This Neuroscience Grand Rounds session, led by Muhammad Osama, MD, provides a pediatric neurosurgeon’s perspective on arteriovenous malformations (AVMs) in children, rare but high-risk vascular lesions that can lead to seizures, neurological deficits, or life-threatening hemorrhage. The presentation reviews the unique features of pediatric AVMs, including their morphology, natural history, and increased lifetime risk, underscoring the importance of early recognition and risk stratification.
Through an evidence-based and multidisciplinary lens, the session explores current management strategies that highlight evolving treatment algorithms and recurrence considerations that shape clinical decision-making in this complex patient population
Learning Objectives:
Describe the incidence of pediatric Arteriovenous Malformations (AVMs) in the pediatric population.
Analyze the morphology of AVM in children.
Identify the various management strategies suggested by the experts.
Review the recurrence rate and treatment algorithm for AVM recurrence.
All right. Well, I guess there's no time like the present, so we will get started. Good morning, everybody. Um, it is my pleasure today to introduce our fellow, Doctor Mohammad Osama. He has been our pediatric neurosurgery fellow for this year. Um, I can't believe the year is drawing to a close already. Um, he earned his medical degree from Dow Medical College in Pakistan, and he completed his postgraduate training in Karachi as well, and then served as an attending neurosurgeon there before joining us here at Phoenix Children's. He's a gold medalist in neurosurgery from the College of Physicians and Surgeons of Pakistan, and his interests include pediatric brain tumor and cerebrovascular pathologies, although I would say he's shown good interest in all kinds of different things in pediatric neurosurgery this year. Um, and I'd just like to say on a personal note, Muhammad, we are very grateful for, uh, the care and thoughtfulness you have brought to taking care of our patients this year. Um, it seems as if you've really tried to understand a great deal of different pathologies in pediatric neurosurgery and incorporate them into your practice, um, and it's been a pleasure to have you. So, we look forward to hearing your talk on brain arteriovenous malformations in children. Thank you. OK. Thank you, Doctor Bristol for a very nice and gentle introduction. So, I am going to talk about brain AVM in children from a pediatric neurosurgeon's perspective. Uh, the objectives today would be to describe the incidence of brain AVM in the pediatric population. Analyze the morphology of ABM in children, identify the various management strategies that are suggested by the experts and how they differ from others in certain circumstances, and review the recurrence rate and treatment algorithm for AVM recurrence. So, I will start from a very basic, broad spectrum slide. Uh, what are actually, uh, What kind of pathologies, uh, are there in Ps in terms of intracranial vascular malformations. So, we have divided them broadly into five categories. Uh, the most, uh, important one to us today is PLAV shunt lesions. That includes both the PLAVMs as well as PLAV. fistulas. We will describe in the subsequent slides what their differences are. The second one are the cave mouths. Third ones are venous malformations that includes developmental venous anomalies, sinus pericrania, and venous dysplasias. Uh, a 4th 1 are telangiectasias, and the 5th 1 are dural aviation lesions. So what are brain aliens? These are arteriovenous shunt lesions. Uh, as I described, these are the subtypes of CNS vascular malformations. Uh, there's a direct connection from arteries to veins with no intervening capillaries in between. And in most of the instances, there is no intervening brain parenchyma in between, as it involves the bile vessels of the brain. Uh, It, the AVMs are particularly surrounded either by CSF in the subarachnoid space or by interstitial fluid in the Worshop uh Robins spaces. So, how does the angio architecture look? Uh, whenever we do an angiogram on a kid who is suspected of having any vascular malformation, uh, the initial thing that we look into is the visibility of the draining vein. In the arterial phase. If we look in the arterial, in the early arterial phase, that there is a draining vein, possibly there may be a PLAV fistula or a PLAVM. How do we differentiate AVM from AV fistula? So, the main differentiating point is the presence of the clustered network of arterialized vein. That is what we call the Nuss, and it is also called as the AV shunt zone. Uh, there may be variable number of feeding arteries in AVM along with variable number of draining veins. But if we compare it to the fistula, number one, there is no niis. There is a single arterialized vein. However, the number of the feeding arteries can be variable. This is how we differentiate the AVM from the AVF. So how ABM forms, uh. As I described that, it is a direct connection from arteries to veins with no intervening capillaries or no intervening functional neural tissue in between. Uh, it's not a fetal lesion. But the recipe for the formation of AVM requires at least two things. One is the developmental vulnerability traits that are determined by the genetic mutations. Either those are sporadic or hereditary. We'll come to that in the subsequent slides. And the second component of the recipe is the environmental angiogenic factors that will trigger angiogenesis. That includes hypoxia. Infection, inflammation, and ischemia. So, if, and both of these components need to be there to form AVM. So, how does it form? It forms in two steps. Step number 1 is angioectasia. For example, if a patient who is having a developmental vulnerability trait in terms of angiogenesis is being hit by any of these angiogenic factors, the first step would, would be angioectasia. It involves the non-sprouting angiogenesis in the capillary bed. There are two types of angiogenesis. One is a sprouting, the other one is non-sprouting. By non-sprouting means that a single capillary lumen is being divided into two. So, it triggers non-sprouting angiogenesis. As we are already having a genetic defect that will trigger the excessive notch force signaling pathway, and that will blur the boundaries in between the arterial and the venous endothelial cells. As we have blurred the boundaries in between the arterial and venous endothelial cells, so these endothelial cells remain lineage undifferentiated. They don't know that they have to become the capillaries. Instead. The the high pressure arterial flow induces the molecular changing that will transform these endothelial cells into