Classification of Epilepsy Syndromes and Role of Genetic Factors Roula N. Choueiri, MD, Michel N. Fayad, MD, Antoine Farah, MD, and Mohamad A. Mikati, MD In this report the types of epilepsy syndromes seen in children in a tertiary referral center in Beirut, Lebanon were studied and the importance of consanguinity and family history in the occurrence of these syndromes was investigated. Records of 230 pediatric patients evaluated during a 1-year period with the diagnosis of single seizure, febrile seizure, or epilepsy were reviewed. Each patient was classified according to the International League Against Epilepsy classification. The occurrence of consanguinity, of family history of febrile seizures or epilepsy, and of other variables was noted. Thirty-six percent of patients were diagnosed with localization-related epilepsy, 21.7% with generalized epilepsy, 11.7% with undetermined generalized or focal, and 24.3% with special syndromes. Twelve percent of patients were diagnosed with idiopathic, 15.1% with symptomatic, and 30.3% with cryptogenic epilepsies. Consanguinity was more common in patients with symptomatic and cryptogenic epilepsies than in patients with idiopathic epilepsies or with incidental seizures (P < 0.05). Family history of epilepsy was more common in patients with symptomatic, cryptogenic, and idiopathic epilepsies than in patients with incidental seizures (P < 0.05). Family history of febrile seizures but not consanguinity was more common in patients with febrile seizures (P < 0.05). We conclude that genetic factors are important not only in idiopathic epilepsies and febrile seizures but also in cryptogenic and symptomatic epilepsies. © 2001 by Elsevier Science Inc. All rights reserved. Choueiri RN, Fayad MN, Farah A, Mikati MA. Classification of epilepsy syndromes and role of genetic factors. Pediatr Neurol 2001;24:37-43. From the Department of Pediatrics and the Adult and Pediatric Epilepsy Program; American University of Beirut, Beirut, Lebanon. © 2001 by Elsevier Science Inc. All rights reserved. PII S0887-8994(00)00231-9 ● 0887-8994/01/$—see front matter Introduction Clinical studies have increased our understanding of epilepsy syndromes and of their courses. More recently these studies have become even more important given the recognition of numerous familial and genetic forms of epilepsy and given the increasing use of the techniques of modern genetics in the investigation of epilepsy. The International Classification of Epileptic Seizures was proposed in 1969 and revised in 1981 by the International League Against Epilepsy (ILAE). The ILAE also proposed a classification of epilepsies and of epileptic syndromes in 1985, which was later revised in 1989 [1]. This classification, which grouped epilepsies as syndromes based on clinical descriptions and electroencephalograms (EEGs), is more precise for diagnosis and prognosis than previous classifications [2]. The execution of studies applying appropriate statistical methods to the problem of epilepsy is desirable and needed despite obvious difficulties [3]. Various studies have been conducted in different countries based on this classification [4-9]. Determining the incidence of epilepsy syndromes in different populations, comparing the results with those of other populations, and investigating the importance of genetic and environmental factors in those syndromes can provide further insights into the problem of epilepsy. This result is particularly true in view of the ongoing revision of the current ILAE epilepsy syndrome classification. The goals of this study were as follows: (1) to determine the types of epilepsy syndromes, and the usefulness of the 1989 ILAE classification, in patients seen in our tertiary care center and (2) to investigate the importance of specific clinical, consanguinity, and family history factors in the occurrence of these disorders in our patient population. Communications should be addressed to: Dr. Mikati; Professor and Chairman, Department of Pediatrics; Director, Adult and Pediatric Epilepsy Program; American University of Beirut; 850 Third Avenue, 18th floor, New York, NY 10022. Received March 8, 2000; accepted August 30, 2000. Choueiri et al: Genetic Factors in Pediatric Epilepsy 37 Materials and Methods This study was cross-sectional. The records of all children (0-13 years of age) seen with the diagnosis of seizures or of epilepsy during a 1-year period in the Department of Pediatrics at the American University