Epigraph Vol. 25 Issue 4, Fall 2023

Autism and epilepsy: Dr. Colin Reilly and Dr. Stéphane Auvin

Reported, edited and produced by Joy Mazur, Epigraph intern

Cite this article: Mazur J. Autism and epilepsy: Dr. Colin Reilly and Dr. Stéphane Auvin. Epigraph 2023; 25(4): 56-64.

Listen below or download the episode.

Find Sharp Waves episodes on Spotify, Apple Podcasts, Google Podcasts, Amazon Music, iHeart Radio, and Stitcher.

Compared with the general population, epilepsy is more common in people with autism—and autism is more common in people with epilepsy. How can autism affect the diagnosis and treatment of epilepsy, and vice versa? Joy Mazur spoke to Dr. Colin Reilly and Dr. Stéphane Auvin.


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

Podcast Transcript

Joy Mazur: Nearly one third of people with autism have epilepsy. For these people, diagnosis and treatment pose several challenges. For example, being autistic can make it difficult to explain experiences and sensations. This means it may be hard to distinguish between behavioral symptoms of autism and seizures. People with autism may also have more sensitive reactions to typical anti-seizure medications or may find it difficult to be in unfamiliar environments, like an epilepsy clinic.

Experts suggest that proper management of epilepsy, autism, and other comorbidities can greatly improve the quality of life of the person with epilepsy and autism and their family.

To learn more about diagnosis and screening, Sharp Waves spoke to Dr. Colin Reilly.

Photo of Dr. Colin Reilly
Dr. Colin Reilly (Sweden)

Colin Reilly: My name is Colin Reilly. I'm a psychologist. I work mostly in Sweden. I work assessing young people with neurological conditions, many of whom have epilepsy. I also work in the UK. Most of my research is about cognitive and behavioral conditions in children with epilepsy. So, I undertake neuropsychological assessments with children with epilepsy. And as I said, my research also focuses in that area. Trying to understand better why children with epilepsy have such difficulties and how one can reduce the impact of them on their everyday lives.

Joy Mazur: You work a lot with the co-occurrence of epilepsy and autism. But what do most epileptologists or general neurologists know about that co-occurrence?

Colin Reilly: I think there's an increasing awareness, but autism is under-identified in children with epilepsy. A number of studies have shown this. Many of the co-occurring neurobehavioral conditions or neurodevelopmental conditions are under-identified in children with epilepsy.

So, it's always important to raise awareness. If we look at autism in children with epilepsy, most studies suggest that between 10% and 20% of children with epilepsy also have autism. We undertook a study in 2014 which showed that although 21% of the children in the study fulfilled the criteria for autism, only 8% had previously been identified.

Prevalence rates will vary depending on the definition of epilepsy, but it's much higher than in the general pediatric population. So, it's much higher in children with epilepsy. And therefore, it's a significant issue. And it's an issue that epileptologists and child neurologists who work with children with epilepsy need to be aware of.

Joy Mazur: So how do you recognize when someone with epilepsy maybe has autism, especially if they don't present a lot of those behavioral symptoms that we usually think of?

Colin Reilly: Well, I think it's important to recognize that autism is quite a heterogeneous condition in terms of presentation. So, it varies significantly in terms of the intellectual function, but also the manifestation of the core symptoms, which are social communication difficulties and repetitive interests or behaviors.

Many children with autism also have sensory sensitivities. In terms of young children, it's often quite clear that they have, you know, autistic behavior from quite an early age, they may be late in talking, they may have social skills difficulties, which are evident in preschool.

For older children with autism, particularly though, for older children with epilepsy, who have, for example, normal cognitive functioning, it may be more difficult to recognize those signs. And they may only become apparent in, for example, early adolescence, when they begin to have more social skills difficulties as the social skills requirements of society increase.

So, for example, understanding teenage talk, understanding more subtle social aspects. So, a lot depends on the cognitive ability of the child. And oftentimes younger children will be identified as having developmental problems, but not necessarily autism. And what can happen with epilepsy is the presence of, particularly of frequent seizures, can overshadow the need to look at the child's behavior and look at the child's neurodevelopment.

So, there is a risk that autism identification is delayed in children with epilepsy, and there is research to show this, that children with epilepsy and autism are often diagnosed later than children with autism who don't have epilepsy. So, it's very important not just to focus on the seizures and hope, for example, that the seizures will go away. For many kids, they won't go away, and even if they do, the child may still have behavioral difficulties, including autistic behavior.

