Upcoming Events
Log In
Pricing
Free Trial

Diffuse Liver Disease, Dr. Kevin Chang (10-8-26)

HIDE
PrevNext

0:02

Hello, and welcome to Noon Conference hosted by Medality.

0:05

Noon Conference connects the global radiology community through free live

0:09

educational webinars that are accessible for all and is an opportunity to learn

0:12

alongside top radiologists from around the world.

0:16

Today, we're honored to welcome Dr.

0:17

Kevin Chang for a lecture entitled Diffuse Liver Disease. Dr.

0:21

Chang completed his radiology residency at Boston University and his

0:24

cross-sectional imaging fellowship at Johns Hopkins.

0:27

He is section chief of abdominal imaging and director of MRI at Boston University

0:32

Medical Center, associate professor of radiology at Boston University's Chobanian &

0:36

Havisian School of Medicine, and an adjunct associate professor of diagnostic

0:40

imaging at Brown University's Alpert Medical School.

0:43

At the end of the lecture, please join him in the Q&A session where he will address

0:46

questions you may have on today's topic.

0:49

Please remember to use that Q&A feature to submit your questions so we can get to

0:52

as many as we can before our time is up.

0:54

With that, we're ready to begin today's lecture. Dr.

0:57

Chang, please take it from here.

0:59

All right. So great. Thank you for letting me give a talk about one of my favorite

1:05

topics. That's diffuse liver disease.

1:06

I've been talking about this, and I think this topic first came up as a resident

1:12

conference back when I was a resident many, many years ago.

1:15

And the talk has evolved a lot since then.

1:18

There's a lot of topics to cover here, so there will be some topics that I will

1:21

definitely not be able to get much in-depth into, but we will try to get through as

1:26

much as we can and cover as many topics as we can.

1:29

These are my disclosures, none relevant to our topic today.

1:34

So definitely, we're using a lot of cross-sectional imaging for diffuse liver

1:37

disease. There's been a lot of advances in CT and MR technology and techniques.

1:41

A lot of these diagnoses still require biopsy to make a definitive diagnosis, but

1:45

imaging is definitely useful for narrowing differential diagnoses, detecting

1:49

complications of a lot of these disease processes, such as portal hypertension,

1:54

monitoring responses to therapy, as well as following up known diagnoses.

1:59

And increasingly, CT and MR and ultrasound are being used as a metabolic biomarker.

2:04

A lot of these findings are used as a biomarker of disease.

2:07

I'm giving the answers to all these cases right up front here in the form of an

2:12

outline. So these are

2:15

not entirely an all-inclusive list of diseases.

2:20

But we're going to touch on cirrhosis first because that's the hallmark of diffuse

2:25

liver process, as well as all the metabolic and storage-related disease processes.

2:28

These are probably the two most common causes of diffuse liver disease that we

2:32

encounter on a daily basis, multiple times a day.

2:36

There's also other causes of diffuse liver diseases like drug-related responses,

2:42

drug-related imaging findings, as well as some hereditary diagnoses like glycogen

2:47

storage diseases and Wilson's disease, and then some more idiopathic processes like

2:52

amyloidosis. And in addition, there are other causes that can look diffuse as well,

2:58

such as infectious causes, hepatitis, schistosomiasis, and then other causes like

3:05

sarcoidosis, diffuse neoplasm, and then a whole array of vascular

3:10

etiologies that can lead to more diffuse, non-mass-like changes or

3:17

imaging findings on various studies, especially the multi-phase

3:22

contrast enhanced studies. General categories in terms of patterns of disease.

3:27

There are some homogeneous patterns, which are predominantly due to parenchymal

3:33

involvement of hepatocytes in the reticuloendothelial system.

3:38

There are more segmental processes as well that are often related to perfusion

3:43

because these affect certain segments of the liver or subtotal portions of the

3:49

liver. So a lot of this is related to the interesting perfusion pattern of the

3:54

liver due to, it's not only its portal venous and hepatic artery perfusion, but

3:59

also third inflow phenomenon of other collateral vessels that affect certain parts

4:03

of the liver, which we'll get into in a little bit.

4:06

There's nodular, more depositional patterns of disease as well.

4:10

So things that can lead to small little nodules throughout the liver, as well as a

4:14

perivascular distribution related to the lymphatic system and Glisson's capsule.

4:21

So let's start first with the findings that we see multiple times a day.

4:26

That's patients with a liver that's very nodular or lobular in contour.

4:32

Sometimes you'll see other manifestations of this disease process, which is

4:37

cirrhosis, like findings of portal hypertension, such as splenomegaly, which you

4:42

can see here with a big spleen.

4:45

Varices, paraesophageal, perigastric, periesophageal, periesophageal, a bunch of

4:50

portosystemic shunts in various typical locations throughout the abdomen and

4:54

pelvis. Presence of ascites. So these are all findings that we see every day in

4:58

cirrhosis. I'm not going to get too much into detail here because we all see these

5:03

changes and are pretty good at least detecting the findings of moderate to severe

5:08

cirrhosis. This can lead to architectural distortion, and it's these

5:13

macroscopic changes that

5:16

we're the best at detecting. But unfortunately, a lot of these architectural

5:20

distortive changes are more evident with more severe cirrhosis rather than mild

5:24

cirrhosis. Mild cirrhosis we may certainly overlook on imaging alone.

5:30

Different causes. These are the typical causes.

5:32

Alcohol,

5:34

viral causes, as well as other causes like primary biliary cirrhosis and

5:38

hemochromatosis, and metabolic-associated etiologies as well.

5:44

And the reason why we get so worried about this is because of the progression to

5:48

hepatic failure, as well as the manifestations and complications of portal

5:53

hypertension.

5:55

In terms of imaging, like I said before, we're looking for surface nodularity.

5:59

That's probably one of our more specific findings, but a later-stage finding of

6:03

disease.

6:05

Architectural changes usually will manifest as atrophy of certain

6:11

portions of the liver, like the medial left lobe and the portions of the right

6:16

lobe, and enlargement of other portions of the liver, like the caudate lobe and the

6:20

lateral segment of the left lobe. And they can even lead to changes around the

6:25

gallbladder fossa, where you have increased fat around the gallbladder, the

6:29

expanded gallbladder fossa sign.

6:32

Other examples of cirrhosis. We're not going to get too much into the details of

6:36

the complications of cirrhosis, such as hepatocellular carcinoma here.

6:40

That's certainly a huge topic in and of itself, which we're not going to touch on

6:43

here. But here you see masses that have some hemorrhage in them.

6:48

That's typical for

6:51

hemorrhagic hepatocellular carcinoma.

6:53

You've got pseudocapsules, you've got ascites here, got washout, all the typical

6:58

findings for HCC in the setting of cirrhosis.

7:03

Other ways to evaluate cirrhosis.

7:06

We've got elastography now that we do both by MRI as well as ultrasound.

7:11

My institution tends to rely more on MR elastography, where we put an

7:16

acoustic transducer on the abdominal wall, and we send sound waves in, and we can

7:21

visualize the speed of the sound wave propagation through the liver.

