0:00

Well, we're about an hour into this course and all

0:04

we've done is go through the anatomy.

0:05

But this is a mastery course after all.

0:08

So you've learned quite a bit of the anatomy of the

0:11

major salivary glands, those including the parotid,

0:13

the submandibular and the sublingual glands,

0:16

as well as the minor salivary gland tissue.

0:19

Let's talk very briefly about scan sequences

0:23

and how we actually image it.

0:25

I've pointed out a few of these cases thus far

0:28

in which we've looked at the technique.

0:32

Now, at my institution at Johns Hopkins,

0:35

we collect and display all of the thin

0:40

section images of multidetector CT.

0:43

So in our analysis of the head and neck region

0:48

or the salivary gland region,

0:50

we will obtain 0.6 millimeter thick slices,

0:58

and we get the whole body of the anatomy at

1:02

0.6 millimeter slices available to us.

1:04

That usually runs about 700 images.

1:08

However, the technologists then reconstruct those images

1:13

in 3 millimeter bytes that we can

1:15

look at for a quick overview.

1:19

So if the 3 millimeters are not of sufficient

1:22

detail, we go to those thin section raw data,

1:27

0.6 millimeter thick slices.

1:31

With those 0.6 millimeter slices,

1:33

obviously, we can create beautiful coronal

1:36

and sagittal reconstructions ourselves.

1:40

However, the technologist provide to us multiplanar

1:43

reconstructions in the coronal and sagittal plane

1:46

at the 3 millimeter thick reconstructions.

1:49

So you get the axial, coronal and sagittal.

1:52

When do we need contrast?

1:55

So if the question is about calcifications

2:00

and calculi,

2:01

you really don't need contrast-enhanced images.

2:04

We can just use non-contrast CT scans to identify

2:08

calcifications and calculi.

2:10

However, if we're looking at whether or not

2:13

the gland is inflamed,

2:14

it's useful to have post-contrast scans that can be

2:18

compared with the pre-contrast scans and can be

2:20

compared side to side to see whether the gland is

2:25

enhancing more on the one side, the pathologic side

2:29

versus the non-pathologic side to suggest that there

2:33

is indeed inflammation of the gland.

2:37

When we are looking for neoplasms,

2:39

however, with CT scan,

2:40

we are obviously giving contrast

2:42

enhancement and in that situation

2:44

we do not do pre-contrast,

2:48

non-contrast imaging.

2:50

We go straight to the contrast-enhanced CT scan

2:53

with the typical sort of venous phase imaging,

2:58

which allows both the

3:00

anatomy of the arteries and the

3:01

veins to be identified,

3:02

as well as to see whether the neoplasm

3:05

imbibes any contrast itself.

3:07

And that also is helpful for inflammatory lesions

3:10

where we're looking for abscesses,

3:11

for example.

3:13

I showed an example of CT sialography.

3:16

I'll do a mea culpa.

3:18

I haven't done sialography in probably 10 years

3:22

because the CT and the MRI scans usually

3:27

solve the questions themselves.

3:29

So we don't do CT sialography except

3:32

in exceptional cases nowadays.

3:35

Let's move to MR of the salivary glands.

3:38

Well,

3:39

since we are doing such thin sections with CT scan,

3:43

there has become a competition, if you will,

3:47

for higher resolution on our MR studies.

3:50

So, you probably are doing 3-5 millimeter

3:56

thick sections for your neck MRI studies routinely.

4:02

However, at Johns Hopkins,

4:06

we will often include thin

4:10

section 0.6 millimetre MR images.

4:14

Why? Because we have to compete with CT, right?

4:17

We want to have as good a resolution as with CT.

4:20

In order to achieve those 0.6 mm thick sections,

4:24

you have to do thin section

4:28

CISS or VIBE imaging.

4:30

These are Siemens terms

4:33

for the constructive interference in the steady state.

4:37

And I've forgotten VIBE,

4:40

but it's T1-weighted sequences,

4:43

and these are also called FIESTA on

4:49

some of the other manufacturers,

4:51

and MP RAGE or MP GSBR for the VIBE option.

4:57

So that allows us to get submillimeter thick sections.

5:00

Routinely, these are your images,

5:02

3 mm thick T1-weighted.

5:04

You need good quality fat-suppressed

5:07

T2-weighted imaging.

5:08

That usually requires inversion recovery sequences,

5:12

either fast spin-echo or STIR sequences with good fat

5:16

suppression and post-gad, T1-weighted scans.

5:20

These may also be inversion recovery.

5:22

Some people do in-phase,

5:23

out-of-phase, Dixon type, post-gad scans

5:27

with 3 mm thick sections.

5:30

You can just do pre-imposed VIBE and

5:34

get thin section images.

5:35

Again, the value of the VIBE imaging and the CISS imaging is

5:38

that you could do reconstructions in the sagittal and

5:40

coronal planes that are very high-quality.

5:44

MR sialography does not require

5:47

contrast injection. In this situation,

5:50

it's a very highly T2-weighted scan,

5:52

very much like a CISS sequence with suppression of

5:56

the tissue around, in which case you can see

6:00

the ductal system very nicely.

6:02

MR sialography has been stolen, if you will, from MR

6:07

cholangiopancreatography.

6:10

So this is basically the same pulse sequence

6:12

that you do for your MRCP.

6:14

You're going to apply to the salivary glands

6:16

in order to see ductal pathology.

6:19

It's pretty uncommon that we do MR sialography.

6:22

That's usually for patients who have chronic sialadenitis ,

6:25

often patients who have, for example,

6:27

Sjogren's syndrome or whatnot.

6:30

So not a frequently utilized application for MRI.

6:36

We have embraced diffusion-weighted

6:38

imaging in the salivary glands.

6:42

Low ADC,

6:44

manifested as bright signal in the DWI sequence,

6:47

is commonly associated with parotid malignancies.

6:51

There is some overlap with Warthin's tumor, however,

6:56

but we routinely now do DWI.

6:59

It also sometimes will be helpful for

7:01

you in looking at lymphadenopathy.

7:04

If we see lymph nodes who have...

7:06

which have reduced ADC,

7:08

we're more likely to suggest that they are metastatic

7:11

lymph nodes than those who have high ADC and are,

7:15

and are more likely to be inflammatory.

7:19

Perfusion-weighted imaging has been advocated in

7:21

academic circles also to distinguish benign versus

7:25

malignant neoplasms in the parotid gland,

7:29

predominantly or inflammatory lesions

7:32

from neoplastic lesions.

7:33

And I'll demonstrate some of the

7:35

findings on that as well.