HCC with arterioportal Shunt|TACE |68|Male
HCC with arterioportal Shunt|TACE |68|Male
Transcript
Related Clips

This is a good case of mine,

hes' a 68 year old male. He has alcohol cirrhosis and esophageal varices, he's not a prisoner. But he doesn't speak English, so just putting that out there. He's a child Pugh A5, he had a total bilirubin of 0.7 his AFP was 43

so not indicative, and his ECOG was 0-1 when he saw me for the first time in clinic and this is what his initial CT scan showed, to from my clinic, he had not been followed up appropriately and just came in with a belly ache and they noted this very large mast in his right hepatic lobe.

Further down you can see that the tumor extends more inferiority and you are getting a very avid enhancement of the portal vein and you can see there's some portal vein thrombosis in there. So the arrow is really pointing at what I was reading as an arterial portal shunt. So knowing that this is what you are going to face ahead of time

sort of can help you plan. So arterial portal shunting diagnosis can very often be made on multiphasic CT scans so you know I'm sure you all get those patients that come from an outside hospital with HCC diagnosis and they have one phase of imaging so it's really important to get that multi phase imaging

CT or MRI because you can make these diagnosis. So this group showed that about 15% of patients can be or actually have this and the diagnosis can be made on CT scan and then they just talked about where these arterial portal shunts are about half of them essential or 24 from a peripheral and then 22 of them are mix. So you can't really tell if they're central or peripheral on the CT

imaging. And then how severe are they, 35% are severe. Severe means that on that arterial phase imaging your portal vein is as bright as your aorta. And 41 or about moderate which is you know.

You can imagine what moderate is and then 24 are mild so you just saying it's within the peripheral areas. Interestingly about half of these patients that do have arterial portal shunting also have portal vein thrombus so even if you just have single phase image and you have portal vein thrombus you can sort of start thinking in your head that these patients may have an arterial portal shunt and they actually

went on to get hepatic and angiography, 22 of these patients went on to get hepatic and angiography and of those patients 86% had arterial portal shunting the author said that they probably all had it they just missed it on angiography. So what's the prevalence really there is not a lot of great data on there you see I went all the way back to 97 to see if I could

actually get a prevalence but overall arterial venous shunting. They said it was about 31%. If you had arterial portal shunting or arterial venous shunting of that, 31% of patients that HCC 92% had arterial portal shunting. So what's the treatment?

So as you can imagine we've all come up with creative things we've been at this meeting all week. And, you go into a room and you can imagine everyone's doing these different things. So, if you can find a single feeding vessel, a coil is a great choice

if you can find it. Glue, if you have a network of vessels and you don't think that glue is gonna get stuck proximally to prevent you from treating the tumor eventually. And particle embolization is a good option. You can use large particles if they don't cross over into the shunting

I had a patient that had a large, we'll get to them in a minute, but it was a very bad complication from outside the hospital. And the other thing is you can try lipidol based embolization and that's not well supported in literature but there are case reports talking about it.

So for my treatment planning, because he had portal vein thrombosis I decided that I was gonna go ahead and do a 190 based on the literature. And so, he showed up to the suites and we had planned for the shunt study. So the initial angiography you can see,

as soon as I'm seeing the arterial phase of imaging. And this is seconds after the injector goes. you can see filling that portal vein, the main portal vein.

And you can see, the filling defect within the main portal vein indicating that he had portal vein thrombus. So, this severe arterial portal shunting was noted. I couldn't find a single vessel, it's just this huge network of vessels that was supplying or communicating with

the portal vein I didn't think I could do glue because I was worried that the glue would embolize the proximal artery and I'll never be able to go back in treat it, and because the tumor was in the right hepatic lobe I just wanted to see is there, I have

had a pulmonary shunt as well as the ulterior portal shunt and so I went ahead and infused the MIA. The lung shunt fraction as is very typical was 20%. And so that brings us to hepatopulmonary shunting. How common is this?