7:27

And measuring the speed of the propagation of the sound waves will give you a

7:32

kilopascal measure of liver stiffness.

7:35

And the stiffer the liver, the higher the kilopascal

7:40

number that you can generate from an ROI on the elastography maps.

7:46

Certainly, I'm not an expert in elastography, but we do an increasing number of

7:51

these, trying to stage the severity of fibrosis

7:55

while trying to reduce the number of biopsies that are performed.

8:01

And at the same time, you can do proton density fat fractions as well to judge the

8:04

degree of steatosis. This can be done by ultrasound as well.

8:09

Basically, similar technology, sending a sound wave in, and you're looking at the

8:13

shear wave, the perpendicular shear wave, the velocity of the perpendicular shear

8:17

wave, and correlating that with the relative stiffness of the liver.

8:25

Next topic, beyond cirrhosis, is going to be the other finding that we probably see

8:30

just as much as, if not more commonly than in cirrhosis, and that's when you see

8:36

hepatic steatosis, otherwise known as fatty infiltration of the liver.

8:42

On CT, typically, you'll see diffuse hypodensity of the liver, and this

8:48

can be easier to detect on a non-contrast CT, where contrast enhancement doesn't

8:55

get in the way of evaluating the liver attenuation.

8:59

Typically, you're looking for an attenuation that's less than 40 Hounsfield units

9:03

or something that's more than 10 Hounsfield units,

9:08

less dense than the spleen, which is your internal control.

9:12

On MRI, we'll use T1-weighted imaging because fat is bright on T1, and if we do

9:18

dual echo, gradient echo sequencing, we can look at the in-phase and out-of-phase

9:21

series, the echoes, to determine the degree of signal

9:27

dropout related to Fat and water protons sharing the same voxel.

9:31

So if there's a fat proton and a water proton in the same voxel, on the

9:35

out-of-phase image, the signal intensities are going to cancel each other out, and

9:39

you're going to see a relative drop in signal.

9:41

So in this case, you've got a very T1 bright liver, which diffusely drops out in

9:47

signal on the out-of-phase T1-weighted gradient echo image.

9:52

So there's diffuse homogeneous

9:55

steatosis in the liver. It can be segmental as well.

9:57

So you can see in this case where the out-of-phase images, you're seeing the signal

10:01

drop out in most of the liver, except for the posterior right lobe here.

10:05

And this is related to the segmental anatomy of the liver for one reason or other.

10:09

The posterior right lobe is being relatively spared from the degree of fatty

10:13

infiltration that the rest of the liver is facing.

10:16

So

10:18

focal fatty sparing. Note that these areas that show these signal changes,

10:23

sometimes they can look mass-like.

10:25

But the key here on whatever the imaging modality you're using, whether it's

10:29

ultrasound, CT, or MRI, is that there's no vascular distortion.

10:32

So you'll see the portal triads and the hepatic veins going through these areas

10:36

without any mass effect from these processes.

10:39

And that can be important when you're looking at cases of a more focal fatty

10:43

infiltration or even nodular mass-like areas of fatty infiltration or fatty

10:49

sparing. This is another example that's a little more focal here in the posterior

10:54

right lobe, segment six area of the liver.

10:56

You can see there's an area of decreased attenuation that looks very wedge-shaped,

11:01

non-mass-like, more segmental or geographic in appearance.

11:06

And this is a typical appearance for focal fatty infiltration or focal

11:11

steatotic liver disease. And on MRI, it can be even more evident here.

11:18

Same patient, same case. On the in- and out-of-phase images here, you can see the

11:24

wedge-shaped segmental or subsegmental area of signal dropout on the out-of-phase

11:29

images compared to the in-phase images.

11:31

You can even appreciate that there's higher signal intensity on the in-phase

11:36

because of the summation of the fat and water-related

11:40

signal intensities there, and then they drop out because of the out-of-phase

11:46

sequence, the echo time for the gradient echo sequence.

11:50

Even on T1-weighted images, if you don't use fat saturation, sometimes it'll look a

11:53

little bit brighter because of the fat content, not the water content.

11:57

And if you do a T2 fat saturated sequence, similarly, that signal intensity can

12:01

drop a little bit compared to the non-fat saturated sequence.

12:05

And sometimes you'll see perfusional changes related to this, and that may be one

12:10

of the reasons for why there's focal fatty infiltration.

12:14

Or conversely, if you see the opposite findings, focal fatty sparing may be related

12:19

to differences in the portal vein perfusion pattern for certain portions of the

12:24

liver. And this pattern can also occur in typical areas that

12:30

have what's called a third inflow phenomenon, which we'll get into in a second.

12:35

Here's another case here on CT where you see diffuse hypoattenuation of the liver,

12:41

except for this tiny little wedge-shaped subcapsular area here.

12:45

So there's a little bit of focal sparing in segment, I think this is seven,

12:51

six or seven. And then on the arterial and the venous phase images

12:57

here of this liver mass protocol CT, you can better see that there's an area that's

13:02

responsible for the sparing in terms of a rounded area of

13:08

peripheral nodular discontinuous arterial phase enhancement that fills in

13:13

centripetally on the venous phase.

13:14

So this is a hemangioma at the apex of this area of focal sparing, which explains

13:20

why this whole segment that's subtended by this hemangioma is being relatively

13:26

spared from fatty infiltration. Maybe because there's more hepatic arterial inflow

13:32

and decreased relative portal venous inflow, or there may be some arterial portal

13:37

shunting that's reversing the portal vein perfusion of that segment of the liver.

13:43

So this is one of the causes of what we call FIDs and FADDs, transient hepatic

13:48

intensity differences or transient hepatic attenuation differences in the setting

13:52

of a steatotic liver.

13:57

On ultrasound, oftentimes we'll see diffusely increased echogenicity of the liver,

14:02

which most of the times we end up

14:06

attributing to steatosis. And then what can make that diagnosis even more

14:12

specific, because fat is not the only thing that causes increased echogenicity in

14:16

the liver, just happens to be the most common cause.

14:20

When you see these non-mass-like, more geographic subcapsular or

14:26

pericholecystic or periportal or peri-falciform areas of

14:32

decreased attenuation, these areas are classic locations for focal fatty sparing.

14:37

And when I see that, I'm more confident that the increased echogenicity is probably

14:41

from diffuse liver steatosis. So this is one area here in the deep

14:48

segment four area of the liver, and then this one is a kind of triangular area next

14:53

to the gallbladder, pericholecystic sparing.

14:56

And you can see the liver is definitely much more echogenic than the right kidney,

15:00

which we use as our internal reference for liver echogenicity.

15:05

Same patient on MRI. You can see on the out-of-phase T1 versus the in-phase

15:11

T1 gradient echo images, that same area in the deep

15:18

segment four area of the liver that is showing less dropout than the remainder of

15:23

the liver on the out-of-phase images.

15:26

Similarly, on the T2 weighted images with and without fat saturation, you can see

15:31

that with the fat saturated out, that area is a little bit brighter than the rest

15:36

of the liver parenchyma, which is fatty.

