More Videos

Grand Rounds-Technical Challenges in AVM Resection: A Review of Intraoperative Videos

Michael Lawton and Aaron Cohen-Gadol

November 21, 2012


- Hello, ladies and gentlemen, and thank you for joining us. This is the second part of the series for resection of arteriovenous malformations, and we're glad to have with us, Dr. Michael Lawton for the first series. He discussed with us his pearls of technique for resection of this difficult vascular lesions. This second session will be a review of my surgical videos with his critical appraisal of the techniques, and we hope that this dialogue will provide a good learning environment for you all. Michael, thanks again for being with us.

- Aaron, thank you for having me.

- We'll go ahead and start with the disclosures. I have none and neither Dr. Michael Lawton. We have previously discussed the importance of wide exposure for arteriovenous malformation, and that's pretty much for two reasons. Number one, is you often need to dissect the sulci within the normal brain to find the feeders to the arteriovenous malformations. And secondly, if you run, God forbid, into intraoperative bleeding or brain edema, you often need to leave your bone flap out as a form of decompressive craniotomy. I'm gonna go ahead and sort of review the slides pretty quickly. The andrioarchitecture on the arteriovenous malformation is pretty classical. It is wedge-shaped towards the enema, and the veins are often on the surface. A choroidal dissection on the veins, leads us to the arteriovenous malformation and coagulating the feeders one by one through the sulci often is the most expeditious way to disconnect the arteriovenous malformation. However, the feeders, the white matter, can be often very daunting because they don't respond to a bipolar coagulation. The next step is really after very wide exposure is the arachnoidal dissection as we initially talked about, getting the feeders fissurely and going around the arteriovenous malformation in a cylindrical fashion. and as Dr. Spencer has previous mentioned, leaving the veins very superficially intact as much as possible and creating the small channels along the AVM and disconnecting the feeders deep, which can be the most difficult part of the operation may be beneficial as much as the veins you leave intact, you leave the flow within the nidus untouched, and that prevents intraoperative challenges with hemostasis. Any other pearls to this point, Michael, please?

- Yeah, and I think your illustrations are excellent. I think this one here showing the circumferential dissection it's very important to really work all sides of the nidus, not work yourself to one or two or three spots, but really to approach and steadily deepen the dissection. You don't wanna get into deep holes where if you were to get some bleeding, you'd have a tough time sort of figuring out where you are finding that bleeder.

- And here is going further deep in the margin between the arteriovenous malformation and the brain coagulating some of the feeders. It's very interesting. Some of the feeders are very easily responsive to bipolar coagulation, some of them burst, and this really maneuver makes their management very difficult. And really you have to recognize the deeper you get into the white matter, the increased likelihood of these red devils, as you call them, causing a hemostatic problems. What are your pearls when you run into these feeders bursting within your bipolar blades?

- Yeah, so a couple of pearls. One is you wanna have the AVM micro clips ready to go because these thin wall vessels they simply won't cauterize, so you really need to use the clips. And then as you're getting ready to tackle one of these, you wanna make sure you have a good length of the vessel exposed, not just a point of the artery, so that if things burst and start to bleed, that you're not chasing it into the brain tissue as it's bleeding. You wanna have a segment exposed that ready to go and deal with any complication that might come.

- Right. And it seems like as you get away from these whites matter feeders toward more of the normal brain, they form a better wall that is responsible or responsive to a bipolar coagulation. However in that circumstance, you're really removing more normal brain, so it's best to have a good segment of them ready to go. And then putting clips, I use permanent aneurysm clips, the smallest you can find, because they're easier to handle rather than AVM clips. And definitely put the clips on both sides before coagulating these ones. And then as you go deeper, the bipolar coagulation still remains an important part of hemostatic maneuvers. Ultimately, as you have gone around the AVM, you put clips on the veins, starting from the smallest caliber one with clips coagulating them and cutting them. Often, there is arterial feeders very much encasing these veins, and can make hemostasis difficult. Getting into the ventricle can be very important. Even though the AVM may not all the way go to the ependyma, they often get feeders from that as AVMs are formed usually early on in life on the ependymal surface as a wedge and fashioned towards the cortex. Any other thoughts there, Michael?

- Yeah, about the veins, the veins, they're really large in caliber often. Sometimes they are a centimeter or even greater in diameter. So I think the clips are nice when you've got one of these enlarged variceal veins that you're nearing the end and you wanna close it down. If you don't put a clip on and control that, you can often get some very brisk bleeding that comes directly from an adjacent sinus, which can be very brisk. And so it's good to use the clips for that last step of dividing that last vein. One comment about clips in general, the illustration shows you how clips kinda get in the field and can sort of clutter things up. I try and minimize things, the usage of clips, to just those that are either the red devils and I use the mini or micro clips for those. And then finally, if you have a large caliber feeding artery, I think an aneurysm clip is okay for those. The ones in between, I think, at least in my experience, generally do well with cautery. You can sometimes put a temporary clip on, cauterize it, cut it, and then take the clip off. That helps to clean out the field. But yeah, I think you wanna just try and minimize clips. What I find is if there are a lot of clips and you bang against the clips with your instruments going in and out to deeper spots, you can stir up some bleeding.

