- Email: [email protected]
Early diagnosis of severe myoclonic epilepsy in infancy
Early Diagnosis of Severe Myoclonic Epilepsy in Infancy Moncef Yakoub, MD, Olivier Dulac, MD, Isabelle Jambaque, MD, Catherine Chiron, MD and Perrine Plouin, MD
Of 329 epileptic patients referred in a six year period with the first seizure occurring in the first year of life, 20 met the following criteria: generalized seizures excluding infantile spasms, myoclonic, tonic or absence seizures, at least one afebrile seizure, normal development prior to the first seizure, normal CT scan, and no etiology. Seventeen of these 20 patients developed the full pattern of severe myoclonic epilepsy in infancy (SMEI). This syndrome was recognizable from the second or third seizure in the first year of life, although epileptiform EEG abnormalities were lacking until the age of I I to over 30 months. Therefore, based on the clinical pattern, the diagnosis of SMEI can be made with quite good reliability by the end of the first year of life. Key words: Severe myoclonic epilepsy in infancy, myoclonic epilepsy, infants. Yakoub M, Dulac 0, Jambaque I, Chiron C, Plouin P. Early diagnosis of severe myoclonic epilepsy in infancy. Brain Dev 1992; 14: 299-303
Generalized epilepsies of early childhood without identifiable brain lesions remain a very confusing field. The first step of nosological breakdown has consisted of the recognition of broad clinico-electroencephalographic entities characterized by the age of occurrence, seizure types, and ictal and interictal electroencephalographic (EEG) patterns. The Lennox-Gastaut syndrome [1,2] and Doose's myoclonic astatic epilepsy of early childhood [3] are major components of this early breakdown, the former being considered symptomatic and the latter idiopathic [4]. However, these entities remain heterogeneous in terms of the age of onset and evolution, and most authors admit that there is overlapping between these syndromes [4, 5] . Recently, the second step focused on more homogeneous groups of patients. Various syndromes were recently defined [6] and are now listed by the International League Against Epilepsy [7]. The so-called "severe myoclonic epilepsy in infancy" (SMEI), first described by
From Service de Neuropediatrie (MY, 00, lJ, CC) et Service d'Explorations Fonctionnelles du Systeme Nerveux Central (PP), Hopital Saint Vincent de Paul, Paris. Received for publication: December 24, 1991. Accepted for publication: June 7, 1992. Correspondence address: Pr. Olivier Dulac, Service de Neuropectiatrie (Pr. Ponsot), Inserm U 29, Hopital Saint Vincent de Paul, 82 avenue Denfert Rochereau, 75674 Paris Cedex 14, France.
Dravet [8] , is one of these recently recognized syndromes. According to several studies [9-12], it begins in the first year of life with clonic seizures that are usually generalized and often associated with fever. Subsequently, clonic seizures recur with a slight or no increase in temperature, and they often involve alternately one side of the body and then the other, with frequent status epilepticus. The myoclonic attacks only occur between 1 and 4 years of life, and absence seizures are rare. Complex partial seizures may also appear later. Psychomotor development is normal during the first year of life, but an ataxic gait and speech delay become evident in the second year. Computerized tomographic (CT) scanning and EEG are normal at the onset, and generalized spike waves appear only in the second year of life: they are isolated or occur in brief bursts with frequent photosensitivity. These patients do not exhibit tonic seizures or slow spike waves on EEG, in contrast with in the Lennox-Gastaut syndrome. The full pattern of SMEI is therefore easily recognizable. But since the first signs seem to lack specificity, several questions related to an early diagnosis remain. What are the most reliable early diagnostic features? At what age does a diagnosis become possible? To what extent, during the first months of the disease, are the clinical, EEG and radiological characteristics of the syndrome shared by other types of epilepsies? Are there borderline forms? What is the relative frequency of this syndrome among epilepsies beginning in the first year of
life? In order to address these questions, we studied an unselected group of patients who had the characteristics of SMEI during the first year of life, and we found that in most of them the syndromic pattern was completed during subsequent years. PATIENTS AND METHODS The study was performed blindly as to their current diagnosis. 