SEARCH

SEARCH BY CITATION

Keywords:

  • Childhood epilepsy;
  • Centrotemporal spikes;
  • Genetics

Summary

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. Disclosure
  8. References

Purpose:  To accurately determine the frequency and nature of the family history of seizures in patients with benign childhood epilepsy with centrotemporal spikes (BECTS).

Method:  Participants with BECTS were recruited from the electroencephalography (EEG) laboratories of three pediatric centers and by referral. Pedigrees were constructed for up to three degrees of relatedness for each proband. All available affected and unaffected individuals underwent phenotyping using a validated seizure questionnaire. The proportion of affected relatives according to degree of relatedness was calculated and phenotypic patterns were analyzed.

Key Findings:  Fifty-three probands with BECTS had a mean age of seizure onset at 7.8 years (range 2–12 years). Thirty-four (64%) of 53 patients were male. For 51 participants, pedigrees were available for three degrees of relatedness. Fifty-seven (2.7%) of 2,085 relatives had a history of seizures: Twenty-one (9.8%) of 214 first-degree, 15 (3%) of 494 second-degree, and 21 (1.5%) of 1,377 third-degree relatives. Febrile seizures were the most frequent phenotype, occurring in 26 of 57 affected relatives. There were 34 relatives with epilepsy: 6.5% (14 of 214) first-degree, 1.8% (9 of 494) second-degree, and 0.8% (11 of 1,377) third-degree relatives. Of 21 affected first-degree relatives: 8 of 21 had febrile seizures (FS), 4 had BECTS, 2 had epilepsy-aphasia spectrum disorder, one had temporal lobe epilepsy with hippocampal sclerosis, 2 had focal epilepsy of unknown cause, 2 had genetic generalized epilepsies, and 3 had miscellaneous.

Significance:  The frequency of epilepsies in relatives and the heterogeneous syndromes observed suggest that BECTS has a genetic component consistent with complex inheritance. Focal epilepsies are the most common seizure disorder observed in relatives, especially BECTS and epilepsy-aphasia spectrum disorder. Additional acquired or environmental factors are likely to be necessary for expression of the seizure disorder.

Benign epilepsy with centrotemporal spikes (BECTS) is the most common epilepsy syndrome in children, accounting for 5–19.4% of epilepsies beginning before the age of 16 years (Eriksson & Koivikko, 1997; Shinnar et al., 1999; Freitag et al., 2001; Shinnar & Pellock, 2002). Onset typically occurs from 3–13 years in a child of normal intellect (Lerman & Kivity, 1986). Seizure semiology begins with an aura of unilateral tongue and/or perioral paresthesia, followed by facial tonic and/or clonic activity that may spread to include the upper limb, often referred to as rolandic seizures (RS) (Lombroso, 1967). Seizures occur mostly during sleep, although children may also experience seizures awake (Lerman & Kivity, 1975). Approximately 16% of children with BECTS have only bilateral convulsive seizures (Kramer et al., 2002). The electroencephalography (EEG) signature comprises centrotemporal spikes (CTS) with a horizontal dipole brought out by sleep.

CTS are not solely associated with BECTS. Between 2% and 4% of all children have CTS, yet fewer than one in 10 has seizures consistent with BECTS (Cavazzuti et al., 1980; Okubo et al., 1994). CTS are also observed in more complex epilepsy syndromes such as Landau-Kleffner syndrome (LKS), epileptic encephalopathy with continuous spike-and-wave during sleep (CSWS), and atypical benign partial epilepsy (Doose et al., 1997). CTS occur in 10–28% of children with autism spectrum disorders (Tuchman & Rapin, 1997; Lewine et al., 1999; Chez et al., 2006) and 6% of children with attention-deficit/hyperactivity disorder (Holtmann et al., 2003). The significance of CTS in these disorders and the general population is not known.

The clinical genetics of BECTS is not well understood and is often confused with the genetics of the CTS trait. Although CTS is necessary for the diagnosis of BECTS, the genetics of CTS is not the same as that for the epilepsy syndrome. A positive family history of seizures in probands with BECTS is reported in 9–59% cases (Beaumanoir et al., 1974; Heijbel et al., 1975; Lerman & Kivity, 1975; Doose, 1989; Degen & Degen, 1990; Shian et al., 1994; Bouma et al., 1997; Doose et al., 1997; Choy et al., 1999; Ma & Chan, 2003; Dalla Bernardina et al., 2005). These numbers are dependent on how thoroughly the family history is undertaken, and often the methods of investigation are not well defined (Ottman, 1989).

