Home > SFARI Community > Community Blog > 2013 > Can epilepsy cause autism?

Can epilepsy cause autism?

Popularity tracker
Greg Boustead
26 August 2013

Last week, we reported on one of the largest prevalence studies of epilepsy in autism yet, which confirms research implicating lower cognitive function as a risk factor for epilepsy among children with autism.

Read the full article here »

The new study estimates that 26 percent of children with autism, beyond age 12, also have epilepsy, compared with an epilepsy prevalence of 2 percent in the general population.

Some common underlying brain dysfunction could explain the high co-occurrence of these disorders. Looking deeper into this connection, however, has led some researchers to consider a more provocative notion: that epilepsy per se can cause autism.

Orrin Devinsky, director of the New York University Comprehensive Epilepsy Center, is quoted in the article and has a working hypothesis of how this progression might unfold. In some cases, he says, “it is likely that the frequent, severe seizures are the actual 'brain injury' that leads to autism — analogous to recurrent concussions causing short-term memory problems.”  

The earlier epilepsy develops, the greater the child’s risk of autism: Children whose first seizure occurs before age 2 have twice the risk of developing autism as those who develop epilepsy later in life.

Although the evidence suggests a causal relationship, much more research is needed to understand how the two disorders are related. A shared genetic root may also explain their interrelated prevalence.

If research shows that epilepsy does contribute to autism, it offers the tantalizing possibility of preempting the seizures to prevent autism.

There are no drugs available to treat the core symptoms of autism. But anticonvulsants such as valproic acid effectively stave off the seizures associated with epilepsy. Intriguingly, in a study last year, researchers found that giving the anticonvulsant clonazepam to mutant mice reverses their autism-like behavior.

Using electroencephalography, scientists can detect bellwethers of impending epilepsy:  short bursts of brain activity resembling the patterns produced by seizures. As much as 85 percent of children with autism exhibit these abnormal brainwave patterns, although only a subset of them eventually develop epilepsy. That means it may be possible to stop the brain damage before it starts.

But this sort of preventive approach raises thorny ethical concerns. Giving powerful drugs that have significant side effects to young children carries obvious risks, especially when it’s unclear whether any of the children will go on to develop epilepsy or, for that matter, autism.

What do you think?                

  • Based on current evidence, do you think epilepsy can lead to autism? How can researchers test this connection?

  • Is it appropriate to treat preclinical signs of epilepsy as a preventive measure?

 
Share your thoughts in the comments section below. Or, to dig deeper, continue the conversation in the moderated SFARI Forum for researchers. Not yet a member? Learn how to register here.

Like us on Facebook » | Follow us on Twitter @SFARIcommunity » | Join our newsletter »

News and Opinion articles on SFARI.org are editorially independent of the Simons Foundation.

Comments

Name: Orrin Devinsky
27 August 2013 - 7:09AM

We really need more systematic, blinded studies of children with the continuous spikes and waves during slow wave sleep (CSWS) and electrical status of sleep (ESES), as well as patients on the autistic spectrum with frequent epileptiform discharges. These are the groups in which we would predict the 'burden' of epileptiform activity is very high and the risk to cognitive-behavioral impairment directly from the EEG activity would be high. The challenge is that these patients are relatively uncommon and any meaningful study would likely require collaboration of multiple sites for adequate power.

For now, we are left with more questions than answers. Some of these patients do appear to respond to 'spike suppressors' such as diazepam, steroids, or traditional antiepileptic drugs, but may do not.

Name: Sarah Spence
27 August 2013 - 11:08AM

I agree with Dr Devinsky that what is needed are more controlled trials. The Simons Foundation has funded our group to do just this, treating a group of children who have autism spectrum disorder and frequent epileptiform discharges on EEG (but no seizures) with a well known anti-convulsant drug that has spike suppression capabilities (valproic acid). Dr Greg Barnes at Vanderbilt University and my team at Boston Children's Hospital are working together and looking at a broad range of outcomes including behavior (ASD specific and maladaptive behaviors), sleep, attention, language, and motor & sensory impairments) and how these relate to the improvement of the EEG.