1) Patients were selected according to the following criteria: first seizure between 1 and 12 months of life, based on clinical description, generalized clonic or tonicclonic seizure pattern excluding infantile spasms, or tonic seizures, and with or without associated focal motor seizures; at least one afebrile seizure (temperature under 38.5°C); normal psychomotor development prior to the first seizure; no etiology, including a personal history; normal CT scanning before 2 years of age; first referral prior to 2 years of age, between January 1980 and December 1985; and follow-up of over 3 years from the onset. Of the 329 epileptic patients with the first seizure between 1 and 12 months of life who were referred to the neuropediatric department of the Hopital St Vincent de Paul, Paris, between January 1980 and December 1985, 195 were excluded because their psychomotor development was considered to be abnormal before the first seizure, or because there was evidence an underlying brain lesion. Of the 134 remaining patients, 95 were excluded because they exhibited infantile spasms (58 cases), or seizure types were restricted to partial, myoclonic, tonic and absence seizures during the first year of life. Nineteen were excluded because of insufficient data concerning seizure types or because of delayed referral or insufficient follow-up. The 20 remaining patients were therefore selected for this study according to the previously mentioned criteria. None of them was included in our 1982 series [12]. 2) For each of these 20 selected patients, the following features were studied: age at onset, characteristics of the first and following seizures, number of seizures, the triggering factors, interictal EEG pattern, and outcome in terms of seizures, and of neurological and behavioural problems. All patients underwent EEG follow-up from the first year of life. Seventeen of the 20 selected patients exhibited myoclonic fits, and they fulfilled the clinical and EEG characteristics of "SMEI" (Group A). The 3 other patients remained free of myoclonias (Group B). Group A Seventeen patients, 8 boys and 9 girls, exhibited a very
300 Brain & Development, Vol 14, No 5, 1992
homogeneous pattern. Age of onset and characteristics of the first seizure The age of onset ranged from 2 to 9 months (mean, 5 months 5 days); the first seizure was generalized clonic or tonic-clonic in 9 cases and unilateral in 8. Six patients had postictal hemiparesis lasting a few hours. The first seizure lasted 2 minutes to 3 hours (mean, 32 minutes), over half an hour in 7 patients. In 14 cases, the seizure was triggered by fever (8 cases) or vaccination (6 cases). Subsequent seizures The second seizure occurred 8 days to 4 months (mean, 40 days) after the first one, the delay being less than 1 month in 11 cases. Again, it was clonic or tonic-clonic, generalized (10 cases) or unilateral (7 cases). In all patients but one the first 2 seizures involved both sides of the body: they were either generalized in at least one instance (13 cases) or hemiclonic involving both sides alternately (3 cases). In only one of the 17 patients, the first 2 seizures were unilateral and involved the same half of the body; in this patient, the third seizure was generalized. All 17 patients exhibited therefore at least 1 generalized clonic seizure among the first 3 seizures. The mean time from the second to the third seizure was 30 days, the time between seizures decreasing for the subsequent fits. Seizures lasting over half an hour occurred 2 to 3 times in 14 patients during the first year of life, and 1 to 2 times in 10 of them during the second year. During the first 2 years of life, postictal hemiplegia lasting less than 3 hours occurred in 11 patients, mainly following unilateral seizures. The mean number of seizures was 9 in the first year and 17 in the second year. By the end of the first year, 13 patients had exhibited unilateral clonic seizures alternately in both halves of the body; this was the case for another 2 patients in the second year of life. Myoclonic seizures They occurred at a mean age of 2 years (range, 9 to 48 months). They were mild, occasional and erratic in 8 cases. They were generalized and frequent in 9 cases, producing drop attacks in 7 of them. Some patients exhibited both types. The myoclonic jerks occurred mainly in the morning, as brief bursts lasting 1 to 2 seconds, or during drowsiness. In eight patients, a clear increase in their frequency often preceded generalized seizures. Three patients had myoclonic status epilepticus lasting, respectively, 3, 24 and 36 hours, at the ages of 14 months, and 4 and 5 years. Other types of seizures Fifteen patients exhibited other types of seizures: - partial clonic seizures involving a single limb in 5
cases (onset between 9 and 39 months; mean, 26) and both upperlimbs alternately inl case (onset at 18 months); - versive seizures in 9 cases (onset between 4 months and 5 years; mean, 2 years 6 months); - complex partial seizures lasting several minutes in all 15 patients, with onset between 10 months and 3 years of age; mean, 2 years; these seizures comprised automatisms in 2 cases (mouthing and swallowing) and vegetative manifestations in 3 others (mydriasis, sweating, cyanosis or erythrosis); - short attacks of isolated loss of consciousness resembling absence seizures were reported by the parents in 2 cases, with onset at 13 to 29 months, but they were not recorded on EEG.
