Hi, everybody. I'm Michael Scott. I'm the Emeritus Neurosurgeon-in-Chief at Boston Children's Hospital, and I'm a professor of neurosurgery at Harvard Medical School. On this webinar about cavernous malformations, I'm going to concentrate on the pediatric cavernous malformations, and I'm going to talk to you a little bit today about their natural history. And at the end of this talk, I hope you'll come away with the impression and maybe the strong clinical indication that many of these lesions, particularly those that are in eloquent areas, should be observed for a bit to find out a little bit about their natural history to determine how they should be managed. I have no disclosures. Now, for the residents who are participating in this webinar, let me just say a little bit about how I got interested in this topic. I finished my chief residency in 1973, and in an era when we had no scanning, no CT scans, and of course, no MRI. In 1973, we had a patient come along with who presented with nausea, vomiting, headache, and a visual field cut, and I was quite sure as the chief resident managing his case that he had an AVM in the occipital lobe, and our way of managing those patients in those days was to do an arteriogram. Here's the arteriogram on that patient in 1973, and it was negative. There was no arteriovenous malformation. The CT scan had just been installed at the Mass General when I was chief resident, and although it was largely being used for research when it was first put in, I was able to get this patient scanned, and here was the scan. It showed this hemorrhage, which you can see here in the parietal occipital lobe on the right side, and you can see some of it's ruptured into the ventricle. This is a very primitive scan. Of course, it's an ME scan that was done in 1973 with huge pixels, but at any rate, it showed us where the problem was. I operated on the patient and found a cavernous malformation. Throughout my entire residency, I had never operated on a cavernous malformation or recognized that this in a busy neurosurgical program, and having this new technology opened our eyes to the presence of this new pathologic entity. I'm mentioning this for the residents because it kindled an interest in me in this particular entity, which has lasted all throughout my career, and it's something I think as residents, if you do find something that appeals to you and interests you, and if you develop it into a specialty, it could become a career interest. So, let me talk to you about a few patients that have been of particular note to me over the years. Here's a boy, Danny, who was screened because of a strong family history. His father was a doctor, and he wanted to have his son screened because there were many relatives with cavernous malformations. We found on his MRI, and this is in 1989, this single lesion. It is extremely tiny. I think you can see it here and here. The question was raised then, of course, well, now we found this single lesion on this boy, what do we do about it? When I lecture in person to an audience, I frequently show this case, and I ask the audience, well, how would you all manage this? The typical response is, and I believe the correct one, we would not operate. This is very small. It shows no sign of activity, but you might guess what happened. He comes to the emergency room eight months later with his very large hemorrhage and edema. He's got a hemiparesis now. We have to do emergency surgery. It was, of course, a cavernous malformation. I show this case to indicate to you how uncertain the natural history of these lesions can be. This tiny lesion resulted in a very large hemorrhage in this particular patient. On the other hand, oh, and by the way, I have a 32-year follow-up on this patient. He's had no recurrences and no new lesions. What has been of great interest to me is that he has gotten married. He had two children, and here they are in this picture. The youngster on the left, both the kids were screened, as you can imagine. The one on the left, Charlotte, had a small cavernous malformation. You can imagine what anxiety this provoked in the parents. We watched this. It expanded just a little bit, but that was enough for everybody. We went ahead with surgery to remove it. Here are the father and daughter here with their paired incisions, the same area, the same lesion. At any rate, the point I would want to make here about these lesions is that the size of a lesion in no way is going to predict whether it's going to bleed or not. This patient, Danny, the first one I showed you, had a very strong family history. I had thought that in this clinical setting, the course of these lesions might be much more aggressive. As I mentioned, when his daughter's parietal lesion began to enlarge, everybody thought, well, let's go ahead and get this out. On the other hand, there are many relatives in this pedigree who have lesions studded throughout their brains and who have never had not only a hemorrhage but have had no clinical symptoms of any type. Here's an MRI. This is done a long time ago, a very primitive MRI, but it shows multiple lesions scattered throughout the cerebellum and brainstem and hemispheres. In this patient, who was age 63 at the time the scan was done, she had never had any symptoms of any type and to this date has never had any. Now, I want to show you some examples of some cases where the lesions behaved in a way that suggests that some of the bleeding from cavernous malformations may be self-limiting and that the patients, particularly with those lesions in eloquent areas, might be watched before making decisions about going ahead with definitive surgery. Here's a youngster who at age 12 presented with hemiparesis and headache. She had this lesion here in the left posterior frontal region. As it turns out, right in the motor strip. Here's her CT scan that was carried out in 1992 and here's her MRI