Dr. Marcus Stoodley at the 2023 Cedars-Sinai Intracranial Hypotension Conference
Dr. Marcus Stoodley, Professor and Head of Neurosurgery at Macquarie University Hospital in Australia, presented this talk on syringomyelia pathophysiology, imaging, and management 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] When Wouter and I started talking about organizing this conference two years ago, the original plan was for this to be half SIH, half syringomyelia, and Wouter said, you do the first draft program and then send it to me, and I did that. And I had half syrinx, half SIH, and then the revised program came back the next day and there was one syrinx talk left. So, so this is it.
[00:00:33] So I’ll go through these things, some pathology and theories of pathogenesis, CSF physiology, and management. Syringomyelia is always associated with something else. It’s not really a diagnosis on its own, can be divided into craniocervical junction and spine, congenital and acquired.
[00:00:49] And this is not a complete list, but there are many associated conditions. But basically you can bring it down to these four things. So the craniocervical junction abnormalities, typically Chiari; spinal cord injury; arachnoiditis; and then the so-called idiopathic. But I think really that just means we haven’t been able to find the associated condition.
[00:01:09] If you look at even modern neurology textbooks, this is the kind of description of syringomyelia that paints a picture of this kind of what they say is cavitation and accumulation of fluid, that paints a picture of this being a very passive process with loss of spinal cord tissue. So normal spinal cord on the left, and a pathological specimen of syringomyelia on the right.
[00:01:30] But of course, we know that is really nothing like the truth. The truth is that problematic syringomyelia, at least, is a very tense high pressure cystic cavity within the cord, and not the picture that you see in the pathological specimen. And we see that both in our dynamic imaging and intraoperatively. So the fluid that we’re seeing there on the right hand side is all syrinx fluid that comes out after opening the cord. It doesn’t always look like that but that I think is illustrative of the issue.
[00:02:00] We know that we can categorize syrinx cavities according to the location. So sometimes they’re in the central canal, in communication with the fourth ventricle. More commonly, they’re in the central canal, not in communication with the fourth ventricle. And sometimes they’re outside the central canal, typically in association with spinal cord injury or arachnoiditis. So the Chiari is the middle type, spina bifida is the first type.
[00:02:27] So in the non-communicating canalicular type, which is the most commonly, most common one that we see in association with Chiari malformation, there is generally not a breach of the ependymal lining. And so this is akin to hydrocephalus. There’s actually no parenchymal damage. And so patients with this type of syrinx are actually asymptomatic, even though there might be an extremely large syrinx, as in this case.
[00:02:51] They only become symptomatic when there is a breach of the ependymal lining and then damage of the spinal cord parenchyma. So, as in this case, where there is the main expansion of the central canal and then a separate cavity where there’s been breach of the ependymal lining.
[00:03:06] And so it’s not the size of the syrinx that matters. It’s the spinal cord damage that matters. So even a syrinx that’s really quite small is very symptomatic because it’s in the dorsal horn, it’s not in the central canal, it’s damaging the cord tissue and damaging the connections.
[00:03:21] We’re taught in medical school that the reason that patients get this cape-like sense distribution is because of the damage to the crossing fibers in the center of the cord. That’s actually not the case. So in, in syringomyelia, the reason that patients get the cape-like dissociated sensory, suspended sensory loss is because of the dissection into the dorsal horns. So commonly it’s actually unilateral rather than bilateral. If it is bilateral, it means that there’s been bilateral dissection. It’s actually dissection into the cord. It’s not actually getting the crossing spinothalamic fibers in the center of the cord.
[00:03:59] So turning to the pathological theories that can be grouped into hydrodynamic and others, and basically comes down to a question of where does the fluid come from? Does it come from the fourth ventricle? Is it coming across the cord tissue from the spinal subarachnoid space? Or is there some other mechanism?
[00:04:16] And I’m sure we’ve all heard of the theories of Gardner, where there’s said to be transmission of cardiac pulsation to the CSF from the fourth ventricle down the central canal into the syrinx cavity. I showed some examples of that. But that’s actually the least common type.
[00:04:33] Williams says it was actually the coughing caused the CSF to go from the spine into the posterior fossa, the tonsils would then plug that posterior fossa, and CSF could only get back into the spine by going down the central canal.
