Transcript
Thank you all. Hello everyone, thanks for being here. I’ve never been here before. It’s an absolutely amazing campus. So, I’ll talk a little bit about – I mean, this was given to me – the updates and challenges of CSF leak surgery for patients who have dural defects. So not per se cysts or fistulas.
I’m from Los Angeles. I work with a fantastic group of people. We talk about CSF leak problems many times every single day.
As far as updates are concerned, I first just wanted to talk a little bit about how common this is. And what I found very amazing is that there have been two epidemiological studies, one from Beverly Hills and one from Olmsted County, Minnesota. You’d think those are really different types of populations, but the incidence is almost identical – it’s 3.7 or 3.8 per 100,000 per year. So, I thought that was pretty amazing.
What I do want to stress is that surgery is really uncommon for people who have spontaneous leaks. When we looked at our study in Beverly Hills, only about 10% of patients actually ended up requiring surgery. And I also think it’s important to emphasize that we’re all here from spinal fluid leak centers. I think it’s really important to treat these patients expeditiously. Dr. Beck has shown recently that if you treat it early on, the outcome is going to be much better. So, don’t waste any time, at least initially. If you have a patient with a leak that’s been there for a few days or weeks, don’t waste your time by trying to get them to a CSF leak center – whatever that may be. You just do blood patches, and then hopefully that’ll work. And if not, then you can work on an expeditious referral.
So, we’ve gone over all of these things. I’ll talk a little bit about ventral leaks. Ventral leaks are almost always caused by a calcification of some sort. And I think if you see those patients early on – the first few days or weeks – you can just do non-directed blood patches, and there’s still a fair chance that the leak will seal. It can seal around the bone spur. If it’s kind of an intermediate type of situation – the leak has been there for, let’s say, 2 to 3 months – then I think a directed approach can be really useful. But if it’s more than a few months, we don’t bother with it anymore. We go straight to surgery. The size of this little hole in the dura can vary greatly. It can be just one or two millimeters, and it can be up to almost an inch in length.
So, for the first update, I looked at the recent literature of different groups who have published their results over the last two years of ventral leak repair. I didn’t want to really look at just a few cases – there had to be at least 10 patients included in the group. And I specifically looked at what is the success rate of surgery defined as a postoperative scan was performed, and it showed that the leak was no longer present.
This is a study from Cedars where we looked at 41 patients who happen to have superficial siderosis – and we were able to successfully treat that in all 41. Again, this doesn’t mean that they’re better after surgery per se – but at least you know that the leak is no longer there.
This is a study also in patients with superficial siderosis from Dr. Beck, also reporting a really high radiographic success rate.
This is a larger study that also included patients without superficial siderosis – also a really high proportion of successful repairs.
This is a study that came out this year from Switzerland. There, they said their success rate was a little less – but they also included a fair number of patients who never had a post-operative MRI scan, so it’s probably quite a bit higher than that.
And then I also wanted to include this recent report by Professor Takai, where they didn’t really say what their success rate really meant, but they defined that the repair was a success in all patients they studied. And usually, I wouldn’t include that in a literature review, but I did because I thought it was really wonderful that Dr. Takai made his own illustrations, as you can see here. I thought that was really beautiful, and you can tell that he put a lot of time and effort into that.
So, I think somewhere between 80%, 85%, and 100% repair success is something you can work for. I think if you do a fairly large number of these surgeries, you should really try to have a success rate of, let’s say, 95% or so. But you know, 95% means not 100%.
So, this is a lady who we saw a couple of years ago. She had a ventral leak. This is her myelogram. It was in the cervical spine, and I did, you know, the regular posterior intradural approach, and the little bone spur that she had was right in the middle, in front of the spinal cord. I really wasn’t able to remove that safely. I did place some artificial dura there, but that was not enough to repair her leak. So, then we went from the front, that Dr. Lennarson just told you about, out through a corpectomy. You can see here on that picture in the left upper-hand corner that that little piece of bone is really lodged in the dura, but it was very easy to take that out from an anterior approach, and it just took one or two sutures to seal that.
Sometimes you and the patient can have good luck. This is a lady who lives in California and had a ventral leak. I repaired. It was easily accessible, but the little muscle graft I placed might have been too small, and she was still leaking following the surgery. She was feeling a lot better, though. She was feeling so good, as a matter of fact, that she became pregnant. After she delivered her son, her symptoms pretty much went away, and we repeated the MRI scan, and the leak had miraculously resolved. There’s at least two years between this post-operative MRI and when she delivered her child. Why that happens, I don’t really know.
As far as complications are concerned, I think you should strive to have less than 1% or 2%percent of any type of permanent neurologic deficits. But that does mean that, if it’s two percent, then if you do have a deficit after surgery, it means that if you want to keep that at 2% percent, your next 49 patients who have this surgery have to have zero deficits.
All right, what about sutures or no sutures? As Dr. Lennarson showed you, if the ventral tear is a little bit off to the side, then we like to suture it – or at least, I used to like to suture it – but if it’s really right in the midline, it’s not safe for suturing. So, I use a muscle graft, but you can use fat or some other material.
This report from Switzerland compared the success rate of surgery of direct suturing versus patching – a graft or putting in a patch – and there was no statistically significant difference between those two.
