Liver Tumors: The Fundamentals of Interpretation with CEUS - HD

Fundamentals of Interpretation with CEUS

Thank you very much.

Today I've entitled my presentation the Fundamentals of Interpretation with CEUS.

I have some disclosures, equipment support, and a research grant from Lathes Medical Imaging.

The approval indication for contrast enhanced ultrasound is liver mass characterization. And this has been the worldwide approval indication. And this answers the question, what is it? Additionally, there is another question about liver mass detection, which answers the question, is it there? And of course we're going to focus on the first of these, although it is inevitable that we will improve our detection of liver masses as well.

The algorithms for diagnosis of focal liver masses on CEUS are similar to those for the non-invasive diagnosis of liver masses on CT and MR scan. And these are based on enhancement characteristics in the arterial and the portal venous delayed phase. CEUS has the unique advantage also of temporal resolution and also exquisite vessel morphology.

The objective of this talk is going to be to highlight the fundamental principles for interpretation of liver masses on CEUS. And we're trying to first of all determine which masses are malignant and then to establish their specific diagnoses.

This presentation is very basic, but I selected it because of the timing at the SRU when this microbubbles have been approved for just over a year. And although I hope that this talk is not too elementary for some of you, I hope that many of you are going to start to do CEUS in your departments or have just started. And this is really what you need to know to interpret these masses.

I've called these ultrasound pearls.

Pearl Number One: Determine Malignancy

First of all we want to determine malignancy. And this is largely based on the observation of the lesional enhancement in the portal venous and late phase relative to the liver with the recognition that washout has a strong association with malignant outcome.

If we look at a patient and do a contrast enhanced ultrasound and CT scan, you can see concordance of the imaging in the arterial phase with a hypervascular scar neoplasm. And then if we look into the delayed or late phase, we can see that both of these lesions show washout. And these lesions are concordant on the two imaging modalities and the interpretation regarding the state of malignancy or malignancy would be the same. The washout suggests that these are malignant lesions.

Compare these with an asymptomatic female who has both CT and CEUS as well where we see a uniformly hyper enhancing mass in the arterial phase. And then when we go to the portal venous phase, we can see that there is sustained enhancement so that the lesion is iso vascular on the CT and still remains hypervascular relative to the liver with a non enhancing scar on CEUS. And this sustained enhancement of course suggests that this will be benign and you recognize that this is a classic focal nodular hyperplasia.

Pearl Number Two: Benign Tumors Specific Vascular Patterns

Benign tumors on CEUS are characterized by very specific vascular and filling patterns in the arterial phase. And these are optimally shown on real time dynamic CEUS. And when I start showing these, you'll recognize that many of these do not show these specific vascular patterns on CT or MR scan.

If we look at the algorithm that Peter Burns and I published back in 2006, if I include just these three common benign lesions, you can see that the gray column in the arterial phase has got little pictures of the specific features that we're looking for. And what's important about this? When we do dynamic realtime ultrasound with CEUS, we're looking for these specific patterns.

Now we always go on to image the lesions further because we must confirm what the lesion does for confirmation by showing its enhancement in the portal venous phase. And you can see there that those white circles almost completely covering every lesion. And of course there's one small gray box at the bottom for adenoma. And what's the importance of that? There are always exceptions in all the rules we set for ourselves. And of course with imaging of adenomas on both CT MR and on CEUS, we're familiar with the fact that a good proportion of these lesions will show weak washout. But nonetheless, the fact that these lesions have this specific arterial phase features and this tendency to sustained enhancement really helps us to be exceptionally accurate for characterizing these benign lesions. And what's the importance of this? Of course, when you're looking in your department and you see all the incidental masses that come through, many of them will be benign.

So let's look at these quickly. Hemangiomas, we know they're characterized in the arterial phase by peripheral nodular enhancement, which tends to show centripetal progression of the enhancement in either the arterial or the portal venous phase and have a tendency to sustained enhancement.

Here's a patient. This was a patient, a young person who came in with a massive pulmonary embolism on the CT scan performed for the pulmonary embolism. A large mass was incidentally noted in the liver. The clinicians had instantly thought this patient might have malignancy with clotting problems. And asked us if we could confirm the nature of this mass. And to their surprise, we show at eight seconds a peripheral rim of enhancement at 20 seconds, centripetal progression of peripheral puddles and at one minute virtually complete enhancement of the mass, a classic hemangioma which showed sustained enhancement.

Now here a patient who's at risk for hepatocellular carcinoma, a female with HBV, you can see a mass, a hypoechoic, ring shadowing looking lesion. On the baseline you can see our images at 11, 20 and 20 seconds. And then at four minutes a classic hemangioma and here this deep lesion showing these peripheral puddles with the centripetal progression of the enhancement towards the center of the mass.

