not be able to see them. You can also have reperfusion injury and shunting, and which will give you high velocities. And but it will give you a normal acceleration time. And we'll take a look at that in just a minute. False negatives, rapid collaterals.

So here's a patient who has rising liver functions post-op transplant. Here's the day one, the normal hepatic artery, and on day three, this is always a bad sign when you see something that says hepatic artery attempt. So you should be able to easily see the hepatic arteries. If you can't, that's a problem. And you can see here when we do our spectral tracing, there is absolutely no flow in the main hepatic artery. So this is an obvious case of hepatic artery thrombosis.

What about this patient? Abnormal liver functions, hepatic artery stenosis. You can see the resistive index here is low 0.33. You can see the sort of tardis pars waveform. It's only in the right hepatic artery. The main hepatic artery still has a relatively okay upstroke resistive index is 0.58, which would be normal. And when we do the angio, which is the way we actually make the diagnosis, you can see here this patient also has an occluded hepatic artery. There's no flow here. And what we're seeing on our ultrasound is collateral flow. This patient, again, PTC shows intense or very, very profound biliary necrosis.

Another patient for you. Another question, questionable case. These are all from our day-to-day practice. This patient has a very low resistive index 0.35, but you can see here the upstroke actually is relatively well preserved, and this is the first day postoperative. And the end diastolic flow is quite high. Velocity again, in the vessels is quite good. And by postoperative day one, this actually resistive index has returned to normal. Again, good upstroke here, good diastolic flow. But now the normal ratio, and this is basically from what we think is from intraparenchymal shunting, as the liver is e equilibrating to its new home and its new location, this patient is a living related donor who had a right lobe allograft and he had hepatic artery thrombosis. But you can see here the resistive indices actually were normal resistive index here, 0.61. But because of his abnormal liver functions, he did go on to the angiogram, which shows basically these very small collaterals that are filling the hepatic artery.

One question that we have. We see a lot of patients with very strange waveforms early on, and we wanted to know if any of these in particular were predictive of hepatic artery thrombosis. So we looked at our patient population and we divided them into these four waveform patterns that you can see here. The normals actually we found about 37%, 38% are normal, which you can see here. This is a tus parvus wave form or res low resistive index wave form. And you can see that it was about 17% a resistive index of one, which you can see here, very high upstroke, but very, very little and diastolic flow. We saw in about a third of patients and then an absent hepatic artery in about 13%. And the question was, which of these is predictive or carries a risk factor for hepatic artery stenosis or thrombosis?

So one of the questions, since we had a lot of these patients with this high resistive index, we were worried were we delivering enough blood to the bile ducts and are we gonna actually end up with long-term problems? And we also wanted to know, does this predict hepatic artery thrombosis? And there'd been some suggestion in the literature that this actually could do that. This patient actually did have a high rri and on follow-up about six weeks later, there was no hepatic artery in an abnormal liver. However, most of the time when we looked at that waveform pattern, the predictive value of that for patients developing hepatic artery thrombosis was really the same as if the patients have normal waveforms. So since then we have, we don't rush our patients to angio. If they have a resistive index of one, we just wait and watch. They usually return to normal and there really is no increased incidence of hepatic artery thrombosis.

So just a couple of examples. Here's a patient with a very high resistance waveform. By day three, it's normal. This patient took two weeks for it to come back to normal, but eventually it did.

What do you do if you can't see the hepatic artery because that is a risk factor for hepatic artery thrombosis. Do you re-scan them? Do you give them contrast enhanced ultrasound? Do you send them all to the an to angiography or do they go immediately to the operating room? All of these have been suggested and have been practiced in various places.

So this is what we decided to do in our institution. So if we have transient non visualization of one or more hepatic arteries, we will re-scan in less than 24 hour, we will re-scan. And if we did that and all hepatic arteries were present, then there was very minimally increased risk of hepatic artery thrombosis. If one or more remains non visualized, the risk was 71%. So in this case, if there's a persistent non visualized, those are the patients that we would send down for angiography.

The other option that you can do if you're gonna use contrast off-label, this is a great article from the literature, you can actually use ultrasound contrast. These are eight patients who had no flow on color Doppler ultrasound. This is from a university of ca Southern California. But six of them were proven to have flow on contrast enhanced ultrasound. And that was actually confirmed either with angio or a normal follow-up ultrasound. Two of them persistently had no flow with contrast and they actually were confirmed on angiography. So the ultrasound sensitivity rose up to basically one. It was pretty much perfect. And so this may be a way to avoid waiting if you have contrast available and are willing to use it off-label.

