Please note: This video cannot be played on this page because it includes surgical images that are age-restricted on YouTube. Please click on the photo above to access the video in YouTube itself.
Transcript
Alright, good morning everyone. Thanks, everyone, for being here. It’s my pleasure to be part of the leak program here at the University of Colorado. Let’s get started.
So just quickly, a little bit about what we’re going to talk about. We’re going to go quickly through some leak types. We’ve already done that, so I won’t spend a lot of time there. A few anatomical considerations for surgery. We’ll go through some case examples, and then, if I have time, we’ll go back and just make some generalizations about these surgeries.
So again, the iatrogenic ones are interesting in many ways, primarily because we, as the people trying to fix these problems, often cause them. And we’ll hear more about those from Dr. Beck. These post dural punctures – spine surgery, of course, is fairly common, more common than you might expect, and can be difficult to repair, as we know. Spontaneous leaks – we’ve talked a lot about. The traumatic leaks – I’m not going to spend much time talking about, but I want to point out here: this is a common scenario that I’ll see in the trauma center, and it doesn’t take a lot to imagine that the dura here is going to be injured. But unlike spontaneous leaks, that we have known so little about for many years, these have been obvious for a long time, and yet they are routinely ignored. It seems that if, after one of these traumas, they’re not, CSF leaking out of the skin – sort of contained – then they’re felt to be not a problem. And yet, we see a lot of chronic issues related to these terrible cases of siderosis, things like that. And when we look at the imaging, the leak’s been obvious all along, so I think they need more attention – but not today.
Spinal fluid – let’s talk a little bit about this, because I have so many conversations with patients, and we’re all talking about leaks, but some of these basic little principles and facts seem to be never discussed. So just in terms of surgical considerations, the spinal fluid here – and this is a simplification – but surrounding the brain and then all the way down the spine. We might have something like 150 milliliters or CCs at any one time within the brain and spine. But we’re making half a quart, or 450 to 500 ml per day. So, it is circulating, it is being reabsorbed, and primarily, we think – and I’ve always been taught – that it’s being reabsorbed up here, back into the bloodstream.
So, in the SIH cases that we’re so interested in, it’s obviously an issue of too little volume or too little pressure, as we’ve talked about. But then we have the opposite problem, right, with our idiopathic hypertension cases, or in a surgical setting, or a treatment setting, with this rebound. So, let’s keep that in mind.
Just looking at the spine a little bit – and yes, as a spine surgeon, I have spines and skeletons in my closet and hanging on the back of my doors. And this is one hanging on the back of my office door. So, you’ll see the thoracic spine – which is defined by that rib cage – is where the majority of our spontaneous leaks are occurring. This is interesting to me because the majority of my non-leak spine surgeries are actually in the cervical and lumbar spines. But really, just to point out here that the lamina is this bone on the back of the spine. And so many of our spine procedures you hear have the word “laminectomy” in them. Whether we remove a portion of it, we might call it a hemi-laminectomy, a full laminectomy, a partial laminectomy, or a laminotomy. That’s all just getting access to the important parts underneath the bone.
If we shift our attention just to the right here, we’ll see ventral leaks are behind the body and in front of the spinal cord, which is here. Our dorsal leaks are just underneath the lamina, which is here – that bony arch. So, that’s our dorsal leak. Then, our lateral leaks and our fistulas occur in this region here, which is the neural foramen, right underneath the lateral aspect of the lamina and the facet joint, which is here.
So, let’s just jump into a few cases, and then we’ll go over some of these principles. So, this is an iatrogenic leak, interesting to me because it was a ventral collection. It started in this 24-year-old at really 5 years previously, when he was 19. He was in an ER, had a lumbar puncture in a workup of meningitis. He had an obvious post-LP headache. He had a blood patch, which gave him some improvement for about six months, but then he had development of orthostatic headaches. And more so, maybe more than orthostatic, it was exertional. So, he couldn’t play basketball anymore. He couldn’t lift weights. He couldn’t do any of these things. And then intermittent nausea, vomiting, brain fog – all those things. And so, he was eventually referred to us. He has this big ventral collection, which you can see nicely on MRI, and on dynamic CT, was localized to the 3-4 level in the region of his prior lumbar puncture.
This is a typical surgery. So, this is a retractor here, holding everything open through a small skin incision. In this case, I did a full laminectomy at that level – the L3 level – because I couldn’t tell from the imaging whether it was going to be to the right or to the left, so I wanted to be able to access it from both sides. Transdural – by that I mean, I had to open the dura dorsally and retract the cauda equina nerves, which you see here, and then found that puncture site, which was in the ventral dura, as opposed to where I typically had thought about post-puncture holes being in the dorsal dura. So, this was in the ventral dura. This pad – we call this a pattie – here, that’s about a centimeter in size, so that gives you an idea of how big that hole was. It was a little bit hard to find actually, but there it was. And that’s repaired with a single stitch. That type of surgery takes about less than 2 hours. He, in this case, developed some immediate rebound – was not so significant – but it did benefit from treatment. So, Diamox, which he didn’t tolerate, as many people don’t, and then Topamax – really Topamax helped him for the first six to eight weeks. He steadily improved, and by 12 weeks, he was back playing basketball and really had a nice outcome.
Let’s talk about a thoracic ventral leak. These are the more complicated leaks to treat, and I won’t spend a lot of time here. But typical SIH symptoms, a number of attempts at patching, dynamic CT myelogram confirmed a ventral leak at 3-4. This just shows a little marker that Dr. Callen and company are able to place for me at the time of their procedures, which makes my surgical localization much more efficient and easy.
