Autism and Epilepsy
Autism spectrum disorders (ASD) and epilepsy are both clinically heterogeneous entities whose co-occurrence has long been recognized to exist at a frequency that is greater than could be predicted by chance alone. This observation has led to interest in the possibility of shared causal pathways for both disorders and raises the possibility of future novel interventions that could impact the course of both conditions. Further, the co-occurrence of the disorders presents several diagnostic and treatment challenges and controversies. This encyclopedia entry will provide an overview of terminology, epidemiology and etiology, clinical expression, available findings on developmental course, issues in differential diagnosis, and treatment considerations.
Terms associated with autism include autism spectrum disorder (ASD), high-functioning autism (HFA), pervasive developmental disorders (PDD), and infantile autism. ASD are behaviorally defined disorders, with literature still utilizing these terms despite the presentation of revised diagnostic criteria in the DSM-5. These terms are defined in introductory chapters in this encyclopedia. Key terms associated with epilepsy include seizure disorder and pediatric seizure disorder. A seizure is commonly understood as uncontrolled electrical activity in the brain, which may produce a physical convulsion, minor physical signs, changes in consciousness, thought disturbances, sensory disturbances, or a combination of symptoms. While the terms seizure disorder and epilepsy are generally used interchangeably, epilepsy is more formally defined as having two or more seizures within a set period of time, most often within 3 years, for which there is no other identifiable cause such as mass lesion, head trauma, infection, toxic exposure, or metabolic derangement (Matson and Neal 2009).
Epidemiology and Etiology
The prevalence of autism in the pediatric population is approximately 14.6 cases per 1,000 children, with a national rate of approximately 1 in 68 cases at age 8 years. Estimated prevalence was significantly higher for boys at about 23.6 per 1,000, than for females at about 5.3 per 1,000 (Christensen et al. 2016). These rates are higher among non-Hispanic white children at about 15.5 per 1,000 than for non-Hispanic black children (13.2 per 1,000) and Hispanic children (10.1 per 1,000); however, it is suspected that these differences in rates are related more to access to care as opposed to true differences in prevalence across racial and ethnic lines (Durkin et al. 2010; Morbidity and Mortality Weekly Report 2012). Further, autistic traits are about 4.5 more commonly seen in males, with a staggering rate of 1 in 42, than for females (1 in 189) (Christensen et al. 2016).
In the general pediatric and adult population, the prevalence of epilepsy is 2–3% (Canitano 2007), with recent estimated lifetime rates of about 10.2 per 1,000 (Russ et al. 2012). The prevalence of having epilepsy and autism as co-occurring, or perhaps comorbid conditions ranges widely from 8% to 30% (Mouridsen et al. 2013; Russ et al. 2012; Spence and Schneider 2009; Tuchman et al. 2010), with children with ASD having a 7- to 10-fold increased odds compared to controls for having epilepsy (Jokiranta et al. 2014; Tuchman et al. 2013). Viscidi et al. (2013) found that the prevalence of epilepsy in children with ASD was about 12–13% for children ages 2–17, with rates rising to about 26% for adolescents age 13 and above. For children with epilepsy, about 30% will eventually have a diagnosis of ASD (Keller et al. 2017), and individuals with epilepsy diagnosed in childhood are at significant risk for later manifestation of ASD (Sundelin et al. 2016). While there does not appear to be a specific form of epilepsy that is present in children with ASD, epilepsy does appear to be present more frequently in individuals with ASD and Intellectual Disability, with the rates increasing with the severity of the Intellectual Disability (Jokiranta et al. 2014). Of significant concern, there also is increased risk for mortality for individuals with ASD and epilepsy, particularly in the presence of an intellectual disability (Woolfenden et al. 2012).
The frequency of co-occurrence of these two disorders has led to interest in the possibility of shared etiological mechanisms in seizure disorders and ASD. Proposed theories of shared causality have been related to the deleterious effects of the seizures themselves and associated imbalances between neuronal excitation and inhibition (Stafstrom and Benke 2015), with a particular focus on impaired GABAergic signalizing as being a common denominator for co-occurring ASD and epilepsy (Kang and Barnes 2013; Tuchman et al. 2013). Additionally, contemporary mechanistic understandings of several key neurodevelopmental disorders have led to new theories about the shared role of impaired plasticity during development (Keller et al. 2017).