In family studies of BECTS, the proportion of affected family members in terms of degree of relatedness has not been examined, and the types of epilepsy in affected relatives are not well studied, although febrile seizures (FS) are most frequently reported (Shian et al., 1994; Ma & Chan, 2003; Vadlamudi et al., 2004). Genetic generalized epilepsies (GGEs, formerly known as idiopathic generalized epilepsies) are also described in relatives (Ma & Chan, 2003; Vadlamudi et al., 2006). However, siblings with BECTS are rarely observed, and the rarity of concordant monozygous twins argues strongly against a major genetic component (Vadlamudi et al., 2004, 2006).

Here, we study the family history of seizures in probands with “pure” BECTS to better understand the clinical genetics of BECTS. We determine the phenotypes found in relatives and the proportion of affected family members according to the degree of relatedness.

Methods

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. Disclosure
  8. References

Recruitment

Participants with BECTS were recruited in two cohorts. (A) The first unselected cohort was recruited from the EEG laboratories of three pediatric centers based on the EEG finding of CTS and RS. They also participated in a neuropsychologic and functional magnetic resonance imaging (MRI) study at Austin Health (Lillywhite et al., 2009). (B) The second cohort had heterogeneous ascertainment from: the investigators’ practices, an unselected first seizure study (Sadleir & Scheffer, 2010), and referrals for genetic studies at the Epilepsy Research Centre.

Proband evaluation

Probands were phenotyped using a validated questionnaire (Reutens et al., 1992) or following assessment by an experienced epileptologist (IES, LS, SFB). Inclusion criteria were RS and/or nocturnal tonic–clonic seizures together with CTS on EEG. Exclusion criteria were other seizure types; significant developmental delay, including speech delay or regression or intellectual disability; or a significant structural abnormality on MRI.

Family study

Pedigrees for each proband were constructed including all individuals up to three degrees of relatedness from the proband. Strenuous attempts were made to invite all affected relatives to participate in the study. Where possible, detailed interviews using the validated questionnaire were conducted with the affected individual and an eyewitness of their seizures, including their mother, where possible (Reutens et al., 1992). Medical records and all relevant investigations were obtained.

Phenotypes were analyzed for each family. For the relatives, a diagnosis of BECTS was categorized as “probable” rather than definite if the clinical picture was typical but CTS was not confirmed on EEG. This was because only an awake EEG was performed, the EEG results were not available, or an EEG was never performed.

Complex phenotypes were identified, including an epilepsy–aphasia spectrum presentation. An epilepsy–aphasia spectrum phenotype was where the presentation resembled BECTS with RS and CTS; however, the child also had developmental plateauing or regression in speech. Their EEG could show multifocal spikes with or without bisynchronous epileptiform activity. There was no evidence of CSWS where bilateral sharp and slow wave activity occupied at least 85% of slow wave sleep.

Data analysis

The pedigrees were used to calculate the proportion of affected first-, second-, and third-degree relatives. Because of a concern that cohort B (see above) could be enriched for familial cases, we initially analyzed the two cohorts separately.

Proportions were determined by calculating the sum of affected individuals compared with the total number of individuals for each degree of relatedness in each cohort. Comparisons between groups were made using a chi-square test.

All participants, or their parent/guardian in the case of minors, signed consent forms to participate in the study. This study was approved by the Austin Health and Royal Children’s Hospital Human Research Ethics Committee.

Results

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. Disclosure
  8. References

Probands

Fifty-three patients with a diagnosis of BECTS were ascertained: 25 in cohort A and 28 in cohort B. The mean age of onset of seizures was 7.8 years (median 8 years, range 2–12 years); 34 (64%) of 53 patients were male.

Fifty-one patients had classical BECTS, whereas two had atypical features. Two had episodes of prolonged RS lasting 30–40 min. Of these, one had a younger onset at 2 years, whereas the other began at 3 years. Their EEG studies showed typical CTS. Both cases experienced resolution of their seizures before 11 years of age.

Family history data

For 51 participants, pedigree data were obtained for up to three degrees of relatedness. In one case, only information on the first-degree relatives was available because the family was lost to follow-up. In one other case, information on second-degree relatives was available with incomplete information on third-degree relatives.

