Understanding Brain Sagging Dementia and CSF Leaks

Spontaneous intracranial hypotension reflects a disruption of the normal continuous production, circulation, and reabsorption of CSF. Diagnosis requires the recognition of common and uncommon presentations, careful selection and scrutiny of brain and spine imaging, and, frequently, referral to specialist centers.

 In this episode, Gordon Smith, MD, FAAN speaks with Jill C. Rau, MD, PhD, author of the article “Clinical Features and Diagnosis of Spontaneous Intracranial Hypotension” in the Continuum® June 2025 Disorders of CSF Dynamics issue.

Dr. Smith is a Continuum® Audio interviewer and a professor and chair of neurology at Kenneth and Dianne Wright Distinguished Chair in Clinical and Translational Research at Virginia Commonwealth University in Richmond, Virginia.

Dr. Rau is an assistant professor of clinical neurology at the University of Arizona, School of Medicine-Phoenix in Phoenix, Arizona.

Additional Resources

Read the article: continuumjournal.com

Subscribe to Continuum®: shop.lww.com/Continuum

Earn CME (available only to AAN members): continpub.com/AudioCME

Continuum® Aloud (verbatim audio-book style recordings of articles available only to Continuum® subscribers): continpub.com/Aloud

More about the American Academy of Neurology: aan.com

Social Media

facebook.com/continuumcme

@ContinuumAAN

Host: @gordonsmithMD

Full episode transcript available here

Interview with Jill Rau, MD

Dr Jones: This is Dr Lyell Jones, Editor-in-Chief of Continuum. Thank you for listening to Continuum Audio. Be sure to visit the links in the episode notes for information about earning CME, subscribing to the journal, and exclusive access to interviews not featured on the podcast.

Dr Smith: This is Dr Gordon Smith. Today I'm interviewing Dr Jill Rau about her article on clinical features and diagnosis of spontaneous intracranial hypotension, which she wrote with Dr Jeremy Cutsworth-Gregory from the Mayo Clinic. This article appears in the 2025 Continuum issue on disorders of CSF dynamics. I'm really excited to welcome you to the Continuum podcast. Maybe you can start by just telling our listeners a little bit about yourself?

Dr Rau: Hi, thanks for having me. I'm really honored to be here, and I really enjoyed writing the paper with Dr Cutsforth-Gregory. I hope you guys enjoy it. I am the director of headache medicine at the Baba Bay Neuroscience Institute at Honor Health in Scottsdale, Arizona. I'm also currently the chair of the special interest group in CSF Dynamics at the American Headache Society, and I've had a special interest in this field since I first watched Dr Linda Gray speak at a conference where she talked about spinal CSF leaks and their different presentations. And they were so different than what I had been taught in residency. They're not just the post-LP headache. They have such a wide variety of presentations and how devastating they can be, and how much impact there is on someone's life when you find it and fix it. And I've been super interested in the field and involved in research since that time. And, yeah. Love it.

Dr Smith: Well, thanks for sharing your story. And as I reflected on our conversation ahead of time and have been thinking about this issue… this is a cool topic, and every time I read one of these manuscripts and have the opportunity to speak with one of the authors, I learn a ton, because this was something that wasn't even on the radar when I trained back in the 1800’s. So, really looking forward to the conversation. I wonder if you could really briefly just summarize or remind for everyone the normal physiology about CSF dynamics, you know, production, absorption, and so forth?

Dr Rau: So, the CSF is the fluid that surrounds the brain and the spinal cord, and it's contained by the dura, which is like a canvas or a sac that covers that whole brain and spinal cord. And within the ventricles of the brain, the choroid plexus produce CSF. It's constantly producing and then being reabsorbed by the arachnoid granulations and pushed into the venous space, the cerebral sinuses, venous sinuses. And also some absorption and push into the lymphatics that we've just learned about in the past year. This is kind of new data coming out, so always learning more and more about CSF, but we know that it bathes the brain and the spinal cord, helps keep some buoyancy of the brain as well as pushing nutrients in and pulling out metabolic waste. And it sort of keeps the brain in the state of homeostasis that's happy. And so, when there's a disruption of that flow and the amount of fluid there, that disrupts that, that can cause lots of different symptoms and problems for people.

Dr Smith: One of the many new things I learned is that even the name of this---spontaneous intracranial hypotension---is misleading. And I think this is clinically relevant, as we'll probably get to in a moment, but can you talk a little bit about this? Is this really like a pressure disorder or a volume disorder?

