Hey, I'm Haley Vance from Vanderbilt Children's Hospital and just want to say welcome to this year's Advanced Practice Provider Seminar for the Pediatric Section Meeting. On behalf of the APP Planning Committee, myself, Andrea, Han, Mandy, and Patty, one of our moderators, we would like to welcome you to today's seminar. We're super excited to have you. Thank you for making this time a priority for those of you who are here live and for those that will be watching the recorded session later on. And so just a couple of housekeeping announcements we wanted to make at the end of today's session. We invite you to join us for the welcoming open reception. If you just go back to your agenda page, it'll give you the link to join us there. We'll have an APP networking table and would love to just interact with you more at that time. There'll be also opportunities tomorrow at the Meet the Leadership meeting for us to also spend some more time networking and getting to know one another and just talking about our seminar and just kind of what's going on in our practice at this time. For CEU and CME credit, on Monday, December 7th, you can log into your MyAANS account online and submit the form there to receive your credit. And then the last thing is on Tuesday, December the 8th, be checking your email. We'll be sending out a survey, just welcoming feedback on this year's seminar and the topics that we are bringing to you today. And so we're very excited. We have a fantastic lineup of speakers. And so without further ado, we're going to get started. All right, thank you, Haley. So my name is Trey McCluggage. I'm one of the faculty at Texas Children's Hospital in Houston. Thank you all for having me. I really appreciate it. I'll be talking to you a little bit about imaging and pediatric spinal trauma. So before we really get into talking about the imaging, it's important to understand some of the differences between the physiologic differences between pediatric spine and adult spine because there's important implications in terms of the injury patterns that we typically see in kids. Kids less than three typically have a much larger head to body ratio than adults. It's about one fourth of their total body mass, as opposed to about one eighth in adults, which means that their cervical spine is submitted to much higher flexion extension forces with injury, which translates to having a higher percentage of cervical injuries that we see in kids, about two thirds, as opposed to about one third, thoracolumbar injuries, where in adults it's pretty evenly split between the three. The pediatric cervical spine and spine in general is inherently more malleable than the adult spine. The bones tend to be less brittle than the adult spine. The intervertebral disc space tend to be more malleable as well. And all this leads to kids having a higher propensity for ligamentous injuries, as opposed to osseous injuries, which we see in older adults. And this generally lasts until about kids are about eight or nine when their spine is more physiologically normal or functions more like an adult spine. It is important to note that most of the imaging we get is static imaging, and so it has a chance of missing some of these ligamentous injuries, which are a lot more common in kids than they are in adults. Some of the other important factors that are unique to the pediatric spine, there are several mimics of spinal trauma that we can see that are due to some of these physiologic differences. One of the common ones we'll see is what's called pseudosubluxation, which is a horizontal displacement of the vertebral bodies. Generally occurs at C2-3, less commonly at C3-4. It's normal in children less than eight years of age. And generally four millimeters or less in that age group of horizontal displacement is within normal limits. At greater than eight years, three and a half millimeters is the cutoff, and anything more than that is probably abnormal, and that goes into adulthood. Swiss Chuck's line can be useful to try to distinguish either physiologic pseudosubluxation or an actual traumatic subluxation. It's a line drawn from the posterior arch of C1 to the lamina of C3, and the lamina of C2 should be within one or two millimeters of that line. So essentially just a normal spinal laminar line at that level. The synchondroses, these are the normal ossification centers of the spine. They're normally fused by about age seven, but that can be highly variable, and these can sometimes be confused for fractures just due to the variability on when they sometimes fuse in kids. Obviously kids can have congenital abnormalities and failures of fusion that can also be confused for fractures. The most common you're probably familiar with would be the anterior posterior arch of C1 failed fusion that can be picked up occasionally just as an incidental finding on imaging and can occasionally be confused for fractures. The best way to kind of distinguish these fractures from actual congenital or physiologic phenomenon is that physiologic and congenital problems tend to have smooth cortical edges where fractures will have sharp edges and cancel this bone as opposed to that nice smooth cortical edge. The other way to distinguish is that synchondroses and the ossification centers are generally stereotyped. The C1 vertebrae has three ossification centers, C2 has C4, so having some idea of where those are can be helpful to distinguish actual fractures from just synchondroses. There's several other mimics that can occur in the pediatric spine that are unique because of these physiologic differences. Also cervical lordosis can be seen in kids with trauma that have muscle spasms in their neck. Prevertebral soft tissue swelling, it's a marker to kind of use in adults for underlying injury. If you ever heard the rule, the six in front of two and 22 in front of six, that's what they're talking about. But in kids, this can be normal just with flexion of their neck, if they're crying, if they're breathing in, they can have some prevertebral swelling that's just physiologic. Prevertebral body wedging, this is also a normal variant that's seen in kids. Most commonly occurs at C3, but just as the normal ossification of the vertebral bodies, they tend to not develop their full rectangular shape until about seven or eight. So this can be normal and atraumatic. Pseudosporetic C1 and C2, again, this is because of the ligamentous laxity that we see in kids. And by that, I mean the lateral masses of C1 having a little bit of an overhang on C2 can be confused for Jefferson fracture, but it's just physiologic in younger children. Schmolz nodes, disc herniations into the end plates of the vertebral bodies can be confused for compression fractures. Schurman's disease is a pathologic scoliosis. It's defined as adjacent level wedge deformities of three vertebrae, end plates abnormalities, and then thoracic apoptosis greater than 45 degrees, but of course, this can be confused for compression fractures or anterior wedge fractures and things like that. And of course, there's a plethora of congenital spinal anomalies that can be confused for fractures as well. So the dichotomy they run into with choosing any imaging modality in kids is that we always, of course, want to identify all the pathologies there. We don't want to miss any potential fractures or instability, but of course, kids are uniquely susceptible to radiation. So we have to try to limit the amount of radiation we're giving kids, but at the same time, try not to miss anything. It's well known that there is a small but not insignificant risk of long-term cancers in kids who've had a lot of radiation when they're younger, especially less than 10. Kids who've had an average of about 50 or 60 milligray, which is about two or three head CTs, have as much as a three-fold increase in their risk for leukemias and brain tumors. So it's not small. As a result of this, there's been a number of protocols that have developed trying to limit this radiation in kids. Most common you've probably seen is the FAST MRI that's used for hydrocephalus patients nowadays, but specific to spinal trauma, the Pediatric Cervical Spine Clearance Guidelines that every hospital kind of has their own flavor of is also geared towards that. So indications for imaging, especially in clearance of the cervical spine. So like I said, every hospital kind of has their own Pediatric Cervical Spine Clearance Protocol. You know, I recommend using what your hospital has, but there is a lot of variation between the different hospitals on what they all include, but for the most part, there's four things that tend to be the branch points in terms of imaging decision-making, and those are number one, GCS and level of mentation. Number two, the status of the neurologic exam. Number three, the nexus criteria, which includes obviously menstation and neurologic exam, but also things like midline, posterior neck tenderness, and distracting injuries. And then the type of injury and whether it's a high-risk injury or not, those are things like clotheslining, diving injuries, axial loading to the skull, and of course, high-risk MVCs, rollovers, high-speed injury, ejection, death at the scene. But if all those things are negative and the patient is awake, alert, and cooperative, then you don't necessarily need imaging. I think that's what's important to understand is that you can clear collars and you can clear the spine too without the need for imaging. So these protocols are made primarily for cervical, but they work just as well for thoracolumbar spine too for the most part. If the kid doesn't have pain, they don't have a step-off, they don't have a deformity, and they're awake, alert, then they probably don't need imaging in their spine. Choice of imaging and spinal trauma. So this is a very loose guideline, and I think the important point to make is that your choice of imaging is going to be directed at the patient, the situation, their exam, the type of injury, and so it's going to be different for every patient. But very general guidelines, kids who have a GCS less than 8, have a poor mental status or intubated, it's probably a good idea to get a CT. You don't have to initially, but for the high-risk injuries, it's probably better when they can't tell you what's hurting or what some of their exam findings are. It's a good starting point. MRI can be very useful in the intended population, mostly to rule out, obviously to rule out cervical spine injury, but also to be able to clear a collar in a patient who's going to be intubated or have a poor exam for a while. Obviously a patient with a neuro-deficit at least needs a CT and probably needs, certainly needs an MRI as well. It's important to not miss injuries like skeuorotype injuries that can have negative X-ray or CT imaging. Midline cervical tendons are distracting injuries in a patient that's awake and alert, otherwise plain films are probably sufficient. And then patients with high-risk mechanisms or polytrauma probably need a CT, but plain films are reasonable if they're somewhat reliable. Plain X-rays are a good start and a decent initial screening. Cervical spine tends to be the most sensitive, but depending on what study you look at, the sensitivity of plain films is somewhat variable, anywhere between 73 and 100%. The AP view does not increase sensitivity, but it is normally part of the protocol at most hospitals. A dontoid view is not generally done in kids less than about five to eight or so, just because it's hard to get a proper open mouth view, you need some cooperation of the patient. Flexion and extension views in the acute period may be a little controversial. Obviously, there's a propensity for ligamentous injuries in kids, so this can be very useful for identifying some of those. I know, I believe it's at primary children's where they actually have flexion and extension films as part of their cervical clearance protocol, but just be aware that obviously some ED physicians and radiologists are a little leery about doing this. I tried to get it in our protocol here and they were not interested in doing that. CT imaging, obviously the modality of choice for a ton of patients, polytrauma, high-risk mechanisms, and if you find anything on just plain film screening, clearly it's going to be better at picking up osseous injuries and ligamentous injuries, and it does, depending on what study you look at, has a higher sensitivity than x-ray. MRI imaging, modality of choice for obviously any neurologic deficits or concern for spinal cord injury. It's useful in a ton of patients to clear a collar, a negative MRI will effectively clear their collar. Obviously, if there's a concern for ligamentous injury, or if you have a known fracture and need to assess stability, it can be very useful. It has the highest sensitivity and specificity, but that decreases a little bit after 48 hours, mostly because edema can kind of wane. Still very useful after that, though, especially if you're just trying to assess stability and basically the ligamentous structure. So if you just want to know, you know, how is this stable, can I clear this collar, is it ALL intact, PLL intact, ligament deflated intact, I guess it'll be useful for that. There is a higher false positive rate with MRI, but that typically refers to things, I'm sure you've seen the reads where it says paraspinal muscle edema or nuchal ligament edema. And while that's obviously a sign of some kind of trauma, that's probably not relevant to the stability of the spine. So lastly, live dynamic CRM fluoroscopy. So this is kind of a unique scenario, and I tend to use this, our program tends to use this in very specific situations. It tends to be kids who've gotten an MRI that's somewhat equivocal. It's mostly for occipital cervical pathology where there's some concern about some ligaments have torn, but the transverse ligament's intact, and maybe the MRI was gotten several days after the injury, so we're unsure about the edema. And so we need to know whether and clear the collar if they truly have an unstable occipital cervical spine or need surgery. We'll actually take them to the OR, intubate it, use intra-op monitoring, and then actually do passive motion with, you know, by hand under live fluoroscopic views. And it's, again, we use it very rarely, I've used it maybe two or three times in the past year, but it can be very helpful, and it's, you know, I've been able to avoid surgery in a couple of kids using this, and finding that they were relatively stable after doing this procedure. Obviously, this may be a little controversial, I'm not sure if other programs do this a lot, there's of course a risk for spinal cord injury in doing it, but it's been our protocol here for a long time, and I found it to be very useful. But again, in certain situations, of course. So, lastly, just in conclusion, you know, I think it's important to understand the unique physiologic differences between the pediatric spine and the adult spine, and just the implication that has in terms of the injury pattern patterns that we see in these kids, and in terms of the mimics you can sometimes see that in pediatric spines that you don't see in adults that can mimic spinal pathology. Any choice of imaging needs to be directed towards weighing the cost of giving a big radiation dose, as well as the need to make sure you're not missing any pathology. And it really should be tailored to every patient, you know, what's tailored to the type of injury, tailored to the neurologic exam, to the history and the physical exam. And so I think those are the really important things to understand, and also to understand the limitations of exams, how they cover for each other, and that, you know, having a negative exam may not clear you or absolve you from involvement. So that's it. Thank you very much for having me. Hi, this is Dr. Douglas Brockmeyer from the University of Utah and Primary Children's Hospital in Salt Lake City, Utah. Welcome to the AANS-CNS section on pediatric neurosurgery in its annual Advanced Practice Provider Educational Session. My topic today is pediatric cervical spine trauma. To begin with, we're going to discuss a few biomechanical principles regarding stability of the cranial cervical junction and cervical spine. Next, we'll talk about clearing the cervical spine and share with you the algorithm we currently use. We'll also talk about specific ways to clear the cranial cervical junction and subaxial regions. Next, we'll talk about a few specific injuries and then discuss the concept of PSIWARA. Physical instability is defined as the loss of ability of the spine under physiologic loads to maintain its pattern of displacement so that there is no initial or additional neurological deficit, no major deformity, and no incapacitating pain. That definition, given in 1990, is still as true today as it was back then. Cranial cervical biomechanics can be a frustrating and complex area to study. However, a significant amount of work done in the last few years have been able to hone in on the areas that are required for injury to cause instability. At occiput C1, the major stabilizing structures include the capsular ligaments and the shape of the joint. Minor to moderate stabilizing structures include the tectorial membrane, the anterior and posterior occipital atlantal membrane, alar ligaments, and apical ligaments. At C1-C2, the major stabilizing structures include the transverse ligament, the atlantoaxial catheter ligaments, and the integrity of the odontoid. Again, the alar and apical ligaments are only minor stabilizing structures. Looking at the cervical spine from the lateral direction, we typically break it down into anterior-posterior columns. The anterior column consists of the anterior longitudinal ligament, the disc space, and the vertebral body. The posterior column includes the posterior longitudinal ligament, facet capsules, interspinous ligaments, and other bones in that area. 40% of the weight of the head is borne through the anterior column and 60% is borne through the posterior column. Looking from front to back, we see that the weight of the head is borne through the occipital condyles down through the lateral masses of C1 to the lateral masses of C2 and on down. All of these structures need to be intact in order to support the weight of the head and provide stability. So what makes the cervical spine unstable? At occiput C1, it includes capsular ligament disruption, abnormal joint anatomy, or malalignment of the supporting bone structures. At C1-C2, an incompetent odontoid, capsular ligament disruption, or transverse ligament disruption can cause instability. In the subaxial region, anterior and posterior column disruption is necessary to cause instability. The ideal pediatric cervical spine clearance protocol does four things. First, it identifies all significant injuries. Second, it minimizes unnecessary radiation exposure to the child. Third, it permits removal of a rigid cervical collar or other method of cervical spine immobilization when deemed no longer necessary. Lastly, it efficiently and effectively utilizes health care and human resources. This is the pediatric cervical spine clearance algorithm we currently use. Below the algorithm is a reference that we supplied from the Journal of Bone and Joint Surgery published in 2019. It was taken from the Pediatric Cervical Spine Working Group, and I encourage you to adopt its use. To clear the upper cervical spine, we rely heavily on the concept of the condylar C1 interval, or CCI. We measure the distance between the condyla and the upper portion of C1. If it's over 4 millimeters in distance, we are highly suspicious for atlantooccipital dislocation. My advice to you is to use the CECI. It works. Again, a CCI greater than 4 millimeters is highly suspicious for an AOD, but not necessarily diagnostic. If you are on the fence and unsure whether instability exists, a flexion extension CT under controlled circumstances or even an evaluation under anesthesia may be necessary. You need to also be on the lookout for occiput C1 and C1-2 combination injuries. Clearing the lower cervical spine is sometimes a confusing topic as well. Again, we use the advice from White and Punjabi in 1990, where they looked at the anterior elements and posterior elements together to determine whether stability or instability was present. If those structures are destroyed or unable to function, then that's a significant tip-off. We also use flexion and extension x-rays and look at sagittal plane translation and rotation in order to help us make our decisions. Here are some specific injuries by level. At occiput C1, you can have atlanto-occipital dislocation or various types of condylar fractures. At C1-C2, odontoid synchondrosis fractures can occur. You can also have osodontoidium or even a type 2 odontoid fracture. Transverse ligament injuries can also occur and also capsule ligament disruption needs to be necessary to cause instability. In the subaxial region, soft tissue injuries occur. These can range from cervical strains to full-blown ligamentous instability. You can also see growth plate fractures or separations such as synchondrosis fractures. And then in older age groups, you start to see true fractures like you would see in adult trauma patients. Combination injuries can also occur. Here are some specific age-related injuries. From 0 to 4 years in descending order, we typically see AOD, C2 synchondrosis fractures and ligamentous injuries. From 4 to 10 years of age approximately, you start to see more ligamentous injuries and fractures and less AOD. Ages 10 and above, you typically see fractures, ligamentous injuries and AOD in descending order. Anatomy plus mechanism dictates injury level and different spinal injury levels are seen in younger patients. A couple of old studies demonstrated that two thirds of all cervical spine injuries in children occur in the upper cervical spine. The lower study showed from 0 to 3 years, over half of cervical spine injuries occur at C1, C2. Whereas from 4 to 12 years of age, only 8% of cervical spine injuries occur at C1, C2. Here are some specific injuries. This is a 5-year-old girl who was involved in a high-speed motor vehicle accident. You can see the distances of the CCI that are measured on the sagittal and coronal CT scans. The MRI scans show the joint taken at the exact same plane as the CT scan and show STIR signal change and abnormality in the occiput C1 joints. We determined the patient was unstable and took her to the operating room despite no posterior ligamentous injury or tectorial membrane injury. Interoperative fluoroscopy, shown here, showed a significant