of Beirut Medical Center were reviewed. Initial and follow-up evaluations of inpatients (including the neonatal care units) and outpatients were included. The following variables were determined: age at onset of first seizure; duration of time between onset and presentation; sex; clinical presentation; etiology; description of the seizures; precipitating factors (if any); family history (in first-degree relatives) of seizures, of febrile seizures, or of epilepsy; consanguinity; extent of seizure control; and developmental status at the time of the latest follow-up during the above year. Computed tomography (CT) and/or magnetic resonance imaging (MRI) results and the results of any metabolic tests or of other investigations were also documented. All available EEG studies, usually performed on an 18-channel machine in our hospital, were interpreted by one of the investigators. In 15 patients was the actual EEG performed elsewhere, and only the report was available. The decision to perform brain imaging or metabolic tests on some patients depended on the physician’s judgment at the time of the evaluation. For the purposes of the study the following definitions were used: (1) incidental seizures: single seizures or seizures occurring only in the context of central nervous system (CNS) infection, head trauma, electrolyte imbalance, or similar precipitating factors other than fever; (2) febrile seizures: seizures occurring in children from 3 months to 5 years of age, with a temperature of 38°C or more and in the absence of CNS infection and of other significant precipitating events; (3) epilepsy: a condition in which there is a demonstrated tendency for recurrent unprovoked seizures as evidenced by the occurrence of two or more unprovoked seizures. Consanguinity was empirically defined to be present if the parents were either first- or second-degree cousins. Clinical, EEG, and other relevant data were used to confirm the diagnosis of the seizure disorder or of the epilepsy syndrome and then to classify each patient’s seizure types and epilepsy syndrome according to the ILAE classification. Each patient was reviewed by at least two of the investigators who had to agree, independently, on the classification. If there was disagreement (which was unusual), then the data were discussed, and, if necessary, further information or investigations were obtained to classify the case. If the patient had a history of more than one seizure type or epilepsy syndrome (e.g., history of febrile seizures in a patient with subsequent temporal lobe seizures), then the most recent active seizure type(s) and the most recent data available on the patient were used for classification, and the history of a prior seizure type was noted. In addition to first classifying our patients according to the ILAE criteria, we found it helpful, based on our experience and on that of previous investigators [4-6], to subsequently add the following subcategories to the ILAE classification. These subcategories (category 1.2 g, subcategories 1.3 a and 1.3 b, and subcategory 2.2 e) were added, for the purposes of our study, to better describe many of our patients and to better classify several other patients (categories 1.2 g, 1.3 b, and 2.2e) that were not classifiable using the ILAE criteria alone (see below and Table 1). 1. To the category of ILAE category 1.2, the following subdivision was added: *Category 1.2 g: “Symptomatic Localization Related Epilepsy, lobe not defined.” Patients were classified in this category whenever they clearly had a localization-related epilepsy syndrome, but the EEG and clinical information available did not allow for definitive localization of the epileptogenic focus. 2. The “Cryptogenic Localization Related Epilepsies” (ILAE category 1.3) were subdivided into two subcategories: *Subcategory 1.3a “Normal Development Cryptogenic Localization Related Epilepsy” and *Subcategory 1.3b: “Abnormal Development Cryptogenic Localization Related Epilepsy.” In these patients, development was documented by the Denver Developmental Scale or by similar measures, and all tests failed to reveal the cause of the epilepsy. These patients did not fit in the Symptomatic Epilepsies (category 38 PEDIATRIC NEUROLOGY Vol. 24 No. 1 1.1.2) because the patients classified under symptomatic epilepsies should have a known etiology. 