Imaging Management of Incidental Focal Liver Lesions - HD

Introduction

Hi everybody.

I'm just a 20 minute talk
and I think a lot of what I'm about to say has been said.
We'll keep it fairly short before lunch.

I'm Suzanne Ryan. I'm a GI radiologist
and I've been asked to talk about the imaging management
of incidental focal liver lesions.

And we're all aware that the number
of examinations we do increases manifold every year, and
therefore the number of liver lesions
and incidental liver lesions
that we pick up increases manifold per year.

Patient Interaction and Strategy

What do you do when you're scanning the patient
and they're staring at your face
and you say, I think I see a liver lesion.
And then you look a bit close and you think, Hmm,
I definitely see a liver lesion.
And then the patient is at this stage hysterical, wanting
to know what's gonna happen to me, doctor?

You have to have some sort of a, an imaging strategy
or a treatment strategy for these patients
before they leave your ultrasound room so
that they don't go off obviously and commit suicide or,
or worse as they wait for their next test.

What should we call them when we're discussing it
with a patient and we've picked up an incidental liver
lesion, we know that most of these liver lesions are benign
and incidental and not ever going to affect the patient.
You want to have a nice sort of, clear name for them
that isn't too threatening.
And I think focal liver lesion is a good name.
It's, it's neither saying it's benign or malignant
and it's saying it's neither solid nor cystic.
It's nice and kind of a
nebulous thing to say to the patient.

It can be difficult straight away on your gray scale imaging
to differentiate benign from malignant
and some non-cancerous lesions will
have malignant potential.
You have to have a sort of a, a treatment strategy
or a spiel to give the patient before they leave.
But the most important thing to remember is
that's probably not going to be something that's ever going
to cause them for the concern.

When you're scanning the patient
as pro limb was saying, you,
you have the patient in front of you
so you can use their age, what their sex,
whether they have used oral contraceptives,
whether they've chronic liver disease
or recent travel to help you try
and diagnose a lesion and prognosticate.

And the useful thing to remember is
that if it's a small lesion,
it's probably going to be benign.
Less than one centimeter is probably going
to be a benign incidental finding.

American College of Gastroenterology Guidelines

The American College
of Gastroenterology has helpfully given us a handy flow
sheet for these patients.

The management of focal liver lesions is recognized
as a problem both in radiology
and in all the medical imaging specialties.
And breaking it down, if the patient has a known risk factor
for HCC or a known malignancy
or elevated tumor markers, then you're pretty quickly going
to get them on to have a CT
or an MRI scan to evaluate the liver lesion
and to look, look for lesions elsewhere
and characterize it as HCC metastasis or other.

I'm not gonna dwell too much in these
because it's not really under the remit
of following up incidental liver lesions.
They'll go for a scan, to look for a primary
or to characterize if you come across a cyst
as you frequently do
and you're happy that it's a nice
simple cyst and then that's fine.
The patient needs no further follow up.

Simple Cysts

If it looks a little bit complex
or worrying, then you can give some contrast ultrasound on
the day, and make your diagnosis.
There are common lesions you can,
you can tell the patients and send them on their way.
Excuse me. If it looks like this
with no enhancement whatsoever, no vascularity, no walls,
this is a simple cyst
and the patient does never, does not need to be seen again.
And we can give them the diagnosis before they leave.

Complex or Hemorrhagic Cysts

If you're concerned about the matrix of the cyst,
you may go on to give, do an MRI,
which will be lower on T one and high NT two
and may show some enhancements.
And if there's hemorrhage, I think that's
where an MRI can be most useful.
Just allowing you to look at differing signal intensities
according to the different heme components of the lesion.
So just a T one and a T two of a hemorrhagic cyst
and you can see it's high signal intensity
and then we can see some layering within that allows us
to be confident that this is all a hemorrhage within a cyst
with no enhancement on the contrast phase.

