History
In this section, the following aspects of acquired epileptic aphasia (AEA) will be briefly discussed:
-
Language symptoms
-
Seizures
-
Behavioral and neuropsychologic disturbances
-
Controversial features
Language symptoms
The first manifestation of the language problem is often word deafness or auditory verbal agnosia. In many patients, auditory verbal agnosia may include lack of recognition of familiar noises; however, alert responses to sound and tonal audiograms are usually normal. In some patients, even the capability of lateralizing and/or localizing sound may be impaired.
Receptive language is often severely or profoundly impaired as a result of an interference with phonologic decoding. Although the primary problem is in the receptive sphere, this syndrome appears in a critical period of language acquisition; therefore, speech production may be affected just as badly as or even worse than language comprehension. This is often the case as the disease progresses.
In some cases, impairment may be most severe in expressive language. In one study, the aphasia was predominantly expressive in 6 of 77 patients who appeared to have AEA.
Reading and writing may be remarkably preserved in children with little speech or auditory comprehension, and these children can be taught lip reading and writing as well. Speech disturbances may include fluent aphasia, use of jargon and paraphasias, asyntaxia, and verbal stereotypies in children who are not completely mute. Some abnormalities may superficially resemble autism or psychosis, common diagnoses given to children with acquired epileptic aphasia.
The age of onset of aphasia is between 18 months and 13 years, but it is usually after 4 years and before 7 years. A few authors include in the definition of acquired epileptic aphasia patients with limited or no language development associated with paroxysmal electroencephalography (EEG). In such cases, being certain about the true age of onset of symptoms is difficult. Other authors have included cases of developmental dysphasia associated with seizure-related fluctuations in speech performance with acquired epileptic aphasia, whereas other authors have not. Further studies are necessary to determine if the response to treatment really differs to justify this separation.
Language deterioration commonly occurs over weeks or months, but acute onset after a seizure has also been described. Intermittent/episodic aphasia may be seen as well. Regarding the course of language dysfunction and its relationship with paroxysmal EEG abnormalities, no consensus exists concerning the relationship between discharges on the EEG and the presence and intensity of language problems. In many cases, continuous spike and wave during sleep seems to precede language deterioration, and improvement in the paroxysmal EEG pattern during sleep often precedes the clinical language improvement.
Several authors have found that aphasia is correlated with unilateral or bilateral temporal-lobe discharges or with periods of 1 or more years of continuous spike and wave of slow-wave sleep (CSWS) when language appears to worsen. [19] Other investigators found this correlation to be far from reliable. Soprano et al observed persistent EEG abnormalities in patients with poor language recovery, but 6 of 9 with EEG normalization remained aphasic. [11]
The relationship between aphasia and paroxysmal EEG may not be an "on-off" response. Several factors limit the reliability of the EEG data. Neurologic deficits do not closely follow the maximal EEG changes in time. Patients with unilateral motor weakness related to a seizure (Todd paralysis) often remain weak for hours, and in rare cases, days after a partial seizure is gone. Repeated and long seizures are most often associated with long postictal dysfunction, which, at some point, may not recover completely.
The assumptions that paroxysmal EEG may or may not be correlated with the aphasia fluctuation also may be flawed, because if epileptic and/or neurotoxic brain damage is present in acquired epileptic aphasia, recovery from this damage may take time or may never happen. Two findings—that some patients improve with the use of corticosteroids or adrenocorticotropin hormone (ACTH) and that patterns on angiograms resemble those seen in cerebral arteritis—suggest that inflammation and vasospasm may play a role in some cases of acquired epileptic aphasia. This phenomenon is probably not universal, because not all patients show EEG or clinical response to steroids, and 2 neuropathologic specimens from temporal lobectomies revealed no inflammatory changes.
