0:01
Welcome to SNI Digital Innovations in Learning in association with UCLA Neurosurgery. Linda Liao, chairwoman and its faculty are pleased to bring you the UCLA Department of Neurosurgery 101
0:22
lecture series on neurosurgery and clinical and basic neuroscience.
0:30
This series of lectures are provided free to bring the advances in clinical and basic neuroscience to physicians and patients everywhere. One out of every five people in the world suffer from a
0:49
neurologically related disease. The lecture and discussion is on the methods to identify the electogenic zones.
1:00
stereo EEG versus sub-dural evaluation. The lecture is Arya Fowler, Assistant Professor of Neurosurgery, Pediatrics and Health Policy and Management. From the David Geffen School of Medicine at
1:21
UCLA, Los Angeles, California, United States of America Which is basically a non-invasive evaluation. That's the MRI, the PET scan, neuropsychology testing, EEG,
1:38
history physical, the important parts of the epilepsy evaluation. Phase two is the invasive EEG portion. So not every patient will get this. So sometimes we say this patient's a skip candidate,
1:52
meaning we skip phase two and go straight to phase three the phase three is actually the definitive procedure.
2:00
as you'll see and will discuss that not every patient is appropriate to skip phase two on. And sometimes we need to do that invasive EEG evaluation. So any sort of implant
2:11
is considered phase two. So have no financial interest, no disclosures. So if you look at the epilepsy evaluation, you see that on the left side going from non-invasive to invasive and
2:24
intraoperative studies and each one of these tests tells you part of the picture. So the important thing with an epilepsy evaluation is that you cannot look at any one modality in its entirety. You
2:38
cannot just look at an MRI, say there's a lesion, so that's causing the epilepsy and I'm just gonna go after that or just look at semiology or PET scan. So you gotta really look at all of this
2:48
information together and each hospital has a different
2:54
workup strategy So for us here at UCLA, you may know that FDG pets. for use in epilepsy was first developed here by Harry Chagani many years ago. It was popularized and it's not being used many
3:08
places. So that the MRI, 3T MRI, PET scan, video EEG, and neuropsych testing for patient six age, the age of six and above is part of our standard workout.
3:22
And really, this is what you wanna get at. So these are the main concepts in epilepsy surgery You may have a structural lesion. There may be a larger irritative zone. So this is the zone where on
3:34
ECOG, you may find spikes, epilepsy form discharges. There's a seizure onset zone and there's anictal network. So there's different strategies for removing the epileptogenic zone. Keep in mind,
3:48
the epileptogenic zone is a theoretical area. We cannot image it. The only way you know you fully remove the epileptogenic zone is that the patient sees you're free many years later.
3:59
you know, a tumor surgery where you can come out and tell this patient, I got it all. You can't do that in epilepsy surgery.
4:06
You have to really wait to see if you got it all. So the strategies here, and you can look at, this is sort of a superimposed, but each one of these tests tell
4:19
you. And, you know, there's ongoing investigation, whether, you know, HFO, for example, I'm not gonna talk about much, but this is sort of a newer biomarker for the zone that generates
4:31
epilepsy, high frequency oscillation. Again, discovered here from UCLA many years ago, and now there's clinical trials looking at the importance of resecting HFO areas in achieving seizure freedom.
4:44
So what are the indications for invasive EEG or phase two when you don't have an MRI lesion? So sometimes you hear the word, we throw around the words, This is a non-lesional case. What that means
4:56
is that the MRI is normal Doesn't mean that there's no underlying Legion. It's usually what we're looking for is a very subtle cortical dysplasia, but we use that term a lot, non-lesional. Or
5:07
another case is a multiple lesion. So tuberous chorosis being a perfect example where you have multiple lesions and you don't really know which one is causing the epilepsy. If your non-invasive data
5:18
is inconclusive or you have divergent non-invasive data where not everything agrees or points to the same region, you may want to do it in base evaluation Other reason to do it is if you want to map
5:29
eloquent cortical function precisely, motor, speech, sensory,
5:36
or secondary indications or to further corroborate the epileptogenic zone. So you may have an idea where the seizures are coming from. In what general area, you may know right frontal lobe, for
5:45
example, or right parietal lobe, but you want to know with a lot more accuracy where exactly. So that's another reason to sometimes place a grid. You want to map out eloquent function
5:58
prognosticate for a patient. It's very different, you know, with epilepsy surgery, there's no guarantee. So it's difficult to consent to patient and say, Look, we're gonna do the surgery. It's
6:07
likely you're gonna be seizure free. It's likely you're gonna wake up with a deficit. Or it may, you may or may not wake up with a deficit. It's a hard thing for some patients to agree to. So for
6:18
some specific situations, you may wanna place invasive EEG just to better delineate the borders of your resection and better inform the patient of their expected outcome. So in terms of
6:30
decision-making, you wanna make sure, first of all, you need to have a reasonable hypothesis. Without a hypothesis, placing electrodes in the brain is not gonna yield any useful information. You
6:42
wanna look at that divergent or inconclusive, non-invasive data and see if it can be explained by a no limitation of the EEG and functional imaging data. For example, sometimes if you have but
6:55
epileptogenic foci deep in the brain. let's say in the interhemispheric region, you may get falsely lateralizing EEG. Keep in mind the surface EEG electrodes are far away from the interhemispheric
7:07
area. So even though they do a better job recording from the cortical surface, if you have a deep lesion or an insular lesion or a singular gyrus lesion or a bit of frontal, you're not gonna
7:18
capture it necessarily directly on the electrodes that are over closest to that area.
7:26
You wanna look at whether there's other non-invasive tests that could be done to eliminate the need for invasive EEG. So these are other specialized tests, spec scan, meg scan, whole host of other
7:36
tests that can sometimes be done to avoid intracranial EEG. You wanna make sure that the added information from the invasive EEG will be likely to change the operative plan. And you also wanna make
7:48
sure that that added information that you're looking to get cannot be achieved with just intraoperative ECoG. That's electro-corticography direct recordings from the brain. Usually it's interictal
8:01
data. It's very rare to actually record seizures during surgery. You also wanna look at the medical comorbidities that may contraindicate invasive EEG. And then there's several, there's a few
8:12
modalities of invasive EEG which we'll talk about. So why face-to-evaluation? It gives you a higher signal-to-noise ratio. The electrodes on the surface of the brain require a large enough
8:25
discharge from the surface of the brain for that to travel through the dura, the skull, the scalp, and finally we picked up by an electrode on the surface of the scalp. But when you go invasive,
8:38
you really eliminate all of that. So there's no muscle artifact, higher spatial resolution, and you may see a seizure pattern before the surface EEG. What are the contraindications invasive
8:52
monitoring? Psychiatric illness. So for patients that, or at risk for psychosis or schizophrenia or depression, you wanna be careful because they don't do too well with the invasive evaluation.
