Epigraph Vol. 25 Issue 4, Fall 2023

Cerebral amyloid angiopathy, seizures, and epilepsy: Dr. Brin Freund

Reported by Dr. Alina Ivaniuk |  Edited and produced by Nancy Volkers

Cite this article: Ivaniuk A. Cerebral amyloid angiopathy, seizures, and epilepsy: Dr. Brin Freund. Epigraph 2023; 25(4): 1-7.

Listen below or download the episode.

Find Sharp Waves episodes on SpotifyApple PodcastsGoogle PodcastsAmazon MusiciHeart Radio, and Stitcher.

This episode discusses the relationship between cerebral amyloid angiopathy and epilepsy through the lens of a recent publication. Dr. Alina Ivaniuk talks with Dr. Brin Freund about this topic and the importance of future research.


Sharp Waves episodes are meant for informational purposes only, and not as clinical or medical advice.

Podcast Transcript

[00:00:00] Dr. Alina Ivaniuk: Hello, everyone. This is Alina from YES-ILAE bringing you another episode of the Sharp Waves podcast. The matter of today's discussion has gained attention only recently: Associations between cerebral amyloid angiopathy (CAA) and seizures with epilepsy. There is not much data on it, but we are gradually getting more and more insights on this topic.

We'll frame today's talk around a paper recently published in the European Journal of Neurology titled, "The incidence and risk factors associated with seizures in cerebral amyloid angiopathy." With me today is the first author of the article, Dr. Brin Freund. 

Dr. Freund, welcome. We are very excited to have you with us today. Could you just briefly introduce yourself to our listeners? 

[00:00:47] Dr. Brin Freund: Thanks so much, Dr. Ivaniuk. I appreciate the invitation. I'm currently about to start my job here in Jacksonville, Florida with Mayo Clinic as an epileptologist and assistant professor in the Department of Neurology.

But in terms of my background, I completed my residency at Johns Hopkins in Baltimore and then subsequently served an active duty as a neurologist for four years in the US Navy, followed by a fellowship in clinical neurophysiology with a focus in EEG here in Jacksonville at Mayo Clinic.

And then subsequently completed my epilepsy fellowship at Cleveland Clinic in Cleveland, Ohio. My research interests and interests clinically relate to EEG and epilepsy and particularly looking at patients with acute symptomatic seizures, as well as understanding better ICU EEG patterns and their meaning in, in prognosis and diagnosis of patients who are critically ill.

I'm also quite interested in patients with dementing processes and the use of EEG as well as the importance of seizures in this population. 

[00:01:43] Dr. Alina Ivaniuk: Amazing. It seems that you're just the right person to talk to about this topic. So, why don't we start? And my first question is, overall, why is it important? CAA is the vascular problem for vessel disease. Seizures are the problem of cortical excitability. Where do those two come together and how can CAA dispose people to have seizures? 

[00:02:11] Dr. Brin Freund: So, I think the importance lies in the recent studies and recent focus on epilepsy in dementing processes and in neurodegenerative diseases, of which obviously CAA is a subtype of that.

A good example would be in the ARIC study where they looked at patients who had a threefold increase in risk for dementia who were newly diagnosed with epilepsy. And also looking at patients who were newly diagnosed with epilepsy, they had a threefold increased risk of dementia. So there's clearly a link, and other studies are pointing to a link between seizures and deterioration of the neurodegenerative process as well.

Specifically regarding CAA and seizures. Firstly, CAA is relatively common, actually. Sixty percent of patients with dementia on autopsy studies have CAA pathology. And this is up to 80% in patients who have Alzheimer disease. So there's clearly a related relationship between dementing processes, in particular Alzheimer type dementia, and CAA.

