Slides
Transcript
Thank you very much for that kind introduction, and I’d like to say thank you to our anesthesia colleagues for coming and giving those excellent lectures. We’ve been doing SIH-related lectures and CSF leak-related lectures for about a decade or so at these conferences, and it’s really nice to see some anesthesiologists come here to talk about this. I know I learned a lot. Thanks also to Wouter for setting up this conference and to the committee for inviting me.
So I was tasked with talking about treatment of post-dural puncture headache, presumably from the neuroradiologist and patching perspective. These are my disclosures. The only one I’ll add here is that Mike Malinzak, who’s one of our teammates that couldn’t be here today, did contribute a lot to this talk. So I want to give him the appropriate attribution.
Some of this was already covered by the two excellent prior lectures. But in general, post-dural puncture headache causes include those of epidural access and intrathecal access. These are iatrogenic cases. There are lots of different things where you might be trying to get epidural access that went awry: epidural anesthesia, steroid injections that some of us do here in this room, spinal cord stimulator placements, and then intended intrathecal access where we are actually intending to breach the dura but leave a hole behind. That includes spinal anesthesia, LPs that are done very commonly, but also transdural catheters and inadvertent ventral punctures, which we’ll talk about in a bit. Who knows what the actual incidence is, right? We’re getting a very wide range here in the literature.
This was already covered, so I’m going to skip over a lot of this because I think it was covered quite well. Some of the things that I’ll highlight again tend to be more common in younger patients. Some of the interesting things that I saw are that patients with IIH are more likely to get it, which I thought was interesting. We see a lot of IIH patients in our clinic as well.
And then harping on this again, the needle really matters. At our group we use the atraumatic needle, we use a 24-gauge needle, and we found that to be really critical for reducing the rate of PDPH. Of course, we have a vested interest in this because when we make a hole and lead to PDPH, we have to fix it. We definitely want to make sure that doesn’t happen, and we’ve actually disseminated that really amongst our healthcare institution and have been pushing for other people to use those needles as well.
Basically, we’ve kind of covered this already. I’ll point out that one thing that was interesting from the literature is that strict bed rest isn’t necessary, is my understanding. A lot of times I’ve heard that before, where patients are put on strict bed rest, and that’s not entirely necessary to do. Be mindful of that. And there were some really excellent points about some risks to doing that too, particularly for our OB patients.
Caffeine can be helpful. The other thing that I thought was interesting is IV fluids don’t help. That’s apparently in the literature, and that was of interest to me as well. And we talked already about how the efficacy may be higher after 48 hours.
I showed this slide before. Basically, we’re trying to cover a hole, right? And that’s kind of the idea. And so, let’s go over some of the anatomy here to sort of talk through this. And this should be basic for everybody in the room, but if there are any trainees here, we can look through this.
These are the basic components of what we’re looking at, right? And essentially what we’re going to be doing is we’re going to be trying to cover a hole that was made in the dura. And so we’re going to take our needle for an epidural blood patch and place it over the dura—that’s the epi portion. And then we’re going to take the patient’s own blood in a sterile manner and inject it into that space, cover the hole, and hopefully seal the leak. That’s the idea.
So, boy, if you really step back and think about it, this is a crazy idea to do to somebody, isn’t it? And so I think the history of this is interesting, and it actually, to my understanding, starts with Dr. Gormley here, who was an anesthesiologist in 1960. I pulled this paper—I think our medical librarian had to go hunt it down for us. I love this. There were a bunch of different patients that he presents here, and this is my favorite part. One of these cases is a personal experience. Forty-eight hours after a spinal tap for a myelographic examination for a possible herniated disc, it was Dr. Gormley himself who ended up with a post-dural puncture headache and then ended up getting a patch from one of his colleagues.
He came up with this idea because he noticed that with bloody lumbar punctures, or bloody taps, there was a less likely incidence of post-dural puncture headache. To my understanding, Dr. Gormley’s really kind of cool paper didn’t really get noticed very well in the anesthesiology literature, and it wasn’t until Dr. DiGiovanni, who was from the Air Force, published this case series in 1970, and then this sort of became the standard of care. We’re still doing it today right here in Amsterdam.
So let’s talk about some of the technical considerations. So I’m going to get into the radiology portion of the talk here where we’re talking about how we are placing these needles. Your first choice is whether it’s blind or imaging-guided, and there are certain instances where you really can’t use imaging. Optimally, you’ll use imaging guidance. One of the reasons for that—this is a paper that’s kind of an oldie but goody—is that the loss of resistance technique leads to non-target location about 25% of the time. So if you compare that to radiologic imaging with the epidurograms in contrast, we see that just using the non-imaging-guided loss of resistance technique leads to us being in the wrong position a fair amount. We see that in our pain injection clinics an awful lot. You’ll get a loss of resistance, turns out you’re in the retrodural space of Okada or other anatomic locations that are kind of interesting. And so if you don’t end up in the right spot, you’re not going to get the blood to the right spot and presumably going to reduce the chances of fixing the problem.