Separate analysis of the pedigree data for cohorts A and B showed no differences between the cohorts across all three degrees of relatives (p > 0.4 for all analyses). Therefore, the two cohorts were pooled (Table 1).

Table 1.   Family history of BECTS probands
BECTS cohortsnProbands with family history n (%)aFirst-degree relatives n (%)bSecond-degree relatives n (%)bThird-degree relatives n (%)bTotal relatives n (%)b
  1. an = number (%) of BECTS probands with a positive family history of seizures.

  2. bn = number (%) of first-, second-, third-degree relatives and total relatives with seizures.

  3. cChi-square was used to compare groups.

Cohort A2515 (60)8/98 (8.2)9/244 (3.7)12/697 (1.7)29/1,039 (2.8)
Cohort B2813 (46.4)13/116 (11.2)6/250 (2.4)9/680 (1.3)28/1,046 (2.7)
p-Value comparing cohorts A and Bc 0.30.50.40.60.9
All cases5328 (52.8)21/214 (9.8)15/494 (3.0)21/1,377 (1.5)57/2,085 (2.7)

Overall, 28 (52.8%) of 53 probands had a positive family history of FS and/or epilepsy. Of a total of 2,085 relatives, 57 (2.7%) had a history of seizures including both FS and epilepsy. Of 21(9.8%) of 214 affected first-degree relatives, 6.5% (14 of 214) had epilepsy and 3.7% (8 of 214) had FS; one relative had both FS and GGE. Of 15 (3%) of 494 affected second-degree relatives, 1.8% (9 of 494) had epilepsy and 1.4% (7 of 494) had FS. Of 21 (1.5%) of 1,377 affected third-degree relatives, 0.8% (11 of 1,377) had epilepsy and 0.8% (11 of 1,377) had FS. One second-degree and one third-degree relative had both FS and epilepsy–aphasia spectrum disorder.

Of the 14 first-degree relatives with epilepsy, focal epilepsies were most frequent, including 4 (19%) of 21 with definite/probable BECTS, 2 with epilepsy–aphasia spectrum disorder, and 2 with focal epilepsy of unknown cause (Table 2). There were two individuals with GGE.

Table 2.   Epilepsy syndromes in each degree relatives of BECTS probands
SyndromesNo. of first-degree relatives (n = 214)No. of second-degree relatives (n = 494)No. of third-degree relatives (n = 1,377)No. of total relatives (n = 2,085)
  1. aOne affected first-degree relative had febrile seizures and GGE, and one second-degree and one third-degree relative had both FS and epilepsy–aphasia spectrum. In these instances, each of these three individuals was counted twice.

  2. bAssociated with head injury, structural lesion, cerebral palsy.

Febrile seizures (probable and confimed)a871126
Epilepsies1491134
 BECTS (including probable)4115
 Epilepsy with focal seizures of unknown cause2103
 Epilepsy–aphasia spectrum disorder/CSWSa2125
 Temporal lobe epilepsy with hippocampal sclerosis1000
 Genetic generalized epilepsya2002
 Single afebrile seizure1102
 Epilepsy secondary to a known causeb1214
 Unclassified epilepsy13711
Total21152157

There were two first-degree relatives with epilepsy–aphasia spectrum disorder. A 20-year-old man had focal FS at 20 months. At 5 years, he developed both focal dyscognitive seizures and RS. Development was normal until speech and behavioral regression occurred at 10 years following a cluster of seizures. His EEG showed frequent multifocal discharges including CTS and high-voltage diffuse slowing. His EEG showed epileptiform discharges occupying 60% of the sleep recording confined to the left hemisphere. By 17 years, he was moderately intellectually disabled. The second individual was a 15-year-old adolescent with onset of RS at 3 years evolving to bilateral convulsive seizures and some episodes of rolandic status epilepticus. The EEG showed frequent multifocal discharges including right CTS, but sleep was not recorded. He had normal early development with plateauing of learning by 5 years of age; he attended a school for the intellectually disabled.