months (mean, 16), but from the first steps the parents noticed that the gait was ataxic, particularly on the days following generalized fits. On follow-up, the patients exhibited cerebellar signs (14 cases), pyramidal signs (7 cases), hemiparesis (3 cases) or dystonia (2 cases). All patients had behavioural disorders which consisted of severe hyperkinesia and speech delay. Twelve patients underwent psychometric evaluation with the BrunetLezine test (French Gesell adaptation) at a mean age of 5 years. The intelligence of no child was within the normal range. The developmental quotient (QD) was ±50 in 5 cases, ±60 in 3, and ±70 in 4. Two patients had repeat evaluations. Before 2 years of age (18 and 21 months), DQ was over 80, but at 5 years it was only 70.
Triggering factors Fever was an evident triggering factor for clonic seizures in all cases, often only a moderate increase in temperature (not exceeding 38.5°C in most instances). In addition, photic stimulation (television, sun or flashlights) in 4 cases, cold water, in 3, and emotional situations, in 2, were also provocative conditions. One patient exhibited autostimulation, moving his head from one side to the other in front of a striped surface until the seizures occurred.
Group B Three patients had never exhibited myoclonic jerks. The ages of onset were respectively, 4,5 and 6 months. One patient had febrile and afebrile generalized clonic seizures. They were of long duration when triggered by fever. EEG showed no interictal spikes, even during sleep or ILS. The child was ataxic, with speech delay and a DQ of 70 at 5 years. Two patients had clonic seizures that were either generalized or unilateral, alternately in one half of the body or the other. In one case, seizures described as absences were observed by the parents from the age of 20 months, but they were not recorded on EEG. Generalized spikes and waves were first recorded at 11 and 23 months, respectively, and they were triggered by ILS at, respectively, 11 months and 4 years. The speech and behaviour abnormalities were similar to those of patients in group A.
EEG
One to 5 (mean, 2.3) standard awake and sleep EEGs were recorded in the first year of life, 1 to 4 (mean, 2.1) in the second year, and at least 1 a year for subsequent years. Fourteen patients had awake and sleep polygraphic recordings (with bilateral deltoid surface electromyographic recording concomitant with EEG). Postictal generalized slowing of background activity was the only EEG abnormality observed until an age ranging from 11 to 30 months (mean, 19). Short bursts of generalized spikes and waves, polyspikes or polyspikes and waves were recorded following this age in 14 cases. They were either isolated or associated with massive myoclonic jerks. Photic stimulation (ILS) triggered these abnormalities in 2 patients, at 17 and 20 months, respectively. Focal spikes occurred in 10 patients. Seven of them had exhibited generalized spikes and these paroxysms persisted at an age ranging from 11 to 57 months (mean, 32). Focal spikes were uni- ( 5 cases) or bilateral (5 cases), and mainly involved the temporal (5 cases) or frontal (3 cases) area.
Outcome The follow-up ranged from 3 years 2 months to 9 years 4 months (mean 6 years 2 months). In half of the cases the frequency of myoclonias decreased progressively; they disappeared at the mean age of 5 years and a half. The patients walked at an age ranging from 9 to 26
DISCUSSION In this study, we addressed the problems of the nosological limits of SMEI and of the reliability of its early diagnosis. An overview of patients with epilepsy beginning in the first year of life was performed by reviewing all the files completed during the study period. It is unlikely that a significant number of patients were misclassified as to the etiology, psychomotor development, CT scan findings or seizure types. On the other hand, patients with SMEI may have myoclonic jerks as the first seizures: such patients would therefore have been omitted from our study group. But these patients do not exceed 10% of the total cases [12] and they do not share the same difficulties for early diagnosis. In addition, SMEI patients with only febrile seizures during the first year could have been omitted, but according to reported series, they represent a minority of cases [11, 12]. After excluding patients with evidence of brain damage or preexisting psychomotor delay, and those with infantile spasms, tonic or partial fits from the first year of life, we found that 17 of the 20 remaining patients fulfilled the
Yakoub et al: Severe myoclonic epilepsy 301