And so, understanding the underpinnings of CAA and the comorbidities that are related to CAA are very important. And specifically looking at seizures, there's not a whole lot in the literature on this, and as you said, this is becoming more of a topic of interest. Again, kind of feeding back into dementing processes and their association with seizures, but specifically related to CAA, there are likely two mechanisms associated with CAA that lead to seizures. One of which is the accumulation of beta amyloid peptide, which in animal models has been shown to lead to recurrent seizure activity in Alzheimer type dementia. 

And then obviously with CAA, there's an increased risk of intracranial hemorrhage. Many of those cases have cortical involvement, as well as cortically based inflammatory process known as CAA-related inflammation, which also would predispose to epilepsy as well as acute symptomatic seizures.

So I think this topic probably needs to be better addressed, given the importance that seizures may have in dementing conditions. 

[00:03:55] Dr. Alina Ivaniuk: Yeah, that sounds very reasonable. And you mentioned that the information right now is scarce. There is not much evidence, not that much epidemiological data. There are some studies. What did they show and which gap did you close with your study? 

[00:04:10] Dr. Brin Freund: So I think the majority of the focus with CAA has been in diagnosis, you know, diagnostic evaluations and the diagnostic criteria for CAA, which is obviously very important. Now that we have that sort of settled, I think the focus is now more on comorbidities related to CAA, and obviously looking at secondary intracranial hemorrhage is an important topic.

But looking at seizures, there was a study that came out last year in Epilepsia looking at a cohort of patients with seizures, and they found similar findings to us, though they looked at a smaller cohort and they were looking more at long-term epilepsy as opposed to acute symptomatic seizures.

In our cohort, we focused more both on the acute symptomatic seizures as well as the long-term risks of epilepsy. So this is something that is, again, as you said, is being talked about a little more, but there's very little looking at a heterogeneous cohort other than that study that I just mentioned.

And there have been other studies looking at more homogeneous cohorts and smaller studies looking at patients specifically with intracranial hemorrhage or subarachnoid hemorrhage, both with and without amyloid angiopathy. And there's clearly a link with seizures, but no one has looked at a heterogeneous population of patients with CAA to define, within that cohort, what are the risk factors for seizures. 

Another interesting topic which we'll probably, I assume, touch on at some point, are amyloid spells or transient focal neurological episodes which also, also are gaining more interest in research and have been studied for quite some time, though are more looked at from the perspective of the vascular neurologist as opposed to the epileptologist.

More recently, there was a study last year in JAMA Neurology that pointed out the importance of prognosis regarding amyloid spells and how they can be a precursor to intracranial hemorrhage and even mortality. So understanding their underpinnings is quite important and I think we'll touch upon our findings and how we are suspicious based on what we found that these spells could actually be more related to an epileptogenic potential as opposed to just a cortical sprain depression, as noted in previous studies, that would more likely be related to a migrainous, aura-type phenomenon that we are all aware of. And so, you know, again, I think from our perspective, looking at more from an epilepsy perspective is something that is more a more recent focus and I think is important.

And I'm hoping that more studies are going to be coming in the future related to that. 

[00:06:14] Dr. Alina Ivaniuk: Yeah. Thank you so much for placing your study in the context of other research that is going on. And I think your study does add to the bulk of existing knowledge. Could you summarize for us the main findings?

What did you find and what are the overall takeaway points from your study? 

[00:06:33] Dr. Brin Freund: So what we found was that in our cohort of 284 patients, all of whom were diagnosed with amyloid angiopathy, either based predominantly on neuroimaging characteristics or with the addition of pathology in a few of the cases, that roughly 20%, or 56 of the patients, were diagnosed with epilepsy at some point during the course of their evaluation, and nearly one-third of those patients who had seizures actually presented with seizures as the index event that led to the diagnosis of amyloid angiopathy.

When we performed a univariate analysis looking at risk factors for seizures in this cohort, we found that lobar intracranial hemorrhage, cortical subarachnoid hemorrhage, superficial sclerosis, and CAA with related inflammation all posed a higher risk of seizures. When we then performed the multivariate study, we found that superficial sclerosis, lobar intracranial hemorrhage, and CAA with related inflammation remained significant.