9:05
Age less than two years used to be a contraindication, but we've sort of pushed a limit and I'll show you how we do that, how we can do these operations and infants. And also, if you don't have an
9:17
adequate hypothesis, again, if you don't have this, you're unlikely to be successful So invasive monitoring has been around for a long time. Peter Krauss, you see, while they're Penfield. A lot
9:31
of what we know about the brain is actually from the field of epilepsy, knowing what the functions of the brain are what different parts do. And there's a huge and fascinating, there's a long
9:42
history that's quite fascinating about epilepsy. I actually encourage you all to read it. There were procedures being done, triphonations and all that for epilepsy and that history is actually
9:55
quite interesting. But a lot of what we know is through this. And this has been around for a very long time. Our techniques have changed, but we've been interested in monitoring the brain for a
10:06
very long time. So some special considerations for phase two evaluation. So this is a non-lesional MRI. Again, there's no particular lesion that you can identify. But if you have a reasonable
10:18
hypothesis that the seizures are coming from a focal or at least lateralized area, you may want to implant electrodes here. The other consideration is focal cortical dysplasia. This is the patient
10:29
with a left insular dysplasia here that spans into the frontal perculum. The challenge with cortical dysplasia, which is the most common ideology that we treat in the pediatric side, is that it's
10:29
difficult to know the full extent of the lesion. So what you may be seeing is the the tip of the iceberg. And it is a mistake.
10:54
to just go and resect what you see on MRI because quite often the area around it could also be this plastic. So in these situations, invasive EEG could be helpful. And the other case, this is the
11:07
most common case of genetic epilepsy in children, tuberous chorosis complex. Children are born with multiple tubers in the brain, often refractory to anti-epileptic medication And one of the
11:20
greatest challenges with tuberous chorosis surgery is identifying the epileptogenic tuber or tuberous. So this is another situation that may benefit from it. So a lot of what we do in epilepsy
11:32
surgery, we spend a lot of time with a neurologist and you really work hand in hand. So
11:37
the one unique thing about this field is as an epilepsy surgeon, you got to try to learn epilepsy as much as you can from the neurologist and they got to try to learn surgery as much as they can
11:50
because if you don't understand the other person, You're not going to be very successful because you're, you know, your success is You're only as good as your weakest link when it comes to epilepsy
12:01
surgery whether that surgery the surgeon neurologist radiologist and so on so an important part of dissecting a seizure simiology which is really the play-by-play of a seizure, right? What is the
12:16
first symptom? What is the second symptom? What is the first sign that we see and you really based on what you see? You can pinpoint where the seizures may be coming from usually to one or two
12:26
regions of the brain with different degrees of reliability so
12:35
you know Some of this you may know, but I would encourage and I can share these slides to take a look at and if you're interested in it you can sort of you know, for exampleictal fear is Hugely
12:48
related to an amygdala lesion with a high degree of reliability nose wiping is a sign of insular seizures. Or if you have a choking sensation, that's a sign of insular seizure. So these are, it's
13:02
really fascinating. I encourage you to read it. And then also with the seizure itself, you wanna look at the different phases of the seizure. The pre-ictal phase, the pro-drome, and the aura, I
13:14
can tell you things about a seizure, theictal onset. For example,
13:20
a patient has aphasia during a seizure, you know that it's their dominant hemisphere
13:25
And, you know, the post-ictal phase, for example, a patient with a Todd's periesis, so a weakness in one half of the body, you can usually infer the seizures coming from the contralateral
13:36
hemispheric region. I will say in pediatric epilepsy, it's much, much more challenging because the seizure immunologies are actually quite bland. And, you know, obviously they don't report any
13:50
auras So you're looking at.
13:55
Sorry, so you're looking at, you know, sometimes something like infantal spasms, which is a very short brief generalized looking seizure that could be coming from a very focal area. And that's
14:07
why the pediatric EEGs are a little bit more challenging to read and try to identify whether seizures are originating from.
14:17
So rules for implantation strategy. So you have this discussion as a group
14:22
This is what you want to target when you implant electrodes. You want to go after the cortex involved in the seizure onset as well as the early propagation. You want to cover your alternative
14:32
hypothesis. You want to delineate a border of the epileptogenic zone. So it's not enough to say it's anterior-singular gyrus. You may want to know how far back on the anterior-singular gyrus or how
14:43
far forward. And you can use stereo EEG or electrodes implanted to try to delineate that border As a surgeon, you need to know where to cut or where to stop.
14:56
And quite often, the neurologists don't give you that information, right? It's sort of based on what you don't tell you that's this general location. And you would have to try to figure that out
15:07
with your EG implantation strategy. You want to cover nearby eloquent area. And of course, if there's any underlying lesions. What do you do after a phase two evaluation? You really want to
15:19
minimize the amount of days within planted electrodes. This is why, when you see me do these cases, I always implant on a Monday and I try to expand on a Friday. I don't like to leave a patient
15:29
with electrodes in their brain over the weekend, where not much is being done. And, you know, we try to know it's very difficult to know how quickly your patients are going to have seizures. You
15:38
need to work very closely with the neurologist. Sometimes you want to cut down some of their medications, even prior to surgery. You really want to optimize. You generally want to get about three
15:47
seizures. Now, if they're a stereotype, then they all look the same. That's all you need. But if you get three seizures and each one looks different, may need more seizures to try to quantify,
15:57
you know, how many seizures look, you know, which way.
16:02
You don't, you also don't want to get clusters of seizures, so three seizures in a row is not helpful. You do want them to be spontaneous, you want them to be their habitual seizures, and you
16:12
want them to be spaced out. You could consider a medication wean as an alternative, if the EEG looks pretty bland after day one or two, you may consider a wean, but you don't want to generally do
16:23
this right away.