Dr Rau: Yeah. It's almost certainly a volume disorder. We do see in some people that they have low pressure, and it’s still part of the diagnostic criteria. But it's there because if you have a low pressure, if you measure an opening pressure and it's below six, if you're measuring it in the spine in the right place, then you have indication that there's low volume. But there's over 50% of people's opening pressure who have a spinal CSF leak, have all the symptoms and can be fixed. So, they have normal pressure in 50% of the people. So, it is an inaccurate term, hypotension, but it was originally discovered because of the thought that it was a low-pressure situation. Some of the findings would suggest low pressure, but ultimately, we are pretty sure it's a low-volume condition.

Dr Smith: Another new thing that I learned that really blew me away is how bad this can be. I did a podcast with Mark Burish about cluster, and I was reminded many cluster patients are pushed to the point of suicidal ideation or committing suicide by the severity of pain. And this sounds like for many patients it's equally severe. Can you maybe paint a picture for our listeners why this is so clinically important?

Dr Rau: A large number of people, even people who are known to have leaks because they've had them before or they've releaked, they have a lot of brain fog and cognitive impairment. They often have severe headaches when they're upright. So, orthostatic headache is probably the number one most common symptom, and those headaches are one of the worst headaches out there. When people stand up, their fluid is not supporting the brain and there's an intense amount of pain. And so, they spend a large portion of their lives horizontal. And there's associated symptoms with that, it's not just headache pain and brain fog. There's neck pain. There's often subsequent disorders that accompany this, like partial orthostatic tachycardia syndrome. We don't know if that's because of deconditioning or an actual sequela of the disease, but it's a frequent comorbidity. We have patients that have extreme dizziness with their symptoms, but many patients are limited to hours, if that, upright per day, combined, total. And so they live their lives, often, just in the dark, lots of photophobia, sensitive to the light, really unable to function. It's also very hard to find and so underrecognized that a lot of patients, especially if they don't have that really clinical symptom of orthostatic headache. So, it's often missed. So, they're just debilitated. You know, treatments don't work because it's not a migraine and it's not a typical headache. It's a mechanical issue as well as a metabolic issue and not found, not a lot helps it.

Dr Smith: So, you know, I have always thought about this as really primarily an orthostatic symptom. I wonder if you can talk about the complexity of this; in particular, kind of how this evolves over time, because it's not quite that simple. And maybe in doing so, you can give our listeners some pearls on when they should be thinking about this disorder?

Dr Rau: A large portion of people do have headache with spinal CSF leak, in particular, spontaneous intracranial hypertension- hypotension, excuse me. And that's something to be thought about, is that there are spontaneous conditions where people have either rupture of the dural sac, or an erosion of the dural sac, or a development of a connection between the dura and the venous system. And that is taking away or allowing CSF to escape. In these instances that patients have spontaneous, there may be a different presentation than if they have, like, a postdural puncture or a chronic traumatic or iatrogenic leak. And we're not sure of that yet, but we're looking into that. Still, the largest presentation is headache, and orthostatic headache is very dominant in the headache realm. But over time, patients’ brains can compensate for that lack of CSF and start overproducing---or at least we think that's probably what's happening. And you may see a reduction in the orthostatic symptoms over time, and you may see an improvement in the radiographic findings. So, there are some interesting papers that have been published that look at these changes over time, and we do see that sometimes within that first three to four months; this is the most common time to see that change. Other patients may worsen. You may actually see someone going from looking sort of normal radiographically to developing more of a SIH-type of picture on the brain. And so it's not predictable which patients have gone from orthostatic to improvement or the other way around, both radiographically and clinically.

So, it can be quite difficult to tell. So, for me, if I have a patient that comes to me and they're struggling with headache… if it's orthostatic, very clearly orthostatic: I lay down, I get considerably better or my headache completely goes away. And then when I stand up, it comes on relatively quickly, within an hour. And sometimes it's a worsening-throughout-the-day type of thing, it's lowest in the morning and it worsens throughout the day. These are the times that it's most obvious to think about CSF leak. Especially if that headache onset relatively suddenly, if it onset after a small trauma. Like I've had patients that say, you know, I was doing yoga and I did some twists and I felt kind of a pop. And then I've had this headache that is horrible when I'm upright but is better when I lay down ever since, you know, since that time. That's kind of a very classic presentation of spinal CSF leak or spontaneous intracranial hypotension. Maybe a less common presentation would be someone who comes to you, they've had a persistent headache for a couple years, they kind of remember it started in March of a couple years ago, but they don't know. Maybe it's, you know, it's a little better when they lay down. It may be a little worse when they're up moving around, but so is migraine, and it's a migrainous headache. But they've tried every migraine drug you can think of. Nothing is responding, nothing helps. I'm always looking at patients who are new daily, persistent headaches and patients who aren't responding to meds even if it's not new daily, but they have just barely any response. I will always go back and examine their brain imaging and get full spine to make sure I'm not missing. And you can never be 100% sure, but it's always good to consider those patients to the best of your ability, if that- have that in the back of your mind.

