Benign Childhood Epilepsy with Centrotemporal Spikes
Natalio Fejerman
Introduction
The recognition that some epilepsies in children have focal clinical manifestations and unilateral electroencephalographic (EEG) discharges with benign evolution has been one of the most interesting contributions to pediatric epileptology in the last 50 years.
The concept of idiopathic and benign focal epilepsies of childhood is relevant not only from the theoretical point of view, but also as a practical tool, because the term implies both absence of structural brain lesions and a genetic predisposition associated with age-dependent seizures. Benign childhood epilepsy with centrotemporal spikes (BCECTS) is the most frequent of the benign focal epilepsies of childhood and represents 15% to 25% of epilepsy syndromes in children <15 years of age. In addition, it is the most frequent epilepsy syndrome in school-age children.25,101,125
Historical Perspectives
Yvette Gastaut was the first to state, in 1952, that “pre-rolandic spikes” could be “functional” instead of an indicator of a cortical lesion.48 In 1958, Bancaud et al.6 and Nayrac and Beaussart91 reported the first series of patients, emphasizing that rolandic or pre-rolandic spikes or spike-waves constituted EEG features that were typical of childhood, and that should not induce “neurosurgical behaviors,” although they did not describe a clear electroclinical correlation. In 1960, Gibbs and Gibbs51 stated that prognosis was much better in children with centrotemporal spikes than in those with spikes in the anterior temporal regions. In the same year Faure and Loiseau39 wrote of “Rolandic spike-waves without focal significance,” referring to age of onset and sleep as a trigger of seizures, and to spontaneous electroclinical normalization in puberty. Clinical features of seizures were not defined, however, because they found generalized seizures in 13 of their 15 cases. In 1966, Trojaborg114 published a longitudinal study in a cohort of 519 children with focal spike discharges. Two hundred and eighty of these patients had cerebral palsy. The main purpose of this work was a detailed analysis of the significance of acute sharp waves in the EEG and their correlation with brain lesions. The author recognized that one prognosis was good in children with centrotemporal spikes, but did not determine their clinical correlates. Also in 1966 the most important series of children with temporal lobe epilepsy was published.97 Among the 100 children followed, 33 constituted a subgroup without pathologic history and with mean age of onset of seizures between 7 and 8 years. It is now easy to imagine that a significant proportion of these 33 children might have had the diagnosis of BCECTS.
One year later, two independent groups reported their series of patients with a peculiar form of epilepsy to be differentiated from other focal epilepsies, mainly from temporal lobe epilepsy. Loiseau et al.78 presented 122 cases with onset of seizures at school age and rolandic sharp waves in the EEG. In 80% of their patients, seizures occurred during sleep and were frequently motor with predominant involvement of the face. Lombroso81 provided a clear description of the seizures, emphasizing somatosensitive symptoms in tongue, oral mucosa, and gums, along with speech arrest, and proposed the term “sylvian seizures,” recognizing also the particular focal EEG features. Both Loisseau et al. and Lombroso stressed the benign character of the condition in terms of the evolution of seizures and normalization of the EEG. In 1972, Blom et al.13 reported a prevalence and follow-up study and proposed that the condition be named “benign epilepsy of children with centrotemporal EEG foci.” Several long-term follow-up studies confirmed the good prognosis.8,9,72,80 Atypical and not-so-benign evolutions have been reported in some patients.1,40,43,44,46,54 This form of epilepsy is now called benign childhood epilepsy with centrotemporal spikes and is placed in the group of idiopathic localization-related (focal, local, partial) epilepsies in the International Classification of Epilepsies and Epileptic Syndromes.21,36
Definitions
The most common name besides BCECTS appearing in the literature referring to this condition is “benign rolandic epilepsy.” The term “benign” has been questioned for other epilepsy syndromes because some epileptologists believe that benign implies a natural evolution to remission of seizures and EEG abnormalities even without treatment. In this sense, BCECTS conforms to the aforementioned concept, even if we must accept that there are exceptions to the rule.