16:26
So with subdural grid evaluations, when we, on Friday, when we go back to remove the electrodes, I like to do the definitive procedure at that time, because you already have a craniotomy,
16:36
whether it's a resection or placing RNS, you can do that at the same time, but with stereo EEG, we try to delay that next stage. First of all, because you don't have to do a craniotomy to remove
16:47
it, you just remove the electrodes at the bedside, and also you want to try to space out the procedure to lower your infection risk. So indications for subdural grids, if you have a courtically
16:58
based lesion with a ill-defined epileptogenic area, again, you need to have a reasonable hypothesis. And also you wanna make sure that you can't obtain your information through an awake craniotomy.
17:09
There's definitely limitations on the pediatric side or how young you can go. There's been groups that have done awake craniotomies in children, can be done, but you really need a very cooperative
17:18
child to do that on. And again, keep in mind, you can map the brain in this region, in this fashion. You cannot delineate your, you cannot map seizure onset in an awake patient. So that if you
17:34
need to know where your seizure onset's coming from, then you wanna do an extra operative e-cog, meaning you implant the electrodes. These are, what do these grids look like? They're very soft,
17:44
they conform to the brain. This is a eight by eight grid, which is one of the most common ones that we use. This is a strip we use very common And these are interhemispheric. grid. So these can
17:55
be slid down the midline if you want to cover those regions.
18:04
Here's an article showing feasibility of a weight craniotomy in pediatric population. In this group, it was mostly done for tumors, but also for epilepsy. And again, there are potentially some
18:16
serious complications that can occur. But again, in OE, well selected patients, especially at a center that has experience doing a weight craniotomies in children, should be able to do this as
18:29
well.
18:31
So in terms of the subdural evaluation, so one of the most important things I spent a lot of time before the case with folks like Dr. Solomon, Andy Frew, Patty Walshaw, creating the case, you
18:43
really want to import all the imaging modalities that you have done and build that case into brain lab, brain lab or whatever neuro navigation software you're using
18:55
'cause if you do a craniotomy and stare at the brain, you're not gonna know what to do next, right? There's really no clue as to where the abnormality is if you don't have those imaging modalities
19:06
fused. You really wanna do a white craniotomy and large brain exposure. This is not, you don't do epilepsy surgery through a superorbital tission
19:16
or eyebrow cut or anything like that. But you really wanna have good exposure of the brain because you're kinda gonna be limited to what you can record if you don't expose the brain So this is
19:26
usually here in trauma. You kinda see this much exposure of the brain.
19:31
You wanna place your grid directly on the surface of the brain. You wanna customize the edges of your grid. You wanna make sure there's no sharp edges pointing onto the
19:42
cortex. Very important to look at the veins and make sure you don't have any venous compression or venous hypertension.
19:52
secure the tails of your electrodes to the Dura. I've seen one case in residency on camera where it's an adult patient in Toronto that pulled out their subdural grid during a seizure, a hyper motor
20:06
seizure. So you don't want that to happen.
20:11
I always do an expand style Duraplasty. You don't want to close the
20:14
Dura on a, especially in a young child that has a very full brain. You place an electrode. You don't want to bring those Dural edges together without an expand style Duraplasty. The bone flap is
20:24
free floating. It allows room for
20:28
pushing the bone flap out if there's any swelling.
20:33
And lots of central attack of sutures, long tunneling of the electrodes to minimize the risk of infection. I place these patients on IV antibiotics for 72 hours and IV steroids for 72 hours again to
20:44
decrease some of the swelling and always send cultures at the time of electrode removal in case you have a delayed infection. really want to understand the venous anatomy, make sure especially your
20:56
larger
20:58
venous structures, again, make sure there's no compression on them.
21:03
So all the patients that get a subdural grid, at least on the pediatric side, they spend the entire time in the ICU, they need Q one hour
21:14
nursing vital checks. You really want to make sure there's no mass effect from the electrodes and hematoma formation We had one patient since I've been here last seven years, it developed a very
21:27
delayed hematoma under the
21:30
subdural grid, actually was asymptomatic, but the way we knew there was a problem was that the EEG signals became, we started to lose the amplitude. So we got a CT scan, this is what it looked
21:42
like, and we had already obtained all our information at that stage, so we actually took the patient to surgery,
21:51
subdural that was underneath the grid. Evacuated actually ended up doing the resection at the same time at 3 am, but. So these things can happen. These are not inconsequential. CSF leak, you've
22:03
got to monitor for that. I don't routinely change head wraps. I know some surgeons like to change it on date, whatever, two or five or seven. But you generally, if you do a good job closing,
22:17
you really don't want to mess with the head wrap at all 'Cause if you're taking scissors next to the head of a patient with epilepsy and grids and electrodes and wires coming out of their head, you
22:27
have a high chance of cutting one of those electrodes, which means we gotta go back to the operating room.
22:33
And then obviously monitor for infection. And sometimes these patients can go into status, especially as you're weaning their medication. So one of the other things I commonly do is I place a pick
22:44
line for these patients during the hospitalization. weaning their medication, sometimes you're doing brain mapping where you're stimulating the brain. And if these patients have prolonged seizures,
22:56
sometimes they can lose their peripheral IVs. So you wanna have a reliable way to stop their seizures if you need to. So I always place a pick line on the Monday where we implant these electrodes
23:10
and then we'll take it out Friday, they're ready to go home.
23:14
So this is a systematic review meta-analysis that was done on subdural grids, 21 studies, 71
23:22
risk of infection, 4 risk of intracranial hemorrhage, 24 risk of increased ICP, and 3 rate of an anticipated surgery. I think these numbers are a little bit on the high side, but it does tell you
23:33
that there's a whole host of complications that can occur with subdural grids and it's not inconsequential. But this is the type of information that you really get and this is what patients and
23:44
families love, where you, this is a cartoon of their brain. with the grids, showing you exactly what the spikes are, area of slowing, seizure onset is these red circles. Fast ripple is a,
23:58
again, a biomarker. It's considered a HFO that we talked about earlier.
