Slides
Transcript
Interesting, interesting case, Hans. We operated a similar patient with an ovarian cyst that was operated by gynecologists transabdominally, resulting in a huge leak, similar case.
Thanks for the next topic: superficial siderosis or cerebral amyloid angiopathy versus SIH. I think superficial siderosis is also one of these diseases where there has been a paradigm shift recently, and there are big names that have changed the disease and how we perceive this disease.
I think you’ve been at a special meeting recently. I think the names are Kumar and Werring and Schievink and many others. But nowadays I think this is not evidence-based medicine, it’s just really subtle good observation. But I think it’s clear that when I was encountering these patients there was extensive whole spine angiography, which took the whole day, very, very complication-rich. And nowadays we start looking for the leak if you see such a patient. So there’s a clear paradigm shift.
And it’s considered recently to be another form of duropathy or arachnopathy and belongs to the spectrum of spinal CSF leaks and SIH, which is, I think, significant—another disease where we have a paradigm shift.
Superficial siderosis is classically discriminated into infratentorial classical siderosis, or into the diffuse type, which is more in the cerebral, not in the cerebellum. And we are basically again looking at the disease that is diagnosed by imaging, by MRI. You barely see it in the T2, but you see it, and you clearly see it in this over-clearly depicting SWI sequences.
Cerebral amyloid angiopathy, which was considered to be the same spectrum, is probably a different disease. And as you can see here, it looks different. It’s not so much focused in the posterior fossa or around the brainstem or around the pons. It’s rather in the hemispheres, and most often you have more or less large or tiny microbleeds, but also signs of sulcal and superficial siderosis.
Cerebral amyloid angiopathy has several distinctive features, how you can probably discriminate it from SIH. It has a different appearance on imaging. There is usually a larger bleed associated with it and many other signs like lacunar infarcts, large ICH, microbleeds, white matter hyperintensities, cortical thinning. There’s also the sign of superficial siderosis, but plus all the other signs here. So you can see it’s probably really a different disease.
What’s the reason for it? It is thought to be that amyloid beta protein is not properly transported from the interstitial fluid via the intramural periarterial drainage, or summarized recently in the glymphatic pathway, that seems to be distorted. And that leads to an accumulation of amyloid beta in the vessels.
This is not so typical for CAA. So you can see there might be difficulties in the differential diagnosis between siderosis due to a spinal leak or CAA. The hallmark is the superficial siderosis in the sulci in the brain adjacent to the sulci. And it’s also very en vogue—superficial—not superficial siderosis. CAA recently, in intracranial hemorrhage and ICH in the arm just this year or a couple of weeks ago, published that the so-called Boston criteria for diagnosing CAA on imaging need to be confirmed with a tissue-based paradigm with biopsies. One can clearly find that there’s also CAA in the patients with ICH. And if you remember when I—the diagram—so probably the big ICH is the lesion that makes the patients symptomatic. So these patients with CAA come to the hospital, to the emergency room, usually with a stroke, with an acute hemorrhagic stroke, but you find the same changes in biopsy.
This is a brain biopsy in one of these papers, and you can see that the amyloid, so to say, is mainly in the vessels, around the vessels, but also diffusely in the brain. So this is a biopsy of an ICH patient with CAA, not with superficial siderosis due to a leak. We also find this very en vogue and published or submitted our results for this very same topic. What is about the tissue-based diagnosis in ICH in patients with CAA, and they have found the same—that if we do remove these ICH hemorrhagic strokes, we take tissue biopsies of the tissue in between and can find the same findings. So, around the arteries but also a little bit in the brain, these beta-amyloid. And we’re trying to build up a center that is using modern technology, stimulated Raman technology, that is out of our neuro-oncology service, where we can measure IDH1, 1p/19q, and ATRX in 90 seconds in the operating theater to finally come to a surgical decision-making—whether to go on with tumor surgery or stop tumor surgery. And we try to duplicate this paradigm also to acute stroke. But this is only a short detour.
