Cerebrospinal Fluid Dynamics - Leaks and Communicating Hydrocephalus, Dr. Jeffrey Scott Pannell (6-1-23)
HIDEInteractive Transcript
0:02
Hello and welcome to Noon Conference,
0:03
hosted by M R I Online Noon Conference connects the global radiology
0:08
community through free live educational webinars that are accessible for all and
0:13
is an opportunity to learn alongside top radiologists from around the world.
0:17
We encourage you to ask questions and share ideas to help the community learn
0:20
and grow.
0:21
You can access the recording of today's conference and previous noom conferences
0:25
by creating a free m r a online account.
0:28
You can also sign up for a free trial of our premium membership to get access to
0:32
hundreds of case based micro-learning courses across all key radiology
0:35
subspecialties. Today we are honored to welcome Dr.
0:39
Jeffrey Scott Pinnell for a lecture on cerebral spinal fluid dynamics,
0:44
leaks, and communicating hydrocephalus. Dr.
0:47
Pinnell is a board certified endovascular surgeon and interventional
0:51
neuroradiologist and director of Neuro interventional surgery in the Department
0:55
of Neurological Surgery at uc, San Diego Health. Dr.
0:59
Pinnell specializes in the minimally invasive catheter-based treatment of blood
1:02
vessel disorders that can lead to hemorrhagic or ischemic strokes and treatment
1:07
of pain disorders of the spine and severs spinal fluid leaks,
1:11
as well as diagnostic neuroradiology.
1:13
We're thrilled he's here today to share his expertise with the radiology
1:16
community. At the end of the lecture, please join Dr.
1:19
Pinnell in a q and a session where he will address questions you may have on
1:23
today's topic.
1:24
Please remember to use a q and a feature to submit your questions so we can get
1:28
to as many before our time is up. With that,
1:30
we're ready to begin today's lecture. Dr. Pinnell, please take it from here.
1:34
Thank you for that wonderful introduction.
1:36
I appreciate you all being here with me this morning.
1:39
Lemme just get my screen situated here really fast and we'll get going.
1:45
So, um, we are gonna be talking about, um,
1:47
surface spinal fluid mechanics and disorders of C S F flow dynamics.
1:51
We're primarily gonna be focusing on spinal C S F leaks and
1:55
communicating hydrocephalus, although I may show other cases just to, uh,
1:59
present basically a counterpoint, um, and, and part of the differential.
2:03
So I have some modest consultancies that primarily pertain to my vascular
2:07
practice. I don't have any conflicts pertaining to, um,
2:11
the treatment of C S F leaks or C s F flow disorders.
2:15
In this presentation,
2:16
I'm gonna provide you guys with an overview of the presentation,
2:18
etiologies and findings and spontaneous intracranial hypotension and
2:21
hydrocephalus.
2:22
And I'm gonna discuss the role of m MRI and screening and assessment of
2:25
intracranial hypotension and hydrocephalus. Um, just to kind of get started, um,
2:30
just kind of review C SF flow. I think most everybody knows this,
2:33
but it's good to review it. Um,
2:35
predominantly most of the ceal spinal fluid is produced in the lateral
2:39
ventricles by the cho plexus. Um,
2:41
large concentrations of the cord plexus are located within the atria of the
2:46
lateral ventricles here posteriorly. Um,
2:48
that is basically secreted and flows through the, um,
2:52
framing of Monroe into the third ventricle. And then from the third ventricle,
2:56
it flows through the cerebral aqueduct into the fourth ventricle.
2:59
And then from the fourth ventricle,
3:00
the surface spinal fluid flows out the fara of e lusca and the Fara of magdi at
3:05
the base of the brainstem.
3:06
And then it flows up and down the spinal column aided by the pulsations of the
3:11
brain. Um, once it is produced, eventually it is reabsorbed. We make and, uh,
3:15
reabsorb about 500 milliliters a day. Um,
3:18
just to go back one slide and show you a couple of other little findings.
3:21
These little areas of T2 hyperintensity within the skull near the
3:26
venous sinuses are called oid granulations.
3:28
That's where the cerebral spinal fluid is reabsorbed.
3:31
So essentially the cerebral spinal fluid is made within the ventricular system,
3:34
excreted flows over the surface of the brain and spinal cord,
3:37
and then eventually is reabsorbed in the arachnoid granulations.
3:41
So let's start with spontaneous intracranial hypotension. Um,
3:45
it's basically just a decrease in hydrostatic pressure that results in a
3:48
postural headache. Most ti most of the time patients will denote that they, um,
3:52
have a headache that gets worse after about 10 or 15 minutes of sitting up gets
3:56
better after lying down for a similar period of time.
3:59
It can result from a C S F leak, excess reabsorption, or excess production.
4:05
Um, and essentially spontaneous leaks can occur in the spine due to aural tear,
4:10
um, or degenerative spondylosis or even just dact,
4:13
tasia resulting in potentially a c SF venous fistula, which we'll cover later.
4:17
Um,
4:18
spontaneous skull base leaks can be due to an osseous de hyphens or an cephalic
4:22
seal, um,
4:23
and most frequently occurs in the temporal bone and it's usually associated with
4:27
oea, uh, or rhinorrhea.
4:28
And sometimes there are also associated perinasal sinuses, ef perinasal,
4:32
sinus effusions, um, particularly the,
4:34
the math to an air cells will fill up when it's the temporal bone. Again,
4:37
we're gonna primarily be focusing on spinal CSF leaks today.
4:42
Typically, um, the patients again present with postural headache.
4:45
A lot of 'em will, uh, complain of nausea, vomiting, neck pain,
4:47
visual disturbances, and essentially, um,
4:50
cranial lower cranial nerve issues like tinnitus and,
4:53
and vertigo and things like that. Um,
4:56
generally clinically you're gonna note that their opening pressure when you do
5:00
an LP is usually less than six centimeters of water. Um,
5:03
ideally this would be done in the lateral decubitus position to get the most
5:06
accurate pressure possible. Um, and essentially, um,
5:10
if they have a skull base leak,
5:11
they should have some history of clear OEA or rhinorrhea. If it's,
5:14
if it's the skull base that we're talking about, um,
5:17
most of them will have pacu, menal thickening and some degree of brain sag,
5:20
if not, um, full-blown tonsil or ectopia. Um, and then, um,
5:24
most of them will also have some degree of pituitary engorgement or, um,
5:27
decreased, uh, um, width of the super cellar cistern.
5:30
A lot of them will also have venous sinus distension.
5:33
Some of them will actually even have full-blown subdural collections and
5:36
superficial cirrhosis. Sometimes the ventricles can even start to look slit,
5:39
like very rarely when you have a very high flow CS leak.
5:42
People can have a full minute course where they actually develop cerebral edema
5:45
and,
5:46
and go into a coma and require a very rapid treatment and correction of this
5:49
problem. Um, initially the way that I normally screen a patient, um,
5:54
that comes into my clinic is I get an m r I of the brain with contrast, uh,
5:58
or with, and without contrast, I should say,
6:00
and we get thin section cts in the temporal bone and or MAC space,
6:03
depending on whether or not they complain of ear stuffiness or rhinorrhea oea or
6:08
if they have fluid behind their eardrum. For instance, when I examine them,
6:11
if there's no dehiscence on the ct,
6:13
then I do a full spine survey with flow compensated fiesta images and fat
6:17
saturated two T2 weighted images. Um,
6:20
and essentially at that point I make a decision between doing a more rapidly
6:25
acquired myelogram or doing a pressure augmented myelogram. And really it's,
6:29
it's more of an imaging, um, uh, acuity,
6:32
like how quick is the contrast gonna leak from the fecal sac?
6:35
And MRI is pivotal in deciding what the next step is. Basically,
6:39
if I'm seeing a pseudo ine where there's an actual, uh,
6:43
collection of fluid outside of the dura, normally those,
6:46
those leaks actually happen very quickly and the communications is very rapid.
6:49
Whereas if I'm seeing perineural sts,
6:51
it's usually more of a slower communication,
6:53
either a CSF venous fistula or a small type two leak. In those cases,
6:57
I'm gonna opt more for pressure augmentation and lateral decubitus positioning,
7:00
which we'll get into a little bit later. Um, if I think it's a skull base leak,
7:04
again, I'll do a nuclear medicine cystogram,
7:06
but that's not really gonna be the focus of the lecture day.
7:07
But that's normally what I would do.
7:08
I do a nuclear medicine syn agram if I think there's a dehiscence and just
7:12
confirm it and make sure that we're not just dealing with sinus disease. And,
7:14
and, um, a patient who has, uh, con commitment C S F hypotension,
7:20
um, one of the things that comes up often is, uh, uh, the burn score.
7:24
So the burn score is actually a screening score scale that we can do for
7:27
patients based on their m r and their are minor criteria and major
7:32
criteria for CSF hypotension or spinal CSF leaks. Um,
7:36
engorgement of the sinuses, pacu, menal enhancement and super cell cistern. Um,
7:41
uh,
7:41
flattening or or CLO closure of the super cell cistern down to less than four
7:45
millimeters, um, is a major criteria.
7:47
So you're talking about two points for each of those in the minor criteria or
7:51
having, um, a subdural collection,
7:53
a prep pontine cistern distance of less than five millimeters or a mammolo
7:56
pontine distance of less than 6.5.
7:58
So this burn score can kind of help put you put the patients into risk
8:01
categories. If their score is less than two, we,
8:04
we generally consider them low risk intermediates between two and four,
8:07
and then greater than five, uh, points on the burn scale, uh,
8:10
is the highest risk, obviously for a spontaneous C S F leak,
8:13
and in most cases they are spinal leaks. Um, patients who have, uh,
8:18
cranial leaks generally don't have as many of these elements on the burn score
8:22
particularly, they don't tend to have the brains bag component of it and their,
8:26
their PACU enhancement can be more focal surrounding wherever the leak is as
8:31
opposed to being more diffuse. And that's,
8:32
that's not an uncommon finding when it's a cranial C S F leak.
8:35
So this is a patient with a typical spinal C S F leak.
8:38
We can see that they have, um, uh,
8:40
diffuse pacu menal enhancement and thickening. They have subdural collections.
8:45
Their pituitary is engorged, their straight sinus is engorged. Um,
8:50
their super cell cistern is very, very narrow, much less than four millimeters.