- Let's go ahead and start a case of a 12-year-old male with history of seizures. This is a rather a classical arteriovenous malformation. It's shed left frontal lobe, not really very close to the speech area. However, we're gonna review the techniques in order to be able to preserve most of the normal brain. Michael, can you tell us what should generally look for on an MRI for arteriovenous malformation.

- Well, the MRI is good for just getting the lay of the land. It shows you where the adjacent eloquence is. It shows you how the veins come to the surface and how they relate to the gyri and sulci. This one's pretty straightforward. There'll be no trouble finding the right spot and identifying where you need to go. And it's got that classic cone shape, so this is a straight from the textbook. Should be very, very simple.

- Obviously an angiogram, we defined the angio as architecture feeders draining veins, really how high flow the arteriovenous malformation is and obviously the opportunity for embolization. Do you routinely embolize all your AVMs as much as possible, or you really say, "You know what, if they can do a good amount, I'm okay. I'm not gonna push my interventional radiologists to hard?"

- I show every AVM case to my team and I let them weigh in as far as what they think they can do safely. We end up embolizing about 2/3 of our AVMs. Ones that are difficult to embolize or ones that it's really not gonna anything to the surgical resection, then you pass.

- Okay.

- So you can see the feeders are, in this case, mostly from AC and the draining vein, obviously located on the surface of the malformation. Here again is additional views showing the exact location of the malformation and the feeders, and we're able to embolize a good portion of this malformation to prevent entropitive bleeding. I think the important factors involved in embolization may not be necessarily just decreased blood loss, the fact that the AVM's slowly changing rather than cellularly and rapidly during an operation. So it add a little some time for the brain to get used to changes in the flow. Do you agree with that concept, Michael?

- Yeah, I think it's particularly important with larger AVMs, the higher grade AVMs, when you've got a lot of arteriovenous shunt flow, and you're gonna be changing that abruptly. But for the lower grade AVMs, I don't think this is as important.

- Let's go ahead and review the video in this case. We use lumbar drain almost routinely on our malformations. Do you use that in order to have a way to decompress the brain since you often don't reach a subarachnoid space expeditiously on cortical AVMs Michael?

- I don't. I very rarely used lumbar drains. I can't think of the last time I used one. It's been quite a while.

- Okay. Here is the generous craniotomy, again, a left frontal craniotomy exposing as much as we can for the normal cortex to be able to identify some of the feeders. The dural opening is really a very classic sort of a hemostasis's key for these micro surgical procedures. You can expose the malformation, the draining vein, often very thickened arachnoid as you mentioned during your previous talk. So we use microsurgical techniques to find the vein and follow the vein towards the malformation. As you can see here, very generous opening of the arachnoid following the vein, obviously preserving it very carefully while you find the nidus of the malformation. Here is what I was discussing briefly before some of the feeders that accompany the draining vein and going through the subarachnoids stage of AVM dissection. We'll go ahead and often go through the pia to be able to expose the nidus. More of the superficial feeders are rather easy to coagulate and disconnect. So here he's going deeper into the malformation and you can readily expose the malformation not necessarily getting neutrenid bleeding, and you can see more of a gliotic margin, and again, preserving as much of the normal brain as possible. Is this the technique, Michael, you usually use, or you try to sort of stay on the brain side of things as you can?

- Well, I think there's always a balance between wanting to stay right at that exact plane between the AVM nidus and the brain, and also wanting to not strain the brain, the risk of deficit versus not getting into the AVM to cause bleeding. So there's a constant back and forth and there're nidus like this one where there's a very clear plane and it's very favorable, very easy to see it. So you're just in the right spot. There, others where it's not quite as defined. You've got a diffused ragged margin and you end up having to make those decisions more step-by-step. But yeah, this is what I would call the parenchymal phase of the dissection, kinda deepening your planes around it and finding those feeders as they plug in.

- Thank you. You can see some of the feeders are very easy. Some of them burst when I try to coagulate this one, and that's where the challenge of the AVM starts. They often bleed pretty quickly, destroy your clean microsurgical field. And sometimes using the micro clips, sometimes you have to follow them more into the normal white matter, where the wall of the artery would alow you to coagulate. And then here he is sort of following the bleeder until a good segment of it is exposed. One pearl that I've heard is that often people keep two bipolars on their surgical field, and one is in a cold saline and they alternate between the two. And this method of always using a very cold tip bipolar is key in terms of avoiding the tips to stick.