amount of injury and instability at occiput C1. Here's a 6-year-old boy who was in another high-speed motor vehicle collision. You can see his CCI is not as high as the previous one. It was 3.5 on one side and 4 on the other. We weren't sure whether he was unstable and therefore treated him in a hospital collar. He came back two weeks later with significant neck pain and a flexion and extension x-ray showed significant motion between occiput and C1. Here's an example of a C2 synchondrosis fracture. This is a sagittal CT image and the red arrow shows demonstration of the fracture through the growth plate at the base of the odontoid where it connects to the body of C2. These injuries are typically treated with halo and not operative fusion. Transverse ligament injuries can also occur. Disruption of the transverse ligament can result in pulling away of the odontoid from the anterior arch of C1 which is seen in the axial CT image on the left. Combination injuries can also occur. This one occurred at occiput C1 and C1 C2. You can see the significant amount of disruption on the CT image in the middle and the significant ligamentous injury on the MRI on the right hand side. This 14 year old boy went headfirst down a slippery slide. He suffered a C2-3 injury with both anterior and posterior column injury. You can see the angulation and disruption of the C2-3 disc space on the CT scan above and on the MRI below. The STIR images show significant posterior ligamentous disruption. He was treated with an anterior C2-3 anterior cervical discectomy and fusion. This unfortunate 16 year old girl fell when a hammock support wall collapsed while she was getting in it. She suffered significant cervical injury with C6-7 spondyloptosis. Unfortunately she came in Asia A and remained that way. The lower CT images show the spinal reconstruction done in both the anterior and posterior directions. A brief word about SIWORA. SIWORA stands for spinal cord injury without radiographic abnormality. It was described 38 years ago on a small number of patients in the pre-MRI era. It was defined as the demonstration of objective signs of myelopathy as a result of trauma with no evidence of fracture ligamentous instability on plain spine x-rays or tomography. This paper changed an entire generation's view of pediatric spinal cord injury and led to significant amount of overtreatment over the years. The biomechanical rationale for SIWORA was that the pediatric cervical spine is quite flexible. In between the positions of flexion and extension the spinal cord could be pinched against the bone. The concept of SIWORA was revisited by the original authors in 2004. They felt that since the introduction of MRI in the 1990s more accurate diagnoses could be made and many of the original SIWORA diagnoses were incorrect. They also felt that the status of the spinal cord seen on MRI was a much stronger predictor of ultimate neurological outcome. Where do we stand in 2020 on the concept of SIWORA? Personally I feel that if a patient comes in with transient neurological deficits and has negative imaging they probably suffer from some sort of spinal cord concussion. Plain x-ray and flexion and extension in MRI is usually all that is necessary for the workup. You should treat unstable lesions surgically. Stable lesions with a normal MRI may be followed and treated symptomatically. Patients with neurological deficits may be braced for a short period while they recover. Follow-up imaging is necessary for lesions of indeterminate stability. I hope you found this lecture useful and I thank you for your attention. Hello I'm Chris Bonfield at Vanderbilt University Medical Center. Thank you for having me talk about thoracolumbar spine trauma in children. I have no disclosures to report. The learning objective today will be recognize the common characteristics of pediatric spine thoracolumbar trauma. We'll go over the epidemiology, some presentations, how you evaluate these children, some radiology findings, classification of the injuries, pathology, and some outcomes. Thoracolumbar spine trauma in children can be a significant challenge and that's due to ongoing spinal growth which can create some substantial morbidity and mortality in these cases which is different than adults. The presentation of thoracolumbar spine trauma is diverse and a variety of treatment options do exist. Furthermore pediatric patients have greater ligamentous flexibility. Their soft tissues can stretch more than their adult counterparts and their relative paraspinal bone and muscle immaturity compared to their adults make individualized treatment decisions very important. Epidemiology, the pediatric spine injuries are less common than in adults and accounts for approximately two to five percent of all spine trauma. Thoracolumbar fractures account for one to two percent of all pediatric spine fractures so it's slightly more common in the cervical spine and in the thoracolumbar area. Risk for neurologic defects decreases with injuries lower in the spine so the cervical spine injury says the most chance of having a neurologic deficit, the thoracic spine next, and the lumbar spine after that. There's differences in ages as well and older children between the ages of 15-20 years they have the highest rate of spine injuries with more males than females as age increases. As the patients are younger the the gender split is approximately 50-50. Also there's most common mechanism is fall in younger children and motor vehicle accidents in older children. It peaks in the summer and the winter months over winter break there's seasonal variation the rate of spine injuries with peaks during the summer months and during break when children are off of school. How do these children's present mainly? There's tenderness which is the most common physical exam finding tenderness over the area of injury and the paraspinous musculature in the area the spine fractures. This is where a physical exam can pick up some of these injuries without even doing any CT scans or x-rays. Other symptoms are also common there's exam findings such as contusions on the skin or in the muscles a step up deformity and neurologic defects. There are also injuries that come along with these spine injuries we can see other injuries to the surrounding structure such as the ribcage, the thorax, the lungs, abdominal injuries of nearly up to 42% or up to almost half of these children especially motor vehicle accidents. The seat belt not only crushes or can create injury to the spine itself but also the soft tissue and the organs that are in front of the spine. They also may have head injuries and furthermore if the patients have spine injury in the cervical spine you might have to look lower to ensure that they don't have spine injuries and other parts of the spine which can be common. So just because there's a spinal injury in one part of the spine we can't forget that we need to continue to look at other areas to ensure that there aren't injuries in those places as well. Spine precautions for suspected spinal injuries are important at first if spinal injuries is suspected the patient should be kept in spine precautions whether that's a cervical spine collar, immobilization, they should lay flat on log roll until injury is ruled out. We can't forget that the ABCs are important so airway first, breathing and then cardiac status, hypotension should be avoided especially if it's a serious spinal cord injury. A complete motor and sensory examination should be performed from cervical spine all the way through lumbar. Other examinations such as general examination and rectal examination should also be included as well as perform reflex examinations. All these things can help us point to notice whether there is an injury and which part of the spine is injured. A more focused examination on the TL spine an examination has been shown to diagnose a fracture with sensitivity of 81% specificity of 68% so most of the time if there is a TL spine you can pick it up on examination whether it's tenderness, step-off or some sort of neurologic deficit. Tenderness is the most common physical exam findings as we talked about, contusions, step-offs, deformity and a neurologic defect can be seen and once again don't forget the concomitant injuries is an important or up to half of the children can have other injuries in the abdomen or the lungs and they may also have head injuries or spine injuries in other areas of the spine that's not seen. A little bit of radiology x-rays are often the first choice because it's quicker and sometimes done in hospitals that don't have other CT scans or MRIs however CT scans especially at the level and trauma centers are becoming much more common. They also pick up many more injuries and are better than x-rays so x-rays are okay if that's all you have but CT scans are certainly better. You can see splaying of the posterior elements, spinous process or for burst fractures, deformities all of these things suggest that you need surgical intervention if these are picked up. The MRIs are generally reserved for those patients with concern for a spinal cord injury, neurologic defect in order to look at the actual neural elements, the spinal cord, the nerve roots. It can also look at the posterior ligamentous complex and injury to the spinal cord. Classification, there are a couple classification systems that are used. The TL-AOSIS, the AOS Spine Injury Classification System is the newest version. It incorporated a couple different classification systems including TLIX, the Thoracolumbar Injury Classification and Severity Score which we'll talk about. I generally use TLIX more often because I feel that it's simple to use. I put these in not to go through the whole thing but so you can see what these different classification systems are. You can refer back to these slides. The TLIX is simple. It's non-surgical management with less points which is zero to three. Surgical management is recommended when it's above four and it's a gray area in between. What we're talking about are the different characteristics of the injury. Compression fractures are less severe and less likely to be operated on. Rotation, translation, distraction, those fairly by itself almost tell you that you're going to need to do surgery. Disruption of the posterior longitudinal complex, if that's split in the back, disrupted, you're likely to go surgery. Then if there's some sort of spinal cord compression, whether it's cauda equina, the spinal cord or nerve root, likely you're going to go to surgery as well. So you take each of these three areas, add them up together and if it's above four, likely it's a surgical procedure if the patient is stable. We're going to go through the different types of injuries fairly quickly here. The most common fracture pattern is the compression fracture. It's caused from axial loading with some flexion forces. So oftentimes when they fall from a height, you land on your feet or your legs or on the spine itself. These are generally stable compression fractures, only one column which is in the front of the spine. You can see in the CT scan on the right about the fourth or fifth vertebral body down, it's wedge-shaped with the front column being compressed. The middle column is still intact and the posterior elements are still intact as well. This is a stable fracture. Most of these injuries do not need surgical intervention and we basically restrict them from having contact, strenuous activities with or without a brace. The data is split on whether or not bracing actually helps, usually for six to eight weeks as these heal. So in summary, it's a one-column injury. It's the most common. It's generally stable and most don't need to have surgery. Burst fractures are compression fracture with the anterior column and the middle column being fractured. So the front of the vertebral body and the back of the vertebral body are both fractured. This is a two-column injury, same sort of axial loading forces. Generally speaking, these stable fractures without having a neurologic deficit can be managed with the activity modification with or without bracing for a few months as well. Unstable fractures can be seen if there is a focal kyphosis, if there's neurologic injury, if there's a lamina fracture or the facet joints have jumped. Unstable burst fracture generally treated with fusion from posteriorly where we decompress the spinal cord or nerves if needed and put pedicle screws to hold things in place. And occasionally, if there's a lot of compression from the front of the bone, we may have to go in from the front of the spine as well to remove that bone and take the pressure off of the spinal cord or nerve roots. So in summary, these are two-column injuries. You have to determine if they're stable or not. And most of the time, it's a posterior approach for surgery if needed. Flexion distraction injuries are chance fractures, which are three-column injuries. It's caused by a distractive force. Oftentimes, these are patients who are sitting in cars having seat belts on. The car stops very quickly. The seat belt blocks their abdomen and the front of their spine continues to go forward and fractures three columns right across. There's abdominal injuries that can become with this because there's going to be compression between the seat belt through the abdomen to the spine with a lot of high energy forces. So approximately 40% of patients can have other injuries. This is three-column fractures, anterior, middle, posterior. Sometimes, it's through the bone only as seen on the top illustration right through the anterior column, the middle column, and the posterior elements there. And sometimes, it's through disc spaces or the posterior ligamentous complex as seen with the three slides on the bottom. Most of the time, these are going to be fractures that are going to need to be fixed surgically because they're unstable because it's a three-column fracture. So this is generally achieved with a posterior instrumentation fusion as seen in these illustrations here. So in general, this is a three-column injury, generally unstable. Frequently, there is spinal cord or nerve injury with them, and most of them do require surgery. Fracture dislocation is even a worse fracture as you can see on these scans. The spinal cord is completely split, and the top half of the spine is in front of the bottom half of the spine. It's caused by torsional forces, distraction forces, shearing forces, again, high energy and high impact. It's commonly with neurologic spinal cord injuries, especially up high in the thoracic spine. This is an unstable fracture, as you can imagine, and it's going to need to have surgery. Generally speaking, from posterior, we have to figure out how to get that spine lined up before you stabilize it with pedicle screw instrumentation. So fracture dislocation, frequently neurologic injury, this is an unstable fracture and likely needs surgery. Skywar is spinal cord injury without radiographic abnormalities. This term was coined before the advent and general use of the MRI scanner, so it should be really actually described as spinal cord injury without bony abnormalities or bone fracture, so CTs and X-rays. It's caused by high energy injury where in children, the spinal column can stretch approximately one to two inches, but the spinal cord itself can only stretch a small fraction of that. So the discs and bones and soft tissue stretch the spinal cord stretches resulting in injury, but when you look at the actual CT scans, all the bones and overall alignment looks fine. This can range from a complete injury to actually no injury as well. It's more common in children under 8 because their soft tissue stretches more than in adults. It's more common in the cervical spine because of the large head. As the body stops, the head continues with injury and stretches the spinal cord. So more common up higher in the neck compared to lower on in the back. The main treatment is going to be collars and braces because these are generally stable fractures and there's generally not any compression on the spinal cord itself. Prognosis is mainly determined on what the patient's status is whenever they come into the hospital. So complete injuries are going to be much worse than patients who don't have any sort of injuries. Outcome, long-term outcome, patient neurologic condition, or presentation generally corresponds with long-term outcome. The function as well, it does improve generally, and it improves better in children than in adults. Complete spinal cord injuries are associated with poor outcomes. Incomplete spinal cord injuries can return to baseline in anywhere from 1 quarter to 3 quarters of patients with spinal cord injuries. So for key points, these are rare injuries, but pediatric thoracolumbar injuries are rare. It can lead to substantial morbidity and mortality. They are different than adults. There's increased stretch in their ligaments and soft tissues. The bones are immature, but as the patient goes to about 8 to 10 years of age, the spine is similar to adults. You have to diagnose them early. Long-term follow-up are important because these children oftentimes will grow as well. So depending on what your fusion is, what your treatment is, you have to keep track of them as they get older. You want to limit their fusion because they often have growth remaining, so we only do what is necessary. And conservative management is typical and most patients can be managed conservatively without surgical treatment. So thank you for your time and we can answer questions at the end. Thank you so much to all three of our speakers for fantastic talks about spine. I know that's something that more and more of us are seeing in our practices now that maybe 10, 15 years ago we didn't see. So just looking at the chat box, there was one question that was submitted and so I know Dr. Brockmeyer and Dr. McClugge are still on and so just a question that was submitted was in regards to cervical collar clearance. I know that there's some variation amongst hospitals and you guys mentioned that some, but the question was really for patients who come in initially after the injury who have their initial imaging is negative but have persistent neck pain and are left in a collar. At what point do you bring them back for cervical collar clearance? One week, two weeks, three weeks? And then do you need to have repeat imaging or not? And so if either one of you guys would like to answer that. Yeah, Haley, this is Doug. Thanks for letting me be part of this program. I'll make a comment. So the default position at our place is come back in two weeks with a FlexX film and you see the patient in clinic. Having said that, we're in the middle of doing a really interesting study and following everybody who goes out of the ER with a collar and it turns out there's a lot of interesting preliminary data. There's a lot of patients who just take the collar off by themselves. There's another group that go to their PCP and get it removed there. And then there's a third group that comes back to see us in clinic. And then we've provided a fourth group where we clear it with telehealth because it's kind of crazy to come back and have a x-ray which is just a couple of minutes and you see the patient for a few minutes and you clear their neck. So yeah, I think there's this default position that people use, but there's a lot of other options that we probably aren't aware of or don't know about. And hopefully we'll have more data to present about this sometime in the near future. Thank you. And then any other questions? We'll monitor the chat to see if any other questions come in. Trey, I don't know if you want to speak to what you guys do in Houston from a cervical collar clearance perspective. Sure. Yeah, we essentially do the same thing. We send them over two weeks and then see them in clinic with the Flex-X film. So essentially the same process. Very good. Well, thank you guys so much for being here with us and we'll keep moving forward with our seminar today. Okay, this is Todd Hankinson from Colorado and I'm just here with some of our awesome support teams. This is Shannoni and Katie and Jenny and they rock and they keep our service working and take care of all of our patients and we would be nowhere without them. Yay! Teach and learn while taking piece. Our advanced practice providers are critical members of the neurosurgery team. They help us incredibly by taking care of patients on the floor in the clinic and in the ICU handling consults and inpatient questions while we are operating in the OR. Hi there, I'm Francisco Perez. I'm one of the pediatric neuroradiologists at Seattle Children's and for the next 14 minutes we're going to talk about brain tumor imaging, some of the highlights and pearls and key imaging features. So when I think about the role of neuroimaging in pediatric brain tumor patients, I think of five critical areas. So first, neuroimaging is fundamental for first detecting the brain tumor, accurately diagnosing the brain tumor. Second, predicting the brain tumor pathology. Third, appropriately staging the brain tumor so patients get the appropriate therapy. Fourth, to assess the extent of resection and then finally on the follow-up MRIs to monitor for tumor response, recurrence, and complications of therapy. So for today's talk, we're going to be focusing on these first three areas, these first three roles of neuroimaging, which is when a patient is newly diagnosed and prior to their surgery. So detecting accurately diagnosing brain tumors, we have two real fundamental options in a child where we're suspecting the possibility of a brain tumor. We've got CT scans and MRI and each have their advantages and disadvantages. With CT, it requires ionizing radiation, which we try to minimize as much as possible in children. However, it's fast and readily available and it does not require sedation in most cases. However, it has a poor evaluation of the posterior fossa, which is important because most of the pediatric brain tumors that we see do occur in the posterior fossa. So we want high sensitivity for imaging that area. One strength for CT is it does identify calcifications very reliably. So our situations where I've got a brain MRI and I will recommend a CT scan to further characterize a tumor because I really like the CT for that purpose. MRI does not require radiation, which is great, but it is less available, particularly when sedation is required and pediatric anesthesiologists might be required. These are long studies, maybe 30 minutes, sometimes longer. So the children need to hold very, very still. So that's why it requires