0:00

Dr. Finazzo, we're on to vascular lesions.

0:01

3 00:00:04,700 --> 00:00:07,439 And now MRI is going to afford us the ability

0:07

to do some histopathologic analysis, some tissue

0:10

typing, and look at some biomarkers

0:13

of cancer, and try and differentiate some of

0:16

the important malignant lesions of the kidney.

0:18

So where do you start?

0:20

We've given contrast here from early to late.

0:23

We've got a T2 on the top.

0:24

This is a contrast-enhanced

0:26

on the far right, and an in-phase image.

0:29

And, I'm sorry, an in-phase and an out-of-phase image.

0:32

Yep.

0:32

So where do we go from here?

0:33

So usually how I start to approach these lesions

0:33

18 00:00:35,879 --> 00:00:38,889 is first I go to my T2 weighted images and

0:38

try to see exactly what the lesion looks like.

0:41

And here on the T2 image,

0:44

it is a heterogeneous lesion.

0:47

So we're not dealing with just a pure cystic

0:49

lesion or a T2 dark lesion, rather heterogeneous.

0:52

And this happens to be a female, 60 years old.

0:56

So first things we think about

0:57

in women is: Could this be an AML?

0:59

So now we're going to try to do

1:00

our exercise looking for fat.

1:02

So to look for both bulk fat and

1:04

intravoxel fat, we go to our in-phase and out-of-phase images.

1:07

So looking at the in-phase and out-of-phase,

1:08

can you identify any bulk fat?

1:11

Well, I certainly can't, but

1:13

I'm not as expert as you are.

1:15

Here's the in-phase, and here's the out-of-phase.

1:17

What we would like to see is substantial

1:20

signal dropout, and maybe an India ink sign.

1:24

Um, neither of which we see.

1:28

So that doesn't help us.

1:28

That doesn't help us.

1:29

That doesn't help us, but we have excluded

1:32

the possibility of this being an AML.

1:34

So now we're leaning towards it being a neoplasm.

1:38

So now we'll go ahead and look at our pre- and post-

1:41

contrast images to see if we can see enhancement.

1:44

So we go to our pre-, our 42-second, our 90-

1:48

second, and our three-minute delay, and what is

1:51

the enhancement pattern that we see is: we see

1:54

a strongly arterial enhancing focus, which

1:58

maintains its contrast on the delayed phase

2:01

imaging, leading us to the hypervascular family of

2:07

lesions, of which I would then consider

2:11

clear cell carcinoma as my number

2:13

one, as being the most common.

2:16

AML would be less likely

2:17

because we did not see any bulk fat.

2:19

One thing we did do is we scrolled up and down.

2:21

We didn't just use one image.

2:23

We scrolled up and down to make sure that we didn't

2:25

have segmental fat, which you can certainly have in

2:28

any one of these lesions, but particularly in AML.

2:32

So, no fat, very hypervascular, it looks

2:35

like it's early, and it persists.

2:39

There may be a little bit of a washout at the

2:40

end, but we haven't really analyzed that, but

2:42

it certainly shows up in the arterial phase.

2:45

So, as you mentioned, AML, renal cell

2:47

carcinoma, chromophobe tumors, and then

2:51

the benign lesion, oncocytoma, which kind

2:53

of overlaps with the chromophobe tumors.

2:55

All of those are hypervascular.

2:57

And for that point, might be

2:59

in the differential diagnosis.

3:00

That's exactly, and this case is very obvious,

3:03

but when we're trying to look for subtle areas

3:06

of enhancement, and if we are not able to

3:09

identify it in our multi-phase scan, that's

3:12

when we really want to look at our subtractions.

3:15

And when we look at our subtractions, just

3:17

understand that there are two basic pitfalls.

3:20

Pitfall number one is any misregistration.

3:24

So what I typically like to do is I like to

3:26

go to the liver and make sure I don't see

3:29

that big rind around the liver because if I do,

3:32

then all bets are off for subtle enhancement.

3:35

This one obviously, we don't see that

3:37

thick rind, so there's probably not

3:40

misregistration, and this is true enhancement.