15:38

And then on the pre- and post-contrast multiphase images, you can see there's

15:42

perfusional effects here as well.

15:44

There's a little bit of a hyper enhancement in that area, which is likely

15:48

perfusional rather than mass-like or mass related.

15:53

So steatosis with focal fatty sparing.

15:57

Here's another dramatic example of steatosis with more central sparing in relation,

16:03

or in this case, it was likely secondary to chemotherapy following

16:09

treatment for pancreatic adenocarcinoma.

16:12

So steatotic liver disease can be either from an overproduction or a faulty

16:18

metabolism of fatty acids or decreased hepatic clearance of fatty acids due to

16:22

hepatocellular injury, which could be related to alcohol, could be post-viral

16:27

hepatitis, could be medication related, such as chemotherapy related.

16:33

And the distribution can be pretty variable, can be diffuse, can be focal or

16:37

subtotal, like I showed you in a few cases, with areas of typical sparing or

16:44

focal fatty infiltration. And this can be explained in large part by what

16:50

I think is the third inflow phenomenon.

16:51

It's one of the theories for why certain areas of the liver are either focally

16:56

spared from steatosis or can be focally involved by steatosis.

17:01

Typical areas, like I said, could be related to cystic vein collaterals around the

17:07

gallbladder. So that could explain pericholecystic sparing or pericholecystic focal

17:12

involvement. The veins of Sappey around the falciform ligament

17:18

can explain a lot of the

17:21

areas around the falciform ligament on our various modalities, as well as there are

17:26

other areas of collateral flow as well around the porta hepatis region.

17:33

So third inflow can also explain a lot of our perfusion-related changes in addition

17:38

to areas of fat sparing and fat infiltration.

17:44

All right, so our next case, we're going to move on from steatosis to

17:50

causes of liver hyperattenuation.

17:52

So this is a non-enhanced CT where the attenuation is 75 Hounsfield units.

17:57

Usually, normal range is 45 to 65 Hounsfield units.

18:01

So in this case, this is a hyperdense liver, and there's a broad differential

18:05

diagnosis for a hyperdense liver.

18:07

Could be from amiodarone,

18:10

which is an iodinated deposition.

18:13

Could be from iron, like from hemosiderosis or hemochromatosis.

18:17

Could be from glycogen storage disease, could be from gold therapy, could be from

18:22

copper deposition in Wilson's disease, could be from thoroughgoing.

18:25

So there's a broad category of multiple possible etiologies for

18:31

increased liver density. MRI is a little bit more sensitive in detecting the cause

18:37

of the increased density. In this case, you can see the liver is super dark on a

18:41

T1-weighted image, as well as on

18:44

the gradient echo pre-contrast images, as well as T2-weighted images.

18:48

So in this case, it's the liver as well as the spleen. The spleen is also dark.

18:53

So in this case, when you have

18:57

susceptibility or paramagnetic effects in both the liver and the spleen.

19:01

We're thinking about something that's beyond just hepatocyte deposition.

19:05

We're talking about the reticuloendothelial system.

19:08

So in this case, this is a case of hemosiderosis or iron deposition in both the

19:12

liver and the spleen, and maybe in the bone marrow as well.

19:16

Sometimes you'll see that on an in-and-out-of-phase T1-weighted gradient echo

19:21

image, such as this case here. So you can see not only is the liver and the spleen

19:26

dropping in signal on the in-phase compared to the out-of-phase, because the

19:30

in-phase has a longer echo time than the out-of-phase sequence.

19:33

So in this case, it's the reverse of steatosis.

19:36

Instead of signal dropping on out-of-phase, when you see the signal dropping more

19:40

on the in-phase, that tells you there's something like iron, something that's

19:44

causing a paramagnetic

19:47

susceptibility effect on the liver and the spleen.

19:50

And in this case, the bone marrow as well.

19:52

So typically, since bone marrow tends to have both red and yellow elements, it

19:57

tends to drop more in signal on the out-of-phase than in-phase because of the mixed

20:01

fat and water protons in bone marrow.

20:03

But in this case, the opposite is happening, which tells you that there's iron in

20:08

the reticuloendothelial system of the bone marrow as well.

20:12

So this is another case of more severe hemosiderosis.

20:17

And similarly, on the T2-weighted images, you just see a huge amount of signal

20:22

drop. And this can be even more evident on diffusion-weighted images, where you

20:27

basically see a signal void in the liver on those

20:31

diffusion-weighted images of the various b-values.

20:37

All right. Another case here,

20:40

just to hammer in the point here. We've got out-of-phase and in-phase T1s on the

20:44

top with signal dropout in the liver and the spleen, as well as low signal on

20:48

T2-weighted images, as well as a lack of signal on the diffusion-weighted images

20:53

here. Another case of hemosiderosis.

20:55

Whereas in contradistinction or in contrast, we're seeing another case here

21:01

of T1 gradient echoes here, where there's signal dropout on the in-phase compared

21:05

to the out-of-phase in the liver, but it's sparing the spleen.

21:10

And

21:12

you also see the low signal on the pre and post contrast images here, as well as on

21:17

the diffusion-weighted images. But when you have iron deposition in the liver

21:21

sparing the spleen, we're starting to think about hepatocyte deposition of iron,

21:26

and that's more characteristic for hemochromatosis, primary hemochromatosis, or

21:32

hereditary hemochromatosis.

21:37

This is another case, again, showing low signal in the liver, sparing the spleen.

21:43

Sometimes you'll see the deposition as well in the pancreas, and this is another

21:46

location where hereditary hemochromatosis may show iron deposition

21:52

within cells in the pancreas. Sometimes you also see the same finding in the

21:56

myocardium, if the myocardium is included.

21:58

If the myocardium looks lower signal than usual.

22:01

That's the typical distribution of iron deposition in primary hemochromatosis.

22:06

So again, another case of hemochromatosis.

22:09

This one is another one here showing in and out-of-phase images here.

22:12

Out-of-phase versus in-phase. Signal drop on the in-phase in the liver, sparing the

22:17

spleen. This liver also shows a nodular or lobular contour, so there's secondary

22:23

cirrhosis as a result of the hemochromatosis in this case.

22:27

And the

22:32

next case I'm going to show you here is the same thing with a little bit more of a

22:37

segmental pattern of involvement. The whole liver is dark.

22:40

Maybe there's more signal dropout that's easier to see in the right lobe

22:46

on the in-phase versus the out-of-phase, although the whole liver is dark in

22:49

signal. Except there's a mass in the left lobe here in a patient with

22:52

hemochromatosis that's developed a secondary neoplasm.

22:57

These primary hemochromatosis patients are at elevated risk for developing a

23:01

neoplasm, particularly primary hepatic neoplasms such as a cholangiocarcinoma

23:07

or a hepatocellular carcinoma. And in this particular case, this ended up being an

23:12

intrahepatic cholangiocarcinoma.

23:17

All right. So iron overload.