sedation. However, it is the gold standard for evaluation of brain tumors and it does provide excellent evaluation of the posterior fossa. MRI can identify calcifications and that can be important to identify different tumor types, but not necessarily to the same degree of specificity and sensitivity as CT. And then finally, MRI is helpful for staging tumors, particularly identifying metastasis in the spine and the brain. So the bottom line is if an MRI is not available, a CT scan is a very reasonable first step to identify tumors and their complications. Here are just some examples of the benefits of CT, some of the strengths. This is a case of a child with hydrocephalus. You can see on this non-contrast head CT, the ventriculomegaly. There's low density in the white matter surrounding the ventricles indicating interstitial edema in the white matter relating to a fourth ventricular mass, which was not shown here, but readily detectable on CT. Here's another case, this child presenting with a posterior fossa tumor, this axial non-contrast head CT showing one important thing is just recognizing that the fourth ventricle is a sentinel. We want to look very carefully at the fourth ventricle to see if it's displaced or effaced, which would indicate a subtle posterior fossa tumor. In this case, the tumor has areas of high density indicating hemorrhage or calcification, very readily seen on CT. In another case, this is a child with hydrocephalus, the sulci or efface in the inferior frontal lobes. And in the fourth ventricle, there's a hyperdense mass, which is really easy to see. And just based on the CT, I know with very high likelihood, this is going to end up being a medulloblastoma. Medulloblastomas are very cellular and they show up as hyperdense on CT. So CT can be very, very helpful for identifying even tumor types. On the other hand, CT has some limitations. Here's an example of that. This was a five-year-old presented to our emergency department with worsening headache, recurrent vomiting, and signs of increased intracranial pressure. And this is an axial non-contrast CT image through the posterior fossa. And we see those dark streaks through the pons. And it's really hard to detect an abnormality there. This was initially read as normal, but if we look carefully, the pre-ponting cistern is effaced. And when the MRI was performed a month later, when the child had persistent symptoms, there's an area of flare signal hyperintensity. We'll talk about sequences shortly for MRI, but flare signal hyperintensity in the pons, which is expanded and encased in the basilar artery consistent with a diffuse midline glioma or diffuse intrinsic ponting glioma. Very hard to see on the CT scan, but much easier to see on the MRI. The other role of MRI and imaging is to classify whether a lesion is a tumor or not. And sometimes that can be really hard. Here's an example. These are two separate patients. The one on the left is a 16-month-old with seizures. The one on the right is a six-year-old boy with left-sided weakness. They both have these flare hyperintense lesions just on opposite sides of each other. One of these is a tumor and one is not. And all the other imaging features can be really, there really wasn't anything specific to distinguish between tumor or non-tumor. Both of them underwent biopsies and resections. The one on the left was a ganglia glioma, and the one on the right was tumor-affected demyelination, so not a tumor. So not everything is a tumor, even if it has mass-like features. So the first takeaways for this section are, one, CT can be really helpful for supporting brain tumor detection and diagnosis, but MRI is almost always preferred, and not every brain mass on MRI is a tumor. Well, the first polling question, this is a four-year-old girl who had a non-contrast head CT to evaluate headaches and vomiting, and what other neuroimaging is essential? Brain MRI, brain MRI, and total spine MRI, brain MRI with MR spectroscopy, or brain MRI with MR perfusion. So yeah, every child who has a brain tumor basically should have a brain MRI and a total spine MRI, preferably before surgery. In our last polling question, we'll show you an example of why it's important to get the spine MRI before surgery, because it can be more difficult to interpret those images after surgery, particularly suboccipital craniectomies. So now that we've talked about detecting and diagnosing brain tumors accurately, and the role of neuroimaging there, we're going to look at its role in predicting brain tumor pathology. And one approach is to look at the tumor location, and then to look at the MRI signal characteristics, and use those two features to create a differential diagnosis and to rank what you think the most likely tumor type is. This is important, so to give the surgeons an idea of what kind of tumor you're thinking about beforehand, if the tumor is in a critical structure, and you think it's a benign tumor, they may be able to take a less aggressive approach up front surgery. However, if we think it's a high-grade tumor, like an ependymoma or medulloblastoma, we're getting as much of the tumor out is so critical for that patient's survival, we can tell them that up front, and they can be more assertive and have a more definitive up front surgery. So in terms of pediatric brain tumors by location, here's just one rubric that people have used. We've got posterior fossa tumors, supertentorial tumors, and within the supertentorial region, we've got tumors that arise in the cerebral space, paracellar space, paraventricular, pineal region, and meninges. For today's talk, we're really going to focus on the posterior fossa tumors, so pilocytic acerocytomas, medulloblastomas, and ependymomas. And then for brain tumor characterization, the MRI signal characteristics are really helpful. This is a lot of information on this slide, but just recognizing how there are different sequences in MRI, and each of these sequences are performed pretty routinely in a brain MRI protocol, and they give us different information about the tumor type and the tissue type. So we can program the magnet to reveal certain pathologies or certain tissue characteristics. We've got a T1-weighted image, which shows fat and hemorrhage, calcification, and contrast enhancement as bright. We've got T2-weighted images, axial and coronal, two different planes, which show CSF is very, very bright. Cysts in tumors and edema can be really bright on T2. We have flare, where we take the bright T2 and we suppress the signal in the CSF spaces. So flare is very helpful for leptomeningeal processes, particularly leptomeningeal disease, spread of tumor. We have diffusion-weighted imaging, we'll talk more in detail next, but this is a measure of how water moves in different tissue types, so it gives us a sense of tissue cellularity. Susceptibility-weighted imaging, which is, we've tuned the magnet to look for areas of hemorrhage and calcification. We have some optional sequences, including advanced imaging like MR spectroscopy, where we look at the metabolites in different tissue types, which can sometimes be helpful. We also do post-contrast imaging. So the gadolinium-based contrast agents that are administered intravenously, they leak through the blood-brain barrier, and then where that leakage occurs, we get enhancement. So where there's neovascularity in certain tumor types, we get enhancement. We use the T1-weighted images after contrast to detect that enhancement pattern, and in some cases, we'll do perfusion to see how well perfused or how much blood volume is within a tumor. When we look at these brain tumors as a neuroradiologist, I have MRI descriptors, so I will look at the different signal characteristics, and I'll answer certain questions, such as the location of the tumor, whether there's edema or not, whether there's calcifications or hemorrhage, the degree of cellularity based on diffusion-weighted imaging, the degree of blood-brain barrier enhancement and breakdown based on post-contrast images, and the presence or absence of CSF metastasis. There's a bunch of other features, but in my mind as a neuroradiologist, I'm looking at all these features, looking at where the tumor is, looking at the signal characteristics, and trying to decide what type of tumor I think it is. So here's our next polling question. A brain MRI obtained in an eight-year-old boy with daily headaches, diplopia, and vomiting, and what's the most likely tumor type? We've got a diffusion- weighted image. This gives us a sense of cellularity, and we have a post-contrast image here showing enhancement. In this case, this tumor, which I'm going to show you in the next section, has facilitated diffusion, so very rapid water movement, and it shows enhancement. So this is a typical appearance for juvenile pyelocytic astrocytoma. So as an example, we've got a posterior fossil tumor, diffusion-weighted imaging, which is one of the MRI sequences, very helpful, where if you look at your PACS workstation, if you look at the sequence called ADC, apparent diffusion coefficient, you can put a region of interest and actually measure the values in the tumor, and what I'm going to show you is that there's low water movement in highly cellular tumors, like medulloblastoma, which are also dense on CT, and there's greater water movement in low cellular tumors, such as juvenile pyelocytic astrocytomas. So here are the three most common posterior fossil tumors in a patient. So we've identified the location, posterior fossil, and with one signal, with one sequence, the diffusion-weighted sequences, we can predict with pretty good accuracy pyelocytic astrocytomas, which have high ADC values of 1500, or facilitated water diffusion, and medulloblastomas, which have restricted diffusion, or very low ADC values, almost half as much as the pyelocytic astrocytomas, based on cellularity. So between location and a single sequence in MRI, we can predict what the pathology is. So two takeaways from this section are specialized MRI sequences are used to determine the brain tumor signal characteristics, which help us predict what the tumor type is, can help guide surgery planning, and then diffusion-weighted MR imaging, in particular, can help us distinguish between posterior fossil tumors, based on the degree of cellularity. And then finally, we need to be able to appropriately stage the brain tumors, and we're going to use MRI for that. This is a post-contrast sagittal T1 image of the brain in a child with a new posterior fossil tumor. So here's the posterior fossa, here's the cerebellar tumor, and we can see that there are all the hot spots for where we find leptomeningeal disease, leptomeningeal spread of tumor. I really like the sagittal T1 images, because some of the most common areas for metastasis are present. This is the floor of the third ventricle, very, very common, the interpeduncular cistern, also a common spot, and then the obex, another common spot. Here are three other examples of where these leptomeningeal metastases most likely occur. These are all MRI sequences. The first here, post-contrast, we see studying along the leptomeningeal folia. We see the interpeduncular cistern, and then realize that it's not just the post-contrast images that are helpful, but other sequences, particularly for medulloblastoma that have restricted diffusion. The diffusion-weighted imaging is really helpful for identifying non-enhancing metastasis. So in this case, we see metastasis in the olfactory bulbs, just adjacent to olfactory bulbs in the recess there, and in the periventricular spaces along the superior cerebellar peduncles, and then here's the fourth ventricular mass. We also do spine MRI. So here's a typical protocol, including axial and sagittal post-contrast T1 sequences of the spine. It's really important to have thin spine MRI slices, typically three millimeters or less, and not to have gaps. So here's a spine MRI. We can see all these leptomeningeal metastases along the spine, and then we have some optional sequences, including diffusion-weighted imaging, which is particularly helpful for medulloblastomas. Here's a typical example in a child with medulloblastoma, just really diffuse metastasis, and also along the spine. Medulloblastoma is pretty high risk for leptomeningeal metastasis, and don't forget that also low grades, like pilocytic acerocytomas, can have metastasis. This is a child represented with a pilocytic acerocytoma. We can see on the ADC maps, the diffusion-weighted imaging, this is very free water movement, but there was a metastasis, sorry, there was a CSF dissemination at the time of presentation, which is why we do spine MRI in all of our children who have brain tumors. So our final polling question, a spine MRI was obtained in a child one day following resection of a posterior fossa tumor, and there's diffuse abnormal enhancement on the spine MRI. Here's the suboccipital craniectomy for the resection of the posterior fossa tumor. So what's the next step? Treat for spine metastasis, perform a biopsy of this enhancing tissue, repeat a spine MRI in three to four weeks, or defer to CSF analysis for spinal metastasis staging. In this case, we're going to repeat a spine MRI in three to four weeks. This enhancement pattern is not uncommon after a suboccipital craniectomy. It can be pretty confusing for people who don't know about that. As a neuroradiologist, when I have an indeterminate finding, I often want to fall back on CSF analysis. However, that's not helpful. CSF analysis and spine MRI findings are not always concordant. You might find metastasis on the spine MRI, but the CSF analysis is negative, and many others in prior studies, the MRI staging is actually more predictive of survival compared to CSF analysis. It's really important as a neuroradiologist to get the spine imaging right, to get the staging appropriate, to provide the best treatment for these patients. So the final takeaway is, so careful MRI imaging of the brain and spine is needed to appropriately stage pediatric brain tumors, even in suspected low-grade tumors like pilocytic astrocytomas. And preferably, we want to perform the spine MRIs before brain tumor resection. So in closing, we really highlighted the three important areas of neuroimaging in a new patient with a newly diagnosed brain tumor. And finally, there's the opportunity to look for extent of tumor resection after surgery, and then look for tumor response, recurrence, and complications. Thank you. Are we ready to start? Oh, great. Hi. I'm Peter Chiarelli. I'm one of the neurosurgeons at Children's Hospital Los Angeles, and today the talk is focused on pediatric diffuse brainstem tumors. So these tumors can come in many forms. This is an example here of a diffuse brainstem tumor, but tumors can also look highly circumscribed, and they can also grow away from the back of the brainstem. These are just a few simple examples, but not all brainstem tumors behave in a similar manner. So understanding the difference between these can really be essential for giving an appropriate prognosis and an appropriate treatment plan, including when surgery is necessary. Before we talk about that, first I'll give you some demographics and some classification schemes for brainstem tumors. And overall, from the most current data that we have available in the United States, brainstem tumors represent 10.9% of new central nervous system neoplasms in kids. By comparison, the same number is 1.5% in adults. There are approximately 500 new cases per year, and this is from the CB Trust Statistical Report that's most recent. And this diagnosis of a brainstem tumor is most common in kids aged five to nine. Masses in the brainstem are substantially heterogeneous from one another by diagnosis, and diffuse tumors are actually the most lethal pediatric high-grade tumors by location. In contrast, other tumors, the ones that are more focal in the brainstem, have a comparatively favorable prognosis. And with the goal of inferring characteristics of histology and prognosis from the data that we have available from radiology, et cetera, people have looked at the tumor location and the appearance on imaging, and they've analyzed them pretty extensively. Over the past three decades, various classification schemes for brainstem tumors have evolved, with names like you see here below. But for today, we're just going to focus on this category of masses, the diffuse masses. The most relevant dividing line remains between masses that are diffuse on imaging and those that are well-circumscribed. And really, a discussion of diffuse tumors is essentially a discussion of DIPG, or Diffuse Intrinsic Pontine Glioma. So DIPG is an invasive neoplasm. It tends to originate in the pons. It's approximately 75% of all pediatric brainstem tumors. It's also the most aggressive pediatric primary brain tumor. And even with the best available treatment, the prognosis, as many of you know, is dismal. Overall, some of these statistics are median overall survival of 8 to 12 months, a one-year progression-free survival, about 20%. Mean overall survival at one year of 45%, and mean overall survival at two years of less than 10%, so pretty terrible. What's more is that over the last three decades, the change in overall survival has been basically zip. Just a few actuarial measures talking about DIPG, the median age for a new diagnosis is around 6 to 7, and the peak in incidence for kids is from around 3 to 9 years. Just to note, a second peak in incidence does occur in adults around the median age of 34 years, although this is thought to be a substantially different tumor in its origin because the progression-free survival and overall survival are about five times greater than when these tumors appear in kids. The male-to-female ratio is one-to-one. So given the limited options for treatment and the challenging nature of this tumor location in the pons or the brainstem, we've long had this motivation to reliably and non-invasively identify DIPG, and therefore the semantic entity or typical or classic DIPG has been very important to understand because historically it's had a role as a surrogate for tissue diagnosis. So the utility of this framework, this typical classic framework, is really its proven positive predictive association with histologically confirmed DIPG, and like I say here below in yellow, in practice, this correlation has been well-documented in cases with these typical features. So then you might ask the question, what actually defines a typical DIPG? So occasionally the definition has been controversial, but a reasonable description includes certain clinical features. And so those clinical presentation features are a short symptomatic course before presentation, cranial nerve signs, cerebellar signs, long-track signs, for example, hyperreflexia or weakness, and extracular movement disorders, which can fit into the cranial nerve signs. Certain radiographic features that are commonly classified as typical or classic include a central location within the pons where the tumor occupies greater than half of the pons, lack of well-defined outer margins on imaging, hyperintensity on T2 and hypointensity on T1-weighted imaging, and lack of significant enhancement. There can be an enhancing region, but most commonly it occupies about less than a quarter of the tumor volume. Also much like Dr. Perez mentioned, the basilar artery is often engulfed by the tumor itself, and that feature can allow it to be designated as classic. Certain features, also sometimes controversial but often cited as atypical, can include prominent enhancement rather than just sort of a minimal amount of enhancement, restricted diffusion, cystic components, and a prominent exophytic component, meaning a portion growing away from the brainstem. So with that in mind, how does characterization of a DAPG as in this framework of typical or classic versus the atypical, how does it impact our decision to do a biopsy of this tumor and our care overall? So the decision-making process for a stereotactic biopsy can differ between neurosurgeons. Now prior to the general availability of MRI and modern stereotaxy, the rates of morbidity and mortality for a brainstem biopsy were actually cited as pretty high numbers. We commonly thought of morbidity of about 30% and mortality of about 4%. These are very huge. So these high rates and, of course, our prioritization of patient safety had led to the reduced availability of tissue for investigating and refining our treatment of DAPG over the last three decades. More modern estimates, given the presence of modern stereotaxy and modern imaging, and these come from large meta-analyses like the citation you see below, can be cited as a 6.7% transient morbidity, 0.6% permanent morbidity, and a 0.6% mortality. The average rate has been calculated of non-diagnostic biopsy for presumed DAPG, and this is around 3.9%. So a reasonable current recommendation is often included as offering a biopsy in the setting of an atypical appearing lesion to rule out pathology that would alter treatment strategy or offering a biopsy for a typical appearing lesion when it serves as the entrance criteria for an available clinical trial. Now this all sounds pretty straightforward, but the seeming simplicity of these criteria for biopsy is tempered by reality. From the study that you see cited below, in response to a survey with brainstem tumor images and clinical scenarios given to pediatric neurosurgeons, a majority agreement of three-quarters or more between pediatric neurosurgeons was only observed in fewer than half the cases, and a median 5% of pediatric neurosurgeons said that they would choose to biopsy a lesion that they themselves designated as typical with a range of 1% to 67%, and an 18% median range with a huge range of 4% to 100% would avoid biopsy of a lesion they themselves called atypical on the survey. So you can see that although we might cite guidelines of when to biopsy, this is highly variable across the country at the present time. So let's talk about the modern role of biopsy. So data from retrospective reviews from biopsy lesions, they naturally contain selection bias because historically the lesions with atypical appearance were the ones that were biopsied. So among these lesions, and again these data that I'm showing on the screen are best regarded as representing the atypical appearing group of tumors, what had been found was a glioma histology or something compatible with DIPG around 84%, and then other tumors such as a PNET, which is no longer a PNET, now it's an embryonal tumor with multilayer rosettes of 3%, pentamoma in the 2% range, other tumors, infection, inflammatory