The advantages of CEUS are really exemplified by this other at risk patient where first of all, a CT scan shows a small mass which is hyper enhancing with no washout on CT. So in a at risk patient this remains an indeterminate lesion. And on MR scan we show the same hyper enhancing mass with no washout. So this patient comes for CEUS and we confirm that the liver is cirrhotic. We show a focal mass lesion in segment two corresponding with the CT and abnormalities, we confirm always the cirrhosis with elastography. And there when we're using bubble tracking technique, we can see that this is definitely a lesion which has peripheral puddles and rapidly fills to the center. And confirm unquestionably that this is a flash filling hemangioma and we back up within that clip and can extract these really exquisite images showing the enhancement pattern of this benign lesion.

Now what about this patient? Another one, with an oncology history. Recently diagnosed brain tumor has a CT scan which shows an unusual exophytic mass which was unexpected and the patient was sent for ultrasound. So this mass is quite indeterminate on the baseline ultrasound as well. However, when we do contrast enhanced ultrasound, you can see the beautiful peripheral puddles of enhancement which are progressing centripetally. This is the one lesion where we tend to move around a bit to enhance our capability to show the peripheral puddles and we can again extract these excellent images showing those peripheral puddles progressing towards centripetal and ultimately complete enhancement, a classic hemangioma.

Focal nodular hyperplasia was nicely touched on by Deb Rubens. Now here we have this stellate vascularity with the centrifugal filling and of course we were looking for the portal venous phase sustained enhancement often with that unenhanced scar. Now have I ever seen an HCC with stellate vascularity? You bet you I have. And this portal venous imaging, which is only available when we do contrast injection is essential.

Here's a 29-year-old asymptomatic woman with an unusual exophytic mass lesion isoechoic to the liver. When we do our contrast injection, as I've said to you, we're looking always for that specific enhancement pattern in the arterial phase. And here with the bubble tracking techniques which are available on virtually all equipment today, we can see those beautiful stellate vessels radiating from the center to the periphery giving us an appearance that the mass becomes larger with time. Again, we can extract these exquisite images from the contrast clip.

Pearl Number Three: Sustained Enhancement for Benign Tumors

Benign tumors generally show sustained enhancement in the portal venous and late phase. Why does this have to be its own pearl? Because of the exception. The exception to the sustained enhancement is for hepatic adenomas and we know that a certain small but definite percentage will show weak and late washout. And these lesions are hypervascular but it is our impression that they're much less hypervascular than FNH. They can have a diffuse or chaotic vascular pattern, but if they do show a pattern it will have centripetal filling in the portal venous phase, they can be variable and up to one half of lesions in different series that we've done have shown washout.

Here we have a 25-year-old female with six months of right upper quadrant pain with an indeterminate CT and MR scan. We show a fatty liver with a well-defined hypoechoic mass. When we do the contrast injection you can see that the mass is hypervascular but not nice and homogeneous like that FNH but still a hypervascular mass. And when we look at the images that we take over the course of the examination at three minutes, there is no doubt that this lesion is washing out. So because of our rule about washout recognizing even that this patient may be at risk for adenoma, we would recommend biopsy or referral for surgical consult.

Let's just look at these beautiful pictures on three different patients showing benign tumors. The upper right lesion is obviously the easiest of them all where we can see that bubble tracking technique showing the stellate vascularity and the centrifugal filling the lesion on the bottom with the peripheral puddles proceeding to the center of the lesion and the lesion on the bottom left showing a centripetal filling mass lesion from the periphery to the center with a more chaotic vascular pattern.

Pearl Number Four: Arterial Phase Enhancement for Metastatic Lesions

Arterial phase enhancement features, which we have just been talking about so specific for benign lesions do not really distinguish metastatic lesions at all. And what is the feature that we're really looking for? We're looking for washout that is most often rapid and marked as the hallmark feature of metastatic tumors.

When we look at our schematic again, there's our metastatic lesions and you can see our gray bar including the appearances that can occur in the arterial phase. These lesions can really have any pattern. They can be rim enhancing hypervascular or they can be hypovascular in the arterial phase and it's not their appearance in the arterial phase which is helpful, it's their rapid washout leading to a very black hole which is often present in the liver even before one minute.

Let's look at three different patients in the arterial phase. All of these patients have metastatic lesions, they bear no similarity to each other and of course there's even many other patterns. The patterns are non-discriminatory and highly variable as above.

Now let's look at a very good patient. This was a patient with a remote history of breast cancer who came to the emergency with flank pain. And while she was having a CT scan, it was noticed that she had a mass in her liver which wasn't characterized. When we look at this patient, this case is a beautiful example of another ultrasound rule. We don't want to do her contrast on the supine position because of the depth of that lesion from the transducer surface. But on the right image we can see that when we place the patient in left lateral decubitus position and come intercostal that we can see this lesion only half the depth of its original space. So that is where we will position the patient to do the contrast injection.