8:54
It'd be very difficult to even measure because the,
8:56
the size of the pituitary theonine distance or prep pontine cistern wit is very
9:01
narrow as well. You can al you, you almost can't even measure it.
9:03
And then the mammolo pontine distance is almost completely closed as well.
9:08
In addition to that, we have an acquired pular ectopia.
9:10
So this is a patient who has a, basically a nine on the burn score.
9:14
There's really, there's there they have, they cover every single point.
9:16
So this patient is extremely high risk for, uh,
9:19
spontaneous intracranial hypertension.
9:21
This is what we would consider a positive screen on M r MRI for spontaneous
9:24
intracranial hyper hypertension.
9:26
They're basically four main types of spinal leaks that we deal with.
9:29
There's the ventral type one leak, which is the most common.
9:32
It's a high flow leak.
9:34
There usually is a ventral pseudomeningocele and it's often related to a
9:37
herniated disc or disc osteophyte complex, uh,
9:41
particularly at the cervical th thoracic junction in those cases.
9:44
And it causes this very long longitudinal epidural collection,
9:48
usually in the ventral, um, epidural space.
9:51
Lateral dal tears are type two leaks tend to be a little slower than type one s.
9:55
Um, they may or may not be associated with an epidural collection,
9:58
but if there is an epidural collection there that you can visualize on m r i,
10:01
those will almost always be more difficult to localize accurately.
10:04
So you have to image more quickly as opposed to doing pressure augmentation.
10:07
So the epidural collection is, is a very important finding to find on mri.
10:12
If you find an epidural collection,
10:13
your myelogram or your myelogram imaging must be much quicker to catch that
10:17
first drop of contrast that goes through the dural defect is particularly if the
10:22
epidural collection is very extensive because those epidural collections can
10:25
parallel the fecal sac and they can almost blend in with a fecal sac on
10:28
myelography.
10:29
And it can be very difficult to see where that leak is if you don't catch that
10:32
first little, um, uh,
10:33
drop of contrast that leaks out of the fecal sac into the,
10:37
the pseudomeningocele. Um,
10:39
it's often related to either a lateral disc herniation or a facet osteophyte,
10:43
and sometimes there can be a small perineural cyst, um, that, uh,
10:47
is facilitating this leak. It is different from the type four leak,
10:50
which I'm gonna talk about in a minute,
10:51
and that it's closer to the margin of the fecal sac.
10:54
So it would be basically within the spinal canal,
10:57
whereas a type four leak is actually a dehiscence of the nerve root sleeve and
11:01
contrast leaking along the nerve without any associated degenerative changes.
11:04
A type three leak is not really an open leak,
11:07
but rather a fistulization of the cerebral spinal fluid space with
11:11
the venous system. So, and it's usually a radicular vein in the neural frame,
11:15
and again, they're slower. Um, and it's also dynamic due to venous washed out.
11:19
So it's,
11:19
it's more difficult to detect these slower type three and type four leaks.
11:23
These are just some illustrations of the things that I was describing,
11:26
but I'll show you some pictures a little bit later on that are of clinical, uh,
11:29
clinical images. But essentially you, you see, you have a,
11:32
a small disc herniation here,
11:33
and then there's aural tear and a collection in the ventral epidural space in
11:37
this patient. This is a type one or ventral, uh, CSF leak.
11:39
This is a type two or lateral CSF leak.
11:41
Presumably this patient would've had a small disc herniation or DIY
11:46
complex out here that resulted in this small tear that's on the margin of the
11:50
fecal sac near potentially a perineural cyst. And, and I'll, again,
11:53
I'll show you a case of this a little bit later on,
11:55
or a couple of cases of this. Um, again, a c SF venous fistula usually is, is,
11:59
is, um, uh, associated with dact to the point where you have a,
12:03
a small perineural cyst or,
12:04
or a perineural cyst that actually visualizes with the venous complex. And,
12:09
and then in those cases, again, the vein watches out.
12:11
So it can be really difficult to detect these. Um, and then lastly,
12:15
type four s are more lateral leaks along the nerve root sleeve where you
12:19
actually have dehiscence of the nerve root sleeve and, um,
12:21
the contrast leaks along the nerve itself. So categorizing them,
12:26
like I said earlier, alluded to earlier into high and low flow is what the,
12:29
the main role of MRI is.
12:31
The other secondary role of MRI is to get a general region because digital
12:35
subtraction myelography, uh,
12:37
and rapid CT myelography for these high flow leaks is kind of the mainstay of
12:41
evaluation.
12:42
You really only get a very short temporal period to detect these leaks.
12:46
So you have to be in the right area, um,
12:48
because you're gonna be injecting only a finite amount of contrast and you only
12:51
inject three grams of contrast into the fecal ack.
12:53
And when you do a digital subtraction image,
12:55
you really only get two shots at seeing the leak.
12:57
So you really need to know about where it is cause you're a detector for your,
13:00
um, your digital subtraction is only gonna be about 20 centimeters,
13:03
so you're not gonna cover the whole spine and,
13:04
and a digital subtraction monogram. Um, so if you can,
13:07
if you can see a pseudomonal that's tracking along the fecal sac,
13:11
you can get a general region where the leak is probably at,
13:13
and then you can focus on that region when you're doing either a DSM or rapidly
13:16
acquired CT igram, which I'll, I'll go through that a little bit later as well.
13:19
They most frequently occur at either the cervical,
13:22
thoracic or thac lumbar junction. The ventral epidural space, and again,
13:25
it's usually caused by DYS, herniation or dico complex, um,
13:28
they vocally leak directly into the epidural space in the acute setting,
13:32
but they're rarely caught in the acute setting because clinicians usually don't
13:35
pick up, uh,
13:36
on c SF leaks in the acute setting unless there's some inciting event like
13:39
trauma. Um, so oftentimes when they present,
13:42
they already have a chronic pseudo low flow leaks. Um,
13:46
both acutely and chronically do tend to persistently leak into the epidural
13:50
space and rarely form chronic pseudomeningocele.
13:53
Type two s occasionally will do that,
13:54
and it does make them harder to detect the exact location,
13:57
but it's easier to detect that there is a C S F leak because a pseudomonal is
14:01
easier to see than basically a focal pinpoint leak into the epidural space
14:05
without a pseudo seal. Um, they typically occur in the lower thoracic region,
14:09
either laterally or posteriorly, uh,
14:11
and they're often associated with a perineural cyst or diverticulum, um,
14:16
and or osteophytes associated with the facet joints. So those are the,
14:19
the different types of leaks, and again,
14:20
categorizing into high and low flow based on the presence of a pseudomonal or a
14:25
perineural cyst is critical for the next steps for those patients. So with that,
14:29
I'm gonna move on to, um, localization of these leaks. Um, again,
14:33
detectability and accuracy, uh,
14:35
of localization is highly dependent upon the rate of,
14:38
of the leak as well as the consistency. Um,
14:40
and the associated extend of the pseudomeningocele,
14:42
the pseudo and enil is very focal, then it's, it's not really a big problem.
14:46
You, you know, about where the leak is, but if it's a really long one,
14:48
it extends along most of the spine, then you really have to be a lot more, uh,
14:52
judicious in your imaging and and image very quickly. If it's a high flow leak.
14:55
Again, high flow acute leaks are easy to detect,
14:58
but clinicians usually don't pick up on them until they're later on in their
15:01
course. The patients don't even suspect it usually. Um, and then, uh,
15:04
chronic leaks, um, that are high flow or easier to detect presence, meaning we,
15:09
we know where they are,
15:10
but accurately localizing them can be extremely difficult just because the
15:14
pseudomeningocele again, can be very extensive. Um,
15:16
both acute and chronic slow C S F leaks as well as intermittent leaks are
15:20
difficult to detect and localize because these patients often have more than one
15:24
abnormality of their fecal sac. They'll have multiple perineural fists,
15:27
so you have to look at multiple different regions and,
15:30
and narrowing it down a little bit can help a lot. Um, so again, um,
15:34
the high flow leaks are typically the type type ones and the type twos. Um,
15:38
low flow leaks, type twos three or fours can be kind of slower, low flow leaks.
15:43
Um, I normally do a screening MRI on, on both of them,
15:47
again to evaluate pseudo and inel and those type two and type one leaks.
15:50
And then looking for peroneal cysts, uh, in the type two s through four s. Um,
15:54
and then if I'm dealing with a high flow leak or suspected high flow leak based
15:57
on the detection of a pseudo and inel,
15:59
then I'm gonna do either a very fast pressure neutral CT mammogram.
16:03
What that means is that I'm gonna put the patient on a bolster in ct,
16:06
do the spinal tap there, and actually put the contrast in,
16:09
pull the bolster out and scan immediately with a large field of view.
16:12
So I can see the contrast as it flows up and I can see the in, uh, the,
16:15
the exact moment or the exact location of the leak without filling the entire
16:19
pseudo and inel and obscuring my my focal area of dehiscence. Um,
16:23
you can also do digital subtraction myography.
16:25
The benefit to digital subtraction is that you can actively watch the contrast
16:29
flow up and, and see where it's leaking from the downside to it.
16:33
Particularly in heavier patients, it can be really difficult to,
16:35
to accurately localize the leak because there there's only, um, uh,
16:39
only so much penetration you can get, um, through the shoulders,
16:43
particularly at the cervical thoracic junction.
16:45
And a lot of these leaks are at the cervical thoracic junction.
16:47
So you can have problems with the shoulders even on oblate, um,
16:51
seeing exactly where it is, particularly on a heavier patient.
16:53
So on heavier patients I tend to lean towards, uh,
16:55
a fast pressure neutral CT milligram. Um, whereas on,
16:59
on thinner patients I tend to go more for digital subtraction myelography and
17:02
watch the contrast come up for slow flow leaks. Um, again,
17:06
I do the screening MRI and I'm looking for, again, perineural cyst. Um,
17:10
and I will normally put them in the lateral recu disposition of the side that's
17:14
most suspicious to me and either do, uh, a pressure augmented, um,
17:17
CT myelogram with plus or minus some delayed imaging and doing everything in ct.
17:22
Or I'll do a digital subtraction and do the same thing.