23:19

Again, to reiterate, primary hemochromatosis is the hereditary autosomal recessive

23:24

disorder of iron metabolism, resulting in excessive deposition in not just the

23:29

hepatocytes of the liver, but the pancreas, the myocardium, and it can also affect

23:33

the joints and the skin. And without treatment, 35% of them can develop a hepatic

23:39

neoplasm. Otherwise, the alternative is the other cause that I showed you.

23:46

First was hemosiderosis or iron overload.

23:48

And this can be either transfusional or it could be related to dyserythropoiesis,

23:54

such as like thalassemia, spherocytic anemia,

23:59

other

24:00

pyruvate kinase deficiencies. So this results in deposition of iron in the

24:04

reticuloendothelial system cells.

24:05

Again, this is going to be the liver, spleen, and bone marrow.

24:09

Usually, it doesn't have the same clinical implications as primary hemochromatosis

24:14

does, unless it's so severe that it's leading to what some people call secondary

24:20

hemochromatosis.

24:24

Here's another case. Again, you can see the T1 gradient echo images out-of-phase

24:28

versus in-phase. There's diffuse signal drop in the liver.

24:31

There's a little bit in the spleen as well.

24:33

But take a look at these adrenal glands. They're super dark.

24:35

So there's a lot of iron being deposited in the adrenal glands in this case.

24:39

And again, on T2-weighted images, you can see similar low signal intensities in the

24:43

liver and the spleen and the adrenal glands.

24:46

And the cause here was due to ferrioxamine deposition.

24:51

So this is more of an iatrogenic cause of iron deposition in multiple organs

24:56

throughout the body. So it's important to note that if somebody's got an anemia

25:00

that they're getting iron therapy for, that if they are going to get an MRI, you'll

25:05

want to try to time the MRI

25:09

to be as far out from the ferrioxamine transfusion as possible, just to reduce

25:13

these artifactual effects of iron deposition, which can be completely

25:19

iatrogenic. In terms of quantifying the iron, so we subjectively looked at iron

25:24

deposition, but we can do quantification by MRI as well.

25:28

And there are two major methods that our institution has used.

25:33

We've basically moved on to the easier-to-perform measure, which is

25:39

generating R2* map and then plugging the ROIs that you get from the R2* relaxation

25:45

times into different formulas that were described by Diego et

25:50

al. in the paper at the bottom there recently.

25:54

So just plugging that number in should give you a liver iron concentration in

25:58

milligrams per gram. And the

26:01

formula differs at 1.5 Tesla versus 3 Tesla.

26:05

There's a vendor-neutral option as well from Yves Gandon, who has some

26:11

free-to-use software online at that address below, where you can plug

26:16

in the relative signal intensities of liver and muscle at various echo times.

26:23

And you can generate the concentrations using the signal decay

26:29

curves from this piece of software as well.

26:31

And this does not require any kind of a vendor-specific pulse sequence to generate

26:37

an R2* relaxation time. I'm not going to get into the details of how it's done, but

26:42

we've got Philips, Siemens, and GE scanners at our institution at various

26:47

field strengths. And basically, these are the slides that I

26:53

use to show the residents and fellows how to measure the ROIs, trying to measure

26:58

areas in the liver with various shapes of ROIs that try to exclude

27:04

major vessels as well as the lung base and

27:07

the adjacent organs. And then basically taking the average of these numbers and

27:12

plugging them into the relative formulas to generate the liver iron concentrations.

27:20

This is Dr. Gandon's piece of software here where you can just plug in different

27:25

ROIs in the liver and the spleen and the muscle and generate these decay curves

27:30

that can separate the relative signal intensity variations related to

27:36

fat infiltration or steatosis, separate from the overall decay curve that's more

27:42

related to the susceptibility effects related to iron.

27:46

And then his pieces of software can even generate the reports for you.

27:51

Good piece of software, but I find the R2* maps to be a lot quicker to measure for

27:57

All right. The next couple topics are going to be a little less

28:03

specific in terms of imaging, so I'm just going to give you a lot of these answers

28:07

right away. But again, we've got another liver here that's hyperdense, measuring

28:11

120 Hounsfield units on the non-contrast phase here, and it's

28:17

affecting the entire liver here. So we can think about that same differential about

28:23

hyperdense liver processes. In this particular patient, it's worth looking at the

28:29

lung bases as well to see if you see any interstitial lung changes.

28:33

Note that this patient also has a cardiac lead, so we know that this patient has

28:37

cardiomegaly and some kind of an arrhythmia.

28:41

In this particular case,

28:43

this patient is on amiodarone therapy, and the iodine in the amiodarone

28:49

can deposit in the liver and give you a hyperdense liver.

28:53

So there's another

28:55

interesting cause of a diffuse liver process that we can see imaging manifestations

29:00

of.

29:02

But in this case, it usually involves looking at the

29:08

medical record to see that they're on amiodarone.

29:14

Okay.

29:16

Next case here, we can see that there's a liver on ultrasound that's more

29:22

echogenic than the kidney. So in most cases, we might think that the patient is

29:27

just another case of a steatotic liver.

29:29

But in this case, if you look at the CT and the MRI, this is not the appearance of

29:34

a fatty liver on CT. No, this liver is hyperdense.

29:39

So what's causing a hyperdense liver on non-contrast CT that's also echogenic on

29:44

ultrasound and very T1 bright on the T1-weighted pulse sequence on MRI?

29:50

So if you compare it to what a normal liver looks like, you can see the liver T1

29:55

signal hyperintensity is even more

30:00

hyperintense than bone marrow, which tends to have some fat in there.

30:04

So this is just another way to gauge the relative T1 signal intensity in liver

30:10

compared to a normal liver. So in this case, this is a glycogen storage disease.

30:15

So this pattern of an echogenic diffuse liver process that's hyperdense on CT and

30:21

T1 bright is a pattern that's probably under-recognized in the setting of glycogen

30:26

storage disease. And patients with glycogen storage disease are predisposed to

30:31

liver neoplasms as well. In this case, this was a

30:35

hepatic adenoma.

30:38

Here's another case of glycogen storage disease, showing you the

30:43

T1 hyperintensity with a focal hyperenhancing mass on the arterial phase here in

30:48

somebody with von Gierke's disease with an adenoma.

30:54

And notice that the adenoma looks like it's dropping in signal a little bit on the

30:59

in-phase versus out-of-phase, but it's actually an optical illusion.

31:04

It's the liver that's dropping a little bit in signal on the out-of-phase compared

31:07

to the in-phase,

31:10

making that adenoma a little bit easier to see in the background of a fatty liver.

31:17

So glycogen storage disease, it's a problem in the use and the storage of glycogen

31:20

due to one of many possible enzyme defects. There's maybe more than 11 now.

31:25

The last time I checked, there was at least 11 known different types of glycogen

31:29

storage disease, and it leads to the deposition of glycogen in the liver.

31:34

And the glycogen deposition in the liver leads to echogenicity in the ultrasound,

31:39

but hyperdensity on CT and hyperintensity on T1-weighted images on MRI.

31:48

Next disease process, which is very nonspecific on imaging, is Wilson's disease.