Coming back to superficial siderosis. Classically it was divided in the localized type of superficial siderosis and in the infratentorial type. And giving you this is what I mentioned earlier. The classic etiology included a spinal pathology, meaning that, Horst, you can tell this better, you had to analyze the whole spine angiographically, which is quite bothersome to find any tiny fistula or lesion. Usually nowadays we did MRI and skipped that, and the rest was repeated bleeding, spinal trauma, AVMs, tumors, ependymomas of the lower spine classically, as opposed to the supratentorial secondary superficial siderosis due to just SAH subarachnoid hemorrhage or AVMs or cavernomas and so on.
Why is this interesting and what’s happening with siderosis? And this is what I’ve learned reading these papers recently. It’s just that siderosis is just a sign of the body to prevent damage. So the real damage is done by free iron and not by hemosiderin. It’s just biochemistry, let’s say. And this is interesting.
You see in these papers it seems to be clear it’s from extraarachnoid fluid collections and from a dural tear, so not anymore from any AVMs or cavernomas, and the blood degradation products are bound with ferritin. And ferritin is produced by the glia of the superficial brain, so to say, and this mechanism has a limit. Once there is continuous and repetitive bleeding, the mechanism is overwhelmed, and then the ferritin and the iron is deposited as hemosiderin in the brain. But this is scavenging, and if the bleeding continues there’s more and more free iron, and free iron is the thing that makes the damage. So we don’t have a good parameter for surveillance. We do not know whether it’s good if the hemosiderin gets less or more, because the damage happens anyway with the iron.
Now coming more closely back to our topic, to spinal leaks, to duropathies, and so on. If I read the literature correctly—and this is so interesting—because here we don’t have prospective trials and many patients. It is just expert-based due to meticulous observation, and still it changes the disease and how we perceive the disease and how we treat these patients.
I think this is really fascinating just with good observation in two or three centers, three or four bright neurologists and neurosurgeons change how we accept 2025 this disease, and this is now Wouter’s observations. I think you’ve published it in 2023. Probably this is the solution so to say. These tiny little vessels, it’s not so overwhelming, but there are highly vascularized epidural base and probably these vessels are leaky, and probably these vessels are leading to an overwhelming, in the end, iron overload of the system. Ferritin is not sufficient to bind it, and this causes toxicity, and it’s an ugly disease. It’s nothing you want to have. You get classically—you have gait problems, you get deaf, and in the long run you get demented. So there is real damage to the brain. We did some studies with superficial siderosis in Bern and in Freiburg a long, long time ago, and same pattern if we have time seems to be important.
So if we have siderosis grade that is increasing, it’s in patients with a longer, longer history, and this is also very strange. I mentioned this earlier this morning, that in patients with a longstanding leak, the pressure is not decreased anymore, it’s not normal, it’s even increased anymore. And in patients with long-standing siderosis, meaning probably a long-standing leak, CSF opening pressure was higher than normal. So this seems to be the same population. And the marker—there is not much published—but one marker is, of course, ferritin, the protein that is produced by the superficial layer of the brain by the glial cells, and it was way higher in patients with visible superficial siderosis on MRI. So I think we can these days clearly add another disease caused by spinal leaks: superficial siderosis. We discussed it earlier, and nobody of us wants to have this. And what to do—this impacts not only what to do, but also what to do with incidental leaks that we find, and it is to my understanding clearly a sequelae of spinal leaks.