cells forming vessels which have slightly larger diameter than that of the capillaries. So that, so, and these are the high flow conduits. So in this way, The Nidus is formed. So, why AVM remodel? Once the AVM is formed, it behaves like a locally aggressive tumor. There are certain factors that need to be considered when we are looking at the remodeling, uh, theory of AVM. So the two important factors are mechanical dilation and the fact that AVM is a high phillo venopathy. So in blood flow from the arteries into the nidus will cause the stretching of the walls that will result in. Remodeling of the AVM and possibly its expansion. The second factor is angiogenesis. The once the AVM is formed, It will trigger the sprouting angiogenesis as well because it will recruit the vessels from the surrounding brain parenchyma, and that is mediated by metalloproteinases and VEGF. And that's why in the, in this modern era, there has been some studies going on on anti-VEGF being a possible treatment of AVM, but there have been no proven benefits of that. Uh, the third important thing is the ischemia and gliosis. The AVM. is the high flow vascular malformation. It drives the blood flow to itself and deprives the blood flow of the surrounding brain parenchyma. This is called as a steel phenomenon. So, that results in ischemia of the surrounding brain, parenchyma, and possibly shrinkage of the brain parenchyma, and that gives the room to expand. Uh The, one of the important things that I want to highlight is that if we keep in consideration the previous slide, this ischemia and gliosis may be a footprint for the development of AVM. If I go to the previous slide, this hypoxia and ischemia is one of the prerequisites for the formation of AVM. So ischemia can lead to AVM formation and AVM can lead to ischemia. So, moving on to how the NGO architecture. Keeps on changing or developing, so there may be a flow-related aneurysm formation in the AVM that occurs because of the higher shear stress on the feeding arteries that results in the formation of aneurysm. The aneurysm may be intranidal, perinatal, or they may be present on the feeding artery. The clinical correlation is that if there is an intidal aneurysm, there is a certainly higher risk of, uh, AVM being, uh, going into pediatric hemorrhage. Uh, the second point is that there is a phenomenon that is called a normal perfusion pressure breakthrough that occurs because the surrounding brain vessels are maximally dilated. There, there is loss of cerebral autoregulation, and if we resect the brain AVM, that will certainly increase the perfusion into the surrounding brain parenchyma, which was previously deprived of the blood flow. That results in increased cerebral edema and possibly there's a chance of developing a post-op hemorrhage as well. Coming to the epidemiology of the pediatric brain AVMs, the prevalence is 14 to 28 per 100,000. The overall incidence is about 1 per 100,000 per year. The overall brain AVM prevalence is 18 per 100,000, and pediatric AVM accounts for 12 to 18% of all the brain AVMs. So, 20% uh 1 1/5 of all the brain AVM belongs to PETs. Uh, there is no gender predilection. Uh, there is no statistically significant racial or ethnic predilection. Uh, 95% of the cases of AVM have got a sporadic. Can of function mutation. As I, uh, as I described in the, uh, AVM formation slide that one of the prerequisites is the developmental vulnerability trait, so it can come either from a sporadic or from the hereditary traits. In a sporadic case, it's a gain of function mutation, while in familial cases, it's a loss of function mutation. Why do children differ from adults? They have got a higher proportion of unruptured AVMs that are detected incidentally, there is a greater lifetime cumulative risk of hemorrhage because due to a longer life expectancy. The simplest equation is the lifetime risk of hemorrhage is equal to 105 minus the age of the patient. So for instance, if a kid is diagnosed at the age of 10 with an AVM, there is a 95% risk of hemorrhage once in life, and the brain plasticity. Allows for more aggressive surgical management. Kids have got amazing brain plasticity. Their functional tissue tends to migrate, so we can tailor our surgical decisions considering this factor, and deep venous drainage is slightly found to be more prevalent in pediatric cases as compared to adults. What is the natural history of ABM and what is the hemorrhagic risk? Uh, the the, the risk of hemorrhage is about 2 to 3% per year. In the Mars study, which I'm going to discuss in the subsequent slide, it's about 1.4% per year for the, uh, overall cohort. If you have got a prior rupture, then there is a 6 to 15% risk of re-bleeding in the first year that comes down. In the subsequent years, and that comes to baseline at the year 5 after hemorrhage. If a kid has got a deep ABM, the risk of hemorrhage may be up to 10% per year. If, God forbid, the kid has got a pediatric hemorrhage, there is a mortality of about 10 to 15% and a morbidity of about 20 to 30%, and that is cumulative. This compounds significantly with each hemorrhagic event in the growing children. So coming to The genetic mutations that are found in the AVM, uh, there is, there needs to be a genetic hit. So, either It is, uh, uh, from a sporadic or from a familial point of view. So we have got the KRAS mutation that is primarily implicated in the sporadic mutations. And in terms of, uh, familial mutation, we have got syndromes like hemorrhagic uh uh hereditary hemorrhagic telangicasia, which have got these mutations that include BRASA1, GDF-2, and ALK1. Uh, this is a paper, uh, from Northwestern. They showed that when they evaluated the kids with pediatric cerebrovascular anomalies and they did genetic testing, their, their hit rate was found to be about 23%. Like they, in about 1/4 of the patients, they were able to detect the genetic mutations. And it determines primarily our screening practices, and it is very helpful in terms of educating our patients. And nowadays, I think insurance is also kind of covering that. I'm not very much familiar with that fact, but Yes. Uh, these are the three main, uh, genetic syndromes that are associated with, uh, pediatric brain AVMs. Hereditary hemorrhagic telangiectasia is the most common one that I want to discuss. Uh, there are 3 known cerebrovascular phenotypes in HHD. One is Nidal type of brain AVM. The other one is PLAV fistula. And the third one are non-shunting microarterial dysplasias. In hereditary hemorrhagic telangiectasias, many of the AVMs are microAVMs like less than 1 centimeter in le size. The animal breeding risk is also slightly less than that of the, uh, general population. It's 