All right, good morning. I also want to thank Errol and Adam for inviting me to be a part of this. I've been fascinated so far. I hope I can contribute to the discussions. These are my disclosures. I do some consulting work for a couple of biotech companies, including Stryker that's a sponsor of this event. And I'm going to start by just showing a few pictures of some of the different lesions that we are talking about at this conference. We've had great discussions this morning. The three talks about the technical aspects of treating aneurysm. Jay is going to speak about endovascular treatment of large vessel occlusions tomorrow. And my colleague Lucas, who was kind enough to switch topics with me, is going to talk tomorrow about our approach to AVMs. So I'll submit to you that obviously these are very challenging cases. They can be technically very difficult. But I would also say that the evaluation of the anatomical challenges and the details of the anatomy are more or less pretty straightforward, certainly with our modern day imaging in the angiography suite and with cross-sectional imaging. I think most of us, by the time any of these three patients get to the angiography suite, they already have some detailed CTMR data where we really know largely what we're about to get into and what we're going to approach. And we probably have thought about how we're going to approach any individual case even before we do a single angiogram. And I will tell you, I think that dural fistula are very different from that. And so first of all, I think they're very difficult to identify and know what's going on based on CT or MR imaging. And then they also tend to be supplied by external carotid artery branches, which is not something we study with every cerebral angiogram. So I don't think all of us, including myself, are always so facile with the external carotid anatomy. And here's a lateral injection of an external carotid artery run. And if you're like me, when you see something that looks like this, there's obviously something sort of interesting going on in this area. But my initial impression is always something like my thought process when I see something like this. It seems sort of cool. It's a little bit bizarre. It's sort of fascinating to look at. And then after that initial impression passes, I realize that I'm not real sure what the hell is going on. And I sort of sometimes feel like this guy over here. So the dural fistula anatomy can be extraordinarily difficult to evaluate and to treat. So when I was asked to talk about angiographic and evaluation and then the techniques of treating dural fistulas, I decided that I wanted to focus a lot on the evaluation itself. Because I think that we should realize that dural fistulas are probably the most complicated cerebrovascular lesions that we're going to treat. Everyone is profoundly different and individualized in my experience. So we treat a lot of ACOM aneurysms, and yes, they're very different. But they all have sort of underlying general themes. I think dural fistulas are very different. And a comprehensive understanding of the anatomy of the fistula of your particular patient, that's what's going to be critical to successfully treating it and most importantly to treat it safely. And so when I think of what are the techniques that I use to treat a dural fistula, I think this understanding of the anatomy may be the most important technique. And yes, we're going to talk about coiling dural fistulas, of treating them with liquid embolic agents and the techniques that I do that, use for that. And I'm going to spend a lot of time with how do I know what the anatomy is, and how does that inform what I think the risk of not treating, and what I think the risk of various approaches are. So probably familiar to most everyone in this room, but we're of course talking about abnormal arterial venous shunts that exist between dural arteries, okay, so most commonly branches of the external carotid, and the dural venous sinuses, meningeal veins, obviously embedded within the dura, or cortical veins, and or cortical veins, so any of these in combination or isolation. And the critical part is to understand of course that there are, I say dural arteries, but in practice are usually innumerable arteries that will supply these fistulas. But instead of anidus that we see with AVMs of course, there's instead a fistula pouch. And that is going to be the critical aspect of the anatomy when we try to eliminate the fistula. But this arterialized pouch then gives rise of course to an arterialized venous outflow, which is where potential complications and symptoms of dural fistula arise. So just to review, here's a sagittal netter illustration, and if we just focus then on the torcula and the transverse sigmoid sinus junction, this is a nice review in AJNR not too long ago. But it just illustrates the basic anatomy of a simple dural fistula, and actually the transverse sigmoid sinus is the most common location for dural fistula. The most common arteries supplying them are these transmastoid perforators of the occipital artery shown here. So in this case, you have these little branches coming off the occipital artery running through the mastoid region and forming a fistula pouch somewhere along the wall of the dural sinus itself, perhaps dumping directly into it. As you all know, a consequence of that is to basically arterialize the sinus. So you're pumping arterial blood directly into the transverse sigmoid sinus junction. Well, what are the consequences of that? Well, in simple fistula, the return of that may be simply anterograde down the sigmoid sinus to the jugular vein. And if that's the case, then the normal venous outflow, which is under normal venous pressure, here's the vena