3. To the category of “Generalized Epilepsy, Cryptogenic or Symptomatic” (ILAE category 2.2), the following subcategory was added: *Subcategory 2.2e: “Cryptogenic not defined above.” This addition was necessary because some patients had generalized cryptogenic epilepsy clinically and by EEG, but their clinical picture did not completely fit in any of the specific cryptogenic syndromes described in the ILAE classification (subcategories 2.2 a-d and 2.3.1 a-b). The presence of consanguinity and family history of seizures, febrile seizures, or of epilepsy were compared among the three groups of incidental seizures, febrile seizures, and epilepsy using the chi-square and the Fisher Exact Probability tests [3]. The occurrence of these factors in patients with various epilepsy syndromes was also investigated using the same statistical methods. Results Patient Population A total of 230 patients (122 males and 108 females) were seen during the study period because of seizures or epilepsy (Table 2). There were 31 children with incidental seizures, 25 with febrile seizures, and 174 with epilepsy. Primarily generalized tonic-clonic seizures occurred in 32.6% of the patients, complex partial seizures in 30.5%, secondary generalized seizures in 20.5%, and simple partial seizures in 6.5%. Many patients (18.7%) had having more than one seizure type. CT or MRI studies of the brain were performed on 142 patients. Eighty-nine patients had normal brain imaging and 53 patients had abnormalities. The most common abnormal finding was “atrophy,” which was noted in 20 patients, followed by brain malformations seen in eight. An underlying etiology was identified in 107 patients (Table 2). Normal development was noted in 29/31, 24/25, and 106/174 patients with incidental seizures, febrile seizures, and epilepsy, respectively. Classification of Epilepsy Syndromes Using the ILAE classification, we were able to classify 205/230 (89%) of our patients (Tables 1 and 3). By adding the four subcategories defined in the Materials and Methods section, 11 additional patients (4.7%) could also be classified. Fourteen (6%) patients were not classifiable because of lack of data: EEGs were not available in 13, and one patient had an unwitnessed seizure and was subsequently lost to follow-up. The most common epilepsy syndromes we found were the “localization-related epilepsy syndromes” (36% of our patients), and in this category, “cryptogenic localizationrelated epilepsy syndromes” were the most common of those patients. Generalized epilepsy syndromes were 21.7%, and those “undetermined if generalized or localization-related” were 11.7%. Special syndromes (febrile seizures, isolated seizures, and seizures associated with Table 1. Classification of patients according to type of epilepsy syndrome Epileptic syndrome 1. Localization-related 1.1 Idiopathic a) Benign childhood epilepsy with centrotemporal spikes b) Childhood epilepsy with occipital paroxysms c) Primary reading epilepsy 1.2 Symptomatic a) Chronic progressive epilepsia partialis continua of childhood b) Seizures with specific modes of precipitation c) Temporal lobe d) Frontal lobe e) Parietal lobe f) Occipital lobe g) Not defined* 1.3 Cryptogenic a) Normal development* b) Abnormal development* 2. Generalized epilepsies 2.1 Idiopathic a) Benign neonatal familial convulsions b) Benign neonatal convulsions c) Benign myoclonic epilepsy of infancy d) Childhood absence e) Juvenile absence f) Juvenile myoclonic epilepsy g) Grand mal on awakening h) Other generalized not defined i) Seizures precipitated by specific modes 2.2 Cryptogenic or symptomatic a) West syndrome b) Lennox-Gastaut syndrome c) Epilepsy with myoclonic-astatic seizures d) Epilepsy with myoclonic absence e) Cryptogenic not defined above* 2.3 Symptomatic 2.3.1 Nonspecific etiology a) Early myoclonic encephalopathy b) Early infantile epileptic encephalopathy with suppression burst c) Other symptomatic generalized not defined 2.3.2 Specific syndromes Table 2. Etiologies according to patient group N 5 230 % Etiology 83 6 6 36 2.6 20 8.6 Intraventricular hemorrhage Neurocutaneous Metabolic Degenerative diseases Brain tumor Trauma Stroke Fever Congenital infections Asphyxia Meningitis Encephalitis CNS malformation Other Unknown Total 1 5 4 4 6 57 35 22 50 22 2 24.7 21.7 9.5 Incidental* n 5 31 Febrile n 5 25 Epilepsy* n 5 174 Total n 5 230 0 0 2 2 0 3 0 0 0 0 2 11 1 2 14 1 0 0 3 0 0 2 2 8 0 0 13 31 0 0 0 25 0 0 0 0 0 0 0 25 1 5 1 0 5 15 5 5 8 3 110 174 1 5 4 25 5 17 7 13 8 3 123 230 * As explained and defined in the text, patients with incidental seizures are those classified under section 4a and 4b in the ILAE classification (Table 2), and patients with epilepsy are patients that fit in ILAE categories 1, 2, and 3. These definitions also apply to the other tables. 