Slides
Transcript
Hi everyone. So, I love the topic that I was assigned. So I get to talk about all the optimistic things coming down the pipeline for how we’re going to diagnose SIH. I want to start off – these are my grants that I have. And I really want to highlight the top two that are in blue because both of those are actually grants for SIH. I think you really have to acknowledge a lot of the other people in the room here who have helped to discover SIH, to explore SIH, and to expand our understanding of it. Because they’re the ones that brought it to the forefront, they’re the ones that are saying “Hey, this is a really important topic” and they laid the groundwork for this. The top one is a spine-specific society, but the really exciting is that second one, the Roentgen Ray Society, because that’s a general radiology society. So, it’s not just neuro-specific, not brain specific – it’s all of radiology. They decided that SIH was a valuable topic that they were going to put their money towards.
Also, I also want to thank Andy Callen and the organizing committee because, obviously, this is a very unique conference in the sense that patients are here, and patients are talking. But it really changes our focus as physicians and how we’re going to present our material.
I want to give you a little bit of an understanding of kind of my thought process on some of this, and just in that spirit, really provide some insight into how I approach these cases. Just to compare this a little bit: if I’m reading a brain MRI for an indication like brain tumor or brain metastasis, I look through the images, and if there’s no tumor, no metastasis, that’s great. I can dictate that, feel really good about that, and move on to the next case. If someone has back pain and they get a lumbar spine MRI, I want to say, “Is there narrowing? Is there stenosis? Is there something pinching the cord where they’re going to need to go have surgery? I don’t see any of that, well, that’s fantastic. There’s no significant narrowing in the spine.” Next case, and move on.
But it’s really interesting to think about it when I’m reading a myelogram. Because if I read a myelogram and I say, “You know, I don’t see a fistula.” I’m much more emotionally invested in those cases. Because for some reason, when I see those cases and don’t see a fistula, I start to think, “Boy, what are we doing wrong with the technique where we’re not seeing it? What am I doing wrong with how I’m looking at the cases where I’m just missing something? Is there a subtle finding here? They could get sent to treatment for that. How do I want to explain this subtle finding?”
I think, just very basically, how we as radiologists approach these cases versus other diagnostic cases is very, very different. I think that highlights a little bit of a difficult problem for us because, some of these cases, in terms of how you look for a fistula, some of it’s about pretest probability. Right? Of all the things that have been done – MRIs, maybe cisternogram, seeing an expert neurologist – what’s the pretest probability that this patient really has a fistula, and how can we improve upon that?
And so, the reality is, when someone’s on the myelogram table and I’m going to do the myelogram, my number one question is: Is there a fistula, right? I think that’s the primary question that a lot of different people who perform these have. That’s the primary question. But I would argue that that’s what we’re trying to get away from that, right? Because, in the ideal situation, through whatever markers, I would know for a fact that this patient has SIH, and when they’re on the table, my only question is, where is the fistula? But the reality is, we’re not there yet, right?
And so, I think we have two related but separate problems. The first problem, which has been clearly stated very well by so many people today, is that SIH is underdiagnosed. This means that all the things we have – clinical history, physical exam, MRI – that we need something with better, and including education. We need better sensitivity, right? Through all these mechanisms, we’re going to be more sensitive to pick this up. But I think it’s also very reasonable to admit that a lot of patients who don’t have SIH and don’t have a fistula are undergoing unnecessary myelos. You could even argue that puncturing the dura is putting them at a risk for getting a leak.
This goes back to the fact that all these tools that we have to make the diagnosis – is that they’re not specific enough. So, we have two very related topics, but you have to approach those questions very differently.
So, here we go. We have two different patients, two different brain MRIs, and I would ask you: Which one has the fistula, right? This one has a fistula. This brain MRI was done for a totally different indication – not CSF, not SIH-related at all.
And so, the thought process in my mind is: how do we go beyond Bern, go beyond SEEPS? Because, you know, clearly, the imaging findings that we look for with those techniques – they’re not going anywhere. They’re going to be a staple of brain MRI going forward with SIH. But how do we move beyond that, and what can we offer?
One of the ways I want to address this topic is to say, let’s look at anterior skull-base CSF leaks. We know very well that the majority of the patients with these do not present with SIH, right? We used to think, as was stated before, that SIH was an issue of low CSF pressure. Then the Duke group, and later the Colorado group, has shown that this is not the case. So, then the mindset shifted: okay, it’s not an issue of low pressure – it’s an issue of low volume.
But if SIH was just an issue of low CSF volume, it doesn’t matter where the CSF is leaking out – any loss of CSF should give you low volume and should give you the symptoms. But we know that’s not true, right? There’s an asterisk here. Over 12 years ago, the Cedars-Sinai group, Dr. Schievink’s group, looked at their very large patient set of SIH patients, and they said, “Huh, that’s interesting – we don’t have any skull-base CSF leaks here. This isn’t a typical patient that presents with SIH.” Very recently, our group kind of took that and focused in. We had a number of patients with verified skull-base CSF leaks, and they didn’t have the classic symptoms or brain MRI findings of SIH.