disease, and other non-neoplastic disease with significant percentages. And of course, a diagnosis of one of these other tumors or inflammatory disease would dramatically affect prognosis and the treatment strategy. For example, a diagnosis of an ETMR, this embryonal tumor with multilayer rosettes, carries a more rapid dismal prognosis. It would require a workup including full CNS imaging, a consideration of surgical resection, and complete craniospinal radiation, which differs a little bit from DIPG. Furthermore, like we were mentioning, non-neoplastic conditions like acute disseminated encephalomyelitis require corticosteroids as first-line treatment. It's very different from DIPG. For lesions with a typical appearance, the biopsy tissue was historically designated with a WHO grade, and those ranged from 2 to 4. However, this is quite significant. The body of literature really failed to find a meaningful association between these histologic grades that we used to grade them as and the overall survival or the progression-free survival of DIPG. And this was puzzling, right? Nobody understood why the histologic grade really didn't correlate with outcome. Often people assumed it was from the location of these tumors and some other things. Often it was assumed that this lack of association was because of sampling error on the biopsy due to heterogeneity of the tumor itself. But really progress in the molecular and the genomic evaluation of DIPG has suggested that the clinical behavior is explained by more in histology alone, meaning that the grades 2 through 4 that we used to apply aren't actually the appropriate grading system for DIPG. And meaningful progress in our current understanding was gained in 2012 upon the discovery of the H3K27M mutation. So let's briefly discuss what H3K27M is. So this is a somatic gain-of-function mutation that's present in more than 80% of DIPG. It's a single amino acid substitution on the histone at the 27th amino acid on histone 3.3 or 3.1. That's where the H3 comes from. And it's from a single amino acid of a polar-charged lysine to a methionine, which is hydrophobic. And this change makes a significant difference because due to this conversion, the interaction of that histone with the complex that leads to methylation and compensatory acetylation of the histone or the DNA in the surrounding area is dramatically altered. Because of the 2016 WHO reclassification, these K27M mutated DIPGs have been designated a specific subset of a whole class of tumor called H3K27M mutated diffuse midline glioma, another thing that Dr. Perez had alluded to in his talk, which by itself, the H3K27M mutated diffuse midline glioma is grade 4 regardless of histology, regardless whether a pathologist would call it 2, 3, or 4, it's automatically 4. The overall survival of DIPG with this mutation is significantly shorter than that of H3 wild type DIPG. And remember that most DIPGs do have this mutation. And it's also been found that this mutation influences prognosis to a greater extent than pretty significant other factors like age, histologic grade, location, and even treatment of the lesion itself. So speaking of treatment, let's briefly discuss therapy. So in the modern era, biopsy, for example, biopsies that we perform at our institution and biopsies that they perform at many other pediatric hospitals, can be performed using adjuncts to improve surgical safety and accuracy, including three-dimensional stereotaxis and robotic neuronavigation. Microsurgical resection is not an advantageous therapy for DIPG. This has actually been studied through meta-analysis, and they found that among tumors that had an attempted resection in any respect actually had a shorter overall survival than those that did not have any attempted resection. So the specific benefit in DIPG is derived from radiation. Now radiation therapy for DIPG is typically delivered as a fractionated focal dose. So in the form of intensity modulated radiation therapy, or IMRT, to a region spanning around two centimeters around the tumor. And common radiation parameters, just for interest, include around 1.8 to two grade daily fractions over about six weeks. And as a result of radiation, the symptoms and quality of life do temporarily improve in about 80% of patients. And there's a stabilization or a reduction of tumor size in around 50% of patients. The overall survival is estimated to increase by about three months or more as a result of radiation. But even with radiation, tumor progression occurs on average three to eight months after therapy. And the time from progression to mortality is around one to four months, so pretty dismal. Modifications of the current regimen have all been attempted in many different forms, hypo-fractionated radiation, hyper-fractionated radiation, all without substantial benefit. So as this implies, improved therapies are dramatically needed in the case of DIPG. Over 200 clinical trials have been performed using very extensive range and combination of chemotherapeutics, for example, intrathecal chemo, targeted therapies, etc., all without substantial benefit. And as of June 2019, when I did a kind of thorough check, there were 28 clinical trials in the U.S. that were recruiting specifically for DIPG and 13 active studies that were still being analyzed but no longer pursuing recruitment. So a lot of academic activity around DIPG. Now there are promising new options that take advantage of this knowledge about KH3K27M. For example, histone deacetylase inhibitors are main candidates in current clinical trials. These are drugs that you might have heard of, like panobinostat or virinostat. There's also bromodomain inhibitors. There's jumanji domain demethylase inhibitors. These are some terms that you might hear thrown out around in neuro-oncology conferences. However, most of the clinical trials that if you pick up a pediatric neurosurgery textbook and read it, for example, targeting of PDGFRA, EGFR, VEGFR, these have actually already shown a lack of efficacy. And so most of what you'll see in textbooks has already been kind of debunked. Immunotherapy is also a current hot topic in DIPG investigation in active clinical trials. And this comes in the form of vaccine therapy, for example, vaccines to a glioma-associated antigen peptide, dendritic cell therapy, CAR-T cell therapy, which has made significant advances in other cancers, and oncolytic viruses. DIPG is a significant target that we're pursuing at our own lab at CHLA. Certain things that we pursue in order to try to solve some of the questions in DIPG are three-dimensional cell culture and modeling of the native DIPG tissue microenvironment. We use patient-derived multicell co-cultures, and we're using human brain-derived scaffolds that are decellularized to better reflect characteristics of the ECM and this malignant phenotype. Our goal is to move towards a high-throughput patient-derived assay, for example, a microfluidic assay for drug screening. We also perform nanoparticle therapy investigation and convection-enhanced delivery. But thank you very much, and I'd be happy to take questions when the question-and-answer session comes up. My name is Ralph Vermoyen, and I'm from the University of Washington. One of the questions that comes up on a weekly basis is, when we have a patient with a pediatric brain tumor, should they get treated with protons or photons? And that's the decision-making process that I want to walk you through today. Here are my disclosures. I work for the University of Washington, where I treat with protons and photons and neutrons and electrons. I'm also a senior editor for pediatrics for the Red Journal. Imagine the Y-axis is a patient's skin, and a tumor is this deep in her body. The radiation beam is coming from left to right, passing through some healthy tissue, then the tumor, and then healthy tissue beyond the tumor. With a photon or X-ray radiation, the radiation affects the tissue between the patient's skin and the tumor, and the tumor and the tissue beyond the tumor. With protons, there's an entrance dose of radiation. Then we deliver the amount of radiation we want to the depth we need to treat the tumor, and then we slam on the brakes, and the radiation stops shortly after the tumor target. What's the evidence for protons in pediatric cancer? Mostly case series. There's a retrospective outcome comparisons. There are a few phase two studies at single institutions, but there are no randomized trials in pediatric patients, and I don't think there ever will be, because to consent a parent of a child, we would have to have that parent agree to a computer deciding whether their child receives more or less radiation to healthy tissue, and most families are not going to sign up for that. The early reports from studies that have been done suggest some favorable side effect profiles for the pediatric patients who receive radiation. The study from MD Anderson showed better neurocognitive outcomes in patients who were treated with protons compared to photons. There are other studies that show benefits in terms of neuroendocrine function, and in terms of hearing, and a variety of other parameters. So why don't we use protons for every pediatric patient who needs radiation? Well first of all, if we're treating with protons or photons, we're going to be aiming the same targets for the same number of treatments and we expect the same results against the tumor. The side effects are what really drive the treatment decisions. It takes longer to treat a patient with protons than with photons. With protons, it takes one to two weeks to plan that radiation. With photons, we can get started within a day or a couple of days. To treat with protons, there has to be an organ at risk that we're trying to protect that's distal to our target and sometimes there isn't. With protons, the logistics of travel and insurance can sometimes be more difficult. Fortunately, insurance usually covers proton therapy for children. However, the logistics of travel are difficult both from a disease perspective, so for example, a patient might have a hard time getting to a proton center from another city after she's recovering from surgery, but also from a family dynamic perspective of a family having to uproot and go to a different city to receive protons over the course of eight weeks. Finally, there's some important biologic questions that I wanna highlight. The University of Florida has really been pioneers in terms of examining the potential increased toxicity of proton radiation compared to photons. So they looked at 300 patients with brain tumors who were treated with protons and they found that of the patients who had posterior fossa tumors, nearly 11% of them had symptomatic brain stem injury or radiation necrosis. And the term I use to describe the side effect to patients is potentially permanent, devastating, and life-threatening, and I say that if a patient has radiation necrosis, all the benefits of protons don't matter. The comparison number for photons is about a one to 3% incidence of symptomatic radiation necrosis when treating patients in the posterior fossa. They looked at a couple of dosimetric criteria, the criteria we as radiation oncologists look at when we evaluate plans. And basically if they found that they had to prescribe to a higher dose than 54 or 55.8 gray, then there was about a 11% incidence of radiation necrosis. And they also found that if they were unable to spare the brain stem of some of the radiation treatment, then there was also an 11% incidence of radiation necrosis. It's important to say we exceed these criteria all the time with photons and don't have an 11% risk of symptomatic radiation injury. So these are criteria that I look at now when I determine whether patients should get protons or photons. Do we have to prescribe to greater than 55.8 gray of radiation? And can we spare at least some of the brain stem that full dose of radiation? So I'm gonna walk through five criteria we use at the University of Washington to decide whether a patient should get treated protons or photons. First, is there excess risk of injury to normal tissue, particularly the brain stem? Second, can we make the timing and logistics work for the patient? Third, are there organs at risk that we are going to be able to protect that need to be protected with protons? Fourth, is the prognosis good enough for the patient so she or he is likely to see the benefits of protons? And the last one is one that I'll allude to with an example I'll provide that sometimes the proton plan will look better on paper than in reality. It's important to say that these are just examples of how we approach the question at the University of Washington. These are not formal guidelines. Other people have different approaches. I will say that we have a relatively high bar for using protons at the University of Washington in spite of the fact that we have protons at our disposal. I get a lot of referral for protons, but in spite of that, about a third of my practice continues to be photon. First, diffuse intrinsic pontine glioma. As you all know, this is a universally fatal tumor with a median survival of about nine to 12 months. When all is said and done, nearly all of the brainstem receives 54 gray. And these patients need to be starved as soon as possible. I have never treated a patient with DIPG with protons, and I won't ever do that for a variety of reasons. First, I think there's excess risk of toxicity because again, we can't spare much or any of the brainstem of that full prescription dose. So these patients are gonna be at elevated risk of symptomatic brainstem injuries. Second, the logistic concerns are paramount. These patients need to start ASAP, and with protons, it's gonna take longer to get them started. And unfortunately, their prognosis is quite poor, and they're unlikely to see any benefits from protons. Average medullastoma is perhaps the opposite example. These patients have a very good prognosis. They need cranial spinal radiation with a boost to the resection bed plus margin, and they should start within one month of surgery in order to not have an excess risk of subsequent leptomeningeal recurrence of disease. I've treated all of these patients with protons since our center opened. I do discuss photons as a reasonable option with patients. We have learned from the Florida analysis how to plan our radiation to spare the brainstem some of the radiation dose, and so I think we can keep the excess risk of injury minimal to any. There are some logistical concerns. I'm fortunate to work with an amazing team that one way or another gets patients here promptly to the University of Washington, and we get them started within about a month of their surgery. We certainly are protecting organs at risk, and these patients' prognosis is good, and there's no particular physical challenges to treating these patients with protons. The pendymoma is more of a gray area. A majority of these patients survive, and the extent of resection is critical. For patients with gross residual disease, they should receive 59.4 gray of radiation. For patients who've had a gross total or near total resection, they may not need quite that much radiation, and the radiation may only need to go to the fourth ventricle if the tumor was confined to that area. However, sometimes this tumor will wrap around the brainstem through the foramen elusca and into the prepontine cistern, in which case we're not gonna be able, even with protons, to spare much or any of the brainstem. So if a patient has gross disease and needs to get the higher dose of radiation, 59.4, I will typically treat them with photons. If the disease wraps around the brainstem and goes to the prepontine cistern, then those patients I also treat with photons. For patients whose disease is not as extensive and in whom have had near total or gross total resections, I'll treat them with photons. We usually can make the logistics concerns work out. We are certainly protecting organs at risk, and again, many of these patients are long-term survivors. Basal skull chordomas, to some extent, are the so-called poster children for proton radiation. These, as you all know, are very aggressive tumors that are technically benign but don't behave that way. A gross total resection is critical when feasible. However, that's very difficult to achieve. And we prescribe very high doses of radiation to these patients, typically in excess of 70 gray of radiation. This is a patient who is a teenage girl with a clival chordoma with extension of the tumor, not only into the cella, but also a soft tissue extension that was displacing the optic chiasm. We have an amazing basal skull surgeon who somehow achieved a gross total resection, but the patient still needed adjuvant radiation treatment. I delivered this radiation plan to her. She is doing well. We maximally spared what we could of the hippocampi and the temporal lobes, hopefully preserving her neurocognitive function. We also spared as much of the brainstem as we could, and she was at excess risk of secondary malignancies because of her underlying genetic syndrome, and we minimized the amount of normal tissue exposed to radiation. However, this patient I did not treat with protons. It may not be obvious, but everything I need to know about this patient's tumor, I could tell from this plain film. No, you can't see the basal skull chordoma in this image, but what you can see is the hardware, and hardware poses some particular challenges when treating with protons. The first challenge is, whether it's protons or photons, neither type of radiation goes through hardware well, but protons have a particular problem going through hardware. But the second is a more subtle problem, which is that when a patient has metal hardware, that creates streak artifacts in CT scans. Radiation oncologists use CT scans to plan their radiation and count on the accuracy of the imaging we're seeing, and streak artifacts are not accurate. So when a radiation oncologist plans on a CT with streak artifact, it introduces the possibility of error in terms of knowing where the radiation is being deposited, or in protons, where it's stopping. For that reason, my bar is very high for using protons with a patient who has metal hardware in my radiation field. So this patient was treated with photons, and that's one of the unique physical challenges that we sometimes confront when treating patients with protons. When it comes to brain re-irradiation, one might say, gosh, if you're worried about radiation side effects the first time around, then the second time around, you should be particularly concerned about it. And indeed, if a patient has indications for cranial spinal radiation, as their second course of radiation, we typically will do that with protons because we think the advantages in terms of sparing healthy tissue are worth it. However, if it's focal radiation that's being done, we typically do that with photons. Our thinking is that if there are some concerns about biologic effects the first time around, then the second time around, those effects and risks are probably magnified. Said another way, if the patient has a injury because of radiation the second time around, we wanna be sure that that was going to happen regardless of the type of radiation that was being deposited. Regardless of the type of radiation that was used and not because we chose to use protons over photons. The bottom line is these cases are complicated and I would encourage you to reach out to your radiation oncologist or to any number of the proton radiation oncologists across the country and get their thoughts. They'll tell you about whether protons make sense or not. And in many cases, photons are the right way of treating a patient. Everyone's goal is just to make sure that the patients get the best treatment. Thank you. I'd like to thank all our speakers for taking the time today to speak to us. I think those talks were excellent and we do have a couple of questions and feel free to ask more. One of the first questions that we have is to Dr. Perez. There's been a little more discussion. Obviously we're avoiding radiation with CT scans. Are you hearing more discussion about contrast being harmful and trying to avoid contrast? Yeah, that's a really good question. So I think there's still a lot that we're learning about gadolinium-based contrast agents, particularly in children. Some of the newer agents are felt to have lower risk, but we're still trying to define those risks. For certain brain tumors, for instance, optic pathway gliomas, we're not routinely doing contrast on those patients because they get so much surveillance scans. So we are moving our practice away from doing contrast unless we really feel it's necessary just because that risk is unknown. It's still felt to be very low risk, but not a zero risk. Thank you. Our second question is also for Dr. Perez. It's regarding the screening spine or the limited spine MRIs. And if you think that is an appropriate test when looking for metastases in a newly diagnosed brain tumor patient? Yeah, I mean, we're constantly updating our spine MRI sequences. Getting the staging correct from the beginning is huge because either they're gonna get radiation therapy or they're not. I mean, it really changes management so drastically that I know that our brain tumor board, when I'm presenting, they will nail me down and they want me to know, they want me to say like, is that metastasis or not? I can't give them a waffle. I gotta say yes or no. So I need the best imaging I can get. And we've seen that better imaging gives us more sensitivity. So thinner slices, certain sequences like diffusion-weighted imaging. So in fact, we're moving away from limited MRIs to even more comprehensive, longer MRIs to really get that staging right for our neuro-oncologists, our radiation therapists, and our surgeons. All right, thank you. Dr. Morian, next question's for you and thank you for coming on live also. One of the questions is, what is the cost difference between proton and photon and does that play into your decision-making at all? Yeah, thank you. It's a great question and really complex. I'll try not to take too much of the time and allow time for other questions. From the perspective of the patient, it rarely makes a difference in terms of cost. And fortunately, protons is almost always covered for pediatric patients, regardless of the insurer, whether it's public or private insurance. What we actually find is what matters much more on an individual patient basis is for a individual insurance basis is the contract the insurer has with whatever radiation center. So