Biliary Cystadenomas

What about biliary cyst adenomas?
Well, we can, prognosticate again according to
where the lesion is, if it's got complex features, what,
what age the patient is
and how it appears on our imaging will depend on the
components of the biliary cyst adenoma.

This is just an incidental,
cyst picture up in a 50-year-old lady on ultrasound
but of through transmission.
And she went on
to have a CT scan which should quickly play through.
And you can see that there are two, cystic lesions,
no significant enhancement,
but she does have biliary dilatation.
You may see polypore and enhancing nodules.
They don't necessarily mean malignancy,
but they can make it more likely to be more be malignant.

We can also look with MRI
and this is the same lady with MRI.
And again, it'll be low on T one
and high on TT depend T two depending on the components
of it with some biliary dilatation.

Now we would recommend removing these lesions
and not really following them up,
add infinitum in a perfect world.
But this is an old woman.
She's got a ous hernia, she's got multiple comorbidities,
she's got rectal cancer
and so we've just been following her up on a yearly basis.

Initially we will follow patients up with MRI
and then onto ultrasound.
If they show lesion stability, you can sample the fluid
and some sources advocate sampling the fluid,
but it's not necessarily going to differentiate
between benign and malignant.
And there is a risk of seeding, with
passing the needle through the liver capsule.
And so we would advocate removing biliary cyst adenomas.
If the patient is not, is too old
or unfit, then we will follow up with MRI followed
by ultrasound with when lesion stability is demonstrated.

Hemangiomas

If we pick up a heman, we will give contrast
and if we show the characteristic enhancement character,
whatever signal, we will then go no further.
These patients do not go for CT or MRI scan.
Once we see our peripheral enhancement pattern,
the patient goes no further.
They need no further imaging and no further follow up.

Fatty Infiltration

Fatty infiltration. A very common finding.
There are some characteristics areas where we will see it
and more often than not, it's fatty infiltration,
but occasionally it can have a very mass like effect.

And so if we look at the gray scale of this patient
adjacent to the gallbladder, here's the gallbladder
and here's our, our fatty infiltration
and it looks quite mass like
and we want to send the patient away.
We don't want to see them again, we don't want to have
to do any other further investigations.

So if we give contrast
and here you can see why it's useful to
split the screen because you may lose the lesion with that.
But here's the gallbladder
and we can see that this lesion as a patient breathes in
and out is enhancing the same as the background liver.

So again, this patient has their diagnosis made,
they do not need any further imaging
and they do not need any further follow up.

Okay? Those are the easy lesions
that we can make her diagnosis
and send the patient away on the day.

Solid Lesions

What happens if we see a solid lesion?
We can't be quite so sure we'll have a well
patient in front of us.
We think it probably is a metastasis.

So if we give contrast ultrasound, it will help us
to triage the patient and their follow up.

If we give contrast ultrasound
and it is, shows enhancement on the arterial
and venous phase, it's more likely to be benign.
And if it shows wash out in the venous phase,
as we've heard, it's more likely to be malignant.

So if we're seeing this
hyper vascularity on the arterial phase with wash out,
they will go on for other imaging
strategies and potentially biopsy.
So we're moving them away now from our follow-up field.

And if we give contrast
and it's enhancing on the arterial and venous phase
and we see a scar, it's more likely to be an FNH.
If it doesn't have a scar, it may be an adenoma or an FNH
or other and need further characterization.

So we will then send them down this pathway
where we'll send them off for an MRI.
We know the vascular characteristics,
we don't have a great need to do a CT scan.
We're fairly sure this will be a benign lesion
because it's holding onto contrast on the arterial
and the venous phases.
And so we'll send them for an MRI plus
or minus hepatobiliary agent.

If we see a scar, we're fairly sure
it's going to be an F and h.
If we don't see a scar, we we need
to keep going a little bit
and we will then use our hepatobiliary contrast agent.
And if on the delayed phase it's holding onto it,
it is usually going to be an F and H.
And if it's not holding onto it, it is usually going
to be an adenoma.