- Usually I have a couple pairs. I did an AVM today and I had a a pair of illuminated bipolars for the deep ventricular spot, and then I had a pair of ISO cool bipolars, which is a bigger tip for some of the more superficial areas. And going back and forth is a good idea just to keep those tips fresh and not have any chaff falling on the tips.

- Okay. So it seems like you use a variety of bipolars. You use, icicles for surface ones where you really need raw coagulation without much necessarily precision. And as you go deeper, you may use sort of finer bipolars that not necessarily have as much force or let's say unstickiness, is that correct?

- That's right.

- And here's the last stage of sort of disconnecting the malformation after everything has been disconnected at the depth of the malformation. So this is, again, an intraoperative angiogram of this case, revealing a good resection of the malformation without anymore being . Let's go ahead and present a slightly more technically challenging case of a 17-year-old female with seizures and a right frontal arteriovenous malformation. As you can see, this malformation is sort of going through more of the critical white matter, deep in the posterior frontal lobe. Also gets close to the middle cerebral artery branches. Any other, and also a deep draining vein, as you can see here close to the brain stem. Anything else that you would say is important to know based on review of these MRIs, Michael?

- Well, not really. This is a formidable AVM. It's probably a Spencer Martin 4. It's a greater than three centimeters. It's got the venous drainage and close enough on that backside to some eloquent areas that it's not a straightforward one, like the first one.

- Okay. And here is after embolization. You can see they did a relatively good job on the non-eloquent part, but really the part that surgeons need most help where the small feeders from the white matter feeding this daunting AVM. Unfortunately there is no much help from embolization and often these are small feeders from the internal carotid artery that are too difficult to embolize. Again, other views of sort of this AVM. And as you very well mentioned, Michael, this does go back close to the area of the cognates, potentially hypothalamus. And you have to be careful as you get close to the more posterior part of this malformation. Let's go ahead and straight to the video. Right temporal lobe, right frontal lobe. Covering the brain with a piece of Telfa to avoid ,any injury from the intensity of the light. Opening the Sylvian fissure very widely to be able to identify the MCA branches and all the important vessel early on to keep them out of harm's way. Here is again sphenoid wing draining vein from the malformation and continuing our dissection until the AVM espederic exposed. Any other pearls you would keep in mind during this AVM, Michael?

- I think it's good to spend some time at this stage of the operation, just opening up the fishers and getting a good survey of the AVM. You don't wanna rush into the AVM resection before you've really exposed all the subarachnoids cisterns and fishers, and really got an idea of where those vessels are coming around from.

- Okay. I agree. And here's the internal carotid artery relatively enlarged on this side. And the AVM is essentially under the retractor and you wanna avoid aggressive retraction. And again, generous opening of arachnoid membrane, which is often very easy around the AVMs, because they sorta create these CSS spaces around them due to their constant pulsation. This is also important because you would like to find the feeders off of the carotid early on and coagulate them, and most importantly, differentiate them from normal feeders or empassage feeders going to them, right? So here again is the carotid opening the membranes very widely, and momentarily, we're gonna go ahead and sorta move over the optic nerve skeletize the carotid artery to be able to see where the feeders are coming from. Here's moving across the chiasm to the other side. And you can see some of the feeders on the surface of the AVM. We're positioning our retractor, and now you can see those little tiny feeders from the carotid artery to the malformation, again, being optic kaizen. We encounter a little bit of bleeding in this part of the operation, and as you can see, we just use a piece of cotton soaked in thrombin to be able to avoid necessary aggressive coagulation on these important vessels. After some of the feeders are coagulated, we come around the cortex. This part of the operation's relatively easy, and we, again, disconnect all the cortical feeders from that malformation. Anything else you would have done differently here, Michael?

- No. Just about your exposure, is that a pterional approach or an orbitozygomatic? I can't tell.

- Thank you. It is a is that a pterional approach. Great question. I should have mentioned that. Very flat to the roof of the orbit. Would you use a orbitozygomatic for this case?

- Yeah, I liked the orbitozygomatic for these AVMs, just because as you go over the top, you end up wanting to look under the brain more and more and getting that orbital rim off gives you a nice view over the top.

- Okay. That's great. And I think here we're just sorta continue to disconnect the malformation from the surrounding area. Here is the A1 artery. You can see is potentially some feeders to the malformation. And again, staying right on the part of the brain that is close to the malformation to preserve any of the empassage vessels. This is getting more to the posterior part of malformation. This is relatively lateral at this juncture, but as we get more medial, we're gonna come to a more eloquent part of the malformation. Maybe you would have probably stayed a little bit closer to the malformation at this stage. Is that correct, Michael?

- I think it's fine. You have to look at where your critical claims are, and for this one, the critical plane is the one posteriorly. So you have a little latitude and getting into some brain on the other sides.

- Okay.

- But definitely on the posterior side, you wanna hug it really tight so that you don't stray into the caudate head or into the other basal ganglia territory.