we've had a couple of cases of patients who we've had to price out both the photons and protons because of some changes in the clinical situation. And it turns out the bill was going to be quite a lot more for photons than protons because of the unique insurance situation. Globally, proton centers cost more, but on a patient-by-patient basis, I can't predict it and they're often surprises. Thank you. We have another question regarding the proton therapy and the risk of vasculopathy following therapy and in what timeframe we would expect to see that. Yeah, that's a great question. So there's some good data from University of Florida that seems to show that the risk of vasculopathy is consistent with what's been seen with photons. There are other studies that seem to show more low-grade vascular events with protons than photons, including some data that was presented at ASTRO last month. I think the timeline we look at is over the course of years. So it's not something that you're going to see early. Okay, are we? And I'll just add to that, even the St. Jude data shows they tend to be low-grade events, not higher-grade events. Okay. Following up from that, is there an age limit with protons? No. We tend to lean on protons more in younger patients, but there's no lower limit. The bigger questions are radiation or not versus which modality to use. When we do treat the youngest patients, we tend to treat them with protons, but the bigger question is for patients under three, what can we do to avoid radiation at all? Dr. Chiarelli, this one is probably for you. Where are they doing CAR T cell therapy for brain tumors and what are your thoughts on that? So, yeah, that's a really interesting question. I'm going to have to actually get back to you in terms of knowing which centers are open for CAR T cell therapy. I would actually defer a little bit of that question to a neuro-oncologist. I know that Nick Vitanza at Seattle has an expertise on those clinical trials specifically. And I actually parenthetically wonder if Seattle is involved in doing CAR T cell therapy, but- We are. Yeah. So we, and Dr. McMurray can probably speak to this a little bit too, but we do have Brainchild currently open that we are enrolling children right now. We have another question that came in through the chat that's for all of the speakers as well as the other APPs, but the question was, what quick guide resource would we recommend for anyone who's a new APP coming into neurosurgery, for tumors, imaging, diagnoses, et cetera? And someone did respond and say fundamentals of neurosurgery, but we just wanted to know if there were any other recommendations for the new APPs. There's a great textbook out and I believe Patty is on with us as well that can speak to it. It's nursing care for the pediatric neurosurgery patient. And I think it gives a fantastic overview. Kathy Cartwright kind of spearheaded that and just a fantastic overview of different diagnoses or kind of recommended treatments, things to look for, typically who has surgery, who does not, how to manage those patients afterwards. And it also has a great chapter at the beginning that talks about a lot of developmental things to be mindful of kind of as the child progresses through age. So that's one and then Greenberg. I agree, it's a good resource. I agree with what Haley said. It is written for to be a resource for the bedside nurse, but it goes into so much in-depth information that it's excellent for new and old providers. I'd like to thank everyone for participating in this chat and our speakers again, thank you for being here and for all this valuable information. We are going to be doing a poll during the break and I encourage you all to participate. One of the things that we are most interested in every year for this discussion is how can we get all the APPs to be in a group that we can have discussions going on or what practices happen at different places and basically get ideas from other facilities. And so in that poll, we're trying to get ideas on how we can facilitate a discussion platform that we can do more of this, basically crowdsourcing all the different facilities around the country. So please participate in the poll. We are gonna go into a 15 minute break. And so we will see you back after 15 minutes and we are excited to keep going. And so thank you for being here. Welcome back, I hope everyone was able to stand up and stretch their legs, get a drink of water during that break. We've had some phenomenal speakers this morning and I'm excited about the speakers we have continuing through the afternoon. Just a quick reminder, if you haven't answered the polling questions, please do so that Mandy mentioned earlier, as we are really trying to reach out and do a lot more networking as a group of APPs across the US and Canada and across the world. So I'm excited about our next session on hydrocephalus. As many of you know, that's the most common diagnosis we see in pediatric neurosurgery. And so this should be an exciting session for us. And so here we go. Okay. So anyway, thanks for having me chat. So I was sort of asked to talk about ETV-CPC or endoscopic third ventriculoscopy and chorioplexis cauterization and for infant hydrocephalus. So just the mandatory disclosure slides. A lot of the work I'll be talking about is the work done by the HCRN. So there's a lot of funders of the network and no personal conflicts. So hydrocephalus, as Haley mentioned, is common or the most common, but it's also costly. This is a study we did, gosh, like 12 years ago now, I think with Tamara Simon that showed hydrocephalus accounts for at least one to $2 billion a year in charges in the US alone and given how old that study is, it's probably creeping up to three. So it's an important thing to study, obviously, and because it has massive clinical impacts. And so just zoom back a bit, like 30 years in the 90s, we're miserable. 60% of shunts were still being revised in two years. That wasn't much better than the 20 years before that. Everyone was making progress. The internet became a thing and became a massive thing. We drove rovers around Mars and we basically stopped the spread of HIV, yet shunts still stalled. So one of the great hopes was endoscopic third ventriculostomy. And so here's a video for the crowd. So that's in the lateral ventricle. Now we're in the third ventricle. You can see the orange is the pituitary. The basilar artery, you can see pulsing in the distance and making a hole from the third, the floor of the third. So CSF can escape the ventricular system into the subarachnoid system. And there you can see the pulsating basilar. And so it works great for older people, older kids and adults, but not so much for infants. And so here's a study at the time from Harvard that showed how miserable ETVs alone did with shunts and basically was abandoned. And so even at Abacus Carney's ETV success score, it tells you, I mean, that anyone under six months, you're getting a low score right off the bat. And then let alone the etiologies are unfavorable. So this is an HCRN slide. This is part of our registry. And this is what hydrocephalus looks like in North America in the HCRN. And you can see that ETV is useful for aqueductal stenosis and tumors. However, aqueductal stenosis does occur in babies, but the tumors usually do not. And so ETV alone is only useful in about a third of etiologies and those etiologies generally aren't infant etiologies. And so again, you see the limited utility of ETV alone. So hence came the ETV with choroid plexus cauterization or coagulation. It was born right there in Uganda, Africa. And there's the father, Dr. Ben Moore, who at Cure Hydrocephalus Hospital, I ran out of shunts and developed this. And there's when I went to visit and train with him. And so you're gonna see this is a very fast clip because it's a long operation. It's hard on the eyes to watch, but anyway, you use a flexible scope, priming the scope, you're going into the third ventricle. It's a myelomeningocele baby, opening this up. And you'll see, we push the scope all the way down and spoke to the foramen magnum. And then you can see some cauterization going on there. I might just replay it. And so that's the difference. Making the hole, that's the same as ETV. So this is all the same, except you can go a little lower and then doing that millimeter by millimeter on both sides. So what did they find? So this is a massive study of 550 children in Africa and in Uganda. And it showed that the ETV versus the ETV CPC in children over one and certainly two doesn't add anything. But when you use it for infants, it can add like 15, 16% of success, which can make all the difference. So that's great where shunts aren't abound, but we have lots of shunts as Jason's gonna show you guys next. So what about North American settings? So Dr. Worf went back to eventually Harvard and brought it with him. And so still had not quite as good success, but 59% success, which is still better than ETV alone because the success score would tell you it should have only been a 45% success. So it looks like the ETV wasn't working. And so this validated the African results on a North American population. But again, the study was done by the father of the surgery, who's also really good at it. But they were also able between Africa and Harvard, they're able to show that it works in myelomeningocele and congenital communicating. Dandy Walker, those are the success rates. And even a little bit in post-hemorrhagic preemie IVH babies, still terrible, 37%. But that number is better than other publishers, which was like zero to 15%. Okay, well, that's the experts doing it. What about other centers in North America? So Miami is a big producer of literature here too. And they had good results, but not nearly as good as the Harvard experience. You see Miami in red, and Harvard in blue. And so was it just that Dr. Worf is that much better than Miami? This study from Al Qaqqarni looked at it and actually zoomed in and found that all the Boston ones were done with the flexible and all the Miami ones were done with the rigid. So perhaps it's a little bit of a technical issue. Dose response with the CPC. So what I'm going to show you here is the first time the HCRN looked at this, just a small retrospective study. But one thing that's important is that there is a dose response. So the more chorioplexes you burn, the higher your chance of success. And a big number is 90%. If you can burn an estimated 90%, you have an 82% success versus 36, so a pretty big gap. That was a small study that was just barely significant. And so the importance, and you'll just see a CPC playing in the background, but the importance of the flexible scope. So remember how important that 90% number is. With the flexible scope, we were able to show that we could get to that number in 90%, 88% of the time. But those that would use the rigid scope, only 14%, because you really have a hard time getting into those temporal horns. So based on that, the HCRN has developed an objective quantification of, we call them the elf boots, but the, because they kind of look like elf boots. And so after each case, you draw on this diagram to see how much choriplexes you got, and we're able to quantify a little bit more, which is important. So from that, you're seeing the story of ETV. Once the HCRN caught the ETV CPC fever, you can see when it got, when they caught the fever about 2011, it skyrocketed. And to this day, we're doing probably 150 cases. This slide says 100, but, and we've done probably, I think we're close to a thousand ETV CPCs by already. So with that large experience, the next study we wanted to do is just dig a little deeper. Is there some populations that it works better on? And we saw that the two main risk factors, so this is a big eight center study, 192 kids, and showed that age. So the older, the longer you can stall. So these are like the one month old is in red, the one to six is in blue. But if you really can get past that six months, you're going to have a much higher success rate. And then the other major thing was etiology. And so you saw that most of the etiologies actually perform equally. However, this depressing purple line is preemie IVH. And just like Dr. Worf's series, ETV CPC is not very successful. And for the most part, members of the ACRN I'll only speak for, have usually abandoned ETV CPC for preemie IVH. There is a subpopulation. If you can make them older than six months gestation, they are extraordinarily high successful, but that's ultra rare. Okay, but given what we've learned now that we take that same diagram, that's the breakdown of etiologies in North America. We already know ETV works here, but with the addition of CPC, it works great for myelomeningoceles and works on a lot of the congenital communicating and a very small subset of preemie IVH. So it's really expanded our reach in infant hydrocephalus. Okay, more on the clinical side of things. Like what about like from an advanced nurse practitioner or PA or a resident or fellow even, our white whale continues to be in the ACRN is believe it or not, when to treat. And then on the same thing is when to declare an ETV CPC a failure. And so that sounds like it's easy. So classic learning, here we have only large or enlarging ventricles, OFCs going up or the head circumference, you see there's a diagram of a head circumference skyrocketing, bulging or full fontanelle, splayed sutures, or there's a picture there, a baby with sun setting eyes. That's all very straightforward, but that's usually with diagnosis. Now, once you do an ETV CPC, you're changing the fluid dynamics of the brain. And so it's not so easy. So when is it failing? Is a huge question. A very common scenarios that you're gonna see if you work with surgeons that do ETV CPC is that babies are doing great, hitting their milestones, parents are happy, fontanelle soft, but the OFC continues to climb. And so when we pull, some people would say, oh, that's a failure, I'd shunt it. And some people would just sit on it a little bit longer, but how long do we permit the head? How big do we permit the heads to get? It's a big controversy. What about babies who are doing great, but the ventricles remain large? That's less controversial. Usually most people sit on that. Another one is myelomeningoceles. Everything's going great with the head, but they get a lumbar pseudomeningocele. Was it a crappy closure of the back that could, or is it a true hydrocephalus? Only the surgeon that closed it really knows. And then also without, so as an easy is agreement to us as a full fontanelle. And so this is something we go through at our center a lot. So the fontanelle, but we did do a study in the HCRN with Jay Wellens did. And there was when two surgeons felt the same fontanelle, there's pretty substantial agreement as what was full and also an almost perfect agreement what were splayed sutures. So we're not too bad with that, even though sometimes it can be subjective. This is a study out of Vanderbilt. Rob Navtel, I really like this study. It's snuck into the literature and I don't think it gets enough press, but this looks at failing ETV CPCs and three things, re-bulging of the fontanelle, progressive head circumference and enlarging ventricles. And if you have all three, it's almost perfect. It's a failure. If you don't have any of the three, it's perfect, but that's not how life works. It did find that fontanelle is the most important indicator of failure, but if you have two or more, then the chances of it being failure about 95%. And if you have one or less, you're probably pretty secure that to sit on them, that your negative predictor value is 83, that should be a P. But anyway, this is a hot topic in the HCRN. So that led to us before the trial, which I'll talk to you in a bit about, is that we have objective criteria, which is all well and good, but if you read the small print, it's like progressive increase in head circumference, and there's not always agreement what that means. Okay, but with every procedure, you got to remember the other side of the coin. So that's who it works on, but what about other complications? And so this is off that prospective study of 192 patients we looked at. On the left, it's ETV-CPC. On the right is probably the gold standard for ETV alone done by Dr. Drake. And so you're seeing very similar numbers. So again, just about a 3.4% CSF leak, 3.6. Seizure rate, 1.7, 1.4. So with all the extra work in the CPC, it appears that there's no, there's still complications, yes, but in terms of more than the ETV alone, not really. And in fact, we're probably getting a little bit better over time. Okay, the article you should all be aware of is New England Journal, Abko Carney, published two years ago now. So number one, this is a randomized control study in Uganda at CURE. And first of all, they looked at, I think it was 50 patients in each group. And you can see that ETV-CPC did a little bit worse than the shunt, but that wasn't significantly different, but that wasn't the main goal of the study, which you're going to see that what everyone's searching for now are patient-centric outcomes, hence the birth of PCORI, but anyway. So the primary outcome was not failure like it always is. It was cognitive scores in the Bayley's three. And so at one year, which kids seemed to be doing well, better cognitively from an intellectual standpoint of view, and there was no significant difference. Now, obviously they're going to continue to follow this group, but this is just important to show you the direction that the research is going. It's all on neuropsych and quality of life outcomes. So what's the future of ETV-CPC? That, well, the future is now really with the ESTI trial. ESTI stands for the Endoscopic versus Shunt Treatment of Hydrocephalus in Infants. So this is a randomized control study sponsored by the NINDS with lots of money. There's 16 North American centers, all the HCRN centers plus two Floridian centers that have been added. And it's just randomized ETV-CPC versus shunt. And again, you'll notice that the cognitive outcomes is Bayley's at one year, the same as the Ugandan study, but also we have permission to study all these patients till they get to kindergarten and grade one. And we're going to do a full scale IQ and really see if there's a difference. We'll of course monitor treatment failure. There's radiological, there's quality of life, lots of outcomes that are going to be coming out when this study first gets published in about five or six years from now. So it has started and four centers, sorry, six centers have started and we already have four patients enrolled. So the future is now and we're studying it very closely, but that's where we're at right now. That's my last slide. Open one time, I think so. So normally I say any questions, but we're going to wait till after Dr. Chu, I think. Sorry about that, I was muted. So thanks, everybody. My name is Jason Chiu, I'm one of the neurosurgeons here at Children's Hospital Los Angeles. And I'm here to talk about what seems to be an eternal debate between programmable and non-programmable shunts. So I have no disclosures here. And so as we know, hydrocephalus is extremely common, and especially common in the types of patients that we see and treat in pediatric neurosurgery. Shunts still represent the most common form of treatment for hydrocephalus. And while they've helped to extend patients' hydrocephalus, hydrocephalus' overall outcome, they are played as kind of being a double-edged sword as things like shunt malfunction and shunt infection represent two of the most common problems that can occur with shunts. And while not much has been done in terms of changing things like the shunt tubing, a lot of focus has been on, at least a lot of the focus over the last few decades, have looked at how changes in the shunt valve may affect our outcomes in shunting. So just a little brief history, the first kind of modern shunt valve was placed in the mid-50s, and it's known as a Spitz-Holter valve, which is a combination of a neurosurgeon at CHOP, Eugene Spitz, as well as John Holter, who is a machinist working in Philadelphia. And John Holter's son, Casey, was born with a myelomeningocele as well as hydrocephalus. And despite multiple treatments for hydrocephalus, Dr. Spitz wasn't able to get it under control. Because of this, Mr. Holter had kind of worked essentially on a way to help control and design a one-way valve to help his son. And Casey was essentially the second patient to get implanted with this valve, and it actually lasted for quite a while, which was really, really revolutionary at the time. So since then, there have been a lot of kind of innovations in kind of shunt valve designs, and even in the mid-90s, there's over 120 different types of shunt valves. The first programmable valve was introduced in the 1980s, and despite all this variation in valves, they kind of fall into two categories. So one is a fixed or a non-programmable valve, which is designed to maintain either a fixed ventricular pressure or a fixed flow rate through the valve. And then, of course, the second one is an adjustable or programmable valve, where there is a mechanism usually manipulated with an external magnet to help change the resistance within the shunt valve to allow for differences in flow rate. So these are some of the most common programmable valves that we see in pediatric neurosurgery. They're kind of made by all different companies. They come in all kinds of shapes and sizes. So what are some of the proposed advantages of a programmable valve? Well, the first and kind of most obvious one is the ability to change the resistance through the valve and help modulate CSF flow and flow rate through the shunt. And the advantage of that is that if a patient is having symptoms of over or under drainage, they're able to change the flow rate to help address those symptoms without needing to do a surgery to replace the valve. Additionally, we've had a little bit of experience with some patients that require a higher resistance or even a virtual offsetting for the valve. So one of the things we'll kind of touch upon a little bit in a few slides is some patients with tumor-related hydrocephalus and the need for intrathecal chemotherapy, very similar to anomyo. So what are some of the disadvantages of a programmable valve? Well, compared to the non-programmable valves, there's definitely more kind of moving parts to it. And so that kind of brings up an increased risk of failure for the mechanical portions of the valve. Additionally, there's an increase in cost between both the programmable valves compared