24:03
In this region, you can see the HFO co-localized nicely with the seizure onset zone, but you can also see them in places further away. But really, you get this type of information. You get your
24:16
neuropsychologist for us, Dr. Walsh will come by and do brain mapping, where we are stimulating different electrodes, getting the patient to do different things. And I encourage actually all the
24:25
residents, at some point in residency, to sit on at least one of these sessions, where they do brain mapping. It's important to see that and how that's done. Usually that's done with the
24:35
neurologists in the room as well, 'cause you can, you look for things like after discharges, patient goes into a seizure, you wanna be able to stop that seizure during that period. But we get
24:44
this sort of map, We try to figure out where the central sulcus is. And then we draw out exactly the boundaries of our resection. So for here, you want to remove the right frontal lobe, anterior
24:58
to the motor area, and the near job is an epilepsy surgeon is to execute that plan, and that's what you get. So this is a specific case example. This is a 20-year-old male, had a previous
25:10
resection in the post-central gyrus. Pathology was chronic meningo encephalitis, but he really had very focal seizures of his left hand, he had chronic movements of his left hand, and that really
25:22
prevented him from driving. That was the most important thing for him. And for a patient like this with rolandic epilepsy, so motor sensory region, really, the best study is a great evaluation.
25:35
So here, because it was a redo, it's hard to get great exposure with a redo without having to dissect an adherent dura from the brain We did a smaller opening. a place to grid in a very focal area.
25:50
This is the central sulcus in
25:53
blue. This is anterior. This is posterior. And you can see very nicely here that electrode 14 is exactly where the seizures start. And they stay in that electrode for a very, very, very long
26:07
time. So in this particular case, we're able to do a very, very focal, small resection right there And this patient has been seizure-free. And we also mapped out the hand function and ensured
26:22
there was no movement right there. So there was no function right there. So we're able to do that. And this patient's able to drive. And it's been seizure-free for a long time. So for this
26:31
scenario, subdural grid evaluation is the better modality. I'm going to transition and talk about stereo electroencephalography stereo EEG. This is a methodology that was introduced by the French.
26:46
And it refers to an anatomical, electrical, and clinical methodology aimed to verify the presurgical hypothesis on location extent of the epileptogenic area. So you really need to consider all
26:57
three. And you need to really have a good understanding of again, semiology and sort of how that seizure is originating and spreading through the brain to be able to determine where to place your
27:08
electrodes. The strengths of EEG is that it's thought of as being less invasive But my particular feeling about that is, again, you guys are all neurosurgeons, you know that just because something
27:22
is less invasive doesn't mean it's less dangerous. When you're putting electrode deep in the brain, insular lesion, region or the hypothalamus or those areas, there's nothing minimally invasive
27:36
about that. The cut on the scalp is small, but you can get some really bad complications from it but it does allow you to record. from deep and varied cortical structures. It allows you to get
27:50
better temporal spatial understanding of epileptic activity. And sometimes electrodes can also be used to stimulate in order to localize eloquent cortex as well as to illicit seizures. You can do
28:00
that with subdural grid mapping as well. So these are stimulation induced seizures that some people like to do. It's controversial. There's some papers saying that if you stimulate and cause a
28:11
seizure in an area at a very low threshold that may be your epileptogenic area, but more classic way is you kind of want spontaneous habitual seizures, not stimulation induced. So you're going to
28:25
hear these words a lot. S-E-G, we're doing depth to you. It may all mean the same thing, or how do you tell one apart? Some people like to call it S-E-G. Some people like to call it depth. And
28:36
some people get pretty angry if you call it by the wrong name. So I'll fill you in on that a little bit So the electrodes are actually different. So SEEG electrodes are smaller in diameter, higher
28:50
resolution, look at all the area of contact compared to the depth electrode. Usually the tip can also record, but there is no internal stylus. So these are, when you place these electrodes, you
29:02
gotta create, you gotta place a stylus before it, create a path, and then place these. What you're seeing that we're using here is we're using depth electrodes, right? So this is, I don't think
29:17
Dr. Fried uses SEG electrodes. He uses depth electrodes. I also use depth electrodes, but in the SEG methodology, I'll show you that, what I mean by that. But these depth electrodes are larger
29:28
in diameter, lower resolution, and commonly don't have electrode right at the tip, but there is an internal stylus. So these do have their own, they can create their own trajectory.
29:43
So when we say depth evaluation versus SEG evaluation, what do we mean? So they both actually originated from Banco and Telac in France. The depth evaluation is an orthogonal approach. So all
29:56
these electrodes are placed orthogonally perpendicular. And it was popularized in the
30:02
United States
30:06
and actually particularly the Mayo Clinic. Its main purpose is to lateralize the epilepsy, particularly the temporal lobes. So it's helpful in telling you, is it the left or right temporal lobe
30:13
and whether it's mesial or lateral onset. These electrodes are usually placed in a very symmetric fashion. You get the same number on the left and right. And again, you want to do that just to
30:25
make sure that you're not biasing yourself towards one side or another. So you can really figure out which temporal lobe. The SEG evaluation is you use various trajectories. They don't have to be
30:38
orthogonal. And that's why I'll talk to you a little bit about the robot and why the robot makes it so much easier to do. to do SCEG evaluations. It was popularized by the French and Italians.
30:49
They've been doing this for a very long time, but recently in North America, a lot of places have, since we've gotten robots, a lot of folks have switched over to doing this. And to us, it's
31:02
very novel, but when we go to these conferences, the French kind of look at us. And we've been doing this for 30, 40 years. This is nothing new. So the
31:11
purpose of this, you can really target all neocortical locations. And generally, your implantations are more unilateral and asymmetric. It lends itself nicely for pediatric epilepsy. Because as
31:23
you know in pediatric epilepsy, we're not usually dealing with temporal lobe epilepsy or mesial temporal lobe epilepsy. Usually in neocortical-based epilepsy. So that's what we mean with the SCG
31:34
versus depth evaluation. Now there's a gray area. There's a lot of folks that sometimes do a depth evaluation, but then put in certain other electrodes or make that asymmetric. So there's
31:47
definitely a gray area as well. But this is the classic definition of when we talk about an SEG versus depth evaluation. So functional mapping in SEG can be done. You can get pretty reliable motor,
31:59
sensory, and visual recordings. Also, they have the possibility of sub-cortical mapping. Speech is also possible to map with SEG electrodes But it's a lot less reliable, and you really need to
32:13
put a high-density amount of electrodes in the speech area to be able to stimulate from electrode to electrode. So in
32:21
general, if you're thinking that your seizure onset zone is close to the speech area, you're usually better off with a grid evaluation, or if you can, awake, intraoperative mapping of the speech
32:33
area. So limitations of stereo EEGs that you really get tunnel vision, right? So again, if you don't have a good implantation hypothesis. you're likely not to get a result that you can sort of
32:44
act on. It is more appetizing to do
32:49
steroid geodept evaluations and patients that may not have a lesion 'cause it's not considered a major or it's a more minimally invasive operation. So you are seeing, especially for non-lesional
33:02
cases, that a lot of folks would say, let's not do a huge craniotomy in a grid, let's just put the electrodes in 'cause it's sort of the complication profile is seen to be lower.
