Dr. Ray Chu at the 2023 Cedars-Sinai Intracranial Hypotension Conference
Dr. Ray Chu, Associate Professor of Neurosurgery, Cedars-Sinai in Los Angeles, CA, presented a talk on surgical approaches for cervical ventral leaks at the 2023 Cedars-Sinai Intracranial Hypotension Conference on July 9, 2023. The conference was hosted by Cedars-Sinai with generous support from the Spinal CSF Leak Foundation in Kohala Coast, Hawaii.
Transcript
[00:00:12] I have been at Cedars for a little while. I joined after Dr. Schievink. He’s still enjoying the fact that we’re in the same decade of life for the next few months, but in that time between when he trained and the many, many years after that I trained, the amount that neurosurgery embraced spine surgery and some of the instrumentation changed a little bit, so even for him, once in a while he needs a little bit of help with some of these things. So we’ll talk about cervical corpectomy and CSF exploration and fusion, and some of our decision-making in how we pick patients and things like that.
[00:00:45] So clearly this is a very narrow set of groups. It’s really this ventral leak and one where we think there’s a cervical component. As you know, many times there are thoracic ventral leaks. There are more thoracic levels in the spine and maybe a little bit more people can get some of these little spurs and things that we’ve seen over the years.
[00:01:02] There is a little bit of a feeling that we may be doing a little bit less of these corpectomies than before. I feel like Dr. Maya and his colleagues are getting better at identifying some of these leaks in the thoracic spine. Perhaps some of the lateral DSMs are identifying some of the lateral tears, whereas before we may not have been able to see it, we might’ve just picked a cervical level where we suspect something. But it’s still a technique that we use.
[00:01:27] So, of course, there are other anterior cervical options. One option would be an anterior cervical discectomy and fusion. So this is what that can look like at surgery, identifying a level and distraction after you remove disc material to open up a disc space, but you have this very narrow window that’s usually typically several millimeters.
[00:01:44] And many times for these, we have a good idea where the leak may be, but we may not have it down to just several millimeters, and it can be a very limited way to kind of look at things. Of course, there’s posterior options for the cervical spine. For the cervical spine, we’re limited by the capacity of the spinal canal, which is different than thoracic spine.
[00:02:04] We’re also limited by the tolerance of the cervical spinal cord for manipulation or retraction. So, many times a leak that’s in the midline or even within five millimeters of either side is very difficult to get to these posteriorly. And there’s obviously individual patient anatomy that could make that possible, but it’s difficult.
[00:02:21] So, when we choose these people, we have sometimes ways we can’t do something. So once in a while, it can be a contraindication if there’s a prior anterior cervical surgery at other levels that interferes with what you might do or makes the spine too unstable. People that have had some type of other surgery or radiation to the anterior neck could be someone where it’s difficult to have that dissection and approach.
[00:02:43] We have to think about people who have preexisting vocal cord paralysis. Many times that is either a cancer-related or a surgical problem. But clearly, if someone had, if your approach is on the right and someone has a paralyzed left vocal cord, you run the risk of paralyzing both vocal cords, which is not a small issue.
[00:03:00] And then sometimes people have bony disorders where we think they may not be able to have the proper healing from this. So we can’t remove this large volume of bone and not create a fusion. You would destabilize the spine. So it has to be someone who can heal from a fusion and can’t be people with bony disorders or people who are vigorous cigarette smokers and things.
[00:03:19] So here’s one of our examples, 34-year-old with a positional headache and dizziness and tinnitus. There was a thought to be a ventral leak on an MRI in Colorado. And so there was an anterior cervical discectomy infusion. Like I said, that limited exposure, they tried to do some repair there, but did not actually find the leak. And years later with persistent symptoms, she actually had a laminectomy.
[00:03:40] So you’re going from posterior to try and repair the leak, but that did not really secure symptoms either. So this is from the MRI after all these things where she has, you can see the MRI artifact here from the anterior cervical procedure before and the defect here posteriorly from the laminectomy, but there’s still persistent ventral fluid collection. And with DSM, we thought we identified the site of that, but it was actually one level higher.
[00:04:05] And so what we decided to do was pursue a C6 corpectomy. So with that, you need basically two anterior cervical discectomies, the one above and the one below, and there’s a section of at least some of the bone in between. And then we explore for a leak. It requires a C6 strut graft, and I’ll show you and talk about what we use, and then requires plating to stabilize all that and lock that into place.
[00:04:30] So this is with a spacer in place there. There’s a natural groove between the carotid artery and the trachea that you can use to approach the anterior cervical spine. So this is part of removing some of that disc material. And this is one disc level that’s already been performed.