- Okay. And I think that's the area you were just talking about. The more posterior part were closer to the malformation. You saw a minute ago, we used thrombin soaked pieces of cotton to cover the areas that are close to the eloquent cortex in order to obtain hemostasis without aggressive electrocoagulation. Here is one of the draining veins against some of the feeding arteries that often hug the draining vein. And as you try to manipulate them, often brisk bleeding can occur. And it's important not to be too aggressively coagulating the bank early on, and rather keep every vein relatively intact. And this is, I've been going posteriorly deep that we saw during the initial review of the images. Again, moving now more posterior part of the malformation, it's often, I think, important to change places. When you have bleeding in one area, try to get hemostasis moved to another area. Gives the surgeon a little bit of rest, first of all, from getting hemostasis, and also gives the brain some time to recover. Here is again, trying to get hemostasis, and as you can see, some of the veins can pop. And hopefully this is at the end of your operation where you coagulate the vein without necessarily causing any radical change in the nidus of the malformation that still owns a numerous number of feeders.

- These AVMs are challenging. We didn't get any clips of the olfactory apparatus, but they often lie right on top of the olfactory track embalming. You wanna try and preserve that as best as possible. The vasculature on the underside to the A1 and the A2 junction is gonna be right against the nidus. Those arteries as well as need to be swept away and protected as you go around those deep Plains. But no, I think this is a pretty classic illustration of these inferior or basal frontal AVM.

- As you mentioned, when you get more medially, you wanna carefully preserve the empassage arteries space appeal in order be able to identify the arteries that are empassage and protect them from harms way on the bipolar and microinsection. Again, this is another view of another draining vein going through the white matter. I'm gonna go ahead and move a little bit faster along here and show ultimately where the malformation was removed. And I'll use thrombin irrigation at the end just to obtain some hemostasis. Often works well. Forms a layer of clot over the white matter where the oozing often occurs, Intraoperative angiogram revealing a good resection of the malformation. I know you don't routinely perform intraoperative angiogram, Michael, for your AVMs. Do you use ICG to confirm a resection if there's any questions or you just wait for your postoperative angiogram?

- Yeah, I generally just use the anatomical cues as you go through the resection. So your vein will become nice and dark, and that will tell you that you've encircled nidus and that there's no arterial supply left. If there's any redness, then it means that you've missed something. I think that's telling us anything in the angiogram. The ICG is not as helpful with AVMs as it is, obviously, with aneurysms, so I use it on a very limited basis. But once in a while, it can help you figure out whether a vessel in question is an artery or vein. It can also give you a clue to where that arterialized vein is on the surface if you can't quite find it. But yeah, I really rely on the anatomy more than anything else.

- Let's go on and enter the world of pial AVMs. 26-year-old male with intraventricular hemorrhage and was subsequently diagnosed with a left sided pontine PLA AVM. These are often around the root entry zone of the fifth nerve On the MRI, you can see the tangle of the vessels that on the surface or the pial of the brainstem here. Any thoughts about looking at the MRI in these pial AVMs?

- Well, you don't see the nidus very well here. What you just get a hint of is a little bit of flow voids in the cerebellopontine angle. So at this stage, just from these images, it's not entirely clear what the anatomy will be like, but it does look favorable. It looks like it's lateral to the ponds, really more in the flakiness of the cerebellum, which is, again, a more favorable place for the AVM than actually in the ponds.

- Here's the angiogram revealing the location of the malformation, the draining vein. And again, you can see the malformation here with the draining vein. One of my very favorite patients. Opening the closer posterior fossa generously, you can see the transfer sinus, sigmoid sinus, skeletonized dural and along the sinuses. I used a lumbar drain for this patient to be able to go around the cerebellum without struggling too much and necessarily avulsing a vein. This patient underwent surgery in a delayed fashion after this hemorrhage about a month or so afterwards, when he underwent a slight amount of embolization. When you do your posterior fossa craniotomies or craniectomies, Micheal, do you necessarily use a different technique, or this is pretty standard for you as well.

- No, I do it just like you've shown. I do probably expose a little bit more down towards the jugular bulb just so that I can get down to the cisterna magna quickly, take off some CSF and relax things in case the cerebellum pushes out at me as I'm opening. But no, I like what you've done here, exposing the sigmoid and transfers sinuses widely rather than shrinking away from those. I think it's nice to really get those fully exposed.

- Okay. We use a piece of rubber dam underneath Cottonoid to be able to slide sorta over the cerebellum without necessarily causing a friction between the Cottonoid and injuring the cortex of the cerebellum. These are very, very standard techniques. Opening the arachnoid membranes early on. Finding the seven and eight cranial nerves. Always staying a little bit more superiorly rather than ending up right on the seven and eight complex. And most importantly, avoiding aggressive retraction of the cerebellum to unnecessarily having immature rupture of the AVM. Again, a wide exposure and arachnoidal dissection of the membranes to avoid retraction on the vestibulocochlear nerve. Here you can see a normal vein, the petrosal vein, a very dark. So we went ahead and sacrificed this very early on to be able to attract the cerebellum, open up the arachnoid membranes, here is the fifth nerve, and the malformation we thought would lie right at the root entry zone of the fifth cranial nerve. Any other pearls in your arachnoidal dissection besides using micro scissors and sharp dissection?