showing this lesion. I had thought that we might be able to approach this lesion through a sulcus here. It didn't quite come to the surface, but perhaps if we came in through this sulcus, we'd be able to get this lesion out. We prepared her for surgery with the aids that we had at that time. This was in 1992. We localized it by then in those days stereotaxis and we used interoperative ultrasound to identify it. We also mapped motor and sensory strips and electrocorticography. We found that by ultrasound stereotaxis that this lesion was exactly in the motor strip and it was not visible on the brain surface. I thought that surgery was going to be risky on this patient with a high possibility that we were going to leave her with a hemiparesis. We backed out and did not remove the lesion. I suspect that not many of you would do this in this day and age, but that's what we did at that time. But look what happened to this lesion. Her neurologic deficit gradually disappeared and I followed her for years. 21 years later at her last follow-up, she has not had a recurrent bleed. She has absolutely no neurologic deficit. She does not have seizures. She takes no medications. Here was her last MRI done. We haven't repeated this one. This was 15 years after her hemorrhage and you can see that all that's left from that lesion on the MRI is a tiny slit with some hemocyturin around it. There's no evidence of any cavernous malformation here. I thought this was a very instructive case and I'm going to show you a few more like it. Let's talk about some of the lesions in the brainstem now. This is the youngster that I followed for the longest period of time. Sherry, she's from northern Maine. She suffered a hemiparesis and cranial nerve deficits with a lateral pontine hemorrhage when she was only 11 years of age. This was in 1984. We had a CT scan on her which I'm going to show you in a second and which demonstrates this cavernous malformation in the right lateral pons. We operated on this underneath the temporal lobe, cutting the tentorium and entering the side of the pons to remove the lesion but unfortunately it was too deep for me to be able to see it all and I knew that the resection was incomplete. We have pathologic verification that this was a cavernous malformation. Here was her initial scan in 1984. Of course no MRIs at that time but here's the scan showing the malformation here, the blood in the right side of the pons. I've followed her since then. Here's a CT scan on her in 1992 and you can see here that there's a lesion that fills almost all of the pons. At this particular point the patient is perfect. She has no neurologic deficit, no headaches, no symptoms of any type, no cranial nerve deficits. Here she is two years after her initial bleed and surgery. Here she is eight years later and here she is in 2019 on the right here with her daughter and her granddaughter. I have now a 37-year follow-up on her. She's had no new hemorrhages and no deficits. I mentioned to you she's married. She has children and grandchildren. I have been unable to prevail on her to get a follow-up MRI. She's very claustrophobic and she just simply will not she does not want to go through an MRI study which would be very instructive I agree but here's a case where there was an incomplete excision of a lesion. There obviously is calcification around where the surgery was carried out. The patient completely recovered her deficits and has never had another hemorrhage now nearly 40 years after her initial bleed. I'm going to show you a few others like her. Here's a 14-year-old boy who presented with a pontine and midbrain bleed from a presumed cavernous malformation. He presented in Mexico with a left hemiparesis and double vision and he was referred up for surgery. Here was his initial MRI study that was carried out after his hemorrhage. You can see this very large lesion again in the right side of the pons not quite coming to the brain surface on these images. By the time he got up to Boston his neurologic deficits had cleared dramatically and in view of the other cases I had had like him I told the family that perhaps we should wait for another hemorrhage before we decided to go ahead after the surgery which I think would have had considerable risk given the fact that he was now neurologically nearly intact after that initial bleed. This is in 1999. I subsequently have followed him up to the present day so we have a 22-year follow-up on him. Here are scans on the left in 2008. It scans on the right in 2011. You can see there's still residual lesion there surrounded by hemocytin but there's been no further hemorrhages. He's made a absolutely complete recovery without any surgery at all. As I mentioned now have a 22-year follow-up on him and he has no symptoms. He's running a company in Argentina completely independent and with no deficits and feeling perfectly normal. Here's a four-year-old Hispanic male who had the subacute onset of right-sided weakness in December of 2003. Here was his scan at that time showing this lesion on the left side of a ponto mesencephalic junction and again presenting with cranial nerve deficits and hemiparesis. You can see the lesion on these initial presenting images. I've of course followed him since then. He's now 22 years of age. We followed him since 2003 and he has had no neurologic deficits. He's completely recovered. He's had no further bleeds now 18 years after his initial hemorrhage. Here he is on the left and here is his scan in 2018. He just actually had another scan this year which is exactly identical to this scan in 2018 showing only a trace of hemocytin. Now a case that is even more interesting I think. This is a youngster who had several brainstem bleeds. I think probably many in the audience would have operated on her. She ended up with no surgery and let me show you what happened. So she had had three hemorrhages into a very