Seizures
The prevalence of clinical seizures in acquired epileptic aphasia is 70-85%. In one third of patients, only a single seizure (or episode of status epilepticus) is recorded. In about one half of affected children, a seizure is the initial manifestation of acquired epileptic aphasia. In some, a few years may pass between the first seizure and the onset of any speech problems, whereas the opposite is true in others. Seizures usually appear between ages 4 and 10 years, and many series show that remission of the seizures before adulthood (often before age 15 y) is the rule.
Clinical seizures are often easy to treat, but normalization of EEG discharges can be challenging. Among patients in whom ictal semiology is well described, 59% had partial seizures, 39% had generalized tonic-clonic seizures, and 16% had atypical absences. Myoclonic seizures involving the face and eyes have been described. About 12% of patients have a family history of epilepsy.
Behavioral and neuropsychologic disturbances
Behavioral disturbances are seen in as many as 78% of the patients. Some children may appear deaf or autistic. The diagnosis of autism is often considered because of the common presence of asyntaxia, parapsias, and verbal stereotypies. Hyperactivity and a decreased attention span are observed in as many as 80% of patients. Aggressive and oppositional behavior, including rage attacks, is not unusual. The aggression and rage may be so prominent that the patients may be admitted to a psychiatric service rather than a neurologic service, either initially or during the course of the disease. Anxiety and avoidant or bizarre behavior may also be seen.
Although behavior patterns are thought to be secondary to the language impairment in acquired epileptic aphasia, some patients may have complex, hard-to-explain, and bizarre behaviors, such as avoidance of interpersonal contact and gestural stereotypies. In some cases, frankly psychotic behavior has been described.
Other aspects of cognition are traditionally said to be preserved, but a discrepancy between nonverbal and verbal skills is sometimes seen. Diffuse neuropsychologic deficits may appear over time. Short-term memory is a debilitating feature seen in long-standing cases of acquired epileptic aphasia.
Controversial features
Some cases initially thought to be benign childhood epilepsy with centrotemporal spikes later develop into a picture of acquired epileptic aphasia. In addition, some patients with early onset benign childhood occipital epilepsy (Panayiotopoulos type) syndrome may have language dysfunction because of the continuous spike-and-wave discharges during slow-wave sleep. [20, 21]
Physical Examination
Mental status examination of patients with acquired epileptic aphasia (AEA) demonstrates language, speech, and behavior problems, as described earlier (see History). A history of a poor understanding of spoken language should be substantiated by performing objective testing of all aspects of the patient's speech and language, such as the child's comprehension, repetition, reading, and writing. Bedside and/or office testing of language skills should be supplemented by formal neuropsychologic testing.
Besides language, speech, and behavior, physical and neurologic examination of patients with acquired epileptic aphasia shows motor clumsiness or, less frequently, apraxia. In some cases, frank abnormalities of tone and motor function are noted, but these findings are the exceptions rather than the rule.
Patients with acquired epileptic aphasia secondary to a tumor, stroke, or head injury commonly have hemiparesis (usually right sided). Signs of increased intracranial pressure, such as papilledema and, in more extreme cases, erratic respirations, bradycardia, and hypertension, should alert the clinician to the possibility of a mass lesion.
Syndromes Related to Acquired Epileptic Aphasia
This section will discuss the following (see also Diagnostic Considerations in the Differentials section):
-
Oromotor-expressive language deficit associated with a centrotemporal epileptic focus (acquired expressive epileptic aphasia)
-
Developmental dysphasia or developmental expressive language disorder
-
Acquired epileptic aphasia [AEA] and autism
-
AEA and ESES
-
Childhood disintegrative disorder
Oromotor-expressive language deficit associated with a centrotemporal epileptic focus (acquired expressive epileptic aphasia)
A rare syndrome has been described in which patients appear to have a primarily expressive language deficit associated with a centrotemporal epileptic focus. Temporary speech and oromotor disturbances may be seen in this syndrome, and voluntary oromotor functions and speech production may be affected depending on the location and spread of the epileptic discharges (more anterior or posterior in the perisylvian region). Similar to acquired epileptic aphasia, these deficits can occur as initial symptoms of the disorder without visible seizures. The range of symptoms in these patients goes from nonlinguistic deficits, such as intermittent drooling to oromotor apraxia, disfluency, and (in severe cases) full-blown anterior opercular syndrome.