[00:04:46] I think we’ve got pretty good evidence now that both of those theories cannot possibly apply because the spinal central canal is really tiny, it’s histologically only just visible, and the distance from the fourth ventricle to most Chiari associated syrinx cavities is too far for that to be possible. Except for those cases that I alluded to yesterday where there is usually a small syrinx cavity close to the fourth ventricle, and we see that intraoperatively with ultrasound that there are communications from the fourth ventricle to that type of syrinx cavity.
[00:05:19] Does CSF get across the cord parenchyma from the spinal subarachnoid space? Is it by coughing and sneezing as Ball and Dayan would tell us? Or by the tonsils acting as a piston and increasing the CSF pressure in the subarachnoid space, pushing CSF across the cord tissue, as Oldfield would tell us?
[00:05:37] But the issue there, of course, is that you can’t get fluid expanding inside the cord from external pressure alone, because increasing the external pressure will only serve to squash the cord. It can’t possibly be that mechanism. So it’s really more complicated than that. And this is an example, a case example, I think illustrates that it’s not just a simple pressure transmission from the subarachnoid space into the core. Because here I’ve treated this patient with a subarachnoid, a syrinx to subarachnoid shunt. And if it was a simple pressure transmission, that would actually make the syrinx get bigger rather than larger, right? So, that’s evidence against that.
[00:06:12] There are some theories that suggest that obstruction of venous outflow causes increased venous pressure and reduced CSF absorption; that there’s a Venturi effect in some theories where there’s increased CSF velocity outside the cord. None of those really stand up to to deep scrutiny.
[00:06:29] One of the questions we have to ask, though: is there always an abnormality in the subarachnoid space? Because even in Chiari malformation, there’s technically an abnormality in the subarachnoid space of the cranial cervical junction, but we know that in cases of tethered cord, for example, the spinal subarachnoid space is completely normal and you can get a holocord syrinx just like that.
[00:06:50] Some cases we really do struggle to find the underlying cause and that would be potentially called idiopathic. So I’d put it to you that we really don’t know the underlying pathophysiology, and there’s certainly no explanation that fits all of the different types of case. But it, I would say that it is not just a simple transmission of pressure from the subarachnoid space into the cord.
[00:07:14] I spoke yesterday about CSF physiology so I’ll just remind you that CSF is produced and absorbed everywhere, that there is now this glymphatic theory. And there’s this nice two photon microscopy work that the Nethergard group have done showing particle tracking in the space, CSF space around blood vessels, showing that when you increase blood pressure, the velocity of those particles actually decreases, and that it is, although it is a cardiac-driven flow, it’s obviously complex.
[00:07:47] I think we’ve shown fairly convincingly that, in our own experimental work, the pathway for fluid flow into syrinx cavities is actually via the perivascular spaces. And we’ve shown that there is a normal perivascular flow from the spinal subarachnoid space through the parenchyma into the spinal cord central canal. And that in animal models, at least, of both canalicular and extra canalicular syrinx cavities, that flow continues. And so that’s likely to be at least the pathway for fluid to get into syrinx cavities, but it doesn’t explain the force that drives that flow.
[00:08:21] Just in summary, though perivascular spaces are the anatomical pathway, the driving force is not clear, but it seems like in the majority of cases, there is some form of subarachnoid obstruction, and we’ve shown at least experimentally that does reduce CSF flow in the subarachnoid space and increases CSF flow into the cord. We’re investigating whether there are phase differences with obstruction of pulse transmission across subarachnoid space abnormalities that might be responsible for the driving force.
[00:08:51] And how does that translate to the way we manage patients with syringomyelia? I’ll show you some examples that, that touches on that. But basically, if we look at the Chiari patients, it’s posterior fossa decompression. Some people in the past have recommended syrinx shunts as the primary treatment for patients presenting with symptomatic syrinx. I don’t think that’s really a current thing.
[00:09:13] For patients with spinal cord injury or arachnoiditis, it’s arachnolysis and shunting of some form. And idiopathic cases generally in our experience are best treated with shunts.
[00:09:24] I’ll just touch over this very briefly. Chiari malformation is treated differently by every neurosurgeon. There’s no clarity around what the appropriate surgical goals are. There’s many variations on the technical aspects of the surgery. But what we do know is that we do expect syrinx cavities to reduce in size with appropriate treatment of the Chiari malformation, even if they’re communicating.