large pontine cavernous malformation. She was referred for surgery or her considerations of surgery and I saw her in referral in March of 2011. She had developed worsening deficits with each hemorrhage with a left facial weakness. She had a left hemiparesis. She had ataxia and diplopia. Her scans were rather dramatic. The scan on the left was carried out in February of 2011. You can see the blood level in the hematoma cavity from the cavernous malformation and you can see the scan in March where the lesion has re-bled and the lesion has even expanded. Now it's interesting for each of us to think a little bit how we would manage the patient at this particular juncture. I thought well she's had three documented hemorrhages. Her neurologic deficit is getting worse. I thought that I saw an avenue of approach to this lesion which I thought could be safely removed underneath the temporal lobe there on the right side as I'd done before but and I presented this option to the family and I suspect that many in the audience would have recommended the same. The family decided that they did not want to undergo the risks of surgery in this situation and wanted to wait despite my advice and the advice of her neurosurgeon who initially had sent her for an opinion. Her bleeding stopped. She recovered completely. I have a 10-year follow-up on her now after that first bleed and she's had no further hemorrhages of any type. She subsequently is married. She's had a baby a normal child via vaginal delivery and this is eight years after her bleed and here is her scan was carried out in 2014 and a CT scan that was carried out later and you can see there's hardly any trace of that bleed. There's no cavity there and as I mentioned the patient looks and feels normal and has a normal exam. Again an example a great example I think of the very unclear history of these pediatric cavernous malformations and why it's worthwhile on occasion with these eloquent lesions watching and waiting. So I think there's certainly a role in these pediatric patients for conservative management in certain patients with brain stem lesions or in eloquent locations despite their initially significant neurologic deficits. However it's very interesting to me that other lesions for example in the hemispheres that are located in very deep areas often can be safely removed even when you think that the surgery is going to be extremely risky. I'll show you a few examples of this is a nine-year-old Hispanic girl again she had multiple episodes of hemorrhage from a lesion in the basal ganglia on the left side and she had intractable headaches. Here is her scan showing this lesion with multiple hemorrhagic cavities in it and deep in the brain and I think I thought at this point that surgery for her was reasonable to consider given her neurologic deficits or repeated hemorrhages and her intractable headache. We localize this lesion using an interesting technique, probably some of you have used this, some may find it not helpful. We placed a ventricular catheter using image guidance down to the lesion, and from then on in, we can get rid of the image guidance. We can follow that catheter dissecting right down around it to the lesion. This was via a frontal approach, and here we can see stepwise fashion heading down to the malformation. Here the malformation coming into view, you can see the purpley appearance of the malformation as the bipolar forceps and suction approach it, and here the malformation beginning to be excised. She had a spectacular result from the operation despite its deep location, and the day after surgery, she was perfectly intact. She was able to lift both her hands and had no neurologic deficit. This has been our experience in several other of these deep basal ganglia bleeds in our pediatric patients, and it has made us a little more aggressive in this particular location. If we can find a safe trajectory to the lesion. Here was her interoperative MRI after the lesion was taken out, and here's her MRI a year later, and I think you can see when you look at this MR, that this lesion, although it appeared to be centered in the basal ganglia, there's a lot of edema and hemorrhage here which really had displaced these structures and now that the hemorrhage has been removed, you can see how much intact neuroanatomy there remains after the surgery has been carried out. We've written up these cases in a paper that Brad Gross was the chief author when it was published in the journal Neurosurgery Pediatrics. Now, one of the other points that I wanted to make in this pediatric series was that cavernous malformations, and I think most everyone in the audience knows this, are very frequently associated with other venous anomalies. We know about their association with developmental venous anomalies, DVAs, but they're also seen in association with other cortical venous anomalies, cortical telangiectasias, and occasionally with arteriovenous fistulae. Let me show you a couple of cases. This case will be familiar, I think, to most everyone in the audience. This is a 12-year-old girl who presented with intractable headaches and had this lesion in the left cerebellum. I think you can see here, if you look at the contrast enhanced study, you can see this very large vascular structure on this scan on the left, and you can see some deeper veins adjacent to the hemorrhage and the malformation, which is right here. You can see that large vein here on the surface of the cerebellum. At surgery, we're looking at the surface of the cerebellum. Here is that very large vein, which is the developmental venous anomaly. At surgery, we're going to follow this vein down. We're going to dissect right along it, keeping that vein intact. Here we're dissecting a little deeper. You can see a tributary of that vein heading down to the malformation. Then here, the malformation uncovered. As per the standard teaching here, we