- No, I think what you're doing is great. I don't use a retractor back here just because when you pull back on the cerebellum, you can put some tension on the eighth nerve in particular and you can compromise sewing. So I'm very much against using any fixed tractors and just using your sucker and your scissors to sort of lay into the brain a little bit for some dynamic retraction.

- That's what we have done recently as well. I think this is a video from a couple of years ago, and I wholeheartedly agree with you that there's no need to use fixed retractors on posterior fossa because a wide arachnoidical dissection ample amount CSF egress really provides a space to be able to use the dynamic retraction of the suction tip. This is the edge of the tentorium here. This is some of the glue that was performed. You can see the arterialized vein. I forgot to mention that there are actually two veins here. One of them was not arterialized. Now we took early on, and this one is actually very much arterialized that was kept intact. You can see the retractor was used just because this AVM is very much on the surface of the pia and located at the root entry zone of the trigeminal nerve. And keeping the vein intact, following the van using arachnoidal dissection and obviously keeping electrocoagulation to a minimal. Besides following the vein, Michael, getting into the malformation, anything else that's important in these brain stem pial AVM's?

- Yeah, well, first you wanna make sure that you've selected an AVM that's right on the pia and not in the brainstem itself. So this one does look like a pial AVM, and it really respects the fifth nerve, which is also nice. Sometimes you've got feeding arteries that are all in and around the fifth nerve, which can make it very difficult to separate, but this one looks, at least so far it looks like it's gonna be clean.

- Okay. And here is against some of the bleeding from the shield feeders to the malformation. And I think there's really nothing here you can do besides being patient and using the bipolar on low settings and dissect the AVM from the pial surface and hope that during this process you are not gonna create too much numbness. As long as you stay away from the root entry zone of the nerve, I think this is a relatively safe area to be able to dissect without causing disturbing numbness of the face. You can appreciate the tiny little feeders on the surface of the brain stem using the bipolar on a no regulation scale and using microdissection. Obviously, bleeding is unavoidable from these pial surfaces. How do you console these patients? Would you say these are now more and more becoming radiosurgery candidates, or still is there a role for these pial AVMs for microsurgery?

- Yeah, well, it's always a tough choice. I think that these are small, but on the surface, they're really decent radiosurgical targets. So I think it's a good option for many of these patients. In the patients that have presented with hemorrhage, then I think that changes the coagulates a little bit and for those, I tend to be a little bit more surgically minded. And other factors are age and just the typical size and Minas anatomy type things that go into the gradings.

- Here, again, sort of running into a little bit of more bleeding as we go on, and I think nothing replaces the patience and avoidance of aggressive coagulation on the pial surface of the brain stem, and so using coagulation judiciously, to be able to disconnect the malformation. We'll go ahead and sort of continue a little further with the malformation. You can see we're dissecting the malformation from the surface. We did have to coagulate the pial surfaces a little bit more aggressively here. Often the feeders here are from the superior cerebellar artery, and you wanna keep disconnecting that until a little bit later in the operation, because often it's deep with as comparison to the malformation. And also you wanna make sure you preserve all of the very critical empassage vessels. And sort of continuing along the pia, coagulating things very carefully. If there is a vessel you can see at the depth of the AVM. That may not be clearly an empassage vessel versus a purely AVM vessel. Sort of keeping it intact until the end. And then here we go further on in this operation. You can see the ABM is almost disconnected. Again, this is the surface of the tentorium. This is the AVM almost completely disconnected. This is one of the branches of the superior cerebellar artery that's being coagulated and cut knowing that all of the other empassage vessels have carefully protected. Fourth nerve at the edge of the tentorium could be sort of very adherent and it has to be also carefully protected. You can see one of the draining veins is now pretty much blue, no longer arterialized, anatomically confirming that the malformation has been disconnected.

- Yeah, that's a nice job, Aaron, that angiogram at the end. The only comment I would make is, these AVMs, it's nice to start as you did along the nerve and peel things away so that you preserve the nerve in that entry zone. I also think, so that you don't get sort of caught going in only one plane in one direction, I always like to get into the backside a little bit, sorta define where that back plane is. And again, try to surround on all sides the nidus and go back and forth. It really helps to, at least in my mind, to kinda form an impression of where the nidus is sitting and also can underneath the AVM so it almost scoops out of the cerebellum and lifts up. It's another nice way to get at these CP angle AVMs.