Further, we found that nearly half of our patients who were clinically diagnosed with amyloid spells were diagnosed later with epilepsy based on both clinical and neurophysiological data. One other thing that's important to note was that the patients that did have seizures did quite well clinically and were reasonably well controlled on either one or two antiseizure medications with a very low incidence of drug resistance.

[00:07:42] Dr. Alina Ivaniuk: Okay, thank you for the summary. Now, let's dive deeper into it. So, in your study, two independent risk factors for subsequent seizures were related to the release of blood product, they either superficial sclerosis or intracerebral hemorrhage. Yet the micro bleeds happened more frequently in a subgroup of individuals who did not have seizures. Why do micro bleeds not appear pro epileptogenic? Is there something to do with the location of the bleed and is ultimately cortical involvement in the bleeding the factor that is detrimental?

[00:08:21] Dr. Brin Freund: I think you're 100% correct. And you've kind of hit on the head there. And it's very interesting. And I think logically it makes a lot of sense. And again, let's just take this into perspective. We're looking at specifically patients with amyloid angiopathy. We didn't compare them to patients without amyloid angiopathy.

So we want to keep that in mind, but when we look at the instance of micro hemorrhages, these are the predominant neuroimaging features or finding that leads to the diagnosis of amyloid angiopathy. When we look at neuroimaging criteria, 91% of our patients had micro hemorrhages on imaging. So they're often found in isolation, obviously, you know, without symptoms.

So they're a common finding in amyloid angiopathy. So we have to keep that in mind. And they're usually subcortical, and therefore as I said, they're usually often subclinical, and therefore, the likelihood of them causing epilepsy is lower, given that they're not irritating the cortex.

On the other hand, intracranial hemorrhage in CAA is often large, and often devastating, and often seen concomitantly with superficial sclerosis and cortical subarachnoid hemorrhage, which we also saw in our cohort. And so these findings, or these pathologies, can lead to cortical irritation and subsequent seizures.

Another pathology that I think we need to keep in mind related to these patients is also the inflammatory process related to amyloid deposition in the cortex. So CAA-related inflammation can actually also pose a higher risk of seizures. And so I think the involvement of the cortex, as you laid out, is the important defining feature that differentiates patients who are at a higher risk of seizures as opposed to those who are not at a high risk of seizures in cases of amyloid angiopathy.

[00:09:49] Dr. Alina Ivaniuk: Thank you for elaborating on this. This makes lots of sense. Let's go further. A quarter of people in your cohort were diagnosed with amyloid spells, right? Transient focal neurological episodes. Half of them later on were diagnosed with seizures or epilepsy upon further evaluation. And it raises a couple of questions.

The first one would be, based on your research and your experience, and other available evidence, how would you relate those amyloid spells and seizures? And second, how would you summarize the practical implications of this interlink, if any? 

[00:10:26] Dr. Brin Freund: This is an extremely interesting and important question that you pose, and something that unfortunately in a manuscript we don't have enough space to get into details. And I think that needs further study, but regarding the findings that we made as well as the data and literature, as well as in my own experience, clinically, I've always been somewhat concerned, disconcerted by the notion of spells. It seems like it's sort of this kind of hand-waving phenomenon where we kind of call them spells and leave it at that, and I know reviewing the cases that we had in our cohort I found that to be the case as well with our clinicians and I think that's something that we see a lot in practice.

There have been studies, though, that have been trying to understand the underpinnings of spells. And particularly, you know, some have suggested that this is related to a cortical sprain depression phenomenon, as I mentioned before, likely, you know similar to migranous auras. But I would argue against being definitive about that, given that a lot of the studies, as we point out in our manuscript, did not really go deep into the possibility of these being seizures. When they studied spells, there were very few that had EEG study, very few that had epilepsy evaluations that were mentioned, and our cohort was the same. There were about 35% of patients who had spells that never had any EEG study. So, you know, I think this is a little odd, given that for patients who present with spells without CAA we often in the differential would consider seizures and would evaluate that as such.