try to leave these developmental venous anomalies intact and not disturb them when the malformation is removed. Now, this is the one extreme which I suspect many in the audience have seen. Here is a second variety, which I suspect you might not have seen, and which took me a long time to recognize the pattern of. This is a five-year-old boy who presented with a hemorrhage. You can see this cavernous malformation located here, adjacent to the frontal horn here on the right side. But look at what's going on here in the hemisphere above it. See this change that's here? I thought when I initially saw this study, that this finding represented perhaps some cortical atrophy or some ischemic change in that cortex related to vascular changes, related to the hemorrhage below it. When we did the craniotomy, and here is the exposure of that right hemisphere with the midline here, and the dural flap reflected to the left. Here's that right frontal lobe. Look at what we found on the surface of the brain. This remarkable concatenation of injected, distended, large veins. Now, one thing I suppose one might ask oneself at this point is, now what do we do? At this point, I said, well, we're operating. I don't know what these veins are. I don't know what they mean, but we're operating here to remove that cavernous malformation. We attempted to enter the white matter of the hemisphere by going actually through some of these veins. I'm sorry, I've just lost my picture here. We attempted to go right through these veins, and when we entered them, they were easily handled. They did not bleed very extensively, could be coagulated easily, and we went right through this area. Here's a picture after the cavernous malformation was taken out, and you can still see some of these vascular structures around the surface of the brain adjacent to the cavity where we removed the lesion. Well, at any rate, this boy did well for five years, and then he started having seizures, and then 10 years after his surgery, he had a massive new hemorrhage. When we scanned him, we found that that area of vascularity on the surface of the brain due to those venules and distended veins was markedly increased, and you can see how expanded it's become here with the hemorrhage subjacent to it. We operated, removed all those vascular structures and removed the hematoma as well. We have a four-year follow-up on him. Now, he's actually due for another scan shortly, but has been doing well, although he has a hemiparesis from that hemorrhage from which he has not recovered. Here's another case with a similar cortical lesion. Can you see it here on the scan? This is located here in the parietal occipital region on this young woman. Here is the susceptibility image showing this gradient echo image showing what appears to be either suffused vascularity or old calcification or old blood. As it turned out, in this patient who was incapacitated with visual auras that were keeping her from doing her work, from studying in school, and were unresponsive to antiepileptic medications, we operated on her and took this area out, which was a confluent area of telangiectasias. Her postoperative scans look like this, followed her through 2015. There's been no recurrence of that lesion, and her visual auras completely stopped. They stopped immediately after the surgery, and I have a seven-year follow-up on her with no recurrence. So we're looking here at what is a cortical telangiectasia. In this particular case, no subjacent cavernous malformation, but part of this juncture of vascular lesions in the brains of children. This is what the surface of the brain looked like in that patient. You can see some of these suffused, dilated, prominent venous structures in the surface of the cortex. Now, I want to finish with this case, which is probably the most extreme of these junctures of vascular lesions that I have ever seen. This is a seven-year-old, again an Hispanic boy, who had the acute onset of a left hemiparesis, a partial right third nerve and six nerve palsies, and ataxia. So clinically, right, we're going to be easily able to localize this lesion, correct? I mean, this is going to be a brainstem lesion, and it turned out that he had a cavernous malformation in the right lateral pons as some of our other children have had. We carried out surgery in December 2004, six weeks after his initial hemorrhage because of no improvement in his deficits, and I thought the lesion was easily reachable because it came right to the surface. Here was his initial scan, and you can see the size of this hemorrhage. This was, again, approached underneath the temporal lobe here. I want you to pay attention to something on this scan, though, which I think is interesting. You see these flow voids here? There are two flow voids here. You'd expect to see the basilar at this particular level, not two flow voids end on here, and one of these is a large vein, and here you can see on the contrast study the vein which is running right down alongside the lesion. When the malformation was taken out, and here's his postoperative scan, was a nice resection. The patient did very well. Here's the lesion removed, and here you can see that large vein just extending all the way around the left side of the pons and midbrain, an incredibly large DVA, but wait, there's more on this case, so I see him again in follow-up. Here he is a year after follow-up with his family. Here he is on the right. He has no new symptoms. He's doing great. We get a follow-up scan on him, and look what's happened on the follow-up scan. That malformation, which I was so happy that I had taken out completely, has recurred. Here it is all back again, and look what's going on here. He's got another one on the other side. He remains asymptomatic at this particular point. What we find eventually on him is that he has an arteriodural arteriovenous fistula, which no doubt is increasing