bulbae, for example, coming down the superior sagittal sinus, that normal pressure venous blood is simply going to join the arterialized blood and continue on its way in an anterograde fashion down the jugular vein. And these are fairly low-risk lesions. They could cause pulsatile tinnitus, given their proximity to the inner ear and the temporal bone, if you will, but they're very low-risk lesions. Well, consider an alternative where the outflow of the sigmoid sinus to the IJ becomes, say, pulverosed and occludes off. Well, now the arterialized portion of the sinus has to direct its blood elsewhere. In this case, now it's going to be retrograde, say, in the sinus. Perhaps most concerningly would be retrograde arterialized blood in a cortical vein, so here the vena bulbae. Now there are multiple different grading systems for dural fistulas that predict their natural history risk. I'm not going to go over those in detail. I want to be much more practical and just talk about the different types of fistulas that we can encounter and how we can approach them. And I like to think of this in the sense of how dural fistulas probably evolve from low-risk to high-risk, although this is sort of hard to prove. But this is the same idea. Again, you can consider this being occipital feeders. This might be middle meningeal feeders coming to a dural fistula along the transverse sigmoid sinus. And again, in this case, all the arterialized outflow is anterograde down the jugular. This would be a low-risk situation. While dural sinuses and venous structures aren't really geared to deal with high pressure and high flow, and one of the consequences over time is often referred to as a high-flow venopathy, where you can get progressive stenosis of the dural sinus, so just distal to the arterialized segment. And when that happens, it's going to start to restrict the arterialized outflow. And at some point, the normal inflow, say, of the vein of the lobe is going to become arterialized and may become retrograde in fashion. Well, you can imagine this continued to evolve where this will thrombose completely. And now you have an occluded distal segment of the outflow sinus. And the arterialized blood in this case has no choice but to go retrograde out the, say, to the torcula and beyond, but again, most concerning in the arterialized vein. So a lot of folks believe that fistulas may start this way and evolve over time to this. There are other hypotheses that suggest that occlusion of the dural sinus, just as sinus thrombosis, may be an inciting event which ultimately leads to a fistula formation. So there's data that fistula can progress from here to here, and I'm going to show you an example of that. But regardless, as you become more and more restrictive in your venous outflow, the risk of venous congestion and of intracranial hemorrhage continues to increase. And the most concerning fistula of all would be where both the proximal and distal outflows of the normal dural sinuses are occluded. Now you have an arterialized segment, and the only outflow of the fistula pouch is retrograde in a cortical vein. And these, of course, are the highest risk for thrombosis. So we need to understand that anatomy. We need to understand that there is a recipient fistula pouch that is located somewhere along the wall of the sinus or in the sinus itself if we're going to be able to treat it successfully. How are we going to treat it? Well, this is another critical take-home point, and again, most of you may be familiar with this, but I'm still surprised that a lot of folks in our field don't fully understand that if you're going to cure a dural fistula, you really have to completely obliterate the fistula pouch. So that means not simply a proximal occlusion here, but somehow getting into the fistula's pouch, whether with a liquid embolic, whether with coils to obliterate the fistula's connection. So if you're going to come from a transarterial approach, and let's say you were to navigate a catheter, and these were the only identifiable feeders to the fistula, you might say, well, could you come in here and simply just coil across here, occlude all of these, and be done with it? And actually, angiographically, immediately afterwards, you may not see the fistula fill anymore. But a good point to make is that proximal occlusions of all the arterial feeders, even if you could get them all, and in fact you can't because there's so many tiny vessels there that we're not going to be able to see, the collaterals are so robust that even if you could occlude them all, that will ultimately fail. So we know that that is not a sufficient treatment for dural fistula. So one other option would be to navigate a microcatheter and then to embolize from a proximal position. And there's lots of experience of doing this with particulates, so with PVA particles. And again, you can achieve an angiographic improvement or cure instantaneously, but we also know that particles ultimately will always fail. These will recantalize. So you need some sort of permanent occlusion of the fistula pouch, and how are you going to do that? Talk about from a transarterial standpoint, the best thing to do is to use a liquid embolic agent. And this is where I think, even though Onyx is not officially approved for fistula treatment, it's for cerebral AVMs, it really sort of revolutionized transarterial approach to dural fistulas. But with something as powerful as Onyx comes issues, right? And so I like to think that danger lurks in tiny branches, should be the name of some mystery novel. But the point is, Onyx, and this was referred to in the previous talk, it's pretty amazing. You can get it to go a lot of places, including into small branches that you may not otherwise recognize. And we have to