Joy Mazur: So who should be doing, I guess, a screening and diagnosis? Is this something that neurologists should be implementing into their practice? Or should they just be on the lookout and be referring them to another specialist?

Colin Reilly: It depends. I think the best way to approach this is true surveillance so that epileptologists can recognize the core features of autism and ask questions of caregivers about those core features, those social skill difficulties, communication difficulties, difficulties with change, so that need for routine, repetitive behaviors, asking those questions and knowing about them.

When formal screening is needed, and it may be needed at different times, depending on the particular age of the child and intellectual capacity of the child, there are a number of screening instruments available. And as far as we know, they work quite well in the autism or in the epilepsy population.

But there isn't so much specific evidence of that, but as far as we know, many of the instruments which are used in the general pediatric population to screen for autism can easily be applied in the epilepsy population, but we do lack a bit of evidence there. One of the things to point out is that there is some evidence that epilepsy might present differently in some children, or some might present differently in some children with epilepsy, particularly children with developmental and epileptic encephalopathies.

So those early onset epilepsies where neurodevelopment might stagnate, in those cases, there might be a period of what appears to be or is normal development, followed by the development of neurodevelopmental symptoms, including autism.

And that means that we might have to, for example, think of screening again, even if a child with epilepsy has passed a screening test quite early on. Fluctuating symptoms in epilepsy in children with autism are not uncommon, and therefore it's very important that we think about autism, even if we feel that children may not have had those symptoms when they were first diagnosed with epilepsy.

Joy Mazur: Are there any other common issues that arise in diagnosis? And how do you usually address these?

Colin Reilly: One of the commonest issues is that there's not enough resources in the general sort of pediatric population when looking at autism, but also in the epilepsy population.

One thing that parents can be concerned about is long waiting times for an autism assessment. So they may come with concerns or a professional might identify concerns, but it may be that they need to be referred to another clinic where our autism professionals may work or a multidisciplinary team, which focuses on the assessment and diagnosis of autism.

I think what parents say to us in our research is they would like that the focus on epilepsy includes a focus on those co-occurring conditions. So co-location of neurodevelopmental services or neurodevelopment assessments in an epilepsy clinic is increasingly being recognized as the most effective way, so that those who work with a child's epilepsy also can work with a child's co-occurring conditions. Because for many, they are a part of their epilepsy. They're a core feature of the child's epilepsy. So that's increasingly being recognized. So it's really important that it does not take too long for the child to get assessed for autism because delay can mean a lack of understanding, but also a lack of supports.

Often an autism diagnosis brings not only understanding, but also means that the child can get extra support in school, for example, so it's very important that that diagnosis is not delayed. And what we advocate is screening and assessment, if needed, but also that that delay is not there and that co-location of neurodevelopmental and epilepsy services is possible and is the best way to go.

Joy Mazur: Do you ever experience any issues with the fact that usually you're asking some sort of proxy to self-report behavioral symptoms and you're kind of navigating, either with a caregiver or parent and not the person themselves?

Colin Reilly: I think a key aspect of assessment in relation to autism in all children is observation.

So it's very important to talk to the parents, but, you know, observing in the clinic or the hospital, and if possible, in the school setting as well. So getting information, not just from the parents, from the school, but actually interacting with the child. So playing with the child, talking with the child if they're verbal. So that is a key aspect in terms of understanding autism. And if you're thinking about autism assessment, it is getting symptoms from parents via screening tests, interviews, but also observations. So observations in a structured way, perhaps in a clinic setting, and there are instruments to do that, particularly the ADOS [Autism Diagnostic Observation Schedule], but also in the school or preschool setting.

So it's very important that you have both information from parents, but also clinical judgment based on observations with the child. And some children will be able to talk about their social difficulties as well. And that's important to include them if they can contribute.

Joy Mazur: Usually we think of autism as male over female and there's evidence that girls are maybe underdiagnosed with autism. Does that play any sort of role in the world of epilepsy as well?

Colin Reilly: I think it's the same in epilepsy, but what we tend to find in epilepsy in research studies where children have been properly screened, we find the gender or sex ratio is equal. There isn't the same difference that you find in the general pediatric population.

So it is close to 1:1, so the same amount of females and males. However, it's always important to remember that females are more likely to mask their autism symptoms, and it's very important to recognize that and recognize that masking can have a very significant impact on their mental health.