characteristics of SMEI. Therefore, there clearly exists a clinically recognizable pattern of SMEI, as first described by Dravet [8]. In contrast, we found that 3 patients satisfying the same criteria for inclusion in our study exhibited no myoclonias; but they had most of the other characteristics of SMEI, including the types of seizures, behavioural and cognitive disorders, and EEG features. These patients may have had borderline variants of this disorder. The possible existence of borderline cases has also been reported by Ogino and colleagues [13, 14] who compared a group of 7 patients without myoclonias to their personal SMEI series and found that there was the same proportion of familial antecedents and similar outcomes in both groups. SMEI was first recognized in older patients at an age when myoclonic seizures are a most disabling feature, because they produce drop attacks as in the LennoxGastaut syndrome. However, many features distinguish SMEI from Lennox-Gastaut syndrome, which is mainly characterized by tonic seizures and atypical absences as the major types of seizures, and by slow spikes and waves on EEG. In addition, it appears that among all the epileptic patients with mainly myoclonic seizures, SMEI must be distinguished from another syndrome, with a later onset and a more favourable outcome, called "benign myoclonic epilepsy in infancy," that involves myoclonic jerks as a single type of seizures [11, 15] . The early diagnosis of SMEI remains a challenge. Our study was not designed to determine diagnostic features among patients with FC. According to our series, patients are most likely to suffer from SMEI when the first seizure occurs between 2 and 9 months, when the first 2 seizures are clonic, either partial alternating in both halves of the body or generalized, either afebrile or triggered by a moderate temperature increase, when there is no personal neurological history, and when development and CT scanning are normal in the first year. Eighty-five percent of the patients in our series who had these characteristics developed SMEI, the others may have had borderline variants of this disease. Only 1 patient out of the 20 had the same kind of partial motor seizures as the first 2 seizures, but the third seizure was generalized and led to the correct diagnosis. Therefore, from the characteristics of the first three seizures, a diagnosis can be made with good reliability. The number of SMEI cases accounted for 5% of all cases of epilepsy in the first year of life in our series, i.e., of the same order of magnitude as that reported by Dalla Bernardina and coworkers [9]. It was seen in three times less patients than cryptogenic infantile spasms. Although moderate, this frequency is quite high when compared to the group of patients with normal development, normal CT scanning, no etiology, and onset in the first year of life: SMEI accounted for over 17% of the patients with
302 Brain & Development, Vol 14, No 5,1992
these characteristics in the present series. An epidemiological study performed in the "Oise department," France, showed that SMEI occurred in 1 of 32 epileptic patients with onset in the first year of life [16], four times less than infantile spasms as a whole, a frequency that is in the same range as that disclosed by the present study. Therefore, between 1/30,000 and 1/20,000 infants suffer from this disease in France. This figure is in the same range as the 1/40,000 suggested by Hurst [17] . In conclusion, SMEI has clear-cut clinical and EEG characteristics. Diagnosis is reliable in most cases from the first year of life, provided the term "myoclonic," that is part of the definition, is considered as a delayed symptom. The occurrence of afebrile generalized clonic seizures in the first year of life in previously normal children with normal CT scanning and EEG is the major clue for an early diagnosis. REFERENCES 1. Gastaut H, Roger J, Soulayrol R, et al. Childhood epileptic encephalopathy with diffuse slow spike waves (otherwise known as "Petit mal variant") or Lennox syndrome. Epilepsia 1966;7: 139-79. 2. Lennox WG, Davis JP. Clinical correlates of the fast and slow spike-waves on electroencephalogram. Pediatrics 1950;5: 626-44. 3. Doose H, Gerken H, Leonhardt R, Volz E, Volz C. Centrencephalic myoclonic-astatic petit mal. Neuropediatrics 1970; 2:59-78. 4. Doose H. Discussion of myoclonic epilepsies and LennoxGastaut syndrome. In: Roger J, Dravet C, Bureau M, Dreifuss FE, Wolf P, eds. Epileptic syndromes in infancy, childhood and adolescence. London·Paris: John Libbey Eurotext, 1985: 101. 5. Aicardi J. The problem of the Lennox syndrome. Dev Med Child Neurol 1973; 15: 77-80. 6. Roger J, Dravet C, Bureau M, Dreifuss FE, Wolf P, eds. Epileptic syndromes in infancy, childhood and adolescence. London·Paris: John Libbey Eurotext, 1985. 7. 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. 8. Dravet C. Les epilepsies graves de l'enfant. Vie Med 1978;8: 543-8. 9. Dalla Bernardina B, Capo villa G, Gattoni MB, Colamaria V, Bondavalli S, Bureau M. Epilepsie myoclonique grave de la premiere annee. Rev Electroencephalogr Neurophysiol 1982; 12:21-5. 10. Dravet C, Roger J, Bureau M, Dalla Bernardina B. Myoclonic epilepsies in childhood. In: Akimoto H, Kazamatsuri H, Seino M, Ward A, eds. Advances in epileptology: The XIlIth Epilepsy International Symposium. New York: Raven Press, 1982: 135-40. 11. Dravet C, Bureau M, Roger J. Severe myoclonic epilepsy in infancy. In: Roger J, Dravet C, Bureau M, Dreifuss FE, Wolf P, eds. Epileptic syndromes in infancy, childhood and adolescence. London·Paris: John Libbey Eurotext, 1985: 5867. 12. Dulac 0, Arthuis M. Epilepsie myoclonique severe de l'enfant.