31:53

This is a rare autosomal recessive disease, often presenting in younger patients.

32:00

The diagnosis is typically more clinical or laboratory-oriented with low serum

32:05

copper and ceruloplasmin levels.

32:08

Copper gets deposited in multiple organs in addition to the liver, such as the

32:13

brain and the cornea. And the liver deposition tends to be periportal and along the

32:18

hepatic sinusoids, invoking an inflammatory reaction.

32:23

Can lead to an acute hepatitis acutely.

32:26

Can lead to steatosis, which can also confound your

32:32

liver attenuation measurements.

32:35

So these patients don't necessarily have a hyperdense liver.

32:37

They can, but not always. And then the more chronic active hepatitis can lead to

32:43

fibrosis and eventual cirrhosis.

32:45

So it's just one of many causes of scarring in the liver, of course, cirrhosis.

32:53

In addition to the variable, sometimes hyperdense appearance on CT can lead to

32:57

hepatomegaly as well. Variable degrees of steatosis and eventual

33:03

cirrhosis. So this is an example from the literature of a very diffuse, kind of a

33:09

multinodular process. Sometimes these

33:14

liver processes can be more diffuse.

33:17

This is a

33:19

patient with multinodular involvement throughout the liver.

33:23

So most of these nodules are

33:26

hypointense. It can be hypointense on T1 and T2.

33:30

This is a patient actually with a different liver process, which is also similarly

33:35

very nonspecific. In this case, this is a patient with amyloidosis.

33:39

Amyloidosis is the deposition of fibrils of protein and mucopolysaccharide

33:45

complexes in multiple different organs.

33:48

The liver is the third most common site behind the spleen and kidney.

33:51

And in the liver, again, these deposition of fibrils are along the hepatic

33:56

sinusoids. The most common appearance is going to be hepatomegaly, and you can get

34:01

these hypodense pseudotumors or little tiny amyloidomas throughout the

34:07

liver, or sometimes you'll see them in the spleen as well as in other organs.

34:11

This is an example of a kind of an area

34:15

of a pseudotumor, or it could even be an area of confluent hepatic fibrosis on a

34:21

T1-weighted image, old image from the literature.

34:24

But here the findings are hepatomegaly,

34:29

you may see ascites, and this pseudotumor area might lead you to believe

34:34

there's amyloid involvement in somebody with a known history of amyloidosis.

34:39

So clinical history is key in interpreting those imaging findings.

34:45

All right. Next disease process here, I'm going to show you an ultrasound through

34:49

the liver. You'll see it looks a little heterogeneous in this case.

34:53

Again, an echogenic liver. Is it fat, or is it something else?

34:58

The liver coarsening, liver hyperechogenicity, liver heterogeneity

35:04

is a very nonspecific finding, which can be related to processes beyond just

35:10

steatosis. In this particular case, you've got some pre- and post-contrast images

35:15

in the liver on MRI, showing you very heterogeneous signal

35:20

intensity and a very heterogeneous enhancement pattern that looks like it's

35:24

following the vasculature in the liver.

35:27

This perivascular appearance is a characteristic finding that we see with liver

35:32

inflammation. In this case, hepatitis, acute hepatitis.

35:36

It can be more segmental as well. In this particular case, on a CT

35:42

scan, you can see there's relative hypoe enhancement involving the medial left lobe

35:47

of the liver with a little bit of capsular retraction, which is a little more

35:51

evident a year later. And this is an area of focal hepatitis that can lead to

35:56

confluent hepatic fibrosis over time.

36:00

On MRI, you can see

36:04

on, again, T1-weighted gradient echo out-of-phase and in-phase images, you can see

36:07

an area of fatty deposition involving predominantly the right lobe of the liver

36:12

with some atrophy of the left lobe here.

36:14

And then on the pre- and post-contrast images on the bottom, you can see a very

36:19

heterogeneous enhancement pattern in the same area along the lateral right lobe of

36:25

the liver in a patient with what looks like hepatitis, in this case,

36:31

possibly related to steatohepatitis.

36:33

So whether the steatohepatitis is metabolic-associated or not, it depends on the

36:41

risk factors of the patient with clinical history.

36:44

So

36:45

hepatitis is basically a nonspecific inflammatory response to hepatic insult.

36:49

Can be from a bunch of different etiologies, infections, autoimmune processes,

36:54

alcohol, drugs, environmental exposures, radiation, metabolic disorders,

37:00

and can be either self-limited, but if it's progressive and chronic, it can lead to

37:04

eventual cirrhosis.

37:08

The scarring and the necrosis inflammation can be focal or diffuse,

37:14

and the acute process can involve the portal tracts, and when it's severe, can lead

37:19

to hepatomegaly and periportal edema.

37:21

But the imaging findings are otherwise relatively non-specific.

37:25

What you're looking for is that heterogeneous enhancement pattern in the acute

37:28

phase can also lead to similar T2 signal intensity differences in a perivascular

37:32

distribution, whereas the late stages will basically resemble early-stage

37:38

cirrhosis. Other findings that can be associated with that would be gallbladder

37:43

wall thickening from edema, ascites, and periportal lymphadenopathy.

37:50

All right. This one's an ANT mini, different disease process.

37:54

You can see what looks like a macrolobular contour to the liver here.

37:58

You can see big, lumpy, bumpy appearance of the liver with areas of hypertrophy.

38:04

But in addition, the

38:07

mini-finding here is calcifications between these segmental, subsegmental areas of

38:12

enlargement, as well as a similar pattern of enhancement along these

38:18

kind of intersegmental areas following contrast administration.

38:23

And then sometimes you'll see similar signal abnormalities on T2 weighted images

38:29

that look like the septations sometimes that you see in the setting of macronodular

38:34

cirrhosis. But the calcifications in this pattern are that giveaway that you're

38:39

dealing with what some people call a kind of a tortoise shell appearance of

38:43

calcifications, something that we typically see with schistosomiasis.

38:48

So schistosomiasis is an infectious cause of a diffuse liver process.

38:53

It's basically

38:58

these adult worms that are water-borne, get into

39:04

the bowel, and then the eggs from these organisms get laid and embolize in the

39:09

mesenteric veins. And these eggs that embolize in the mesenteric veins basically

39:14

shower the liver and end up being deposited in the periportal distribution.

39:19

And one particular schistosomiasis organism is the one that incites this

39:24

calcification kind of a response.

39:26

The schistosomiasis japonicum

39:29

organism. But there's other ones like mansoni and haematobium as well that can

39:34

cause this kind of a

39:38

pattern of involvement. So the periportal

39:43

deposition can lead to the granulomatous reaction and fibrosis and

39:48

calcification in that particular japonicum variant of schistosomiasis.

39:53

So tortoise shell calcifications, schistosomiasis.

39:59

Another very non-specific imaging

40:02

finding of diffuse liver disease is sarcoidosis.

40:05

I'm giving an answer right up front before I show you the images, but it's

40:08

basically a multi-systemic disorder of unknown etiology from non-caseating

40:12

granulomas in almost any organ.

