Dr. Wouter Schievink at the 2023 Cedars-Sinai Intracranial Hypotension Conference, Day Two
Dr. Wouter Schievink, Professor of Neurosurgery at Cedars-Sinai in Los Angeles, CA, presented this talk on surgical approaches for thoracic and lumbar ventral leaks at the 2023 Cedars-Sinai Intracranial Hypotension Conference on July 8, 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] So, you know, like Dr. Chu mentioned for the cervical spine, I think it’s important to, you know, just have everything in your armamentarium. So you can go from the front, you can go from the back for these ventral leaks, which is, you know, which is a large part of our practice when it’s in the thoracic spine.
[00:00:29] Of course you can approach that anteriorly, but you know, who wants to have their chest cracked open. But I think it’s really important to realize that even though we talk about these you know, surgical approaches, there’s a lot of referral bias, right? So in our medical center, about half of the patients we see end up with surgery. But when you look at it at a community base level, it’s really about at most one out of 10 because, you know, most leaks can be fixed with either patches or just some conservative treatment. And I think, you know, especially at these meetings, we kind of, you know, lose sight of that.
[00:01:09] So we’re talking about ventral leaks. We went over that yesterday. So if you catch them early, you just do blood patches and almost always that will cure the leak. But if it’s more of a chronic leak, then you have to go to surgery. This is sort of my own timeline for the surgical treatment of leaks.
[00:01:29] And then, of course, the main difference was made back in 2009, when we started doing DSM. So before we did DSMs, only rarely did we know exactly where the leak was, or I would just approach some sort of thoracic osteophyte, or sometimes we’ll do that with a thoracotomy or with endoscopic approach.
[00:01:52] But after DSM, we knew exactly where the leak was. We could direct our treatment. And in the beginning, like the first two years or so, I used an extradural approach. So, for an extradural approach you do a laminectomy, partial facetectomy. You usually don’t need a fusion, even though we almost always did that.
[00:02:11] And then you go around the thecal sac. Right. And it’s really difficult to identify the actual tear, when you do an extradural approach. Only if it’s pretty off the midline, we can be able to see it. And then you would just, you know, you would put some artificial dura, some glue at that level.
[00:02:29] And in the summer of 2011, I thought, well, let me look at the actual results of that extradural approach. And the results were not very good. I think we were only able to cure about two out of five patients with ventral thoracic leaks. So in the summer of 2011, I started doing it just purely through an intradural approach.
[00:02:52] Now as far as finding the exact site of the leak, like I said, you know, we use DSMs a lot. The drawback of that is that you can’t really take that out, you know, for more than maybe one or two minutes. So then you can go to dynamic CT myelogram. Sometimes you can look at an MRI scan if you have really high quality MRI scan.
[00:03:12] And if you’re lucky, you can actually see the little dural defect in front of the spinal cord. We haven’t been that successful with that. But if the only thing we can go by is a little bone spur, that can be really helpful. But as I will show you later, oftentimes that’s actually not where the leak is located.
[00:03:33] So these little bony spicules, they usually look like the picture on the left. So there’s like a, it’s almost like a little scalpel that has gone through the dura sometimes into the spinal cord. This is one that that’s a little bit off to the side. And then after surgery, you know, you can remove that disc through the dural defect.
[00:03:55] The size of this little piece of bone really can vary, right? So like on the left side here, you can see a tiny little bone on the left, and then the CT, the post myelo CT on the right side—or actually this is just a plain CT—you can see, you know, that’s like a big osteophyte. You have to be careful again, right? If you try to use the bony defect or the the little spicule as the source of your leak. Because we think even though maybe up to 90, even a hundred percent of patients with ventral leaks have a little bony spicule, it’s probably not always the case. And certainly it’s not always the largest.
[00:04:32] So this is an example of somebody we saw a few years ago, and I was pretty much convinced that the leak was at that level where you see this large osteophyte. That was not where the leak was located. And the DSM showed that it was actually located at a level where there was just a tiny little fragment of bone.
[00:04:50] And then of course, over time, we’ve known that for a while, that fragment of bone, the osteophyte, can just absorb. And I do think it preferentially absorbs at the site of the leak. So once in a while we’ve been lucky, and we have been able to look at all CT myelograms and we see that one of the little spicules is no longer there. And then we would direct our treatment to that part of the spine.
[00:05:17] So, I’m just going to show you a whole bunch of these dural defects, right? So this is all looking from the back. So, we drill off some bone from the back, I open the dura, and as you will see early on in these pictures—I, really I’m showing these because Dr. Chu still hasn’t gotten me the wallpaper of this that he promised me quite a long time ago. But if you look at some of these pictures, you’ll see that there are little sutures like in the dentate ligaments because I sometimes would use sutures to kind of rotate the spinal cord away from me, but, you know, as you can imagine, that has some risk to it. And I know they all look kind of similar, but they’re really, they’re all unique patients. So, then after a few years, I kind of gave up on that and I kind of used a technique where you use a little instrument, you go into the dural defect and you kind of sweep that dura away from underneath the spinal cord. So then you really don’t even have to touch the spinal cord in most instances.