2 3 1 14 13 2 6 5 15 11 5.6 6.5 1 10 4 3. Undetermined if generalized or focal 3.1 With both generalized and focal seizures a) Neonatal seizures b) Severe myoclonic epilepsy in infancy c) Epilepsy with continuous spike-waves during slow-wave sleep d) Acquired epileptic aphasia (Landau-Kleffner syndrome) e) Other undetermined epilepsies not defined above 3.2 Without unequivocal generalized or focal seizures 27 18 11 1 11.7 7.8 5 9 3.9 4. Special syndromes a) Febrile seizures b) Isolated seizures or status epilepticus c) Seizures associated with metabolic or toxic events Not classifiable 56 25 28 3 24.3 10.9 12.2 1.3 14 6.1 1 * Subdivisions added by us, for the purposes of our study, to the ILAE classification (see text for details). metabolic or toxic events) constituted 24.3% of our study population. Of 174 patients who had epilepsy, 32 (18%) had a personal history of a prior febrile seizure(s). The details of the distribution of the different types of epilepsy syndromes are included in Table 1. Table 3 demonstrates that a positive personal history of febrile seizures occurred more commonly in patients with temporal lobe epilepsy than in those with primary generalized epilepsy, and more commonly in the latter group than in patients with incidental seizures (P , 0.05). Role of Genetic Factors Consanguinity. Patients with epilepsy had a higher consanguinity rate than those with incidental seizures, who themselves did not differ from those with febrile seizures (19.5%, 0%, and 4%, respectively; Table 4). Patients with cryptogenic and symptomatic epilepsies had higher consanguinity rates than those with idiopathic epilepsies or with incidental seizures (24.3%, 22.9%, 3.4%, and 0%, respectively; Table 5). Family History of Febrile Seizures. Patients with febrile seizures were more likely to have a family history of febrile seizures (20%) than patients with epilepsy (1.1%; P , 0.0001) and, possibly, more than those with incidental seizures (3.2%, the difference approached, but did not achieve, statistical significance, P 5 0.079; Table 4). Family History of Epilepsy. Patients with epilepsy were more likely to have a family history of epilepsy than Choueiri et al: Genetic Factors in Pediatric Epilepsy 39 Table 3. Personal history of febrile seizures in patients with incidental seizures, primary generalized epilepsy, and temporal lobe epilepsy Personal history of fibrile seizures present Total Paired Comparison (P) INC* PIGE* TLE*† 0 (0%) 3 (6%) 5 (35.7%) 31 50 14 INC vs PIGE and INC vs TLE both P , 0.05 PIGE vs TLE 5 0.0097 * Patients included in this table were older than 3 months of age. † These 14 patients included the five patients with 1.2c (symptomatic localization-related) and nine patients with either 1.3a or 1.3b (cryptogenic temporal lobe epilepsy). Abbreviations: INC 5 Incidental seizures PIGE 5 Primary idiopathic generalized epilepsy TLE 5 Temporal lobe epilepsy, includes patients from categories 1.2c and 1.3 in Table 2 patients with incidental seizures, who themselves did not differ from those with febrile seizures (20.1%, 3.2%, and 8% respectively; Table 4). Family history of epilepsy was equally present in symptomatic, cryptogenic and idiopathic epilepsies and was more commonly found in those epilepsies than in patients with incidental seizures (28.6%, 21.4%, 25.0%, and 3.1% respectively; Table 5). Concordance. In families with multiple epileptic patients, we investigated the concordance in epilepsy syndrome types in different patients within the same family. In six families, at least two members of the family could be fully studied by one of the investigators to definitively determine their specific epilepsy syndromes. We found that in three of the six families, there was a lack of concordance in the epilepsy syndromes occurring in different family members. Table 4. groups Discussion ILAE Classification We found that the ILAE system was useful in classifying our patient population. We were able to classify most, but not all, of our patients by using that system alone. A difficulty we encountered in using the ILAE classification was that many of our patients clearly had cryptogenic epilepsy but could not fit in one of the clinically described cryptogenic epilepsies. These patients had developmental delay and multiple seizure