Onset during childhood with hemifacial motor, seizures, speech arrest, and sialorrhea, occurring mostly during sleep, along with distinctive centrotemporal spikes in the EEG are well-defined features that allow a prompt diagnosis and assure a good prognosis, although subsets of atypical cases with slight compromise of neurological functions are increasingly being reported.102,120,122
Epidemiology
Benign childhood epilepsy with centrotemporal spikes accounts for between 15% and 25% of all epileptic syndromes in children between the ages 4 and 12 years.19,25,101,121 Its annual incidence has been reported to be between 7.1 and 21 per 100,000 in children <15 years of age.56 Because nocturnal seizures can be overlooked in diagnosis, this disorder may be even more common than generally suspected. There is a slight male predominance.82
The prevalence of epilepsy is much higher among close relatives of children with BCECTS than in a matched control
group.17 In one study, 15% of siblings had seizures and rolandic spikes, 19% of siblings had rolandic spikes without attacks, and 11% of the parents had had childhood seizures that had disappeared by adulthood.57
group.17 In one study, 15% of siblings had seizures and rolandic spikes, 19% of siblings had rolandic spikes without attacks, and 11% of the parents had had childhood seizures that had disappeared by adulthood.57
In an epidemiologic study of epilepsy in a cohort of 440 consecutive children, benign rolandic epilepsy of childhood accounted for 8% of patients (neonatal seizures were excluded from the analysis).67
Etiology and Basic Mechanisms
The high incidence of a positive family history for epilepsy and focal EEG abnormalities indicates the importance of genetic factors in the etiology of BCECTS.13,17,27,82 Most authors speak of an autosomal-dominant trait with variable penetrance.17,57 This type of inheritance was also suggested by studies of monozygotic twins with rolandic discharges64 and of HLA antigens and their haplotypes.34 However, in another study of clinical and genetic aspects in children with benign focal sharp waves, including 134 probands with seizures (24% of which had typical rolandic seizures), the findings agreed with a multifactorial pathogenesis with “benign” focal epileptiform sharp waves.32 Epileptic seizures appear in only 25% or less of individuals with this EEG trait.82 Expression of the gene may be influenced by other genetic and environmental factors.79 Linkage to chromosome 15q14 was found in 54 patients of 22 families with benign childhood epilepsy with centrotemporal spikes.92 However, in a study of 70 families with BCECTS in Italy, the same linkage could not be found.104 Approximately 10% of patients have a history of febrile seizures, and this also supports a genetic predisposition for febrile seizures expressed at earlier ages in children with BCECTS.63,71
A family with nine affected individuals in three generations was reported showing the features of rolandic epilepsy associated with oral and speech dyspraxia and cognitive impairment.109 Similar seizures and the EEG phenotype of BCECTS were found in three children with de novo terminal deletions of the long arm of chromosome 1q, and the authors suggested that it could be a potential site for a candidate gene.118
Although the pathophysiology of BCECTS is unknown, and there is no associated structural lesion, the typical ictal clinical behavior and EEG discharge indicate a disturbance in the sylvian and rolandic areas.37 Electrophysiologic studies, however, fail to demonstrate a discrete generator, and a large, shifting area of regional cortical dysfunction may be present. In some patients with BCECTS, the occurrence of generalized spike-wave EEG discharges, as well as focal spikes in other areas, suggests a relationship between BCECTS and the idiopathic generalized epilepsies, as well as with other idiopathic localization-related partial epilepsies.73,82 Between 10% and 20% of patients with centrotemporal spikes may also have sharp slow wave complexes in other cortical locations.100
Combined recording of interictal spikes and somatosensory-evoked potentials led to the conclusion that in some patients multiple simultaneous neuronal populations are active within the central region.7
Magnetoencephalographic (MEG) analysis of generator sources and propagation of rolandic discharges in BCECTS using three-dimensional dipole localization suggested that rolandic discharges are generated through a mechanism similar to that of somatosensory-evoked responses.86 A localization analysis of spontaneous magnetic brain activities also suggested the value of MEG for pathophysiologic studies.65 Six children with bilateral centrotemporal synchronous discharges were studied using MEG and EEG with equivalent current dipole modeling. Results implied cortical epileptogenicity in bilateral perirolandic areas.75 Interictal spikes were recorded in seven children during functional magnetic resonance imaging (fMRI) acquisition using a MR-compatible digital EEG system. Spike-related activation in the perisylvian central region was found in three of them.14 High-resolution EEG and MEG and a realistic volume conductor model were used to study the spatiotemporal aspects of the sources of spikes in children with benign rolandic epilepsy. Results for EEG and MEG were different. Both high-resolution EEG and MEG revealed that in some cases there were multiple sources for spikes that were well separated in space and time, whereas in other cases only single-source activity was present.61
Recent papers considered BCECTS, Landau-Kleffner syndrome, and electrical status epilepticus in sleep as a spectrum of disorders with a common transient, age-dependent, nonlesional, genetically based epileptogenic abnormality, involving a perisylvian epileptic network. Halasz et al. refer to “mild to severe epileptic encephalopathy limited to the perisylvian network, where the cognitive impairment is caused by epileptic discharges interfering with cognitive development.”55
Clinical Presentation
We discuss the main clinical features of BCECTS, assuming that this is an idiopathic focal epilepsy that appears in children with normal neurological development, although this syndrome has also been reported in patients with nonevolutive brain lesions.108
BCECTS begins between 4 and 10 years of age in 90% of patients, and the median age of onset is approximately 7 years. There are no reports of BCECTS occurring during the first year of life or after age 15 years, and cases with seizure onset before the age of 2 years are extremely rare.25,42
BCECTS is seen more frequently in males, with a male-to-female ratio of 3:2.