[00:09:48] I’ll show you an example here of a patient with a communicating syrinx. With Chiari malformation, where the underlying pathophysiological problem is this membrane at the foramen magendie that is obstructing the outflow of CSF from the fourth ventricle and allowing CSF to be driven down the central canal and causing that high cervical syrinx.
[00:10:11] So my own experience with Chiari associated syringomyelia, we’ve been pretty successful. I’ve had only one patient where the syrinx hasn’t collapsed. This is a young boy and I accept that there’s a skull base abnormality, and so there’s probably some complexity there. I’ve had the opportunity to treat many patients who’ve had their original Chiari surgery done elsewhere. And the things that are common: Duragen or other very inflammatory producing graft materials, or sometimes even having the posterior fossa dura left open, as is a British tradition. This is a Duragen graft attached to the cerebellum. And so I make a case for not using Duragen for this purpose.
[00:10:53] Our experiences for revising posterior fossa decompression for these patients has been pretty good. In over 30 patients, they’ve done reasonably well and the syrinx patients also done reasonably well, but we’ve had to revert to shunts in a few of the patients.
[00:11:09] Some patients present with craniocervical junction arachnoiditis. This is a case where the cause of that was birth trauma. So, hemorrhage at the time of birth. And then this patient’s now in his twenties and presenting with syrinx cavity. And that arachnoiditis as a distinct membrane and obstruction of the subarachnoid space is evident at surgery. And I tend to put a shunt catheter across that space because I can’t pretend to think that scar tissue is not going to reform. And I think that a catheter like that helps to keep that open. And as I say, I think it’s about the transmission of pulsation more than the actual transmission of CSF flow. I think in those cases, the fourth ventricle to spinal subarachnoid space catheter is very useful.
[00:11:55] We published that recently the craniocervical junction arachnoiditis, this is not redo Chiari’s, this is where there’s been either surgery for something else or trauma or birth trauma, birth hemorrhage, and being pretty successful with that.
[00:12:11] And turning to idiopathic syringomyelia, that’s not really received a lot of attention in the literature. There’s not really very many large series. Obviously one of the difficulties is the imaging technology that we use to try and find the site of arachnoid obstruction, because we tend to think that in most of the cases that are classed as idiopathic, there will be an arachnoid pathology. It’s just a matter of finding it.
[00:12:34] This is from our experimental work demonstrating that in a region of obstruction in the subarachnoid space, there is an increase of CSF flow into the spinal cord. So, it’s likely, I think that’s the underlying mechanism for syrinx cases where there’s an arachnoid pathology. You’ll all be familiar with the cardiac gated phased contrast MRI as a method of studying CSF flow. We do use that, but use that also in conjunction with the cardiac gated heavily weighted T2 scans, such as the balanced fast field echo here, that is very useful at demonstrating arachnoid membranes, such as this case, much more clearly than on the plain T2 scans or on the phased contrast scans. And that’s that patient at surgery where we find intraoperative ultrasound extremely useful at imaging this kind of pathology. And usually there is clear separation of the CSF spaces above and below that kind of membrane. It’s dorsal membranes that seem to be the problem rather than ventral membranes. And that’s the patient after treatment.
[00:13:43] For some patients, even without imaging, we can’t really identify the underlying arachnoid pathology. And we treat those patients with shunts. I’ve shown this case because this is a patient, we’ve got quite a reasonable number of these patients where there are two cavities. So the top panels at presentation. So, a separate cervical and thoracic cavity, and then I’ve put in a single shunt in the thoracic cavity and both cavities will collapse.
[00:14:11] I’ve seen that in all the cases where we’ve done this with two separate cavities. And I’ve highlighted here the fact that I’ve put the catheter going upwards. And you might think that intuitively you should point the catheter downwards because you’re worried about the CSF pulsing into the syrinx cavity.
[00:14:27] The reality is that although there is a pulse transmission in the spine, the wavelength of that pulse is over a meter. And so it doesn’t actually run down the spine, like a wave coming towards you if you’re watching it, the whole thing goes up at one time. So it’s not like you might think with a wave coming towards you. It’s quite complex.
[00:14:48] So sometimes people think that this protrusion is a cause of syringomyelia. I think that’s very rare. This is a case where you might’ve thought that. So, when she presented at our clinic, she had the syrinx and that disc protrusion, it might’ve been reasonable to think that was the cause, but when we went back to her previous scans, it was present in 2010 without the disc protrusion. And that’s with a. Treatment after a shunt.