0.4 to 1%. What's the screening protocol? Uh, we do the brain MRI screening in all the kids with HHD at diagnosis, and then we repeat it after every 5 or 7 years, and it is accompanied by DSA as well. The two other syndromes are CMAVM syndrome that include the RRASA1 mutation in which there are capillary malformations along with intracranial, as well as the spinal AVMs. Uh, the third one is Weber-Mason syndrome, that, that includes a retinal AVM along with an intracranial AVM. How do brain AVMs present? The most common presentation and the most dreadful presentation, unfortunately, is the pediatric hemorrhage. It can occur in any of the three compartments with intraparenchymal hemorrhage being the most common and subarachnoid and intraventricular hemorrhage being the less common. If a kid comes to us with ALOC in the ED and there is hemorrhage on CT scan, we need to think about AVM versus any vascular tumor, and we need to order a CTA right away. The second most common presentation is seizures. Uh, they may precede the hemorrhage by years. They may be focal or generalized. Uh, and this is, again, something like ischemia or gliosis. Seizures are an angiogenic trigger. They can result in the formation of ABM and AVM can result in development of seizures, so it's, it's the dual-edge thing. Uh. There is a very good seizure-free rate once the AVM is resected. And if the AVM is present in the temporal, uh, region, there is slightly more chance of seizure than in any other region of the brain. The other phenomena that are, uh, uh, chronic in, uh, in terms of AVM presentation includes progressive, uh, neurological, uh, deficits that occur primarily due to, uh, uh, vascular seal phenomena. There are concerns that if a kid is presenting with migrainous headache that are reflected to treatment, we need to investigate them. Uh, the kids can have cognitive decline and developmental delay if they have got an AVM, and that may be a result of venous, uh, encephalopathy, which I'm going to discuss. And there may be a direct mass effect from the large ABM which is slightly uncommon. So, uh, rare clinical syndromes of pediatric brain AVM includes two important things that I want to discuss. One is cerebral proliferative angiopathy. The pathogenesis behind this is The Steel phenomenon, and it works as the same way as moyamoya disease. The brain parenchyma is deprived of blood flow. It will trigger the sprouting angiogenesis, and that will promote the subcortical and transdural angiogenic proliferation. Uh, 88% of the patients presents with seizures, headache, or ischemic stroke symptoms. Unfortunately, the treatment we have available for the cerebral proliferative angiopathy is only palliative. The second one is venous, uh, developmental encephalopathy. What is the mechanism of this formation? So, whenever there is direct flow from arteries to vein, there is chronic venous hypertension, and that decreases the arteriovenous gradient and impairs the cerebral perfusion. That if, if the cerebral perfusion pressure is decreased, the ICP tends to increase and that results in disruption of the brain, blood-brain barrier. Uh, that will cause white matter edema, and clinically, it will cause seizures, developmental delay, encephalopathy, cognitive decline. These are some of the rare clinical syndromes that, that occur if the kid presents uh in the neonatal age. Uh, there may be macrocephaly, there may be cranial bruiton auscultation of the fontanels, and there may be development of congestive heart failure, like the high output cardiac failure if the kid is having AVM. Coming to the investigation, there are 3 major investigations that need to be done in every patient with AVM, and each of them has its own merits and demerits. Uh, the CT scan along with CTA, it is the most readily available investigation to us. It is best for intracranial hemorrhage. It identifies the large AVMs. It can also identify the major feeders and as well as the major draining veins. There, we may be able to identify calcifications that. Drives our thinking process that this kid may be having AVM and it guides. Our urgent management that something needs to be done or not. What are the limitations? We all know that there is radiation exposure. We cannot determine flow dynamics and there is poor visualization of the small AVMs. And one more thing that I want to point out is that we can have a very good idea of the vascular malformation in relationship to the bony, uh, landmarks in, uh, if, if we are doing CT or CTA. The next one is MRI. MRI is the most important study that is used for AVM localization. It defines the nitus relative to the eloquent cortex, and it is also able to detect the prior hemorrhage. It is able to detect the chronic ischemia and steel phenomena, and If the AVM is localized in an eloquent location, we can do the functional MRI and we can determine if there is remapping or functional reorganization. This is one of the most important implications when we are dealing AVM in a loquent location. What are the limitations of MRI? It usually requires sedation in young children. It cannot reliably differentiate terminal fetus versus in massage arteries, and there is limited depiction of the detailed angio architecture. Coming to the most important investigation, uh, it is DSA, digital subtraction angiography. It is almost mandatory before any definitive treatment until unless there is an emergency situation going on. It defines all the feeding arteries. It defines the angio architecture of the nitis. It defines the feeding veins. Uh, it is best for the, for detecting the aneurysms associated with AVM. Whenever we look at the AVM as Dr. Lawton says in his book, we need to think of it as a box. So, what we do as pediatric neurosurgeons, we correlate this box that shows the boundaries of the AVM in Relationship to the DSA, what picture we are looking at the DSA, we try to correlate both of them. Uh, limitations of the DSA are. It's invasive. Uh, there is radiation exposure. There is contrast exposure, and it can be falsely negative in acute hemorrhage. So if we get a negative angiogram in the acute hemorrhage, it doesn't mean that the kid does not have AVM. We need to repeat the PSA at 4 to 6 weeks. So, what kind of angiographic evaluation we get by the DSA? Number one is the Nidal morphology. We need to determine whether it is compact versus diffuse, and we need to distinguish the Nuss from the angiomatous changes. Uh, in my next slide, I'm going to describe how do we differentiate the Nuss if you look at the bottom of this slide. This is the angiomatous change, and this is the nitus. And the differentiating point is that the nitus drains into veins. The