Here's a question for you. Has this patient have hepatic artery stenosis? So what are the things that we see the right hepatic artery? The resistive index is just a little bit under five, it's 0.48, the left is 0.42, but the main is normal. So do we wait the findings in the parenchyma or do we actually follow what's going on in the main? And I will tell you that my criteria is all I need is one of these to be low, and I will go ahead and recommend that this patient probably has hepatic artery stenosis. This patient actually had an a stenosis at the main hepatic artery, but we really didn't see the findings of it until we were further downstream. So this patient went on to balloon angioplasty, but unfortunately you can see that not all angioplasties fix things. This patient six weeks out, the resistive index is falling in both the left and now in the main as well. And you can see they've tried a stent. This is sort of the second repair here, and so we've got a stent across it. But even with that, a month later you can begin to see hypoechoic spaces in the portal regions. And you can see again infarct here in the left lobe as well as periportal necrosis on ct and again, the shaggy biliary necrosis on PTC six months later. This is what happens. These bile ducts just don't recover. And these patients go on to develop myelomas, which become infected with abscesses.

Venous Complications: IVC and Hepatic Veins

Let's turn now to the venous side, the IVC and the hepatic veins. And actually those are rarely a problem in cadaveric transplants, but in the live donors there's a smaller caliber vessel that's gonna be anastomosis. And so that it's increased in this living donor population. These patients can present with abdominal pain, ascites, decreased liver function, and on doppler it's the same diagnosis that you would expect in the hepatic veins, you get a monophasic waveform, which is basically proximal to the obstruction. And a 10 millimeter pressure gradient across the stenosis is considered significant.

So just some examples you can see here. This is a patient with a pig piggyback anastomosis, but you can see that there's a lot of clot actually in the native IVC, and tight area here as the piggyback is coming down onto the recipient IBC may, the hepatic vein proximal to this is got a dampened wave form. And at the IBC junction here is the hepatic vein IBC junction, you can see a high velocity of 176 centimeters per second. This is an early case. Again, this was an unsuspected clot. This patient has just a very small remnant of the IVC, which is patent a fairly large thrombosis. But if the patient's asymptomatic, they may not do anything in this patient. This was just left. They didn't even anticoagulate her and she did well.

Again, here are the hepatic veins. This is in a native, so right hepatic vein, very, very damped. Middle hepatic vein actually is not. So it's a solitary stenosis just in the vein. This was repeated a couple of times, but there were no clinical symptoms and so nothing was done. If the patient is symptomatic and has elevated hepatic wedge pressures, then the patient will go on to a revision. So this patient has a very distended hepatic veins and has, you can see here, proximal to the stenosis has a low velocity at the stenosis, at least three times what it was approximately. And this was actually revised with an IVC revision.

Again, this is a patient who's a living related donor and he had symptoms for three years after his transplant. You can see here the IVC is dilated proximal to the stenosis up here. And nothing really worked. Here's an MR for you just to show you how tight this was. This is above the stenosis. It's opened up a little bit, but here it is just pinhole, just a couple of millimeters. So radiology, we tried from below several times to dilate this, but were unsuccessful and he eventually was su successfully stented from above by the cardiothoracic team.

Portal Vein Complications

The portal vein actually rarely has a complication. As I say here, stenosis is common, but it's usually asymptomatic and really requires no therapy. Thrombosis, fortunately is relatively rare because that would cause a patient to possibly to lose the graft. Hepatic artery to portal vein fistula are common usually from liver biopsy. And what we see there is low resistive indices in a feeding hepatic artery and it arterialized shunt flow that then goes back into an en large portal vein. These usually don't require therapy unless they're symptomatic.

So just some examples on the portal venous side here. This is a patient with portal vein thrombosis. You can, she was initially normal, but at three months she came in. You can see here is the main hepatic artery. And you can see very, very hypoechoic portal vein actually almost an coic. And if you did not have the color, you would not know that this was thrombosis. But you can see here on the ct, obviously you can see the collaterals are again a pacified here and the absence of any contrast in her portal vein. Unfortunately, her liver function deteriorated rapidly. She required retrans plantation because they could not salvage this.

