Idiopathic Generalized Epilepsies: Do Sporadic and Familial Cases Differ?

Authors

  • Regula S. Briellmann,

    1. Epilepsy Research Institute and Department of Medicine (Neurology), University of Melbourne, Austin and Repatriation Medical Centre, Melbourne, Australia
    Search for more papers by this author
  • Yvonne Torn-Broers,

    1. Epilepsy Research Institute and Department of Medicine (Neurology), University of Melbourne, Austin and Repatriation Medical Centre, Melbourne, Australia
    Search for more papers by this author
  • Samuel F. Berkovic

    1. Epilepsy Research Institute and Department of Medicine (Neurology), University of Melbourne, Austin and Repatriation Medical Centre, Melbourne, Australia
    Search for more papers by this author

Address correspondence and reprint requests to Dr. S. F. Berkovic at Epilepsy Research Institute, Neurosciences Building, Repatriation Campus, Heidelberg West, Victoria 3081, Australia. E-mail: s.berkovic@unimelb.edu.au

Abstract

Summary:  Purpose: Genetic factors are the only identified cause of idiopathic generalized epilepsies (IGEs), but the majority of cases do not have affected first-degree relatives. Here we investigate whether subjects with sporadic and familial IGE differ in terms of antecedent events and clinical and EEG features. Differences would support the hypothesis of a different etiology for sporadic cases, which has implications for choice of subjects for genetic association studies.

Methods: We analyzed 98 patients with IGE, diagnosed on clinical and EEG criteria. All patients and, if possible, one relative were interviewed, with special emphasis on potential antecedent events and family history. Patients with first-degree relatives affected with epileptic seizures were regarded as “familial,” and the other patients were regarded as “sporadic.”

Results: Of the 98 IGE patients, 32 (33%) patients were familial. The risk for seizures was 13.2% for siblings, and 7.7% for parents. The distribution of the IGE subsyndromes, the presence of antecedent events, and other electroclinical features did not differ between familial and sporadic IGE groups.

Conclusions: No differences were found between familial and sporadic IGE patients. This does not the support the hypothesis that sporadic and familial IGE cases have separate etiologies.

The idiopathic generalized epilepsies (IGEs) are a group of partly overlapping epilepsy syndromes (1,2). The age at onset and distribution of seizure types define the well-known subsyndromes of childhood absence epilepsy (CAE), juvenile absence epilepsy (JAE), and juvenile myoclonic epilepsy (JME) (1). A residuum of cases have generalized tonic–clonic seizures (GTCSs) alone and are referred to here as GTCS–IGE.

Genetic factors are the only identified cause of IGE, yet the majority of cases appear sporadic, with no immediate family history. This is consistent with complex inheritance believed to underlie IGE (3). Genetic research on diseases with complex inheritance is difficult because of factors including genetic heterogeneity, low penetrance, and existence of phenocopies. Phenocopies mimic the genetic disorder, but have another etiology.

Complex inheritance may comprise oligogenic or polygenic transmission, in which the disorder is due to a few or multiple genes acting together (3). Alternatively, it can be due to a multifactorial etiology in which, in addition to a major genetic component, acquired factors such as brain injury may contribute (3,4). The possibility that acquired factors contribute to IGE is supported by the observation that monozygotic twins show <100% concordance for IGE (5,6).

We therefore sought to determine if there were clinical differences between sporadic and familial IGE, which could support a separate etiology.

METHODS

Subjects

One-hundred five patients with IGE were studied (mean age at examination, 27 years). Possible candidates were identified from all patients seen at our center between 1990 and 1998, based on an EEG database, and on clinical files suggesting the presence of IGE. Our center is part of a large metropolitan public hospital. The candidate population included patients with newly diagnosed epilepsy and some with long-standing disease. None was referred because of our group's interest in genetics. Patients were included based on the following criteria: a clinical history of at least two lifetime seizures of the following types: myoclonus, absences, or GTCS; at least one EEG showing generalized epileptiform discharges of 3 Hz or faster; onset of seizures between ages 2 and 20 years; and absence of gross neurologic abnormalities. Seven of the 105 patients were excluded from further analysis because they had atypical IGE (two patients), intermixed partial seizures (one patient), or did not allow analysis of family history (four patients). The International League Against Epilepsy (ILAE) Classification of Epilepsy syndromes was used to determine the subsyndromes of CAE, JAE, and JME (1). Where adolescent patients had both myoclonus and absences, the predominant seizure type was used to classify the case, rather than adding an additional syndrome of “juvenile myoclonic and absence epilepsy”(7).