If you’re going to use imaging guidance, you have two choices. Your choices are basically fluoro or CT, and that was talked about a little bit before in terms of the different imaging guidance.
The other thing that we harped on this morning is whether or not you do a targeted or non-targeted, right? And these are some of the general considerations that we need to consider. The other thing we need to consider is what we’re going to put in there. The standard is blood, right? That’s what we started with, but many are using an off-label use of this fibrin glue. And what fibrin glue is is basically just the ends of a clotting cascade that comes in two different syringes that Lal showed us earlier, creating these fibrin glue monomers that look just like this. Quite similar to physiologic fibrin. So that’s the concept here. This is what it comes in. It’s got two different syringes and this Y connector where it enters and mixes inside the patient. That’s usually how we do it. This is our setup here.
Okay, so let’s talk about this problem. Here we have somebody where the physician did a lumbar puncture, maybe to get CSF to send it off to the labs. And what happens here is we’ve left that hole that was nicely shown in those images in the prior talk. And so now we’ve got a hole.
So what are we going to do to fix it? Well, we under CT guidance are going to place our needle adjacent to where this hole is, right? And so the first thing is we’ve got our ligamentum flavum, our epidural fat, and our thecal sac here. And this is my needle where I now have it in the ligamentum flavum. And here’s where I’m going to get that resistance feel. Then I’m going to slowly advance the needle through that space with constant pressure on my contrast-filled syringe until I get into the epidural space. The nice thing about CT is that we can see this very well here. And there’s a really nice paper by Dr. Callen that showed that sometimes you can see the contrast go in through the hole. We’ve seen that on occasion in some of our cases as well.
These are the different fluoro-guided approaches. This is what it would look like under fluoro. Another thing that can happen that I think people don’t think about—and I think this was mentioned before at this conference—is that the dura is actually a cylinder, isn’t it? And so if you advance your needle too far, you can puncture the ventral dura. And if you think back, if any of you were in the talk this morning where I was talking about spread to the ventral epidural space, it’s harder to get spread to the ventral epidural space than it is to the dorsal epidural space. And so I think the treatment for that is a little bit different.
What we tend to see at Duke are an awful lot of patients that are refractory to multiple prior blood patches, and this technique has been helpful for us in those cases because presumably what’s happened is the patient has had a ventral disruption of the ventral dura. And so here we use CT guidance in order to place needles to the ventral aspect of the epidural space in the spinal canal. This technique was first published a long time ago for the thoracic spine for SIH and trying to treat disc osteophyte leaks.
The keys to this are you have to be mindful of where the nerve root is. This is actually the nerve root here. That’s the edges of the thecal sac. And so I’ve got to come in here with my needle in between those two things, right? So you have to, a) know what you’re doing, and b) be able to see that stuff, right? And so CT guidance I think is kind of critical. And so we place it into that location. I certainly don’t want to transgress the lateral aspect of the dura and create another leak, right? And so that’s the challenge here. And you can see very nicely we’re able to place it bilaterally and get spread all the way across. So that’s been successful in many of our patients.
But post-dural puncture headache occurs for a variety of reasons, not only after lumbar punctures, right? And so sometimes we have to do needle placements in the thoracic spine or higher. The different approaches for patching are the same as they are for a lot of pain injections. There’s transforaminal approaches, interlaminar approaches, this ventral approach that I just showed you. But you can also do other clever things under CT, like go through the facet joint to get through specific spots, go through the costovertebral joint, or Dr. Gray did this one where she actually went through the uncovertebral joint to get to the front.
And I think if you’re going to use fibrin glue, one of the things that matters is to make sure that you’re not intravascular. There’s a higher risk of anaphylactic reaction with fibrin glue if you are in the vein, and so what we do under CT to look for that is we basically inject some contrast and we note that there’s a curvilinear structure that looks like it’s in a vein. We step again and we note that it washes out, and that washout indicates to us that we’re in a vein.
Okay, there was a movie that is not going to play. Sorry about that. But basically, what you’re looking for is washout of that contrast.
Where are the locations that we can see this? Well, this can occur in a variety of different locations in the spine. This is in the foramen. This is in the anterior soft tissues. It can be in the basivertebral plexus where you end up seeing uptake. You can see posteriorly here, and then you can see it even in the contralateral side. You have to have your field of view set up appropriately.
We’re looking for that because if you then subsequently inject fibrin glue with impunity, particularly if there was a prior exposure to fibrin glue, you increase your rate of anaphylactic reaction, which is not fun, particularly when you’re a radiologist and you have to go call a code, right?