angiomatous change never drains into veins. The second is we can, we, we need to determine the feeding arteries and we need to. differ what are the perforating end feeders versus what are the in passage vessels because we cannot sacrifice the in passage vessels while doing surgery. DSA determine what kind of draining veins are they? Are they, are they draining deep? Are they draining superficial? Is there, yeah, any venno restrictive disease or not? And are there any flow-related aneurysms associated? So, what do I mean when I'm saying in passage vessel? In passage vessels means that the vessel is supplying the AVM, but digitally, it is also supply supplying some part of the brain paradigm. It differs from the feeding arteries that are directly supplying the nidus and not supplying any part of the brain panym. So we try to preserve it surgically as much as we can. Now, coming to the presentation, the intracranial hemorrhage is the most critical presentation in pediatric brain AVMs. We have already discussed. It can occur in any of the three compartments, and there are certain numbers that I have already discussed in the previous slide. What are the high risk features that Preclude the development of hemorrhage in the pediatric population. Prior hemorrhage, number one, no, nothing debated. It is the most important risk factor. Deep location of the AVM, deep venous drainage. By deep venous drainage, when I'm calling this a deep venous drainage, I don't mean uh something deep that is. Deep to the surgical cavity. By deep venous drainage, I mean either the subapendymal or, or drainage into the telemostriad vein exclusively. If there is a single draining vein. There are high intraidal pressure associated with intraidal aneurysms. If the AVM is located in the periveventricular location and it has got a small needle size, these are all the factors that increases the risk of hemorrhage. Many of them are independent factors, and some of them are risk modifiers, and that is determined by the Mars study. I'm going to discuss that. So, how do we manage these ruptured brain AVMs? Number one, if the patient comes to us in the ED with a ruptured brain AVM, what do we need to determine if this is neurologically unstable or a neurologically stable clinical situation. If it is unstable, naturally, we have to do something, either decompress or put an EVD in. And in very few of the uh clinical scenarios, we may end up taking out the AVM. Along with the decompression, but that occurs in very limited circumstances. For example, if the AVMitis is superficial, draining, uh, uh, vein is superficial, feeding arteries are easily accessible, we may end up taking out, but in case of AVM surgery. This is one of the most difficult procedures to do and technically challenging procedures to do, so we don't want to go unprepared while encountering those. So in most of the circumstances, we just end up decompressing on maximum, taking out the hematoma. If the patient is neurologically stable without significant mass effect, then there are two thoughts. One is the Helsinki concept. It says that we need to deal with the AVM early, within the 7 days of immediate diagnosis. What advantages does it give? It Gives the the most important advantage is that it eliminates the risk of further hemorrhage. The second is that the hematoma gives us a pathway to reach the AVM. Like it forms a cavity around the AVM. It's even the deep AVMs can become accessible if the hematoma is present. Uh, this is the concept which usually we now employ. And the second concept is the Steinberg principle that says that if the patient is stable, there is no mass effect. We have. Some time to deal with it. So, we delay the definitive treatment by about 6 to 12 weeks. That allows the brain swelling to resolve, that allows the hematoma to liquefy, and cerebral autoregulation to restore. If the cerebral autoregulation is restored, that decreases the chances of normal perfusion pressure breakthrough. So we can employ either of those, like the both of, both of them are comparable. Coming to the other clinical scenario, what happens if a patient with AVMs comes to us in the office rather than in ED? Uh, there are 2 questions that every surgeon asks himself. What, what happens if we observe? The answer is given by the Mars study. And if we decide not to observe, should we treat it surgically or proceed for multi-modal strategy? The answer is given by the Spe Martin scale, uh, supplemented by locking in scale. So, whenever the kid comes to us in the office, some of the pediatric neurosurgery considerations that needs to be taken into account are the blood volume reserves in kids, their potential for rehabilitation, their lifetime cumulative risk of hemorrhage, and the possibility of recurrence. So, I will give the answer to my first question. What happens if we observe? This is, the answer is given by the Mars Study. This is Multi-center Arteriovenous malformation Research Study that is the largest multi-center ABM natural history cohort consisting of about 3000 patients' data. It determines that the annual risk of ICH is 1.4% rather than 2 to 3%. It has identified the factors which I already discussed in the previous slide, which are associated with increased risk of hemorrhage. Some of them are independent, such as increasing age, associated arterial aneurysms, Des, or cerebellar AVM location. Some of them act as a risk amplifier. For example, deep venous drainage was not an independent predictor of pediatric hemorrhage, but it can act as a risk amplifier when it is associated with. Aneurysms, deep AM location, or venous outflow restriction. The size of the AVM, seizure presentation, and gender, they are not independently associated with hemorrhage. But the critical pediatric caveat is that mass history includes the majority of the adult population. Uh, there are, there were a small number of pediatric patients that were that were included in this, in this cohort, so they did not account for the lifetime cumulative risk of hemorrhage. They did not account for higher flow and angiogenic biology and venous hypertensive encephalopathy in kids. So whenever we are making a decision of not going for surgery in any patient, we have to take into account these three factors rather than the overall picture which we usually see in adults. So what happens if we decide to proceed for surgery? This is the famous Epe Martin grading scale. All the neurosurgeons know that it serves as a lexicon or the crystal ball in AVM management. This includes three components size, eloquence, and venous drainage, and it has got a grading from 1 to 5. But Doctor Lawton. has described an other scale because this is something that is determining our surgical risk, and surgical risk should not depend only on these three factors. So, he describes more, more of the three factors that should be taken into consideration while determining the surgical risk, age. If the AVM has ruptured previously or not, or if the Niss is compact, it has got a very nice clear margins versus if it has got a very lacy network and it's diffused and it's difficult to take it out. So, in the neurosurgery practice, we usually combine both of them and then determine our surgical risk. So this is a, uh, a table that I copied from Dr. Mike Lawton's book in which he described the four possible scenarios. Two of them are easy. Like, if the patient is having low Spesler-Martin scale and low Lawton-Yung scale, by low, I mean Speer Martin 1 to 3, and by low Lawton and Young, I mean Lawton Young classification 1 to 3. So, if both of them are low, The decision is surgical management. 