17:24
Basically I'll watch it with pressure augmentation over time, um, and,
17:28
and look for, um, uh, those slower leaks. Um,
17:31
you can also do contrast enhanced mri my myography with or without pressure
17:35
augmentation. You can basically inject galine in the fecal site,
17:38
which is also a useful tool. And I, I'll go over that in a minute and how,
17:41
how I do it. Um, you can also do nuclear medicine sonography,
17:44
but I I find that to be of limited value unless you're dealing with fairly large
17:48
perineural cyst. If they're very large and there're, there're multiples, um,
17:52
nuclear medicine actually can be very useful, um, for those cases.
17:56
But for smaller cysts it can be really,
17:58
really difficult cause they kind of blend into the fecal sac on nuclear medicine
18:01
stenography. But in patients with massive delicia like,
18:04
like those with marfa syndrome, things like that,
18:06
nuclear medicine can be helpful when it can detect it. Um,
18:10
so let's start with the ultrafast, uh, pressure neutral CT myelography. Um,
18:14
it does have a significant advantage and again,
18:16
those high flow leaks over conventional delayed myelogram cuz you can get that
18:19
pinpoint accuracy of the,
18:20
of the exact location of leak when you have a large complex pseudo and inusal
18:24
that otherwise might be obs might obscure the actual location of the leak. Um,
18:29
we perform at start to finish at U C S D and CT with both the LP and the
18:33
myelogram on the, in, on the same table.
18:35
The head is lowered right after the contrast inject is injected. Normally,
18:38
I I perform the CT fairly quickly after, um, I do use, uh,
18:43
a smart prepped to actually detect when it's getting into the region and then
18:46
it, it triggers and it'll scan the entire region of interest basically. Um, uh,
18:51
and this is just sort of the, the, uh, extensive details.
18:54
I'm not gonna read all of this to you,
18:56
but I tend to use a coaxial system to do the puncture so that I make a smaller
19:00
puncture. I use a 25 gauge for the dural puncture.
19:02
The reason you can get away for get away with this is because you're infusing
19:06
the entire contrast bolus in the, in the luo sacral spine with the patient's,
19:10
um, uh, head elevated.
19:12
So you can infuse the entire thing at whatever rate you want.
19:14
You don't have to infuse it as quickly as you would for A D S M. Um,
19:18
and normally I use omnipaque two 40 for these higher flow leaks. You'll see I,
19:22
I I talk about using a denser contrast, omni pick 300 for those. Uh,
19:26
for those slower leaks it is a little bit more helpful,
19:28
but normally I use two 40 for these, um,
19:30
higher flow leaks and normally I remove the needle before I even get ready to
19:34
scan. I usually remove the bolster and then again do a single slice, um,
19:38
and trigger a smart prep to, to actually, um, do the scan.
19:41
It typically takes about 15 seconds for the contrast to reach the CS sponge.
19:45
So if you just wanna trigger it based on pine,
19:47
you could do it after about 15 seconds, you'll probably catch most of them. Uh,
19:51
and generally I only perform a a, a region where I'm,
19:54
I have concern based on the M r mri, like just the region of the,
19:57
the pseudo and a seal that I'm concerned about. Um, and if no leak is detected,
20:02
I will often roll them into the prone position and immediately repeat the scan
20:05
and just take a look and see if there's anything else that I'm missing.
20:08
So I'm gonna present a case here so you can see the utility of this and,
20:12
and you can hopefully understand what I'm talking about a little bit better and
20:14
it can reinforce this concept.
20:15
So this is a 32 year old female with postural headaches and this patient has a
20:19
pretty obvious type one leak.
20:21
So it doesn't require a flow compensated FIEs or anything special to see that
20:25
this patient has a fairly decent size pseudomeningocele in the ventral epidural
20:29
space at the thac lumbar junction associated with, um, this, uh,
20:33
small disc herniation. Uh, and, um, essentially right below the, the, uh,
20:38
thesis six, uh, level. So in this case, um,
20:42
the pseudomona in seal is big enough that you might not get an accurate
20:45
localization because you're looking at four or five levels.
20:48
So you do need to scan the entire thing, uh,
20:50
and catch the exact moment that the contrast leaks out.
20:53
So this is an example of, uh, of me doing this ultrafast, uh,
20:58
pressure, pressure neutral CT myelogram. I'm doing basically my, uh,
21:01
lumbar puncture. Uh,
21:03
and I I tend to do it in either the lateral to cubitus or prone position and
21:06
I'll put 'em on a bolster and then I'll roll 'em and drop their head and then do
21:10
the smart prep.
21:10
But basically this is the initial immediate CT mammogram showing this tiny
21:16
pinpoint dural tear right at the C6 level, um, just inferior to the C 56 disc.
21:21
This shows that same patient in a delayed image,
21:24
you can see the entire pseudo inic seals filled.
21:26
So you can't see that the leak is exactly right here and you're probably
21:29
thinking in your mind what does it matter if if,
21:32
if you localize it exactly here versus over here?
21:35
It matters for the surgeon based on which way they're gonna approach cuz they're
21:37
gonna approach anteriorly most likely, and they're, in this case,
21:40
they're gonna a approach from the, from the, uh,
21:42
from the right side as opposed to the left. Um, and you can also, um, you can,
21:46
you can also, uh,
21:47
imagine that once you start to look at this in reconstructions,
21:51
it's gonna be a lot harder to see. So I'm gonna demonstrate that. Um,
21:54
but this is the mag view of that. You can see this is the,
21:56
the focal area where the leak occurred.
21:58
This is the delayed conventional myelogram where the leak, the,
22:01
the actual tear the durrin blends in basically with the entire pseudomonal.
22:06
This is the, uh, a level slightly lower around the T1 level.
22:10
So you can see the pseudomeningocele hasn't even filled yet on this ultra fast
22:13
myelogram, whereas on a conventional myelogram, the pseudomeningocele is filled.
22:16
And I just wanna really drive this point home by showing you the,
22:19
the recons of this patient. So when you look at the reconstructions,
22:23
you can see on the conventional myelogram here on the left,
22:25
the entire pseudomeningocele is filled. So you,
22:27
you can't see where the leak is there, there's no way in this blended sort of,
22:31
uh, homogenous, uh, appearing collection in the ventral epidural spaces.
22:34
You're gonna know exactly where the hole is. Um, but on the ultra fast, um,
22:39
myelogram here on the right where we scan immediately after the injection,
22:42
you can see the pseudomeningocele is not even filled yet.
22:45
We just had this tiny little pinpoint focus of extravasation contrast right
22:48
below the C 56 disc, and that's exactly where the leak was.
22:51
The surgeon did actually a full corpectomy in this patient and did find the leak
22:55
and did repair it, basically. Um,
22:58
and it was a very focal pinpoint link that they just oversold and the patient's
23:01
doing well now. So, but that illustrates the, the, the, um,
23:04
the value of localization, um, based upon, uh,
23:07
those ultrafast CT mammograms.
23:09
And without having the m r I beforehand to actually know where the pseudo men
23:13
inusal was and where to focus, it would be a lot harder for me to detect, um,
23:17
exactly where the leak is because I would miss the exact point of where it's
23:20
leaking. Because you can't just scan the entire spine. You have to,
23:22
you have to focus on one area and you have to kind of know ahead of time about
23:26
where the leak is. Um, with that we can move on to pressure.
23:29
Augmented CT myography.
23:31
It has a temporal advantage over conventional myelography and detective slow CSF
23:35
leaks.
23:35
You can perform it entirely in CT or you can split it between fluoroscopy and
23:39
ct. Um, when I do that, I tend to do a DSM in, in, um, uh,
23:43
in fluoroscopy and actually do that first and then I'll send the CT and use the
23:47
CT as sort of my delayed image. Um,
23:49
I usually raise the pressure with a manometer and I, I, I utilize,
23:52
utilize both contrast and Elliot B solution. Um,
23:55
I put them in the lateral decubitus position on the side of most interests.
23:58
So the side with the most suspicious looking, um, uh, perineural,
24:02
cyt on the M R I, I'll put 'em on that side.
24:04
So if the left side's most suspicious,
24:06
I'll put 'em left side down or left lateral decubitus, uh,
24:08
prep their skin and essentially I'll access 'em with a 22 gauge needle in this
24:12
case because I wanna be able to get a really accurate pressure.
24:14
So I need to get some flow of C S F back and forth to kind of get an idea of
24:18
what the pressure is dynamically as I'm injecting additional, um, solution. Um,
24:22
but normally I inject 10 mls of Omni 300, and if I'm doing it under ct,
24:27
I just put 'em on a bolster N CT with the, the bolster underneath their hips,
24:30
uh, which is in contradistinction to what we were talking about earlier. The,
24:33
the rapid CT mammogram we're injecting and everything with their head elevated.
24:37
I would normally have the bolster with those rapid cases with the,
24:39
the bolster under their shoulders. In this case,
24:40
I'm gonna put the bolster under their hips so that the contrast goes ahead and
24:43
flows basically all the way up the spine. And as you,
24:47
you might imagine if they're on their side,
24:49
what tends to happen is that contrast just sort of runs into and drips into each
24:54
one of those perineural cysts almost like the, the, um, uh,
24:58
reservoirs in an ice cube tray.
25:00
And then after I finish dripping the contrast into all of these,
25:04
the next thing that I will do is I will raise their pressure withs.
25:07
You can also use some, uh, um, normal saline. I mean, it,
25:11
it's a little bit more painful for them, but you can use it. Um,
25:14
it causes kind of a pressure sensation for some patients. Um,
25:17
but basically I raise their pressure to between 30 and 35 centimeters of water.
25:21
Some people just wait, some people just, um, uh, essentially, um, uh,
25:26
raise their pressure until their headache is gone. I, I,
25:28
I raise it to 35 to give me my, my best chance of seeing it. Um,
25:32
and then I complete the scan and that position,
25:34
and then the delayed scan is usually performed an hour later.
25:37
So this is a patient who had both a conventional myelogram in the past and had a
25:42
pressure augmented myelogram with us. Um,
25:44
and you can see basically they had a suspicious looking perineural cyst at T
25:48
nine. Um, but which one of these is the one that's leaking it,
25:51
it is really difficult to tell. In this case. After pressure augmentation,
25:54
we could see that it was clearly, um, the one here on the right that is leaking.