33:15
You know, and I'll just quickly tell you, in the past, when we didn't have RS and we were implanting subdural grids, I remember it's a very difficult decision because sometimes I remember this one
33:24
case in residency at tickets, it was a 14 year old boy who had a Rolanda Keplepsy and we implanted a grid, took a couple of weeks to record the seizures, But it was coming exactly from his dominant
33:40
hand. And he was a hockey player, he was a goalkeeper actually. So we were in situations where, for example, we would offer a resection in that area and he wouldn't want that because hockey was
33:53
very important to him and we didn't have RNS. So really, some of those patients would have to just explain the grid and then do nothing. So that was a very difficult thing to do. You've basically
34:03
operated twice on a patient with no therapeutic value. Nowadays, it's a little bit easier again with RNS and we have other things to offer. Back in the day, we were doing MSTs. Some of you may
34:16
have heard of this. This is multiple appeal transactions, really falling out of favor. It's quite traumatic to the brain. There were some patients that woke up in status, but really the idea was
34:28
for eloquent cortex epilepsy, you could, if you vertically make these vertical cuts, you can stop the spread of seizures, at least in a palliative way You're not gonna get seizure freedom. but
34:41
lower the seizure frequency, but preserve function in that area. We're also seeing it now, MSTs being performed in dominant hippocampus. So there's folks that are going in, in situations where
34:54
you cannot remove a dominant hippocampus, you do MSTs of the hippocampus. Again, this is being done mostly outside the United States that they don't have access to the RNS, but that's one
35:04
technique that we were doing mostly in the past
35:08
In terms of SEG and its complications, this is a large meta-analysis. A lot of these SEG studies, they're gonna count the number of electrodes. They're gonna say 22, 000 electrodes with their
35:21
complication rate is per electrode, but you gotta be careful 'cause you're putting 20 electrodes in a patient.
35:28
Patients don't count their complications by each electrode. They get a bleed from any of those 20 electrodes. That's a bleed, that's a stroke So we've got to kind of look at how these studies are
35:40
being reported. but essentially the surgical risk of morbidity is low and the risk of permanent neurological deficit is low. This can happen. If you do enough of these cases at some point, you're
35:52
gonna get an intraprenechimal hemorrhage or large stroke, there's patients that can die from this as well. So sub-droid hematomas and epideral hematomas are a little bit less common than
36:03
intraprenechimal hematomas. Now I will say the majority of hematomas you get are tracked hematomas that will, the brain will absorb over time. There's usually nothing to do, but you will get this
36:13
every once in a while. And so it's important to be aware of it and also make sure that you really need to do this extra step. There's been plenty of cases. You know, we just, there's a LPP
36:24
surgery techniques meeting where it's sort of, it's a global consortium of surgeons and we discuss cases and sometimes they're patients that really didn't need a phase two evaluation and you get
36:35
something like this. So it's really important to make sure that you require this extra step. So in terms of SEEG planning, again, you wanna look at the anatomy, see meology, and EEG to determine
36:46
where you want your electrodes. You wanna get a volume metric MRI. This is a 3T Tesla scan with thin cuts. In terms of blood vessel imaging in the past, you would get a DSA. But nowadays CT and
36:59
MRI are so good that you can just get CT within without contrast or MRI with and without GAD and use that to avoid your electrodes I will say if you have a vascular lesion or AVM or something that
37:15
causes more abnormality in the blood vessels, you probably still wanna get an angiogram
37:21
over CT or MRI. In terms of trajectory planning, you wanna maximize the contacts over the gray matter, the neocortical contacts. It's not very helpful to have the majority of electrodes just
37:32
sitting in white matter.
37:35
the ventricle, CSF spaces, and intraoperative CSF leak. If you start getting a CSF leak, all your trajectories will be off. You want to avoid the salsai, be at least two millimeters away from
37:46
the blood vessels, especially the peeled surface blood vessels, 'cause those are the blood vessels that are stuck between the skull and the brain, and they're not gonna move. The ones that are
37:56
deeper inside the brain are usually a little bit more mobile. If you want to avoid thin areas of the bone, particularly the squamous temporal region in young children, you want to be as
38:08
perpendicular to the skull as possible. You also want to make sure that the electrodes, they have some logical plan. You don't want to implant crisscrossing electrodes with different trajectories
38:20
and then have the neurologist figure out where the seizures are coming from. It gets very, very challenging and difficult to interpret that EEG.
38:29
And you also want to minimize a number of electrodes If you can place one electrode that will. can record from two areas of interest, that's beneficial. And also there's been a lot of studies that
38:41
once you get a certain number, I think it's 14 or above electrodes, you really don't have a good hypothesis and those patients aren't going to do well. So you really want to make sure that you're
38:51
minimizing the number of electrodes. This is what a typical SEG plan may look like. And again, this is done well in advance. So for those that are interested in epilepsy, I would definitely
39:01
encourage you to try to come to the CMP conference. And then, you know, you can email me or Dr. Fried, Dr. Barrie. And when we plan our electrodes, you want to be there at that meeting as well.
39:14
It's much more beneficial for you than just doing the operative steps. So in terms of SEG placement, patients are placed under general anesthesia. I keep them all paralyzed, so there's no movement.
39:28
A frame is placed. We get some intraoperative CT scan for co-registration. you wanna put your electrodes in a logical sequence. So in general, anterior to posterior and superior to inferior. The
39:41
reason why is if you get a CSF leak from the back of the brain, the whole brain anteriorly, it will sag down. So all your trajectories will be off. And you also wanna put your deep electrodes
39:52
first, 'cause those are most important in terms of, you know, so your insulin electrodes or your dynamic electrodes, if you're gonna do that. You wanna obtain your trajectory, pin, stab,
40:00
incision. Again, we use the arc, most places use a robot, small drill, you place your bolt, and your bolt is what's gonna give you the trajectory. So that's the most important part in terms of
40:11
your trajectory. When you measure the depth and you, I use brain lab on top of all this as a double check system, it appears to Dura, place the lead to the target, tighten the cap to prevent a
40:26
CSF leak, and at the end you wanna obtain a CT or MRI for colocalization This is generally what it looks like.