[00:04:43] So you have this bone in between.
[00:04:49] So this is for preparing the actual corpectomy. So for this, we were resecting the bone in between those two disc levels. So this is one of those tricky things that it’s redundant for the neurosurgeons in the audience, and I have run the risk of making the neurologist nauseated, but this is what we have.
[00:05:03] So we often use a spacer that we use for anterior cervical discectomies to both plan how we’re going to plan the discectomy, but also have a template for how much of the bone, anterior cervical bone, to remove. So for the vertebral body, we don’t really want to resect the entire vertebral body side to side.
[00:05:21] One reason for that is the soft tissue injury that could be associated with that. The other is that if we have a template and resect enough such that it’s just a bit bigger than the size of the future strut graft, is that we have more bony surface for fusion. And also we have a little more limitation of rotation of this graft.
[00:05:38] If we have this wide exposure and this thing floating in space, there’s a little less resistance to rotational motion. So we have a template, so we can drill and resect the proper amount. It’s nice to take some bone with a larger instrument like a rongeur, because what this does is it allows you to harvest autologous bone graft.
[00:05:57] So clearly there’s a way to harvest bone graft from the hip, but we try not to do that, and it’s another pain site for patients. And if we can harvest bone graft here, even when Dr. Schievink’s getting impatient, it allows us to use that bone later. Then there’s always a part that you have to wax venous channels in the bone and drill.
[00:06:15] So clearly we drill this to that width, like I was showing you with the template, and expose posterior longitudinal ligament and then dura.
[00:06:27] So here, as we get closer, we use a match head bit that’s actually a little bit dull at the tip and try to resect little bits of bone and reveal ventral dura and ligament. Sometimes we, you know, find some of the fluid into the pseudomembrane, but the point is to try and remove some of this bone and look around and find the actual leak site, which sometimes we can do.
[00:06:51] And this is some of the little spots of dura that are exposed and we just keep exposing more of them until we see what we need to see. This is more work with a curette instrument trying to remove these little pieces. For the cervical leaks, like the thoracic, we use spinal cord monitoring and want to make sure we’re not putting undue stress, although this is really one of the less areas we have stress. We clearly manipulate the spinal cord to touch more for thoracic leaks.
[00:07:21] And part of what we face when we’re doing this, we’re moving these little bits of bone and ligament, we’re always releasing CSF and encountering engorged venous plexuses and have to coagulate some things there.
[00:07:37] So, once we’ve revealed a lot more dura, we can turn it over to Dr. Schievink to explore more and find more of the areas that are an issue. Once in a while, we don’t find an exact leak site, which is what this case is like. We just think we’re very close to it. Once in a while, that could be something that’s a little more lateral that we can expose even with this, but this clearly posterior, we can always expose a little bit more bone if we need to, but it’s hard to get totally lateral around the ventral dura there.
[00:08:06] And so if we can’t really find much, then we use fibrin glue and some dural substitute and try to create the best epidural blood patch of life. This is some of that durograft. There’s, there were comments before about one of these absorbable processes like DuraGen®, right? That’s been kind of our experience that’s not really the best graft.
[00:08:25] We do like this more solid modified bovine collagen Durepair that you can come back on other surgeries in the future and usually there’s something still there, as opposed to DuraGen® that is basically resorbed.
[00:08:40] And sometimes this is a part where I’m out of the room and Dr. Schievink encourages the residents to resect more bone and I have to deal with it when we get back. This is what one of the strut grafts looks like. So there are a few different versions of this. Most of these are surgical grade polymer, like PEEK, polyether ether ketone.
[00:08:55] This tray is a little bit nicer. They have a bunch of ones that are sterilized and you can try different ones and pick the right size. So these are under distraction. And so with that, when you release the distraction with compression, it helps stimulate the bony healing process.
[00:09:11] So these are a little bit nicer than one of the other types. And so we size this and kind of counter tamp it and make sure we have a large ventral surface of dura exposed for you to explore, right? If you get almost three centimeters exposed and then kind of repair things afterwards.
[00:09:27] So for that, to hold everything in place, it requires anterior cervical plating with screws that go into the vertebral bodies above and below. And then there’s usually a locking mechanism to try and prevent screw pullout. The other part about preventing screw pullout is the trajectory of the screws, right?
[00:09:43] So some of this is neurosurgical carpentry, right? If we had four screws all with totally straight trajectories, the pullout strength is a little bit less. But if we have the Top and the bottom screws medialized because of that triangular pathway, there’s more pullout resistance. And if the upper screws aim more cephalad and the bottom screws aim more caudate, again, you have more resistance to pull out.