Journees Parisiennes de pediatrie. Paris: Flammation, 1982: 259-68. 13. Ogino T, Ohtsuka Y, Amano R, Yamatogi Y, Ohtahara S. An investigation on the borderland of severe myoclonic epilepsy in infancy. Jpn J Psychiatr Neurol (Tokyo) 1988;42: 554-5. 14. Ogino T, Ohtsuka Y, Yamatogi Y, Oka E, Ohtahara S. The epileptic syndrome sharing common characteristics during early childhood with severe myoclonic epilepsy in infancy. Jpn J Psychiatr Neurol (Tokyo) 1989;43: 479-81. 15. Dravet C, Bureau M, Roger J. Benign myoclonic epilepsy in infancy . In: Roger J, Dravet C, Bureau M, Dreifuss FE, Wolf
P, eds. Epileptic syndromes in infancy, childhood and adolescence. London·Paris: John Libbey Eurotext, 1985: 51-7. 16. Luna D, Chiron C, Dulac 0, Pajot N. Epidemiologie des epilepsies de I'enfant dans Ie departement de l'Oise (France). In: Jallon P, ed. Epidemiologie des Epilepsies, Journees d'Etudes Medicales de la Ligue Frallfaise contre l'Epilepsie, Paris Ie 30 Novembre 1987. London·Paris: John Libbey Eurotext, 1988: 41-53. 17. Hurst D1. Epidemiology of severe myoclonic epilepsy in infancy. Epilepsia 1990;31:397-400.
Yakoub et al: Severe myoclonic epilepsy
303
Of 329 epileptic patients referred in a six year period with the first seizure occurring in the first year of life, 20 met the following criteria: generalized seizures excluding infantile spasms, myoclonic, tonic or absence seizures, at least one afebrile seizure, normal development prior to the first seizure, normal CT scan, and no etiology. Seventeen of these 20 patients developed the full pattern of severe myoclonic epilepsy in infancy (SMEI). This syndrome was recognizable from the second or third seizure in the first year of life, although epileptiform EEG abnormalities were lacking until the age of I I to over 30 months. Therefore, based on the clinical pattern, the diagnosis of SMEI can be made with quite good reliability by the end of the first year of life. Key words: Severe myoclonic epilepsy in infancy, myoclonic epilepsy, infants. Yakoub M, Dulac 0, Jambaque I, Chiron C, Plouin P. Early diagnosis of severe myoclonic epilepsy in infancy. Brain Dev 1992; 14: 299-303
Generalized epilepsies of early childhood without identifiable brain lesions remain a very confusing field. The first step of nosological breakdown has consisted of the recognition of broad clinico-electroencephalographic entities characterized by the age of occurrence, seizure types, and ictal and interictal electroencephalographic (EEG) patterns. The Lennox-Gastaut syndrome [1,2] and Doose's myoclonic astatic epilepsy of early childhood [3] are major components of this early breakdown, the former being considered symptomatic and the latter idiopathic [4]. However, these entities remain heterogeneous in terms of the age of onset and evolution, and most authors admit that there is overlapping between these syndromes [4, 5] . Recently, the second step focused on more homogeneous groups of patients. Various syndromes were recently defined [6] and are now listed by the International League Against Epilepsy [7]. The so-called "severe myoclonic epilepsy in infancy" (SMEI), first described by
From Service de Neuropediatrie (MY, 00, lJ, CC) et Service d'Explorations Fonctionnelles du Systeme Nerveux Central (PP), Hopital Saint Vincent de Paul, Paris. Received for publication: December 24, 1991. Accepted for publication: June 7, 1992. Correspondence address: Pr. Olivier Dulac, Service de Neuropectiatrie (Pr. Ponsot), Inserm U 29, Hopital Saint Vincent de Paul, 82 avenue Denfert Rochereau, 75674 Paris Cedex 14, France.