Dr Smith: So obviously, goes without saying, this is something people need to have on their radar and think about. And then we'll talk more about diagnostic tools here in a second. But how common is this? If you're a headache doc, you see a lot of patients who have intractable headaches. And how often do you see this in your headache practice? Now you’re- this is your thing, so probably a little more than others, but, you know, how common will someone who sees a lot of headache encounter these patients?

Dr Rau: If you see a lot of headache, I mean, currently the thought is it's about 5 in 100,000. That was from a study before we were finding CSF venous fistulas. I think a lot of us think it's more common than that, but it's not super common. We don't have good estimates, but I would guess between 5 and 10 for 100,000 persons, not “persons who come to a tertiary headache clinic with intractable headaches”. So, it's hard to gauge how frequent it is, but I would say it's considerably more frequent than we currently think it is. There's still a group of people with orthostatic headaches that we can't find leaks on; that, once you treat other things that can cause or look for other things that can cause orthostatic headaches. So, there may be even still a pathophysiology out there that is still a leak type. Before 2014, we didn't even know about CSF venous fistulas. And now here we are; like, 50% of them are CSF venous fistulas. So, you know, we're still in a huge learning curve right now.

Dr Smith: So, I definitely want to talk about the fistulas in a second. But before moving on, one of the things that I found really interesting is the wide spectrum of clinical phenotype. And we obviously don't have a lot of time to get into all of these different ones, but the one that I was hoping you might talk about---and there's a really great case, and you're on bunch of great case, a great case of this---is brain sagging dementia, not a term I've used before. Can you really briefly just tell our listeners about that, because that's a really interesting story and a great case in your article?

Dr Rau: Yeah. So, brain sag dementia is a… almost like an extreme version of a spontaneous intracranial hypotension. Where there is clear brain sag in the imaging---so that's helpful---but the patients present kind of like a frontotemporal dementia. And when this was first started to being determined, you could turn the patient into Trendelenburg, and sometimes they would improve. There are some practitioners that have introduced fluid into the thecal sac and had temporary improvement. Patching has improvement, then they leak again, sometimes  not. But the clinical changes with this have been pretty tremendous to be able to identify that that's a real thing. And in some cases, out of Cedars Sinai, you know, who does a lot of the best research in this, they've had lots of cases where they can't find the leak, but there's clear brain sag that fits with our clinical picture of CSF leaks. So, we're on a learning curve. But yeah, this- they really present. They have disinhibition and cognitive impairment that is very similar to frontotemporal dementia.

Dr Smith: Well, so let's talk about what causes this. You mentioned CSF venous fistulas. I mean, that was reported now just over a decade ago, it's pretty amazing. That accounts for about half of cases, if I understand correctly. What are the other causes? And then we'll talk more about therapy in a minute, but what causes this?

Dr Rau: So, within the realm of spontaneous, you know, we say it's spontaneous. But the spontaneous cases we account for, they can be tears in the dura, which are usually sort of lateral tears in the dura. They can be little places that rubbed a hole, often on an osteophyte from the spine. They can come from these spinal diverticuli. So, I always describe it to my patients like those balls that have mesh and squishy, and you squeeze them in the- through the mesh, there's the extra little bubbling out. If you think of like the dura bubbling, out in some cases, through the framing of the spine, right where the spinal nerve roots come out, they should poke out like wires from the dura. But in many cases they poke out with this extra dura surrounding them, and we call that spinal diverticuli. And if you imagine like the weakening of where you squeeze that, you know, balloon through your fingers, in those locations, that's a very common place to find a CSF leak, and you can imagine that the integrity of the dura there may be less than it would be if it were not being expanded in that direction. And that's often the most common place we see these CSF venous fistulas.

So, you can get minor traumas; like I said, it can be spontaneous, like someone just develops a leak one day. It can be rubbed off, and it can be a development of a connection between the dura and the venous system. There are also iatrogenic causes, but we don't consider them spontaneous. But when you're considering your patients for spontaneous cases, you should consider if they've ever had chronic---even long, long time ago---had any spinal implementation, procedures near the spine, spinal injections, LPs in the past, and especially women who've had epidurals in pregnancy.

Dr Smith: All right, so we see a patient, positional severe headache, who meets the clinical criteria. Next step, MRI scan?

Dr Rau: Yeah. So, the first thing is always to get the brain MRI with and without contrast. Most places will have a SIH or a spinal CSF leak protocol, but you should get contrast because one of the most pathognomonic findings on brain MRI is that smooth diffuse dural enhancement. And that's a really fantastic thing when you find it, because it's kind of a slam dunk. If you find it, then you will see other findings. It almost never exists alone. But if you see that, it's pretty much a spinal CSF leak. But you're also looking for subdural collections, any indication of brain sag. We do have these new algorithms that have come out in the past couple of years that are helpful. They're not exclusionary---you can have negative findings on the brain and still have spinal CSF leak---but the brain MRI is extremely helpful. If it's positive for the findings, it really does help you nudge you in the direction of further investigations and treatments.