One investigator (R.S.B.) performed personal interviews with the patient and if possible a relative, usually the mother, to establish the family history and potential antecedent events. All available medical records, including childhood records, were acquired and scrutinized.

Potential antecedent events

The following were regarded significant antecedent events: severe perinatal events, such as severe asphyxia, severe head injury, viral encephalitis, bacterial meningitis or cerebral abscess, and complicated febrile seizures. Febrile convulsions were diagnosed when convulsions associated with fever occurred between the age of 6 months and 6 years. They were regarded as complicated when their duration was of >15 min, when there were focal signs, or when there were two or more febrile convulsions in 24 h.

Family history

Patients were split into the two groups, “sporadic” and “familial,” based on presence or absence of first-degree relatives (parents, siblings, or offspring) with epileptic seizures. Epileptic seizures in second- to third-degree relatives were noted separately. Epileptic seizures in relatives were divided into (a) epilepsy, when there was evidence of two or more unprovoked epileptic seizures in lifetime; if possible, the epilepsy syndrome was diagnosed; (b) febrile convulsions; and (c) single or provoked afebrile seizures. For each patient, the number and gender of affected and unaffected siblings, parents, aunts/uncles, and grandparents were counted.

Statistics

The clinical features of sporadic and familial IGE were compared by using χ2 tests and t tests. Level of significance was set at 5%.

The study was approved by the hospital Human Research Ethics Committee.

RESULTS

The 98 IGE subjects had a mean age at examination of 27 (±10) years. There were 41 males and 57 female subjects. Eighteen (18%) patients had CAE, 28 (29%) patients had JAE, 30 (31%) patients had JME, and 22 (22%) patients had GTCS–IGE.

Affected first-degree relatives in IGE patients

Of the 98 patients, 32 (33%) had affected first-degree relatives, and this group was defined as familial IGE. Two patients had one affected child, 22 had one to three affected siblings, and 15 had one affected parent. Twelve of the 32 patients had more than one affected first-degree relative; the maximal number of affected first-degree relatives was four.

Of the siblings of all 98 IGE patients, 13.2% were affected with epileptic seizures, whereas 7.7% of parents had a history of seizures (Table 1). There was no significant effect of gender in affected siblings or parents, although 60% of the affected parents were mothers. The diagnosis in affected siblings or parents was in the majority epilepsy, with the syndrome in most cases being consistent with IGE. In nine siblings febrile convulsions were present; three of them also had afebrile seizures. Provoked seizures occurred in a third of the affected siblings.

Table 1.  History of seizures in siblings of IGE patients
  %
  1. Three of the 14 siblings with epilepsy had febrile convulsions as well.

  2. IGE, idiopathic generalized epilepsy.

Total siblings235 
Affected siblings3113.2
 Male208.5
 Female114.7
Split for diagnosis  
 Epilepsy146.0
  IGE83.4
  Possible IGE41.7
  Non-IGE20.9
 Febrile convulsions only62.5
 Provoked seizures114.7

Clinical features in sporadic and familial IGE

The distribution of the subsyndromes was not different between these two groups (Table 2). The number of patients affected by one, two, or three different seizure types also was similar in both groups. There also was no difference in any of the other electroclinical features, such as a history of febrile convulsions, significant and minor antecedent events, response to treatment, or presence of photosensitivity.

Table 2.  Clinical features of idiopathic generalized epilepsy
 Sporadic IGEFamilial IGE
  1. No difference between “sporadic” and “familial” IGE (χ2 and t tests).

  2. AED, antiepileptic drugs; CAE, childhood absence epilepsy; JAE, juvenile absence epilepsy; JME, juvenile myoclonic epilepsy; GTCS, generalized tonic–clonic seizure; sig, significant.