types but had not yet developed all the clinical criteria of the known cryptogenic epilepsies, such as Lennox-Gastaut syndrome. Neuroradiologic examinations often demonstrated nonspecific findings such as atrophy, which did not help much in the classification of the epilepsy syndrome. However, at other times Consanguinity and family history in the incidental seizures, febrile seizures, and epilepsy INC FEB EP Consanguinity present 0 (0%) 1 (4%) 34 (19.5%) Family history of febrile seizures present 1 (3.2%) 5 (20%) 2 (1.1%) Family history of epilepsy present 1 (3.2%) 2 (8%) 35 (20.1%) 31* 25 174 Total Paired Comparisons (P value) INC vs. EP 5 0.007 FEB vs. EP 5 0.104 INC vs. FEB 5 0.446 INC vs. FEB 5 0.079 FEB vs. EP 5 0.000 INC vs. EP 5 0.940 INC vs. EP 5 0.043 INC vs. FEB 5 0.848 FEB vs. EP 5 0.238 * Patients included are older than 3 months of age, and the patients with febrile seizures were all older than 3 months of age. Abbreviations: EP 5 Epilepsy FEB 5 Febrile seizures INC 5 Incidental seizures 40 PEDIATRIC NEUROLOGY Vol. 24 No. 1 Table 5. Consanguinity and family history of epilepsy in the incidental seizure group and in different epilepsy syndromes INC CRY IDI SYM Consanguinity present 0 (0%) 17 (24.3%) 1 (3.6%) 8 (22.9%) Family history of epilepsy present 1 (3.1%) 15 (21.4%) 7 (25%) 10 (28.6%) 31 70 28 35 Total Paired Comparison (P values*) CRY vs. INC 5 0.0026 CRY vs. IDI 5 0.0100 SYM vs. IDI 5 0.0273 SYM vs. INC 5 0.0041 SYM vs. INC 5 0.0057 CRY vs. INC 5 0.0209 IDI vs. INC 5 0.0206 * All other comparisons were not significant (P . 0.05) Abbreviations: CRY 5 Cryptogenic IDI 5 Idiopathic INC 5 Incidental seizures SYM 5 Symptomatic the changes were specific and quite helpful (such as the presence of medial temporal sclerosis on MRI). The other difficulty was that the ILAE classification was easier to apply to patients who were more extensively studied. It was more difficult to use for ambulatory patients with less extensive investigations. For example, most patients in the category “localization-related epilepsy, lobe not defined” that we added could probably have been classified under the appropriate lobe if video EEG monitoring had been used on them. However, because this technique, even when readily available, is not routinely used on all patients, the exact lobe may not necessarily be identified. It will be helpful to have, as part of the ILAE classification and in addition to its current electroclinical system, a classification system that is exclusively based on the clinical description of the seizures. This recommendation is supported not only by our experience but also by the experience of several other investigators [5-9]. Types of Epilepsy Syndromes We found that the most common epilepsies in our population were localization-related epilepsies. In both localization-related and generalized epilepsies, the symptomatic and cryptogenic syndromes were more frequent than the idiopathic syndromes. These findings are consistent with previous reports that have generally demonstrated higher frequencies of localization-related and symptomatic syndromes in developing countries than in developed countries. In fact, our results are intermediate between those two categories of countries. Our study, like many previous studies, reflects the patient population seen in a tertiary referral center rather than one seen in the general population. However, it does have the advantage of having used both EEG and clinical criteria in classification of the patients, which was not the case in many, but not all, of the previous studies [5-11]. The differences between developing and developed countries have been attributed to the importance of perinatal insults, to CNS infections, and to other environmental factors in the former countries. In addition, in developed countries the lower rates of cryptogenic epilepsies, relative to symptomatic epilepsies, could be the result of the more detailed monitoring and investigations that patients receive in those countries. Role of Genetic Factors We found that genetic factors are important not only in idiopathic epilepsies and febrile seizures, as one may have expected, but also, perhaps even more so, in symptomatic and cryptogenic epilepsies. Parental consanguinity was more common in symptomatic and cryptogenic epilepsies than in patients with idiopathic epilepsies and in patients with incidental seizures. This finding, to our knowledge, is a new one. In a recent study reported by Ottman et