always remember that there are a lot of dangerous external to internal anastomoses. The occipital, as everyone knows, has a suboccipital knot connection to the vertebral artery. There are connections to the meningohypophyseal trunk and inferior lateral trunks of the cavernous carotid external branches, ophthalmic, et cetera. We have to be aware of those so that our embolic materials are not going places we don't want to, to the intracranial circulation. But even more important to that, I think, are we have to remember that many of the external carotid branches that are going to be supplying dural fistulas, that we're going to want to go in and embolize to cure the fistula, many of them are supplying really important cranial nerves. And so if you were to use particulate PVA particles of a certain size, say greater than 250, 300 microns, you're going to have particles that are so big that they can never penetrate the vasovasorum of the supply to cranial, to a nerve. You're never going to harm them with a particulate. But with agents like Onyx that can penetrate even these extremely small vasovasorum supply to cranial nerves, we can do a lot of damage and not particularly be aware of it. So we need to know what arteries are safe and what arteries are going to be high risk for embolizing fistulas. From a transvenous approach, and I trained at UCSF where the team was one of the teams that pioneered early transvenous approach to dural fistulas, and so I have a decent number of experience in my fellowship training with this, is another interesting way to try to approach the fistula. So for example, it would be possible to navigate simply up the arterialized sinus from the IJ and to simply come in and coil across the fistula's pouch and eliminate it. But one important point, as we'll come to again and again, is that we know in this case, in this scenario, the normal brain, the vena libae, the superior sagittal sinus, it continues to use this outflow for normal venous drainage. And so if you go in and occlude the fistula, you may occlude the fistula, but now you're going to be rerouting normal outflow of the brain. And you may get away with it, but you may not. And so my general rule is that if there is normal venous outflow already present, we need to try to preserve that. In this scenario, where this is already occluded, no normal outflow is going through this arterialized segment. In fact, the only outflow present is high-risk arterialized outflow. And so you could come up and across the other transverse sinus through the torticula into this space, and you could coil off the fistula and eliminate the pouch, because as a general rule, no flow in a cortical vein is much safer than high-risk retrograde arterialized flow. But there, I think the danger lurks in incomplete occlusion. And so if you're going to attack a fistula in this manner, if you were to be a little imprecise with the coil placement, you could do sort of the worst possible thing, and that would be to essentially create now an isolated fistula pouch where the only outflow is retrograde through the cortical vein. So those are a few issues that we think about when we're trying to say, can we approach this fistula, are we going to do so from a transarterial or a transvenous approach, or perhaps both. And there are some cases where both may be useful. Now, to get to the point of deciding, I think this is where we really, again, have to understand the entire anatomy. And that requires a complete six-vessel cerebral angiogram. Now, we're not always routinely doing external carotid artery injections in our aneurysm cases, diagnostics, et cetera. It's probably the most important angiography you're going to do in the evaluation of a dural fistula case, and that is selective external carotid arteries injections. Again, we've all probably seen examples of patients coming transferred to your institution with unexplained intracranial hemorrhage that had an outside angiogram, and it was a common carotid artery injections, perhaps selective internals. But without visualization of the external carotids, you can miss an obvious high-risk dural fistula. So that's critical. What are the goals of doing the angiogram? Well, obviously, you want to identify all the arterial feeders that you can to see what is the best target for possible embolization. We want to also identify the red flags. What are the potential dangerous players with external to intracranial anastomosis? But this, again, is where I'll come back to again and again where I think is really important as we use ONIX more and more is that we need to be aware of some of these critical dural branches that are supplying cranial nerves. Now, there are more than two, but it turns out that the most common external branches supplying dural fistulas are the occipital first, the middle meningeal second, and the ascending pharyngeal third. So at least we should start with those to know what are our dangerous red flags. And for these guys, there's the petrosal branch of the middle meningeal artery, which runs in the internal auditory canal with seven and eight, supplies the facial nerve. And then the neuromeningeal branch of the ascending pharyngeal, which I'll show, basically supplies a lot of the lower cranial nerves. These are important. Patients want them to work. And we need to be careful to preserve them. Obviously, we want to map out the fistula pouch and its venous outflow. Is there anterograde or retrograde flow within the dural sinus? Most importantly, I think, is because this is really the key determinant of grade and the risk. Is there retrograde cortical venous drainage? And I've shown the example of the vena libae. So that is critical to look for because all of those patients are at now elevated risk of