So, we can think a little bit about why is that? If you take a patient with a skull-base CSF leak, this is a drawing right here showing a CSF leak in the temporal bone right by the ear. We postulate that the CSF right by the skull-base or right by the brainstem has normal equilibrium in these cases. But if we take someone with a spinal CSF leak – being a SLEC, where you see that little osteophyte or there’s a fistula – well, you’re losing CSF in a low spot in the spine, and it has this “sump effect” where it pulls everything down. So, you’re going to see more pronounced downward flow by the brainstem, and you’re going to see all the other manifestations that we talked about. The pituitary gets big, brain sag, the tonsils are going down.
The Freiburg group did a fantastic job where they used MRI of the spine to evaluate patients with SIH, and they said that there’s abnormal CSF flow in the upper cervical spine, in C2-C3, and that the cord moves more too. So, this is using MRI to get objective data to say can we see who has SIH?
Our group wanted to use something called MRE, or MR elastography, to study SIH patients. MRE is a technique that uses an MRI scanner. It was developed at our shop, at Mayo, by the body imagers several decades ago. Because people with a stiff liver or liver fibrosis, the way to diagnose that is to put a needle into the liver and then look at what’s going on in the liver under pathology. They wanted to say “boy, we should be able to have another way to diagnose this ‘stiff liver’ without putting a needle in.” So, they interpreted MRE their thought process behind it was “Can we palpate the liver with MRI?” So, we want to use that same technique that’s been around for decades, but use it on the brain in patients with SIH.
So, what does it look like? That top picture in the middle – that’s just a standard MRI head coil, and there’s a little pad that goes under the head when they’re in a normal magnet. And there’s air that goes in and out of there, so it vibrates in the back of the head. When I put myself in the magnet, it feels similar to if you’re in the airport or in the mall and you’re on one of those La-Z-boys that you know you put coins in and it vibrates. So, it’s not this really jarring vibration, but it’s subtle, and it’s definitely there.
And what did we find? We imaged people with SIH and we did controls, and you can use fancy math with that to say “how stiff is the brain?” and you get a thing called a damping ratio. What’s important here is that the orange and yellow part – that’s the part where part of the brain where people with SIH have abnormal brain stiffness compared to controls.
Let’s talk about this for a second because this is really interesting. When we talk about a normal brain MRI, whether you want to use Bern or SEEPS, you’re looking at the surface of the brain. Are there fluid collections there? Is that dura thick? And then you’re looking centrally in the brain, so you’re saying “is the brainstem smashed?” But where the difference is here – those other mechanisms basically ignore the brain parenchyma themselves. This technique is showing that hey within the meat of the brain, there are abnormal changes that we can detect in a non-invasive way with MRI. If you look at the little dot plot kind of on the left side there, the upper right of that is abnormal stiffness, the bottom left of it is equivalent of normal stiffness. The circles are normal controls – people without SIH – and the triangles are people with SIH. The really interesting thing is that, we didn’t have large numbers with this study. But the green triangles were Bern zero. So even in the patients that were Bern zero, they had this abnormal stiffness pattern. This is something we’re actively enrolling for and we’re looking at more patients.
One of the other topics we really want to explore is the goal target would be to get for a biomarker, right? To get something that can be tested just by a blood draw. So, we looked at this protein called beta-trace protein. Out of all the proteins that are in the CSF, this is the most abundant one that’s made in the CNS. So, we looked at the peripheral blood in patients that have fistulas – again small numbers, but this is a start. We showed that their values there, and we did the normal controls. And it was a statistically significant difference of this protein marker in the peripheral blood of patients with fistulas versus controls. So, I definitely think we’re on the right track here, and we’re excited for the progress that’s to come.
I really wanted to give this talk. It’s something that showcases optimism, right? Because so many people here are patients who have suffered, and so many people here have taken care of patients who are suffering. But let’s talk about something that’s coming on the horizon – some bright points that are coming, the work that is being done, the bridges that we’re building. I have no doubt that if you look in the history of humanity and of SIH, that this is the best time in history to ever have SIH. You can say that a lot about most diseases, but even more so about SIH because in the past decade, the rate that our knowledge has expanded about what SIH is, what’s causing it, how do we diagnose it, how do we treat it, it’s really taken off at a much more rapid pace. But the reality is that going forward, what do we need? We need a universally accessible tool to identify SIH before they get on the myelo table. The reality is that if someone – if a patient and the provider thinks that the patient has SIH, it is a much easier conversation to have to say “We found the fistula – it’s right here.” It’s really a hard conversation to have to say, “We don’t see anything.” I think that the future of this, as we develop different biomarkers, whether it’s by MRI or blood tests, is that hopefully we’ll be able to rule people in having SIH to get a myelo with one test. More than likely, it’s going to take at least two tests to say “Hey, I don’t think you have SIH” because that is such a difficult conversation to have.
And so, what’s my prediction? I wanted to show you just a brief snippet in saying that this work is being done. We’re going to get there. It’s likely going to start with imperfect tools available just to some. We’re going to improve the tools and we’re going to have better tools available to some. But then the goal is to take those improved tools and disseminate them to everyone. So just realizing that these are the normal steps you have to take and just because we get to the point where we have imperfect tools available to some doesn’t mean it’s not a good method. It just means that this the natural progression that we have to go through. So, I kind of had a high-level overview of some of these things. I’d be happy to talk in more detail about any of these topics, but thank you for your attention.