When we inject the contrast, you can see instantly the mass is hypervascular instantly even before there's any contrast in the liver. And now as the liver is enhancing normally 26, 27 seconds, the liver's now lovely enhanced and you can see that the mass has already completely washed out. And the transient hyper vascularity is not evident. And this explains of course the discord that shown in the arterial phase often for metastatic lesions.

When we look at this patient, we see the baseline image on the left, the central image is the arterial phase uniformly hyper enhanced mass. And by well under one minute the lesion has already become a black hole in the liver. Very confidently we can suggest that this is likely to be a malignant lesion or very suspicious and recommend a biopsy. And this biopsy shows unsuspected metastatic carcinoid tumor.

Pearl Number Five and Six: Detection of Metastases

Contrast enhanced ultrasound in the portal venous phase obviously will improve the detection of metastases over baseline in that it improves the conspicuity of the lesions and therefore we will see both more and smaller lesions on CEUS than we will see on baseline ultrasound.

In a study that Peter Burns and I participated in with the Europeans way back around 2000, this is the publication which I cite in this publication we had several hundred patients and we had results that were equivalent to CT scan and superior in some circumstances in looking at metastasis in patients with CEUS.

Pearl number six follows this, we always try to perform portal venous phase sweeping of the liver and what are we doing? We're looking for washout lesions. This is the optimal time and the technique for detection of liver metastases.

This is my most shocking example ever and I've had this for a long time. Some of you will have seen it before. And the reason I still use this same case is because this is the best case ever and I don't see this every day, this is infrequent but I show it to you for impact.

This patient came to our emergency department with abdominal pain and one of my colleagues called this liver normal on ultrasound. And when I look at this picture I think, that's a pretty good call. It looks pretty normal to me too. What happened was the patient was taken to the operating room for emergency surgery because they thought his abdominal pain was related to a leaking aortic aneurysm. And it was the anesthetist who phoned me and said, Stephanie, can you come with your machine and those bubbles to the operating room? I would like you to look at this liver before I anesthetize this patient.

We trundled off to the operating room and there's the injection of contrast and you can see that this liver has got multiple lesions in the liver and when we sweep in the portal venous phase, you can see the tremendously increased conspicuity. I showed that for effect. Thank heavens this does not happen daily, weekly, or even monthly but it can happen.

The timing of washout is discriminatory. Metastases as we've shown wash out very fast often within the timeframe of the arterial phase and they tend to wash out completely.

Let's look at a patient with breast cancer. We can see on the baseline scan a very hypoechoic mass in the liver. The low MI image interestingly for those of you doing contrast, often times we show additional lesions as here where we can see different lesions with just with pulse inversion imaging. But the first images that I will show you here show that dominant mass with increasing enhancement, so both the liver and the lesion are increasing in enhancement up to around 22 seconds. However, from here we now start to have decline. We have to judge washout relative to the level of enhancement that the lesion acquired in the arterial phase. But this lesion washes out and of course in the portal venous phase when we would be sweeping the liver you can see that the conspicuity would be greatest and we would detect the largest numbers of lesions.

This was a case that was in Peter Burns and my very first series of patients that we scanned when we looked at 176 patients with contrast ultrasound compared to CT or MR scan. All patients had an unenhanced scan and then a contrast enhanced scan. When you look at these two baseline images, they're both quite horrid actually. But you can see that the ability of ultrasound to resolve lesions is obviously far superior to CT scan which depends on contrast to increase contrast conspicuity in the liver.

I'd like to suggest to you that nobody would interpret that CT scan with contrast and certainly I would like to suggest to you also that nobody should interpret that ultrasound scan without contrast either. And I think Deb Rubens was really wonderful to us in her presentation. She made that point over and over that contrast really helps.

There's this patient on their baseline scan and then here they are at the portal venous phase showing this peak of difference between the multiple metastatic tumors that really riddle this liver and the background parenchyma. And my conclusion from this many, many years ago was really that both CT and MR and ultrasound require contrast agents for detection and characterization of liver masses.

Pearl Number Seven: Timing and Intensity of Washout

Malignant tumors show washout, we've been discussing that. And the timing and the intensity of the washout are discriminatory on CEUS. On MR and CT scan we tend to document the presence of washout and on CEUS we want to document the presence of washout its timing and its intensity.