So, it is important. And generally speaking, the screening instruments are quite good at picking up on autistic symptoms in females. So doing formal screening is one way of reducing the risk of missing females, but it's important to say that the ratio is different. Perhaps the autism in epilepsy is slightly different than autism in the general population with respect to the gender ratio and possibly in respect to mechanisms and background factors.

Joy Mazur: Are there any other notable demographic differences either with the general population or just anything particularly notable within the epilepsy population?

Colin Reilly: I think the autism population is heterogeneous and the epilepsy population is heterogeneous.

As I said earlier, there may be that risk that the autism symptoms do not appear perhaps as early or in the same way as in the general autism population. In the epilepsy population, I mean, there might be slightly later manifestation or there might be slightly different manifestation.

There is some suggestion, for example, that the social skills difficulties that are prevalent in autism without epilepsy, may be slightly more subtle in children with epilepsy. But again, it is so varied and there are so many epilepsy syndromes, and there may be differences between epilepsy syndromes as well.

So it's important to think about that and understand the epilepsy and understand the child's autistic symptoms in the context of their epilepsy. So for example, if a child is having an awful lot of epileptic seizures, so seizure not well controlled or taking a number of anti-seizure medications, it's important to consider that.

But at the same time, not wait too long, waiting for the epilepsy to be well controlled because that might not actually happen. So it's very important to sort of, you know, wait and try and optimize treatment for epilepsy, but not wait too long.

Joy Mazur: And as far as research, what is the state of research or future research with epilepsy and autism currently?

Colin Reilly: If we take screening, I guess the key thing is basically to evaluate whether the instruments we have in the general epilepsy or general autism population work as well with children and even adults with epilepsy. So it's important to establish that.

And we're doing some work within the ILAE looking at these instruments and exploring their psychometric properties. However, very few studies have looked specifically at the epilepsy population. And there are no specific autism instruments for the epilepsy population at this moment in time.

In terms of the wider sort of autism and epilepsy picture, I think developments, particularly in children with epilepsy around treatments and precision medicine, mean that we might be able to better understand why autism develops in children with epilepsy and why it might develop in particular epilepsy syndromes. And in that way, in terms of treatments that are coming down the line, can we actually reduce the impact of epilepsy on newer development or basically reduce the impact of the underlying disease on both seizures and co-occurring conditions.

So trying to better understand, you know, what can we do for these children with more targeted treatments and hoping that those targeted treatments will not just reduce seizures, but will also impact co-occurring conditions. And I think we're going to see in the next few years with the development of these target treatments and precision medicine a good impact, not only on seizures, but also hopefully on the neurodevelopmental aspects of pediatric epilepsy.

Joy Mazur: And what's next for you and your work?

Colin Reilly: At the moment, we're looking at particular epilepsy syndromes. You know, we're looking at Dravet syndrome and we're including focusing on autism there. There's a very high prevalence of autism or autistic symptoms in Dravet syndrome.

We're also doing a study on tuberous sclerosis. So, we're trying to understand the impact of autism on quality of life, both the child and the family in those particular epilepsy or neurogenetic syndromes. And not just autism, but many of those other co-occurring conditions. I think it's probably fair to say that if you have autism and if you have epilepsy, you probably have other conditions or symptoms of other conditions as well.

I'm thinking of sleep problems. I'm thinking of ADHD, motor problems. It's rare to find a child who just has autism. And it's probably rare to find a child who just has autism and epilepsy and doesn't have learning problems or executive functioning problems or attention problems.

So really, although we talk about autism, it's often much broader in terms of the impact on the child's everyday life and autism may be useful in terms of understanding the child's particular difficulties or diagnosis of autism, but it's very important to also look at the wider picture and see what other difficulties that child might have. Because as I say, these things often go together and having just autism is probably rare.

And we have these categories and these diagnoses, which can be very helpful, but oftentimes, as I say, children may have difficulties in an area without reaching the full criteria. So it's important that we map out the child's difficulties across the neurodevelopmental spectrum so that we can support them, but also think about families and how these co-occurring conditions can impact on the families as well.

And I think we know, for example, that parents of children with epilepsy are at higher risk for mental health problems, particularly moms. But we also know that by better supporting the children in terms of both their epilepsy and their neurodevelopmental problems, we can probably reduce that burden and reduce those mental health difficulties.