For example, portions of the temporal lobe of the brain and associated neural pathways are likely to be key brain regions in the complex network that has been described as “the social brain.” The temporal lobe has long been a suspected region of importance because of the relative frequency of temporal lobe epilepsy both among patients with epilepsy with social challenges and among those with ASD and epilepsy. Animal research using mouse models has demonstrated that mice with induced temporal lobe seizures exhibited less social behavior than control mice (Marin et al. 2008). Neuroimaging studies in patients with temporal lobe epilepsy have provided evidence showing damage to other recognized social brain structures in this network, such as the hippocampus (Dager et al. 2007). Further, such studies also have begun to show a linkage between aberrant neural migration over the course of neurodevelopment (Blackmon 2015) and general neurological vulnerability (Gilby and O’Brien 2013) to seizures in individuals with ASD.
Another set of examples wherein a potential shared mechanism for both ASD and seizure disorders has begun to be explored comes from the study of several recognized genetic syndromes that are associated with both autism features and seizures (Keller et al. 2017; Lee et al. 2015). In this regard, fragile X, tuberous sclerosis complex, and Rett’s syndrome all have been proposed as possible models of overlapping causality in ASD and epilepsy/seizures. For example, Rett’s syndrome is a neurodegenerative disorder that affects girls and is currently understood to be caused by mutations in the gene encoding methyl-CpG binding protein 2 (MeCP2). Rett’s syndrome is characterized by regression of verbal skills along with repetitive hand motions that usually begin to occur between 6 and 18 months of age (Brooks-Kayal 2010). Up to 90% of Rett’s syndrome patients develop seizures (Canitano 2007). Tuberous sclerosis has been associated with both epilepsy and autism (Jeste et al. 2016). The prevalence of tuberous sclerosis in the general population is around 1 case per 10,000 to 20,000 (Sherpherd 1999). Around 1% of children with autism will have tuberous sclerosis (Harrison and Bolton 1997), and approximately 80% of patients with tuberous sclerosis will also have seizures (Canitano 2007). With respect to epilepsy, tubers are thought to be foci of epileptic activity, and many of the ASD symptoms have been linked to tubers found in the temporal lobes of the brain (Bolton et al. 2002). Finally, fragile X syndrome is the most common form of inheritable intellectual disability, and frequently manifests with co-occurring autism and seizures. This syndrome is caused by excessive CGG trinucleotide repeats on the X chromosome, methylating either in whole or in part the Fragile X Mental Retardation gene leading to many of the phenotypic features associated with fragile X syndrome (Brooks-Kayal 2010). With approximately one third of individuals with fragile X syndrome showing co-occurring ASD, this syndrome provides a clear single gene disorder for examining not just autism but its related comorbidity.
While exact mechanisms for the behavioral manifestations remain unknown in each of these disorders, there has been an expanding knowledge base relating to presumed causal genetic defect(s) and their downstream molecular effects. Resultant impaired inhibitory/excitatory regulation and impaired neuroplasticity have been proposed as a possible common explanation for seizures and ASD-related behaviors (Brooks-Kayal 2010). Further, a number of other gene mutations have been associated with ASD, Intellectual Disabilities, and epilepsy/seizures including the genes encoding neuroligins, neurexins, arestelles region X-linked (ARX), and neuropilin-2 (Brooks-Kayal 2010).
Clinical Expression and Pathophysiology
There are a number of ways that the co-occurrence of seizures and ASD can be examined in terms of clinical expression and variables associated with seizure pathophysiology. These include: type of seizures, seizure location, epilepsy syndromes, age of seizure onset, level of intellectual functioning, and developmental course.
Type of Seizures
There are several classification schemas for seizure types and epilepsy. The most commonly used classification is based on the broad categories of generalized seizure onset versus focal onset, each with subcategorizations based on various clinical features and origin of seizure activity. Further, there are numerous recognized epilepsy syndromes. Seizure types in individuals with ASD are highly variable, and multiple seizure types in the same individual are not uncommon. It is important to note, though, that the prevalence of particular seizure types among those with both disorders does not seem to differ significantly from the distribution of seizure types in epilepsy patients in general (Sternberg 2003).
There is a suggestion that seizure location may point to a relationship with autistic features or autism. In epilepsy in which the seizure activity manifests from the frontal lobe, behavioral changes may include irritability, altered mood, subtle changes in alertness, associated attention dysregulation, and cognitive rigidity features often associated with ASD (Fohlen et al. 2004). Seizures originating in the temporal lobe may be associated with autistic features or autism (Hamiwka and Wirrell 2009) in that the individual may present with affective blunting, odd or impaired language functions, including impairments in core language functions or pragmatics, and poor recognition of faces.