We won't always see a scar in F
and H, generally if they're bigger, we will see a scar.
If it's smaller, we may not see a scar.
It doesn't mean it isn't one, it just means we need
to follow it up a bit more closely.
For the, for the beginning.

Differentiating FNH and Adenoma

The difficulty in differentiating between FNH
and adenoma is that they are pretty much in the same group
of patients with predominant female,
occurrence occurring in the third to fourth decade of life.
There may be a history of oral contraceptive pills.
The fact that there are multiple lesions is not that helpful
because 10 to 20% of F and hs may be multiple,
but if you see Hermans, then
that might veer you towards being an FNH.

And you can also get multiple adenomas as well.

I think this is a useful statistic.
FNH is the second most common benign hepatic tumor occurring
in 8% of primary hepatic tumors,
whereas hepatic adenoma is occurring in
one in a million patients.

So if you're a betting person
and you're trying to decide what's this lesion going to be,
it's probably going to be an FNH.

It is important to try and differentiate them
because as we've heard, there is an instance of hemorrhage
and malignancy within subtypes of hepatocellular adenomas,
but it tends to be with the larger lesions.
So if you have a smaller lesion,
you probably have a bit more time to follow up with this
and not necessarily ru rush into treatment
or excision or biopsy.

The delayed phase imaging, the hepatocytes,
phase imaging and MRI is crucial in helping to differentiate
and prognosticate for the patient.
And FNH because of its, protein transporters
because of its matrix tends to hold on
to the hepatocellular agents on the hepatocellular phase
where adenoma tends not to.
And this is thought to be, secondary to the ot,
protein transporters.
It, it's not a hundred percent,
but it's, it's, it's fairly high up there.
And so that's a useful way of deciding
between the lesions if you're not going to biopsy
and if you're going to follow up.

Subtypes of Hepatocellular Adenomas

Now, this is a busy slide
and I would bear in mind that hepatocellular adenoma counts
for 1 1 1 in a million of our, of our lesions.
Well, not one in a million lesions,
but one in a million people
and not get too hung up about it.

But they've worked out the genetics
and the, the phenotypes of these lesions
and their character, characteristics and MRI.
And so if you're, if you're trying to be clever, we can look
and see that some of them will show more fa
and so it will signal loss on chemical shift.
Some of them will be more likely to bleed
and hold onto contrast.
Some of them will be have show a washout.
So it's not definite, but you can go
through this algorithm with your MRI.
I'm not gonna go through it now.
I think you, you'd probably look it up as you were doing it.

And try and prognosticate which lesion it is.
H and F1 alpha counts for 30 to 35% of adenomas
and inflammatories are the most common.
The beta cain is the one with the malignant potential.

And the um, inflammatories can bleed
and the beta cat containers can become malignant.
The unclassified accounts for 10%
and the thought that they may be unclassified
'cause a lot of these have hemorrhaged
and so the imaging characteristics are
difficult to work out.

F and H has different cellular types as well just
to show that it's not been left behind in its complexity.
So again, they'll all have general imaging characteristics,
but not every lesion will show them for definite.

Imaging Examples for FNH and Adenoma

So if we come across, we're scanning
and we come across this well circumscribed lesion
and we give some ultrasound contrast
and we can see that it is going
to avidly enhance relative to the background liver.
Here it is and there's the background liver on the arterial
phase and we can see it's holding onto contrast
on the portal venous phase
and looking at that arterial phase, it doesn't have
that spoke qual enhancement characteristic.
We know it's likely benign, it doesn't have a scar,
it's probably going to be,
or it may be hepatic adenoma If we're scanning this patient
and our gray scale shows us again a well circumscribed
lesion, the doppler is not that useful.

And then we go on to give contrast
and we see on the arterial phase
and the very arterial phase, this spoke wheel
spoke wheel ish spoke wheel enhancement characteristic
and then it holds onto contrast on the port venous phase.
Again, we think this because it has a scar and
because it has the SP enhancement,
we think this is likely to be an FNH.