[00:15:16] Arachnoid pathology can be complex. This is a man who had a spontaneous spinal subarachnoid hemorrhage, developed arachnoiditis, and then a syrinx several years later. And our imaging, dynamic imaging studies suggested there was arachnoiditis here. He eventually came to surgery and I put in a shunt. I operated over this region that we thought was the arachnoiditis pathology, put in a shunt in the syrinx, but also in the subarachnoid space, but he got worse in the very early postoperative period. And what I had done was missed that there was a lower region of arachnoiditis. There was still an obstruction in the subarachnoid space.
[00:15:56] And this was just an example where if you put a shunt into a syrinx cavity, you’ve got to put the distal end of that shunt catheter below any arachnoid obstruction, not above the obstruction. And so he eventually got better when we addressed that arachnoid pathology. Another example of a way our dynamic imaging is very useful: so, on the plain T2s with a bit of scoliosis, it’s difficult to see the arachnoid pathology, but our dynamic scans have demonstrated that very clearly. And with resection of that membrane, the syrinx collapses.
[00:16:31] Some are more complicated. So this is a more extensive arachnoiditis pathology related to subarachnoid hemorrhage that we were able to treat surgically and she did get some improvement, but I can tell you that’s a very difficult problem to deal with. And in some cases like that, I think it’s essentially untreatable.
[00:16:52] Post traumatic syringomyelia, I’ll just touch on quickly. I use shunts as the mainstay here rather than trying to clear the entire subarachnoid space of arachnoid pathology, because I think the scar tissue will tend to reform. The key is to get the distal end of the catheter caudal to any residual arachnoid pathology, rather than just into subarachnoid space that can remain affected.
[00:17:15] I showed this case yesterday where at an outside institution, the distal catheter was placed into the subdural space rather than the subarachnoid space. And that means the CSF coming out of the syrinx had no prospect for being reabsorbed. And so it created this separate cavity. This is that case reopened. So I’ve opened the dura, that’s the CSF coming out of the subdural space where there’s a new arachnoid membrane, but basically an encapsulated arachnoid cyst in a sense, and then reopening the real arachnoid space and putting the catheter into that space resolved the syrinx.
[00:17:49] We published our syrinx shunts a few years ago, and it’s currently, I think, the largest sort of modern series in the world and showed that the results are actually very good.
[00:18:00] I’ll go through the technique. So I think I’ve got a little bit of time to show the technique. So just finding the midline, I usually do a midline myelotomy unless there is clear presentation of the syrinx to the dorsal root entry zone, and I’ll go through there, but otherwise it’s in the midline. And I find that a careful approach through the midline only very rarely produces any neurological symptoms.
[00:18:28] And then the catheter goes in like this. You might think that I’ve got a bit of a tremor, but I’m actually just wriggling it, wriggling the catheter in. And then it’s a matter of which way it goes. If there’s no arachnoid pathology, this is the case where I’ve, it just seemed to slide better caudally and the distal end of the catheter is directed rostrally, but I’m not concerned about that.
[00:18:50] So our results in that series we published a few years ago were that all the syrinx cavities are that were stable was where they were post traumatic cases, and there’s arachniditis around the cord and it kind of holds it open. But in the majority of cases, the syrinx cavity actually reduces. I had a couple of cases where the shunt catheter pulled out, and I changed the way that I secured the catheter from a finer gauge suture to an 8-0 suture that seems to be more robust and holds it in place. I had one patient, one post traumatic syringomyelia patient who’s paraplegic, who dislodged her catheter while she was jet skiing, which I hadn’t anticipated.
[00:19:32] So, just finally, the challenges with syringomyelia management is detecting the subarachnoid space pathology, particularly in those where you think it’s idiopathic; the cases with extensive arachnoiditis I think are untreatable; the small idiopathic syrinxes are really difficult if there’s no identified arachnoid pathology. And fundamentally, I think the real problem here is we don’t actually understand the underlying pathophysiology. So it’s going to be difficult to make any big advances.
[00:19:56] And I’d just like to acknowledge this research support we’ve had for our laboratory research and the big team behind that over the 20 years that I’ve been doing this.
[00:20:06] Thanks for your attention.