Seizures are clearly related to sleep, whether during the night or the day. This is seen in 80% to 90% of patients. Seizures during waking hours are more likely to occur shortly after awakening,82 although on many occasions, during the early morning the child wakes up with a seizure that really started during sleep. Seizure frequency is usually low, and around 10% of cases have only one seizure. In about 20% of the children, however, seizures are frequent and may even occur several times per day.25 Most patients have a single type of seizure, but 20% to 25% of children experience more than one type.79 Loiseau and Beaussart77 described 35 signs or components of 275 seizures analyzed in 190 children with BCECTS. We can reduce this number to a small group of characteristic manifestations of seizures:
Orofacial motor signs, especially tonic or clonic contractions of one side of the face, with predilection for the labial commissure (contralateral to the centrotemporal spikes). There are also contractions of the tongue or jaw, guttural sounds, and drooling from hypersalivation and swallowing disturbance.
Speech arrest, most probably due to tonic contractions of pharyngeal and buccal muscles, constituting anarthric seizures. In fact, the patient cannot speak during the seizure because either he or she wakes up with hemifacial contractions or while awake opens his or her mouth with the intention to speak but stays blocked in that position.
Somatosensory symptoms, namely unilateral numbness or paresthesia of the tongue, lips, gums, and
cheek, are frequent, but sometimes have to be looked for in the context of recalling previous events.
The three mentioned groups of manifestations are related to an epileptic activation of lower rolandic motor and somatosensory areas. In terms of sialorrhea, it is not clear whether it corresponds to increased salivation, a swallowing disturbance, or both. In general, these seizures only last a few minutes. Although partial seizures are characteristic of this disorder, generalized seizures are not infrequently observed, particularly in younger children.71,82 The initial event is often a nocturnal hemifacial convulsion, which may spread to the arm and the leg or may become secondarily generalized. It is highly probable that in these cases a focal seizure begins during sleep with a rapid generalization and loss of consciousness, which makes it difficult for the child to remember what happened. This is in line with the fact that almost all seizures that start while the person is awake are focal.
Behavioral and learning problems are less frequent than in other forms of childhood epilepsy.58 It is often mentioned that children with BCECTS are free of neurological and psychological impairments.45,46,71 In recent years, however, several reports based on meta-analyses of published series, retrospective analysis of patients, or prospective cohort studies have reported a higher incidence of learning and/or language difficulties in these children. In a study of 40 children with centrotemporal spikes with and without seizures compared with 40 healthy controls, patients were significantly impaired in IQ, visual perception, short-term memory, and psychiatric status. The deficits in IQ correlated more with frequency of EEG spikes than with the frequency of seizures.122 Similar findings were reported in 19 children with this syndrome.29 An increased frequency of rolandic spikes has been reported in children with attention deficit hyperactivity disorder.59 One study found a consistent pattern of language dysfunction in 13 of 20 children with BCECTS, suggesting interictal dysfunction of the perisylvian language areas.113 A longitudinal study of one boy with acquired epileptic dysgraphia was reported. Most probably, in this case, the acquired regression of graphomotor skills was associated with an increase in spike frequency as happens in cases with atypical evolution of this syndrome.33 In a more recent report, written language skills were compared in 32 children with typical BCECTS with 36 controls. As a group, the patients with BCECTS performed significantly worse than controls in spelling, reading aloud, and reading comprehension, presented dyslexic-type errors, and frequently had below-average school performance.102 Language assessment in 16 children with BCECTS also showed that the domains of expressive grammar and literacy skills were affected in a significant proportion of the cases.89 A comprehensive study of neuropsychological and language profiles of 42 children with BCECTS selected using strict clinical and EEG criteria showed that the patients have normal intelligence and language ability, although a specific pattern of difficulties in memory and phonologic awareness was found. No correlation between EEG features and the aforementioned impairments was demonstrated.94
Atypical features in benign childhood epilepsy with centrotemporal spikes can be seen both clinically (daytime-only seizures, postictal Todd paresis, prolonged seizures, or even status epilepticus) and in EEG features (atypical spike morphology, unusual location, or abnormal background). In a retrospective case series, atypical clinical features were seen in 50% of patients and atypical electrographic features in 31%.123 In a follow-up study of 74 children with typical rolandic epilepsy and 14 with atypical features, a significantly higher percentage of learning and behavioral disabilities was found in the second group.120Related posts:
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