intracranial hemorrhage. And then, of course, do we even have access for transvenous embolization? Is that an option for treatment? And then, again, the reason to do a complete six-vessel angiogram, you might have a left-sided transverse sinus fistula supplied by the left occipital. You do a left internal. There's no obvious supply from the cavernous branches. You might not feel the need to do a full diagnostic angiogram. But you need to know, what is the normal venous outflow of the posterior circulation of the right anterior circulation? Because if you identify that, again, you're going to want to preserve it. So I can't stress enough the need to do a full diagnostic angiogram. And then, finally, I would stress that 3D rotational angiography is standard. We use it every day, basically, for the evaluation of aneurysms and AVMs and how we plan our intraoperative projections. And I use that, of course, as well. But I found this most useful in combination with a flat panel CT scan and overlaying them as defining the fistula anatomy. Where are your most likely arterial inputs that you can access? What are your transvenous pathways? Where actually is the fistula located? What if you can't easily cure this? Where do you want to tell your surgeon to look? I find it extremely useful. The other thing I find useful, and again, to come back to this notion that external carotids are the main supply of these guys. And again, they're not something we're super familiar with, because we don't evaluate them all the time. So we need good reference material. So here's one of our bibles, of course, by three giants in our field. And if you look at their beautiful overview, this is the common carotid. That's the internal. It's been sort of lopped off. So they're showing the many branches of the external carotid artery beautifully identified. I think I can name four of these. So the point is, in normal anatomy, you often don't see these at all. But when they tend to supply dural fistulas, they become markedly enlarged. So the anatomy that we see with dural fistulas, again, it's just not something we're used to. It's not something I'm used to every time. And so I find references like this extremely useful. My favorite, actually, is this one from, again, another giant in the field. Microsurgical Anatomy of the Dural Arteries, published in Neurosurgery about 10 years ago. I would definitely, this is like a 41-page tour de force. It may be my favorite reference article period ever, because I love anatomy. But there are these beautiful dissections of all the dural arteries. And these unbelievable maps of everywhere along the skull base, along the dura, what artery of the external carotid artery system supplies that particular supply. So if you have a dural fistula running along the skull base at this point, you have a good idea what arteries you should be looking at. Do they supply cranial nerves, yes or no? I find this extremely useful and unbelievably well done. There are two arteries I'm gonna highlight before I get to cases, and this is one of them. So this is the patrosal branch of the middle meningeal artery. So this is the external carotid artery illustration. So here's the middle meningeal artery entering the foramen spinosum right about this point. At the entry to the foramen spinosum, it gives off a patrosal artery. And again, this is an artery that's gonna be small. You're not gonna see it, right? You're probably not gonna identify it. We just have to know that it's there. But it's gonna run in and supply the facial nerve. There's a branch coming up the stylomastoid from the posterior auricular artery that is also gonna supply the facial nerve as it's exiting the skull. But we need, again, the MMA is one of the most common suppliers of dural fistulas. And if that supply is arising in and around this area, we need to be very careful. There's a case report, if I have time I'll get to, of embolization of a dural fistula in this region. And the patient had a postoperative Bell's palsy that never resolved. And an institution actually tried a facial nerve decompression. It was very easy to visualize onyx in the vasovasorum running along the facial nerve. So this is, in my mind, a big red flag if we see supply arising in this area. The other important one, I think, to know about is the ascending pharyngeal and, in particular, the neuromeningeal trunk. So this is an illustration of the ascending pharyngeal. Here's the pharyngeal trunk with the palatine arteries running forward. Here's the neuromeningeal trunk running forward. If you look down into the foramen magnum in the skull base, after it's entered the foramen magnum, it's running up. It gives several branches, the hypoglossal, the jugular, and the internal auditory canal. Well, look where they're running. Well, we know seven and eight run there, nine, 10, and 11 run there, and 12 run there. These branches are gonna supply those cranial nerves. And if you're embolizing with onyx in and along the ascending pharyngeal, particularly in the neuromeningeal trunk, there is a real risk of cranial neuropathy that's not going to improve. So I find these as big red flags, and I personally am very cautious about any transarterial embolization in their territories. Okay, let's actually go through, I'm actually gonna go through like two or three cases, okay, because again, I wanna focus mostly on anatomy. But let's talk about, start with this one. This is a 61-year-old woman who had a seizure and awoke with postdictal aphasia and right hemiparesis. And this was her CT scan where there's some unusual sort of multifocal, fluffy-type hemorrhages on the left hemisphere. There was an MRI scan that was done that showed also edema in this area and in the temporal lobe. And the CT angiogram, I don't