Here's another patient, day one. You can see that we can see a little bit of flow in the portal vein quite a bit of flow here in the hepatic artery. But the rest of this portal vein has echogenic material within it consistent with thrombus. And you can see on day two, they were, they put the patient on anticoagulation, but it didn't open up. And you can see the thrombus actually is extended. Power doppler shows basically the same thing. So this patient actually went to the OR for a thrombectomy because there was concern that this would continue to progress and liver functions were deteriorating. And obviously, thrombectomy worked main portal vein is now open and normal.

So one question that a lot of people have is what about the anastomosis of the portal vein and what happens if you don't have a good fit? And is this significant? So this has been defined basically by angiography as an eight millimeter gradient of pressure gradient across this anastomosis. The literature will tell us that a stenotic velocity of 155 centimeters per second is a good criteria and a three to one ratio between the stenotic segment and the proximal segment will lead a 73% sensitivity for stenosis. But I think the most important thing to remember is that most of these are asymptomatic and they actually resolve spontaneously over time. The anastomosis dilates up and the mismatched areas and sort of smooth out and really need no therapy whatsoever.

So just here's an example. Is this a portal vein stenosis? Well, it looks quite narrow here and you can see some turbulent flow in this post stenotic area dilatation, but the velocity is 120. So, and the ratio was normal, so it's not a three to one ratio and the patient was in asymptomatic, which is the most important. So no therapy was done. Here's another patient, high velocity over 155, and a good ratio here. But three months later, velocity is normal and even though there's still a ratio, patient's asymptomatic and nothing needs to be done on this case.

However, there is a portal vein stenosis, and you can see that velocity is 168 here. And the ratio is five to one. So this patient if they're symptomatic, you would go on and treat. And you can see here on the CT this very, very tight stenosis. Here you can see there's this waist here between the donor and recipient portal vein. So this patient did go on to be revised and who was stented. And you can see now the velocity in the stent. You can see the nice pulsatility in the portal vein and the velocities are now perfectly normal.

Sometimes you encounter waveforms and you just don't know what to do about them. So the question here is, is there portal vein thrombosis? And again, it's always bad when your sonographers say something area that means they can't find it. So if you can't find the main portal vein and you see the artery quite well, what could be the problem is it thrombosis and you need to look through the rest of the liver because here the left portal vein actually is reversed and the right portal vein is also reversed. And if it doesn't, there's reverse flow in the portal veins, it's gotta be getting out somewhere. Can't see it in the main portal. Vein doesn't make sense. So you need to go on and do something else, which means you need probably to get a ct.

So this patient gets a ct. And the cause of the problem, actually you can see this very, very narrowed portal vein right here. And when you get down inferior to the liver, there's a large, large hematoma. So this was a compressed portal vein and there was so much compression that there was basically no flow coming in. And the flow was exiting the liver trying to exit out of the left and right portal veins. So when this went to the or and was evacuated and decompressed, you can see the portal vein, the main portal vein is now well visualized and is normal and you can see it as well on ct, it's normal.

Other Vascular Abnormalities

Again, here's a patient who is asymptomatic but had very low resistive indices. And again, left hepatic artery is a completely normal resistive index, right hepatic artery, very low resistive index. And here is the main hepatic artery, also low. So does this patient need to go to angiography? Is a hepatic artery thrombosis and just involving a couple of vessels or is there something else going on? And you need to look around because one of the causes of a low resistive index is actually excess outflow. And when you look here, there's some blue vessel here that's right next to your hepatic artery. So when you look more carefully, you can see this turbulent flow and this very, very turbulent flow in the portal vein. And this is an hepatic artery to portal vein fistula. So this patient was symptomatic. You can see some clips here from an embolization attempt, which unfortunately didn't work particularly well. So this is a common complication. As I said before, most of them are asymptomatic. We are, they're seen in up to 50% of patients within a week post biopsy if you look for them. But most of them will not persist beyond a week. So 90% will close spontaneously.

The other um, vascular abnormality you should be aware of is hepatic artery, pseudo aneurysm. They're usually extra hepatic. They occur at the anastomosis. We often miss them at ultrasound if we don't look for them carefully. CT angiography is much better at finding them than we are at ultrasound. These are certainly at risk for rupture. And you can see this one here is fairly large. And these need to be embolized with coils or bypassed with a stent.