Is it a prevalence of overall again arterial venous shunting is 31% and of those hepatopulmonary shunting is much less so it's 8% of those patients. And all patients with HCC they say the prevalence is about 2.4%. Again in my clinic it's probably much higher than that. So the typical shunting in HCC this is Ron Gaver's/g group who talked

about what's the prevalence based on lung shunt fractions they said about 56% of patients are gonna have no lung shunt fractions to less than 10%. 10 to 20% of lung shunt fraction you got 30% of patients with HCC and those higher lung shunt fractions are going to be 14% of your patients that you are treating.

So they also found that there is correlations how can you sort of predict this on imaging sometimes you can't see this hepatic pulmonary shunts on imaging so if you do have an infiltrate of tumor that was correlated with hepatic pulmonary shunting if you had greater than 50% hepatic tumor burden, if you had portal vein thrombosis or

portal vein compression so the tumor is actually pushing on the portal vein you are going to see a higher degree of hepatic pulmonary shunting. If you have arterial portal shunting that's interesting that also correlated with the hepatopulmonary shunting. And if your tumors are hypervascular,

all tumors are hypervascular so that's not gonna be very helpful again how do you treat this? So embolization again if you can see that single feeding vessel that's fantastic and you can do it again glue is possible particle and hepato-based embolization is also then reported. Systemic therapy with Sorafenib has some literature saying that

if you start patients on Sorafenib you can reduce the shunting, this is the case serious with couple of patients, that they showed that the mean shunt fraction in the patients was 26.5 prior to starting Sorafenib therapy, and post sorafenib therapy, went down

to 7.5. But again the time it takes to treat patients with Sorafenib to get those shunts to come down sometimes can be longer than the patient would survive otherwise. And what about if we do the Y90 what if we,

go ahead and do Y90 therapy is there real risk of radiation pneumonitis? So this is out rehabs group and he found that of his 403 patients now these are patients not only with a HCC but all comers 58 of those patients or 14% actually received greater than three gray accumulative lung dose and of those just so I wanted to highlight that HCC population

74% of those patients with those elevated or high lung shunt fractions were actually patients with HCC with sort of a similar to the data that I showed you earlier. And the mean shunt fraction in the HCC patients was 20% and the mean accumulative

of lung dose is 54 Gy to those patients and his report he had no cases of imaging or clinical real issue of pneumonitis. And so you can quote that to your patients that you know a very large series has shown that there's no risk of having any risk issues with the lungs other patients or other series have sort of talked about may be having radiation pneumonitis.

So really the things to consider when I was treating this patient are you have severe arterial portal shunting on the angiogram because we're seeing filling of that main portal vein and now he's also has a hepatic pulmonary shunt which is almost 20% and so do I go on to do Y90? With an arterial portal shunt you have to really think about where

the beads are gonna go so if they cross over into the main portal vein, where are they gonna lodge? So they are gonna lodge any where in the liver and we have started to do or what I've started to do is if you see these large arterial portal shunts you can do CT spect with your MAA and then you can see where the bead is gonna go because that really is a good indicator.

I had a patient a couple of weeks ago that we did that and it actually shunted to the contralateral lobe, so I felt why don't we just treat the whole liver except all of the beads would have got the contralateral lobe not the lobe where the tumor was. So I would have had an effective dose and I would have just cause atrophy of the wrong lobe of the liver.

So that sort of a trick you can think about. With hepatic pulmonary shunting you have to think about can I get a dose high enough to effectively treat the liver without having it just bypass around and go into lungs. And so if that's the case, do I have to worry about this cummulative/g

lung dose? Do I, do I not or can I do some sort of fancy technique in the end of my hepatic vein and do an occlusion during my Y90 administration to get those beads to lodge in, take out the balloon afterwards get an effective treatment.

And really, is it affectatious, do I go ahead and treat and is this gonna affect these patients, appropriately? So again, here's the image this huge tumor. Here's the shunt.

Here's my hepatopulmonary shunt. So, I'll bring my first team question up. [BLANK_AUDIO] So given the clinical and imaging characteristics below, how would you treat the tumor?

Anyone change the words from a monograph. So a, systemic chemotherapy, percutaneous ablation, c, full dose radioembolization d, conventional chemoembolization or e, reduced dose radioembolization.