So, I think that this is a problem with defining what spells are. And so, regarding what we found, when you look at previous studies, there have been suggestions that that there's an anatomical correlation with spells. So, for instance, there was a study that showed that on neuroimaging when a patient presented with motor or sensory phenomena, you could actually look at the MRI and you could localize it to the contralateral hemisphere in the sensorimotor region.

So, so clearly there's a link there, right? And we know that spells have been related to superficial sclerosis and subarachnoid hemorrhage in previous evaluations and analyses. And so, you know, as we point out in our study, obviously superficial sclerosis and cortical subarachnoid hemorrhage are also risk factors for seizures, not just spells, as pointed out in previous evaluations.

So I think this is probably the pathophysiological link between spells and seizures. But again we need further study to kind of better understand this because you know, our assessment of patients with spells leaves a lot to be desired. 

[00:12:46] Dr. Alina Ivaniuk: That is very interesting and I hope that further research can clarify what those spells are and how they relate to seizures. But I think that we are getting more and more information on that. And that's something that is important to keep in mind when you work up people with amyloid spells. Knowing what they are from the pathological perspective and whether or not it implies any interventions, whether diagnostic or therapeutic afterwards. 

[00:13:13] Dr. Brin Freund: I think when you discuss the practical implications of that, I think the implications, as I said, would be that when clinicians are encountering patients with amyloid spells, they should probably take a step back and analyze the case similarly to how they would with any patient presenting with a transient neurological event. And this would include evaluating for the possibility of seizures. And even considering, if the events are frequent, maybe even trialing an anti-seizure medication in the appropriate setting would be something to think about.

[00:13:43] Dr. Alina Ivaniuk: Thank you so much for giving this perspective on the practical implications because this is what everybody wonders about. Those are epidemiological data, but what do we do with that and how do we frame it in our clinical practice? 

Let's talk about another practical point. For more than one-third of your cohort, the seizures were the heralding sign, the first sign that led to the diagnosis of CAA. And do you think it has any anything to do with the approach to workup of individuals with late-onset epilepsy? If individuals present with late-onset seizures, would you consider working them up for CAA? 

[00:14:25] Dr. Brin Freund: Well, I think any patient who has late-onset epilepsy is likely going to undergo an MRI study, so I think you're probably going to get your answer with that. And my concern or my thoughts on the diagnosis of amyloid angiopathy as a cause of seizures is really not just with being definitive about the cause of the seizures, but then being able to counsel the patient as well as being able to involve other specialists that are important in preventing secondary injury in patients with amyloid angiopathy, particularly related to secondary intracranial hemorrhage from the use of antithrombotics, which often is an issue with these patients given that there's comorbid vascular disease. So getting our stroke specialists or vascular specialists involved and then obviously screening for cognitive impairment and the possibility of an underlying neurodegenerative process that's comorbid.

And then again, making sure that we are getting our cognitive specialists involved, if necessary, to then rule out the possibility of comorbid dementia. So I think really the diagnosis is important, obviously, in any case of seizures in a later age or elderly age. But I think in particular with amyloid angiopathy, we just need to make sure that we know that this is a distinct possibility as a cause and then being able to relay that and communicate that clearly to everyone involved,

[00:15:42] Dr. Alina Ivaniuk: Fantastic. There's another practical point that I would like to raise. In your study, there was no increased mortality in the seizure cohort. The seizures were not treatment resistant. So most of them were reasonably controlled and or on one or two antiseizure medications, as you mentioned, yet you do report status epilepticus in 5 individuals out of 56, that's roughly 9% of your cohort.

How does this rate relate to other causes of late-onset epilepsy? And how would you wrap this together and present it to your patients when discussing their prognosis? 