- Thank you. Let's go ahead to the last two cases. The first one is another pial AVM. This one is on the medial temporal lobe, a deeper area, more difficult to handle. 52-year-old female with a left uncal pial arteriovenous malformation. She initially presented with hemorrhage, as you can appreciate on the MRI, and on CTA, she had these tangle of abnormal vessels with the draining vein on the CT angiogram. As we were preparing her for having surgery, unfortunately she had another hemorrhage intraventicularly, but she recovered very well. And as you can see on the angiogram, the malformation, again, get it's feeders from the internal carotid artery, nothing really special to necessarily embolize. Before we go to the video, any other information you would extract from these images or pearls in terms of approaching this issue?

- Tough to really read these two angiographic images, but typically, or classically, they are fed by antrichoroidal And sometimes on that medial temporal lobe, you get those posterior temporal arteries of the PCA. This one's very forward, so it's probably not involved in this one, but choroidal is a big vessel feeding this and very tough artery to embolize. So you're kinda on your own in terms of taking this out without any endovascular help.

- And since its non-uncus and tip part of the hippocampus. I absolutely agree. Here we expect a PCA would feed that. Unfortunately, I don't have the angiogram here, but you're absolutely right. PCA was a big feeder for this uncal AVM. So to orient our colleagues here, you can see this is the right frontal lobe, right temporal lobe opening up the arachnoid membrane. Previous history of hemorrhage staining the brain. Here's the optic nerve, carotid artery, MCA, and, again, a wide exposure of the fissure. So on the left side, you wanna avoid removing as much of the lateral neocortex as much as possible. Here again is the posterior part of the carotid artery, identifying some of the antrichoroidal branches in order to prevent your sacrifice during disconnection of the malformation. Here's the third nerve, really a beautiful anatomy of the carotid optic sisters. You can also appreciate a draining vein over the surface of the uncus So we're using these at the transylvian corridor to expose the uncus and remove it. How would you have approached this, Michael, using orbitozygomatic, using the tangential technique you have described? I remember reading an article about it, or what are the pearls here, please?

- Yeah, so I love these AVMs. They're really nice. They're really just like an approach for the basilar apex. I'd do an orbitozygomatic for these. I get right along the third nerve and the tentorial incisura and just follow the choroidal artery straight back. 'Cause again, that's the major supply, that's the trajectory that the orbitozygomatic gives you and you just skeletonize or peel away those little feeders from the choroidal as they dive into the nidus and it gives you a perfect shot.

- So you would have not necessarily come as much lateral trajectory, you would have come from front to back, removing the roof of the orbit and using a tangential trajectory. Is that correct?

- Yes. But just like you're doing it, I can't tell. Is this the orbitozygomatic or just straight pterional?

- This is straight pterional, Micheal.

- Yeah. So they're the same trajectories. Orbitozygomatic just gives you a little bit more opening, a little bit more illumination down there, and you can, I think, mobilize the temporal lobe a little bit more aggressively because it's not as locked in by that bone of the interior and anterior temporal bone.

- And I think that you saw a moment ago, skeletonizing the third nerve in expectation for manipulating some of the feeders from the PCA along the medial temporal lobe, Obviously you don't wanna dissect around the third nerve as there is risk bleeding. You wanna do all your dissection early on, identify everything, make sure all the important structures are sorta known where they are, rather than saying, "Well, there he was." One of my mentors said it's better to say a thousand times, there it is, and be wrong, better than once say, there it was, and be right. And I think AVM surgery with brisk bleeding, surgeon becoming very nervous, it's best to everything safely, identify all the structures, and then continue with the dissection. Here is the draining vein on the surface of the uncus. Here is entering the uncus itself. You can see some of the previous hemorrhage in different stages of evolution. Here is moving along the anterior tip of the temporal lobe at all times, keeping this fissural draining veins intact, and now we're on the dura over the medial temporal lobe. How would you maneuver your dissection here after you tangential exposure, Michael?

- Well, the tangential exposure gives you your feeding arteries right off the bat. So you take care of the choroidal feeders. You can get back to the PCA and pick off those feeders off of the PCA, moving along the tentorial incisura. And once you've dearterialize things, then you can start to skive in like you're doing for that lateral plane and work your way behind it. It does make it a little challenging to see the posterior border, because it's all the way in the back and you often don't get a good view of that anatomy, including the basal vein, which is typically what drains this until the very end.

- Okay. So you believe that a tangential view would give you early exposure of the feeders?

- If you go from a more lateral trajectory, like many people would do, you come on those feeders at the very end, which I think is not as favorable.