Number6632
Gender male30 (45%)11 (34%)
Age at examination (yr)27 (±11)27 (±10)
Age at onset (yr)13 (±4)12 (±5)
Diagnosis  
 CAE11 (17%)7 (22%)
 JAE20 (30%)8 (25%)
 JME23 (35%)7 (22%)
 GTCS–IGE12 (18%)10 (31%)
Number of different seizure types  
 One sz type16 (24%)12 (39%)
 Two sz types40 (60%)14 (45%)
 Three sz types11 (16%)5 (16%)
Antiepileptic treatment  
 No treatment10 (15%)3 (9%)
 Seizure free with AED34 (52%)13 (41%)
Photosensitivity14 (21%)7 (22%)
History  
 Febrile convulsions9 (13%)2 (7%)
 Sig antecedent events14 (21%)6 (19%)

Second- to third-degree relatives in sporadic and familial IGE patients

A family history in second- to third-degree relatives was noted in 35% of the sporadic IGE and in 47% of the familial IGE (Table 3). The frequency of a history of seizures was about double for uncles/aunts and grandparents of familial IGE compared with that of sporadic IGE patients. The difference between sporadic and familial cases did not reach statistical significance.

Table 3.  Second- to third-degree relatives in sporadic and familial IGE
 Sporadic IGE
(n = 66)
Familial IGE
(n = 32)
n%n%
  1. Risk to distant relatives was not different between sporadic and familial IGE.

Family history    
 Any FH in distant relatives23351547
 More than 1 affected1218825
Risk to distant relatives    
 Uncles and aunts    
  Total460 186 
  Affected112.484.3
 Grandparents    
  Number260 124 
  Affected62.375.6

DISCUSSION

Research on genetic disorders and identification of candidate genes relies on the correct classification of patients. Attempts to find genes by association studies depend on the assumption that the collection of unrelated cases is reasonably homogeneous, or at worst contaminated by only a small percentage of cases with alternate causes. It is known that certain symptomatic generalized epilepsies (e.g., ceroid lipofuscinoses) can mimic IGE at least for a period in their evolution, but such cases are believed to be rare (8). It is not known if the common sporadic cases of IGE are due in whole or in part to nongenetic etiologies.

Our data showed no differences between sporadic IGE, defined as absence of first-degree relatives affected with epileptic seizures, and familial IGE in clinical features and presence of identifiable antecedents. A post hoc calculation was performed to assess the power of the study (9). Based on Table 2, 10 degrees of freedom (independent variables) were used. The effect size was 0.4 for a sample size of 98, and with an alpha of 0.05, the power was 0.8. Therefore our sample size provided enough power to test for a major difference between sporadic and familial IGE. The absence of demonstrable clinical differences between sporadic and familial IGE does not support the hypothesis that they have different etiologies. For the single patient, it is possible that similar clinical manifestations may be caused by different etiologies (genetic and nongenetic). Overall, however, there was no indication of a difference in the etiology between the two groups.

These findings are not surprising, as sporadic cases are expected in disorders with complex inheritance. The proportion of affected first-degree relatives is much lower than that in conditions with simple mendelian inheritance. As complex disorders are due to the effects of more than one (and perhaps many) genes, with or without environmental effects, the affected individual may be the only one in the immediate family with the combination of factors needed for expression of the phenotype. Additionally, the potential role of reduced penetrance, the possibility of new mutations, and the question of postfertilization genetic changes (epigenetic effects) could all result in the observation of sporadic cases (3). Genetic heterogeneity is likely to be present in IGE, with different genes causing similar clinical phenotypes. Such heterogeneity is likely to be represented in both familial and sporadic IGE cases, as defined here. In our study, we defined familial as the presence of any febrile or afebrile epileptic seizures in first-degree relatives as a pragmatic distinction. It could be argued that single unprovoked and provoked seizures not be included in the analysis, as such attacks may have a different etiology from that of the familial seizure disorder. However, it is likely that seizures in relatives are underreported, particularly in parents, where the early childhood history is not easily accessible, and we chose to include such seizures as putative markers of familial seizure susceptibility in this population.

The frequency of potential antecedent factors was the same in familial and sporadic cases. This suggests that environmental factors have no increased role in the etiology of sporadic cases and favors an oligogenic or polygenic model over a multifactorial one. These findings suggest that sporadic cases should be included in molecular genetic studies of IGE, where appropriate to the experimental design.

Acknowledgments: This research was supported by a grant from the National Health and Medical Research Council of Australia. R.S.B. was supported by the Brain Imaging Research Foundation, Australia.

We thank the following doctors for providing us with their patients: Geoff A. Donnan, Mark A. King, Carla Marini, Mark R. Newton, David C. Reutens, Ingrid E. Scheffer, and the A&RMC First Seizure Clinic.

Ancillary