[BLANK_AUDIO] Good answers. I love it. [LAUGH] I don't think there's a right answer. I chose the answer D,

conventional chemoembolization, and if we can go back to my slide. I said it was incorrect. You can do Sorafenib and I don't think that's the wrong answer it's just a matter of how long are you willing to wait and how often are you gonna re-image these patients to try to shut down their

shunts again. The studies showed that their follow up was up to 270 days, that's almost a year that the patients need to out to be able to get these shunts to shut down. You're never gonna ablate a tumor like this.

It's too big. C. I said is incorrect because the Y90 is gonna disperse throughout the entire liver unless of course you do that spect imaging and then you can sort of determine that it's really gonna go to where you want it to go. So I said I probably wouldn't just go ahead and treat the whole

liver if the patient's is cirrhotic. In the patient that has liver metastasis since they may be willing to or able to tolerate it just depending on how bad your patient's overall liver function is. I thought D was the correct choice,

patient is candidate for conventional chemoembolization. Cuz it really provides treatment while potentially reducing the shunts and allowing the future radioembolization. The other thing I like about lipidal based chemoembolization is you can see it.

So the shunts crosses over, you can see where it's going, so you know if it's going only to the left at the time we can see if it's going only to the right if you're not doing your pre Y90 ahead of time instead of beads where you sort of can't see where they're going and don't really have a good handle on it at the time.

And then C, they've talked about in a literature doing a load of stuff that guys at Stanford published trying to decrease your dosing when you do Y90 and they actually show that it had decreased efficacy of treatment. So that's probably not the best choice according to the literature that we have now.

So I went ahead and I proceeded with TACE I did the lipiodol based TACE. We don't have any support in Wisconsin I think that's probably universal in the United States these days. So I used Doxorubicin and mitomycin I mix my lipiodol thicker so it would get stuck into those blood

vessels so it would trap it so it wouldn't cross over into the portal veins so I did a two to one and sometimes I'll even do a three to one just to get that lipiodol thicken and sludging in instead of just crossing over into the portal vein,

and now embolized into the back end with 3 to 5 PVA. So this is what my angio looked like at the end. It was sort of shocking that we had shut down that shunt with the lipiodol based TACE. We always get CT scans the day following.

I would show you the picture but really, you can't appreciate where the lipiodol is because it's so defused throughout the entire liver, I thought this is never gonna work. I called the medical oncologist and I said can we start the Sorafenib?

And I told the patient and his family I didn't have very good hope that this was going to work but we were gonna try to take care of him. On follow up imaging about a month later, this is what it looked like so you can see that that huge mass is gone,

I thought I had the wrong patient. Study pulled up and then, did you scan the wrong patient? I looked at the other anatomical structures, those were his kidneys,

no, this is this guy. It really melted away that tumor. You can see some residual lipiodol there and he did have a residual viable tumor that you could see. His portal vein thrombus had melted away so that was no longer there and on the arterial phase imaging there was no longer any appreciable

or arterial coral shunting. So we decided why don't we go ahead and repeat the shunt study and see what's happened to the hepatopulmonary shunt and see if that was also effectively treated. This is the second study. I took these pictures and I told my fellow this was the most traumatic

case that I had ever seen in my life and everyone should go back and look at the pre-images because I've never seen this that you have zero arterial portal shunting after one session of TACE, is a very dramatic change in his imaging. I did a CT spin just to

see if I could figure out the distribution of was there anything going to that left hepatic lobe you can see this is residual viable tumor enhancing here and nothing is going to that left hepatic lobe again we did hepato pulmonary, a lung shunt fraction calculations and

it came down to 7.4%. We went on to do a Y90 radio embolization of the right hepatic artery and on seven month follow up the patient is disease free, he has no residual viable tumor he has no portal vein thrombus and he tells me that he prayed and I told him to tell me to whom he prayed

cuz I wanna give that to all of my other patients.

So what are the focal therapies we could use?

Well surgery's been used for years, and there's certainly clinical evidence for this in a number of scenarios. Pts who have colorectal metastases to the liver. They undergo a partial liver resection and they live longer. They are long-term survivors from that.