know how well this projects, but there's no anitis, there's no obvious abnormal external branches, there's no obvious abnormal internal branches. And yet there's a bunch of sort of corkscrew-ish, tortuous, everywhere you look, and the more you look, you'll see all these sort of bizarre corkscrew-y type vessels in and around the middle cerebral branches. And this is sort of a crappy 3D reformat. But again, there's just too many little notes in here with a bunch of corkscrew tiny guys. And those should be sort of red flags of some arterialized, small cortical veins. And if you don't see an obvious arterial venous shunt, a dural fistula should be high on the list. And again, so this is the angiogram, this is the left external carotid artery injection. I showed you this earlier. And so what we see, again, is here's the middle meningeal coming, the superficial temporal IMX obviously running forward. Is there something clearly not normal going on in this general area? And now these look like cortical veins that are running sort of retrograde. If I go back, you'll see that those look like cortical veins that are running on the surface within the left hemisphere. Okay, so suspicious for arterial venous shunt at the skull base. I'll be honest, I'm not exactly sure where that is right now. We do an internal carotid artery injection, okay, looking at the other venous outflow. And one thing you'll notice is that this is probably the inferior lateral trunk coming off the cavernous carotid, coming down. And it's also contributing, you actually don't see the high-risk cortical venous drainage because of just the competitive inflow to the fistula, but you know that that is where this is going, okay. So we've identified, again, an important branch coming off the cavernous carotid. Now, this is where, again, where I find three-dimensional rotational angiography so useful. This is the 3D angiogram of the left external carotid artery injection. And I'll just rotate it a little bit. And you can see there's this sort of, I don't know, it looks like a small little dog turning around there, which is actually the fistula pouch. And then there's a connection, looks like just a high-grade stenosis connection into what is a cortical vein near the sphenobasal sinus region. Now, what I find useful then is to overlay that with a flat panel CT, and that's what's shown here in a sagittal reconstruction. So here's what we're looking for. I wanna know exactly where is this located. And then this is overlaid where the 3D, obviously, is projected in pink. The white is the 3D angiogram we did of the internal circulation. And I think you can see sort of beautifully that this sort of bizarre dog-shaped thing is actually a fistula pouch that's in the sphenoid wing at the skull base. And as we go back, you can see right here is where there's a little hairpin connection where it becomes intracranial. And these are the cortical veins that are now running sort of retrograde. So if you couldn't approach this from an endovascular standpoint, I think this has a lot of useful information for the surgeon if you will talk to Adam and others if they would ever attack this surgically. But the point is, I wasn't totally sure where this fistula was, exactly where the feeders were coming from. And the technology we have in the angiography suite gives us unbelievable detail of sort of what we're dealing with. So I think we've identified, so stop there. So there's the summary of what we're dealing with. Now we know the fistula pouch. We know the venous outflow is isolated. It's cortical venous drainage. There's not gonna be really a venous access to this. The question is, what's the arterial supply? Well, there's an interlateral trunk coming down. I know there's something going on in the external carotid, but we haven't defined those. And here's where I would advocate very strongly that you do super selective angiograms with microcatheters in the usual suspects. And I started with a microcatheter in the common trunk of the middle meningeal. And this is just a lateral projection. And again, you can see that this is, the fistula actually is located here. This is sort of the venous pouch. Hopefully you can recognize now the cortical venous outflow. But again, this is the area, that thing right there, of the petrous branch of the MMA that we know is gonna be supplying the facial nerve. And so I'm not actually super excited about the notion of embolizing this fistula from this location. Yes, you could watch to see if you have any flow to it, but again, onyx, there's a lot of penetration of tiny vessels that you probably can't visualize angiographically to know when to stop. So I'm not a huge fan of that. I looked a little distal on the IMAX into the accessory meningeal, and I felt like we could get fairly close to the fistula. Again, there's the venous pouch. The fistula actually is located sort of more near the base. This is the fistula pouch, and then there's the outflow to the cortical venous drainage. But again, I'm just immediately downstream of this thing, and I still don't wanna embolize the fistula and have any concern for retrograde flow in the petrous branch. Yes, you have a question? Sorry, can you point out again, what tells you what part is the fistula and what part is the pouch? Sure, so that's a great question. For that, I unfortunately don't have the entire angiographic run here. So you have to look in the earliest. I do this like you're filming an AVM, so I do this on fast filming, as fast as possible. And you wanna look as early as possible in the angiographic run to see where does the fistula actually form, and then you can see what is the actual outflow of the fistula. My next case, if I can get to it, has a better example of that. But you can show them