25:58
So pretty straightforward to see, um, on the pressure augmented mammogram,
26:02
whereas a conventional mammogram, we could not see it at all.
26:04
And without kind of having the, the, um,
26:07
the M r MRI beforehand to show us where the perineural CSTs were, we,
26:10
we didn't even know where to focus. So, but this is really,
26:12
really helpful and I,
26:13
I can kind of show you some more of these lateral decubitus images as well, um,
26:17
as we go on. But th that was the, that was the final result after doing a D S M.
26:22
Um, this is the patient with a type four CSF leak.
26:25
So you can see basically we've got contrast all along this, uh,
26:28
T1 nerve root extending, um, into the intercostal space.
26:31
So very similar situation.
26:34
Both of these patients actually just underwent blood patches, uh,
26:37
foraminal blood patches and did fine, didn't have any more symptoms of,
26:41
of CSF leak. Um, but without doing these, um, uh, um, uh,
26:45
more specialized, uh, pressure augmented myelograms,
26:48
I would not have seen these C S F leaks. Um,
26:51
with that we can move on to contrast enhanced Mr. Myelography. It usually, um,
26:56
it, it involves an off-label injection of gatum and the fecal sac.
27:00
It has a temple advantage over, uh, um, over, uh, um,
27:05
uh, CT myelography and slow and intermittent leaks and over many other forms of
27:09
myelography.
27:10
Normally I perform the injection and the fluoroscopy suite and I usually make
27:13
sure that I'm in the fecal sac, uh, by injection of, uh, iodinated contrast. Um,
27:18
and then, um, I normally will raise the,
27:21
raise the pressure again using a manometer and lap solution.
27:24
We do a delayed MRI to allow time for the CSF leak to happen. Um,
27:29
uh, in general, um, I do it the same way that I, I do the, uh,
27:34
ultrafast, uh, pressure augmented, um, or not, sorry,
27:37
ultra fast CT myography in, in that I use a discogram set,
27:40
I use the coaxial set with a 20 gauge needle and a 25 gauge puncture needle so
27:45
that I'm making the smallest hole in the fecal sac possible. Um, and in general,
27:49
I just, I give it a little bit more time. The problem is,
27:51
is you're giving 'em so much delay time if you puncture them with a larger
27:54
needle than a 25, it, it,
27:56
it is painful to raise the pressure with a 25 gauge needle,
27:59
but it's worth it in this case cuz you're giving 'em so much time to leak. Um,
28:02
a larger needle can make a pretty impressive leak into the C S F or into the
28:06
epidural space, um, if you use a larger needle to do the lp.
28:09
So I do use a smaller needle for this and I use 0.3 MLS of, of gadolinium, uh,
28:13
and essentially, uh, dilute it, uh, to a total volume of three mls,
28:17
either with Elliot Bs or with autonomous csf,
28:19
and then inject it directly back into their fecal sac. Um,
28:22
and then we do fat sat, T1 and mri. So this is a patient who had, um,
28:27
a postural headache, 64 year old lady, and essentially she had perioral cys.
28:31
They were pretty suspicious on ct, but we really didn't see anything.
28:33
It looked like it was leaking. Um, and then i, I did do, uh, a fat sat,
28:38
um, T1 after injecting gadolinium,
28:40
and we did see a tiny leak from the one on the right here. Ultimately, um,
28:44
she did end up undergoing blood patch for this type four, um, leak and, and,
28:49
you know, did well, didn't need any additional treatments after that. Um,
28:51
here's another similar example except this is actually a type two leak.
28:55
So this patient underwent R Myography as well. Um,
28:58
she did have a suspicious looking nerve root on the, on the Fiesta imaging.
29:03
There was just some irregularity, uh, of the, uh, nerve root sleeve itself,
29:07
and then a small sort of collection outside the nerve root sleeve.
29:10
And then we injected glan, we could see a, a small leak into that, um,
29:14
into that area. And then when we did the digital subtraction myelogram,
29:16
it was clearly extravasated from the fecal sac around the T 11 nerve route here
29:20
on the, uh, on the right side. So this patient again, underwent blood patch,
29:25
did well, didn't need any additional treatments. Um, but that I'll, I'll move,
29:28
move on to digital subtraction. Myography, uh, again,
29:31
digital subtraction myelography is really heavily reliant upon MRI ahead of time
29:35
to kind of know whether we're dealing with a high flow or a low flow leak.
29:38
And as we'll put them either in the prone position if it's a high flow leak or a
29:43
lateral decubitus position. If it's a, uh, a slower leak, I,
29:49
I usually use as you need to inject very quickly, uh,
29:52
because the patient's either gonna be under general anesthesia with a breath
29:54
hold or you're gonna be having them hold their breath and injecting quickly and
29:57
watching the contrast flow up with a mask, uh, or with, uh, a negative roadmap,
30:01
either one. Um, I normally use omni two 40 again for those high flow leaks,
30:06
but I use 300 for the slow flow leaks just to get a little bit better
30:09
visualization. Um,
30:10
and then you can raise their pressure with Elliot BS if it's one of those slower
30:13
flow leaks. Um, but again, the position is critical.
30:16
So if they have perineural cyst on the m I put 'em in the lateral to Cuba
30:20
position in the most su the sort of focus on the most suspicious regions on the
30:23
mri. And then if they have a high flow leak or a, um,
30:27
a pseudo enga seal detected on mri,
30:29
then I focus on that area and I put them in the prone position. Um,
30:32
usually cause the, the, the pseudomonal is, uh, in the, um,
30:36
in the ventral epidural space. That being said,
30:38
sometimes PDO cells are lateral with type two leaks,
30:40
so sometimes we'll put 'em in the, uh, lateral decubitus position, uh, if,
30:44
even if it is a high flow leak,
30:45
if we believe it's one of those lateral type two leaks. Um,
30:48
so this is the patient who had previously undergone a, um, uh, fat sat,
30:53
uh, t2, M R i and it showed basically, uh,
30:57
extensive stranding in the epi within the epidural space in a small ventral
31:01
epidural collection, which we can see here in the lumbar spine.
31:05
This happened after an LP at an outside hospital. And, um,
31:10
essentially we weren't sure where the leak was.
31:12
I think the natural presumption is that it's at the clinical level where the
31:16
lumbar puncture was performed,
31:17
but she had numerous blood patches at L four five and it just did not resolve,
31:21
she continued to have severe postural headaches. Um,
31:25
and ultimately we did do a digital subtraction myelogram with her in the prone
31:29
position. And here's the injection.
31:30
You can see basically here's your initial injection,
31:32
you're just filling the fecal fact. Um, just so you guys are aware,
31:35
this is the S one level, L five, L four, L three, and then this is L two.
31:39
And then we see a little bit of contrast extravasation at L two three, uh,
31:43
on the initial, uh, or, or the second image here. And then on the third image,
31:47
we can see it clearly flowing backwards into the pseudo seal.
31:50
So the puncture clearly occurred much higher than you would suspect based on
31:54
where it should be clinically.
31:55
So it should have been clinically where I put the needle at L four five,
31:58
but it was actually closer to L two three and this was born out in surgery.
32:01
This patient had undergone so many blood patches,
32:04
she just skipped straight the surgery. Dr. Sh I think Dr.
32:06
Shain did the surgery on this patient. And, um, ultimately the,
32:09
the leak was right at two three,
32:10
right where we localized it on the digital subtraction myelogram.
32:13
And he repaired it without any, uh,
32:14
issues and she did well and has had no issues since. Um,
32:18
here's another patient with a type one leak. Sometimes they do present this way.
32:22
This is 74 year old guy with hemorrhagic bilateral subdural collections.
32:25
And the presumption is that this is obviously, uh, gonna be a, uh,
32:29
a subdural hematoma. But this patient had no real, um,
32:33
symptoms referable to the subdural. A patient didn't have any focal deficits.
32:38
He did report a several year history of tinnitus and strange postural headaches
32:42
and he was a very high performance tennis player despite being 74 and a lot of
32:47
twisting, bending when you, when you play tennis.
32:49
And he had a lot of degenerative changes in his spine.
32:52
So our spider senses were tingling with this larger collections.
32:55
And you can see he had the couple of discs here in his cervical spine.
32:58
So the presumption is gonna be that the leak must be up here, right?
33:01
It has to be, it can't be anywhere else.
33:04
This is just a standard TT weighted saal, M R i.
33:06
But this is gonna show you the value of a flow compensated fiesta.
33:09
So the flow compensated fiesta.
33:11
So actually that the epidural collection was down here behind the t1, uh,
33:15
vertebral body.
33:15
And this helped us really focus when we did the digital subtraction myelogram.
33:18
As I had alluded to earlier, when you get around the cervical thoracic junction,
33:21
it gets a little bit dicey with digital subtraction. But this is the T1,
33:25
T2 vertebral bodies that I had marked, um, here.
33:27
And basically the leak did occur right where we predicted it,
33:29
it was going to occur.
33:30
So we have a split in the dura here on the digital subtraction myelogram,
33:33
exactly where we saw the, uh,
33:35
the small epidural collection on the flow compensated Fiesta mri. Ultimately,
33:40
I did again do a blood patch on this patient and he did really, really well.
33:43
The subdurals completely resolved, um, within just a few weeks. So, um,
33:47
presumably we were correct, uh, in, in all of these assumptions, um,
33:50
based upon the mri. So the M R MRI was, uh, you know,
33:53
o obviously extremely helpful in this case. Um, with that, that's,
33:57
that's all I have to say about CSF leaks for now, but I,
33:59
I'm gonna look at the chat really quick before we move to hydrocephalus just to
34:03
see if, uh, anybody's asked any really important burning questions.