- Okay, thank you. And here you can see the MCA, and I think this actually is one of the arterial branches. Now you can see the vein is a little bit more medial. And try to protect the arterial branch, which is empassage here. The AVM is underneath one of the anterior temporal branches. and here is these really unknowing bleeding from the whites matter areas. And that's why it's so important to keep the whole area exposed, making sure everything is. Put your suction over its bleeding, keeps the field dry. So coagulate the point of bleeding rather than just haphazardly coagulate and hope that it would work. And use high magnification to keep the normal surrounding structure out of harm's way of the bipolar blades. So we're moving a little bit more, again, medially from lateral to medial up here. We initially did somewhat of a tangential view, Michael. And as we go more medially, we're gonna probably run into some of the feeders from the PCA. Third nerve is okay and a little more medial. And you're working on the two sides of one of the anterior temporal branches. It's on the left temporal lobe, the dominant temporal lobe, so it's very valuable piece of cortex that has to be carefully protected. And I think this is really using that tangential trajectory you were talking about. And if there is bleeding deep, rather than being aggressive close to the cranial nerves, trying to use a piece of thrombin soaked cotton or gel form to avoid snow ooze. And here is, again, moving more posterior in a tangential view, and you can see some of the feeders from the posterior cerebral artery. And again, exercising caution in coagulating and cutting them, making sure they're not empassage vessel. And this is the edge of the tentorium that we're following along the uncus, third nerve carotid.

- That tentorium is a very nice guide, 'cause as you pick up that tentorium and follow it back, you're gonna get every PCA branch that comes over the edge and gets to the nidus. So it's a really helpful landmark to follow.

- And I think here is, again, the posterior part, the arachnoid over the PCA that was left intact and over the brainstem. And at all times trying to keep the empassage vessels intact, the third nerve carotid artery. And again, the depth of our dissection, keeping as much possible that I can always attack. Using a Previn soaked cotton to make sure the vessels that have been most manipulated are not undergoing too much vasospasm. So relatively straightforward AVM and the post off of the angiogram reveals a good resection. This is our last case, Michael, and this is something where we really went into trouble with AVM. I thought we should review in order to get familiarized with complications. 22-year-old female with headaches and a left parietal malformation. This one is obviously a deep. Gets close to endyma. A very large size AVM. Very high flow, as you'll see momentarily. Here is the angiogram feeding from ACA and MCA. Very cone-shaped. We did employ embolization, but really didn't do too much here. When you look at this case, what is it that most important comes to your mind.

- Well, you're helped out by a couple of things. If you go back a slide, what I didn't see was any feeders from the intrachoroidal artery. The intrachoroidal is right here, very small. Doesn't look like it connects to the nidus, which is a huge help, because then if that's feeding the AVM, it often means that you've got some life to the AVM all the way down to the ventricle. And of course, bleeding in the ventricle can be really difficult to manage. So that's not an issue. And the other thing that's absent is any large lenticular strides that go through the deep white matter to complicate things. So I think the absence of those two feeding, on the flip side, you've got some major supply from two territories, ACA and MCA. We haven't seen the PCA injection yet, but you might even have that as well. But you got a long AVM, goes all the way to the ventricle. These are tough because working down those long corridors makes it hard to see at the end.

- That's exactly like working down in the corridor for this patient was really a very difficult task and everything started well. This is a prial craniotomy with a midline located right here with this precise . Here you can see sutures in the dura. Started early with a subarachnoids section that we very well described, and then continued through the pial dissection, part of this AVM. And these AVMs can be often very tasking and very long. Do you have a threshold when you would say, "You know what? I have worked enough, it's going okay. I'm gonna go ahead just stages because I'm tired." Or do you just potentially work through the night and after midnights, Michael? Can you tell us some guidelines about that?

- Yeah, and as soon as it gets to be around seven o'clock, if I realize that I've got a long way to go, then I usually stop at that point. But I pace myself and if you give yourself 12 hours to tackle an AVM and you still have a long way to go, then I think it's best to stop and stage it.

- I rarely have seen AVMs that require 12 hours straight, unless the efficiency can be an issue because most AVMs, if you manage your plannings and move ahead relatively expeditiously, you should be able to remove them in a relatively good timing before 7:00 PM. Don't you agree?

- Yeah, I agree. I can't remember the last time I was spending more than 12 hours on an AVM. When things go bad, it usually makes things actually finish faster. If you're working that long, you have to wonder whether you should have taken that on in the first place.

- I think that's very well said. Oh, was it an operative AVM that you were tackling? You see my clamp squeezing the vein here. I thought maybe I disconnected good amount of these feeders, but you see how the AVM saw and started swelling. I know this is not necessarily a very safe technique, but it gives you an idea about how much of the AVM is still feeding. Do you use that technique just to evaluate how much feeding coming through AVM nidus still?

- I don't. I think color is so valuable. You know exactly where you stand just by looking at the color of those veins. And as long as they're red, you've got something going. And it doesn't take much to keep red, but when you've gotten around it, you'll see the color change. It gets nice and purple and then a nice and dark.