So, intravascular injections here, this is what it looks like under fluoro. This is planar imaging, and sometimes it’s a little bit more readily apparent if you know what you’re looking for. I often get the question for patching, well, when do I stop, right? And I think that this is kind of a little bit of a balance. On the one hand, we want to have adequate leak coverage. Similarly, and I think maybe even aligned with that, is this volume concept which comes from the anesthesia literature—this magical number of 20, which I’m not sure how much I believe in. But the reality is, I think we can agree that it probably makes sense to cover the leak, right? That’s the whole point to cover the hole.
On the other end, you have the issue of pain and deficits and mass effect. Using planar imaging, or if you’re using blind non-imaging guided blood patches, you can’t really assess mass effect very well. But under CT guidance, you can. I’m going to show you that. That can be quite helpful. The other thing that people do, of course, is they will stop when the patient’s uncomfortable. Actually, that’s the way I was taught it: oh, just keep going until the person tells you to stop, right? Which is kind of a very interesting thing to be taught. But that’s kind of the standard.
And I think what we do, at least at Duke under CT guidance, is we have a little bit of contrast in our blood, which some of us have all talked about here. The benefit of that is we can see that there is progressive mass effect on the thecal sac. The tip of my yellow arrow there is just showing where we’re going. And we can kind of get to the point where we realize that we’re compressing a bit too much.
So I want to give attribution here to Dr. Malinzak. And this is his, I think, framework here of different types of post-dural puncture headache, which I think is a nice way to think about this. It splits it into acute and chronic, and then uncomplicated and complicated.
So for acute uncomplicated cases, they usually have that orthostatic headache and the neck pain, but you can have actually really debilitating cases, and I think you guys mentioned this as well, with cranial nerve palsies or subdural hematomas. We’ve even had patients admitted to the intensive care unit that were obtunded.
In chronic situations, these are quite challenging cases to deal with. I think these are long-standing headaches that are unresponsive to epidural blood patches. They can have other issues with them too, as we see here, including myelopathies and stuff like that.
This, I think, is maybe one thing that I hope I’m contributing to people in the audience here. The issue that we have at Duke—we see an awful lot of people with post-epidural for childbirth or post-lumbar puncture come to our clinic for treatment—and the problem is that almost always the brain imaging is negative, and almost always our myelograms are negative. Not all the time, but you can see those little dural blebs, those arachnoid blebs that we talked about, but the vast majority of cases, I would posit maybe more than 99% of the time, everything’s negative. And so at the end of the day, our decision here about whether or not post-dural puncture headache is actually the cause of the headache, and it’s not one of these other things that you guys are mentioning. It’s the story, right? And the story is often lost, particularly in the chronic setting when it was a year or two ago. That is one of the challenges.
Brain imaging is usually not helpful. I think it’s helpful to diagnose sort of complicated cases, where we’ve got patients who have all these other symptoms and problems. Getting a brain MRI can be helpful to say, oh, there’s really horrible brain sag, this is a problem, we need to do something. That’s unlikely.
Myelography, again, we don’t usually see it, and in the acute phase we don’t really think it’s necessary. Identifying the level is the key, and the level is where the band-aid was if it’s acute. That’s really the goal—to put it in that location.
I think myelography, in my view, is actually best used when you have definite post-dural puncture headache that’s refractory to epidural blood patches, and then you might as well give it a try in that scenario to see if you can find a little bleb that you could send a surgeon in after. In my experience, at least at our shop, it’s quite difficult to convince a neurosurgeon to go in if you don’t have something on imaging. So that’s another reason why we might do that.
Let me run through a couple of cases here. This was a 28-year-old who underwent a spinal cord stimulator placement and immediately postoperative ended up with a severe positional headache. An epidural blood patch was requested by our teammates in neurosurgery on postoperative day four. We did a myelogram because we thought there were multiple potential sites to leak, right? Because they’re running these things up along the dorsal epidural space. And you can see here that this was the location of the placement of catheters in the interlaminar space, and there was a little contrast that leaked out there that we thought was a leak. And so this is what we’ll typically do in this case. We’re trying to cover all these different sites, and you can see that CT guidance I think is quite critical for this. Needles are being placed at multiple different locations up and down the spine to try and get spread.
And I’ll point out this one. This is not uncommon for us. We’ll do this a fair amount of time in these spinal cord stimulator replacements. We’re just trying really hard not to contact the leads and basically get spread up and down the spine. So this patient had pretty diffuse spread and did well after the procedure.
And so what we elected to do in this case was to basically do that 360° circumferential patch at multiple levels. And you can see we patched in all of these different locations in order to get coverage 360°. And you can see here that this is the thecal sac. That white stuff is our contrast and patching material surrounding the entirety of the sac. So this is the 360° or circumferential approach to cover all sides of the thecal sac. And this was successful.
We’re going to, in the interest of time, skip over this one and move on to case three.