34:15
These look all like form responses. I don't see any real questions yet,
34:19
so I'm gonna go ahead and move on then to hydrocephalus. Um, as promised,
34:24
I am gonna show, um, a,
34:26
an obstructive case to start just cuz I don't want people to miss, um,
34:29
obstructive hydro thinking. They're dealing with NPH and,
34:32
and we're gonna get to NPH eventually. But this patient was sent to the,
34:35
the first patient that, that, that I'm gonna show you was sent to me with, uh,
34:38
concern for NPH H and, um, it, it was not nph, but at any rate,
34:43
obstructive hydrocephalus is just an increase in intraventricular volume of C
34:46
sf, um,
34:47
due to impaired drainage of c SF from the lateral ventricles resulting in an
34:51
increase in intraventricular intracranial pressure. So in other words,
34:54
the c SF is being produced normally in the ventricles,
34:56
it just can't get out of the ventricles to be absorbed by the erect
34:58
granulations, which are along the surface of the brain or,
35:01
or along the surface of the skull. And dura near the,
35:03
the venous sis as we talked about earlier, normally presents with headaches,
35:05
visual disturbances, cranial neuropathies, poor balance, ga disturbance,
35:09
sleepiness and seizures. Um, etiologies can include, um,
35:12
aqueduct stenosis adhesions due to prior infection or hemorrhage. Um,
35:17
particularly basler meningitis or prior subarachnoid hemorrhage can do this kind
35:21
of thing can also come from extrinsic compression or entrapment due to a tumor
35:25
like, uh, you know, quadri, genital plate cistern kind of lesion, pineal lesion,
35:29
those sort of things can obstruct the aqueduct. Um,
35:32
there are congenital webs that, that occur in the aqueduct that can cause this,
35:36
um, imaging findings usually, um, include a significant enlargement,
35:40
the lateral in the third ventricle, um,
35:42
without the sort of normal flow artifacts that you see, uh, you know,
35:45
through the, the, the cerebral aqueduct and the fourth.
35:47
And usually the fourth is, is quite a bit smaller than you'd expect to be,
35:51
but you have to keep in mind the fourth is kind of confined by the posterior
35:54
fosus, so it's not gonna get that big to begin with.
35:56
So it might not be that sensitive of the screening, uh,
35:58
evaluation for obstructive hydrocephalus. But one of the things it is,
36:01
is if you see Boeing of that recess of the third ventricle right above the optic
36:05
chiasm or right behind the optic chiasm. And then if you see Boeing, um, uh,
36:09
superiorly of the, um, uh, of the corpus coum, uh,
36:12
that's usually indication that you're dealing with probably obstructive
36:15
hydrocephalus and it can be treated with either a VP shunt or an endoscopic
36:18
third ventriculostomy. So this is a patient who was sent to me with, um,
36:22
gait disturbances and cognitive issues and they,
36:25
they told me that they thought it was MPH and I looked at the patient, he,
36:29
he just, he did not strike me as being someone with MPH cuz he was just really,
36:33
really sharp, um, despite having some minor, you know, issues with,
36:38
um, word finding and, and, and other things. Um, so at,
36:42
at this point my spider sense was tingling,
36:43
so I sent him for a C S F flow study.
36:46
So the C S F flow study we do here at U C S D includes a sagittal fiesta just to
36:51
kind of get a lay of the land beforehand.
36:52
And as you can see immediately we can see that he has an aqueduct web and
36:56
there's dilation of the cerebral aqueduct. There's Boeing of the callosum,
37:00
there's Boeing of the recess of the third ventricle, um, right,
37:02
right behind the optic chim. So all these things are suggestive that,
37:05
that he has obstructive hydrocephalus due to an aqueduct web. Um,
37:08
at this point I stopped them from doing our traditional flow study,
37:12
which is an axial flow study through the aquatory measure,
37:14
basically the velocity and the, the, um, the,
37:17
the rate of flow through the aqueduct. In this case, I asked them just to do,
37:21
um, a, uh, uh, a flow through the, uh, through the frame of magnum,
37:26
the Statal plane with five, 10,
37:27
and 15 banks and just to kind of look and see if there's any flow through the
37:31
aqueduct. Um, there's no flow through the aqueduct on this patient.
37:33
As you can see, there's only, um,
37:35
movement of C S F in the pre pontine cistern and maybe through the frame of
37:39
Magen, a little bit out of the fourth ventricle,
37:41
probably due to just two and fro flow from pulsation of the cerebellum most
37:44
likely. Um, but you know,
37:47
I I think one of the things you have to keep in mind is you wanna set your vex
37:50
on the lower end for something like this,
37:52
because presumably if you have aqueduct stenosis,
37:55
you're gonna have a lower flow rate if it's significant. So you need to,
37:57
you need to bump your vex lower and just make sure, um,
38:01
that it's not more than 10 or 15% above what you think the velocity's gonna be
38:05
through there, which is, it's gonna be pretty slow presumably if it,
38:08
if there's that bad of aqueduct stenosis. Um,
38:10
so that's what we ended up doing in this case. And,
38:12
and I sent this patient for an ET T V and after the ET T V all of his symptoms
38:16
resolved and his ventricles returned to normal thyazide and he didn't have any
38:18
issues. But again, don't,
38:20
don't go down the rabbit hole of NPH until you've at least seen a Saal Fiesta on
38:24
these patients. With that, we're, we're gonna move to the what you came to see,
38:27
which is normal pressure hydrocephalus.
38:28
It usually affects older adults in the sixth and seventh decades of their life.
38:33
Um,
38:33
it results from an increased production of C S F volume without obstruction or,
38:37
or, or elevated pressures.
38:39
Presents with gait disturbances and cognitive decline and incontinence in some
38:43
cases. Um,
38:44
some of the etiologies are speculated to involve sort of changes in compliance
38:48
of the brain that result in kind of compression of the peri ventricular white
38:51
matter and result in emia um, imaging usually demonstrates enlargement,
38:55
the lateral and third ventricles out of proportion of the degree of volume loss
38:59
with some resultant reduction in the, um, uh, colossal angle.
39:02
So you get basically this acute colossal angle at the level,
39:05
the posterior comme.
39:06
One thing to keep in mind though is that this colossal angle is not specific
39:11
nph, I've heard people say, oh,
39:13
it must be NPH cause the colossal angle is narrow. That's not true. It just,
39:17
it's,
39:17
it's a way of determining whether you're dealing with hydrocephalus or just
39:20
plain old volume loss.
39:22
So if you measure the cosal angle and it's narrower than 80 degree, uh,
39:25
80 degrees, um, uh, and, uh, the patient essentially, uh,
39:30
has, you know,
39:31
findings that are concerning for NPH H and they don't have ACR ductal stenosis,
39:35
then probably is NPH h but at that point, you're really just distinguishing mph,
39:40
or sorry, you're just distinguishing hydrocephalus from volume loss.
39:42
You're not dis you're not really making the diagnosis of mph,
39:45
you're just suggesting that there's some form of communicating hydrocephalus at
39:48
that point. Um,
39:50
hyperdynamic flow through the cerebral aqueduct is often present in most of
39:53
these cases, but the diagnostic criteria for this is still pretty controversial.
39:57
Um, a lot of the, the work on, um, C S F flow dynamics was done here by,
40:02
um, a former chair. His name was Bill Bradley,
40:05
and I was actually here when we did the updated studies on our GE machine.
40:09
But originally all this was done on a Siemens, uh, m r MRI machine, uh, with 10,
40:14
20 and 30, uh, van um, velocities or van van, um, settings.
40:19
Um, and essentially what they found was, uh,
40:21
stroke volume of greater than 42 microliters was, um,
40:25
correlated with a higher probability of shunt responsive nph.
40:29
And then the same thing was found on the GE flow sequence, but,
40:32
but only for greater than a hundred microliters. Um, that being said,
40:36
these were small numbers of patients that we took and, um,
40:40
I think with that low of an in number, you have to be a little bit skeptical.
40:43
Same thing for the flow rate of 24 milliliters per minute. Um, that,
40:47
that also could indicate nph.
40:49
The way I think about it though is that the gold standard is still temporary
40:52
aversion of C S F to actually, um, see if the patient's symptoms resolved.
40:56
So if you do a high volume LP or a three day lumbar drain trial and they,
40:59
they improve, then you know, they, they probably do have nph.
41:04
I look at the, um, the,
41:05
the flow study as sort of an adjunct to help me make a decision in those
41:09
patients where we have some equivocal issues. So things that can cause your,
41:14
your, um, uh, gold standard temporary C S F diversion to be somewhat,
41:19
um, uh,
41:20
more dodgy or difficult to trust is patients who've been immobilized for a long
41:24
time and can't walk. So you can't really do a get up and go test,
41:27
you have to rely on their M M S E, uh, or their MOCA test, um, after diversion.
41:32
And honestly, those things can be a little bit variable from day to day,
41:35
even with just normal dementia patients.
41:36
So they could just be having a good day.
41:39
And I think that the C S F flow study can really help in those categories.
41:42
It can also, you know, help with screening too if you,
41:45
if you're not really sure you're dealing with NPH and um,
41:48
you just wanna see if there's hyperdynamic flow and it kind of push you in that
41:51
direction, it can be helpful to do that. Um, but I, I think in the end,
41:55
you're still gonna have to, to,
41:56
to do an LP or a lumbar drain to be 100% certain if that's what you're dealing
42:00
with before you shunt the patient. And again,
42:02
it's treated by either a VP or LP shunt. I, I tend to like, um,
42:06
VP shunts for these patients because the LP shunts just not as easily controlled
42:09
and as, as older patients often do,
42:12
these patients usually have con commit volume loss, and if you do an LP shunt,
42:15
it's harder to control it and you can easily over drain these people with an LP
42:19
shunt and they can end up with subdural collections and or, or subdural HROs,
42:23
you can end up with a a different problem going forward.
42:25
So this is an 82 year old female that that came to me with, uh,
42:29
concern for mph and basically she had progressive gait disturbances, uh, and,
42:34
uh, memory issues. And as you can see, her ventricles are enlarged. Um, and,
42:39
and it is a little bit out of proportion of the degree of her volume loss
42:42
overall. Um,
42:43
you can see she does not have any aqueduct stenosis in her colossal angle is
42:46
narrow suggesting hydrocephalus rather than just volume loss in this case. Um,
42:50
so we did do a, uh, a flow study, um, and essentially the, the first thing,
42:55
I always look at these aile flow studies,
42:57
I wanna see at least that there's two and fro flow through the aqueduct on the
43:00
initial image acquisition. So I wanna see, um, you know, basically the, the,
43:05
the bright flow jet and then the reverse flow jet giving me sort of a dark
43:09
signal as it as it goes backwards. Um, and essentially you're gonna have, um,
43:13
forward flow and backwards flow.
43:15
The stroke volume is actually the flow in one direction,
43:18
not the flow in both direction. So, um,
43:21
the stroke volume in this patient is around, uh,
43:24
or just a little over a hundred microliters.