Dravet [8] , is one of these recently recognized syndromes. According to several studies [9-12], it begins in the first year of life with clonic seizures that are usually generalized and often associated with fever. Subsequently, clonic seizures recur with a slight or no increase in temperature, and they often involve alternately one side of the body and then the other, with frequent status epilepticus. The myoclonic attacks only occur between 1 and 4 years of life, and absence seizures are rare. Complex partial seizures may also appear later. Psychomotor development is normal during the first year of life, but an ataxic gait and speech delay become evident in the second year. Computerized tomographic (CT) scanning and EEG are normal at the onset, and generalized spike waves appear only in the second year of life: they are isolated or occur in brief bursts with frequent photosensitivity. These patients do not exhibit tonic seizures or slow spike waves on EEG, in contrast with in the Lennox-Gastaut syndrome. The full pattern of SMEI is therefore easily recognizable. But since the first signs seem to lack specificity, several questions related to an early diagnosis remain. What are the most reliable early diagnostic features? At what age does a diagnosis become possible? To what extent, during the first months of the disease, are the clinical, EEG and radiological characteristics of the syndrome shared by other types of epilepsies? Are there borderline forms? What is the relative frequency of this syndrome among epilepsies beginning in the first year of
life? In order to address these questions, we studied an unselected group of patients who had the characteristics of SMEI during the first year of life, and we found that in most of them the syndromic pattern was completed during subsequent years. PATIENTS AND METHODS The study was performed blindly as to their current diagnosis. 1) Patients were selected according to the following criteria: first seizure between 1 and 12 months of life, based on clinical description, generalized clonic or tonicclonic seizure pattern excluding infantile spasms, or tonic seizures, and with or without associated focal motor seizures; at least one afebrile seizure (temperature under 38.5°C); normal psychomotor development prior to the first seizure; no etiology, including a personal history; normal CT scanning before 2 years of age; first referral prior to 2 years of age, between January 1980 and December 1985; and follow-up of over 3 years from the onset. Of the 329 epileptic patients with the first seizure between 1 and 12 months of life who were referred to the neuropediatric department of the Hopital St Vincent de Paul, Paris, between January 1980 and December 1985, 195 were excluded because their psychomotor development was considered to be abnormal before the first seizure, or because there was evidence an underlying brain lesion. Of the 134 remaining patients, 95 were excluded because they exhibited infantile spasms (58 cases), or seizure types were restricted to partial, myoclonic, tonic and absence seizures during the first year of life. Nineteen were excluded because of insufficient data concerning seizure types or because of delayed referral or insufficient follow-up. The 20 remaining patients were therefore selected for this study according to the previously mentioned criteria. None of them was included in our 1982 series [12]. 2) For each of these 20 selected patients, the following features were studied: age at onset, characteristics of the first and following seizures, number of seizures, the triggering factors, interictal EEG pattern, and outcome in terms of seizures, and of neurological and behavioural problems. All patients underwent EEG follow-up from the first year of life. Seventeen of the 20 selected patients exhibited myoclonic fits, and they fulfilled the clinical and EEG characteristics of "SMEI" (Group A). The 3 other patients remained free of myoclonias (Group B). Group A Seventeen patients, 8 boys and 9 girls, exhibited a very
300 Brain & Development, Vol 14, No 5, 1992
homogeneous pattern. Age of onset and characteristics of the first seizure The age of onset ranged from 2 to 9 months (mean, 5 months 5 days); the first seizure was generalized clonic or tonic-clonic in 9 cases and unilateral in 8. Six patients had postictal hemiparesis lasting a few hours. The first seizure lasted 2 minutes to 3 hours (mean, 32 minutes), over half an hour in 7 patients. In 14 cases, the seizure was triggered by fever (8 cases) or vaccination (6 cases). Subsequent seizures The second seizure occurred 8 days to 4 months (mean, 40 days) after the first one, the delay being less than 1 month in 11 cases. Again, it was clonic or tonic-clonic, generalized (10 cases) or unilateral (7 cases). In all patients but one the first 2 seizures involved both sides of the body: they were either generalized in at least one instance (13 cases) or hemiclonic involving both sides alternately (3 cases). In only one of the 17 patients, the first 2 seizures were unilateral and involved the same half of the body; in this patient, the third seizure was generalized. All 17 patients exhibited therefore at least 1 generalized clonic seizure among the first 3 seizures. The