And then thrombo, there are a couple of unusual things that can happen in transplants. And I wanna tell you about them so that you can be aware of them. First is something called an arterial steel syndrome. We don't really know what causes it, but we think it may be due to excess portal vein flow. The reason that these patients get in trouble is they're not getting enough hepatic arterial flow into the transplant. And if you suspect this really the way to go is arteriography, it shows low flow into the allograft, ultrasound will show, can show a couple of things. It can show high resistive indices with low velocity, or sometimes it can even cause a tardis parvis wave form so they can be variable. As I said before, angiography really will be the way to demonstrate it. And what you see is increased flow to the splenic artery or the gastroduodenal artery. And the way to treat it is to embolize the splenic artery.

So I'm just gonna show you a case of ours. You can see this very abnormal waveform in the left hepatic artery. However, the right hepatic artery has a bizarre waveform as well. It's also tardis parvis. This one's high resistance, this one's low, doesn't make sense, something's going wrong. And you can see that the portal vein flow actually is quite exuberant here. So this patient went on to angiography in part because I think originally we may have suspected that they had a hepatic artery stenosis or thrombosis. And when we inject, nothing goes really over into the right upper quadrant. Everything is going over the spleen. You can see here the spleen is filled out. You can see big splenic vein. Again, not much happening over here and post coiling. Now when we do the our injection, we get good flow into the hepatic artery and all the resistive indices actually return to normal, not immediately, but certainly within a day. They're much better.

The second thing that I wanna mention to you is something called portal vein steel. And this occurs because you have some residual varis that have been not been taken care of at the time of the transplant. And these cases actually were given to me by Mindy Horo. So thank you to Mindy. Portal vein flow in the perioperative period should be normal. And this patient had very poor portal vein flow. So they took the patient back to the or. You can see here day one they've got flow that's very pulsitile and not continuous forward flow. Same thing here on day two. And so they revised the anastomosis at the portal vein, but their intraoperative flow also was very poor. And so what they had to do in this patient actually was look around and there were large splinter renal varis that were shunting the flow from the portal vein. So they ligated the varix and that you can see the varix right here. And that improved the portal vein flow. So now the portal vein flow is what you want it to be. It's nice and continuous forward flow.

And here is just an a reminder that you can see these things on preoperative imaging. So if you're in a transplant center and reading out your cases, you wanna make sure that these surgeons are aware of these 'cause these really should be ligated at the time of transplant if possible to prevent this kind of a problem.

So this is another patient who presented five months post-transplant with a portal vein thrombosis. And this is a classic appearance here you can see the hepatic arteries portal vein relatively hypoechoic, no flow through it and just noise on spectral tracing. You can see here is the clot as the patient's being injected. And that means that of course that there's acute thrombus. And the other thing to recognize here is again, a large coronary varis, which were basically diverting the blood away from the liver and allowing the slow flow in the portal vein. So once they embolize the varice and did a thrombectomy of the portal vein to reestablish flow, you get again a nice normal waveform in the main portal vein.

Case Example: Delayed Detection

So here's my last case for you. This is again, all these are problems that we've encountered in taking care of these patients. So here's a patient day zero liver main hepatic artery. Looks fine. Point six nine, relatively good upstroke. Same thing for the left hepatic artery on day two. The upstroke is still fine over here, but the velocity is dropping. So what would you do? And the question is, does this patient need an angiogram? Do you follow carefully? The resistive index is dropping a little bit in the left hepatic artery as well, but again, that upstroke looks okay right here. So well they elected to follow and unfortunately they didn't follow very closely. And so on day four we have the same bad outcome here. Main hepatic artery area, that means we're getting no flow, just noise there. And we can see an infarct here in the right hepatic lobe and the injection shows nothing coming through the jump craft. These are renal arteries. Here's the jump craft as well. So the patient unfortunately required re-transplant because this could not be thrown ized.

Conclusion

So in conclusion, doppler ultrasound is central to the management of hepatic transplants. It's critical for the diagnosis of arterial and venous thrombosis and stenosis as well as abnormal flow patterns. And it will identify post interventional sequela including AV fistula and pseudo aneurysms. And I encourage you to use it and to be confident with it. And if you don't know what's going on, it's good to sometimes use a CT and go on to angiography. And with that I wanna say thank you. And it's been a pleasure working with you.