43:26
It's very common to see patients have slightly more forward flow than
43:31
backwards flow because they're making C S F. So you presume that, you know,
43:34
they're secreting some out of the brain to go to the arachnoid granulations and
43:38
be reabsorbed. So having a slightly higher forward volume is, is typical,
43:42
but if there's a big discrepancy between forward and backward volume,
43:46
I tend to not really believe the stroke volume. So this is a, a,
43:50
a case where technically the stroke volume was telling us that this patient
43:53
probably had nph. Um, the flow volume was not that impressive,
43:57
though it was only six mls per minute. Um, whereas, you know,
44:00
24 has been more closely correlated with, with patients who have nph.
44:05
That being said, you know,
44:06
we felt very strongly about this patient potentially having mph and we did do,
44:10
uh, a, a high volume LP and this patient recovered dramatically and,
44:13
and did very, very well with a shunt. Um, so it just goes to show you the, the,
44:17
the, the, the, um, flow study may not always give you the answer. Um,
44:21
it really is meant to mostly predict shunt responsiveness rather than really
44:25
diagnosed mph. Um, with that we can move on to intracranial, or sorry,
44:30
idiopathic intracranial hypertension or pseudotumor cerebro,
44:33
however you wanna say it,
44:34
and that that's along the same spectrum as venous sinus or or venous outflow
44:38
obstruction, jugular venous obstruction also like from, uh,
44:41
either cranial cervical instability or, um, uh, uh,
44:45
basically venous eagle syndrome if they have elongated styl processes. Um,
44:50
idiopathic intracranial hypertension though typically affects younger,
44:53
more obese females in the kinda the 20 to 40 age range. Um, and again,
44:57
risk factors, obesity, retinoids, like vitamin A overdose, um, uh, you know,
45:02
things like that, venous obstruction contributes to it. Uh,
45:05
in some cases hyper hypoparathyroidism hearing deficiency,
45:09
cushings hormone supplementation and tetracycline can also be risk factors for,
45:13
um, idiopathic intracranial hypertension. Um,
45:16
typically there's an increase in intracranial pressure and opening pressure when
45:20
you do an LP without really an increase in or without a significant increase in
45:24
ventricular volume,
45:25
may typically will present with PA edema and or double vision palsy.
45:29
A lot of 'em do have tinnitus and headaches as well.
45:32
And the imaging findings typically include sort of slit like ventricles and
45:36
empty cell dilated optic nerve sheaths. Uh,
45:39
and typically there's some degree of diffuse sinus narrowing. Um,
45:43
focal venous sinus narrowing, jugular venous stenosis and or, um,
45:47
cerebral venous sinus thrombosis may potentiate or mimic IH in certain select
45:51
cases. Whether it's a mimic or a potentiator is really more, um, uh,
45:57
a judgment call based upon the patient's phenotype.
45:59
So if you're dealing with a thin person who you know,
46:01
has a very low B M I and they have very severe venous stenosis,
46:05
then it's more mimicking I h whereas if you're dealing with someone who's closer
46:09
to the phenotype that you expect for i h then a focal venous sinus stenosis may
46:13
be potentiating it and making it worse and they could benefit from
46:16
revascularization, but typically you'd have to do venous pressure measurements,
46:20
which I'll, I'll go through that in just a minute.
46:21
But the typical treatments would invo involve, involve weight loss diamox, um,
46:26
VP shunting or LP shunting. And, and they, they both,
46:29
they respond to both of those. Again, VP shunts are easier to control,
46:32
but it's harder to put in a VP shun on one of these patients because the,
46:35
the ventricles are more slit like, um,
46:37
you can do venous stenting as I alluded to in select cases. Um,
46:41
this is one of those select cases.
46:43
This is a 46 year old lady who came to me with edema. Um,
46:46
her b m I was on the higher side, but it wasn't really that suspicious. I mean,
46:50
and she didn't really look that big to me. I mean, I'm, I'm from the south.
46:53
I mean she seemed like normal kind of, I mean, so I was like, I don't really,
46:57
I don't, my spider sense was tingling.
46:59
This is more of a venous sinus problem on her. Um,
47:02
but you can see in this patient she does have slit like ventricles. Um,
47:05
she does have, um, dilation of the optic nerve sheath.
47:07
It's easier to see on the right side here. Um,
47:10
she also has a partially empty cell,
47:12
and what we can see is that she has basically diffused sinus narrowing with a
47:16
little bit of superimposed narrowing at the sigmoid transverse junction
47:19
bilateral, which is a common location for arachnoid granulations.
47:22
And that's usually the, the, um,
47:24
con commitment contributing factor in these patients.
47:27
She did have a high opening pressure, I believe it was, um,
47:30
in the mid to high thirties, if I'm remembering right. Um,
47:33
so we did an M R V on this patient,
47:34
and you can see she does have some degree of diffuse sinus airing with some
47:37
distension of the cortical veins. Uh, and then she has actually, um,
47:41
focal arachnoid granulations at both sigmoid transverse junction, um,
47:44
causing superimposed, um, uh, um, uh, stenosis,
47:48
but still because of her b m i being a little on the higher side, we,
47:51
we would have to call this more potentiator as opposed to a, a mimicker.
47:55
And in this case, she definitely needs venous sinus pressure measurements,
47:57
which is what we did.
47:58
We catheterized the venous sinuses all the way across the toula,
48:01
usually from one jugular, and essentially measured the pressures on a pullback.
48:04
Um, she had a six to one gradient with an absolute gradient of 21 millimeters
48:09
mercury, which is very high. Um, anybody with a,
48:12
a gradient of greater than three to one or eight millimeters,
48:14
mercury usually predicts, uh, sh uh, stent responsiveness or,
48:17
or revascularization responsiveness.
48:19
So she ultimately did undergo stenting and didn't require anything else
48:22
required. No more dialogue pressures dropped to normal, uh,
48:26
PA edema resolved everything, um, resolved, and she,
48:29
she's doing well with no headaches. Now at this point. Um,
48:31
I usually treat these patients just like arterial stents. I put 'em on aspirin,
48:34
Plavix for three months, and then after that, just lifelong baby aspirin. And I,
48:38
knock on wood, I haven't had any patients who've had, um,
48:41
significant repeat narrowing. I,
48:42
I typically just for those who are interventionalists, I, I, um, I,
48:46
I typically use, um, the silver biliary stent. I mean, it's just,
48:50
it's really easy to open. And if you have trouble getting there,
48:52
I normally just use our stroke catheters to get a guide up there.
48:55
So I'll use like, um, you know, like a,
48:58
a neuron max guide catheter or B M X 96, and then I'll put like, um,
49:02
either a, uh, red 62 or 68 through it, and then some sort of, you know,
49:07
microcatheter wire. And I, I typically use the road runner wire though,
49:11
but once you get the neuron max into the head, um, it,
49:14
it's very easy to deploy one of these tenents if you, if you,
49:16
if that's what your goal is. So, um,
49:18
but the stroke catheter system is basically just used to get the guide there and
49:21
then the road runner exchange wire will, will easily take the, uh,
49:24
silver stent into position if that's, if that's what needs to happen. Um,
49:29
so this is that same patient after, um, uh, a few months.
49:32
We did an r v prior to taking her off the aspirin PLAs, you can see the stents.
49:35
Why it be patin? It's harder to see the signal through the stent, um,
49:38
because obviously the stent, um, obscures MRI a little bit, but it,
49:41
it's a pretty good, uh, result overall.
49:43
And you can see the stent appears patent, uh, also on the, on the, um,
49:48
uh, the, uh, 2D time and flight. But, um,
49:51
I think that one thing that is a little bit easier if you do a a ctb rim,
49:55
you can see it a little bit easier, but I tend not to like to do, uh,
49:58
ionizing radiation in these younger patients and,
50:00
and give 'em contrast if I don't have to. So I,
50:02
I typically start with an R v and if I'm not,
50:04
if I'm not worried based on the r v and the symptoms,
50:06
I usually just stop at that point. But then I'm,
50:09
I'm gonna move to Keri malformations, which are,
50:11
are more cranio cervical obstructions, um, and can result in hydrocephalus.
50:14
So we, we'll talk about those and then we'll, we'll conclude and I'll,
50:17
I'll take a look at the questions and answers again and see if there's anything
50:20
that I can help people understand a little better or answer any more, you know,
50:24
um, philosophical questions if, if those arise. Um,
50:27
so keary malformations are congenital lesions that are usually associated with a
50:30
small posterior fossa. Type one s are the most common.
50:32
There's usually tons utopia and the tons usually, um,
50:36
as they herniate through the frame of magnum take on like a peg like appearance.
50:39
And that doesn't always happen. But that's, that's what we're looking for,
50:42
sort of a peg like appearance, um, is the classic keary two s are, um,
50:46
displacement of the cerebellar tonsils and the brain stem through the frame of
50:50
magnum with usually a luo sacral pseudo or luo sacro myelo meninge.
50:54
And then type three s are similar to keary two s with either a high cervical or
50:58
occipital myelo meninge.
50:59
And they're typically associated with both hydrocephalus and, um, sargo hydro.
51:03
My Ilia, um, imaging can demonstrate, um, crowding the frame of magnum again,
51:07
like with those peg light, um, uh,
51:10
cerebellar tonsils that you can see a searing sometimes, um, low line conus,
51:14
especially with the Q R E two, um,
51:16
because of the myelo steel and the tethering of the cord. Um,
51:19
the ventricles can be enlarged in some cases if they,
51:21
if they have hydrocephalus. And you can also see, um,
51:23
decreased C S F flow around the frame magnum and brainstem C SF flow studies on
51:28
m r I can be very helpful in determining which patients are really symptomatic
51:31
due to their curi and, and help guide surgical management. Um,
51:35
this was a 29 year old lady with severe headaches that were triggered by
51:38
coughing. Um, and you know, I,
51:41
I was asked to render an opinion on this case because the tonsils really weren't
51:44
peg, like they really weren't lying that low,
51:46
but there was some crowding in the frame of magnum and you,
51:48
you can see they're a little bit low here.
51:49
They're about seven millimeters below the frame of magnum.
51:52
But the patient had a pretty large snx. Um,
51:54
what was interesting about this patient is,
51:56
is she didn't really have any cord signs of any kind,
51:58
which I was kind of shocked at.