85% of the patients either improved or unchanged, but 15% of them were worse or or died. If there is higherpecier Martin and higher supplementary grades, the answer is not going for surgery. But what happens in the mismatched risk protection? For example, if the patient has got a low Speer Martin grading system, but he has got a higher supplementary lot in Young classification. 41% of the patients either become worse or they died after surgery, and this number is higher than the Speler Martin grade 4 in which we usually don't go for direct surgery. If, if we, if we go for direct surgery in a Speci Martin grade 4, the rate of surgical mor mortality and morbidity is about 35%. But if we consider this scenario and we are going for in. In this clinical scenario when there is low Sperla martin but high Lotton-Young classification, then there is a 41% rate of surgical mortality and morbidity. In contrast, if there is a higher Speror Martin rate and a low Lawton-Young rate, the surgical mortality and morbidity is 29%. This is less than dealing with a grade 4 or grade 5 AVM. Surgically because the combined mortality and morbidity of grade 4 and grade 5 AMM is about 35 to 40%. And in this case, it's 29%. So, this table, the, the, the, the bottom line is that the Lawton Young classification is very much helpful. In determining the surgical risk, and it should not only be the Special Martin classification. These are a few of the other skills that I have put in my presentation that are important in determining what are the outcomes of SRS. We use the Virginia radiosurgery skills and One of the scales or classification is Heidelberg classification that takes into account the genetic component and the clinical utilities that it helps to guide us about the genetic counseling. It impacts the family screening and surveillance protocol. I will not go into the details of this. So, once we have decided we need to proceed surgically, either we need to do surgery alone. Or we need to take into account the other uh treatment strategies as well. So this is again, a table that I have copied from Doctor Lawton's book. In his experience of 1000 patients dealing with brain AVM, 41% of the patients were dealt with a dual strategy, embolization along with microsurgery, and they got some amazing results if we compare those results to the older series. What does embolization do? It has got a great role in ABM management. If we embolize the feeding. Arteries or the initial aneurysm, it greatly reduces the blood loss. It reduces the operative time. It reduces the surgical morbidity, and we are very much able to identify the perforating arteries because the embolic material is there within the perforating arteries when we are dealing with these cases surgically. Only 8% of the patients underwent microsurgery alone, and the majority of the cohort of those patients included the ones that presented to the emergency department. The second scenario was the patients who had superficial arterial feeders, superficial venous drainage, like they were very much easy to access surgically, and the third clinical scenario was there was a poor endovascular excess. For example, if there's a distal vascular territory or if the parent arteries are perpendicular to the feeding once there is a poor endovascular excess, so those patients were dealt with microsurgery alone. Radiosurgery, 14% of the patients underwent radiosurgery alone, and these are the patients whom we were not able to deal with surgically. So, the bottom line of this table is that multimodal treatment always favors and benefits the patient and it should be considered in every case. And then unless there is some other clinical scenario that is precluding those uh multimodal treatments. This is kind of a summary of the different treatment modality outcomes. I want to highlight the fact that embolization plus microsurgery has got an angiographic arbitration rate of 95 to 100%, with the neurological morbidity of about maximum 12% and mortality of less than 2%. So, when, when dealing with the uh pediatric brain AVM we have to take into account this factor. This is the most important one, that is the kids have got amazing brain plasticity. Their functional tissue tends to migrate. Language function can migrate to contralateral hemispheres, and in one of the series they documented migration of up to 40% in the contralateral hemispheres. Motor cortex Gets reorganized to the surrounding perillesional areas. However, the visual cortex is least likely to migrate. The clinical implication is that we have to evaluate the patient who have got an AVM in the eloquent location with the uh functional MRI. Uh, children may tolerate resections much better than adults. If there are post-operative deficits, there is certainly a higher chance of improvement. With intensive rehabilitation. And the earlier we do the surgery, the maximum plasticity benefit we can get. So this is just a summary of what kind of management strategies we have discussed. I will not go into the details of those. So, we'll highlight the merits and demerits in particular KVS regarding different treatment strategy, surgical treatment. What is, what are the principles of AVM surgery? The AVM must be removed completely. If you have left a part of it, the risk of bleeding is almost the same as that of an untouched AVM or non-treated AVM. The feeding arteries should be controlled before the draining veins. We need to preserve the draining veins until the end. We need to identify the endassage arteries and try to preserve them as much as possible. And considering the phenomena of normal perfusion pressure breakthrough, we need to define our map pool. Post-operatively. What are the perio and post-operative, uh, considerations? Number one, blood