mean time from the second to the third seizure was 30 days, the time between seizures decreasing for the subsequent fits. Seizures lasting over half an hour occurred 2 to 3 times in 14 patients during the first year of life, and 1 to 2 times in 10 of them during the second year. During the first 2 years of life, postictal hemiplegia lasting less than 3 hours occurred in 11 patients, mainly following unilateral seizures. The mean number of seizures was 9 in the first year and 17 in the second year. By the end of the first year, 13 patients had exhibited unilateral clonic seizures alternately in both halves of the body; this was the case for another 2 patients in the second year of life. Myoclonic seizures They occurred at a mean age of 2 years (range, 9 to 48 months). They were mild, occasional and erratic in 8 cases. They were generalized and frequent in 9 cases, producing drop attacks in 7 of them. Some patients exhibited both types. The myoclonic jerks occurred mainly in the morning, as brief bursts lasting 1 to 2 seconds, or during drowsiness. In eight patients, a clear increase in their frequency often preceded generalized seizures. Three patients had myoclonic status epilepticus lasting, respectively, 3, 24 and 36 hours, at the ages of 14 months, and 4 and 5 years. Other types of seizures Fifteen patients exhibited other types of seizures: - partial clonic seizures involving a single limb in 5
cases (onset between 9 and 39 months; mean, 26) and both upperlimbs alternately inl case (onset at 18 months); - versive seizures in 9 cases (onset between 4 months and 5 years; mean, 2 years 6 months); - complex partial seizures lasting several minutes in all 15 patients, with onset between 10 months and 3 years of age; mean, 2 years; these seizures comprised automatisms in 2 cases (mouthing and swallowing) and vegetative manifestations in 3 others (mydriasis, sweating, cyanosis or erythrosis); - short attacks of isolated loss of consciousness resembling absence seizures were reported by the parents in 2 cases, with onset at 13 to 29 months, but they were not recorded on EEG.
months (mean, 16), but from the first steps the parents noticed that the gait was ataxic, particularly on the days following generalized fits. On follow-up, the patients exhibited cerebellar signs (14 cases), pyramidal signs (7 cases), hemiparesis (3 cases) or dystonia (2 cases). All patients had behavioural disorders which consisted of severe hyperkinesia and speech delay. Twelve patients underwent psychometric evaluation with the BrunetLezine test (French Gesell adaptation) at a mean age of 5 years. The intelligence of no child was within the normal range. The developmental quotient (QD) was ±50 in 5 cases, ±60 in 3, and ±70 in 4. Two patients had repeat evaluations. Before 2 years of age (18 and 21 months), DQ was over 80, but at 5 years it was only 70.
Triggering factors Fever was an evident triggering factor for clonic seizures in all cases, often only a moderate increase in temperature (not exceeding 38.5°C in most instances). In addition, photic stimulation (television, sun or flashlights) in 4 cases, cold water, in 3, and emotional situations, in 2, were also provocative conditions. One patient exhibited autostimulation, moving his head from one side to the other in front of a striped surface until the seizures occurred.
Group B Three patients had never exhibited myoclonic jerks. The ages of onset were respectively, 4,5 and 6 months. One patient had febrile and afebrile generalized clonic seizures. They were of long duration when triggered by fever. EEG showed no interictal spikes, even during sleep or ILS. The child was ataxic, with speech delay and a DQ of 70 at 5 years. Two patients had clonic seizures that were either generalized or unilateral, alternately in one half of the body or the other. In one case, seizures described as absences were observed by the parents from the age of 20 months, but they were not recorded on EEG. Generalized spikes and waves were first recorded at 11 and 23 months, respectively, and they were triggered by ILS at, respectively, 11 months and 4 years. The speech and behaviour abnormalities were similar to those of patients in group A.
EEG
One to 5 (mean, 2.3) standard awake and sleep EEGs were recorded in the first year of life, 1 to 4 (mean, 2.1) in the second year, and at least 1 a year for subsequent years. Fourteen patients had awake and sleep polygraphic recordings (with bilateral deltoid surface electromyographic recording concomitant with EEG). Postictal generalized slowing of background activity was the only EEG abnormality observed until an age ranging from 11 to 30 months (mean, 19). Short bursts of generalized spikes and waves, polyspikes or polyspikes and waves were recorded following this age in 14 cases. They were either isolated or associated with massive myoclonic jerks. Photic stimulation (ILS) triggered these abnormalities in 2 patients, at 17 and 20 months, respectively. Focal spikes occurred in 10 patients. Seven of them had exhibited generalized spikes and these paroxysms persisted at an age ranging from 11 to 57 months (mean, 32). Focal spikes were uni- ( 5 cases) or bilateral (5 cases), and mainly involved the temporal (5 cases) or frontal (3 cases) area.