[00:06:18] Now, you know, we talked a little bit about monitoring. We still, we’ll still use monitoring when we do this type of spinal cord surgery. But the monitoring can be super, super sensitive. So certainly early on, when I started doing this, we would sometimes abort the surgery because the monitoring really showed that there was complete loss of motor evoked potentials. But then the patient would wake up completely normal, with normal strength, you know, within an hour of these motor evoked potentials going to to zero. So you kind of have to, you know, take that into consideration.
[00:06:56] Now the size of these little ventral tears can be really variable. So the smallest that I saw over the years was about one or two millimeters, as you can see there on the left. And then the one on the right was, it really was a really large ventral tear. That lady had superficial siderosis, even though she was pretty young, she was in her thirties. But this leak occurred when she was about 12 or 13 years old. Of course, by that time, whatever little bony abnormality there was had absorbed. So, the tear can be of different sizes. And this is just a an example of how you can remove that little piece of bone through the approach from the back, through the opening in the dura.
[00:07:39] So this is just a a very sharp little knife. It’s the 11 blade and the spinal cord is there on the top. And you know, when you do that, you do need a steady hand, because you don’t want to puncture the spinal cord. And then you can remove that, it’s kind of, you’re kind of debulking and pulverizing the disc fragment.
[00:08:03] So sometimes, you know, a patient is really anxious in the pre-op area. Certainly I would be, if I ever would need surgery. And then, you know, once in a while they say, Oh, hey, doc, you know, how about your hands? Are they steady this morning? And then if I feel like, you know, the patient, you know, has a certain sense of humor, then I sometimes will kind of reenact this this short fragment of “Blazing Saddles”.
[00:08:30] [Movie clip]: “Look at that steady as a rock. Yeah, but I shoot with this hand. I knew you would.”
[00:08:41] And then, you know, but once in a while I totally misjudge the patient. I mean, that’s one of the problems with, you know, seeing people from out of town.
[00:08:49] Anyway. So, certainly early on, I would always try to suture these tears just because, you know, surgically it’s certainly more satisfying. But if the, if that tear is really in the midline, right in the midline, in front of the spinal cord, and you don’t want to touch the spinal cord, then oftentimes, at least in my hands, it’s just prohibitively risky to suture it. So then at some point, I don’t remember exactly what year I started using these little muscle grafts.
[00:09:19] So you just get a tiny little fragment of muscle from the paraspinal musculature. It’s a tiny little percentage of the muscle. And then you place it through that dural hole, in sort of a configuration that there’s a, the bulk of the muscle graft is extradural and the bulk is intradural and at the site of the dural tear, that’s where it’s at its smallest.
[00:09:43] And you can you know, sometimes I tell patients it looks like a ribosome, right? Like if you have a, if you have like a biochemistry professor, I’ll say it’s a ribosome. So ribosome, or I’ll say it’s like an hourglass, or a snowman or kind of a corset. And then I used to put, you know, some glue on there, quite a bit of glue. Now I really try to minimize the use of glue because you can get a scarring. when you use that intradurally.
[00:10:11] Now very rarely I’ve explored patients who really don’t have any evidence of a ventral leak, right? Meaning they don’t have any CSF outside of the dural sac. But there’s a little bony spicule, there’s just one, and I think, you know, I really think that’s where the leak is. And they still have leak symptoms, even though we don’t see anything on the on the myelogram. And this is just an example of three patients. So if you look at those upper panels, you can see that there’s the dural hole, and the body has done a really great job sealing most of that over the dural hole.
[00:10:49] So it’s not just arachnoid, it’s actually like a very, it’s kind of flimsy, but it’s definitely scar tissue. And then on the lower panel, you can see after I have removed that flimsy scar tissue, or the one on the right, you can see on the right upper panel that there’s a bony fragment that the dura has kind of closed around. But they still have symptoms. And it’s difficult to really go ahead with that decision for surgery, because of course, we also know that there are patients, and we’ll talk about that later today, there are patients who have an extensive leak, you fix their leak, but they still have symptoms. So, you know, out of these three patients, right? I don’t know how many of those would have had symptoms anyway, whether or not you repaired the leak.
[00:11:35] So, a little bit about the technical aspects. You can do a full laminectomy. You can do a hemi laminectomy. You can do a very small laminotomy. You open the dura, you do have to divide the dentate ligament, right? So even if you feel that you don’t need to do that, that you have really great visualization of your dural tear, I still divide the dentate ligament because I think that’s safer, because that just allows the spinal cord to move more freely around that area where you do the operation.
[00:12:10] And then as far as the results are concerned. So these are the results of the first 325 patients with ventral thoracic repairs I’ve done. But it’s just for people where we were able to identify the exact site and I was able to do a repair, right? So for example, if this was a patient 10 years ago, we knew where the leak was I saw it at surgery, but the motor evoked potentials went down, I would have said, okay, you know, we’re going to back out. We’re just going to put a little, you know, dural substitute there, but it wasn’t really a repair. So those few patients are not included in that. The radiographic success rate is high. It’s like better than 95%.