pressure management. Uh, the goal is normal tension and evolemia during surgery. For large and high-flow Avians, there are slightly lower blood pressure goals post-operatively. Oh, I'm sorry, uh, to prevent the normal profusion pressure breakthrough. The second is in the pediatric population, we should be well aware of the fact that they've got a very limited blood volume. We cannot afford to lose. A lot loss of even one quarter of the blood volume can result in hemorrhagic shock. We need to confirm our surgical resection by postoperative angiography. Some of the centers do it right away after surgery. Some of the centers do it within 24 to 48 hours. And if there is residual, the kid may need to go to OR again. What are the complications that can occur in AVM surgery? The most important and the most dreadful complication is, again, intraoperative hemorrhage, and the main reason is that you go into the nitis. You don't move, hover around the Nis, but instead you go inside it. And once it tends to bleed, it is catastrophic. So, If you are unable to control, we can apply a temporary clip on the feeding vessels and reassessess the anatomy once the nitis is kind of cooled down. Uh, there is a risk of normal profusion pressure breakthrough. I've already gone through the details of this. That you may end up taking a vessel that was supplying an important or eloquent brain parenchyma. That is why it is very important to preserve the in-passage vessels. The kid may end up getting the neurological deficit either from the brain manipulation, either from the postoperative brain edema, or either from the inadvertent vessel occlusion. Incomplete resection is one of the most important complications that we have to take into account. We have discussed that, and the recurrence rate. In children. So, coming to the recurrence, there is a phenomenon that is called as rays, delayed recurrence after angiographically confirmed surgical elimination. Why does it occur? Because of the mist or dormant remnants or de novo AVM formation in the penumbral region. The prerequisites of reformation of AVMs are the same, and they will Be present forever in the patient with AVM. We cannot eliminate those prerequisites. So we have to be very vigilant in terms of uh surveillance angiography once we are done with the surgery. When does it present? The vast majority of the patients recur within 2 years. 50% of the patients present with hemorrhage, and the prevalence of this trace is about 5%. And what are the risk factors? In kids, due to young age, they have got immature vessels, and whenever they are undergoing puberty, there are hormonal changes that serves as the angiogenic driving factors that results in the recurrence of AVM. And if they have got a deep venous drainage, that has got a greater chance of recurrence. And the third factor is that if, uh, the kid has got a previous pediatric hemorrhage, there is always a greater chance of recurrence in comparison to a patient who did not get a hemorrhage. So there are different protocols that are used to, uh, uh. That are used for surveillance in uh terms of uh the AVM recurrence. Uh, Cincinnati protocol says. Immediate postoperative angiogram, 3 months. MRA and DSA, then 6 months and 1 yearly MRA and DSA, and then an annual MRA and then graduation DSA should be done at 20th birthday. But Dr. Lawton recommends that the follow-up should be done until 25 years. He recommends DSA every 5 years until 25 years. Uh, Glasgow protocol says DSA every 5 years until 20 years. Theory protocol says DSA at 1 year and 5 years. So these are different protocols which have got a comparable, uh, outcome, but the most common one that needs to be followed is Cincinnati plus locking protocols. What are the treatment alternatives we have? We already discussed that embolization serves as a great management strategy in pediatric brain AVM management. Whenever we are embolizing any AVM, we need to think what goal we need to achieve. Is it a surgical uh is it an AVM cure that we need to achieve versus we need to embolilize only the associated aneurysms? Is the embolization being done to cut down the, uh, the power of blood loss, or it is done. Pre-radiosurgery to reduce the volume of the AVM. So these are different goals. These vary. If we determine them. Regarding different clinical scenarios, so in most of the cases, the embolization is used to reduce the blood loss, to embolize the associated aneurysms or the feeding vessels. This is a Buffalo score that they. Uh, that, uh, the people at University of Buffalo, uh, uh, classified, and it was used to stratify the risk of endovascular treatment of AVM regardless of whether the intent was complete obliteration or the intent. was to reduce the blood loss. So there are 3 factors that they took into consideration number of the arterial particles, diameter of the arterial particles, and the Niss location. The higher the score is, the more risk it is associated with the endovascular treatment. Coming to the third management strategy, stereotactic radiosurgery, it serves as a stand-alone treatment in many of the circumstances when the AVMs are surgically inaccessible. There is a high risk in terms of the, uh, the, the general condition of the patient if the lesions are small, less than 3 centimeters, and the minimum age at which the SRS can be done is 3 years. So, SRS efficacy in the children, it largely depends upon the dose which we are administrating. 1818 grays of uh SRAS, uh, at least 18 grays of SRS are required for, uh, uh, the lesion to be cured, and the effect is plateau at 25 degrees. For example, if we have got a kid who has got a deep AVM, size is greater than 3 centimeters. Uh, and it is kind of surgically inaccessible, so the cure rate is low as compared to the ones who are having a reduced needle size less than 3 centimeters. But, however, this can also be treated via volume fractionated SRS rather than dose fractionated. We divide the AVM into different compartments and then we do SRS on one of the compartments. One session and the other compartment in the other session, but we don't reduce the dose and it has got obliteration rates, uh, like 80% obliteration rates at 36 months. Uh, what are the shortcomings of SRS? Uh, one of them is the latency period. It takes about on an average at least 3 years to get rid of the AVM if we do SRS alone. So for those 3. Years, the kid has got the same risk of hemorrhage as compared to an untreated AVM. There is a risk of radiation exposure in the kids and that can lead to cognitive decline, radiation-induced brain surgery. And there is always a risk of incomplete obliteration that is associated with any of the treatment strategies. So