What are our objectives? We know that hepatocellular carcinoma is characterized by weak and late washout and that non hepatocellular malignancy by comparison has early and marked washout as we've just been looking at on those metastatic lesions. If we make a two by two table, we can see that the late and weak washout we were thinking of HCC whereas the early and marked washout we're thinking of metastatic lesions and other non hepatocellular malignancy. And what do we do with the lesions that come up in the other squares where we're gonna characterize them as suspicious for malignancy requiring further evaluation.

Let's look at a patient who came to our emergency department with no risk factors that were known and he had a mass found on his ultrasound that's hypoechoic concerning looking. We do contrast at that time and you can see that we're showing a hypervascular mass. As we've already said, we don't know for sure what this mass is. A hypervascular mass is characteristic of many different things when we see it in the arterial phase. What do we do? We continue to look at the mass.

At one minute this is a critical picture, this is the time difference between rapid and more and late washout. At one minute the lesion is invisible so it's isovascular and at one and a half minutes it's still invisible. We don't see it so it's isovascular. Do we make the interpretation that the patient has sustained enhancement? Yes they have sustained enhancement for 90 seconds but what does this actually mean? To avoid missing the diagnosis of hepatocellular carcinoma, we have to watch for a longer time and with the contrast agent that I'm using, it's five minutes for sure.

Here we're still looking, you can see just before four minutes is there vague perhaps early washout on this lesion And at four and a half minutes it's unquestionably there and you can see that this is not a black punched out lesion rather there are multiple bubbles which remain within the lesion. This is weak washout at over four minutes. What is this? This is a classic hepatocellular carcinoma and this is confirmed in this patient by biopsy.

Now what about this patient? A 73-year-old male with a non-cirrhotic liver but he's got hepatitis B viral carrier status. We can see this lovely mass in his liver when we do his contrast injection. You can appreciate that this robust technique lets us see the enhancement regardless of his respiration. We confirm easily a hypervascular mass at the peak of the arterial phase, a hypervascular mass at one minute, sustained enhancement at two minutes sustained enhancement but at three minutes our intermittent scans to preserve the bubble population allows us to show weak late washout a classic HCC.

Pearl Number Eight: Rim Enhancement for Non-HCC Malignancy

Non hepatocellular malignancy often shows rim enhancement and rapid and marked washout similar to metastatic disease. Here's an 82-year-old female not with risk factors. She has mid abdominal pain and a very large mass in her liver. We can see surface retraction with much greater difficulty on ultrasound than on CT MR scan but well shown here. And when we do the contrast enhanced injection, you can see that we're looking at a rim enhancing lesion with a relatively hypovascular center. Definitely big rim on this lesion. And when we look at our subsequent imaging by 47 seconds we can see that that rim is disappearing. There's rapid washout which becomes more marked by one and a half minutes.

I'm looking at the later images, even more striking marked washout and you can see multiple satellite lesions on our sweeps through her liver. What do we suggest? Non hepatocellular malignancy and the patient has biopsy with peripheral cholangiocarcinoma.

Pearl Number Nine: Microbubble Contrast Agents Purely Intravascular

Microbubble contrast agents are purely intravascular. The contrast agent for CT and MR scan by comparison has a well-recognized interstitial phase, which is especially evident in malignant tumors with permeable endothelium and fibrous center. The result is that washout is correctly shown in the portal venous and late phase on CEUS for a malignant lesion whereas CT MR scan can show sustained or even increasing enhancement in the portal venous and late phase. And when does this happen? Especially in non hepatocellular malignancy and in our at risk population we're always suspicious for cholangiocarcinoma.

Let's look at such a patient. This is a young man who has a mass in his liver, which is heterogeneous and enhancing in the arterial phase on CT with increased enhancement in the portal venous phase, indeterminate result. This patient goes on to have MR scan which shows the same observations, so an indeterminate CT and MR scan when the patient comes for ultrasound, the arterial phase is identical with heterogeneous hyper enhancement and in the portal venous phase complete rapid washout, so malignant on CEUS and biopsy in this 29-year-old male shows cholangio hepatoma.

Why do we have this discordance with ultrasound showing washout and CT and MR showing increased enhancement amount and intensity. This is related to the purely intravascular bubbles on CEUS shown at the top and related to interstitial leakage showing pseudo enhancement on CT and MR scan.

Conclusion

The ultrasound pearls that I've shown you are all reproduced in my syllabus and if you're interested they're there and I'm just going through them right here. In the interest of time I won't repeat them, but I believe if you understand all of those things you'll do exemplary. Peter Burns and I made a mistake on everything when we were starting kind of trying to wrap out all of these rules and I hope that my sharing them with you will be helpful.

CEUS is obviously an excellent modality for characterization and detection of focal liver masses and although it has excellent agreement with CT and MR scan, the instances when there is discordance show the advantages of purely intravascular microbubble contrast agents.

Thank you for your attention.