Oromotor apraxia and speech problems may be congenital, or they may develop or worsen with episodes of sustained spike and wave discharges during sleep. Seizures are nocturnal and either orofaciobrachial partial or secondarily generalized. The ages of onset, progression, and recovery of the deficits are variable but depend on the degree and duration of epileptic activity. The electroencephalogram (EEG) in this syndrome shows rolandic (ie, centrotemporal) discharges, which are commonly bilateral.
During sleep, continuous spike and wave discharges may be seen on the EEG, and that pattern may be correlated with clinical deterioration. One family with this syndrome had autosomal dominant transmission with anticipation for the seizure disorder, oral and/or speech dyspraxia, and cognitive dysfunction, raising the possibility of a triplet repeat syndrome. [22] Antiepileptic medication and other agents (see Treatment), may affect the course of the disease in some cases.
Developmental dysphasia or language disorder
Developmental dysphasia is a syndrome in which language acquisition does not occur despite normal intelligence and the lack of brain or hearing pathology. The Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, Text Revision (DSM-5-TR) refers to this syndrome as language disorder. [23]
Overnight sleep recording in patients with developmental dysphasia may show epileptiform discharges; in one study, as many as 30 of 32 cases studied had these discharges even though half of the patients studied never had a seizure. The EEG abnormalities in developmental dysphasia may be more prominent during sleep, but a study found only minimal or no worsening in the transition from sleep stages 1-2 to 3 (slow-wave sleep).
Other investigators have described patients with developmental expressive language disorder and epileptiform EEGs, which suggests that these cases are "congenital variants of the Landau-Kleffner syndrome (AEA)." As can be concluded from the name of this disorder, the main problem is with expressive language, whereas in classic acquired epileptic aphasia the primary problem is in the receptive sphere. Nonetheless, differentiation between this disorder and acquired epileptic aphasia may be difficult for the following reasons:
-
In acquired epileptic aphasia, the primary problem is in the receptive sphere, but speech production may be affected just as badly or even worse than language comprehension.
-
An expressive-aphasia variant of acquired epileptic aphasia is known.
-
Cases of acquired epileptic aphasia and developmental dysphasia can be seen in the same pedigree, including a report of discordant monozygotic twins, suggesting a similar genetic abnormality with different phenotypes. [2]
-
Many patients with developmental dysphasia also have EEG abnormalities, and 3 cases seemed to respond to adrenocorticotropin hormone (ACTH) therapy.
Because of these arguments, clinicians should be open to the possibility that some patients with developmental expressive language disorder may have abnormal EEGs and that they occasionally respond to treatments used for acquired epileptic aphasia. Differentiation of developmental expressive language disorder (developmental dysphasia) and acquired expressive epileptic aphasia (oromotor-expressive language deficit associated with centrotemporal epileptic focus) with an early onset of symptoms is difficult and may be impossible in many cases.
Acquired epileptic aphasia and autism
Autism is a strong consideration in patients presenting with an acquired epileptic aphasia-like picture. Not all patients with acquired epileptic aphasia have seizures, and some patients with autism may have EEG abnormalities with or without seizures. As mentioned earlier, many families change their perception of the patient's history over time. In these cases, consulting the child's initial medical records to obtain the correct information is helpful. In that sense, retrospective analysis, especially if the initial consultation notes are not accessible, may overestimate the incidence of language regression in these patients.
The diagnosis of autism (autistic disorder) based on criteria of the DSM-IV-TR is divided in 3 subgroups: (1) impairment of reciprocal social interaction, (2) qualitative impairment in verbal and nonverbal communication as well as imaginative activity, and (3) markedly restrictive repertoire of activities and interests. Pervasive developmental disorder (PDD) is diagnosed when qualitative impairment of reciprocal social interaction and verbal and nonverbal communication skills is present without fulfillment of the criteria for autistic disorder, schizophrenia, or schizotypal or schizoid personality disorder.