[00:12:51] We had, you know, fairly low infection rate. Neurological deficits like lower extremity weakness or paresthesias, hypoesthesia, . It’s about 2 to 3%. Most of those patients had prior surgery, intradural surgery. So then there’s, you know, an excessive amount of scar tissue. So if somebody said prior intradural surgery you know, I certainly tell them that there’s a chance that they’ll end up in the hospital for more days, or they might have to go to rehab for a few weeks.
[00:13:23] The pseudomeningocele, I think most of the time what happens is that you fix the initial leak, as a result, the pressure builds up, it’s like rebound high pressure, but they don’t get the headache, but they start leaking from the suture line of where you have opened the dura. In the beginning, I would always take those patients back to the OR, but then oftentimes you wouldn’t even see the leak anymore. So now I usually just treat those patients with high dose Diamox.
[00:13:53] And then, you know, these are the kind of things you like to see, right? A lot of patients send you these nice pictures of what they’re able to do after surgery. So that’s, you know, I would say that’s really the most satisfying part of of what I do. These are some other pictures. And this is a video. This is actually a German patient. And he does this kind of sport. I mean, I don’t ski, I’m very risk averse, but they take a helicopter to some mountain and then they go out of the helicopter and they ski down the mountain.
[00:14:29] But, you know, this is what we like to talk about. But what’s disturbing is that, you know, 10, 20 percent of these patients who you know, I feel really good about, right, that we fix their leak, surgery went perfect. They get their scans after surgery, that extradural collection of CSF is totally gone, but they still have symptoms. And we’ll talk about that later at the end of the day today. That’s a real conundrum what to do.
[00:14:59] So recently I looked at the, I guess the first 21 patients where I did the surgery with the intention of fixing the ventral tear, but we didn’t really know where that leak was located, right? So these are people where the myelogram did not show me the level of the leak.
[00:15:17] And then the vast majority of these 21 patients, I think 19 out of the 21, I just targeted what I thought was the offending bone spur. And I think on two of them, I went by the distribution of the contrast. You know, some of these people are from quite a few years ago. And, you know, as you can imagine, the results of that surgery are not as good, right?
[00:15:40] So for example the radiographic results are not as good. So whereas with, when you’re able to see it on the myelogram and you can fix it better than 95% as a radiographic cure, when you just go by these, you know, somewhat localizing, but not precise type of methods, the success rate of surgery was you know, 12 out of 21, so that’s not that high.
[00:16:05] And, you know, of course, the numbers are really small, but two of those 21 patients did wake up with, you know, numbness or some weakness in one lower extremity. And even though it was transient, right, that really put a dent in the expectation after surgery. I think one of these patients had to go to rehab for a few weeks.
[00:16:25] Again, these are small numbers, but I think that’s related to, you know, when you think, you know, where that tear is located, but you don’t see it, you tend to be a little bit more aggressive, right? Maybe take off a little bit more bone, maybe do your hemilaminectomy on the other side, and then just, you know, look for the leak. But again, right, these, I’m sure that’s not statistically significant.
[00:16:46] I’ll just quickly go over some of these more unusual post operative complications, right? So for people who do a lot of this, you know, you might’ve seen that. This is a gentleman who had a ventral leak. We repaired the ventral leak, and about six months later, I remember this really well because I’m standing in line at Universal Studios and it’s like a Friday evening. I get a call from the radiologist, who says, “Oh, I just did your follow up MRI of your patient, and there’s like a giant herniated disc at the site of her surgery.” You know, this was quite a few months later. So I didn’t, you know, remember exactly what it looked like before surgery, but I thought, Oh, I thought with this patient, we didn’t really even see any disc.
[00:17:32] So anyway, you had him come back and then you see that at the level, right? If you look at those lower panels, you see that level where we did the surgery. There’s this large calcific disc herniation. And then when I talked to him, he said, “Oh yeah, you know, a few months ago, I started having some numbness or tingling in my lower extremities.”
[00:17:52] So, even though his leak is no longer there, we had to come back to the operating room to remove this calcified disc. Sometimes, and when I saw this the first time it was somewhat nerve wracking because it looks like here, right? So that’s an you can’t see that, but the one on the left, that’s a MRI scan and you see how the nerve roots, right, the cauda equina, are being compressed by these fluid collections in front and behind the cauda equina, and those are called subdural hygromas. And basically what happens is that when you do the surgery, sometimes the arachnoid kind of gets sheared off the dura, right? That’s a virtual space, but that’s just self-limiting, and that will go away without any type of treatment. We’ve seen that a few times now.
[00:18:39] And then this is a picture after surgery. This is, I think, 10 years after surgery for repair of a ventral leak. And back then I used to, you know, kind of coat that whole dural graft or even the suture line with glue. So I don’t do that anymore, just because I think that’s really the cause of this scar formation around the spinal cord.
[00:19:02] Luckily, even though radiographically it can be progressive, these patients, I think we’ve had about three or four of them, they have no or just very minimal symptoms. So, you know, this is really the kind of surgery that is, it’s really a zesty enterprise for the CSF leak surgeon. All right.
[00:19:22] Thank you.