The relationship between ASD and seizures also can be understood by considering the presence of an epilepsy syndrome. There are numerous epilepsy syndromes, and those that are believed to contribute to progressive disturbance in cerebral function may be termed “epileptic encephalopathies.” These disorders begin early in life and are often associated with regression of cognitive, language, and other neurodevelopmental functions. Many children with these disorders may present with features of ASD or they may in fact meet diagnostic criteria for an ASD (Nabbout and Dulac 2003: Nabbout and Dulac 2008). Among these syndromes, infantile spasms (IS), Landau-Kleffner syndrome (LKS), and epilepsy with continuous spike-waves during slow-wave sleep (CSWS) are most strongly associated with ASD symptomology (Ballaban-Gil and Tuchman 2000).
In IS, the association with ASD may as high as 35%, and this risk seems to increase in the presence of a severe intellectual disability, structural brain lesions, and ongoing epileptiform activity in frontal brain regions (Kayaalp et al. 2007; Saemundsen et al. 2007, 2008). LKS and CSWS have overlapping symptoms in relationship to seizure presentation, and both manifest features that overlap with ASD symptoms. The failure or regression of language development in these disorders often leads to confusion with autistic regression that is reported in children with and without underlying seizure disorders (Canitano 2007).
Age of Seizure Onset
The relationship between ASD and seizures/epilepsy can also be investigated by considering the age of seizure onset. It has been theorized that epilepsy with a late onset during adolescence is brought on by the hormonal fluctuations associated with puberty (Gillberg 1991). One study of children with autism showed that seizure activity peaks between 3 and 10 years of age (Matson and Neal 2009). Other studies, however, have suggested that epilepsy has two peaks in children with autism: one during infancy and another during adolescence (Volkmar and Nelson 1990). The peak during infancy may correlate with the peak of seizure activity that is seen in children with epilepsy without autism, while the second peak during adolescence may be unique to children with autism (Nomura et al. 2010). Recent data have challenged this bimodal distribution, suggesting that the primary peak occurs by 6 years of age (Jokiranta et al. 2014).
The range of the overall level of intellectual functioning in individuals with ASD is quite large and variable; however, it has been well established that in populations of children with epilepsy, the risk of autism or autistic features is increased among those with the lowest intellectual functioning (Hamiwka and Wirrell 2009; Jokiranta et al. 2014). Among populations of children with ASD, those with severe intellectual disability, severe receptive language deficits and motor dysfunction (i.e., those with more severe autism symptoms) have the highest risk of epilepsy (El Achkar and Spence 2015; Mulligan and Trauner 2014; Tuchman et al. 2009); and, conversely, children with ASD and epilepsy manifest more cognitive and neuropsychiatric difficulties than those without epilepsy (Viscidi et al. 2014; Weber and Gadow 2017).
When considered independently, the developmental course, severity, and outcomes of individuals with ASD and epilepsy are highly variable and dependent on numerous factors. To date, there are scant empirical data related to the moderating or mediating effects of epilepsy and ASD on one another in relation to developmental course and outcomes. In general, children with comorbid or co-occurring ASD and seizures/epilepsy have lower IQ, lower adaptive behavior, more emotional problems, and have more frequent use of psychiatric medications (Matson and Neal 2009). Also, a higher rate of seizure activity has been linked to decreased intellectual functioning (Jokiranta et al. 2014; Matson and Neal 2009) but is unclear how medications or other factors (e.g., other neurological factors) may be contributing to this suspected association. Additionally, the presence of temporal lobe seizures has been described as a poor prognostic indicator in relation to social adaptation among individuals with ASD and seizure disorders (Matson and Neal 2009). As noted above, the notion that children with comorbid ASD and seizure disorders have more pronounced social impairment when compared to children with ASD who do not have seizures has been proposed, but this issue is only beginning to be evaluated (Tuchman 2013).