in many of the circumstances, the, uh, if there is a large AVM, we need SRS to shrink it down and then take it out surgically. These are some of the special scenarios that includes the thalamic, uh, how to deal with the, uh, brain stem AVMs or the deep-seated AVMs. For example, if we discuss basal ganglia or thalamic AVMs, these are surgically eloquent locations. We can't go surgically there. So, SRS is considered as the first line of management, and that may be augmented by embolization. Surgery is only considered for life-threatening hemorrhagic evacuation. If we consider brain stem AVM, again, this is the area which a surgeon fears to go in, or this is a no go area for the surgeon. So either we go for conservative management or either we go for SRS or SRS plus embolization. If we are dealing with a giant AVM, we need to employ a multimodal strategy so we can embolize first, then go for SRS, and then go for surgery for the residual AVM. This has already been discussed. So, coming to the future, there has been some uh research going on regarding medical treatment of AVM, but as I said initially in my discussion that there has been no proven benefit of any medical therapy in AVM management. There has been research going on regarding Avastin or regarding MAC inhibitors, but There has been no proven benefit, so our focus should be those three lines of management rather than any medical management for you. So take-home points are PLAVM is not a fetal lesion. Pediatric AVM warrants treatment due to lifetime cumulative risk of hemorrhage, except for certain scenarios. Pediatric brain AVM requires a multidisciplinary approach, and multi-modal therapy is better. If Spetzler Martin plus Low and Young is better than a Spetzler Martin alone in terms of determining the surgical risk. Recurrence after treatment in pediatric brain uh brain AVM is not, not uncommon. So, it mandates a long-term follow-up, and the way we deal with the recurrent AVM is almost the same as we deal with the virgin AVM. There has been no proven rule of any medical therapy for brain AVMs. So I have got some cases to discuss as we have got a few minutes left. So, this one is a six year old right-handed female that came to us with hereditary hemorrhagic telangiectasia. Uh, she was, uh, diagnosed because she came with shortness of breath, and the workup revealed multiple pulmonary AVF. Then her brain MRI was done that determined, uh, multiple brain AVMs. There was no history of headache, no history of seizures, or any of the other complaints. So, This was the MRI that showed a superficial draining vein with this, with, with anitis that is located almost in the eloquent region of the somatosensory cortex and the motor cortex, and there was a small intranatal aneurysm associated with it. We'll see this aneurysm much in detail when we go to the DSA of this patient. So, the questions which a surgeon needs, needs to ask himself, what happens if you observe? This is a six year old kid. Lifetime cumulative risk of hemorrhage is 99% if I consider that equation 105 minus H. So, is observation. Uh An option, yes, it is, but is it like something very much uh into patient benefit? I don't think so. So the next question, if we decide to proceed surgically, what is the risk? The risk is determined by the Spetzler Martin and Lawton-Young scale. In this case, it was Spetzler Martin 2 and Lawton-Young 1. This is a patient with HHD. Any specific considerations that may be taken into account while dealing with patients with HSD who are having brain AVM. So, we need to uh deal with macro AVM, the ones that are greater than 1 centimeter in size right then and there. Regarding dealing with the microwavVM, if we can, if it's easily accessible in the same surgical. Session, then we can deal with that. Otherwise, we'll just either keep on sur uh surveillance or uh we may send the patient for SRS for dealing with microwavia. So, this is the box that Doctor, that, that I have copied from Doctor Lawton's book that uh shows the cubic structure and the usual vessels that are employed in these kind of frontal AVMs. So, if you look at these pictures, uh, we have, we see that there is the supply from the middle cerebral artery. The supply is the, the, the feeders are usually superficial and can be easily taken out. This is the 3D reconstruction which we made, uh, after the DSA. So This is the DSA picture that shows supplies, again, supply from the middle cerebral artery. It shows a draining vein, so. We ended taking out the AVM surgically. We did not do any preoperative embolization because of the fact that it was superficial, and, uh, the feeding arteries were superficial. The draining vein was also, it was draining into the superior serittal sinus, so it was superficial. So we just ended taking it out surgically, and the kid did well post-op. One of the scenarios, this is a case of a nine year old who presented with bithalamic AVM with a progressive movement disorder. What do we do with that? Is there an option for observation? This kid is symptomatic, so observation may not be an option. Is microsurgery an option? For bialamic AVM, microsurgery is the kind of no go area. So what are the treatment alternatives? In this case, this, uh, they did SRS. This is not our case. I've, uh, got it from another center. So, they did SRS and the kid did fairly well after 3 years of surgery. Uh, the movement disorder was fairly well controlled. Another case, 16 year old right-handed female presented to Edie with this intraventricular hemorrhage. She had diminishing level of consciousness after the onset of severe headache. Number one, is it an emergency? Yes, it is. Do we need to do something? Yes, we need to get CTA done. I'm following the same rubric which I have tried to develop, uh, in my presentation. This patient needs CTA. This, does this patient need decompression with EVD placement? For sure, yes. So, We ended up doing decompression. decompressive hemicranniectomy along with EVD placement. The second question, does the AVM need to be treated right then and there versus we can delay this. So, in the in the CTA it was shown, I, I, I, I'm, I, I'm sorry, I did not put the picture of CTA. I've got the picture of MRI. It is a deep AVM. Dealing with a deep AVM right then and there in the emergency situation may not be a wise idea, so employing the Steinberg principle. It was a very wise idea in this case. They allow the AVM, uh, sorry, they allow the brain to cool down. So, during this time period of hemorrhage, till the definitive pre-treatment is completed, there must be some temporizing measures. So, we embolized the feeding arteries. We embolize the associated aneurysm with the ABM and The