And then I just threw it in another FNH
because I couldn't resist.

So you can see this is sort
of a nice spoke wheel enhancement characteristic holding
onto the contrast and the scar.
So most likely an FNH,
M on M-R-I-F-N-H will have, imaging characteristics
that are, are, only for it.
It will show that it has the scar
with MRI you may get delayed enhancement of the scar,
whereas with contrast ultrasound,
the scar doesn't enhance at all
because it's a truly intravascular, contrast agent.

And it will hold on to contrast
and it will hold on to
our hepatocyte specific contrast as well.

So they will have their scan
and then go on to have probably an MRI,
this is just a gadolinium MRIT one T enhancing,
arterial and continuing to enhance.
And then you can see delayed enhancement of the scar.

And this is taken from ravioli's paper which shows,
enhancement and then holding onto the contrast on the
delayed phase with a scar centrally.

And this is again from grazioli with biopsy proven,
FNH holding onto contrast on the hepatocyte phase
and the adenoma biopsy proven adenoma losing the contrast on
the hepatocyte specific phase.

We've heard that there is some crossover
and we have to be careful between hepatocellular adenoma
and HCC.
A small percentage of the inflammatory HCAs will,
show arterial enhancement and rapid wash out.
They can mimic each other basically. Nothing is definite.
We do have to follow them up.
There will be lesions we have to biopsy,
but in general there are broad imaging characteristics
that they will conform to.

The risk factors for malignant transformation
for hepatocellular adenoma is the male sex
glycogen storage disease steroid use and the beta contain
and mutated subtype
and also size greater than five centimeters.

Follow-up for FNH and Adenoma

Our follow up then is if we see an FNH on contrast
ultrasound, we will go on to our MRI usually
to confirm the diagnosis.
We don't routinely biopsy if we're happy that it's an FNH,
we don't, stop people getting pregnant
or stop them taking the pill or their steroids.

And we don't really think they need any intervention if
they are going to continue with the pill
or if they're getting pregnant,
we may do an annual ultrasound just to follow it up there.

This is supposed, this started off as an emoji,
but I dunno about transmission and kind of a monster.
This is supposed to show, women
that they're more common in women and they're larger
and women present with earlier lesions than men.

And but so they thought maybe they will had something
to do with it, but they found with increased use
of the pill over time, it hasn't led
to an increased prevalence
and they don't particularly increase the pregnancy.
So there isn't great evidence to say don't get pregnant
and stop the pill.

Okay, if we suspect, a beta contain
and hepatocellular adenoma,
then obviously we will refer for resection.
We may biopsy, we need
to watch these closer than the other lesions
because they have a chance of, becoming malignant.

And they won't, shouldn't show signal dropout
on chemical shift imaging.

Our advice for patients in whom we think have a he
hepatocellular adenoma is avoid the pill
or any hormones we will biopsy if the imaging is
inconclusive or it's going
to make a change in our treatment.
We don't contraindicate pregnancy if they're less than five
centimeters, but it is as always an individualized approach.
And if it's greater than five centimeters,
we definitely would recommend intervention.

They will initially have a follow up MRI not a CT at six
to 12 month intervals.
And then as the lesion becomes stable,
we will follow them up with ultrasound.

Again, just to put in the caveat,
some hepatocellular adenomas will continue
to increase in size even if the pill is stopped
or the steroids are stopped.
And some there are reports
of HCC developing in these lesions,
even if they get smaller in size.
So it is important to follow them up
by whatever imaging modality you choose
by either MRI or ultrasound.

And then these are just some of our local images.
This is the arterial phase of a biopsy proven FNH,
the port venous phase.
And on the coronal we have the hepatocyte phase.
You don't see a scar because it's small.

And then this was just from, this week,
this is a patient I think who has multiple hepatic adenomas
and this is an in and outta phase scan
coming up to the outer phase.
You can see their lipid rich.
So they are going to lose,
they've lost signal on the outer phase scan.