Evaluation and Differential Diagnosis
Issues in Differential Diagnosis
Early diagnosis and treatment of both epilepsy and autism are crucial in order to maximize development and quality of life (Tuchman et al. 2010). Early identification and treatment allow for the optimal usage of all therapies and resources available. The potential co-occurrence of these disorders does raise several important issues in differential diagnosis. For example, the mainstay of evaluation in seizure disorders is the electroencephalogram (EEG), but a seizure evaluation also can include metabolic and genetic components. It is important to note that abnormal EEG activity can be seen in 7–28% of children with autism, but without any other symptoms of epilepsy (Youroukos 2007). On the other hand, high-functioning individuals with autism may be missed when presenting for epilepsy treatment (Matsuo et al. 2010). The association between autism and seizures has led the Committee on Children with Disabilities of the American Academy of Pediatrics to recommend prolonged sleep-deprived EEG in children with ASD showing developmental regression or in those where there is a high suspicion of subclinical seizures (American Academy of Pediatrics 2001). Due to the current dearth of empirical knowledge about subclinical epileptiform activity and its treatment, universal screening via EEG for all children with ASD has not yet been recommended as a standard of care (Johnson and Myers 2007), but its routine use has been suggested (Swatzyna et al. 2017).
Another important area of concern relates to the convergence of sleep problems in the populations of children with ASD and epilepsy/seizures. Sleep difficulties are common among individuals with neurologic disorders in general as well as in those with ASD and seizure disorders (Malow 2004). Screening for sleep problems and formal sleep evaluations (based on clinical need) are often important for individuals presenting with comorbid ASD and epilepsy (Accardo and Malow 2017). Sleep disorders have significant implications for behavioral functioning and quality of life beyond challenges associated with the underlying disorder (Clarke et al. 2005), such as creating daytime sleepiness, increased irritability, less efficient cognitive functioning (potentially in addition to cognitive impairment), and decreased seizure threshold. Further, sleep studies in children with ASD and sleep problems in rare instances may elucidate a previously unrecognized seizure disorder related to sleep (Accardo and Malow 2017; Malow 2004).
Early recognition of ASD and co-occurring epilepsy is important in that it is hoped that developmental outcomes can be improved via early treatment. Medication is a first-line treatment in children with epilepsy. The chief goal here is to eliminate (or lessen) all seizure activity while minimizing medication-related side effects such as behavioral problems or weight gain. In autism, psychosocial and behavioral interventions are commonly used as first-line interventions for behavioral symptoms. In autism, medication treatment is used as an adjunctive therapy to lessen symptoms of inattention, hyperactivity, repetitive behaviors, impulsivity, irritability, and aggression (Tuchman et al. 2010).
Antiepileptic medications (AEDs) are the mainstay of treatment in epilepsy. Several AEDs are used commonly in general psychiatric practice due to beneficial psychotropic properties, most notably in mood stabilization and the mitigation of aggression. Examples include valproic acid, carbamazepine, lamotrigine, and levetiracetam. While a full discussion of this class of medication is beyond the scope of this chapter, the aforementioned AEDs have been evaluated in the ASD population with and without epilepsy in several case series or small open-label trials. At present, AEDs seemed to have had equivocal results in terms of benefit with irritability, aggression, or behaviors associated with the core features of autism such as repetitive behaviors (Hirota et al. 2014; Tuchman et al. 2010), and concerns always are present for the medications to create affective blunting and/or to negatively impact cognitive and social capabilities. Formal evaluation via large randomized clinical trials in the ASD population with seizures is lacking and will be an important future step in guiding the care of this population (Tuchman et al. 2010).
Epileptic encephalopathies associated with ASD, such as infantile spasms (IS) or Landau-Kleffner syndrome (LKS), are treated early and aggressively with AEDs, adrenocorticotropin hormone (ACTH), steroids, the ketogenic diet, or surgery. The main focus of these interventions is to improve seizure control. Outcomes of these practices as they relate to mitigation or prevention of ASD features are unknown (Crumrine 2002; Kosso et al. 2005; Trevathan 2002; Wheless 2004).
The treatment of epileptiform activity on EEG, without the presence of clinical seizures, is an area of considerable debate. This debate is most relevant among those with ASD showing cognitive regression, but without a clear epilepsy syndrome or epileptic encephalopathy. Approximately 30% of children with ASD present autistic regression, which is understood as a loss of verbal and nonverbal communication skills between approximately 12 and 24 months of age. The relationship between regression and epileptiform activity noted in this subgroup has been postulated, but remains unclear, and treatment recommendations for this subgroup remain without a clear evidence base (Baird et al. 2006; Venkateswaran and Shevell 2008).