A history of language regression is not unusual in autism and is obtained retrospectively in as many as 39% of children with autism and prospectively in one third of patients with either autism or PDD. Language regression occurs equally among children with autism or PDD with or without epilepsy. Children with low cognitive function are more likely to have undergone regression (34%) than those with better cognitive skills (20%). The age at which language regression occurs (ie, before or after 2 y) makes no difference in the proportion of children with epileptiform EEGs.
The frequency of overt epilepsy among patients with autism or PDD is 7.6-25%. This range partly depends on the definitions of autism and PDD and on how strictly these criteria are applied. Of these patients, epilepsy is more common among males, and the seizures start in the first year of life in more than 80% of the children. Tuchman and Rapin found that epilepsy was present in 14% of autistic children after they excluded patients with Rett syndrome. [24] Other authors have reported higher frequencies than this among patients with autism.
The relationship between autism, epileptiform EEG, epilepsy, and language regression is complex and only partially understood. In a study in which 60% of an autistic population underwent EEG, about 22% had epileptiform abnormalities. In approximately one half of the children in whom EEG demonstrated epileptiform discharges, the discharges were located over the centrotemporal region, regardless of whether the child was epileptic or had regression. Two explanations for this finding are possible: (1) Patients may have comorbidity of benign (rolandic) epilepsy with centrotemporal spike-EEG trait with autistic symptoms (eg, benign epilepsy with centrotemporal spikes, one of the most prevalent epileptic syndromes) or (2) There may be a cause-effect relationship between the epileptiform abnormalities and the autistic and language regression symptoms.
Tuchman and Rapin prospectively studied language regression in patients with PDD and found that, in nonepileptic autistic children, a history of regression was associated with a 2-fold increase in the incidence of epileptiform EEG compared with those who had not undergone regression and had no seizures. [24] The proportion of children with epilepsy or epileptiform EEGs who had regression before or after age 2 years did not differ. [24]
The types of seizures most commonly associated with autism are infantile spasms, complex partial seizures, and generalized tonic-clonic convulsions. About one third of autistic patients with epilepsy have (or had) infantile spasms or myoclonic seizures. Severe mental retardation and motor deficit appear to be associated with an increased incidence of epilepsy in autistic patients. A high incidence of epilepsy occurs in patients with deficit in oral comprehension or verbal auditory agnosia. This finding is not unexpected, because it is probably due to inclusion of patients with acquired epileptic aphasia and its variants, which all can demonstrate autistic features.
Acquired epileptic aphasia and ESES
One of the main differential diagnoses of acquired epileptic aphasia is the syndrome of continuous spike-wave during slow sleep, or ESES. In fact, some authors consider acquired epileptic aphasia as part of the ESES spectrum. A few features may help differentiate these 2 syndromes, as follows:
-
In ESES, nocturnal spike and wave discharges by definition occupy more than 85% of slow-wave sleep (stages 3 non–rapid eye movement [NREM]), especially during the first sleep cycle, and become focal or disappear during REM sleep.
-
In acquired epileptic aphasia, generalized spike-and-wave discharges may continue to be maximal or may be seen only during REM.
-
The EEG in ESES may show frontal or frontocentral location of focal discharges, but in acquired epileptic aphasia the focal discharges often show temporal or parietal distribution.
The use of the source localization of the focal discharges for the differentiation of acquired epileptic aphasia from ESES has limited value, because well-documented ESES cases may have primary parietal generators for secondary generalized discharges. In many cases, the generalized discharges seen in ESES may represent secondary bilateral synchrony from a consistently unilateral focus, which can be located in either hemisphere. The seizures seen in patients with ESES are similar to the ones of acquired epileptic aphasia, but drop attacks and myoclonic and unilateral clonic seizures may be more common in ESES. The nature of the cognitive deterioration is more diffuse in ESES than in acquired epileptic aphasia.