40:36
And when you go back to do a craniotomy, these are the serieg holes in the skull that you can see. And this is what it looks like as it pierces the brain.
40:47
So in terms of the SEG workflow, we place our elect, again, I also do this on a Monday, Friday, place electrodes on a Monday, patient goes to the EMU. Again, I try to keep all our patients on
40:58
the ICU side. But if you have a good step-down unit, they can probably be in the step-down as well. We don't have an EMU on the pediatric side, so everything is mostly done in the
41:10
ICU. We localize where the electrodes are, do the EEG, surgery too is to remove the electrodes, and then in a delayed fashion, we would go back to do either a resection, RNS, LIT, so on. In
41:22
infants, you can place electrodes without bolts. So you can do bothless SEG implants would basically suture the electrode right to the scalp.
41:34
You know, sometimes the bone is really thin. For example, in this area, you probably can't get a bolt in there. It's a little bit less accurate than having the bolt in there but allows you to do
41:43
SCG in very young children. The newer thing we're doing now as we're doing a lot of thalamic stimulation for epilepsy is we have been, during SCG our evaluation, I have been implanting the
41:57
anteronucleus and centromedian nucleus of the thalamus at the same time of these SCG evaluations The reason why is, if they're not a good candidate for a localized resection or therapy, I actually
42:10
can obtain information as to what the best target is for their epilepsy stimulation if we're gonna go through a target in the thalamus. So this is being, I think there's very few groups doing this.
42:23
I think there's an adult series that was published, but it allows you, we can nowadays do this as well The insula is a very interesting area. ignored for many years, but in recent years, I see a
42:37
lot of talk about insula, whether it's for glioma or epilepsy. There's a renewed interest, I must say. And actually, probably a common cause of failed temporal lobe epilepsy surgery is that
42:49
insular origin. And obviously, because it's deep from the surface of the brain,
42:55
it's a little bit harder to localize in your surface. The GE electrodes aren't going to be able to give you direct recordings from the insula But there's many ways to implant the insula. There's an
43:06
orthogonal approach through the
43:11
PI. You can just push the electrode through the PI across the other side that some people like to do, especially if they're doing orthogonal approaches.
43:18
You can split the sloven fissure and go trans orthogonal. That way, you can do a SEG approach from a parietal trajectory or a frontal trajectory. Or you can split the sloven fissure and put
43:30
electrodes from the apex, so from the Lyman insula. up backward. So there's many different ways to sample the insula. This is a quick case of an SEG, three-year-old right-hand dominant female who
43:43
had developmental regret, catastrophic epilepsy, infantile spasms, and had on MRI, polymicrojibri on the right side, but also by lateral peripheral heterotopias,
43:57
FDG pet showed decreased metabolism on the right side So we didn't really know if this patient required a full hemispherectomy or we could do a localized resection. So we implanted invasive SEG
44:09
electrodes. Here you can see predominantly on the right side, very little on the left. And again, it takes us a long time to do these just using a frame. If you have a robot, you can do this in
44:22
a couple hours.
44:24
This is what our depth electrode, sorry, I should say, SEG electrode, not depth. Electroplacement looks like, and we. identified seizures coming from both the frontal lobe and the insula, and
44:37
there's really the only way to address both these regions, especially in the kid who has neuroplasticity to do a hemisprectomy, 'cause you're not gonna be able to resect insula, frontal lobe, and
44:49
it's probably not, you know, it's probably this whole region. So we did a peri-insular hemisprootomy here and patient recovered quite well. No isolated finger movements, but the strength mostly
45:03
all came back. And most importantly, in terms of cognition development, the child really started doing well. This is a good summary slide that was put together, basically the strengths,
45:14
limitations, and indications of ECOG versus sub-dural recordings versus SEEG. And again, I'm happy to share this with you to sort of, it helps sort of conceptualize and put this into good
45:28
frameworks for your understanding determine why we use which modality.
45:34
Again, this is something I put together. So in situations favoring subdural grid is where your epileptogenic area is more focal, where it's closer to eloquent areas where you want to map a function.
45:45
It's not very good for paralymic and limbic structures, very difficult to with grid evaluations. If your imaging shows a superficial lesion, this is helpful in first time cases. It's helpful,
45:57
it's very difficult to, especially for a big exposure to dissect the dura away from the brain in a redo situation. Children and infants with
46:07
thin skulls of drill grids are easier to do. And it's a simpler surgical,
46:17
I don't know what I mean by that last comment, but
46:21
favoring SEG, if you have less focal epileptogenic care. So if you want a sample, let's say funnel lobe and occipital lobe and then the other side, You can't do that with a grid, or it's much
46:32
more challenging to do with. with a grid.
46:35
If you're looking at non-alloquine areas, this is better. It's definitely better for limbic and paralimic structures. If you don't have a lesion or you have a very deep lesion, this is the way to
46:45
go. In re-operation cases, SEGs are easier to do because you don't have to dissect a dura way. And children infants with dense gall, I say no. But again, we're pushing the limits of what can be
46:57
done
47:01
There are many groups - this is a group in London, Ontario, which is, again, a very -
47:07
it's an epilepsy group with a rich history. And you can see, over time, there's subdural electrode evaluations in SEGs. It's almost 90, 95 SEG now. And this is the trend at many hospitals now,
47:22
also in the United States. And this is a systematic review comparing subdural grids to SCEGs,
47:31
the writing small there, essentially what it tells you is that the risks are both low for both of them.
47:40
And they're very comparable. There's not a big difference. So if you know how to do each one of these well, that shouldn't really make the difference. This is the easier to read slide. Again,
47:52
comparing SEG with subdural grid evaluations. With the subdural grid, you can see the epileptogenic zone was located more often and resection was done more often. But in terms of the ones that got
48:06
resection, the seizure freedom rates were the same. Now, any study comparing SEG to subdural grid is really comparing apples to oranges. You're not, it's not the same patient population. And
48:19
these two evaluation techniques are largely complimentary. So what SEG is not good at, subdural grids are good at, and vice versa. And quite often, to be honest with you, nowadays, we're doing
48:30
a lot of hybrid evaluation. So this is a patient where we exposed, first put interhemispheric grids. This is what they look like. Then placed depth electrodes. This is a rolandic epilepsy. So
48:44
this is motor sensory area. And then on top of that, we place a grid.