to the non-programmable valves. And this is kind of a natural average with some of the Medtronic valves. So you can see that the programmable valves can cost almost as much as five times as a non-programmable one. The metal within the programmable valve does cause MRI artifacts. And so this can be somewhat problematic in patients, once again, with something like tumor-related hydrocephalus, where they're receiving very frequent surveillance scans and the artifact may obscure the area that you're looking for. And then lastly, the majority of these programmable valves do require some form of external programmer to help interrogate as well as change any of the settings. And while that's very good for patients that stay at your own institution, if the families do move to a different location or different area and the new neurosurgeon that they set up contact with doesn't have that programmer, that may cause some trouble. So is there any data to suggest that programmable valves are superior to non-programmable valves or vice versa? Well, the very first randomized trial was done actually about over 20 years ago. They actually looked at this. And while it wasn't designed to look at efficacy, what this group had found was that essentially there's no difference between programmable and non-programmable valves for shunt survival at about two years. So they had concluded that these programmable valves are very safe and could be used. Another group had actually taken that data and did a post hoc analysis to actually look at the role of subdural hematomas or hygromas and the role of the programmable valves in terms of treating that. And overall, they had found the incidence of subdural hematomas or subdural hygromas at about 6%. And it was pretty evenly split between the two groups. But what was kind of became very apparent was that in about 58% of the patients with the programmable valves, those subdural hematomas or hygromas were able to be resolved just by changing the valve pressure. And that was something that they proposed as a very big advantage to these programmable valves. But since then, a lot of other studies have come out with essentially clinical equipoise. So this is another big group that has suggested that programmable valves can actually fail quite frequently. In their series, they had nine valves failed. And that came out to be about 11% per year, while there's no non-programmable valves that had failed. Another group had suggested that the programmable valves may actually help prevent proximal shunt obstruction with almost a twofold reduction in risk for the need for shunt revision in the future. And this is actually Dr. Rupert Canberra's study from the ACRN that looked at over 1,000 pediatric patients undergoing initial shunt placement. And in this study here, the ACRN group was able to find any difference in terms of the primary outcome, which is time to first shunt failure between programmable and non-programmable valves. So what's kind of the bottom line? Well, there is a very nice systematic review of literature and evidence-based guidelines that was first put out in JNS in 2014, and actually just recently updated, published this month, which, taking all the evidence together, suggests that there's kind of insufficient data to recommend the use of programmable versus non-programmable valve. And both of them are very viable options. And once again, there's the data that we currently have suggests that one is not superior to the other. So who may benefit from a programmable valve? So patients, I think, with severe, very massive untreated hydrocephalus, like this patient, this first picture here, where they have very, very large ventricles, macrocephaly, and kind of a very, very, very thin cortex, I think, may benefit from a programmable valve, because what you want to try to avoid draining all that CSF very, very quickly, and it may be a kind of stepwise or gradual process where you kind of start a little bit higher and eventually kind of get them down to a reasonable level. Other patients in this kind of second picture here that are either being over-shunted or under-shunted may benefit from a programmable valve, once again, adjusting the valve setting to help address the over-shunting and under-shunting issues. And other patients, like, that we don't see as commonly in pediatric neurosurgery, but those with normal pressure hydrocephalus or idiopathic intracranial hypertension may benefit from programmable valves to help address their symptoms. One of the things that we have been, we've had some experience here at CHLA is tumor-related hydrocephalus in patients that require some form of intrathecal chemotherapy. In those patients, we've placed a programmable valve with a virtual offsetting where the chemotherapy can be delivered through the shunt into the ventricular system, and the valve is turned to the virtual offsetting to allow the chemotherapy to permeate and work. And then once after a certain defined amount of time that the shunt is turned back on to allow for the treatment of hydrocephalus. And finally, you know, are there any other considerations for programmable valves? Well, once again, you know, they are, they do require an external magnetic field to help interrogate and change their settings. So there's been a lot, a lot of other investigators have looked at, well, are there any, are there any other external fields created by everyday objects that we use, toys, headphones, you know, phones, things like VNS magnets that, that may affect it. And the kind of bottom line is that some of these external fields do, can affect the shunts and can affect the programming of the shunts. So the FDA has actually recommended that these shunt or external magnetic fields be kept approximately two inches or five centimeters away from, from the valve. So in conclusion, programmable valves can help non-invasively change CSF shunt flow rates to meet clinical needs. Currently there's insufficient evidence and data to suggest that one type of shunt valve programmable or non-programmable is superior to the other. Certain patients that do require some form of CSF diversion may benefit from the presence of a programmable valve. And then patients with programmable valves should be aware that external magnetic fields may affect the valve setting. Thank you. Hi everyone. My name's Todd Hankinson, and this is just a quick lecture that's really clinically based talking about evaluating patients who have a CSF diversion needs. So we'll just start off with jumping straight into the cases. The first one is an 11-year-old boy who has a myelomeningocele who presents to you in the ED. He's from another state, and in his life, he's never required a shunt revision. The reason he was transferred up is that he's been having a couple of weeks of really bad headaches and ultimately went to his local hospital where they got him a CT scan, and it demonstrated the following. Where they got him a CT scan, and it demonstrated the findings that you see below. The scan from today is on the left, and a baseline scan that they had obtained as part of his usual management is on the right. So the question is what to do. And we'll go straight to poll questions. And since this is a pre-recorded talk, the poll question features a little bit different. So I had my sons and pets jump in and give us the answers here. So there's the poll results. So the next best step for this patient, fix the shunt. Well that is most likely true. This patient does appear to have a shunt problem that's going to need to be fixed, but we can probably get a little bit more information about what's going on with the shunt before we dive right in, and that can lead to a safer and more effective management. So I wouldn't do that as the next step, although it's likely that we will be doing it. B, tap the shunt for infection. This patient's 11 years old. He's probably 10, 10 and a half years out from a shunt revision or shunt operation. He hasn't had any signs or symptoms of infection and so while people have very different thresholds for tapping a shunt for various reasons, tapping a shunt for infection in this context is probably not worthwhile and is more likely to stir up problems if you end up with a contaminant or some problem than it is to give you useful information. Get an MRI of the brain. Generally speaking, in many institutions an MRI is the baseline imaging study that we try to get. Ours is included. However, in this particular patient, the CT scan is pretty high quality. We have a nice comparator CT scan and what we care about really is a ventricular anatomy, which we can see. So I think getting an MRI in this context is probably a little bit superfluous. D, getting an x-ray of the shunt system. Needless to say, the crowd thought this was a good answer and I agree with them. Getting an x-ray gives you a lot of information and should really be part of any evaluation of a potential shunt malfunction. It can tell you, in cases like this, what kind of valve you're dealing with. It can tell you where your catheters are. It can tell you if you may have a break or disconnection of some variety and that can really inform your operative management. So it's really important to include an x-ray of the shunt system in any workup of a potential shunt malfunction. E, send the patient back to the hospital. Well, looks like my nine-year-old was playing a funny little joke there and so we'll just skip on that one. So here's the shunt x-ray and what you may see initially is it's hard to see much and that's often the case in children who are a little bit older who have had their shunts for a long time without recent intervention. What you can sort of see here is it looks like the catheter may be backed out a little bit and if you zoom in what you can see is you don't really see a lot of tubing right here. You can start to see the tubing here. Maybe there is a halo of where a valve or something used to be but it's really unclear and it's tough to tell. So what this says to me is you may be dealing with some kind of disconnection or a catheter back out and that can really impact how you approach things operatively. In this particular patient it was interesting if you look closely you can see on his previous CT scan on the right that there is the tubing visible in the right retrorhicular region but you don't see that tubing and the scan from today. It turned out he did have a disconnection that was fixed and we replaced the position of his ventricular catheter and he did. All right, so pro tips on evaluating a shunt. First, listen to the parents. I didn't expressly mention this in the last case presentation but especially experienced parents can very often let you know what things look like. Does this look like the patient's previous shunt malfunctions? Does it not? If an experienced family is coming in two, three, four times and getting sent back out of the emergency department that should be a sign that you should really be looking closely for something that's going on. It may not be the shunt but it may be and you really need to take that seriously. Use comparative imaging. It is incredibly important to know what a patient's baseline scan looks like. To know if their scan shows increase in ventricular size when they malfunction or if they're one of the less common patients that doesn't have any increase in their ventricular size when they malfunction. Does the patient always have large ventricles or are these ventricles really big? In this day and age where we can transfer images digitally relatively among hospitals we really should be making an effort to always have comparative axial imaging whenever we're evaluating a patient for any kind of shunt or post endoscopy malfunction. I put consider x-rays. I get x-rays religiously every time somebody presents for a malfunction but that doesn't mean that you need to get x-rays for standard clinic follow-ups and things like that. It's just really when we're looking to see if somebody has had a potential problem with her or his shunt. And then I mentioned I'll mention it again in a minute the infection window. So institutions do vary a little bit. At our institution we tend to really think about infection if we're within 12 months of a shunt operation and someone presents with a malfunction. But some places do more like six months. And then logistics. Think about do I need to make this patient NPO? Are there any labs I need to check? Do I need to get that COVID testing done? Those are the types of things that can really make or break your day if you're gonna have to take a patient to the operating room for a shunt problem. You may be able to do it during regular business hours versus have to do it in the middle of the night because the patient's not NPO or you're waiting for a COVID test to come back. So those are the types of things that just help the service run better and a little bit more manageable. So next patient. This one's a little different. A nine-month-old who's an X 30 week premature infant who has post hemorrhagic hydrocephalus. She was shunted six months ago and presents with a couple of weeks of irritability and vomiting. But she has a g-tube and so it's not totally clear what's going on there. Again we have our baseline imaging and our comparator imaging and the comparator imaging does show an increase in ventricular size. So back to the crowd with our poll question. Here we go. What do we do? So 10% said go straight to the shunt. Same as before. We're probably gonna need to do that but we can get a little bit more information. B tap the shunt for infection. Now this patient is different than the last one because she's still well within the infection window. She's a high-risk patient because she's a chronically ill patient with post hemorrhagic hydrocephalus and she just had her shunts on six months ago. So tapping the shunt for her is very much appropriate. C get out the scope. It's time for an ETV. Well you know that there has been a ton of enthusiasm for ETVs. I'm sure Dr. Riva Cameron gave an absolutely riveting lecture about all the awesomeness around ETV. No I'm sure you did but but in this particular case this particular population has been demonstrated to do very poorly after endoscopy procedures. So patients with post hemorrhagic hydrocephalus of prematurity are not even eligible to be enrolled in our current HCRN randomized clinical trial that's comparing ETV CPC versus shunt because we know that they have really poor outcomes. So in somebody with post hemorrhagic hydroendoscopy generally speaking is not something that we offer for CSF diversion. D x-ray the shunt system. As you know I'm a big fan of that and would certainly plan to do that. E just vent the g-tube and ask pediatrics to deal with it. Again my funny little kid getting in there and voting I guess he's got a friend now because it's 2%. So what I would do here is tap the shunt and I would also get the x-rays. So here are the x-rays. In this case everything looks fine. Things are connected. We put the shunt in so we know what kind of valve is there. Following up the shunt tap the wah wah. So this is basically demonstrating an infected shunt. So what do you do then? Well it's time to buckle in and get ready for your journey with this patient because they're gonna be in a hospital for a little while while we get their shunt out, treat their infection, and put the shunt back in. So what are some of the baby pro tips when you're dealing with an infected shunt? Well first if you can try to do your shunt tap before the patient gets antibiotics. This is often a topic that we get from a pediatric service or somebody calling asking if they want us to tap and it's important to try although I don't know how much it really impacts the yield of cultures. When you do see positive bugs consider whether there's contaminants there but generally speaking they're probably gonna be real and you need to take them seriously. When that happens the goal is to get all the shunt hardware out of there and there are really only very rare exceptions of bugs that can infect a shunt and can be treated through without removing the shunt. And again logistics, does the patient need to be NPO, labs, COVID? Coffee break? Anybody have any questions? So this is our third and last case a little bit more sophisticated one. This one is a patient who initially presented at two months of age with a large head circumference and potentially a down gaze palsy. He was making pretty good developmental progress but had kind of a complicated social situation. So you can see on the left here his head circumference growth chart which was quite high and then in the middle you see images and as you can tell from the fact that I'm showing you images over time we did not initially intervene for him. So we just watched him closely his ventricular size never really changed his head circumference growth started to slow down a little bit developmentally he did well. But after following him for almost two years he had an eye exam that showed papilledema. So that was obviously very concerning and I believe we placed an ICP monitor that demonstrated elevated pressure so then we went ahead and moved towards CSF diversion. We had a long conversation with his adoptive parents or his grandparents about whether to go with a shunt or an endoscopy procedure. They were very much in favor of endoscopy so that's what we moved to. So now we can see some examples of that procedure. Ladies and gentlemen so here we go with our ETV video which we're gonna do gamer style. So I stole my son's gamer headset and microphone and let's party. As you can see we have already entered into the third ventricle which in this patient is quite narrow as demonstrated by the MRI scan in the top left corner. We're just using the forceps to penetrate through the floor in front of the mammillary bodies and behind the infundibular recess. As you can see the floor is fairly thin and we're dilating it out a little bit. Resident made a mistake got a little bleeding there but that should be pretty easy to control with just a little bit of irrigation. You can see now we have a nice opening that goes into the prepontine space and we'll dilate that out using the instrument as well as additional instruments which will show next. The next instrument is going to be a neuro balloon that we enter down the same passageway again through that fairly narrow third ventricle. You can already see the basilar artery and the branch points at the top of the basilar and this balloon holds about one milliliter of air or fluid. We're dilating it out to again try to stretch the walls of our ostomy in order to minimize the risk that this will collapse down. This device is called an echo myriad. It's basically a suction device that also has a cutting edge and you can use it to very carefully trim down those edges to try to dilate out that opening which you see here. Now you have a really nice view into the prepontine cistern with the basilar artery and top of the basilar. So he did quite well after that operation and his papilledema resolved. Here you can see scans again across time and what you'll see is that from 2014 to 2019 his ventricles remained the same into late 2019 again they stayed the same but lo and behold he showed up with papilledema again. So this represented functionally a failure of his ETV. So again we had a conversation with his grandparents and what we decided was that the best option would be to do an exploratory endoscopy and if it looked like the ETV was closed that we would reopen it. If it looked like it remained open that we would go ahead and place a shunt and so we were prepared to do either of those things at that operation. So what happened? Boom. So while he ended up with a shunt he's done very well since then. His papilledema has resolved, he continues to make a developmental progress, he hasn't had complications related to his shunt and hopefully he will not knock on wood. So quick check on what to do when evaluating an ETV. It's important to be conscious of the fact that with ETV ventricular size may not change or it may be much more subtle than what we tend to expect with patients who are shunted. So we've got to keep that in mind when we're evaluating those axial images across time. Rates of ETV success can be as high as 50% so it's really important to talk to family because some families are more than willing to take that chance to avoid having hardware. Other families are less comfortable with that idea. So it's really important to communicate with your families and be honest about what you think the ETV success rate is going to be in this particular case. Think about your anatomy. Is it going to be really hard to get into that ventricle? This one was a bit of a challenge for an ETV as we showed in that video. Is the floor of the third ventricle really thick? Is there anything dangerous about the anatomy? And have a plan B. If my ETV doesn't work I'm going to go ahead and go with the shunt. So quick summary. It's important to remember that CSF diversion is a fantastic group of procedures. It saves kids lives. It saves adults lives. It improves quality of life. It's one of the best things that we do despite the fact that it can be a challenge sometimes. Usually speaking if you have some basic principles in your back pocket when you go to evaluate one of these patients you'll be able to get through that evaluation with no trouble. Certainly you'll initially be able to get the ball rolling. Some of the key things to remember are listen to your families. Make sure you look at the patient and have a sense of how sick they are. That'll help you determine how quickly you may need to get to the OR. Get to know their ventricles. Get to know their history. Make sure you have comparison scans. Understand what kind of shunt you're dealing with and if it's infected it's probably going to have to come out. Just have a great time taking care of kids with shunts. Thanks a lot. I hope the meeting goes well and have a great day. Thank you for all our speakers for presenting. We do have some questions. The first question I think is for everyone. In terms of shunt x-rays, if you have a patient who presented to the ED multiple times within a month or so, how often would you repeat that shunt x-ray? For me, it would depend on the context of how they're presenting. So I tend to certainly would get one at that first presentation. But if that's reassuring and the presentation with the repeat visits to the ED is similar or not, you know, suspicious for a fracture, then I wouldn't continue to repeat the x-rays. I agree. If they are really close in time, but I do encourage all of our house staff