al. [12], genetic factors were important in cryptogenic but not in symptomatic epilepsies. However, those authors were studying a different patient population in the United States, which, presumably, also had lower consanguinity rates. Our findings strongly favor the hypothesis that, at least in our patient population, symptomatic epilepsies are not necessarily because of prior exogenous, usually environmental, insults to the brain, but that such insults are more likely to result in epilepsy (cryptogenic or symptomatic) if they occur on a genetic background favoring the development of epilepsy. Some of these genetic factors are probably inherited as autosomal-recessive or multifactorial traits (as evidenced by the increased rates of parental consanguinity in patients with cryptogenic and symptomatic epilepsies) or as dominant traits (as suggested by the increased family history of epilepsy in those patients). Exposure to the same or similar environmental factors by members of the same family is another factor that could Choueiri et al: Genetic Factors in Pediatric Epilepsy 41 have contributed to those findings. We hypothesize that “epilepsy predisposing genes” may increase the risk for the development of symptomatic and cryptogenic epilepsies, and the interaction of those genes with environmental factors may determine further the development of epilepsy and its types. This hypothesis is supported by the following evidence: (1) in many of our patients, there was lack of concordance in the type of epilepsy syndromes manifested in different members of the same family; (2) Berkovic et al. [13] found occasional, although not frequent, lack of concordance in the types of epilepsy in twins; and (3) the Italian League Against Epilepsy found concordance for the same clinical form of epilepsy in only 38.9% of the studied families with $ three cases of idiopathic epilepsy [14]. It is known that several idiopathic epilepsies, such as benign familial neonatal epilepsy and benign epilepsy with centrotemporal spikes, are thought to be transmitted as autosomal-dominant traits. This factor may explain our observations of an increased rate of family history of epilepsy and the lack of an increase in parental consanguinity in idiopathic epilepsies (as compared with the comparison group). Consanguinity contributes to the occurrence of multifactorial and autosomal-recessive traits, whereas dominant traits are more likely to demonstrate a positive family history. The inclusion of various cryptogenic epilepsies in one group and similarly of the various idiopathic epilepsies and of the various symptomatic epilepsies each in a separate independent group is justified by the fact that epilepsy syndromes within each of these groups share common clinical and susceptibility factors despite differences that do occur in their pathophysiology and outcome [12]. We found a significantly increased rate of positive family history of febrile seizures in our febrile seizure patients. However, rates of consanguinity and of family history of epilepsy were not different from those of the comparison group of incidental seizures. This is consistent with the single major-locus model of inheritance with a dominantly inherited susceptibility to febrile seizures proposed previously [15-17]. Our data also demonstrated a high rate of occurrence (18.4%) of prior febrile seizures in children with epilepsy. Sofijanov [17] found a rate of 20.3% and Camfield [18] a rate of 14.9%. The limitations of our study must be emphasized. This study was retrospective, and was thus prone to bias in data collection and to referral bias. In addition, the analyses we performed did not correct for variations that may have existed in family size or in other confounding variables in the different subgroups we studied. We did not make corrections for the multiple statistical comparisons that were performed, and our comparison group (patients with incidental seizures) was not a completely homogenous group itself. Also, the rate of consanguinity found in this group probably represents an underestimate of the real consanguinity rate of our general population because it was based on a retrospective review of data collected at 42 PEDIATRIC NEUROLOGY Vol. 24 No. 1 the time of the clinical encounter. The possibility should also be raised that some of the differences we found in the family history and consanguinity rates between different groups may also be secondary to the methodology we used rather