Same thing with resection of lung metastases, even adrenal metastases. Radiation therapy is certainly used for this in certain areas, particularly of the spine. Embolization is certainly used as a local regional therapy

for metastatic disease, particularly into the liver, and currently it's being used in patients who have more than oligometastasis, several metastases. Focus ultrasound is being used, it's really in the experimental stage now for actually developing local control,

not just in the uterine fibroid here or benign tumor, but in bone metastases.

But I'm really gonna focus on percutaneous ablation because it's particularly well suited to this application, minimally invasive for these potentially frail and elderly patients,

as well as high kill rate with tumors of many different histologies. So when we're choosing, this is the technique, so how do we do it? If we were facing a metastasis in the scapula like this, we can treat it with heat,

radiofrequency, or microwave ablation, or we can treat it with cryoablation, extreme cold temperatures, extreme cold or extreme heat, they'll both kill the tumor. How do we decide? Well, if we compare cryoablation versus microwave ablation

or radiofrequency ablation, ease of use, the heat-based therapies are certainly easier to use. They're generally faster, so the procedure duration is quite a bit shorter, but the energy transmission into bone is better with cryoablation.

It'll go through the cortex, whereas heat is limited in that regard. The predictability of the ablation zone, the cryoablation. As you can see in that scapular picture, we can actually see the edge of the ablation with several different modalities, CTMR and ultrasound.

Our ability to monitor that ablation then and prevent it from escaping into adjacent collateral structures. And then the ablation zone size, we can usually treat a larger area with cryoablation, and patient tolerance, their pain scores are generally less

after a cryoablation than a heat-based therapy. So in general, most of us who are treating for local tumor control would use cryoablation. These factors are a little less true these days where there are newer bipolar radiofrequency devices that are designed specifically for bone

so have better ability to control tumors within these sites.

This is basically kind of your anatomy in hemodynamics.

Celiac axis gives supply to the spleen, gives supply to the hepatic artery, and there's increased flow to the spleen for some reason. Then flow goes back from the gut and spleen back to the hepatic graft by the portal vein. This is kind of your general circulatory

kind of hemodynamics. What the Japanese for years have been talking about was a small for size graft, a completely different syndrome, called small for size graft. In other words the liver is actually too small

for the amount of portal flow that's coming to it. It's an undersized graft. So for example, a transplant in a kid put in an adult. The liver is too small for the amount of portal flow that's actually coming at it.

And what they found out was that there was slow flow in the hepatic artery as well, and they didn't know how that went about. So for years, the Germans were talking about splenic steal, and the spleen is stealing blood flow, the Japanese were talking about

there's a small size graft and there's increased portal flow, but as a coincidence we're looking at hepatic or slow arterial flow. Then, kind of the Americans came along and actually put it all together for them. This is probably a constellation of syndromes and problems

that are occurring at the same time. One can predominate, one can be the only sole problem, or it could be a group of problems. What happens when you increase flow to the liver by the portal vein, there is actually something

called the HABR, the hepatic artery buffer response. This is a partly-reciprocal, poorly understood relationship between the portal vein and the hepatic artery. If you increase portal flow to your liver right now when you eat, if you take a big meal,

portal flow will increase to your liver and your hepatic artery's gonna slow down, because 20% of flow to the liver is from the artery, and 80% goes from the portal vein. If you eat a lot, your portal vein flow increases, your hepatic artery compensates by dropping down to

maintain that flow to the liver. That's kind of a partly-reciprocal relationship. So when you put a small graft in a patient, with relatively high flow, too much flow for the portal vein, the hepatic artery slows down. Okay?

So this is not just a steal phenomenon, this may also be a response or a reflex response to a high flow situation. The graft- and just to add a little bit more detail to this- the graft could be small, or the graft could be non-compliant and poor.

A poor graft, as well, can do the same thing. So it could be a big graft that's not functioning that well, and/or stiff, that would do the same thing as well. Okay? So increased portal flow, decreased hepatic arterial flow.

There are more videos in this playlist...
Upgrade to an unlimited account to access full playlists & more!