40:16

More often than not, you're not going to see the hepatic involvement from

40:19

sarcoidosis, but sometimes it'll look like tiny little nodular areas of deposition

40:24

in the liver as well as in the spleen.

40:28

More commonly, you'll see the pulmonary manifestations as well.

40:32

So if you see multi-organ involvement, definitely think about sarcoid, particularly

40:38

if the pulmonary findings are more specific in making that diagnosis.

40:43

I think it's easier to recognize this in the lungs than it is in the liver and the

40:48

spleen. But on CT, you'll again may see some heterogeneity in the attenuation

40:54

as well as hepatomegaly and lymphadenopathy.

40:57

You may see some pseudotumors or sarcoidomas in the liver and spleen.

41:03

And similarly, on MRI, you'll see hepatomegaly and

41:08

variable signal intensity. So this is an example of a liver-spleen involvement.

41:12

You can see these tiny little hypodense nodular areas scattered throughout the

41:18

spleen, as well as maybe more subtly in the periphery of the liver here.

41:23

In the same patient in the lungs, you can see your typical kind of a

41:28

periseptal, perifissural, subpleural distribution pattern of

41:34

nodularity and interstitial involvement in a patient with sarcoidosis.

41:38

And then in the same patient on a scrotal ultrasound, you can see a bunch of

41:42

hypoechoic nodules in both testes.

41:46

So this is presumably from diffuse sarcoidosis.

41:53

And this is another patient, again, showing you a bunch of little hypodense nodules

42:00

diffusely scattered throughout the periphery of the liver, maybe some subtle ones

42:03

in the spleen as well. You can see them on ultrasound as well as these little

42:07

hypoechoic areas. And if you saw your typical pattern of pulmonary involvement,

42:13

maybe some lymphadenopathy as well, then you can presume, or you

42:19

can put forth the possibility of sarcoidosis involvement in the liver and the

42:24

spleen.

42:26

This is another case of known sarcoidosis leading to eventual cirrhosis

42:32

in the liver. So you can see the macrolobular pattern of the capsule, as well as

42:38

the expanded gallbladder fossa sign.

42:39

In this particular case, the cirrhosis was from sarcoidosis.

42:45

And then another etiology that can also lead to a very similar pattern of kind of

42:49

multiple small nodules or even a pseudocirrhotic appearance of the liver with

42:54

capsular nodularity and modularity would be diffuse neoplasms as well as treated

43:00

neoplasms. So the typical

43:03

processes that are best known to do this would be metastatic disease, specifically,

43:09

especially from breast cancer, as well as other neoplasms that can affect multiple

43:15

organs like lymphoma and other proliferative processes.

43:19

So

43:21

the poster child of diffuse metastatic involvement is breast cancer, particularly

43:27

after treatment. A lot of these areas of metastatic involvement can tend to atrophy

43:32

and give you this pseudocirrhotic appearance of the liver.

43:38

But there are other diseases that can metastasize and cause a diffuse

43:43

liver disease appearance of the liver, as also does melanoma, islet cell or

43:48

neuroendocrine metastases, pancreatic

43:54

adenocarcinomas, and even colon cancers.

43:59

This is an example. By ultrasound, you can see there's a little bit of coarsening

44:03

of the liver architecture. We know how specific that finding really is. Not very.

44:08

But on CT, multiphase CT, you can see similarly some very subtle hypodense,

44:14

hypoenhancing areas in the left more than right lobe of the liver, as well as

44:20

a kind of a nodular contour to the liver.

44:24

And on PET, you can see there's an FDG avidity in this patient with

44:27

pseudomembranous breast cancer.

44:29

This is another patient with diffuse involvement from breast cancer.

44:34

You can see there are multifocal, ill-defined, hypoenhancing masses in the liver

44:40

pre-treatment. But particularly post-treatment, you can see how a lot of these

44:44

areas atrophy and give you the areas of capsular retraction that can mimic a

44:49

cirrhosis. And you can see the PET here two years later shows no FDG avidity

44:55

following treatment.

44:57

So pseudocirrhosis.

45:02

I'm going to lightly touch on this liver pattern.

45:06

This is an old case where you can see on T2 weighted images, kind of a

45:12

heterogeneous T2 intensities in the periphery of the liver, sparing the central

45:16

liver, whereas on post-contrast, you can see some delayed enhancement centrally

45:19

here.

45:21

You can see filling defects in the portal vein and the IVC in this patient with

45:24

Budd-Chiari syndrome. Here's by CT.

45:30

With the late stages, you can get findings that look just like cirrhosis as well.

45:34

The key being that this heterogeneous enhancement pattern, sometimes you'll get

45:39

that flip-flop enhancement pattern between the peripheral liver and the central

45:43

liver. But if you never quite see the hepatic veins

45:48

and/or the IVC, then start thinking about Budd-Chiari syndrome, particularly if you

45:52

have a phase where you should be seeing the enhancement of the hepatic veins or the

45:57

intrahepatic IVC. So Budd-Chiari syndrome is an obstruction of hepatic venous

46:02

outflow, leading to progressive hepatic failure, can lead to manifestations of

46:08

portal hypertension as well, including ascites.

46:11

In the acute phase, you'll see the heterogeneous enhancement pattern as well as

46:15

areas of portal edema and periportal edema.

46:20

And then in the subacute to chronic phases, you basically see findings of

46:24

intraparenchymal collaterals as well as architectural changes that look like

46:29

cirrhosis. And again, the most specific finding is looking for enhancement

46:35

or lack thereof in the hepatic veins or intrahepatic IVC.

46:40

This is something that we see fairly commonly. This is a CT scan with IV contrast.

46:45

You can see there's a giant heart, particularly the right side of the heart is

46:49

enlarged. You've got right atrial and right ventricular enlargement.

46:52

You've got enlargement of the IVC as well as the hepatic veins.

46:57

So this is contrast reflux from right heart failure.

47:00

So this is another cause of a congestive enhancement pattern in the liver, but in

47:05

this case it's not because of thrombosis in the hepatic veins or IVC, but the

47:12

IVC and the hepatic veins are widely patent and more dilated than usual.

47:16

So this is another cause of a relative decrease in hepatic venous outflow.

47:21

But in this case, it's from the

47:24

failure of the right heart, which the IVC empties into.

47:28

So the enhancement pattern that you can see with this, oftentimes, depending on the

47:33

contrast phase that you are imaging the liver in, is this kind of a

47:39

nutmeg appearance. People call this a nutmeg because if you cut a nutmeg in half,

47:42

it kind of looks like this. So you've got these

47:46

areas of heterogeneity in a perivascular distribution, again, from

47:52

congested hepatic sinusoids. So passive hepatic congestion,

47:57

oftentimes related to right heart failure, but

48:01

leading to this mosaic-like heterogeneous enhancement pattern, oftentimes with

48:07

reflux of the contrast from the right atrium into the IVC and the hepatic veins,

48:10

which remain patent. Sometimes you'll see that perivascular edema pattern as well

48:15

on MRI.

48:19

Okay.