[00:10:06] And so this is different patient. Actually, I think that was one of these examples of we worry about subsidence, whether something can collapse a little bit. Part of what we try to do is keep a lot of that bony end plate of the level above and below intact. And partly the the footprint of these strut grafts is thicker than it used to be.
[00:10:28] So it doesn’t telescope into the next vertebral body and collapse. So this is what that looked like. And she was that one that had that part ACDF, anterior cervical discectomy and fusion, before, and this sits on top of it, but the graft has markers so you can see and tamp it just into place and really have it, the anatomy the way it restored and really repair that leak. Of course, this is why the podiums are made of wood. I have to knock on wood because I do worry about her a little bit with her multiple cervical surgeries, but she did better.
[00:10:56] Another example is this 43 year old neurologist from 2013 who had these sudden symptoms and subdural hematomas. He had a blood patch. Those things help, but he still didn’t have perfect resolution of symptoms. By the time he saw us, he had less subdural hematoma. But still persistent fluid. So fortunately, we were able to identify the leak site. There is that kind of bony spur, that little spicule that we thought. As you can see, it’s fairly close to midline.
[00:11:29] It’s a little bit difficult of approach from posterior. And so with a corpectomy, we actually were able to find the leak site and directly repair, which is not super common, but it happens, and clearly, you feel very confident when you have that kind of reconstruction.
[00:11:44] This is a different kind of graft, strut graft, that’s maybe a little hard to see, but in here, it’s, this is more like one that has a couple pieces as a centerpiece, and you assemble two caps in either end, so it gets a little tougher if you kind of measure a certain amount and the fit isn’t quite right once you’re in there, you’ve already opened these implants, as opposed to the other one that’s like, a preset tray with multiple sizes where you can try any one you want.
[00:12:07] Also the footprint of this is a little thinner, so it looks like it would telescope into the vertebral body a little bit more than the other one. And this is what that looks like at surgery with the anterior cervical plating and things like that. This is showing the difference in the fluid collection and a little bit of durability response months and months later.
[00:12:30] So for that patient, he improved. In fact, one of the good signs after a surgery like this, if someone does have rebound intracranial hypertension and requires acetazolamide. He did have some numbness and tinglings in the hand, which wasn’t, we thought, directly from surgery, but it was improving. So.
[00:12:45] This is what one of those strut grafts kind of looks like with ridges to kind of hold it in place. There are other ones that are more of an expandable cage. You can put in a place and dial up to the size that you want. But there is some worry about whether you can have a mechanical failure. If we have a failure of that screw system and it collapses, there’s always this worry if you have the same tension as you have a nice graft placed under distraction in the first place. This is someone else’s procedure where they did a partial corpectomy. So you get a little more space to see the ventral dura than a single anterior cervical discectomy infusion. But the worry for this is, you get drilled through that bony end plate of one level and the next level with that cancellous bone, I worry that a strut graft too can collapse and have a problem. So this is not our preferred method of doing it. You also then have less vertebral body available for your anterior plating and screws. It requires a more careful trajectory.
[00:13:47] I definitely have a worry about an older way people used to do this with a metal cage, and these metal cages typically have a point and then can telescope through their vertebral body. And this one too feels like it might have been a partial corpectomy or the screws here— this is not my surgery, it’s from an article— that maybe were not perfectly placed in the vertebral [?] and part of why there may be some motion afterwards.
[00:14:13] So they’re, obviously you get complications with any surgery, there definitely are risks with the spinal cord that we worry about and try to alleviate that with monitoring and increasing our mean arterial pressure during surgery and use of steroids. Any of these surgeries can have a bleeding, infection, or recurrent leak.
[00:14:30] I mentioned some of the things with subsidence of the graft and collapse of the arterial bodies. Adjacent segment disease is when we worry that when we immobilize a segment of the spine that we put extra stress to the level above and below. That’s certainly an issue for this kind of thing. But we hope that if the construct is good and if patients do a little bit of cervical therapy, then maybe we can keep those levels pretty stable.
[00:14:52] So far, patients that have had corpectomy have not really had recurrence, but I think that’s a very carefully selected population, which is in our favor. So ventral leaks can be approached via cervical root. It feels like a corpectomy is a better way to approach it than an anterior cervical discectomy, it allows a lot more area and sometimes we can even find a leak that way.
[00:15:14] Although there is this part where we may be doing a little bit less of them as we can find lateral tears with lateral DSMs. Thank you.