So when looking at children with epilepsy and supporting them, it's very important to think about the family and how epilepsy and how those co-occurring conditions impact on the child and the family. And by supporting the family and not just the child, I think we can maximize quality of life for both children and caregivers.

Joy Mazur: To learn more about supporting people with autism and epilepsy and their families through treatment, Sharp Waves spoke to Dr. Stéphane Auvin.

Stéphane Auvin (France)
Dr. Stéphane Auvin (France)

Stéphane Auvin: I'm Stéphane Auvin. I'm a child neurologist, child epileptologist in Paris. I'm working in Robert-Debré Children's Hospital and Paris City University. I'm involved in the care of epilepsy patients.

Of course, we have also on our site a center for rare epilepsy, but something that is special, we have a department, a very large child psychiatry department that is involved a lot in research of autism. And then we have a lot of interaction, and they are referring a lot of patients.

Fifteen years ago, I started to have a special interest in these patients that have both conditions, and sometime it is a patient with autism spectrum disorders that is starting epilepsy. I mean, some of our patients, in particular when they are drug resistant or in the case of developmental epileptic encephalopathies, they may develop autism after the epilepsy.

Joy Mazur: And can you tell me a bit about the epidemiology of epilepsy as it relates to autism and the relationship between the two?

Stéphane Auvin: We used to say they are a dual relationship because of the different scenarios I mentioned. Children with autism spectrum disorder are more likely, unfortunately, to have epilepsy compared to the general population. With two peaks of incidence, early childhood or in adolescence.

And the other way around, we know that compared to the general population, patients with epilepsy, in particular with early onset, with developmental epileptic encephalopathy or with drug resistance, they are more prone to develop autism spectrum disorders and the incidence is definitely higher in both scenarios if you start with epilepsy or if you start with autism spectrum disorder.

Joy Mazur: Can you tell me how you go about treating and working with people who have epilepsy and autism?

Stéphane Auvin: If we improve the seizure condition or the epileptic burden, we might improve also behavior and cognition. And this is when this part of the epileptic encephalopathy is very predominant. It's when we decide to treat kind of very intensely with anti-seizure medication because our goal is to rescue part of the cognition or the behavior.

While there is a path to what we call developmental epileptic encephalopathy, meaning that the development or the underlying cause of the disease is also involved, for example, let's say a brain malformation or a gene mutation that is not only resulting in seizure, but also resulting in change in behavior.

In that case, we are very cautious because if you push too hard with your anti-seizure medication, you get more side effects because the part that is developmental is not accessible for changes based on the anti-seizure medication.

And it's something I think it's important to know that, for example, it has been mentioned, maybe, you know, we have some patients with autism spectrum disorder, they have an EEG abnormality. And it's clear now that treating this EEG abnormality with anti-seizure medications, unfortunately, won't solve the autism spectrum disorder features.

Joy Mazur: My understanding is that it can also exacerbate some behavioral symptoms of autism. So can you tell me how you approach that?

Stéphane Auvin: I will take the example of the patient with autism spectrum disorder that developed epilepsy. Then we can go further on how the anti-seizure medication can act on behavior in the epilepsy patient as well.

There is a study that was published a few years ago, the methods are not very robust, but I like the methods because they ask the parents. Of course, there is no verification of the diagnosis. There is no flowchart. There is no correlation with anything. Then it's just based on the claim of the parents. In that study, they look at different strategies— anti-seizure medications, but also the ketogenic diet, and they asked the parents on a scale, like, is it better, a bit better, much better, very much better, or same thing, a bit worse, much worse, or very much worse.

And then when they look for seizures, you know, when you're treated with anti-seizure medication, not a big surprise, then there is an improvement. But when they look at all their skills that are crucial in patients with autism spectrum disorder, like communication, interaction, language, sleep, it always scores in the worst parts. Always. It seems that they are more sensitive to side effects, and in particular the side effects on cognition and behavior.

Anti-seizure medications are targeted in the brain, then it's not a big surprise that they can affect cognition and behavior, but because of the condition, the autism spectrum disorder, maybe they are more sensitive towards that. In the case of patients with epilepsy that we are treating with anti-seizure medication and they have a diagnosis as well of autism spectrum disorder, we know that some anti-seizure medications are more prone than others to induce behavioral trouble. And in that case, we are very cautious.