Update number three is minimally invasive surgery. This just lists, from most invasive to least invasive. When I started doing these, in I think 2009 to 2010, we would do facetectomies, a fusion. I would try to repair it from outside of the dural sac – that’s, you know, kind of an invasive procedure – and in my hands at least, the success rate of that was not very good. So then, I changed to this intradural repair. You can do a full laminectomy with or without a laminoplasty. I still think that’s probably the most effective and safest way to repair these. But you also can do a hemilaminectomy, which is, you know, almost as safe and effective, and definitely, the recovery is a little faster.
You can also do it through a tubular retractor. Patients stay less long in the hospital. But for me at least, I find it challenging to repair the dura with sutures just working through the tube. I think there have been a few patients who’ve been treated with a pure endoscopic approach – not putting an endoscope in after you do a laminectomy, so endoscopic outside of the dura.
This is from Dr. Beck’s older paper already, and you can see the hemilaminectomy defect.
And then this is from a French publication, and you can see that the amount of bone removal to repair this ventral leak is really minimal. It’s only 1 by 1 centimeter. I think that’s maybe a little bit too minimal.
And then Dr. Beck and Dr. Massicotte in Toronto are really the two neurosurgeons who started doing repairs through a tubular retractor, as you can see here. The amount of bone removal – whether it’s done in France or in Germany – is pretty identical, and it’s made quite a bit of a difference. So, this is from a patient I operated on in 2009 when we did facetectomies. That is a big incision, and now we try to do it through a tube.
And then, this is this endoscopic repair. I’ve never done that before, so we’ll skip that.
Lateral leaks – so we saw a little bit of that as well. Most of those are in the axilla of the nerve root. Some of them are at the shoulder of the nerve root, as you can see here. And it depends a little bit. It’s nice to have that knowledge before you do your either needle procedure or your surgery. And then, sometimes, they’re not related to the nerve root sleeve, and it’s just at the level of the pedicle itself.
The results of surgery for lateral leaks – I think the lateral leaks have a little bit of a higher miss rate, so to speak. But I think, you know, again, if you approach about 95%, you’re in really good hands.
Another update is that the group at Hopkins has developed an animal model for CSF leaks, and they’ve used knockout mice who miss a certain gene that we have found is prevalent in patients with spontaneous leaks. Cassie Parks, who you might have met at our last meeting in Hawaii, has totally developed this model by herself. She does a C1 through C8 laminectomy in little mice and she places a tiny catheter and infuses them with saline.
Now, some of the challenges – what to do with, you know, the osteophyte. So, some people call it an osteophyte, a microspur calcified disc herniation. Oftentimes, they’re really small – like you can see there, it’s like just a little dot of calcification, and it can have migrated intradurally, as you can see on that left-hand side. If the disc is a little bit larger, you kind of like to remove that, and you can do that intradurally or you can do it extradurally.
There’s this same paper from Dr. Takai again, who believes that a lot of these ventral leaks – it’s not really a calcified disc, it’s just little calcifications in the ligament. But clearly, sometimes, you know, you do have to remove the disc, as you can see here. And so that, you know, that’s actually pretty straightforward, but you have to be, you know, kind of careful – there’s a very sharp knife right next to the spinal cord. I just show this to keep this out of the hands of orthopedic surgeons, and they’re not really interested in that anyway.
Sometimes, they’re a bit more difficult. This is an ER doctor I saw this week who has this calcified disc, and I wasn’t quite sure if I could remove that intradurally, so we plan to see if that’s possible. So then, if that’s possible, then I go ahead and do it, and if it was not possible, then I would have used a different approach to remove that from the front. Luckily, we were able to do that.
This is a different patient – he’s an F22 fighter pilot, and he had the subdural as a very extensive ventral leak. But he had this rock that’s poking right into the spinal cord, and I didn’t really feel confident I could fix that intradurally. I also thought it certainly wouldn’t be safe to do it. So, two approaches I’ve used in the past are either a thoracotomy – you do that with the cardiothoracic surgeon, but that’s a really invasive procedure and apparently is very painful afterwards. So sometimes I do this – it’s called a costotransversectomy – also a very invasive procedure. You have to remove a large part of the rib and a large part of the spine, and also there’s a lot of bleeding. So whenever – and I don’t do that approach, but one of my colleagues does that – but then they also want to leave a drain in, and I don’t like leaving drains in when you have repaired the dura because that can cause a higher risk of CSF leaks.
For this patient, we did this minimally invasive retropleural approach that I’d never seen before until two weeks ago. And it’s really – it’s pretty neat, and you know, they have their own instruments. I’d never even seen that, and it’s all done by the neurosurgeon – all done by the spine surgeon, but it takes a long time. It took him about five hours to get me to where the dural tear was, but the result was really good. You can see the pictures before and after surgery. The patient does not need a fusion. You can, you know, safely remove this rock poking into the spinal cord.
And then really the most important challenge is, you know, a lot of patients have successful repair – meaning, we don’t see the leak anymore – but they still have symptoms. So, it’s not these types of patients. It’s, you know, patients who are still suffering. I’ve estimated that in the past, somewhere – you know – it’s pretty high, somewhere between 10 and 20%. So that’s a real problem, and actually, Dr. Beck is publishing this article, or maybe it’s out already – I don’t know – where they looked at a fair number of patients, 57, and 20% of them, even though the leak was successfully repaired, still had persistent headaches. Alright, thank you very much.