So it’s called nowadays in the reviews, in the papers another form of duropathy, a duropathy. I think it’s probably also related to the arachnoid. Again, this wonderful publication, also just meticulous observation by MacRobert in 1918, that the arachnoid plays a crucial role, probably only whether the dural defect is able to heal or not. So it could also be a kind of an arachnopathy, superficial siderosis. And don’t blame me for that. I believe membranes are important, and probably the membranes are the source of the bleeding, because the surface is big. And Wouter, correct me in these forms of postoperative pseudomeningocele, where you also have this large defect with a big surface of non-physiological tissue, you have a high risk of getting superficial siderosis. So it can also be a membranopathy, so to say, if you want, because we believe in membranes. And there’s also another form of membranes, this actually Wouter describe to me also very often, it’s the pseudo dura. And I remember the best spine surgeons I was operating with SIH cases calling me, now I’m, I’ve opened the dura, please come to the OR, let’s proceed, and it was not the dura. Very, very, very experienced spine surgeons. Because this dorsal membrane, I haven’t touched it in the earlier talk. These dorsal membranes look completely different. They are not these thin translucent arachnoid-like membranes. They look like dura and they are highly vascularized.
And this is a photo from a surgery and this is the rendering. So it’s really under the dura there’s another dura, the dura propria, and this is highly vascularized. Maybe this is also a source or possible source of the erythrocytes that leak in the long run to superficial siderosis. And as I told you, it’s a thick, a very thick membrane, the neodura, the pseudodura, with a lot of vessels. And if you cut in it, it really bleeds a lot. And this is probably also the cause that we get some patients operated on by other centers, that the dura is highly vascularized. You cannot suture the dura. I believe, I have no proof, they only opened this pseudodura and not the dura propria.
A little bit more science. 1,598 patients from the Cedars center, and 57 patients had SIH and superficial siderosis. And Wouter analyzed this, and it was mainly, unfortunately you didn’t make a table, so I had to read the whole paper and find the numbers. Please correct me if it’s wrong, but 10% had ventral leaks, so it was a clear predominance of ventral leaks. 4% percent with dural ectasias, and even, and I think you are the only one who found that, with CSF fistula. So there might be a backflow from venous blood into the subarachnoid space, and that’s it basically.
Again it was a long latency, 126 months, if you can estimate from the history the onset of orthostatic headache for instance, then you can estimate roughly the latency. It is 10 years or more. And if you look at these patients vice versa, the other way around, not all the patients, 80% of the patients with superficial siderosis had ventral leaks. So graphically you can say you have an increasing and ever-increasing risk of developing superficial siderosis with time.
And I might add a personal note, because we started to discuss what to do with an incidental leak with a SLEC-positive patient, young patient, who is asymptomatic. The risk of getting superficial siderosis is in the long run quite, quite high. And superficial siderosis is the end stage. Do you want to have over 10–15 years, as a young person, a continuous toxic overload of your brain cells that is somewhat exhausted, and once it’s exhausted it’s irreversible? So I think we should consider probably treating incidental spinal leaks and probably with registries, and we have to follow up these patients. We have to combine and write down our experience.
It’s not that we deliberately shed some light into this fancy disease. Another thing is that probably we can discuss tomorrow, since Lalani so aggressively said that percutaneous and endovascular procedures are good in spinal leaks. This is an example for ventral leaks, and none, zero, none of these leaks could have been treated with any percutaneous procedure in these patients, and they have all been followed up so we know the results. Surgical repair resulted in resolution of all the CSF leaks and was associated with 5% risk of injury.
So this is an old series published recently, but we started of course before this was all so crystal clear. I remember with my friends in Bern, when we had the first siderosis patient we said, this must be a leaker, this must be a leaker, all these SLECs and so on, and so we gathered these patients. In the meantime, all these fantastic papers have been published, and it seems to be clear. But this is what Amir [El Rahal] put together that starts nowadays, so to say. And we have a patient with superficial siderosis, SLEC- positive, and then we find old pictures. We look back 10 years history, and 10 years earlier there was also SLEC, no superficial siderosis, but we found orthostatic headache. So we think, okay, that’s a clear hint. It’s the same, it’s the same disease. It’s a long-term sequelae of the same disease.