that, so just on that picture, show them. One of the things you can ask yourself when you're analyzing this is what on that picture is artery, what is vein, and then start to narrow it down between those two routes into where the fistula is. Absolutely, right. Yeah, so this obviously is the main trunk entering the foramen spinosum running forward, but there are tiny little branches that are coming off here, but then there's a caliber change right about in this location, where this is switching to the venous pouch. So the fistula has to be located more in here, and this more is that venous pouch that's in the spinoid, ultimately running with the outflow to these cortical veins. I agree with that analysis 100%, and that's a useful thing to do because you know for an AV fistula, you need to get that venous pouch where the fistula's going to cure the patient. You know for AVM, the last thing you want to do is take the vein early. Absolutely. They're opposite thought processes, but you can't even engage that thought process until you've gone through sometimes really confusing extensive analysis of the anatomy. Right. And Joey's doing a really good job of sort of taking you through that, but I can tell you sometimes you start to analyze it and it's just daunting. Yeah. You're just looking at these different vessels going what the hell am I looking at? At least in my shop, we spend a lot of time arguing with each other and saying well this is what I think is what I think I'm trying to figure out. Yeah, I completely agree with that. I'm showing, you know, we probably did selective angiograms of six to eight different vessels, and I'm just going to show a couple of them, but I completely agree with that. So this was looking at the accessory meningeal, which, again, is just a little downstream on the IMAX, where we were able to visualize, again, the fistulas connection, which is in here. Again, this is the part of the venous palx in the outflow. We've hijacked your talk a little bit. That's okay. So there's an expression, measure twice, cut once. Right? In arbitrary. Well, we're not working on a two by four. Yeah. the data, we scrub out, we sit down, we go over this in sort of great detail before we decide where we're going to attack it. I agree, a hundred percent. There's so many things going there. You need to take your time. You cannot be under time pressure. You cannot just say, well, can we put some on somewhere and start an NBCA? No, you can't. You have to have a plan. You have to understand, yeah, I'm going. What's my goal? I don't know yet. So yeah, I agree. I'm glad that we're all on the same page. And I mean, this is my basic point where I think the most important technique of treating these fistulas is the evaluation of the anatomy. And that's sort of what we're going through. Without showing all the other angiograms, we felt like we had reasonable access here with the left accessory meningeal to try to attack the fistula, which we believe by all of our runs was based right here at the base of this pouch. But we were still nervous about any retrograde flow that could get to, because we're really just basically going to be embolizing right in here. We wanted to avoid taking out the petrous branch of the minimeningeal artery. And so I'm curious as to other thoughts. But what we decided to do was simply to coil off the MMA. If we occlude this here and thrombose it, we're not going to be able to retrograde fillet with onyx. And so again, remember, the collaterals are so robust. You can take down the MMA with impunity. That's no problem. What we're trying to do is prevent downstream penetration of the vasovasorum by a really powerful liquid embolic agent. So that's actually what we did. So we just simply put a 2.3 French microcatheter. I use a Prowler Plus, the various ones that you can use. I use it because you don't need to be super elegant for this. You can use inexpensive, fibered, pushable coils, 18 coils. And so we simply coiled off the MMA at that point. And then we went back to the accessory miningeal. Now you can see our coil mass. You can see how close we are actually to the MMA. And we're going to use onyx from here. Again, before we start with the onyx, we obviously know where we want to be. We know what the venous outflow is. So we know what our target is. But we're more concerned with the onyx going places we don't want it to go. So hopefully, we've eliminated. This is the MMA that we've now coiled off. We hopefully have eliminated that. We still have to realize that there are collaterals that go back to the miningeal hypophyseal trunk in the ILT into the cavernous sinus. We have to avoid that. We know this guy is coming downstream. And we could easily retrograde and embolize onyx directly into the internal carotid. So you just have to be aware of all these things before you start. Now, I use a 6-inch guide catheter system just in the proximal ECA. I don't use any heparin at all for this. The case I'm going to show you, I've used a 1.5-inch marathon catheter, either over a mirage wire or an expedient wire, an 08 or a 10. There's a new microcatheter now out now by Covidien. It has a detachable tip called the Apollo. They may have that downstream. Jay mentioned earlier using the scepter balloon for dural fistulas. There are lots of different techniques that are sort of beyond the scope of this right now. Hopefully, we'll get to do them down in the lab. And I like to use onyx 18 for if I'm in an extremely small branch and I still have a little ways to go to the fistula. And most of you probably have used onyx, and we'll get to use it in the lab. But it's important that we just use serial road maps, blank road maps, along with the raw florals. We're building our plug to then gradually sort of push it