48:50
And this is what we get. So we get recordings from interhemispheric region We get depth electrode recordings and surface and really allows us to evaluate this patient in 3D. The only thing with a
49:01
craniotomy when you're placing depth electrodes is that you're gonna be a little bit less accurate. The moment you open the head and open the door, that's gonna be a little bit of brain shift. So
49:11
you're gonna lose a little bit of accuracy. So in conclusion, phase two evaluations often required in epilepsy surgery. Both techniques have relative advantages and disadvantages And really, your
49:24
surgical epilepsy team needs to pick the right approach for the right patient. And as an epilepsy surgeon, you need to be comfortable with with both techniques.
49:42
Let's see, Joe, I could run it on the deaf and secondary chair.
49:46
Um, which one? Yeah.
49:53
Right. Frontal and saliva. Yeah.
49:57
That's one.
49:59
Why do we use SCG? Oh, because this is a, again, so we're dealing here with a. We're pretty confident It's usually coming from the right side. I call it metabolism on the right. You can see
50:12
that here. The polymite region is the right formula. It's kind of small, hard to see there.
50:20
And infantile spasms. So they're generally looking epilepsy, but they can be coming from local regions. So just because they're infantile spasms, they don't necessarily mean they're coming from
50:29
the whole brain. Could be coming from a region. So we were pretty confident we had good laterality So an SCG
50:37
type evaluation, meaning it's asymmetric. Right. And also we're not going in, you know, it's not a bi-tempo. So we would call that a SCG. And the electrode, if you look at the,
50:52
you know, these look more orthogonal, but this one, for example, is going more parasaginal. So they're not all in this orthogonal plate. So I call this SCG variation. Because I have a question,
51:01
sorry. I did look for some sausages. Is that you typically use depth, right? Yeah, the reason I use depth is
51:10
because we can put the thank you for your research electrodes in the so. So those are dead electrodes, but you're using a likeness. Yes. Oh, it just looks like they had so many contact. Yeah,
51:22
yeah, some of them are a lot, but yeah, they're all like the depth style. Yeah, I'm using
51:31
I've read before that epilepsy is a cortex. And I've always wondered if that's because you know we do Thank you. and it's on the surface and then procedural. But my question is, with depth of
51:44
electricity, you'd find like some cortical origin zones? Generally not. Generally, it's neocortical,
51:53
it's on the surface. But again, depending on what your, for example, FCD-type 2, usually have a transbandyl sign. You can have this glass of neurons all the way from the something almost zoned
52:06
all the way to the surface
52:09
It's
52:12
a
52:14
great question. So we are actually looking at, because now we have a decent series of salamic SEG.
52:28
And actually, let me add it up. Let me turn this up. So now we have a decent number of salamic EEGs.
52:33
And I can tell you from the 13 or so patients we've studied, The thalamus is almost. always simultaneous onset with the seizures or neurosimultaneous. From those patients, we haven't seen a lead
52:47
from the thalamus
52:49
that spreads to the cortex, but it's one of those fascinating areas. There's people looking into that. Especially now as we have, we're doing more and more thalamic type recordings. The idea is
53:02
certainly for certain seizure types and primary generalized epilepsy, whether there's a central pacemaker that's causing on, whether that's a thalamus. There's
53:13
some papers on there. I will tell you though, and this is indirect and
53:20
anecdotal. But Dr. Matherne told me he had at least a couple hemistrectomy patients that failed hemistrectomy and all he did is he went back and he resected thalamus and they became seizure-free.
53:34
It's unpublished
53:37
but for what it's worth, that could be an example. Maybe there was this plastic neurons there, but
53:44
yeah, there's nothing written about it.
53:48
Yes. Great timing. Yeah. Go ahead. But oscillational surgeons. Yeah. When does a patient present with seizures? When does a Lycenol
54:02
patient become an epilepsy patient versus an oncologic or even a neurovascular, you know, cab mal or something like that. Is it the chronicity of the seizures? Is it the, you know, sometimes
54:16
it's a, who know why I refer this patient to, let's say they come in with a cab mal, where they refer to the neurovascular surgeon or the epilepsy surgeon when they have seizures.
54:29
Yeah, great question So
54:34
I will say, I mean, we, what I see is the failures of, I mean we what I see is the failures of tumor resection and a patient that still has epilepsy. And I, granted, I'm not seeing, you know,
54:46
the denominator, right? So, but I will say for regional epilepsy, commonly related to a focal, a low grade tumor or a cavernous malformation, those patients do really well to just a full
55:02
resection of the lesion. If you get a gross total resection, that patient will likely do very well And
55:09
if you talk to epilepsy folks, they'll say they need an epilepsy evaluation, all that.
55:13
But especially if it's seizures, they haven't failed more than two, it's just they've had a couple seizures and here's a lesion, those patients will do really well. And I think for the vast
55:23
majority of places, they don't require all this invasive, or not invasive, all this evaluation. Same for a calf mal that is presented with a seizure or two seizures and maybe just on one
55:36
monotherapy It's really the patience that I've.
55:49
had epilepsy for a very long time and probably failed more than two medications. Sometimes those patients may have kindling, they may have dual pathology now, especially if it's a temporal lobe
55:50
calf mal, now they have hippocampal sclerosis on top of it. Those are patients that would sort of benefit from a epilepsy evaluation. So I look at the chronicity, maybe
56:01
the numbers, how severe the seizure burning, 'cause you know the patient populations that we operate on are kind of different The tumor patients have had a seizure in the last year or a seizure in
56:13
the last couple of months. Our patients have seizures every day. It's a very different patient population. But yet there's enough data out there that if you just fully resect the region and for
56:26
calf mouths, you get the hemocidarin ring and you can do that all safely, I think those patients will do really well.
56:37
The critical thing is really the history, you know, if it's a patient and in the adult population, we see patients who have had seizure for many years. And this kind of situation, you know, just
56:48
to be a resection of the lesion will not be sufficient. And also, especially if there's proximity to the hippocampus, you know, there's really a tendency for the hippocampus to become excitable on
57:01
its own merit So this kind of patient really needs epilepsy evaluation, but the patient will present with one or two seizures, first time, you know, probably a gross resection of the lesion is
57:13
sufficient.
57:17
Dr. Fowler, I have one question from you. You mentioned that in some centers, they use weight craniotomy in children. How do they call the child? Do they bring the parents to be with them during
57:29
the surgery? Because it's very, very difficult to calm the child. to do the right, even though I don't almost, not everybody wants to know what's going on. You know anything about that? Yeah,
57:43
so these are, when I say child, they're sort of the youngest is sort of probably around 10 or 12 years of age, and they have to be evaluated by a psychologist to make sure that they are mature
57:55
enough and they can handle it. Now, there's always a fallback plan. So if the patient is not cooperative or is having issues we can always intubate, we can always place the grid. But you're right.