There are general rules at the individual level, we should be always very cautious. When we prescribe a treatment, the aim is not only reduction of seizures or seizure freedom. It's how the seizures are changed. Does it improve the quality of life? And in addition, for what consequences? Is the patient sleepy? How the behavior is? You need that feedback and to be on alert, making sure that you are going back and decreasing or withdrawing anti-seizure medication that is worsening the full condition of the patient.

Joy Mazur: There's also, as I understand it, a lot of comorbidities with things like ADHD, anxiety, sleep disorders. How do you manage that? What is your approach there?

Stéphane Auvin: What is very important is to try to have like a full view on the comorbidities and try to address them. Of course, the idea is not to put like a drug for any of these symptoms. Otherwise, you end up with 10 drugs. That is not the idea. But you have to look globally to the patient, look what is more predominant, what are the consequences. And then sometime treating the patient, improve one condition, but can work on another one. And at the end, it's always a benefit-risk ratio at the individual level that should guide you how to manage the treatment.

But let's say in some patients, the predominant feature is ADHD, and we know that we can use specific treatment like methylphenidate, for example. One question that is coming frequently is because in the summary of the product of many of the neuroleptics, it's written, be careful with seizures worsening.

That doesn't mean that you should not use neuroleptics, like for example risperidone or Aripiprazole. It's clearly written, but it's mostly based on screenings that were done in animal models. I'm not saying that there is never any worsening like, for example, with methylphenidate. But we know that for most of these patients, the worsening is not frequent.

Then even with an active epilepsy, you can use methylphenidate. You can use risperidone, or Aripiprazole, same thing. But my advice is if you use an anti-seizure medication, and then some behavior changes appear, you should first withdraw the new anti-seizure medications that might provoke these new features instead of adding, for example, a neuroleptic. Because in that case, you'll go for polytherapy that is most likely due to your first treatment, then you have to avoid that.

Joy Mazur: And so if someone also has maybe multiple doctors, someone you're seeing is working with other physicians, OT [occupational therapy], anything like that, how do you work with them to provide comprehensive care in that multidisciplinary manner?

Stéphane Auvin: Comprehensive care is very important. And I mean, in this complex condition to go in the good direction, it's very important.

At least in our setting, epilepsy nurses are very important. Of course sometimes we have communication directly between physicians, but because of the time adjustment, sometimes it's not the same schedule.

Joy Mazur: What is your advice to families and caregivers when they're helping their loved one and how to work through various symptoms or struggles they may have?

Stéphane Auvin: I mean, if there is any emerging symptoms, anything that is new and you don't know, maybe you can record what is worrying you. And I mean, videotape with your cell phone, for example, now everybody has a camera in his pocket. And then you can share during the clinic or even now there are more and more apps or devices that can help to send to the healthcare system for advice. Or tele-expertise, for example. I think, you know, it's always very difficult to describe a behavior or to mimic even worse because, you know, we are not actors. The movie recording is very helpful.

And then never hesitate to mention something that is new, in particular when it's related to a new prescription.

Joy Mazur: And so then on the flip side of that, what are some of the main concerns that you hear from epilepsy professionals who maybe don't work with autism a lot? What kind of questions do they have?

Stéphane Auvin: Let's speak about my two scenarios again.

For the patient with early onset epilepsy, developmental epileptic encephalopathies or drug resistance, we need more screening. Because the profile of autism spectrum disorder is not looking like the historical picture of what we think is autism spectrum disorder. Then screening is very important because it leads to some help to dedicated care and it's very helpful for patients and families. And this is my advice for this first scenario.

In the other way around, movies are very important to do diagnosis of epilepsy in patients with autism spectrum disorder. Because if you think about the main features of what we are looking for as care professionals for epilepsy, we will ask, okay, did your kid suddenly stop his activities? Is your kid having like repetitive movement? Does your kid change abruptly of behavior or they are not responsive? I mean, if you ask these very simple questions that usually lead to the idea, okay, maybe it's seizures, or 100% of the family of patients with autism spectrum disorders, they may say yes.

Then it's more complicated than the usual interview for an epilepsy diagnosis, meaning you need expertise, and at some point it's very important to start by recording the video. And not only the video, you should advise the family when the behavior that you think it's a seizure, it's happening, try to interact, not only call the patient. Touch or take the toys or put your finger in front of the eyes, then to see what is the reaction. Let's say if it's seizures with impairment of awareness, the patient that you interact with will continue to have an impairment of awareness. But if it's not seizures, then it's most likely that they will react.