forward. And the key about using a guide catheter system that will accommodate your microcatheter but also to allow you to do surveillance angiograms as you are in progress. And then, of course, with onyx, whether it's an AVM or a dural fistula, patience is sort of critical. I'm just going to show some selective images from this. And so our typical catheter is about here. We've had some reflux that we had to be patient to allow. Ultimately, we were able to get this going forward into the fistula pouch. You can see the different septations here that correlate with the fistula pouch. We thought we had done pretty well at this point. We do a surveillance angiogram. And again, you can see that the cortical venous drainage is still present. So the fistula is still filling. It allows us to continue on. We didn't see any worrisome reflux or onyx going where we didn't want. And then with the follow-up angiogram, this appears to be now completely obliterated. Again, the dural fistula, we believe, is in this location. We filled the venous pouch completely. Because again, you want to saturate the fistula's connection and get it into the main venous outflow if possible to completely eliminate it. And this is after. These are the coil mass and the MMA and the onyx cast, which hopefully you can see a little bit. That was cured. So I'm going to, in the interest of time, I'm going to go through. I want to do at least one other case. I have a couple others. But just to emphasize a slightly different approach. So this is a 76-year-old man who was referred to us by our neurosurgical colleagues. Had an ugly occipital hemangioparasitoma. And they asked us to do a preoperative embolization at the time. And we did a diagnostic angiogram which showed that this thing had occluded the distal superior sagittal sinus, the torcula. Both transfer sinuses were out. We did an embolization. But we also saw, interestingly enough, again, not predicted by the cross-sectional imaging. But when we did the ECA run to look at our targets for embolizing the tumor, this is sort of what we found. So this is a lateral view. This is the ascending pharyngeal coming up. This is the pharyngeal branch here. This is the neuromeningeal branch coming back. This is the occipital artery. And you might notice that both of those are filling much faster than the other external branches. That's because they're going somewhere. And the place they're going is the fistula. And this gets back to the question you asked earlier. Initially, I was wondering, when I first saw this, I'm like, well, is the fistula in here? Because there is this caliber change. But actually, if you look closely, you can tell these are arteries. Right there is a caliber change that you can see that this is a venous pouch. And actually, if you look up along through here, there are multiple little tiny, little vessels. So it may be that the fistula pouch is all along through here. But certainly, the caliber change is here. There are transmastoid perforators coming from the occipital running also here into the transverse sinus. Remember, the proximal transverse sinus is already occluded. We know that by the tumor. And then if we look at the outflow of this, we see two things. This is retrograde cortical venous outflow here and here. All these are cortical veins. It's a little busy because this is actually going to end up being tumor supply, which you'll see in a minute. But then there is outflow down the sigmoid sinus down to the IJ. So initially, I thought that was the sigmoid sinus. It's not. It's a parallel channel running along it. But as you can see, that the sigmoid sinus is patent, although it's nearly occluded right at this point. But there is anterior grade outflow to the jugular. And here you can see the tumor blush, which is separate. So this is an intermediate to high-risk fistula because there's retrograde cortical venous outflow. So the patient was actually symptomatic with a field cut of this tumor. And that was actually handled. We embolized the tumor. He had a surgical resection and then was going to come back about three weeks later for us to do a treatment of his fistula. Now, what are the options? Well, we could come transarterial. So we could come up the ascending pharyngeal. This would be the neuromeningeal branch. Hopefully, I've convinced you to at least have second thoughts about doing that. The transmastoid occipital feeders would be quite good targets for this, I think, as long as you could avoid reflux down the ascending pharyngeal connections. But again, I want you to think about transvenous access. So again, I haven't showed you all the angiography. But this is an isolated pouch within the transverse sigmoid sinus junction. It doesn't go anywhere except for retrograde cortical venous drainage and down a sort of dysplastic, nearly occluded transverse sigmoid sinus, IJ. And again, the full angiogram we did told us there was no other normal venous outflow using the system. So you could take it down with abandon. It should be fine. So you should be able to come up, retrograde access via the IJ, come up through the sigmoid sinus, get into the venous pouch, and to simply coil it off. So that actually was going to be our goal. This was the initial angiogram, just again, to remind you of the anatomy. And when we did our follow-up angiogram, which again, I think was about a month after his initial surgery, this is sort of what we saw. So the fistula is still filling, ascending pharyngeal, occipital artery is coming up. Here's the fistula pouch running through here. As we go forward, though, you'll see something has changed. And what has changed is that the sigmoid sinus, which was almost occluded at the time, now is fully occluded. And so now the only outflow of this is, all of