than real differences. However, because the same method was used in all of our groups (incidental seizures, febrile seizures, and epilepsy including idiopathic epilepsy, cryptogenic epilepsy, and symptomatic epilepsy groups) the comparisons we made and the conclusions we derived are justifiable at least as initial hypotheses about the importance of genetic factors in epilepsy. Future studies should investigate these hypotheses prospectively [19]. We conclude that the ILAE is useful in classifying our patient population. Modifications of the current ILAE system are, however, probably needed. The relatively high frequency of localization-related epilepsies suggests that environmental factors are important indicators of epilepsy in our patients. However, genetic factors are also important and contribute not only to the occurrence of febrile seizures and idiopathic epilepsy syndromes, as one may expect, but also to the occurrence of cryptogenic and symptomatic epilepsies. We believe that further prospective investigations into the above clinical and genetic factors are bound to improve our understanding of the biology and of the clinical phenomenology of epilepsy. Acknowledgments The authors thank Dr. Iman Nuwayhed for his helpful comments and review of the manuscript. Supported in part by a grant from Sanofi-Sadco. References [1] Commission on Classification, and Terminology of the International League Against Epilepsy. Proposal for revised classification of epilepsies and epileptic syndromes. Epilepsia 1989;30:389-99. [2] Aicardi J. epileptic syndromes in childhood. Epilepsia 1988; 29(Suppl 3):S1-5. [3] Glantz S. Primer of biostatistics. New York: McGraw Hill, 1981. [4] Viani F, Beghi E, Atzo MG, Gullota MP. Classification of epileptic syndromes. Advantages and limitations for evaluation of childhood epileptic syndromes in clinical practice. Epilepsia 1988;29:440-5. [5] Shah NK, Rajadhyaksha B, Shah VS. Experience with the International League Against Epilepsy: Classifications of epileptic seizures (1981) and epilepsies and epileptic syndromes (1989) in epileptic children in a developing country. Epilepsia 1992;33:1072-7. [6] Oka E, Ishida S, Ohtsuka Y, Ohtahara S. Neuroepidemiological study of childhood epilepsy by application of International Classification of Epilepsies and Epileptic Syndromes (ILAE 1989). Epilepsia 1995;36: 658-61. [7] Eslava-Cobos J, Narino D. Experience with the International League Against Epilepsy. Proposals for classification of epileptic seizures and the epilepsies and epileptic syndromes in a pediatric outpatient epilepsy clinic. Epilepsia 1989;30:112-5. [8] Al Rajeh S, Abomelha A, Awada A, Bademosi O, Ismail H. Epilepsy and other convulsive disorders in Saudi-Arabia: A prospective study of 1000 consecutive cases. Acta Neurol Scand 1990;82:341-5. [9] Loiseau P, Duche B, Loiseau J. Classification of epilepsies and epileptic syndromes in two different samples of patients. Epilepsia 1991;32:303-9. [10] Ohtsuka Y, Ohno S, Oka E, Ohtahara S. Classification of epilepsies and epileptic syndromes of childhood according to the 1989 ILAE classification. J Epilepsy 1993;6:272-6. [11] Kramer U, Nevo Y, Neufeld MY. Epidemiology of epilepsy in childhood: A cohort of 440 consecutive patients. Pediatr Neurol 1998; 18:46-50. [12] Ottman R, Lee JH, Risch N, Hauser WA, Susser M. Clinical indicators of genetic susceptibility to epilepsy. Epilepsia 1996;37:353-61. [13] Berkovic SF, Howell RA, Hay DA, Hopper JL. Epilepsies in twins: Genetics of the major epilepsy syndromes. Ann Neurol 1998;43: 435-45. [14] Italian League Against Epilepsy Genetic Collaborative Group. Concordance of clinical forms of epilepsy in families with several affected members. Epilepsia 1993;34:819-26. [15] Rich SS, Angeres JF, Hauser WA, Anderson VE. Complex segregation analysis of febrile convulsions. Am J Hum Genet 1987;41: 249-57. [16] Rantala H, Uhari M. Risk factors for recurrence of febrile convulsions. Acta Neurol Scand 1994;90:207-10. [17] Sofijanov N, Sadikario A, Dukovski M, Kuturec M. Febrile convulsions and later development of epilepsy. Am J Dis Child 1983: 137. [18] Camfield P, Camfield C, Gordon K, Dooley J. What types of epilepsy are preceded by febrile seizures? A population based study of children. Dev Med Child Neurol 1991;36:887-92. [19] Berg AT, Shinnar S, Levy SR, Testa FM. Newly diagnosed epilepsy in children: Presentation at diagnosis. Epilepsia 1999;40:44552. Choueiri et al: Genetic Factors in Pediatric Epilepsy 43