58:10
I mean, it's sort of,
58:13
I'll say it can be done, but in practice and thinking about the issues that you presented, and also it adds a lot of OR time. And quite often, a lot of times, even if the seizures are coming that
58:27
close to the language areas in a child, I'd probably wanna map it. and do that in an extra operative setting and do, so sometimes things that just 'cause we can do it doesn't mean we should. So
58:42
it's an option. They use FTG only for the seizure, but
58:49
they're all dopaminergic studies that they do in the adult. I don't know in the children. They send them to one university unfortunately, because they don't have the isotope here. In the past,
58:59
that was the routine Do they use other receptors for
59:05
checking on the children or only the FTG is useful physiological studies?
59:11
Yeah, so in terms of for epilepsy, we typically using FTG, there's groups that use AMT for
59:22
TSC, but those are the only two isotopes. I don't, maybe Dr. Fried or someone can weigh in as well. I don't know if there's any other isotopes being used for
59:32
the epilepsy world.
59:35
A couple of patients that have multiple seizures a day and then what I bring up here have none. First couple of days you have an anesthesia effect but it's like a week after and they're still not
59:45
having despite like taking off their meds. How common is that and why is that? Yeah, I think like you mentioned, one of it is for long anesthesia time. I mean, our patients are 10, 12 hours in
59:57
anesthesia. Again, just because we don't have a robot 'cause if we, if you do have one, you can plan everything ahead of time and you don't have to change your trajectories on the fly, especially
1:00:07
with your SCG electrode. When you use a like cell frame and you're doing all these different trajectories, quite often you're gonna have collisions and you can't use their trajectories. You gotta
1:00:17
plan it a lot of times while the patients are asleep. But to your point, sometimes we're putting thalamic electrodes, I've seen that slow down seizures, at least temporarily, there's a lesional
1:00:30
effect and just. Yeah, just the effect of general anesthesia could be quite a, you know, keep in mind the younger the kids are they're more fat their body has and that's where the anesthesia gets
1:00:41
stored so it can take a few days to just exit their system. But yeah, it's a good point you want to really minimize the number of days, patients in hospital with these electrodes. Yes, my
1:01:02
mind Yeah, so Toronto, we had a lot of experience with Meg. We're trying to get a Meg. I think it's helpful. If we had it, we'd probably use it on every patient. It's the type of thing. So if
1:01:15
you have,
1:01:18
depending on how deep the focus is, it may or may not be helpful. What you're looking at with Meg is you want to find the cluster. So if you have a cluster of dipoles, there's studies showing that
1:01:30
cluster rectumies or that, you know, that can be helpful.
1:01:35
So we, because we don't have it, we send our patients to San Francisco
1:01:41
to get those evaluations, but we do it on an as-in basis. So if there's enough, enough of
1:01:49
the phase one data is non-conclusive or it's divergent or giving us, and we kind of need the mega, sort of a tiebreaker, those are situations where it could be helpful, and sometimes it could be,
1:02:00
sometimes these patients are going to have a phase two anyway, but the mega may be helpful in determining where exactly the place the electrodes.
1:02:08
So if we had it, we would use it,
1:02:12
but we just don't.
1:02:16
Navigation as well. You definitely create objects that can merge the mega dipoles directly onto your navigation, but I will say there's not a lot of data with single dipoles, so So you don't want
1:02:29
to treat it as a, you know, unless you see a cluster, that's sort of And quite often we don't see a cluster.
1:02:38
Yes, you said psychiatric illness is a contractations of phase two. Is that because of, like what you mean by that is inability to emotionally tolerate the leads? Yeah. So it seems like there's a
1:02:52
certain amount of coincidence of psychiatric disease and epilepsy. And in some cases, like Friday has a paper in the last year or two showing phase two for depression or for psychiatric disease. So
1:03:05
I guess my question of why is psychiatric disease a contraindication?
1:03:12
Yeah, I mean, that's a good question. I mean, I'll answer it two different ways. There's quite a lot of data with patients that are depressed and have epilepsy. If you get rid of their seizures,
1:03:29
they're at risk for suicide for some of them Um, because you've got to think about, these patients have had a chronic debilitating illness. They've had a lot of attention because of this illness.
1:03:41
And all of a sudden you take this away. And sometimes it also empowers more energy to carry out what they want it to. So if someone's suicidal ideations or anything like that,
1:03:55
so you've got to look at that, 'cause those patients are at increased risk of suicide You also, just because they have psychiatric disease, doesn't make them not a candidate. So you do want to
1:04:06
work very closely with a psychologist or psychiatrist prior to surgery to make sure they're emotionally ready. It's a very difficult stage of their treatment to be in hospital with these electrodes
1:04:19
in. So again, you need a lot of cooperation with them. Again, we don't have a ton of experience on the pediatrics side. It's more an adult phenomenon
1:04:30
Our kids usually don't have psychiatric disease. but yeah, it's more just emotionally being able to tolerate it. And also if they've had prior suicidal ideations or major depression, that's
1:04:41
something you wanna counsel them on, prior to just even undergoing this evaluation.
1:04:49
Yeah, just in the adult population, it's not an absolute, in the adult population, it's not an absolute contraind indication. In fact, many of the patients that we operate on and we're including
1:05:00
depth electrodes, it may be an anti-depressant, but the main thing is having some security follow-up that can accompany them in the process.
1:05:16
Okay, thank you, everyone, I'd say. We hope you enjoyed this presentation. The material provided in this program is for informational purposes and is not intended for use as diagnosis or
1:05:31
treatment of a health problem.
1:05:34
as a substitute for consulting a licensed medical professional.
1:05:43
Please fill out your evaluation of this video to obtain CME credit. This recorded session is available free on snidigitalorg.
1:05:56
Send your questions or comments to osmondsnidigitalorg from
1:06:07
UCLA Neurosurgery and supported by the James I. and Carolyn R. Osmond Educational Foundation, owner of SNI and SNI Digital and the Waymaster Corporation. Producers of the Leading Gen Television
1:06:26
Series, silent majority speaks, role models and medical news network
1:06:39
Thank you.
View Transcript
Listen to Podcast