3:42

Then I also look at the noise in the

3:45

background, and just remember, noise is additive.

3:48

So if we see a lot of background noise, that could

3:52

also be misleading and give us pseudo enhancement.

3:56

Not in this case, though.

3:58

And, you know, when you say misregistration,

4:00

what you mean is a bad subtraction.

4:02

That's correct.

4:03

They didn't register it properly.

4:05

You're always gonna see, for the newbies

4:07

and youngins in the audience, you're always gonna see

4:09

a little bit of signal anteriorly due to respiration,

4:13

but the subtraction here has been done beautifully.

4:15

Another area of misregistration that can sometimes

4:18

be confounding is the aorta can sometimes

4:21

smear side to side or front to back, depending

4:23

upon how the phase and frequency are arranged.

4:26

And you can also get peristalsis that projects

4:29

over the renal cell carcinoma that simulates

4:31

an area of hypervascular enhancement.

4:34

So where do we go from here?

4:35

So you were asking, you were talking

4:37

about the differential, and again, our

4:38

goal is to do some histologic subtyping.

4:41

So a lesion this size.

4:43

Sometimes people would say, oh gosh, could this be

4:45

an oncocytoma or a chromophobe neoplasm.

4:50

And so the question is how can we try to

4:53

help narrow down the differential in these subtypes?

4:58

So the first thing I do is I look at T2.

5:01

T2 is the key.

5:03

90 percent of the time, chromophobe

5:07

neoplasms, or chromophobe renal cell cancers

5:10

are hypointense on T2 weighted images.

5:14

So when you see, even though

5:15

they're hypervascular.

5:16

Even though they're hypervascular, which we're

5:18

going to go to next, we'll talk about hypo

5:20

T2 dark lesions.

5:25

But oncocytoma is pathognomonic, would be

5:28

90 percent of the time you'll see T2 dark.

5:30

So the fact that this is T2 bright doesn't

5:33

help us in the differential, but if it was

5:36

dark, you could throw that in the differential.

5:38

Secondarily, you can look for a central scar.

5:42

Oncocytomas are known to have a central scar.

5:47

It doesn't happen all the time, but

5:49

when it does, you can clear that.

5:52

Central necrosis.

5:53

Central necrosis you can see in all lesions.

5:56

Both renal cell cancer, oncocytomas,

6:01

and chromophobes, so that doesn't help you.

6:03

So T2 is really key, and an article was

6:06

just written last year describing that.

6:09

The, the third point, or the third, uh,

6:15

segmental enhancement inversion.

6:19

And what segmental enhancement inversion

6:21

is looking at the enhancement pattern of

6:24

a lesion during the arterial phase

6:28

and then looking at it in the delayed phase.

6:30

And what that means is, when something strongly

6:33

enhances in the arterial phase, that area

6:37

will de-enhance on the delayed phase, and

6:41

the area that's hypo-enhancing in the early

6:45

phase will enhance in the delayed phase.

6:49

And that's seen only in oncocytomas.

6:52

So I'm going to illustrate that for you.

6:53

That's a great point.

6:55

Of course, I'm a little older and more

6:57

simplistic, so I'm going to use my little

6:58

drawing tool here, and I made a pie, a pizza pie.

7:02

Let's take one segment of the pizza pie.

7:03

Let's take this segment.

7:05

This is the hypervascular segment.

7:06

So in that segment, the contrast

7:08

comes early and then it washes out.

7:12

Now let's go to another segment of the pizza pie.

7:14

Let's go to this yellow segment, which was

7:16

initially hypovascular and keeps on coming.

7:20

So while this one's disappearing over

7:22

here, this one's getting brighter.

7:24

So one segment comes up and goes down

7:26

quickly, the other segment continues to rise

7:30

very slowly, and their behavior inverts.

7:32

This one was bright initially,

7:34

now this one is bright late.

7:36

And that's what you mean by segmental inversion.

7:37

That's exactly right.

7:38

Great.

7:39

So let's move on, shall we?

7:40

Mm-hmm.

7:40

Okay.