Slides
Transcript
Well, thank you for that very kind introduction and thank you very much to Wouter for giving me this talk, which I’m clearly qualified to be giving because I’m a physiologist and an expert on the skull base. So Wouter asked me to give a talk on skull base CSF leak physiology, and I decided to just talk about some other things as well. There’s a lot of, I think, interesting physiology related to the differences between the skull base and the spine. I’m going to talk a little bit about that. It may or may not make sense at the end, but hopefully it’ll be educational in some way.
This is a Norman Rockwell painting that is called Freedom of Speech. And the title of my talk is “What Happens When You Stand Up and Differences in Physiology of Skull Base and Spinal CSF Leaks.”
So when we think about things that control CSF physiology and hydrodynamics, there’s sort of three main categories of things that influence that. One is arterial inflow into the head, which is controlled by cerebral autoregulation.
There’s venous outflow from the head, and then there are issues of venous — CSF hydrodynamics related to the container that it is in, the compliance of that container, and the effects of gravity. Now, we’re not going to spend a lot of time talking about arterial inflow because actually cerebral autoregulation is very good under most circumstances at maintaining constant levels of arterial inflow. But we are going to talk about venous outflow and CSF hydrodynamics.
And so one of the things that I find very interesting when it comes to venous outflow is that venous outflow from your head is not the same when you are laying down versus when you are standing up. And this has been well demonstrated in a number of physiologic experiments, both in humans and in animals. But basically, when you are laying down, the vast majority of your venous drainage is through the internal jugular vein system.
And they’re distended, and all the other venous channels, particularly the venous channels in the suboccipital region and around the cervical spinal canal, are constricted. But when you stand up, your jugular veins collapse, and there is a shift in venous drainage into the vertebral venous plexus. And you will see those physiologically increase. And this actually changes the way that blood flows, not just out of the head but also changes the hydrodynamics of your spinal canal because those veins will dilate when you stand up. And that pathway, that venous pathway, is a relatively high-resistance pathway compared to the jugular veins. And in fact, if the jugular veins didn’t collapse, you would connect the intracranial pressure to your intrathoracic pressure, and you would have sort of a suction effect that would cause really big shifts in venous pressure between your head and your thorax. So the collapse of the jugular veins is a way of buffering against that really big shift and isolating it from the thoracic venous pressure. So here you can see the difference in these veins between laying down and standing up. And how do we actually know that this occurs in humans?
Well, there have been a number of animal experiments, but this is an experiment that I found that was relatively recent that I quite like. And they took CT scans of the neck in patients who were laying down and standing up. And you may say, how did they take CT scans of the patients who were standing up? Well, they actually have an upright CT scanner that they developed, I think it was Philips maybe, as sort of a one-off for this group in Japan so that they could actually take CT scans of patients in the upright position, which I would really love to have because I think there’s a lot of really interesting questions. But what you can see is that the internal jugular veins, when you’re supine, are quite distended, and they collapse substantially when you stand up, and so you get this shift of flow away from your jugular veins. And where does that flow go?
Well, as I told you just a moment ago, it goes to your cervical epidural venous plexus. These are subtracted images from that same paper, and this is in the supine position. You can see the distended jugular veins, and then when you stand up what you see is this shift towards these cervical epidural veins, which are not opacified and not substantially distended when you’re laying down.
This has also been shown on MR, and this is a very nice figure from a guy named Noam Alperin, who did a lot of work for NASA down in Miami at the University of Miami. He had this open MR where people could lay down or be put upright. What you can see is, in the supine position, venous drainage is through the jugular veins here, but a very different picture when you’re upright. That venous drainage shifts into the vertebral cervical venous plexus.
So again, the vertebral venous plexus is a very rigid conduit, whereas the jugular venous system is a very compliant conduit. So what happens is that you get a much higher resistance or pressure in flow when it’s trying to go through the vertebral venous system, and that prevents sudden drops in intracranial venous pressure, which helps maintain CSF pressure when you’re in the upright position. It also causes this anatomic distension of the vertebral venous plexus, which has the effect of squeezing the thecal sac a little bit like a blood pressure cuff. And you can imagine that in states where you’re hypovolemic or you have a decreased venous return to the heart, such as with dysautonomias like POTS, those vertebral venous plexuses do not fill, and so you don’t get that same squeeze. Patients who are in prolonged microgravity, like in space for instance, also have this effect where they don’t have the shift into the vertebral venous plexus, and so they can actually get a lot of problems in terms of swelling, intracranial swelling.