So we asked the questions at the time, is there association of spinal leaks and superficial siderosis? And that’s probably more important now since we know the etiology – does surgery help? Only 12 patients, only ventral leaks. We were not so sophisticated, we didn’t even find CSF leaks at the time the study started, so only ventral leaks. Six women, six men, 60 years of age, all SLEC-positive, and all were microsurgically sealed, as in Wouter’s series, all were operated on and all were closed. Blood patching didn’t help at all. And the interesting thing is that we know that, okay, long latency again, 10 years or so. But the interesting thing is it’s a small series, but it’s very encouraging. 70% of our patients did improve, so not only we stopped the disease, but they did improve, and this is very encouraging I think. And who did improve? The patients with a shorter history. Makes kind of sense.
And all patients with SIH improved. The SIH symptoms, the orthostatic symptoms, improved earlier. So this is the same story in a graphic. Patients where we thought the history is longer than 10 years, even 30 years, they didn’t improve. The progression stopped. But patients with a shorter history, they were at least, when we did the follow-up, better than before. This is what I already told you.
So to summarize, 70% of the patients improved, 30% were stable, low complications. So we clearly favor surgical sealant, surgical closure of any leak in these patients. And it will turn out whether it’s really a duropathy or arachnopathy or a membranopathy. It’s again a disease where we should or must work together to shed light on this, on the pathomechanism.
And also this another groundbreaking work from Cedars, this brain sagging dementia. We don’t know what’s the role of this longstanding disease. There’s iron, of course ferritin, hemosiderin, but there’s also brain sagging. There’s also distortion of the brain, and this might also cause cognitive decline, because in the end, if you don’t treat superficial siderosis, the patients get cognitive decline and dementia. And there’s even a more subtle form of cognitive deficits in patients with spinal leaks that also add on to the chemical toxicity of siderosis, which we call mild cognitive impairments. Sometimes we call it spinal dementia. Horst coined the term. I love the term, but usually we get really burned by neurologists. “There is no spinal dementia. You are crazy. Don’t use this term.” Probably they’re right. I don’t know, but at least they wake up. And it’s not good. A spinal CSF leak changes your cognition.
And I don’t know whether it’s published or not yet, Katharina, but at least we measured in a couple of patients, just a handful, over the last year or so. And you can measure with some standardized tests the day before surgery and even immediately after anesthesia the next day, a highly significant impact on cognition if you close the leak. This all adds up to that spinal leaks are chemically, mechanically combined, really nothing that should be there. And this is also one argument if you discuss what to do with an open asymptomatic leak. Probably there is no headache, but have you really neuropsychologically assessed your patient? Was there a job change, or is there a divorce, or do the relatives say, “He’s a little bit slow recently”? I think a spinal CSF leak is not beneficial for your cognition.
And the same is that we of course know the super severe brain sagging syndrome now, but there are subtle changes probably not visible with the eye that we also measured with AI, and there is a clear distortion of the brain even if you don’t see it on MRI.
Please start the video to bore you once again with the superficial siderosis video. I think this summarizes a lot we’ve discussed today. I’ve showed you this, I’ve shown you this video this morning, but it summarizes everything. The SLEC positive, you have infratentorial blooming of the SWI images, easily detectable. It’s mainly in ventral leaks. It’s longstanding, huge SLEC. You can do easily surgery; blood patching doesn’t do anything in these patients, to my opinion. And then again, what I would like to show you, I cannot forward it here, but 30 seconds and you can see it. It’s probably all due to these membranes that form, so we need to open it, seal the leak, and I don’t think that these membranes are any good either. It’s just a protective mechanism.
You’ve seen it then, and only if I cut into the membrane CSF is oozing out, and then you see finally the tiny little hole, the CSF leak that is causing all this problem, even superficial siderosis. And to see the signs of siderosis also in the spinal canal. And sometimes if you operate these patients, the spinal cord is just yellow. And here you can see a little yellow rims and some chromia around the dural tear. You can see in a second. So, opening the dura, the green stuff without fluorescence mode, it’s even greener. And then this, I find this super fascinating. This is also the reason for superficial siderosis. This tiny little hole you can see in a second. So this here, and if you look closely there’s yellow stuff around it. Okay, good. So again, thanks to my team and thank you for your patience.