0:00

Our technique for evaluating the spinal cord

0:03

is similar to that which was described in the

0:07

discussion of demyelinating disorders

0:10

of the brain and spine.

0:12

That is, we rely on sagittal T1-weighted,

0:16

sagittal T2-weighted, sagittal STIR images,

0:20

axial gradient echo scans for the cervical

0:22

spine, which, as you recall,

0:24

I said all often shows demyelinating plaques

0:26

better than axial fast spin echo T2-weighted scan.

0:31

However, if we were just scanning the thoracic spine,

0:35

we would do it without the gradient echo sequences.

0:39

Most instances of cord lesions that cause

0:43

myelopathies, we will administer gadolinium.

0:47

And this is usually scanned in the sagittal and axial plane.

0:51

Sometimes, we will apply fat saturation

0:54

techniques for the axial plane.

0:57

Now, some people have advocated using the Vibe

1:00

technique, which allows us very thin section

1:03

T1-weighted scans, both pre and post contrast for

1:09

excellent depiction of the anatomy and more

1:12

subtle depiction of gadolinium enhancement.

1:15

As of right now, this is still being used only in academic

1:19

centers rather than widely,

1:21

and we just don't know how good Vibe is as a

1:25

replacement for standard spin echo T1-weighted

1:29

post-gadolinium enhanced sequences.

1:32

Now, there are certain special circumstances with

1:35

which we will scan patients using

1:38

other pulse sequences,

1:40

so we do not routinely do diffusion weighted

1:43

imaging for the spinal cord.

1:45

However, in that instance,

1:47

where one is concerned with ischemic

1:49

injury to the spinal cord,

1:51

you can apply diffusion weighted imaging,

1:54

or in most instances, diffusion tensor imaging

1:58

and still have an ADC map that can identify

2:03

spinal cord infarction.

2:06

Similarly, if one is suspecting a vascular malformation

2:11

of the spinal cord, you can do MRA of the spinal cord.

2:16

This is quite difficult and has been best

2:19

described by publications by Brian Bowen

2:22

and the University of Miami group.

2:25

If you want to do those sequences,

2:28

I highly recommend you try to duplicate

2:30

their pole sequences.

2:32

Part of the best protocol is a dynamic twist

2:36

sequence, in which you inject the contrast and

2:40

scan at the same time in order to see the

2:43

different phases of the arterial venous phases of the MRA.

2:50

This may help you in better defining

2:52

the vascular malformation.

2:54

Because the spinal cord is relatively narrow

2:57

in a coronal plane, we usually do the

3:00

MRAs in a coronal plane using that twist MRA technique.

3:05

This is a demonstration of a patient who has

3:09

an intramedullary lesion, which is seen quite

3:14

nicely on our T2-weighted scans.

3:16

This being the T2-weighted scan

3:18

and this being the STIR scan,

3:22

and not as well demonstrated on T1-weighted scan

3:27

and not showing enhancement on

3:30

our gadolinium enhanced sequence.

3:33

But these are the stereotypical sequences that

3:36

we would do, T1-weighed, T2-weighed, STIR.

3:40

And because this is in the cervical spine,

3:43

we will perform the gradient echo scan.

3:47

If it's anywhere else,

3:49

we're just doing T2-weighted fast spin echo axial scans

3:53

to show the spinal cord.

3:56

And this is the T1-weighted post-gad in the axial plane.

4:05

So this is our workhorse series of seven pulse

4:08

sequences for evaluating cord lesions.

4:13

Here are the two special sequences that I described.

4:17

Here we have on the left, the DWI and the ADC map.

4:25

Now, this DWI may be performed as a diffusion

4:28

tensor imaging sequence rather

4:31

than a straight DWI sequence,

4:33

depending upon your software package.

4:35

And this is a normal looking cord.

4:37

You notice that the quality of the images

4:40

is somewhat decreased.

4:42

However, all we're really looking for is a bright area

4:46

within the spinal cord to identify a cord

4:48

infarct. So we don't need high resolution,

4:51

we just need to see whether it's restricting

4:54

diffusion. And obviously, if it was bright on the DWI,

4:58

just as in the brain,

5:00

you're going to see decreased signal intensity on

5:02

your ADC map. To the right, we have the raw data

5:08

and the reconstructed data and an axial

5:13

reconstructed image of a patient who has a

5:18

vascular malformation of the spinal cord, and

5:21

this is demonstrating a varix in the spinal canal.

5:28

So this would be our twist sequence.

5:31

In this instance, we would see both the

5:33

arterial, as well as the venous and

5:35

then the delayed venous phases on the twist sequence.

5:39

We actually see the contrast coming in.

5:41

You notice that we have pulmonary arteries

5:43

and aorta all showing up, and this is a

5:46

reconstructed image with also the intercostal

5:49

arteries being demonstrated.

5:51

So, very nice looking study.

5:53

May not have been done at Hopkins.