Besides substantial language problems, patients often have reduced temporal-spatial orientation and memory function during the active phase of ESES. As measured by using the Wechsler intelligence scales for children (WISC), cognition (ie, intelligence quotient [IQ]) severely declines. Verbal scores on the WISC are affected more than performance scores. The term disintegrative epileptiform disorder has been used in reference to patients with normal development in whom deterioration of language, sociability, nonverbal communication, and cognition occurs after age 2 years in association with an epileptiform EEG, often with an ESES pattern.
Genetic factors in epileptic aphasias
Lesca et al found the following: "About 20% of cases of LKS, CSWS, and electroclinically atypical rolandic epilepsy are often associated with speech impairment and can have a genetic origin sustained by de novo or inherited mutation in the GRIN2A gene (encoding the N -methyl-D-aspartate (NMDA) glutamate receptor α2 subunit, GluN2A). The identification of GRIN2A as a major gene for these epileptic encephalopathies may provide crucial insights into the underlying pathophysiology." [25]
Rare pathogenic deletions that include GRIN2A can be implicated in neurodevelopmental disorders. Carvill et al found the following: "[They] sought to delineate the pathogenic role of GRIN2A in 519 probands with epileptic encephalopathies with diverse epilepsy syndromes. They identified 4 probands with GRIN2A variants that segregated with the disorder in their families. Notably, all 4 families presented with epileptic aphasia syndromes and accounted for 9% of epilepsy-aphasia cases. They did not detect pathogenic variants in GRIN2A in other epileptic encephalopathies (n = 475) or in probands with benign childhood epilepsy with centrotemporal spikes (n = 81)." They reported the first monogenic cause for epilepsy-aphasia. " GRIN2A mutations are restricted to this group of cases, which has important ramifications for diagnostic testing and treatment and provides new insights into the pathogenesis of this debilitating group of conditions." [26]
Childhood disintegrative disorder
Regression of language, cognition, and behavior after age 2 years has been classified as disintegrative disorder. The definition of the term is still evolving. Rapin suggested that the term disintegrative disorder should be used for patients with autistic regression with onset after age 2 years. [27, 28]
In the DSM-5-TR, childhood disintegrative disorder is subsumed under the diagnosis of Autism Spectrum Disorder. [23]
The term disintegrative epileptiform disorder has been used to describe patients with normal development who present with deterioration of language, verbal and/or nonverbal communication, sociability, and cognition after age 2 years associated with an epileptiform EEG.
The EEG abnormality is often ESES with frontal predominance. The latter may be differentiated by widespread deterioration of cognition, as opposed to the picture of acquired epileptic aphasia, which affects mostly language, at times with secondary behavioral changes. Differentiation between disintegrative epileptiform disorder and ESES may be impossible on clinical and EEG grounds, owing to the variability of the behavioral and cognitive findings in ESES.
Table 2, below, summarizes the numbers of patients with abnormal EEG findings among autism, dysphasia, and epilepsy from studies by Tuchman and colleagues.
Table 2. Epileptiform EEG Findings in Autism, Dysphasia, and Epilepsy (Open Table in a new window)
Source |
Diagnosis |
Number of Patients |
Number of Patients with EEGs |
Patients with Abnormal EEGs (%) |
Tuchman et al (1991) |
Autism with epilepsy |
42 |
40 |
75 |
Autism without epilepsy |
160 |
139 |
8 |
|
Dysphasia with epilepsy |
19 |
19 |
58 |
|
Dysphasia without epilepsy |
218 |
66 |
9 |
|
Tuchman and Rapin [24] (1997) |
PDD or autism |
585 |
392* |
NA |
With epilepsy |
NA |
66 |
59 |
|
Without epilepsy |
NA |
66 |
59 |
|
Without epilepsy but with history of regression |
NA |
155 |
14 |
|
Without epilepsy and without history of regression |
NA |
364 |
6 |
|
EEG(s) = electroencephalogram(s); NA = not applicable; PDD = personality developmental disorder. * Sleep EEGs. |