New information about shared genetic and molecular causal pathways may provide new insights about the management of children with epilepsy and autism. For example, in fragile X syndrome, mouse models have provided evidence that FMRP dysfunction may lead to behavioral and cognitive deficits as well as seizure formation (Brooks-Kayal 2010; Penagarikano et al. 2007). A key target in this dysregulation may be the metabotropic glutamate receptor (MgluR). Modulation of MgluR in mouse models has provided promising results in terms of behavior, cognition, and seizure formation (Brooks-Kayal 2010). Several molecules that modulate the function of this receptor are currently in various phases of development. Their role in epilepsy treatment and treatment of any ASD feature remains to be seen, but it is clear that much is to be learned from conditions where ASD and epilepsy coexist (Brooks-Kayal 2010).
This encyclopedia entry provided an overview of the interesting association between autism and seizures disorders. This is an intriguing area for clinical inquiry, but it is also an area ripe for scientific investigation. With the prevalence of seizure disorders in the general population being approximately 2–3%, the rate of seizures in the population of individuals with autism is arguably as high as 22 times as much, with about one third experiencing at least one seizure by adolescence. With the recently documented prevalence of autism in the population, this combines to create a significantly large number of individuals with comorbid ASD and seizures. As a subgroup of individuals with ASD, however, this area has only begun to receive scientific scrutiny. Increased understanding of the type of seizures, identifiable neurological contributors, other associated conditions, and developmental course all should contribute to improved seizure management. Key to this understanding is early, comprehensive evaluation and associated differential diagnosis. Also, recognizing that at least one-third will manifest a seizure by adolescence implicates the need for routine and thorough developmental surveillance for seizure manifestations by an interdisciplinary group of trained professionals (Eom et al. 2014). Ultimately, coordinated multimodal treatment approaches will be critical to maintaining a good quality of life for individuals with ASD and comorbid seizure disorders. Although treatment of epilepsy is a medical necessity, it typically will not be enough to address the additional symptoms related to ASD and related social and cognitive functioning (Tuchman 2013).
References and Reading
- Center for Disease Control. (2012). Prevalence of autism spectrum disorders-autism and developmental disabilities monitoring network, United States. Morbidity and Mortality Weekly Report, 61(SS03), 1–19.Google Scholar
- Christensen, D. L., Baio, J., Van Naarden Braun, K., Bilder, D., Charles, J., Constantino, J. N., Daniels, J., Durkin, M. S., Fitzgerald, R. T., Kurzius-Spencer, M., Lee, L., Pettygrove, S., Robinson, C., Schulz, E., Wells, C., Wingate, M. S., Zahorodny, W., & Yeargin-Allsopp, M. (2016). Prevalence and characteristics of autism spectrum disorder among children aged 8 years – Autism and developmental disabilities monitoring network, 11 sites, United States, 2012. Morbidity and Mortality Weekly Report. Surveillance Summaries, 65, 1–23.PubMedGoogle Scholar
- Durkin, M. S., Maenner, M. J., Meaney, F. J., Levy, S. E., DiGuiseppi, C., Nicholas, J. S., Kirby, R. S., Pinto-Martin, J. A., & Schieve, L. A. (2010). Socioeconomic inequality in the prevalence of autism spectrum disorder: Evidence from a U.S. cross-sectional study. PLoS One, 5, e11551.CrossRefPubMedCentralPubMedGoogle Scholar
- Morbidity and Mortality Weekly Report Surveillance Summary. (2012). Prevalence of autism spectrum disorders. II. Autism and developmental disabilities monitoring network, 14 sites, United States, 2008. Morbidity and Mortality Weekly Report Surveillance Summaries, 61, 1–19.Google Scholar
- Mouridsen, S.E., Rich, B., & Isager, T. (2013). Epilepsy in individuals with a history of Asperger Syndrome: A Danish nationwide register-based cohort study. Journal of Autism and Developmental Disorders, 43, 1308–1313.Google Scholar
- Sherpherd, C. (1999). The epidemiology of the tuberous sclerosis complex. In M. Gomez, J. Sampson, & V. Whittemore (Eds.), Tuberous sclerosis complex (3rd ed., pp. 24–36). New York: Oxford University Press.Google Scholar
- Sternberg, B. S. (2003). Autism. In M. Harris & E. Thackerey (Eds.), The Gale encyclopedia of mental disorders (Vol. 1, pp. 97–102). Gale: Detroit.Google Scholar
- Youroukos, S. (2007). Autism and epilepsy. ENCEPHALOS Archives of Neurology and Psychiatry, 44, 200–203.Google Scholar