- Okay. Here is this part of AVM surgery that I just don't have a good solution for. You do good coagulation, you're staying within the white matter. Suddenly everything looks clean and dry. You go to another location, and some of the brain starts bleeding spontaneously. Is that a normal pressure breakthrough or is it just, again, it's just brain that is not having a good time changing its full patterns so quickly? What's the best way to do it when you run into those circumstances?

- Well, if you've got an AVM that just starts to go bad on you, then it usually means that either there's one little artery that's gotten away from you and you gotta hunt it down and control it. And I'd say that in this case right now, we're probably seeing just one little vessel that's gotten away. That's a very controllable situation if you just kinda keep your sucker in there, keep focused on that one spot, and just chase it down. When the whole nidus changes and things start to pop and rupture and swell, then that usually means that you've made a mistake. You have to ask yourself, if you've come across a vein inadvertently, did that artery that you thought you were coagulating, was that really just a vein in disguise? Those are the things that I start thinking. And the other thing, sometimes your planes that you selected are wrong. Perhaps you've tried to stay too close to the AVM and you've come across some feeders that are really part of the AVM rather than in front of it. So you have to kinda reevaluate your margins. And if things are getting away from you, sometimes you have to just take a quantum leap back from the AVM, get a little bit more into the brain, and redefine a border that's clearly away from the nidus.

- I think that's a great point. What happened is in this case, and I went back on the video, as you can see, we got into problem. Here is me getting into the ventricles. So excited that I'm close to the opening to the ventricle. But the moment we got into the area of the ventricle suddenly the whole brain started swelling. Again, I didn't see any arterial feeder or any vein that was mistaken for each other, and thought the brain was sort of becoming, at this point, very much non-compliant. You can see there is just oozing everywhere, Michael. Brain is very much swollen. Not really a good way to obtain hemostasis. It's just oozing everywhere. You can see the brain is coming a little bit above dura, the AVM is almost above the dura. We went ahead and closed the case, I'm gonna show the postoperative scans momentarily, came back about a week later. The AVM was much easier to handle. You can see the bleeding from the first surgery. Patient didn't have any ontour consequences from the first surgery, but again, these AVMs can be really challenging as they go bad and becomes really a commando operations as you very well mentioned during the first talk. If things start going bad, you have to sometimes take out of the AVM very quickly. In this situation, I felt the brain was too swollen. We did not proceed with removing it, hoping that the whole brain would calm down. Again, this is the second part. It's really easy when you come back a week or two earlier. It's as if this is an entirely different operation. The brain has relaxed. Everything is very much calm. How often do you stage AVMs and run into this kind of problems would you say in your large practice?

- Well, I don't know, it's not very common. I think a couple of reactions to this case when you're in the ventricle or right at the ependymal layer, you have to be super vigilant for bleeding, because what can happen is even one small bleeder can drop into the ventricle out of your view and all of a sudden you've got an intraventricular hemorrhage that you can't see and all it does is it just pushes the brain out towards you. So you get this externally herniating brain from a source you can't see. So whenever I get to the ventricle, rather than just have that kind of birdhouse view into a small opening, you wanna really widen that, get widely into the ventricle so that you can see the choroid, you can see the ependymal widely. You can find the colloidal vessels and keep everything well-visualized and under control. I view that ventricular spot is just a red alert where you can really get in trouble. If things go bad, I think stopping and staging is fine. The problem is that if you've got some uncontrolled bleeding, it's really hard to leave that sometimes. And I think you have to make a decision whether you're gonna get away with just closing and staging or whether you just have to do a commando resection, where you just fasten your seat belt, buckle up, and just get that thing out. And I think more often than not, when I'm in trouble, I go for growth and I try and get it out.

- Okay. I think that's a great point, and that's I think how the things got out of control. This is the post-operative CT scan after the first stage. You can see excess amount of bleeding, just like you mentioned intraventricular hemorrhage. So now discussing this case with you, I 100% agree that at the ependymal layer things got out of control. I wasn't able to adequately obtain hemostasis, and that is evident by the excessive amount of ventricular hemorrhage. She ultimately had a PVD placed. Did really well. We did an angiogram about a week and a half later, and the whole AVM was lost. But we still felt that to avoid any chance of recanalization it's best to go back. We had to put the bone flap back anyways, and she did well. But again, AVM's are challenging cases. They require ample amount of patience, technical expertise, and remembering that being efficient is important, but being safe is more important. Michael, I wanna thank you for your expert pearls advices, and I've really enjoyed working with you on these two webinars.

- Well, thank you, Aaron. Congratulations on some really tough cases and good results and a nice technique. Excellent.

- Thanks. Thanks, Michael.

Please login to post a comment.

You are leaving the Neurosurgical Atlas

Full 3D Models are available outside the Neurosurgical Atlas through an Atlas Meditech subscription.

You can make a difference: donate now. The Neurosurgical Atlas depends almost entirely on your donations: donate now.