51:59
She just had these headaches that were just terrible,
52:02
that were triggered by coughing and sneezing. So I was like,
52:05
this is not a main reason I was asked to render pain.
52:08
Cause someone thought there could be maybe a CSF leak. This is acquired Kia.
52:11
And I was like, ah, I don't think so. Um,
52:13
we did a contrast enhanced brain r i and there there was no secondary signs. Um,
52:17
but in, in the end, this patient did undergo a flow study.
52:20
So this is the Saal Fiesta from that flow study.
52:22
You can see there's crowding the frame of magnum with very little room around
52:25
the brainstem, and there was literally no flow through the frame magnum on the,
52:29
on the sagittal flow study here. Um, and, and again, at this point we,
52:32
we turned our veins very low just in case, um, that we were missing something.
52:36
But we,
52:36
we saw basically no flow through the frame of Magna at that point we decided
52:40
she, she needs decompression. So she got a suboccipital, cranny neuroplasty,
52:44
and um, a few months later the steerings is completely gone.
52:48
So it, it, that had to be what the problem was,
52:51
even though the tonsils weren't peg, like, I mean, and I,
52:53
I show that because the, we, we do fixate on that peg like tonsil thing.
52:56
And I think the flow study can really help you decide if,
52:59
if it is a significant Kiara or not, if you don't see those peg like tonsils.
53:03
Okay. Um, with that, I'm gonna check the questions and answers again, da da da.
53:09
What is the cause of subdural effusion and hypotension? Um,
53:14
that's a good question.
53:15
So presumably there's actually negative pressure in the
53:20
subarachnoid space so that it's actually exerting sort of like a vacuum on the
53:24
PACU meninges to the point where it's actually pulling basically the dura and
53:28
causing the subdural effusions to occur. So that's,
53:31
that's what we believe is happening basically,
53:33
and that's what we believe is causing the enhancement as well.
53:36
It's causing inflammation.
53:37
Do you stent vocal venous sinus stenosis? The patient only have tinnitus. Ooh,
53:43
that's a really good one. So, um,
53:46
I will often do their full workup. I'll do their lp, see what their pressure is,
53:51
and if their pressure is very high to the point where I'm worried about them,
53:56
um, then I will often offer them the, the, um,
54:00
venous sinus pressure measurements and I do all this awake. So I'll,
54:04
I'll basically do an lp. I'll tell 'em, Hey, your pressure's 30, you just,
54:07
you're lucky that you don't have pa edema. Um,
54:09
let's go ahead and do venous sinus pressure measurements if you're okay with it.
54:12
And I'll just flip 'em over and do the venous access. And just quickly,
54:15
you know, I, I use a very small microcatheter,
54:17
so I don't even give 'em that much heparin, so it's safe to do 'em the same day.
54:20
I give 'em like 1500 to Heparin and I just throw in like a five French, um,
54:23
diagnostic catheter, like usually a Bernstein or vert or something.
54:26
And then I'll throw like a prowler select plus through, and then I'll just,
54:29
I'll measure the pressures all the way across. I mean,
54:31
it takes like 10 minutes of added time. And then in, in most cases,
54:34
like if the pressure gradient is there,
54:36
I'll just tell 'em it's there and then we'll have a big conversation in clinic.
54:39
But I can tell you I've never personally stented anyone who just had tinnitus.
54:44
I've stented someone who had tinnitus and bad headaches,
54:47
but never just tinnitus. Okay. Next question.
54:51
What is the,
54:53
what is the specific sign in imaging studies for NPH H and not any
54:57
kind of hydrocephalus? Ooh. Um,
55:02
I don't think there really is a specific sign for MPH and imaging, um,
55:07
that collosal angle, um, that we were, that we were talking about.
55:12
Um, really just predicts, you know, atrophy versus hydro.
55:16
I don't think there is an imaging specific finding.
55:18
It's just a general gestalt that there is, um, enlargement of ventricles.
55:22
It's out of proportion of the degree of overall volume loft.
55:24
And that causal angle is narrow to the point where we suspect hydrocephalus.
55:28
It really is a clinical diagnosis.
55:29
And that's what I fall back on anytime someone tries to push me on, uh, on mph.
55:34
That's generally, um, what I do.
55:36
So someone's asking what the stroke volume cutoff, is there any role, uh, oh,
55:41
sorry, this thing's, man, I'm getting a lot of questions real quick.
55:44
So the cutoff that I use is a hundred cause we have GE machines.
55:47
What is your opinion about complex kiari um
55:52
com? What, um, I'm not sure I understand what they're asking about Complex Kia,
55:57
r e Um, maybe, uh, maybe they could ask another question,
56:01
but maybe define what they're asking exactly. Um, and then, um,
56:06
in which side or C SF Venus officials more frequently and in what level? Ooh,
56:10
that's a good question. Um,
56:13
I find them to be more common actually on the, uh,
56:19
right side of the patient and more common at the thac lumbar junction.
56:24
And I think the reason that that happens is because the pressure actually,
56:29
yeah, it, I think it's just because the pressure is a little bit lower. I mean,
56:33
there's less interference with fistula effor cause you're closer to the ivc,
56:37
closer to the, uh, uh, also closer to the, uh, AGAs hemi AGA system,
56:42
et cetera. Um, let's see. What about agen of the transverse sinus? Oh,
56:46
that's a good question. So, um, agen of the transverse sinus,
56:50
I've never seen bilateral,
56:52
I've seen it unilateral in the setting of a a a a contralateral stenosis.
56:57
And in those cases, those,
56:58
a lot of those patients do end up benefiting from a stent. Um, but yeah,
57:02
agensis is a good thing to think about when you're,
57:04
when you're talking about stenosis.
57:05
So that's where the gradients come in when you're actually measuring the
57:08
pressure gradients, you, you need to make sure it's significant. Um, man,
57:11
these are, these are really good questions. They're coming so fast though.
57:13
I don't know if I'm gonna be able to answer 'em. Let's see here. Um,
57:17
regarding follow up hydrocephalus,
57:19
is there any objective score that can be used?
57:23
I assume they mean for nph H Um, no,
57:27
there's not really an objective score that, that I've used. Um,
57:31
I'm not aware of one. Um, that being said, um,
57:34
somebody's saying Crocker k r ad scale, I've not used that. I just,
57:39
I look at the colossal angle and it oh, KIARI 1.5. Yes. Um,
57:43
very nice presentation. I think that,
57:45
I think that's most of the questions actually.
57:48
How can we determine normal pressure hydrocephalus in the background of
57:51
generalized atrophy? Oh, that, that,
57:53
that's actually gets back to the colossal angle. So that,
57:55
that's a good question too. I mean,
57:57
that coastal angle tells us basically that it's probably more likely hydro and
58:01
not just generalized atrophy,
58:03
because the normal coastal angle is actually usually greater than a hundred.
58:07
If it's less than 80, it's more likely to be, um, mph.
58:10
But I don't use any other scoring system because I don't think anything's
58:14
reliable. I, I just as a gestalt I look at it,
58:17
if the ventricles are bigger than I would expect for the degree of volume loss,
58:20
then I just measure the colossal angle. If it's positive,
58:23
then I just say this is suggested to MPH in the appropriate clinical setting.
58:26
And then the, the neurologist evaluates the patient and they say, yes,
58:28
it looks like mph, then they send 'em back to me.
58:31
Then I end up doing the LP or a three day float of, of a three day, uh, uh,
58:36
lumbar drain trial. And if it's positive, then they go to shunt occasionally,
58:39
like I said, I'll use, uh, I'll do a flow study and, and just see if, um, if,
58:43
if, if it jives with my, um, with my flow diversion, uh, or sorry,
58:47
my temporary flow diversion, particularly if they're immobilized and they,
58:49
they can't walk. Um, let's see. Do you use CT and Mr.
58:53
Milo in the same setting? How much Mr.
58:58
Max contrast can be used? Okay,
58:59
so I personally never use more than 0.5 intrathecal
59:04
gadolinium. I usually stick to 0.3 as far as I'm aware,
59:08
there's no f d a approval for that. I normally just, uh,
59:11
tell 'em it's off label and actually put it in the consent that this is
59:14
off-label. Um, and that's, that's usually what I will do. Um,
59:18
do use the MPH scale from Scandinavian. Now I don't use that NPH scale. Um,
59:23
let's see about intracranial hypertension.
59:27
What do you think is the most reliable MRI finding? Ooh, um, honestly,
59:32
optic disc cupping and, and optic nerve sheath. Um,
59:36
dilation I think is probably most reliable in intracranial hypertension.
59:39
I see empty sellers all the time and it means nothing. Um,
59:44
the other thing that's pretty reliable is diffuse sinus narrowing in the
59:48
appropriate clinical setting. Along with those, um,
59:51
with those findings within the opting nerve, she,
59:53
so if you have diffuse sinus narrowing, um, particularly, um,
59:56
along the SP sidel sinus as it transitions into the transverse sinuses,
59:59
if you have narrowing right there,
60:01
a lot of those patients have have NPH now patient 50 female CSF
60:06
leak history of direct head trauma last 20 years.
60:11
What suitable imaging needed? So from noses? Oh, okay.
60:16
Um,
60:17
so I normally do a super thin section max face to see if I can see a dehiscence.
60:21
And then I actually go to nuclear medicine pretty quickly.
60:24
And when I do the nuclear medicine scan for those patients that I'm looking for,
60:28
uh, C S F rhino, I actually raise their pressure substantially.
60:31
I take 'em to 35 centimeters of water with Elliot B pollution, um, right before,
60:35
sorry, right after I inject the nuclear medicine tracer.
60:38
So first thing I do is I do my lp,
60:40
I confirm I'm in intrathecal with a little bit of, um, uh, iodinated contrast,
60:44
then I inject the radio tracer, then I inject a lot of Elliot bees,
60:47
or you can inject saline and get 'em up to 35 just intermittently monitoring.
60:51
Also, you, you kind of tolerate to their,
60:52
that you kind of titrate with their tolerance. Some people won't get to 35. Um,
60:57
if you can get 'em to 30, even 25, that makes a difference.