to always make sure they get those x-rays. But again, if it was like a week later, the chance of a fracture is pretty slow. Yeah, I agree with both Dr. Cameron and Dr. Hank. Another question we had actually is about repeat ETV. And so we wanted to know kind of like everyone's thoughts, starting with Dr. Cameron's thoughts on doing a repeat ETV, like when would you try again? And we've also had some patients who like had a shunt that had been working or non-functional and then they do an ETV like later on in life. So we wanted to know across all the different facilities, thoughts on repeat ETV. Sure. So I had a slide, took it out, but after an ETV CBC, there's pretty good, some really good evidence from Dr. Wharfe that showed that to go, you know, you do the films and if the hole is closed to go back in and then that has about, you can rescue about 25% of your failures, which is nice. And that was both in Africa and in Boston. So that's for the ETV CBC. Re-ETV for older patients, a very great option. The biggest thing is if it's a very early failure, then it's probably, it's not from closure of the stoma. So it won't probably work. And then personally, in terms of a shunted patient and then subsequently doing an ETV, when I finished my fellowship, I came out like gangbusters and started doing those a lot and I had a miserable results. Other people like Jay Wellens has presented some pretty good results, but I haven't. So I don't really offer it. Once you have a shunt, I never convert to an ETV, but see what the other two say. Yeah. I mean, our experience has been, I think, pretty similar to Jay's. There was a paper that, that Paul Clemo led up a year or two ago that looked at ETV after shunts. And I think they, they showed about like a 60% success rate, but I think every single one of the failures, but one was from our hospital. So like we have not had great luck with it. So generally speaking, I'll talk to families about it, but generally, you know, we, we don't do an ETV in a patient who's been shunted previously. And with regard to re-ETV, again, I usually have a conversation with the family and give them an estimate. Usually it's between 25 and 50% that a second look ETV is going to work. And then we talk about whether or not they want to take that, those odds or just go straight to a shunt. Yeah. I think I fall in the same camp as both of you two. And then we do have a question for you, Dr. Chiu, are there any known differences between infection rates in programmable versus non-programmable valves or does it have no effect? Yeah, that's a, that's a good question too. And you know, for my combing through the literature, I have not been able to find any, any difference in terms of the two. I don't know if Dr. Ever-Cameron or Dr. Hankinson has, you know, any anecdotal experience about that or not. Oh, like Tamara has that, Simon has a big study on infection and valve type fell out very quickly as a risk factor. And then I think we have one last question. I think some people wanted to know about the recruitment for the ESTHI trial. So some patients or some parents do come into clinic wanting to talk that they want to have like an ETV for their child. Like how do you talk to these families, like objectively about both ETV and shunt in order to enroll them in this study? Yeah, it's a, it's really tricky because like sometimes people have pre, you know, with Facebook, Instagram, all that, a lot of people come and want, want me to do the ETV CBC. And so like it's, or, or rarely the other way, the shunt, oh, I only want a shunt. And so we thought long and hard about that in the HDRN. And so one way we've dealt with that is the patients come in and then if they are a candidate for both, then we send a PowerPoint to this expert expertise panel and of which the three of us are on. Plus I think, I don't know, I forget, I think it's like 30 and then you have 24 hours to answer and then you get all these experts from around North America that say that patient's a good candidate for both procedures and you can present that to the family and say there's equipoise or there's an evenness. And so all of North America thinks either one's a good choice and then the parents can decide to go in the trial or not. But it still will be a problem because people sometimes seek us specifically out because we do offer that procedure, you know, it's one of the reasons Boston is not in the trial. Yeah, I mean, I think you have to, you know, part of being an enrolling site for a randomized trial like SC is that you have to honestly believe in your heart that there's equipoise regarding which procedure a patient should have. And my experience with both with this and another context with enrolling patients is that when you explain that to families, you say, look, we really don't know and everybody has their own biases one way or another, but truly we don't know and that's why we have to do this study. And some families will accept that and some will not. And that's fine. It's their prerogative. But that's kind of the way that we try to approach it at least. And we're going to be measuring that like those that refuse. We still can collect a little bit of data on because that will be a big question is are we only attracting rich people or a certain segment of population and we'll adjust slightly for that or report on that anyway. I think we have time for one more question. So someone wrote, how often would you repeat a baseline scan? So for example, if the shot was placed in infancy and your scan is a simple post-op scan, when would you repeat the scan for it to be a true baseline scan? Jason, you want to take that? Yeah, I mean, I would probably get kind of some interval pictures in a couple months after the immediate post-op. But I mean, I think usually about a year or so after the shot has been placed, I would consider it a little bit more of a baseline. Yeah, for me, I always do it like two years, you know, your brain is 80% grown in two years. So two years is a big landmark. So when they turn two, I like to do a picture. And I try not to do one within the first six months as their ventricles are changing. But that's my practice. But everyone's different. Todd, what do you think? Yeah, I think there's a ton of variability. And Jay, you may remember this better than I do. I think that Jerry Oaks and Shane maybe did a study where they surveyed experts to see what their regimens were, and they were just all over the map. And this was with regard to follow up in general, and also imaging. So I tend to image a lot less as kids get older, but we'll still usually do something at about a month after surgery, and then in a little kid, maybe six, and then once a year for a couple of years, and then I go out to multiple year intervals. And, you know, I think that Kurt Rosell also wrote a paper that showed that it's pretty rare that you pick up a malfunction in a totally asymptomatic patient just at a clinic visit with a scan. So I think that we can probably, at least me, do fewer images than we do. But there's not a standard. It's funny with the advent of the fast MRI, we're a little faster and looser. We order more. But I bet you our imaging is skyrocketing, because we're not as scared of the radiation anymore. Thank you for everyone for participating, we greatly appreciate it. And now we have a short little video. When you're feeling low, and there's no one around, when it looks like it's over, and life's got you down, hold on to me brother, I'll be here when you need, cause there's a brighter tomorrow, this I truly believe, that everything, yeah, everything is gonna be alright, everything, yeah, everything is gonna be alright, cause when you're feeling low, and you can't see the light, everything, yeah, everything is gonna be alright, yeah don't you cry, it'll be alright, but everything, yeah, everything is gonna be alright. a little bit today about neuroimaging of pediatric epilepsy I have no disclosures so I find it a little bit helpful in contextualizing imaging of pediatric epilepsy to understand the overall evaluation of drug-resistant epilepsy in children and where imaging can fit in and by and large I'll be concentrating on the topic of drug-resistant epilepsy and epilepsy surgery which will be one of the major areas of involvement for advanced practice providers in the field of neurosurgery so there's clinical exams that are used to evaluate patients with drug-resistant epilepsy semiology neuropsychologic testing that can help to localize and lateralize and obviously neurophysiologic examinations including EEG and some venues and EEG and then structural imaging including a seizure protocol MRI and functional imaging with either perfusion spec or FAT and FGT PET which are the mainstays of imaging involvement in evaluation of drug-resistant epilepsy and attempting to identify the epileptogenic focus and I understand it in this approach the zonal approach that was described by Luters back in the early 2000s I find this helpful to kind of understand how to integrate all the data streams that come together if you're sitting in an epilepsy surgery conference and you're hearing presentations from lots of different players. So Luters described a number of zones. The epileptogenic zone is the amount of cortex that needs to be resected to achieve seizure freedom and that's really the that's the goal. That's what we want to identify so that surgery can be effective but it's not defined by any single imaging or physiologic test and the extent has to be inferred. The seizure onset zone is the area of cortex that initiates clinical seizures and that's derived from ictal neurophysiologic testing and in some cases ictal SPECT. The epileptogenic lesion is really where imaging is key and that's looking for any macroscopic lesion that is causative of epileptogenic seizures or at least putatively causative of epileptogenic seizures and we'll discuss that a little bit later. The functional deficit zone is the area of cortex that's functionally abnormal during the interictal period and that can be inferred from neuropsychiatric testing or interictal EEG as well as interictal PET or SPECT functional studies. And then the irritative zone is the area of cortex that generates interictal spiking. Interictal neurophysiologic testing is the is the mainstay of evaluating the irritative zone and that's the area that a lot of epileptologists believe most closely correlates with the epileptogenic zone. And ultimately we're using all of these data streams in concert to try and identify what we think the most likely area that's causing seizures and needs to be resected in order to achieve seizure freedom is. The goals of structural MRI and how it fits into that, well one we wanted to identify an epileptogenic lesion in cases of structural generalized or focal or localization related epilepsy. And as a secondary goal in some cases we're looking for sensitive mouse. We're looking to support a diagnosis of localization related epilepsy in children where that's not obvious by EEG. Functional imaging has a different subset of goals and it's often as a confirmatory test. And there are a number of clinical scenarios that can come up but in most cases what we're looking to do is either confirm a suspicion of epileptogenicity based on neurophysiologic evaluation and MRI or to guide intracranial evaluation whether that be grids and strips placed by neurosurgeons or at EEG done by neurosurgeons which is coming up more frequently nowadays. In terms of the optimal functional imaging study for pediatric evaluation of epilepsy in children, these are abstracted data from across the literature and one thing you'll find if you look into the literature on this topic is that it's very heterogeneous and very inconsistent. But this is sort of generalized numbers that I've taken and these are sort of average true that ictal spect generally outperforms interictal PET but similarly and both well perform much better than interictal spect and this is for cases of temporal lobe epilepsy specifically and the literature generally addresses temporal lobe and extratemporal lobe separately. If you look at those same types of numbers for extratemporal lobe epilepsy, what you'll see is that ictal spect and interictal PET, sorry I'm having difficulty with advancing slides here, both perform similarly and well outperform interictal spect. The things to be aware of are that seizures in children more rapidly generalized than in adults and extratemporal seizures more rapidly generalized than temporal lobe epilepsy and both of these features together make it very difficult to get accurate ictal spect and that's the reason that interictal PET is sometimes preferred in many institutions for functional evaluation of pediatric epilepsy patients just because it's very technically difficult to get an ictal spect examination you end up with a peri-ictal or post-ictal spect. Multimodality imaging can be very important and we can use that to take all these data streams and assimilate them in one 3D space so we can take a post-operative CT for example and we can extract the grids and strips data and we can label those and then superimpose that on MR data and functional data so that that's all visualized in one space for the neurosurgical team to help optimize approach. If we look at the pathologies that underlie drug-resistant epilepsy in children, the single largest area of pathology is in malformations of cortical development or cortical malformations and this is in contrast to an adult in which almost half of cases will be related to hippocampal sclerosis. Hippocampal sclerosis is a relatively uncommon etiology in children less than 10% so we really want to concentrate on our imaging on malformations of cortical development and detecting subtle malformations of cortical development and this next section will sort of be about what is it exactly that the team is looking at when they put the MRI up and they're talking about these malformations of cortical development. What should I be seeing is the subtitle of this next section. So malformation of cortical development broadly categorized into abnormalities of neuronal proliferation, migration, and organization and examples of each of these would be cortical dysplasias or megalencephalies, heterotopias and leasencephalies, and other variants of dysplasia and polymicrogyria. And these exist on a spectrum within each category such that some of them are very focal and some of them are more diffuse. And the diffuse variants are very much less likely to be the underlying substrate of a focal epilepsy whereas the top row the focal variants are the ones that we're really looking for as supportive of what might be the substrate for drug-resistant epilepsy that's focal. So the imaging of focal cortical dysplasia type 1 is very subtle. This is an axial fluid sensitive MRI. This is the subcortical white matter in the temporal lobes and the temporal lobe on the left has slightly increased signal compared to the contralateral side. This is the PET and it's regionally decreased metabolism on the PET image. And these are the imaging features of type 1 focal cortical dysplasia which is extremely difficult to identify on imaging. Again a different case of focal cortical dysplasia type 1. This is the fluid sensitive sequence and the subcortical white matter in the right frontal lobe here where my mouse is is asymmetric to the contralateral side. Only apparent on this fluid sensitive sequence. The T1 sequence here for comparison is normal. This is frontobasal right-sided focal cortical dysplasia type 1. You can see the subcortical white matter is slightly abnormal and signal too bright compared to the contralateral side on both of these fluid sensitive sequences. So subtle low bar volume loss subcortical white matter hyperintensity and very regional or broadly distributed hypometabolism are the imaging features of type 1 and what we're looking at when we display in conference. This is in contrast to focal cortical dysplasia type 2 which is very focal. This is a right frontal focal cortical dysplasia. You can see abnormal cortical signal. This is too bright on this flare sequence. And there's focal hypometabolism on the PET sequence. This is brought out by you doing a fused image where you can see that this the normal cortical metabolism is missing from this area here. It would be very easy to miss this if you were just looking at the grayscale images. This is a nice application of fusion of imaging here. This is a variant called a transmantle for cortical dysplasia where you can see abnormal white matter radiating from the ventricular margin out to abnormal cortex. The cortex will be too bright on T1 and too bright on fluid sensitive sequences. The subcortical white matter too dark on T1 too bright on fluid sensitive sequences. Here's another example. This is called a bottom of dysplasia sulcus where right at the bottom of the sulcus there's this sorry bottom of sulcus dysplasia right at the bottom of the sulcus there's this abnormal blurring of the gray white junction and abnormal subcortical white matter hyperintensity. So these are the features that we're looking for and talking about when we look at the MRI for focal cortical dysplasia. And just a point that in unmyelinated brain under the age of two focal cortical dysplasias will look different than they do in the myelinated brain. In this case on the fluid sensitive sequences they're too dark and then this patient ages and now we see it's abnormally bright in the cortex and subcortical white matter. That's the imaging appearance of focal cortical dysplasia type 2. You can also get abnormal hypermyelination early but once the patient myelinates that abnormally early laid down myelin is abnormally bright. So gray matter and white matter signal abnormality and blurring of the gray white junction, asymmetric sulcation and focal rather than diffuse hypometabolism are the features of focal cortical dysplasia type 2 that we're looking at. And just to know that there's a sort of a spectrum of abnormality in these transmantle dysplasias where you can go from focal to more lobar, quadratic or several lobes together or even half the brain can be involved. And that leads into megalencephaly which is usually more frequently a hemi version, hemimegalencephaly where half the brain is involved but can also be dysplastic megalencephaly where the whole brain is involved. You can see that the cortex is too dark so we're looking at the signal intensity in the cortex. The volume of white matter on this affected side is too great. The volume of fluid in the ventricle is too great and the sulcation pattern is abnormal, looks like polymicrogyria. And on functional studies what we'll see is mixtures of abnormal increased and decreased metabolism. This can be very difficult to detect in the early stages so this was a patient that was missed early on, came back with a quadratic dysplasia but looking back you can see that there's subtle decreased signal on the T2 from the first examination. So we're looking at unilateral enlargement, abnormal white matter, abnormal cortication, and variable hypometabolism when we talk about megalencephaly or hemimegalencephaly. Polymicrogyria, what we're looking at on MRI is this over folded appearance to the cortex here. It's sort of nodular or stippled at the interface between the gray matter and white matter. Shows up very nicely on the sagittal T1 sequence out parasagittally because it has a predilection for the perisylvian area and that area shows up nicely on the sagittal. Here's an unmyelinated brain where this technically doesn't have abnormal signal, the cortex is normal signal, but because it's over folded and increased in density and in each voxel it looks dark. So we're looking at this abnormal dark over folded signal here with its stippled interface between gray and white. And it can have very variable metabolism. So the same patient with polymicrogyria at two different time points on a PET had normal metabolism and hypometabolism. So the imaging features of polymicrogyria to review over folded cortex, apparent cortical thickening as that cortex is heaped up on itself, a stippled or nodular gray-white junction, and variable metabolism or perfusion on the functional imaging. Heterotopias come in a variety of forms. They can be periventricular, band, or subcortical nodular heterotopias, but they basically will follow gray matter on all sequences. It's gray matter where it doesn't belong. And these often show up really nicely on diffusion sequences. These will stand out. My diffusion sequence won't display right now for some reason. So they follow gray matter on all sequences, no enhancement, preserve sulcation, and they are isometabolic to overlying cortex and functional evaluation. Least encephaly means smooth brain and it's synonymous with age area package area. Age area meaning no gyration, package area meaning these broad gyrations. And least encephaly just is the combination of those two and it implies an abnormality in migration. So these brains are very diffusely abnormal, often with a gradient, either posterior gradient or anterior gradient of involvement. And these will very rarely be the underlying substrate for stereotyped focal epilepsy. Age area or package area, true cortical thickening, typically with a gradient. So some take-home points for you. Imaging is just one of many complementary investigations in the workup of drug-resistant epilepsy. F18-FDG-PET is the single most useful adjunctive imaging in children. SPECT does play a role, but it's very difficult to get accurate ictal SPECT imaging in children because they have a propensity for extratemporal epilepsy and because they rapidly generalize. And cortical malformations are the most common etiology of drug-resistant epilepsy in children. I'd be happy to take questions during the breakout or by email. Thank you so much for your attention. Hello, I'm Jeff Ogeman from Seattle Children's Hospital and University of Washington, and I want to talk for the next few minutes here about what's new in epilepsy surgery. So since there's a lot new, I will focus first on some of the things we're doing to map out function or plan surgery, some of the different steps, some of which Dr. Wright briefly mentioned, to localize where seizures are coming from as part of our surgical treatment, especially around stereo EEG and other invasive monitoring tools, and then look at some of the ways that we can do treatments more safely, especially around laser ablation, and then finally look at some of the newer palliative stimulation methods that are available. No disclosures, and now many of these stimulation methods