So that procedure, again, I would typically do just a hemilaminectomy from one side – left or right – and again have to open that dura to get to that ventral space. It typically takes less than 3 hours. So, here’s a nice intraoperative picture of one of those classic ventral leaks right here that you can see through my dural opening posteriorly. In this case, I chose to patch that with a fat graft – so that’s a piece of fat harvested from the patient from the same incision site. Many people would put muscle in there, and then there are other patch materials as well that we can use. You see the edge of the spinal cord there. And then again, clinically, I typically keep those folks in bed overnight. She was mobilized the next day, cleared by our physical therapy team, discharged home without a headache. Two-week visit – a little soreness from the incision but not requiring medication. And then by four weeks, follow-up scan shows resolution of all that epidural fluid.
Another ventral leak just to show you – we commonly talk about these bone spurs. So, here’s a nice myelogram picture on the left of the leak, and then that little bone spur there that I’m pointing out was the culprit here. You can see that here intraoperatively. That’s the spur that was sticking through the dura, and once that was removed, that’s the extent of that dural opening. So, this one – again, hemilaminectomy – you can see the laminotomy window there and the dura. That’s what I have to open to mobilize the cord, which, below here, you can see the cord. And then this is a fibrin collagen type duraplasty patch, and this is really nicely described by Dr. Beck in one of our Neurosurgical journals from a couple of years ago. I encourage you to look at that if you have more interest in that. This is just the day after surgery after this gal was in bed literally for a month at another hospital before coming and cautiously taking her first steps the day after surgery.
I want to show you this real quick – some have already seen this. This is just an endoscopic view of a ventral leak, and I want to show it because it really gives you a better sense of what I get to see in surgery. So that’s the spinal cord that you’re looking at. That’s the dentate ligament there, and there’s the hole that I will show you as I retract just gently. That’s a typical ventral leak live in surgery, and then you can see as I’ll be scraping that little osteophyte and roughing up those edges to facilitate healing of that hole once patched.
I think we’ll – well, most of these ventral leaks I don’t end up repairing primarily. This is an example of one that I could reach. I’m holding the spinal cord out of the way with one instrument and then sewing with the other one, and you’ll see bringing those edges together nicely. And that’s most satisfying, but it’s often not possible.
Let’s go on to another case here. So, this is a case of a lateral leak – typical SIH changes, a high Bern score, multiple patches that did not solve the problem. Quick look at the leak, and then, interestingly, the view I get in surgery. So, this is a close-up, obviously under the operative microscope. This is my laminotomy window. I’m working through one of these small tubes in a minimally invasive fashion. The nerve root is here, just under this edge of bone. It’s hard to see, and this is in that axilla, or sort of the armpit, of the nerve. And this is a diverticulum that was actually leaking from right here. But that’s the diverticulum – a challenging thing to repair in some cases. In this case, I chose to put a little clip across the neck, almost like treating an aneurysm, and that cured the leak. I then put patching material over that as well.
So, this shows what that’s like working through that little tube, and again, took less than two hours. I admit these folks overnight, generally doing well the next day. By two weeks, resolution of symptoms other than some rebound when lying flat. By six weeks, through that period, and that’s the surgical incision.
We’re going to hear a lot more about fistulas today. I like this picture because it illustrates well how this fistula here comes out of the cyst here and why our most common treatment then surgically is to clip that root or divide that root just proximal to that cyst. And if there’s no cyst anymore – there can be no fistula. So, it doesn’t require that we actually identify the fistula, although I do attempt to do that.
Here’s an example of that again through this tubular retractor, and you see the cyst here. That’s the proximal nerve, and then in this case, there’s a clip across that. I more commonly use sutures rather than clips. Clips are easier, but I prefer suture when possible. I also patch that as well, and again, a two-hour type procedure. In this case, because it’s in the neuroforamen, I’ve not only removed some of the lamina and just a little bit of it, but I have to take away some of that facet joint, which in the thoracic spine is acceptable and does not require any sort of reconstructive surgery after that. Again, typical overnight stay and a fairly quick recovery.
I should point out I said here “non-eloquent” because in the thoracic spine, we really can get away with taking those nerves and not creating any issues. In fact, I’d say most of my patients really have limited symptoms at all – not pain, not numbness. There’s no weakness to be concerned about, so they tolerate that well. But some of these fistulas occur either in the lumbar spine or the lower cervical spine where those roots are important, and we don’t want to clip them – we call those eloquent nerves. And here’s just two examples where I’ve identified a fistula in an area where I did not want to sacrifice a nerve.
Both of these patients – this one had had an embolization procedure already, a transvenous embolization, and so the only vessel that didn’t get embolized was the actual culprit, and so it was easy to identify. This is one where it was just easy to demonstrate, and both of these patients did well and had resolution of all their brain findings and their SIH symptoms with that simple clipping.
Just thinking about all that, most of the patients that I treat are really in the hospital for a day or two, some maybe as long as four days. The activity restrictions sometimes, I feel, are much less restrictive than patching. So, I’m typically getting patients up within a day – not three days of bed rest. They’re usually on the ward. They’re not in the ICU. They may or may not deal with rebound, as we’ve seen, and then the post-op pain is typically quite manageable.
I would recommend you partner with the anesthesiologist, because I use other blocks, antiemetics, steroids, non-steroidals, even ketamine – a lot of things to help make the patient and the experience better from a pain standpoint.
Just one last mention then – cervical leaks. Because we access the cervical spine from the front, it can be reached from the front at times rather than through the back. It’s nice that we don’t have to work around the spinal cord in that case. But on the other hand, it often requires removing a whole vertebra, which we call a corpectomy, and then it requires a spinal reconstruction when we’re done, whereas the posterior procedures do not.
I think I’ll stop there in the interest of time. This shows a surgical setup with our operating scope, which allows me to take all those nice pictures. Thank you.