subsequent surgery was done for the small residual that was left after the embolization. So, this kid did fairly well and like the, the scenario in which he, in which he came versus in which he is right now is like totally 180 degrees apart. This is the final case. I want to present. It's a 2 year old female with a history of hereditary hemorrhagic telangiectasia, having, having non-galenic choroidal AV fistulas, for which she underwent embolization. On workup, it was found that she is having a right-sided parietotemporal lanal type of AVM. So this is the picture of the avian. Number 1 question, what happens if you observe? Uh, this patient has got HHD. She is 2 years old, so observation may not be an option or maybe a less considerable option. If you decide to proceed surgically, what is the risk? This is a superficial AVM. So, the, the, the risk in terms of Sperel Martin and Totten and classification was low. So we decide to proceed surgically. Is there any role of embolization before surgery? For sure, there is. Uh, In this case, we ended up embolizing the feeding vessels as well as there was an aneurysm that was associated with the vein, so we ended up embolizing that. So there was certainly a role of embolization in this case as compared to in the case of six year old which I described previously. So this is again the rubric, the cube, the, the, the cube that Doctor Lawton describes in his book if we are considering uh a frontal parietal labium. So, whenever we are dealing with A any AVM, we have to take into consideration at which location this is located, and there is a certain blood supply and, and draining vein for every AVM that has been defined. So, that helps to determine. The surgical planning very efficiently. So, thank you. I'm open to any questions. This is my email address and I'm, I, I especially want to thank Doctor Todd A Russo for helping me out in making this presentation and Doctor Bristol, Doctor Hoffman, Doctor Veronica, Doctor Shaffron, and my faculty, Doctor Chiarelli for like keeping me here and letting me learn throughout the year. So, thank you all. I'm open to discussion or any questions you have. Doctor Abruzzo, Doctor Hoffman, do you have any comments? Yes, can you hear me, Mohammed? Yeah, yes, sir. Very good. Excellent, very comprehensive overview. You covered every aspect of cerebral arteriovenous malformations in the pediatric population. Uh, you've made excellent points, you know, concerning surgical management, um, and, uh, there, there really isn't a, a corner that you left, uh, uncovered. Um. I did wanna say something though. Remember we talked uh recently with Doctor Hauptmann about concepts and Sue, right? That, uh, it's not an all or none thing, right? So we're not just trying to turn off seizures, right? If we have a seizure reduction, that's still beneficial to the patient and the concept uh of an all or none, either we eliminate the entire AVM. Uh, and the bleeding risk goes to zero versus, um, a reduction in the bleeding risk with over time with radiosurgery. So our old concepts were, uh, based on limited data sets that when we treated patients with radiosurgery and we're sort of in the latency interval, uh, as the AVM is obliterating, that there's no reduced bleeding. Risk or bleeding frequency in that latency interval, but I think that the accumulation of data over time actually shows that there is a reduction in the bleeding risk over time, uh, slowly, right? So there's an incremental reduction. I think some of the more recent data shows that, right? That, um, before you achieve total AVM obliteration. There is um a uh reduced bleeding risk during that latency interval. And, um, but, uh, it, it's, it's not a widely accepted concept that I, I'm not sure that everyone sort of has embraced that yet, but, but the data is showing that. OK. Yeah, I, I, I know we're almost at time here. Mohammed, beautiful job, I think, doing a sweeping overview of the data. Um, I just have one question, you know, when you think about how you're gonna use this in your practice, and a lot of people, and, you know, I sort of, a little bit tongue in cheek, say this, but a lot of people like to use the Spetzler, and then the Lawton Young classifications, and then apply them to their own practice and experience. And I think that the challenge there is that if you look at Stetzler's original series with Neil Martin, and then the updated classification with Lawton and and Jacob Young, um, These are single surgeon. Large volume series, single surgeon with a, with a, with a very unique clinical experience, and I, where I struggle a little bit is how do you apply that. You're gonna be in practice for just a few years, and you're gonna see some of these challenging AVMs that may or may not require resection. How do you sort of adjudicate that with the family, when you're using his classification and your clinical experience? I think about this with my own clinical experience too. What are your thoughts there? So, um, when I'm dealing with a pediatric brain AVM, the most important concentration is the age factor, right? So, the cumulative risk of hemorrhage that should be discussed with family, right then and there. And what, uh, Doctor Lawton described in his, uh, supplementary classification is kind of, uh, Uh, like, it, it, it does make sense. Like, if the kid is having a prior hemorrhage, there is a greater risk of hemorrhage again. Secondarily, if the nitus is compact, like, it defines our surgical margins very efficiently in comparison if the nitus is diffused. So that all of these factors can help me make the clinical decision. Regarding the fact that is, is it a high-risk patient versus is it a low, low-risk patient? What kind of help do I need from my neurointerventional colleagues if I need to send the patient to SRS or not? So, like, it, it, it should be a rubric that is dependent on not only on, only on these classification but also other factors when I am dealing with the pediatric brain MA. Thanks, Mohammed. Uh, thanks everybody. That's terrific, Muhammed, and we got a lot of Um, Great comments here. Just one question from Doctor Agapan. Uh, how early can an elective resection be done? I'm assuming she means following a rupture event. Yeah, we try to do it as early as we can as the workup is complete. So previously, what Doctor uh Abruzzo told me, uh, uh, when Doctor Dave Edelson was here as the CP followed that his time of principle of letting the brain cool down. But since Doctor Hoffman's concept is like mimicking the one with Doctor Lawton, so we usually, we can end up taking it out as early as within a week. All right, thank you, Mohammed. Everyone have a terrific week. Thank you. Bye.