have been rapidly changing in what they're approved for. VNS is now down to age four, and the others will focus on age 18 and above, which are their current approval. In epilepsy surgery, we've learned a lot in the past 10, 15 years about who is a candidate, in part because we know from prospective studies, both in America and overseas, that after two drugs that have been attempted to control seizures, you are now medically refractory and surgically amenable lesions should be considered. We probably can be even more aggressive in identifying patients to at least be candidates, because we know that if you present with seizures and a lesion, you actually have a very high chance in the next five to 10 years of becoming intractable, and this is from data that's now about 15 years old, it came out of Penn, but we also know that surgery is beneficial, and we have seen improvements in quality of life that appear quite early after surgery. In children that undergo temporal lobectomy, for instance, quality of life will actually normalize after three years. So the importance of early surgery is clear. There is a cost to surgery, and it is minimizing that that is an important advance. An example is if we look at memory in children who have temporal epilepsy and surgery and compare their change pre- and post-surgery. So what you see is that for story recall, which is a verbal task, you hear a story and you have to remember details of it, those with left-sided surgery do poor after surgery by a degree, but those with right-sided surgery actually do better after surgery, and the reverse is true for visual memory and surgery on the dominant side. So what you see is overall the changes in memory subserved by the operated side are small, sometimes existent, but considerable and measurable improvement in memory subserved by the unoperated side. So these ideas really underscore the importance of surgery, but in fact, that has not taken hold. Now this is now about 10-year-old data, but it's still the case that if you look at the age of seizure onset of people who ultimately undergo surgery, it's all in the first couple years of life, but the duration is actually linked more to your age, which means that if you see an 18-year-old with seizures and a 5-year-old with seizures, it's probably the case that the 18-year-old has had a seizure disorder that started at a similar age, and so there continue to be delays in potentially inoperative surgery, and one reason that you may delay would be an example here. So this is a child who had, a 17-year-old, who had a lesion, wound up being a low-grade tumor, and it was very close to language areas, so we actually mapped them at the time of surgery in both English and Spanish, so we were able to remove the lesion, but that's a big deal, obviously, especially for a child who would be much younger, almost impossible. So one of the methods that we use as a functional MRI, and especially including the structural imaging, and then merging the two, so the ability to use advanced imaging and advanced representation of images allows us to bring a functional MRI, in this case of language, to use that and superimpose it to where the lesion is, and then guide our surgical placement of electrodes. So in this case, you can see in the bottom panel, a representation of the grid array that was placed over the lesion, as you can see here, the functional MRI activation of language, and then the fibers that we can trace using diffusion imaging that connect to that area. So these approaches have gradually evolved to where we can, in fact, get data, not just from when a child is mildly sedated, and then passively move their hand, but actually get information from the resting state. So I want to talk a little bit about resting state imaging, because it really has advanced our ability to offer surgery. So in this case, this was an adult that had a tumor, and you can identify where you think motor cortex is going to be, right at this small green dot here, based on the anatomy, but the anatomy is distorted. Or you can have them do a task and get a functional MRI task, in this case of tapping their fingers. Or you can get the imaging just from them sitting there over about a six to eight minute period and extract from that areas that are connected to each other. What you find is that there are these resting networks where, over about once a minute, the activity oscillates, but it does so in combination with other parts of the brain. There's a motor network that's very reliable, and you can identify those even in the setting of disruptions for pathology. There's also a visual network, and then a so-called default network, which is relevant to epilepsy surgery, because it also reliably includes the posterior hippocampus. So you can start to use these imaging findings to map out both function, in the case of motor, and potentially disruption of the networks. This was an example of a child who should have had bilateral connectivity from the center parietal lobe to both the left and right side, but the lesion on the right side causing seizures had disrupted this normal connectivity. Now, what was really neat about this child is after the surgery, they were seizure-free, but actually also regained the ability to draw in perspective, and then he wound up to become a semi-professional photographer afterwards. So the functional significance of these findings are relevant, and perhaps will take us beyond our conventional understanding of brain function, beyond moving and talking to some of the more sophisticated and life-enjoying skills. We also know that seizure activity can disrupt these networks, and we're beginning to learn about how that influences our localization. So in these more severe cases, like Dr. Wright was showing, where half of the brain is disrupted, you can see that on the disrupted side, these networks are quite smeared, if you will, whereas on the healthy side, you still have the semblance of very focal connections and involvement, but no representation on the impacted side. So there's no bilateral connectivity, and you see that very starkly in the visual cortex. So we've seen that pattern, and we are able to take these imaging studies, in this case a functional MRI study again, superimpose the invasive monitoring, and then allow us to get the information that we get from invasive monitoring and guide surgery. And our traditional ways of doing it are still very valid, if you need to cover a broad area and understand where motor cortex might be relative to your resection. But if we're just trying to survey a variety of different areas, then the discovery or rediscovery more accurately of stereo EEG has really impacted function, and many of your programs are certainly moving towards it, if they haven't already. And the idea of placing depth electrodes, which is what stereo EEG is, is not new. It's been around, especially in Europe, for a very long time, but what is different is with the robotic technologies is you can place many more electrodes in a very reasonable amount of time. And in fact, once you do the setup, the incremental time cost to place an additional electrode is quite small. So much is made of the arm, and it's certainly a useful tool, but it is a fairly straightforward stereotactic instrument in terms of its concept. You define the trajectory, and then the calculations are made, and the arm brought into position that point you towards your target. And then from the imaging, you can avoid important vascular structures. And then at the end of the day is, again, showing the operative setup a little more broadly, and your final result are these images that have the stereo EEG, and then you can do schematic reconstructions as well. So it allows you, as you can see here, to sample many different areas, both sides, with less risk compared to the broader electrode array. We have really taken a hybrid approach where we offer the electrode combination, whether it's stereo EEG, depth of strip electrodes, or both, really depending on the needs of the child and what our evaluation has found. And certainly, we've not hesitated to use both in sequence if our stereo EEG finding shows that we really do need to proceed with the targeted electrode array. And I think it's greatly reduced the barrier to considering depth electrode array-based monitoring in children, in part because the infection rates are quite low. It's not something to take that lightly, as there are incidences of hemorrhage, including potentially symptomatic hemorrhage, any time you're sticking multiple probes into the brain. But it has been well tolerated. And again, I am still a fan of the combination. You can see an example of this post-resection case where we use both depth electrodes and grid electrodes to map the deeper areas of the brain for their seizure onset, but also understand where cortical function is located. Now, for deeper targets, we really had a big advance recently in the use of laser ablation, which we've done in our institution for the past seven years. And this is through just a single twist drill hole. You can place a catheter into a target, even a tumor, and then take the child to the MRI scanner, where you can measure the heat that's delivered. And that's the big advances with the fiber optic technology and the MRI imaging technology. We can actually see that we've delivered adequate treatment to a target. This was really useful in temporal lobe surgery, including in adults. This is the processed image that's created, and you can see on the right panels the calculated area of destruction. That's been validated. And so, in this case, you can do a hippocampal ablation through a single twist drill hole. Now, this has had an efficacy that's close to resective surgery in adults and kids, especially kids with suspected dual pathology or pathology outside of just the hippocampus alone. It may wind up being not as effective, and there are risks involved. We've had a couple instances where the inflammation from the laser ablation gives symptomatic ventriculitis. We've tried to reduce that with steroids. We think the placement of the catheter is important. This laser ablation was not successful, and it probably was not anterior enough or medial enough in its ablation in the hippocampus and cervical area. So the localization is important, and perhaps we need to be more aggressive with doing more than one laser. Although, at some point, your risk will start to approach that of an open surgery. Finally, we've had examples where things really don't go as expected. This is what the kind of source image you see, and you should see the probe. This little dusky black right there was curious and then evolved. We stopped the procedure immediately seeing that and saw there was actually a hemorrhage formed during the ablation. This individual adult was completely asymptomatic from this, but did require additional observation. So laser ablation, new, very good for structures like hypothalamic hematomas and other deep structures. Its efficacy is still being explored as it now has about 5 to 10 years of experience as opposed to the 50-plus years that we have with resective surgery. And then to give another perspective on where we're going with epilepsy treatment, I want to just review a couple of the stimulation options. And so vagus nerve stimulation, which is now FDA approved down to age four, has been around since 1995. Original approval was for age 12, but nevertheless, we were using it at that time off label, no longer off label for the younger kids. Let's talk about the RNS as well, but for the VNS, what we found in several other groups as well is that about a third of the kids had a 50% reduction, a third did not have a significant reduction. But then there was this group that had a very large reduction in seizures. And so it's really trying to get into that group, which will benefit from medication reduction or other benefits. And now the newer devices actually have a cardiac detection feature, which lets the tachycardia that occurs with seizures be found without involvement of the family and then a treatment given. And most importantly, the data is now coming out that VNS treatment actually reduces the risk of sudden death. So the sudden unexplained death of epilepsy that has a rate that actually makes epilepsy a significantly morbid disease or fatal disease in those with chronic epilepsy actually do benefit with about a 40% reduction in SUDEP rates with long-term use of VNS. So that's motivated continued interest in that. In those cases that either aren't VNS candidates or seem to have a very clear target that is not eligible for resection, then responsive stimulation has been used. It's been used in younger ages, but approved for age 18. And that involves placement of electrodes around the targets that you've identified, typically by invasive monitoring with the decision that resection is not justified because of the cognitive risks or reduced risk of seizure freedom. The battery and detector are placed in the skull. And we have seen in adults now, the groups have published very high rates of at least six to 12 month reduction against seizure freedom being quite rare with a minority that have no benefit or increase. So again, something that like a drug has a degree of benefit, but unlike a drug seems to maintain it over the years. And then finally on the horizon is a deep brain stimulation, which was recently approved for adults. And that does not require knowledge about the localization of the seizures, but just like DBS for other purposes, has a lead placed into the subcortical structures and then stimulation accordingly. So we're still very much in the process of learning who benefits and who doesn't. Finally as surgeons, we want to put ourselves out of business. Our epilepsy tissue research is showing that there are mutations in many of the tissue specimens, and these might be targeted with drugs. So we've actually opened a medical trial for those with surgical refractory epilepsy who failed surgery and are having initial very exciting results that Dr. Jay Halpert's work on using a rapamycin liposome that's injected intravenously in these very difficult to control cases. So as we look forward, we're going to continue to work on more precise mapping function, imaging and interoperative technology, and increasing palliative options with the hope that our future research will eventually allow us, spare us from these destructive approaches to the disease. Thank you and look forward to the questions. Thank you, Dr. Wright and Dr. Ojiman. Great presentations. We have a few questions. The first one is for Dr. Wright. We were wondering if you could touch on the benefit of magnetoencephalography in the pre-op evaluation of the epilepsy patient. Can you hear me? Yeah. MEG is used in certain centers, and it really has to do with epileptologist's preference as far as I can tell. Some places really value it and put a lot of stock in it for source localization, and other places prefer doing an EEG or a dense-ray EEG approach. I think they're complementary, and if both are available at an institution, they're frequently used, and it's often institution-dependent and depending on practice in the local environment. Thank you. Do you have any comments, Dr. Ojiman, to add? No. Yeah, sorry. Sorry to get excited there. Yeah, it depends in part, too, on whether you have one. Yeah. And the other big limitation is you can't do ictal studies, so you're really limited to situations where the child can hold still. So we have referred those out on occasion, but have found them to be of limited value compared to the functional studies that Dr. Wright mentioned. Okay. Our next question is for you, Dr. Ojiman. What is your personal cutoff for how long you will leave electrodes in place? We all have our kids who have multiple seizures every day, and magically you implant the electrodes and they stop. How long will you leave them in? Yeah, so there's no absolute cutoff. It's always a combination. Well, most are approved for 28 days, so you will get into unlabeled use beyond that, which is fine. It's really just a combination of how the child's doing psychologically, whether things are changing in a way interictally that will lead you to think that another day will matter, and then how important that seizure is or how many seizures you'll need to make an answer. So if you're going in without a very strong pre-test probability of an area involved, one seizure probably isn't enough to do a highly invasive destructive procedure, whereas if you have a lesion and really just need, for whatever reason, your infirmatory work and your interictal is very focal and consistent day-to-day, then you might have a plan without an ictal study and might stop once the child's starting to get very anxious. So that's dancing around it a little bit. I think my record is somewhere in the six-week range for strip electrodes. It's not the world record, and that was not pleasant for anybody. So I would say after about 14 days, you really are getting a diminishing return. Yeah. Fair enough. And, Dr. Wright, there was a question for you on what are some of the imaging findings that you see post-ictal, and how are those different from what is their baseline anatomy abnormality that you're seeing on the imaging already? How do you differentiate between the two? Yeah. If I understand the question, it has to do with imaging findings that are as a consequence of the seizure as opposed to as a cause, and there are a few that are well-described. There can be abnormal fluid-sensitive sequence signal and diffusion restriction in the cortex, and that can be variable in its location, but there's also often ipsilateral, thalamic, and mesial-temporal signal abnormality both on fluid-sensitive sequences and diffusion sequences. And if you see that thalamic and mesial-temporal distribution, you can often assume that the cortical abnormality is related to seizure effect. That can be transient, and in cases where it's long-standing, the seizure is long-standing, you can see long-term sequela with ipsilateral hippocampal volume loss on that same side. So those are some of the things that you might see just as a consequence of the patient recently having had a seizure. All right. Thank you. And Dr. Oetjeman, if you could just touch a little bit more on laser ablation and if there's any difference in the criteria that you use for epilepsy patients versus when they're used for tumor patients. I would say in general, no. The concept is the same as surgery in that you're making a destructive lesion. I think that you're hampered by the fact that you have no ability to record or tactile feedback or other ways to define a boundary. So that's an important consideration in choosing. It makes a lot of sense conceptually if you're targeting a well-circumscribed deep lesion as opposed to a field-defined superficial lesion. I think you have to be aware that the destructive lesion can disrupt function if it's in a spot and although the heat will in general follow tissue planes, it's not an absolute protection. So gray-white junctions, tumor-non-tumor junctions, CSF all in general deflect heat, but we've certainly seen injury from either direct thermal radiation or from the heat itself where catheters were too close. So in terms of the decision tree, I think we're at the point where we can say that for most conditions, ablation is somewhat less effective than resection. Hypothalamic hamartoma is maybe the exception, I think there, especially those that do large numbers have had outcomes that rival the open approach with lower morbidity, but not zero. But I think for things like hippocampal disease, especially in kids where the concept of isolated hippocampal sclerosis is probably a minority of what we see, whether we realize it at the time or not, I think in those you have to consider it's reasonable to try, but we'll not have the efficacy of an open operation and also don't have hesitation to offer open operations later. That said, the cognitive benefit in non-dominant temporal lobe surgery is also real and needs to be considered in the decision tree as well. So it really just goes back to presenting these options to the families and helping them make an informed decision. All right. I think that's all the questions that we had. Great session. Thank you again. Thank you. Thanks, everybody. So I'm just going to conclude for us at this time. I wanted to thank everyone for participating in our conference. We did get some great poll results. So thank you for everyone looking into that. You're welcome to look at the poll results on the side now that they've closed it. I think one of the greatest parts is seeing where we're all from. We literally have every region of this country, including Hawaii, Canada, well, not this country, North America, I should say, Canada represented. And so it's really exciting to see how many APPs are on this call. I think it's also exciting to see how many of us actually do want to form an APP consortium. And so some of the work that we really need to look at is what platform to try to put that on so that we can try to get all of us on the same page and really start to learn from each other. So we did create a Facebook as part of the preparation for this conference. We've included that name in there. This is just a starting jumping point for people to start to talk. We had a lot of discussion amongst us on should this live in a LinkedIn or where these communications could happen. So feel free to join us in the opening ceremony where they have the tables where you can come sit and have a discussion about this, because we really do want to start a discussion on where does this happen and how do we have this communication. I don't know if any of the rest of my team has anything to add, and I'm so thankful for them all for making this happen, too. Thank you so much, Mandy. And just to echo what Mandy said, on behalf of our entire APP planning team and moderating team, Patty, Mandy, Andrea, Han, and myself, we just want to say thank you so much for participating today, and we look forward to continuing to network with you guys. I will leave my email in the chat, and so if you have any questions or want to reach out to us, please feel free to reach out via email, and I can get you connected with the rest of the team. And we look forward to seeing you guys at the opening reception in just a little bit.

pediatric diffuse brainstem tumors

differentiation

prognosis

treatment planning

surgery

characteristics

shunt series X-rays

evaluation

shunt system

symptoms

conference participants

APP consortium

communication

Facebook group

organizers