[00:16:21] Dr. Brin Freund: So I think relating to the incidence of status epilepticus when you look at other studies specifically looking at or analyzing the pathologies that we looked at, particularly with intracranial hemorrhage or subarachnoid hemorrhage in the acute setting, the numbers can be as high as 8% regarding those presenting with those clinical histories and then subsequently being diagnosed with status epilepticus.

So if you take that into account, our numbers are relatively on par looking at the whole cohort, being roughly 2% having status epilepticus. Again, we have a very heterogeneous population. We have patients with intracranial hemorrhage, patients that had ischemic stroke, both acute and chronic. We have patients that just had micro hemorrhages. We have patients that had CAA- related inflammation. There was a study, as I've mentioned earlier, last year in Epilepsia looking at CAA and seizures. They had similar findings, including the low rate of status epilepticus as well as the low instance of drug resistance.

I think kind of going back to your point about prognosis regarding these patients with seizures who we end up diagnosing and decide to start on anti-seizure medication. When I counsel a patient, I would tell them that the good news is that most patients with amyloid angiopathy and seizures are responsive to one or two medications and often are seizure free, which is not something that we can say for most patients with epilepsy. If you look at the data, around a third of patients that we know of don't respond to anti-seizure medications and require evaluation for possible surgery, given that they are drug resistant. Looking at our cohort, obviously the number is much lower. So I think from that perspective, that's a good sign. And something we can tell our patients that's reassuring. 

The one thing I would bring up would be, and patients may not ask this question, but something that would maybe be in the back of their mind as to, you begin the diagnosis of seizures and you get in the diagnosis of amyloid angiopathy and they're going to see that that's related to Alzheimer's disease. So what does that mean for the patient? Do they have Alzheimer's disease? And are these seizures going to affect their cognitive function? And I think we still need to sort that out. 

So our study was retrospective and we had a follow-up period with a median of about 35 months. So I think we'd have to be very, we'd have to temper our counseling regarding that and tell the patient that, you know, we still need to sort that out with further study in the future.

But overall I would have a pretty hopeful outlook regarding their seizures. 

[00:18:34] Dr. Alina Ivaniuk: That is very reassuring. And although longer follow-up studies are needed to highlight or to clarify that link to Alzheimer's disease, at least the seizures appear to be well controlled.

With that, which questions are still unaddressed in the field of CAA and seizures and epilepsy, and do you have any plans on investigating this area further? 

[00:18:57] Dr. Brin Freund: So I have two big burning questions which we kind of touched on. One of which would be the association between seizures and long-term neuropsychological functioning in these patients.

This needs prospective evaluation and a longer term follow-up to better understand how seizures are affecting patients with amyloid angiopathy and whether or not it can have an impact on their outcome and prognosis from a cognitive perspective. And I think the second question relates to amyloid spells and what's the underpinning of them.

And I think this needs to be at the very least addressed in a more rigorous retrospective study, including a cohort of a similar size to ours. And I think to better analyze the implication of spells and their possible link to seizures, this needs to be done. And really with a longer term follow up, to better delineate the incidence associated with seizures and spells.

I think the risks of misdiagnosing spells as TIA or other causes can have grave consequences. And as I mentioned, in the prior study in JAMA Neurology last year, there's a significant association between spells and the risk of intracranial hemorrhage and mortality. So this, the diagnosis of spells and the management of spells is something that we need to better understand with further study.

And this is a topic that I actually plan on looking at in the future. 

[00:20:08] Dr. Alina Ivaniuk: That is fantastic, Dr. Freund. I'm very much looking forward to talking to you a couple of months or probably years later about this and what you have found.

And thank you so much for making time to talk to us today. We greatly appreciate your insight. Yeah. And we are looking forward to discussing more results from you in the future. 

[00:20:28] Dr. Brin Freund: Thank you so much for inviting me. It was a lot of fun. And I look forward to hearing this on the podcast.