this is high-risk retrograde cortical venous drainage. And so we've gone from bad to worse. This is a very high-risk lesion. And our venous access has now been apparently cut off. So the question is, is transarterial via these transmastoid occipitals, is that the only option, or can we still go retrograde? And so again, I train by doing a lot of transvenous approach to these and navigating through occluded dural segments to get to the fistula pouch. And as a general rule, if the occlusion is recent, you can probably still tunnel through it quite well. And you might say, well, how do you know if it's recent? Well, we were lucky because we had a pre- and a post-angiogram to know where we are. But as a general rule, if patients become acutely symptomatic from a dural fistula with venous hypertension or a hemorrhage, it's hard to prove. But there's a hypothesis that that might have been because of a very recent complete occlusion of their normal venous outflow that led to a high-risk situation. And so transvenous approaches to dural fistula are possible. And they're extremely safe if you can get there. You're not going to run into this risk of possible cranial nerve injury, ECA to ICA anastomosis, that you're going to be putting things into intracranial circulation. So when I do a transvenous approach, I still do an arterial access to put a diagnostic catheter. That's obviously going to give you your information of where you are and where you're going. You can put it in the external. If you get reasonable pacification just from a CCA injection, I'll just leave my fried French diagnostic there. That's obviously very safe. For venous access, I still come transfemoral through the venous. Some people would puncture the IJ directly. But I come through the femoral vein. I use a six-fringe guide catheter. Obviously, you just navigate up through the right heart. You might want to let your anesthesiologist know that you're going to do that. Or you can just see if you cause any sort of arrhythmia and they look up from their crossword puzzle. Usually tell them you're going through the right heart. So I use a 2.3-fringe micro catheter. And again, the whole point is I'm going to use detachable coils when I'm in a critical area where I know I want really refined placement of the coil, usually in the high-risk cortical venous drainage. Once I get more comfortable that I've secured what I want to secure, and now I just need to pack the hell out of the thing, you can switch to cheaper, pushable coils. But you can only do that in fibered coils. You can only do that if you have a micro catheter that will accommodate it. So I would use a larger micro catheter. And if you're going to navigate through an occluded venous segment, you're probably going to need a stiffer micro wire than you might use for catheterizing your aneurysm. I like the transcend. I use an EX platinum for a lot of my aneurysm work, or synchro too. I like an EX for this work because it's a little more robust. So here's we know what the anatomy used to be. So what I've done is navigated again. There's a guide catheter in the internal jugular at the skull base. There's a micro catheter that I've navigated down into the sigmoid sinus, or what's left of it. And I do a dual run where I inject the arterial catheter, which is going to opacify the fistula. And here's the fistula pouch. And at the same time, I inject my micro catheter, which is in the sinus. And that allows me, I can tell that basically there is the short segment occlusion that occurred in the interval. Right there, that's what's occluded. And so without a whole lot of difficulty, I was able to navigate a stiff micro wire just through this point of occlusion into this fistula pouch. So there's the roadmap for that. And this just shows now the micro catheter. I'm doing a run through the micro catheter, obviously, to show that I'm in the fistula pouch. And I see the isolated cortical venous drainage. This is the lateral view, obviously. This is an AP view. Just to point out, the micro catheter tip is here. The origin of the cortical venous drainage is right there. That is what you want to eliminate. You need to pay close attention to that. And so we navigated the micro catheter into the origin of that and almost treated it like an aneurysm, using small, soft, detachable coils to get into that. And again, coming back on the lateral view, you can see there's the coil mass that we started with and essentially coiled off the entire fistula pouch and was able, I guess I only have one shot left. And the coil mass is here coming back through the fistula pouch to cure the aneurysm. Do I have time for one more? You hijacked me for five minutes. We did. All right. So. But during lunch, we're going to show cases. Do you want to show a case during lunch? Yeah, I'll show it. Yes, I'm happy to. So save that case and do that during the work. So I'll show the case. So I will end just by saying, I think what's already been echoed by a lot of the discussion, which is I'm glad it went that way, is that for these cases, we can talk about technique. How do you navigate to the fistula pouch? How do you use onyx? That's important. But there's little doubt to me that the main technique, the most important thing we do, is spend a lot of time figuring out the anatomy of what we're dealing with. These are very complicated cases. They're beautiful cases, actually. But I agree with what everyone has said, that we need to know what we're doing. I rely a lot on the 3D rotational angiography with the CT overlay to help figure that out. So sorry if I went over a little bit. And thanks for your attention. Thank you.

dural fistulas

anatomy

cerebral angiogram

arterial feeders

venous outflows

transarterial embolization

risks

case studies