Dr Smith: And what about those further investigations and treatments, right? So, you see that there's findings consistent with low pressure, and I guess I should say low intracranial CSF volume. Be that as it may, what's the next step after that?

Dr Rau: Depends on where you are and what you can do. I almost always will get a full spine MRI: so, C spine, T spine, and L spine separately. Not, you know, we don't want it all in one picture, because we want to get the full view. And you want to get that with at least T2 highly- heavily T2 weighted with fat saturation in at least the sagittal and axial planes. It's really helpful if you can get it in the coronal planes, but we have to have- often have good talks with your radiologist to get the coronal plane. I spoke about the spinal diverticuli earlier, and I want to clarify a little bit of something. The coronal image will show those really nicely. It's interesting, but 44% of people have those. So just having the spinal diverticuli does not indicate that you have a leak. But if you have a lot of those, there may be more likelihood of having leak than if you don't have any of those.

So, I will get all of those and I will look at them myself, but I've been looking at them myself for a long time. But a lot of radiologists in community hospitals, especially not- nonneuroradiologists, but even neuroradiologists, this isn't something that's that everybody's been educated about, and we've been learning so much about it so rapidly in the past ten years. It's not easy to do and it's often missed. And if it's not protocoled properly, the fat saturation's not there, it's very hard to see… you can have a leak and not see it. Even the best people, like- it's not always something that's visible. And these CSF venous fistulas that we talked about are never visible on normal MRI imaging. Nonetheless, I will run those because if I can find a leak---and 90% of the ones that are found on MRI imaging are in the thoracic spine. So that's where I spend the most of my time looking. But if you find it, that's another thing to take to your team to say, hey, look, here it is, let's try and do this, or, let's try and do that, or, I've got more evidence. And there are other findings on the spine; not just the leak, but other findings, sometimes, you can see on spine that maybe help you push you towards, yes, this is probably a leak versus not.

Dr Smith: So, your article has a lot of great examples and detail about kind of advanced imaging to, like, find the fistula and what not. I guess I'm thinking most of our listeners are probably practicing in a location where they don't have a team that really focuses on that. So, let's say we do the imaging of the spine and you don't find a clear cause. Is the next step to just do a blood patch? Do you send them to someone like you? What's the practical next step?

Dr Rau: Yeah, if your- regardless of whether you find a leak or not, if your clinical acumen is such that you think this patient has a leak or I've treated them for everything else and it's not working and I have at least a high enough suspicion that I think the risk of getting a patch is lower than the benefit that if they got a patch and it worked, I do send my patients for non-directed blood patches, because it currently does take a long time to get them to a center that can do CT myelograms or any kind of advanced imaging to look for sort of a CSF venous fistula or to get treated outside of a nondirected patch. You know, sometimes nondirected patches are beneficial for patients, and there's some good papers out there that sort of explain the low risks of doing these if done properly versus the extreme benefit for patients when it works. And, I mean, I can't tell you how many people come in and tell me how their lives are changed because they finally got a blood patch. And sometimes it works. And it’s life-changing for those people. You know, they go back to work. They can interact with their kids again. Before, they didn't know what was wrong, just had this headache that started. So it's worth doing if you have a strong clinical suspicion.

Dr Smith: Yeah. I mean, that was great. And, you know, to go back to where we began, this is severe. It's something like 60% of patients with this problem have thought about suicide, right? And you take this patient and cure the problem. I feel really empowered having read the article and talked to you today. And so, I'm ready to go out and look for this. Thank you so much for a really engaging conversation. This has been terrific.

Dr Rau: Thank you. I appreciate it. I enjoyed being here.

Dr Smith: Again, today I've been interviewing Dr Jill Rau about her article on clinical features and diagnosis of spontaneous intracranial hypotension---which I guess I should say hypovolemia after having talked to you---which she wrote with Dr Jeremy Cutsworth-Gregory. This article appears in the most recent issue of Continuum on disorders of CSF dynamics. Please be sure to check out Continuum Audio episodes from this really interesting issue and other interesting issues. And thank you, our listeners, again for listening to us today.

Dr Monteith: This is Dr Teshamae Monteith, Associate Editor of Continuum Audio. If you've enjoyed this episode, you'll love the journal, which is full of in-depth and clinically relevant information important for neurology practitioners. Use the link in the episode notes to learn more and subscribe. AAN members, you can get CME for listening to this interview by completing the evaluation at continpub.com/audioCME. Thank you for listening to Continuum Audio.