I guess what I want to summarize in terms of venous outflow is that when you’re horizontal the predominant outflow is through the internal jugular vein, about 75%, and that almost completely reverses when you lay down. You have only a very small percent going through your internal jugular veins. They collapse and they shift to these higher resistance vertebral veins. And this disconnects the intracranial pressure from central venous pressure and prevents against sharp falls in intracranial pressure.
One of the things that I thought was very interesting just talking about the CSF pressure venous connection is this test called the Queckenstedt’s test, and you may not have heard of it, but before the advent of cross-sectional imaging this was how it was determined whether or not a patient had a spinal block. What they would do is they would do a lumbar puncture and connect it to a manometer, and with the patient laying down they would compress the jugular veins essentially like this, I guess, and what you should see is a rise in pressure in the lumbar cistern. And in patients who had spinal blocks you would not see that rise.
For those of you who are thinking about positive pressure myelograms, maybe you just choke your patient a little bit on the table and you can better than infusing CSF. I mean maybe there’s something to it, right? And actually on my flight over here, I read an interesting paper looking at the effect of head position on lumbar opening pressure. And it turns out that if you rotate your head and you compress your jugular veins, you can influence—you can cause increases in your lumbar opening pressure. So a potential source of error when measuring lumbar puncture.
Why don’t we hear about Queckenstedt’s test anymore? Well, because you can see this timeline of the evolution of the test, and then right around here CT was introduced, and shortly after CT was introduced, the test became obsolete.
So we’ve talked a little bit about the venous outflow. Now we’re going to talk about CSF hydrodynamics. And there’s this very wonderful paper by this guy Bjorn Magnaes, who’s a Swedish physician. And he did all these experiments where he actually measured physiology in terms of CSF pressure and the difference between laying down and standing up under various different conditions. And he described two points. One is called the hydrostatic indifference point, or the HIP. And this was the point that was the same whether you were laying down or standing up. And I think another, maybe conceptually easier point to understand, is the ZPS, or the zero pressure in sitting point. And in this position, the pressure of the CSF at this level is equivalent to atmospheric pressure. So above it, it’s negative relative to atmospheric pressure, and below it, it’s positive relative to atmospheric pressure. And in a normal person, the distribution of the ZPS looks like this. So the ZPS is usually somewhere in the upper cervical spine. But this is the occipital protuberance. And you can see in the majority of patients it is not above that level, which means that when you’re standing up, intracranially your CSF pressure is less than atmospheric pressure, and under certain conditions this can change.
So this was a patient, these were a series of patients who had CSF leakage. Now some of these were skull base CSF leaks, but some of these were post-laminectomy leaks. So this is a combination of spinal and intracranial CSF leak, and unfortunately Dr. Magnaes did not discriminate between them. But if you look what happens to the zero pressure point, it shifts downward in patients who have CSF leakage, which means that there’s a larger area, a bigger area, and a bigger gradient of negative pressure relative to the atmosphere above. And you see the same thing happening with the hydrostatic indifference point. It also shifts downwards.
In patients who have hydrocephalus, you have sort of the opposite effect in the pre or in the after shunting treatment. So once they have their hydrocephalus treated, the ZPS, or the ZPS, is down here, whereas before they have treatment it’s up here. So they have elevated intracranial pressure that then shifts downwards. It sort of mimics what happens when you create a spinal leak, right? You’re shunting fluid out of the head, and you create this bigger pressure gradient. Same again with the hydrostatic indifference point.
In patients with subarachnoid block, you also get a caudal shift in that ZPS. So again, you get this downward shift, and this is again making a larger negative pressure intracranial with respect to the atmosphere.
And so when you have a skull base CSF leak, a leak that occurs at this level under normal conditions—oh sorry, if the intracranial pressure is less than atmospheric pressure, you’re not going to drive CSF out at that level. In fact, you may drive air inward, which is why in a lot of patients, particularly immediately after trauma, what we’ll see is pneumocephalus air that goes into the cranial.
And I was trained to do cisternograms as a resident. And when they teach us how to do a cisternogram, you want to see the spot where the CSF leaks. So what do they tell you? They tell you to tell the patient to put their head in the position where they can provoke the leak. And almost all the patients will do the same thing, which is they put their head down between their knees.
So you drop your head down between your knees, and now you’ve inverted the effect of gravity, and all that intracranial pressure is positive relative to the atmosphere. But what it means is that everything in your spine, or almost everything in your spine under normal or pathologic conditions, has a positive pressure with respect to atmosphere, and everything intracranial has a negative pressure with respect to atmosphere. And so they behave differently with regard to CSF leaks.