Name: Sarah Spence
27 August 2013 - 11:30AM

Your second question is indeed provocative. There are certainly data from other childhood epilepsy syndromes showing that the EEG discharges themselves (rather than the seizures) are associated with poor developmental outcomes. But there are those who believe that rather than being a causal relationship, this is simply an epiphenomenon of the underlying neuropathophysiology and that treating one will not necessarily treat the other.

To me and other clinicians who work with these children every day, I would say this is an empiric question that begs an answer. If indeed these EEG changes are a biomarker in this disorder (that doesn't have many biomarkers) then they could represent a viable and important treatment target. But we need to know if treatment does change the overall patient profile. Certainly many children with ASD have epilepsy and treating the epilepsy does not usually make the ASD phenotype go away so this is not a simple relationship.

In addition, the ethical considerations of presymtomatic treatment are enormous and much more information needs to be known before that type of study would be justified. But I am very encouraged by the renewed interest in the connection between epilepsy and autism spectrum disorders as evidenced by increased awareness in both research communities of the overlap.

Name: Eric Morrow
27 August 2013 - 12:53PM

I agree with Dr. Devinsky and Spence. Dr. Spence is a co-author on our study. The field would benefit from a large (likely multi-site) study that should include a genetics component. Dr. Spence's study funded by SFARI is exactly the sort of studies we need to address some of these questions.

As Dr. Devinsky indicated there are many questions that have not been resolved. To be clear our study published in PloS indicates what clinical correlates we see with epilepsy in autism but we are not able to discern causative relationships.

Regarding the questions above, I can contribute some clinical and scientific observations that we have made. We find that the experience of epilepsy in autism is very diverse. As Dr. Devinsky indicates, there are some children in whom epilepsy precedes autism; however, I also find that there are many, many children who have autism that precedes epilepsy. For example, our study and others show that a large proportion of those participants with co-occurring autism and epilepsy have a juvenile onset of epilepsy. Of course, we cannot rule out the notion that these children may have had sub-clinical brain activity consistent with epilepsy that is not observed and only emerges as epilepsy in adolescence. In addition to those participants wherein autism precedes epilepsy, we also see a large, large number of children who have co-occurring autism and epilepsy and the epilepsy is fairly mild -- only several life-long clinical seizures that are well controlled, while the autism and related conditions may be relatively severe.

The above clinical observations and several genetic observations suggest that at least a subset of people with autism may have a common biological cause for the autism and epilepsy. For example, many studies (including the SSC studies) are finding mutations in autism in genes that have been traditionally considered as epilepsy syndrome genes, such as SCN2A and others. In many of the participants in whom these gene mutations are found, the child is found to have autism and no sign of epilepsy.

One important question that my lab and others have pursued for several years is: Does the autism with epilepsy subtype reduce the heterogeneity of autism to such an extent that it can aid in mutation discovery? Of course, my lab, Chris Walsh's lab, Joe Gleeson and others have been pursuing this question. It would be valuable to form a genetic collaborative involving many of the labs that have been enrolling these patients specifically.

In conclusion, parents with children with autism and epilepsy commonly say to me that a major cause of distress and morbidity is managing the seizures. Understanding the cause(s) for autism with epilepsy and developing new approaches to treatment is important. Epilepsy may provide clues regarding causes and treatments in these cases. Regardless, this is a common and important problem -- for a subset of families with autism, the epilepsy is among their top concerns.

Name: Eric Morrow
27 August 2013 - 1:19PM

One additional comment: the important (although perhaps inevitably overlysimplified) hypothesis of increased excitatory:inhibitory neurotransmission (Rubenstein JL, Merzenich MM, 2003; and others) in autism may relate to the notion of a shared susceptibility to autism and epilepsy.

Some animal models wherein changes in e:i properties have been shown -- these models largely based on autism-related genes may show epilepsy but sometimes they appear not to have clinical or overt epilepsy suggesting that this imbalance may be associated with abnormalities in social behavior even without overt epilepsy (Yizhar et al, Nature 2011 to cite one of several relevant papers). Admittedly, we need to study more and more genetic models with this and other hypotheses relevant to autism in mind to understand the potential pathophysiologic links further.

Name: Elaine Chemistruck
28 August 2013 - 1:35AM

This possible connection interests me a great deal. My son, who is now, developed a seizure disorder (complex partial seizures) at age 8. We didn't figure it out until he was 9 and the seizures became generalized. It was at this point that he began to show signs of autism. At age 12, he was diagnosed with PDD-NOS. He continued to regress and, by age 15, he looked more like he had classic autism. At age 17, he was diagnosed with CDD.