And then on the arterial phase scan,
we will begin to see this one up.
Sorry, I can't really see my screen here.
So they're almost eyes are enhancing
to the background liver on the arterial phase scan.
Remembering that we don't get such a great arterial phase
with MR as we would with contrast ultrasound.

And then on the portal venous phase scan there,
they're losing quite a bit
of signal on the portal venous phase scan,
that's another one by the IVC,
on our diffusion weighted imaging.
We're not seeing them at all, which is reassuring
or not seeing them very showing much restricted diffusion.

And then lastly on our prima scan,
we can see that they're not holding onto primas.
So we've been following these multiple adenomas for years.
They're biopsy proven.

Cirrhosis and New Nodules

Okay, so onto the very last bit of the talk
before lunch cirrhosis, a new nodule we've heard about this
and size is the important thing here.

If the lesion is less than a centimeter, it's unlikely
to be malignant and we would follow it up at three months
earlier than our normal follow up time.
If it's greater than a centimeter, then we're concerned.

And we've seen a similar flow diagram in the last chart
just show talk showing us that it's all to do
with the vascularity of the lesion
and the tumor invasion of the nodule.
Ultimately eating up the portal venous flow
and the hepatic venous flow so
that you get arterial at the end of your cancer stage.

And hypovascular on the portal venous phase,
but it's a bit in between that we are trying to catch.

So if we find a lesion greater than a centimeter
or if the patient has an elevated alpha protein
or we're concerned that there may be HCC, then we need
to perform, characterization with CT or MRI.

Most of the HCCs will show this characteristic
enhancement pattern on the arterial phase with washout,
but not all biopsy is can be used,
but there is a risk of seeding
and it's useful to use a coaxial biopsy technique
to minimize that risk.

It's also important to note
that hypervascular nodules can come and go in cirrhotics
and there's a wide spectrum of what they can represent.
So a hypervascular nodule does not necessarily mean a HCC
and in one study up to 25% of these lesions regressed.
So we have to maybe not biopsy everything
and keep a close eye on them as they come along.

So this is a regenerative nodule about a centimeter in size
and given contrast ultrasound,
and we can see that it's enhancing the same
as the background liver and the arterial.
And more importantly, on the venous
and delayed phase, it's holding onto contrast.
So this is a regenerative nodule.

And we can follow this up with ultrasound,
however, with the HCC, when we come up in our gray scale
and we begin to see larger lesions, we give the contrast
and we can see multiple lesions on this late arterial phase
scan or with the smaller,
this is just an early arterial phase scan
showing a rapid wash in of a, of a lesion.

So when we find a focal liver lesion in cirrhotics,
we usually give them contrast to try
and help us to characterize it as we know that
because we've got this broad imaging inspection
of these active nodules that are changing,
usually we will see,
a characteristic enhancement pattern,
but there will be some overlap.
We would have a, a low risk for follow-up and for biopsy.

Again, just to scan from this week.
So here's the lesion on the arterial phase.
Again, not a brilliant arterial phase Mr compared to
what you'd get on a contrast ultrasound.
Here's the washout in the portal venous phase,
here's the hepatocyte specific phase,
and here's restricted diffusion.

So the American guidelines say less than a
centimeter, a peak in three months.
Any change then off you go
with all your other investigations.
And if it's stable, go back to routine follow up.
Great in the centimeter, you need your four phase
with delayed contrast imaging, whether that's CT or MRI.
The Americans at the moment don't recommend using contrast
ultrasound because of the overlap
between hepatocellular adenomas and HCCs.

If there's any, question,
then they ultimately go to biopsy.

Conclusion

So that's it pretty much.
We have an algorithm so that we can say to the patient,
you have a cyst, you have Herman, you're fatty sparing,
you have a solid lesion, it's likely an FNH or an adenoma.
And then the end stage is happy patient
and an alarmed radiologist.

Okay, so thanks very much. Enjoy your lunch.