Now, I told you that we weren’t going to talk about arterial inflow. I’m just going to make one comment about this, and that is normally your CSF pressure, when you transition from sitting to lying down and back, there’s an initial brief spike and then a relatively stable plateau phase. But in patients who have orthostatic intolerance, for a variety of reasons—and this would include patients with dysautonomia conditions like POTS—what you can see is that that baseline is not stable. So there’s an initial sharp increase, but then the CSF pressure actually begins to drift downward and the patient begins to feel faint as their blood pressure drops off, and then they actually may pass out.
So one of the things that happens when you have orthostatic intolerance is that your CSF pressure actually decreases, which is why a lot of patients, I think, with dysautonomias may have symptoms that are very, very similar and sometimes indistinguishable from patients with CSF leak because they’re actually experiencing a drop in CSF pressure.
Now, when it comes to skull base CSF leaks, these are nice illustrations from a paper from the Mayo Clinic. And you can have a variety of different skull base CSF leaks. You can have paranasal leaks through the cribriform plate or into the ethmoid air cells. You can have leaks into the sphenoid sinus. You can have leaks through the tegmen into the mastoid air cells or the middle ear.
But all of these things share one thing in common, which is that it’s leaking into the external environment. So all of these spaces that it’s leaking into are atmospheric pressure, and that’s very different than spinal CSF. So again, when you have a leak above this line when you’re standing upright, it’s typically negative relative to the environment, and so you’re not going to be leaking CSF under most cases. In some cases, you’ll see pneumocephalus. Very frequently, patients will present with meningitis. It’s outside coming in, causing meningitis, and you get provocation in that head-down position.
So here are some examples of trauma cases that I’ve seen. Here’s an anterior skull base CSF leak, and what do you see? You see a little pneumocephalus over the site. Here’s a patient with a temporal bone fracture. What do you see? You see pneumocephalus over that site because air is being drawn in.
This is an exceptional case. This was a case of a 10-year-old that we saw relatively recently at my institution who presented with a history of recurrent meningitis—so multiple episodes of meningitis. Didn’t really have headaches but kept getting meningitis. And what you can see, he’s got this large meningocele going into his sphenoid sinus.
Now he had a surgical repair, a nasoseptal flap to cover this area, this meningocele. And when that happened, he developed orthostatic headache. We were a little confused by this. So we did a myelogram to make sure that there was no spinal leak. And you can just take my word for it, there was no spinal leak. But he did have a skull base CSF leak. And here you can see that meningocele, and you can see contrast coming out and coming down into the nasal cavity. This is the pre-contrast, and you can see that there—that’s not bone, that is in fact contrast. And here you can see contrast in the oral cavity from where this was leaking down. Interestingly, this was his brain MRI. This was pre-repair, he had venous distension.
So he had signs of intracranial hypotension even though this is a skull base CSF leak. This is the exception rather than the rule. After repair, his venous distension went away, and then when he redeveloped the orthostatic headache it enlarged again. So this patient was apparently leaking just enough CSF that he became volume depleted and did develop orthostatic headaches. But like I said, this is the exception rather than the rule. He never had dural enhancement as part of this. And we’ve seen these cases, case series reported, of course the big case from Cedars that everybody references, but also a nice case series from the Mayo Clinic. And what they found—this is a lot of text—but they basically said in the Mayo paper, the one patient with a skull base CSF leak who had dural enhancement had an infratentorial CSF leak.
And before repair of the skull base CSF leak, he went for a myelogram that didn’t show anything. Three years later, the patient was found to have a ventral CSF leak from a disc osteophyte complex, so it was a little bit of a muddy case. But in both of these series, the bottom line was that but for this one exceptional case, none of these patients had dural enhancement.
And so we think that patients who we see with dural enhancement do not really have skull base CSF leaks that communicate with the environment. Occasionally, we’ll see some of these cases where they’re leaking into the soft tissues below the skull base, or they’re leaking into venous fistulas or things like that, but they are not associated with CSF rhinorrhea or otorrhea, and that’s because of the physiology and this zero-pressure point.
So, in summary, the CSF hydrodynamics change between supine and upright positions. Those changes make CSF volume depletion less likely in skull base CSF leak. It’s not impossible, but it’s certainly much, much less likely. It is the exception rather than the rule. And skull base leaks with intracranial hypotension do happen, but are definitely uncommon. And so you should not think of that first. So thank you very much.