the venous pressure in the back of his brain, in the posterior fossa. Here's the injection that was carried out through the external carotid and its branches, mental meningeal branches, and this was embolized and completely eliminated, and it's been eliminated now for years, but unfortunately he's had one new cavernous malformation after the other developed de novo in his brain. Here he is in September. This is three years after his initial bleed, and he's developed a new hemorrhage in the cerebellum. I'm going to show you a whole series of these films where there's going to be new bleeding. Here he is after his initial surgery. The scan doesn't look so bad, but now you see a new bleed here, here, so we operate on this lesion, take it out. Then in January 2009, he develops another one, back further in the cerebellum, and then April 2009, another one, and you can see at this point it's pointless to continue operating on these. They keep forming de novo, bleeding, patient becomes symptomatic, and then recovers from them. Here he presents with a new left facial weakness, and you can see here this lesion that develops in the pons in this area. Here he has a stable scan a year later, and then look at all these multiple lesions that have occurred in his case. Now here he is nine and a half years after his initial bleed. He's got a stable exam with no new clinical or imaging re-bleeds for four years, and then 15 years after his initial bleed, he now presents with a confluent expansion of these lesions in this left cerebellar hemisphere, and you can see this right anterior pontine lesion that's now expanded, that's worsened his gait and given him great ataxia. So now at age 24, he's independent, but he's mobile only in his wheelchair. He goes to college, but he's got severe truncal ataxia. This is now 17 years after his initial bleed. His arteriogram has been repeated several times. The arteriogram remains normal with no evidence of recurrence of his dural arteriovenous fistula and remains obliterated. He had a new right third nerve palsy a couple of months ago, but the MRI showed regression of many of the older lesions. Of course, there's no point in operating on any of these, I think, they just keep coming and going, and he's a real mystery case for him, and if I had some nice anti-angiogenesis medication, that probably would be the best treatment for him. Here is a family portrait. This is the initial one I showed you with him on the right, and here he is again right back here, this picture taken just a couple of years ago. So, I mean, I think you all know a great deal about cavernous malformations, and particularly their relationship with venous anomalies. I think one of the important points for the neophyte operating on these lesions, when you're looking at your preoperative scans, look for these venous anomalies on the preoperative scans carefully, and the teaching that we've all been given is that we must preserve these at the surgical resection to avoid venous infarction of the adjacent brain, and I believe that's true, although I also believe that these venous anomalies probably are responsible for focal increases in venous pressure in the brain at their distal point, leading to the development of these cavernous malformations, which are probably repeated bleeds that instead of getting absorbed, get endothelialized and bleed, much as chronic subdural hematoma membranes continue to bleed. Of course, I have no direct definite evidence of that, but that makes the most sense to me, and there have been some papers describing this phenomenon, and the Journal of Neurosurgery Pediatrics, five years ago, one author discussed the de novo formation of a large cavernoma that was associated with a congenital torcular arteriovenous fistula. So my conclusions here and takeaway for this audience is that we've reviewed, I hope to your satisfaction, how variable and unpredictable the natural history of cavernous malformations can be in the pediatric patient. I've shown you very small lesions that have ruptured and bled and caused serious deficits, and I've shown you large lesions that have bled significantly and have completely resorbed. So I think the neurosurgeon has to very carefully weigh the risks and benefits of operating on cavernous malformations in the pediatric patient, particularly when the lesions are located in these so-called eloquent locations. I'm particularly thinking here about the brainstem. I know that others in this webinar are gonna talk about and have talked about innovative approaches to lesions in difficult areas in the brainstem and how they might be removed safely, but I wanna emphasize to you that not all of these lesions need to be operated on and that you need to let the patient's lesions' natural history declare itself before forging ahead with a very difficult and risky surgical procedure. We discussed also the various manifestations of cavernous malformations that are associated with other vascular anomalies of the brain, and I'm thinking particularly of these very bizarre cortical telangiectasias with subjacent cavernous malformations. You need to be aware of that. And also the presence of these cortical telangiectasias in and of themselves, vascular anomalies that can cause symptoms much as a brain tumor might cause with seizures and headache. I mentioned these again, the confluent telangiectasias and the cavernous malformations that are associated with DVAs and dural arteriovenous fistulas. It's been a pleasure participating in this webinar today with you, and I hope you enjoyed the presentation. I'd be pleased to answer any questions that any of you might have, and you can certainly contact me by email if you wish. My email address is michael.scott at childrens.harvard.edu. Thank you.

Dr. Michael Scott

neurosurgeon

pediatric cavernous malformations

natural history

observation

management approach

venous anomalies