So this is chronic uncomplicated. This is a 43-year-old female with an 8-year history of positional headaches after an OB for an epidural. And I see our anesthesiology friends shaking their head. This is really common for us. We see this a lot. And it’s tough. Really tough.
Her imaging is normal, and her myelogram is also normal as well. And so we elected to effectively just do circumferential patching at every level. The problem with this scenario is we’re eight years out.
And so if you ask mom where did they do the epidural eight years ago when you were exhausted, right? No one’s going to remember that. So we just make an assumption it’s somewhere in the lumbar spine. And so we try and cover both the ventral and the dorsal surface throughout the entirety of the lumbar spine.
Optimally, if we had an imaging exam that could more reliably show where the dural defect is, that would be our first step. But we have not found a way to do that successfully, at least at Duke.
So case number four, this is a 13-year-old female. She had six months of occipital and positional headaches after an LP for meningitis. Here I’m showing you this because this is, I think, the exception to prove the rule. There’s our little dorsal arachnoid bleb, right? Our little bleb there. That’s what we’re hoping to see, at least, because now we know we have a target.
Here we’ve done a myelogram, and right at that level where they did the LP, we can do our blood patch, which we’ve done here. The interesting thing about this is that the blood patches each provided excellent temporary relief, and she would do well for a while, but then basically would get a recrudescence of all of her symptoms.
I will say I’ve had many patients where you’ll do a patch for one of these types of situations and you’ll get incremental progress with each patch. They’ll report, oh, I’m 25% better from baseline. Now I’m 50% better from baseline. Now I’m 70% better from baseline.
So I’ve had a few patients where we’ve had to do four to five patches to get them to a point where they’re 90-plus% better and can go about living their lives again. This case, due to the patient having a positive response and the known imaging finding here, I sent for surgery, and I think this is from Dr. Schievink. Perhaps he did the surgery, I do not know, but did great after that. So that’s what we’re hoping for when we decide to do a myelogram, but again a lot of times we don’t see it.
Case number five, this is acute complicated. This is a 21-year-old, and she had four attempts at OB epidurals for childbirth and then ended up with a really bad headache. So they tried four times. Now she was a larger person. So I think it was a hard epidural placement for OB for childbirth. What ends up happening with her is she rapidly progresses to seizures and decreased consciousness. This is a scenario where she ends up getting the MR after that, and she has horrible brain sag. She’s got dural enhancement, venous distension sign, all the things that we’re talking about for intracranial hypotension. And that’s uncommon with post-dural puncture headache. So I think there was probably a pretty big leak here. Plus we know she’s not doing well. So this is a complicated case of that.
I thought this one was super interesting because she ended up in the intensive care unit. She was essentially obtunded. We brought her down and did one of these circumferential patches. And I’m not exaggerating when I say she basically woke up and was fine. And that was really pretty cool. This is another circumferential patch here. I’ll try to use the little arrow here. You can see all the white stuff is the contrast that’s mixed in with the patching material. And so, we kind of tried to get the ventral and the dorsal surface, and that worked really well. And then three days later, she went home and her brain MRI was normal again. So that was pretty cool.
I think this is the last case here. This is a chronic, complicated case. This is a 23-year-old female, also epidural for childbirth, limited self-limited headaches. But then you end up getting progressive bilateral upper extremity numbness and weakness with a right foot drop. So she ends up getting an MR, and what do we see on the MR? We see basically all the signs of intracranial hypotension, reduced mamillopontine distance. But we’ve also got crowding at the foramen magnum, and we’ve got a pre-syrinx formation here, which if you continue down into the thoracic cord, you see that there’s extensive pre-syrinx and edema and that kind of stuff. And so that’s why she’s got these myelopathic symptoms.
The challenge here is now she basically had this done, this really bad situation. So we essentially did a myelogram, and the myelogram was quite interesting because we end up having a clear durotomy here with contrast leaking out all over the place. This was one of the few cases where the CT myelogram was positive, and it showed us the site of the leak so we could patch directly at that location.
And she did quite well, and 8 weeks post-patch looked better. So I guess what are the take-home messages? Yes, epidural blood patching is the standard of care for refractory post-dural puncture headache.
For those of you, though, that aren’t on my side of it, the thing that I think is really challenging is when you have these chronic cases, oftentimes we don’t know what the level is, whether or not it really was a leak. And people, I can understand, are reluctant to say whether or not it was a wet tap, but that’s really helpful information if that’s in the chart, right? Then we know where to go.
A lot of times the imaging is negative, and so because of that, it’s really dependent on the story, and that makes it challenging. And then finally, multiple patches may be required. We’ve had plenty of situations where you’d get incremental progress and then send the patient to a surgeon if it fails, particularly if the imaging is positive and you find a dorsal bleb.
Thanks a lot for your time and attention. Appreciate it.