60:59
But if you leave 'em with a, a c SF hypertension, uh, or,
61:03
or even a normal pressure, you might not elicit the leak. So I,
61:05
I typically try to get 'em around 30 if I can, um, even for that. And then,
61:09
then I normally have my EMTs packed the nose with pledges right before and we
61:13
actually count the pledges. Um,
61:15
you can also do thin section T2 s through the anterior skull base and try and
61:20
see if you can see the C S F leak. Um,
61:22
there are some cases where you can see that hiss it a little bit better on thin
61:26
section, um, T2 cube or Fiesta Imaging, um, compared to,
61:32
um, uh, compared to just, uh, a plain, um,
61:37
maxillofacial ct. But that being said, seeing the desistance,
61:39
you're relying upon seeing a tiny, usually a sci,
61:43
a tiny ence seal basically is what you're looking for.
61:45
How to differentiate small perineural cyst from SPH leaks on plain
61:50
skin. You can't, um, it's impossible if the patient doesn't have, um,
61:54
if patient has a prior MRI of the brain and they don't have a, um,
61:58
they they do not have a positive burn scale score and they don't have symptoms
62:03
of CSF leak, then it, it clearly is just a normal peroneal cyst.
62:06
But if they have signs of CSF leak on MRI and or clinical,
62:10
then you just have to report that there are perineural cyst in all these
62:13
locations. Um, and in general, if you've already done, um,
62:18
that you could, you could ask your tech, you know,
62:20
to go ahead and try and do a fat sat t2,
62:23
but that's gonna add a lot of time to your scan.
62:25
The fat SAT T2 is really useful.
62:27
If you go back or if you remember back earlier when I was showing a fat sat T2
62:30
of that young, uh, lady who had had the lumbar puncture,
62:33
you could see stranding in the epidural space. So basically on,
62:37
on the fat sat T2 around a perineural cyt,
62:39
you can see stranding in the epidural, uh,
62:41
epidural fat when there's a positive D S F leak and that,
62:44
that can help you kind of focus on one perineural cyst. I,
62:47
I think that probably is what you're asking, basically a fat sat t2,
62:50
but on a standard scan you can't tell if a perineural cyst is,
62:53
is gonna be a leak or not.
62:54
You need the fat sat T2 at minimum to be able to tell if it's,
62:58
if it's really suspicious threshold for stroke volume to be
63:03
CSF flow study. Uh,
63:08
I'm not sure what this, I, again,
63:10
I go back to the a hundred microliters for GE or 42 microliters for Siemens. Um,
63:14
but I still rely ultimately on, on a, um, on a temporary float, uh,
63:18
temporary float aversion meaning like a, um, a lumbar puncture or a, uh, or a,
63:23
um, a lumbar drain restricted diffusion and optic nerve head and pseudotumor.
63:27
Yeah, that is,
63:28
that is actually a finding that we see not infrequently or optic disc cupping
63:31
or, or restricted diffusion. That does happen for sure. Um, let's see.
63:38
Is there any significance using needle gauge size when doing seal? Yes,
63:41
there is significance. So, um, when you're doing a high flow,
63:45
when you're looking for a high flow leak, um,
63:47
somebody asks a question about what size gauge needle to use.
63:50
So if you're doing a high flow leak,
63:51
you need to be able to inject the contrast really quickly if you're doing a dsm.
63:54
So I use a 20 gauge needle to be able to inject it quickly. Um, that being said,
63:58
if you're doing a high flow leak in ct,
64:00
presumably you got the head up for a while and you're actually injecting slowly.
64:05
The, the sort of, the distinction is, is that in, uh, in A D S M,
64:10
the patient can't be moved. So in A D S M you already have the head down,
64:14
so you have to inject quickly and wait for it to flow up.
64:16
Whereas in CT you can inject the entire volume and drop the head.
64:20
You can't drop the head and do a DSM as easily. Um, you, it's,
64:24
it's theoretically possible if your table moves really,
64:26
really quickly for you to drop the head and quickly subtract and then before the
64:30
contrast runs in. Um,
64:32
but I tend to use the smallest gauge needle I can for the low flow leaks because
64:36
I don't want my puncture to obscure the leak basically.
64:40
So I try to use the smallest thing I can. Um, the, the,
64:43
the good compromise though is the gerdy set. So the,
64:46
there's a gerdy set out there that I think has a 20 gauge coaxial.
64:49
It's about an inch and a half long and then a 22 gauge ping cam that goes
64:54
through it. Um, I've had a lot of luck with those. So, um,
64:57
for those sort of leaks where I'm not sure if it's high flow or not,
65:00
if I think it may be a type two lateral leak and it may be kind of high flow,
65:03
um, then I'll use that gerdy set. It's a good compromise.
65:05
But when I think it's a low flow leak,
65:06
I tend to use a 25 basically a discogram set, more or less. Um,
65:12
why should we need to use diagnostic lp? Oh,
65:14
what needle should we use for diagnostic LPs? Um, I, I tend, like I said,
65:19
I tend to use that gerdy set for a diagnostic LP if I'm worried about C S F
65:23
leaks. Um, if I'm not worried about C S F leaks,
65:26
I'm trying to do a very quick, um,
65:29
assessment of whether or not someone has nph. I tend to use a bigger needle, uh,
65:34
cause I don't wanna sit there all day. So I,
65:35
I tend to use a 20 gauge needle and I have this 33 inch one meter
65:40
tubing that I hooked to the 20 gauge needle and I actually raised the head of
65:43
the bed up and I actually dropped the tubing down lower. And um,
65:47
generally I can finish a high volume LP in about 15 minutes. So,
65:52
um, our epidural C s F leak surgical emergencies, like hemorrhages,
65:56
they can be if they're really fast. So if an epidural leak is extremely fast,
66:01
um, to the point where the patient's comatose it can be an emergency. Um,
66:05
I've run into those situations where I don't even have time to localize 'em and
66:08
I tell you exactly what I do.
66:09
Usually the epidural space is so engorged because of the leak and it's so
66:14
soupy for lack of a better way of saying it.
66:16
And you can put almost anything into the epidural space.
66:18
So I'll take a vascular forefront sheath and actually just use like a toy
66:21
needle, uh, and get my toy needle into the epidural space.
66:24
I take a really steep angle into it and then once I get in there,
66:27
I'll put like either an oh three five or a nitrix wire into the epidural space
66:31
and I'll put either a four French radial or, or just a small four French, um,
66:35
vascular sheath into the epidural space. And then I'll put like a vert,
66:38
a four French Burt or a four French Bernstein catheter over an oh three five
66:41
wire. And I will run a catheter all the way up the dorsal epidural space to the
66:46
cervical thoracic junction and I will patch them with a hundred,
66:48
125 ccs of blood.
66:50
Generally injecting about two to four ccs per level titrated into the patient's
66:54
ability to tolerate it. Um, but that's how I deal with,
66:57
with emergent leaks like where I don't really have time to localize it like the
66:59
patient's comatose. I I need to do something fast,
67:02
so I'll put in a catheter and just inject a lot of blood. Uh,
67:05
what is the anticoagulation policy for venous stent and follow up?
67:08
So I normally will give them aspirin and Plavix for three months.
67:12
I do check levels for those. Um,
67:14
so I check P t y 12 level and Aspirin Verify now,
67:18
and I send out full blood platelet aggregation for a D p and um,
67:22
and for the AA level, um, because they're more reliable,
67:25
but they're a 24 hours send out lab for us.
67:27
So if the PT y 12 level is less than 1 94 and the aspirin's less than five 50,
67:32
I normally feel comfortable proceeding. Um, and then I'll wait for the,
67:35
the final numbers, but generally I like them to to be below 50%, um,
67:40
when I'm, when I'm putting in a venous stent on the a d P activation. Um,
67:44
and then for aa I like them to be about the same below 50%.
67:49
What is flow compensated fiesta for the spine? Um,
67:54
oh, so basically, um, essentially we,
67:58
we put on EKG leads and we're actually timing the patient's heart and we're
68:03
compensating for flow based on when their heart beats.
68:05
Cuz when their heart beats their brain pulsates essentially, I mean, and we're,
68:09
we're actually compensating for essentially when their heart beats and when the,
68:13
when the, when the CSF is moving. So it's almost like gating, I mean, and,
68:17
and you can do it that way. Um, another way there,
68:20
there are a couple other ways to do it, but I'm not a physicist full, full, um,
68:22
full disclosure, there are other ways to do flow compensated fiesta,
68:25
but you can gate based on the heartbeat. Um, and then there are,
68:28
there are a couple other ways to do it, but in in general, um,
68:30
flow compensated fiesta is helpful because it removes a lot of different
68:35
artifacts.
68:35
Fiesta itself removes truncation and then the flow compensation removes the flow
68:39
artifacts so that you can see those epidural collections much more easily.
68:42
But you can either, you can either, uh, gate to the cardiac cycle, uh,
68:45
or there's a couple other tricks. And again, I'm, I'm not a, I'm not a,
68:49
I'm not a, uh, a physicist side. I don't know all of the, the tricks to that,
68:53
but I have a very good physicist here and they,
68:55
they do really nice low compensated fiesta for us. And uh, and that's,
68:59
that's how I see a lot of my CS athletes. Okay.
69:02
Looks like we have all of our questions and answers, um, done. Um,
69:07
I really appreciate everyone sticking around to the very end and I,
69:10
I hope this was helpful.
69:11
I know it's a lot of information to throw at y'all at once and it, it's,
69:14
it's as comprehensive as I can get about the specific topics that I'm talking
69:18
about in this amount of time.
69:20
Yeah. Dr. Pinnell, thank you so much for that lecture.
69:22
Thank you so much for answering all of those questions and for everyone else in
69:26
the audience for asking such great questions. We really appreciate it.
69:30
You can access the recording of today's conference in all our previous noom
69:33
conferences by creating a free m r I online account.
69:36
And you can join us next week on Thursday, June 8th at 12:00 PM Eastern,
69:40
featuring Dr.
69:41
Jenny Ben Carino for a lecture on overstretching injuries of the calf
69:45
musculature tennis leg, and more.
69:47
You can register for this free lecture@mrionline.com and follow us on social
69:51
media for updates on future noom conferences. Thanks again and have a great day.
Jeffrey Scott Pannell, MD
Director of Neurointerventional Surgery
University of California San Diego
© 2026 Medality. All Rights Reserved.