48:20

Next disease entity I'm going to show you here is, this is an MRI, and you can see

48:25

there's T2 weighted images on the left here.

48:28

There's coronal and axial T1 post-contrast images here showing you an interesting

48:33

pattern of

48:35

this fluffy,

48:38

perivascular T2 hyperintensities as well as similar subtle areas of

48:44

enhancement, which follow the blood pool in the liver.

48:48

So these represent little congested or dilated hepatic sinusoids

48:55

filled with blood that's enhancing similar to the hepatic vascular trees.

49:00

So this is not a very specific pattern, but it is a pattern that you should

49:05

recognize for something called peliosis hepatis.

49:09

So these areas are little, tiny, globular areas of vessel-like enhancement.

49:15

They can enhance centripetally or centrifugally.

49:18

They tend to show more continuous enhancement to differentiate them from

49:23

hemangiomas. They don't tend to have much mass effect, and they can be either

49:27

diffuse or focal in the liver, and it's related to sinusoidal dilatation,

49:33

multiple blood-filled Kuner spaces.

49:35

The pathogenesis is a bit uncertain, but oftentimes, the

49:41

Bartonellosis in an immunocompromised patient can explain many of these cases of

49:46

peliosis hepatis. Could also be congenital or medication-related, but at least half

49:51

these cases go unexplained.

49:55

And this diagnosis tends to be associated with patients with chronic wasting.

50:03

Okay. Next disease I'm going to show you here, and we're getting into other

50:09

vascular etiologies for diffuse liver processes and changes in

50:15

imaging appearance. We've got a multiphase CT here.

50:18

This is a dual energy CT, so I'm showing you the iodine map here just to accentuate

50:22

the

50:24

changes in the arterial phase enhancement pattern of the liver.

50:27

So you've got something that looks like a cyst in the right lobe, but around the

50:32

remainder of the right lobe of the liver here, you've got a bunch of areas of

50:36

heterogeneous enhancement

50:39

that disappear on the venous phase.

50:42

So this looks like a transient process in the arterial phase, and the key here is

50:47

noting that the portal vein branches in the right lobe are poorly visualized.

50:53

So in this case, if you can follow the portal vein branches on every venous phase

50:58

CT or MRI that you look at, you'll be able to better explain any perfusion

51:04

changes that are related to that area that's poorly perfused by the portal vein.

51:08

So in this case, this is an example of portal vein branch thrombosis leading to

51:13

arterial phase hyperenhancement from compensatory inflow of the hepatic artery.

51:16

So

51:18

you'd think that a thrombosis would lead to hypo-enhancement, but that doesn't

51:22

happen. That's not the case in the liver.

51:25

It's because the liver has a dual main source of perfusion from

51:30

not just the portal vein, but the hepatic artery as well.

51:33

Oftentimes, when you have portal vein thrombosis, you'll get compensatory earlier

51:39

enhancement from hepatic arterial, recruited hepatic arterial inflow, and that

51:45

tends to be transient and best seen in the arterial phase.

51:48

So this is portal vein thrombosis of the right portal vein branch leading to

51:53

transient hepatic arterial enhancement, which is another

51:59

explanation or another cause of a THAD or a THAD, the intensity difference or the

52:03

attenuation difference that's transient.

52:07

This is another case

52:09

where, again, you can see some more subtle hyperenhancement in that right lobe.

52:14

This patient does have cirrhosis, and there are branches of the portal vein that

52:18

are not well-visualized. Sometimes the branches of the portal vein, subsegmental

52:23

branches of the portal vein, may be difficult to visualize because the caliber

52:29

may narrow over time. Best way to detect it, in

52:35

addition to seeing the transient arterial hyperenhancement of the involved area, is

52:40

to compare with older priors if you have them.

52:44

If you can identify a patent portal vein branch on a prior study that's no longer

52:50

present on the current study, then that's probably the best way to diagnose a

52:54

branch portal vein thrombosis and the THAD or THAD that accompanies it.

52:59

This is a not great quality MR, but showing you the same finding

53:05

on that same patient. These areas of non-enhancement in the branches of the right

53:09

portal vein, as well as the lack of signal voids on the T2 weighted

53:15

images involving those same segments.

53:17

So you can look at multiple pulse sequences to confirm this finding on MRI.

53:25

Okay. And then I think this is my last vascular case.

53:32

Here's an MRI of a patient with a history that I am going to withhold.

53:37

We've got some T1 weighted multiphase images here, pre- and post-contrast.

53:43

This is pre-contrast arterial phase, venous, and then a

53:49

hepatobiliary phase in the setting of Yovist or gadoxate contrast use.

53:56

I'm showing you the T2 weighted image in the upper right, as well as a MIP image

53:59

from that arterial phase here. You can see there's multiple little areas of early

54:05

arterial phase enhancements scattered throughout the liver, as well as maybe you

54:11

can see better on the MIP here in the pancreas.

54:15

Sometimes you'll see it in other organs as well, maybe even in the kidneys and in

54:19

the lungs if you had a chest CT. And on the venous phase, you don't see those areas

54:25

quite as well. These areas of transient hyperenhancement tend to blend in with the

54:30

rest of the liver. And then you see contrast retention on the hepatobiliary phase

54:36

in multiple nodular areas throughout the liver.

54:40

So if you saw this gadoxate retention in the hepatobiliary phase, you'd be

54:45

thinking about focal nodular hyperplasia.

54:49

And focal nodular hyperplasia is one diagnosis, but one of the causes of focal

54:54

nodular hyperplasia, or supposedly the main etiology for focal nodular hyperplasia,

55:00

is arteriovenous malformations underlying the FNHs.

55:06

In this case, this patient had a known history of hereditary hemorrhagic

55:10

telangiectasia, and these are multiple little areas of telangiectasia involving

55:14

multiple organs, including the liver, leading to focal nodular hyperplasia.

55:22

All right. So that was a whirlwind through a bunch of diffuse liver processes, some

55:28

of them more specific than others. We went over cirrhosis briefly.

55:33

We went over metabolic and storage-related etiologies of diffuse liver disease,

55:38

including Steatotic liver disease.

55:41

We didn't get into all the different subcategories of steatotic liver disease.

55:44

That's another whole lecture in and of itself.

55:46

But

55:48

I want to remind you what the current terminology is. That was recently updated.

55:51

It's now known as metabolic associated steatotic liver disease, previously known as

55:56

non-alcoholic fatty liver disease, as well as metabolic dysfunction associated

56:00

steatohepatitis. There's a metabolic associated alcoholic liver disease, as well as

56:06

your old-fashioned alcoholic liver diseases.

56:10

There is causes of iron overload divided into two main categories.

56:15

Hereditary hemochromatosis or primary hemochromatosis, which affects just the

56:20

liver, maybe pancreas, maybe myocardium.

56:23

Versus hemosiderosis, which is reticuloendothelial deposition of iron in the liver,

56:30

spleen, maybe bone marrow. We've got drug-related processes, particularly iodine

56:35

deposition from amiodarone use, chronic amiodarone use.