In addition, one second-degree and two third-degree relatives had epilepsy–aphasia spectrum disorder or CSWS, including a father–son pair. The son had several FS from 6–18 months and an afebrile bilateral convulsive seizure at 3 years. His EEG showed frequent right CTS. He had speech delay with significant fluctuation. His father had frequent FS from 20 months to 4.5 years. His EEG showed left mid-temporal spikes and bilateral epileptiform discharges. He had speech delay and stuttering until 8 years and continued to have reading and writing problems at school. The remaining third-degree relative was a 12-year-old girl who had afebrile focal dyscognitive seizures from age 2 years. Although developmental milestones were delayed, she showed significant regression and fluctuation of skills from 4 years. Her EEG showed CSWS. She was treated with antiepileptic medications and required oral steroids over several years.

Discussion

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. Disclosure
  8. References

There has been considerable debate about whether the most common focal epilepsy syndrome in childhood, BECTS, is a genetic disorder. Here, we examined the family history of seizures in patients with BECTS. Our study differs from previous BECTS studies, where the denominator of total relatives (affected and unaffected) has not been defined and thus the frequency of affected relatives was impossible to determine (Beaumanoir et al., 1974; Heijbel et al., 1975; Lerman & Kivity, 1975; Doose, 1989; Degen & Degen, 1990; Shian et al., 1994; Bouma et al., 1997; Doose et al., 1997; Choy et al., 1999; Ma & Chan, 2003; Dalla Bernardina et al., 2005). Our data provide a clearer picture of the frequency and nature of the family history of seizures in BECTS by looking at the seizure types and the proportion of all relatives with seizures according to the degree of relatedness to the proband. Although this is the largest number of probands included in a family study to date, it should be acknowledged that our 51 completed families may not be representative of a large cohort of BECTS families.

We looked at all relatives up to three degrees of relatedness for 51 of 53 BECTS probands, and found that 9.8% of all first-degree relatives had seizures, falling to 3% and 1.5% in second- and third-degree relatives, respectively. A lower frequency of seizures in third-degree relatives than in the general population was observed; however, given that our third-degree relatives were drawn predominantly from the older generations, this may be explained by age-related recall bias where reliable early history is not available (Ottman et al., 1995). A further limitation of the study is that not all unaffected relatives were personally interviewed, although strenuous attempts were made to speak to family members who had a history of seizures or learning difficulties; therefore, underascertainment of affected individuals may have occurred. These frequencies in relatives are typical of a complex trait, and are similar to the findings for focal epilepsy overall (Ottman, 1989).

Although the frequency of affected relatives overall was low, previous studies have provided little information about the nature of the seizure disorders in families of probands with BECTS. In our study, the most frequent phenotype in affected relatives was FS. Eight (38%) of 21 first-degree relatives, 7 (47%) of 15 second–degree, and 11 (52%) of 21 third-degree relatives had FS. These numbers simply reflect background population rates with <3.8% of all relatives in each grouping having FS (Table 2) (Nelson & Ellenberg, 1976). Compared with previous studies that reported a family history of FS in 4 (4%) of 94 and 8 (16%) of 50 probands (Shian et al., 1994; Ma & Chan, 2003), 12 (22.6%) of 53 probands had a family history of FS.

Of interest, epilepsy with focal seizures was the next most frequent phenotype, affecting 9 of 21 first-degree relatives with seizures. Four had BECTS and two had an epilepsy–aphasia spectrum disorder. In the second- and third-degree relatives, there were two cases of probable BECTS, two cases of epilepsy–aphasia spectrum disorder, and one CSWS. The epilepsy–aphasia spectrum phenotypes bore similarities to BECTS with RS and CTS; however, the individuals also showed developmental plateauing or regression in speech with multifocal spikes with or without bisynchronous epileptiform activity in sleep. Their sleep EEG did not fulfil the criteria for CSWS, as they had <85% epileptiform activity occupying their non–rapid eye movement (REM) sleep EEG. Similarly they did not satisfy a diagnosis of LKS, as they showed milder fluctuation in speech. Therefore, we have used the term epilepsy–aphasia spectrum disorder to encompass these cases that fit along the spectrum between BECTS at the mild end, and LKS/CSWS at the severe end (Doose, 1989; Rudolf et al., 2009). GGE occurred in only two relatives. The paucity of GGE is in accordance with the concept that distinctive genetic factors underlie generalized and focal epilepsies.