Name: Shafali Spurling Jeste
30 August 2013 - 4:41AM

I am so happy that this dialogue regarding epilepsy and ASD continues. Thank you Dr. Morrow for this important and hypothesis generating paper. I am greatly intrigued by the relationship between epilepsy, abnormal EEG’s and ASD.

The only way to truly address the issue of causation (epilepsy-->autism) is to study children longitudinally. Prospective studies of high-risk infants (namely, siblings of children with ASD) have not yet included clinical EEG, however several studies, including our own at UCLA, have investigating resting state oscillations in high risk infants, with early data (ie Tierney and Nelson) demonstrating differences in resting state gamma activity between 6 month old high and low risk infants. The findings in these high risk infants support an electrophysiological biomarker of risk that requires further exploration. While differences in resting state oscillations do not equate to precursors of epilepsy, it is certainly possible that EEG signatures in early development represent a biomarker of aberrant development or, at least, a risk for aberrant development.

The potential for prophylactic spike suppressors raises tremendous ethical and clinical safety isuses, but it is compelling given that EEG abnormalities are quantifiable and therefore could be monitored as an intermediate outcome measures.

We need studies that couple clinical EEG with behavioral measures to better characterize risk for ASD. Only then can we consider prophylactic measures (ie AED's) for the prevention of ASD.

Name: Caren Haines
6 September 2013 - 12:32AM

Please visit the website of Dr. Fernando Miranda www.brightmindsinstitute.com and watch the ABC news clip showing two children who were incorrectly diagnosed with autism when they actually suffered with a treatable form of epilepsy. We need a better test because currently the standard EEG often misses the spike activity leading physicians to the wrong conclusion!

Name: Alma Dzib Goodin
11 September 2013 - 10:49PM

I have done research about perinatal triggers of neurodevelopment issues, and I found the epilepsy related, however I found something else, there is a complete corpus of data about neurovirology and this field explain that epilepsy is the brain answer of virus or bacterias, on some cases. Now, since autism diagnoses are done basically through the DSM criteria, many other situations are behind, since metabolic syndromes, virus, and other situations.
I add another find, the presence of cytokines, which can be a key. I have wrote about it in scientific magazines in Latin America.

Name: Jennifer Taylor
3 March 2014 - 3:46AM

I am convinced that there is some kind of correlation. My daughter had her first tonic-clonic generalized seizure at 38 hrs old. She was on meds for 3 years when they tapered her off her meds. She only had one febrile seizure at age 5. No further seizures until after puberty. At age 12 her ADD increased and we put her on Vyvanse, which helped. Her doctor believed it was hormone related. At age 13 severe migraines developed with her monthly cycles and grew progressively worse until she was hospitalized for them. She had her first teenage tonic-clonic seizure at age 14. They put her through a series of tests, but the epileptic activity grew worse around PMS and cycles. She always had the tonic clonic during her cycle. After they got her on seizure meds and hormone therapies, our daughter stopped having seizures, but other behaviors developed that were not an issue before. She was diagnosed with Aspergers. She refused to eat Oatmeal and Applesauce, two things she never had a problem with before the seizures. If we altered her routine unexpectedly, she would have a meltdown - behavior that was very childlike for a 15 year old. Certain clothing was more of a problem than before. Communication had always been a problem due to delayed development from infantile seizures, but it grew worse after the teenage seizures. She has been seizure free for 17 months, but now our biggest issues are the Aspergers issues. I KNOW there is a connection somewhere.

Name: mahmud
25 April 2014 - 7:19PM

Today, 2.3 Million Americans live with epilepsy with more than 1 million of them experiencing uncontrolled seizures. CBDs (Cannabidiol) which are non-psychoactive have been proven to help reduce the frequency of these seizures.

Gevitta is working hard creating a legal CBD spray that is derived from HEMP. We are starting with a crowdfunding campaign set to launch shortly in order to raise the money needed for production as well as further studies on the effectiveness of CBDs on other diseases.

Add a Comment

You can add a comment by filling out the form below. Plain text formatting.

Question: What is 10 + 4 ?
Your answer: