Slides
Transcript
Okay, thank you for the kind introduction. Yes, glymphatics and SIH. Why does that matter? Well, let’s begin with the original description from Maiken Nedergaard. You know the glymphatic system is a para-arterial influx of CSF, then a connection to the interstitial fluid by aquaporin-4 channels located in astrocytic endfeet, and then flow through the brain and the venous outflow again. So simply spoken, it is just a CSF flow through the brain. We can have serial gadolinium cisternography at different points of time after injection to see this CSF flow through the brain.
These are T1 relaxation time maps, difference between pre- and post-contrast, and you see there is a difference in the basal subarachnoid cisterns and then in the perivascular CSF. So we can show that CSF flows through the brain. The first who did that was Eide and Ringstad, a neurosurgeon and a neuroradiologist from Oslo, Norway.
And we tried to replicate that, but we could not replicate it in the first patients, and we included all patients in which we did intrathecal gadolinium application. Well, there were patients who had no leaks. There were patients who had idiopathic intracranial hypertension and there were patients with spinal CSF leaks. The contrast patient with two leaks has a tegmen tympani and cribriform plate, similar to what Peter Kranz showed us before. That is a hypertension IIH patient, and here we have a SIH patient with a CSF-venous fistula. So putting all these patients together did not give any results.
So we said we have to separate the patients with SIH from those with no leaks or IIH. Then you see at the basal cisterns there’s a signal drop, the T1 relaxation time decreases, and then it goes through the brain, and it makes a difference whether the patient has SIH or whether they have no SIH. That means the original description by Nedergaard says the arterial pulsation drives the CSF into the brain. It’s not only a matter of the arterial pulsations, it’s also a matter of the CSF pressure, the intracranial CSF pressure. And this is interesting because we know that the glymphatic system is more active during sleep. But one reason why it is more active during sleep is because the CSF pressure intracranially is higher when lying down. That is not the only reason, because recently Maiken Nedergaard showed that the vasomotion, the power bed, pumps the CSF into the para-arterial spaces, is also different between sleep and non-sleep.
So that’s not only the reason that we are lying down, that we have a better glymphatic transport during sleep. So the CSF inflow, in summary, is influenced by different things: cardiac pulsations, respiration, that is what we use for detection of CSF venous fistula, posture, and gravity. We know when lying down it’s higher, but we also know that’s higher in chronic obstructive lung disease, in sleep apnea. And what we have learned from the study, CSF pressure makes a role. But what is about the venous outflow? Can we image that?
We know the venous outflow is from CSF in arachnoid granulations in the venous system, and there is a second way which is relevant for the glymphatic system, that is in lymphatics, and these lymphatics are either located in the parasagittal dura or in the nasal mucosa. That is a schematic illustration, the CSF in the parasagittal dura.
That was a recent paper or former paper from Maria Absinta which used FLAIR after gadolinium to show the tiny lymphatic vessels. We tried to replicate it, but we could not really replicate it, and we say okay, we have a better imaging modality, that is gadolinium cisternography. These are the schematic drawings, and that is what we can show. We have the nasal mucosa. We can have signal intensity time curves in the nasal mucosa and in the parasagittal dura. But what you see on these images is also interesting. You have the arachnoid granulations. But we have not to measure in the granulations, but in the parasagittal dura itself. I’m showing these images because years ago I asked Jürgen Beck and also our anatomist to show me the spinal granulations. We see them intracranially, but we don’t see them in the spine. That is an illustration by Peter Kranz years ago. Arachnoid granulations—why don’t we see them on imaging with high-resolution imaging? Maybe the reason is that they are different on a spinal level. Here we have clear-cut 8 mm large arachnoid granulations. In the spine we have really that are interconnected with the veins, so small that we likely can’t see them.
Okay. So we took care that we did not measure the signal intensity in the arachnoid granulations but in the parasagittal dura, and then you can see that there is a signal increase related to the signal in the vitreous bodies of the eyes. So this is clearly a glymphatic transport into the parasagittal dura and into the nasal mucosa. These are the signal intensity time curves. To make the story short, there is a difference in the nasal mucosa and in the parasagittal dura. The parasagittal dura occurs a lot later than the CSF drainage in the nasal mucosa.
Some words to the DTI along the perivascular space index. We have tried to replicate these findings. The idea behind it, that the glymphatic transport perpendicular to the association fibers displayed in green, is lower. We have, I think, a strong physics department. No one could duplicate it. So here we signal intensity measurement here, and a complex formula we could not replicate it. So we have a similar method, and this similar method is maybe superior to the DTI-ALPS index. We decompose the diffusion signals in three components. One component is very short diffusion distance, and this is only possible in the neurons or length along the neurons. The second one represents the axons, represents the neurons as a neuropil. And the third component represents the free interstitial fluid. We have applied this method for several diseases.
For example, here in COVID patients who have neurological symptoms after the acute phase, and we can show there is a water increase and increase of the interstitial fluid in this patient. We also can combine that with molecular with ultrastructural histological findings, here in a patient with temporal lobe epilepsy and anterior temporal lobe blurring. Of course, the next step is to show that in patients with spinal CSF leaks before and after treatment, but it is not easy tasks because probably related also to partial volume effects. But what you can see from this image is in the hippocampus, where we have the strongest effects, there is a difference pre and post and pre and compared to healthy controls.
So why that is interesting for us? Well, we want to understand that brain fog associated with SIH. Only a few patients have really spinal dementia that needs such activities of daily living. They are handicapped. Most of them have a mild cognitive impairment—we could say that. And it is unlikely that you can solve that just from the CSF perspective or from the venous distension or whatever. It may be that is related to what parts of the brain are displaced. That is my favorite impression at this time. So we have that many patients with epilepsy, and that is probably related to the suction of the fronto-orbital portions and to the suction of the hippocampal head. But we haven’t understood. We have some patients who have apathy. We have a lot of patients who are not a worry, who have a disinhibition.
You can go into the patient and you see that he has hypertension, and if you have it takes 30 seconds to find out that this is a clear patient with disinhibition. But we want to find out, it’s maybe just microscopically related to the deformation of the brain, but it also could be related here as spinal MCA patients, just to that interstitial glymphatic fluid transport or a glymphatic transport.
The summary at this point is not complete. It just shows with different MRI sequences: gadolinium cisternography, T1 relaxometry, a high-resolution compressed black blood space sequence, and diffusion microstructure imaging, that there is the CSF flow lower in patients with SIH. Thank you for listening to me.