Genetics of BECTS and CTS

Until recently BECTS has been considered an idiopathic focal epilepsy, implying that hereditary features play a key role (Commission on Classification and Terminology of the International League Against Epilepsy, 1989). However, the genetics of BECTS versus the EEG signature of CTS alone has been intensely debated. Studies have shown variable findings partly dependent on their methodology (Ottman, 1989; Neubauer et al., 1998; Vadlamudi et al., 2004, 2006; Bali et al., 2005; Strug et al., 2009).

The initial hypothesis that BECTS was an autosomal-dominant disorder was based on EEG studies in siblings of children with seizures and temporal sharp waves (Bray & Wiser, 1964). Follow-up of 12 families in which the siblings had temporal sharp waves led the authors to postulate that the EEG trait followed an autosomal dominant inheritance pattern with age-dependent penetrance (Bray & Wiser, 1965). These studies have skewed current concepts of the inheritance of BECTS. First, only some of these patients had BECTS, as the study included all cases with temporal sharp waves. The syndrome of BECTS had not yet been recognized (Lombroso, 1967). Second, their theory related to the EEG abnormality and not the seizure disorder. Third, their focus on the 12 of 40 positive families led to their theory of autosomal-dominant inheritance being inappropriately applied to the clinical syndrome of BECTS overall. Subsequent studies have added support for an autosomal-dominant inheritance pattern of CTS (Heijbel et al., 1975; Degen & Degen, 1990; Bali et al., 2007). Our study was not intended to address the inheritance of CTS; this would ideally require EEG studies in childhood and adolescence in all family members. This approach is not practical with family studies, given the inability to study older generations retrospectively for the age-dependent expression of CTS.

Further evidence against a major genetic influence in BECTS can be drawn from twin studies. Interrogation of twin populations, where bias toward familial cases is absent, has failed to find a convincing genetic basis for BECTS; 18 monozygotic twin pairs showed absolutely no concordance for BECTS (Vadlamudi et al., 2004, 2006). Although we argue that probable complex inheritance underlies BECTS, the effect of acquired factors in contributing to the seizure disorder has not been assessed. These observations led to the recent reclassification of BECTS as an epilepsy syndrome of unknown etiology rather than the older system in which it was referred to as idiopathic (Berg et al., 2010).

Several groups have attempted to identify genes predisposing to BECTS. Neubauer et al. (1998) mapped 22 small families to chromosome 15q14 in 1998 using the presence of CTS as well as BECTS; however, no mutation has been found. Recent studies by a U.S. group also studied CTS and, despite promising logarithm of odds (LOD) scores, no genetic mutations have been identified (Strug et al., 2009; Pal et al., 2010). Other studies of larger monogenic families have focused on more complex phenotypes with RS and additional features (including paroxysmal exercise-induced dystonia, writer’s cramp, speech dyspraxia, and speech and sound disorder), although gene identification has been unsuccessful to date (Scheffer et al., 1995; Guerrini et al., 1999; Kugler et al., 2008).

The phenotypic pattern of seizure disorders in our relatively unselected cohort of families of BECTS probands is complicated and does not fit into a well-recognized pattern. Families show phenotypic heterogeneity, with the most common phenotype being FS, presumably reflecting a genetic basis for a reduced seizure threshold. The next most common phenotype is that of focal seizures, specifically BECTS and epileptic–aphasia spectrum disorder. This has been described in other families where overlap between epileptic encephalopathy with CSWS, atypical benign partial epilepsy, and BECTS has been observed (Neubauer et al., 1998; De Tiege et al., 2006). These observations have led to the concept of a spectrum of epilepsy–aphasia disorders with BECTS at the mild end, the broad less well-defined group of epilepsy–aphasia children in the middle, and classical LKS and the syndrome of CSWS at the severe end (Cole et al. 1988; Hirsch et al., 1990; Deonna et al., 1993). The finding of overlap in these families supports a genetic basis for this spectrum. What factors determine whether one family member has BECTS and another has epileptic–aphasia spectrum disorder is unclear; it is likely to be due to additional genetic determinants, but environmental factors may also be implicated.

Acknowledgment

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. Disclosure
  8. References

The study was supported by National Health and Medical Research Council, Australia.

Disclosure

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. Disclosure
  8. References

None of the authors has any conflicts of interest. We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines.

References

  1. Top of page
  2. Summary
  3. Methods
  4. Results
  5. Discussion
  6. Acknowledgment
  7. Disclosure
  8. References