56:39

Got copper deposition from Wilson's disease, as well as diffuse liver processes

56:44

like amyloidosis and glycogen storage diseases, which are less specific by imaging.

56:52

But good to recognize if you know the clinical history.

56:55

There's infectious causes of various types of hepatitis.

56:58

I showed you one and many appearance of infectious liver process,

57:04

schistosomiasis. And then we went into a little bit of sarcoidosis, as well as

57:09

mimickers of diffuse liver disease, like diffuse neoplastic processes.

57:15

And then we went over the spectrum of vascular findings in the liver that can lead

57:21

to vascular patterns of disease.

57:25

Some of them related to the third inflow phenomenon, leading to areas of focal

57:29

fatty infiltration or focal fatty sparing, as well as other vascular etiologies of

57:35

transient arterial perfusional changes in the liver from a bunch of causes like

57:42

pas de deux hepatic congestion and branch portal vein thrombosis.

57:47

So with that,

57:49

diffuse liver disease is definitely, I find it more challenging than focal liver

57:53

masses. But definitely it's important to recognize these because they

58:00

have important prognostic implications as well as associated risk factors for

58:05

malignancy for some of these processes.

58:08

It's important to increasingly recognize the metabolic syndrome and the

58:13

risks that this

58:16

leads to. And a lot of these also have cardiovascular risks as well.

58:22

So with that, I would be happy to answer any questions you may have.

58:27

Thank you so much for that really in-depth lecture and case review.

58:31

Appreciate that. There are some questions in that Q&A box if you want to pop that

58:35

open.

58:36

Okay.

58:37

It might be at the top of your screen right now.

58:40

So I see four major questions.

58:44

One question is, what is my protocol for measuring hemochromatosis?

58:48

I

58:49

went over our measurement of iron deposition.

58:54

I'm not sure if that's the question you have, but again, right now we're

58:59

using some quantitative MR sequences that differ by vendor, but

59:06

all of them basically will give you an R2* map that you can put an ROI on

59:12

and measure the R2* relaxation time, which we plug into

59:18

that equation that I showed you just to spit out of the iron concentration.

59:22

Or if you don't have access to those quantitative MR sequences, basically you can

59:27

just do a grading echo sequence with eight to 12 different echo times, successively

59:33

longer echo times, to monitor the signal degradation over

59:39

longer echo times, as well as to separate that signal drop effect from

59:45

iron versus the signal drop from the in and out-of-phase effects of fatty

59:51

deposition. And that way, you can actually measure the iron concentration as well

59:55

as the fat fraction. Second question was, how long should you wait after

59:59

ferrioxamine administration before obtaining MRI?

60:03

We don't use a specific cutoff time. I would say the longer the better.

60:07

Maybe at least a couple of months, that these effects can be pretty longstanding,

60:12

but

60:14

we don't have a specific guideline, but that's an excellent question.

60:17

We can certainly look that up online to see what other places are doing.

60:21

We don't encounter it all that often. So we haven't developed a policy around it.

60:28

Another question is about whether we use a T2* sequence for liver iron load.

60:33

You can, and certainly if you can get an R2*, you can get a T2* as well, but the

60:37

formulas that were published are R2* ones, or at least the ones that we're using at

60:42

our institution are related to R2*. But there are ways to do it by T2* as well.

60:47

I would have to look up those formulas.

60:54

And then another question is, why does atrophy only occur in certain segments of

60:58

the liver? Good question. That just seems to be the pattern of disease.

61:01

I'm sure there's an answer to that. I'll have to look that up.

61:05

I'm a pattern recognizer. I'm not great at

61:10

knowing exactly what the etiology for the areas of the relative atrophy are, but

61:15

I'm not sure if it's perfusion related or not.

61:21

And then next question is, does the degree of dropout on in-and-out-of-phase

61:25

imaging correlate to the percentage of steatosis? It can, yes.

61:28

There's certainly some formulas that you can use to try to calculate it, but I

61:33

don't think having just two phases alone or two echo times alone is sufficient to

61:38

really give you a good measure because

61:42

just having two echo times alone doesn't always separate the effects of iron

61:47

deposition from the effects of

61:50

fatty infiltration. So that's the reason why a lot of these sequences

61:56

use way more than two echo times, I think eight to 12 echo times even.

62:01

So while it's feasible and there are some formulas out there, I don't think it's as

62:05

accurate as actually doing the proper proton density fat fraction measurements.

62:11

And then another question is, how do you quantitatively measure hepatic steatosis

62:15

at MR? Good question.

62:17

There was a slide-- I didn't show you that the same quantitative MR sequence that

62:22

we use for iron quantification will also spit out a fat fraction map.

62:27

So we're basically looking at the proton density fat fraction map and just

62:32

generating an ROI over the liver. And it gives you a percentage fat

62:38

fraction for the liver. So that's how we quantify the degree of fat

62:43

by MR. So that's basically what we do.

62:48

I'll usually put in an ROI in the succu fat too, just as a control.

62:52

You know you're looking at the right sequence when you put an ROI over the fat and

62:55

it's close to 100, and then you put it in the liver and you can generate a

62:59

percentage, which you can look up in the tables on that paper that I referenced

63:04

earlier in the talk to quantify or correlate the

63:10

percentage to the relative severity of steatosis.

63:17

And then, yeah, the last question we have here is how would a liver with both

63:21

steatosis and iron deposition behave on in-and-out sequences?

63:24

So that depends on whether there's more fat or more iron, right?

63:27

That's why two echo times alone is not really sufficient to isolate one versus the

63:33

other. And maybe the iron would cause a signal drop on both the in and

63:39

out-of-phase.

63:41

And if there's a lot of fat in addition to the iron, then you may not be able to

63:46

appreciate as much of the iron, the signal drop on the in-phase versus the

63:49

out-of-phase. So it's an excellent question, and the reason why we break out those

63:55

two quantifications into separate

64:00

quantification sequences like the R2* map versus the proton density fat fraction

64:06

map. And I would refer you to the paper I showed you before for more

64:11

details in terms of the physics behind how that works and how they generated the

64:18

values using historical

64:22

biopsy results.

64:25

And I think that's it for our questions.

64:27

I think you got them all. Thank you so much. Thanks so much for being here, Dr.

64:31

Chang, and giving that excellent lecture.

64:32

Oh, thank you. My pleasure.

64:34

Yeah.

64:34

Thank you for the opportunity to share.

64:36

Of course. Yeah. So thank you for everyone else for participating in our new

64:39

conference and asking such great questions.

64:41

Be sure to join us next week on Thursday, October 15th at 12:00 PM Eastern, where

64:45

Dr. Lacey McIntosh will deliver a lecture entitled Atypical Lung Cancers.

64:49

You can register for that at medallia.com and follow us on social media for updates

64:54

on future new conferences. Thanks again for learning with us and have a great day.

Report

Faculty

Kevin J. Chang, MD, FACR, FSAR

Section Chief of Abdominal Imaging & Director of MRI

Boston University Medical Center

Tags

Gastrointestinal (GI)

Body