Outcome The follow-up ranged from 3 years 2 months to 9 years 4 months (mean 6 years 2 months). In half of the cases the frequency of myoclonias decreased progressively; they disappeared at the mean age of 5 years and a half. The patients walked at an age ranging from 9 to 26
DISCUSSION In this study, we addressed the problems of the nosological limits of SMEI and of the reliability of its early diagnosis. An overview of patients with epilepsy beginning in the first year of life was performed by reviewing all the files completed during the study period. It is unlikely that a significant number of patients were misclassified as to the etiology, psychomotor development, CT scan findings or seizure types. On the other hand, patients with SMEI may have myoclonic jerks as the first seizures: such patients would therefore have been omitted from our study group. But these patients do not exceed 10% of the total cases [12] and they do not share the same difficulties for early diagnosis. In addition, SMEI patients with only febrile seizures during the first year could have been omitted, but according to reported series, they represent a minority of cases [11, 12]. After excluding patients with evidence of brain damage or preexisting psychomotor delay, and those with infantile spasms, tonic or partial fits from the first year of life, we found that 17 of the 20 remaining patients fulfilled the
Yakoub et al: Severe myoclonic epilepsy 301
characteristics of SMEI. Therefore, there clearly exists a clinically recognizable pattern of SMEI, as first described by Dravet [8]. In contrast, we found that 3 patients satisfying the same criteria for inclusion in our study exhibited no myoclonias; but they had most of the other characteristics of SMEI, including the types of seizures, behavioural and cognitive disorders, and EEG features. These patients may have had borderline variants of this disorder. The possible existence of borderline cases has also been reported by Ogino and colleagues [13, 14] who compared a group of 7 patients without myoclonias to their personal SMEI series and found that there was the same proportion of familial antecedents and similar outcomes in both groups. SMEI was first recognized in older patients at an age when myoclonic seizures are a most disabling feature, because they produce drop attacks as in the LennoxGastaut syndrome. However, many features distinguish SMEI from Lennox-Gastaut syndrome, which is mainly characterized by tonic seizures and atypical absences as the major types of seizures, and by slow spikes and waves on EEG. In addition, it appears that among all the epileptic patients with mainly myoclonic seizures, SMEI must be distinguished from another syndrome, with a later onset and a more favourable outcome, called "benign myoclonic epilepsy in infancy," that involves myoclonic jerks as a single type of seizures [11, 15] . The early diagnosis of SMEI remains a challenge. Our study was not designed to determine diagnostic features among patients with FC. According to our series, patients are most likely to suffer from SMEI when the first seizure occurs between 2 and 9 months, when the first 2 seizures are clonic, either partial alternating in both halves of the body or generalized, either afebrile or triggered by a moderate temperature increase, when there is no personal neurological history, and when development and CT scanning are normal in the first year. Eighty-five percent of the patients in our series who had these characteristics developed SMEI, the others may have had borderline variants of this disease. Only 1 patient out of the 20 had the same kind of partial motor seizures as the first 2 seizures, but the third seizure was generalized and led to the correct diagnosis. Therefore, from the characteristics of the first three seizures, a diagnosis can be made with good reliability. The number of SMEI cases accounted for 5% of all cases of epilepsy in the first year of life in our series, i.e., of the same order of magnitude as that reported by Dalla Bernardina and coworkers [9]. It was seen in three times less patients than cryptogenic infantile spasms. Although moderate, this frequency is quite high when compared to the group of patients with normal development, normal CT scanning, no etiology, and onset in the first year of life: SMEI accounted for over 17% of the patients with
302 Brain & Development, Vol 14, No 5,1992
these characteristics in the present series. An epidemiological study performed in the "Oise department," France, showed that SMEI occurred in 1 of 32 epileptic patients with onset in the first year of life [16], four times less than infantile spasms as a whole, a frequency that is in the same range as that disclosed by the present study. Therefore, between 1/30,000 and 1/20,000 infants suffer from this disease in France. This figure is in the same range as the 1/40,000 suggested by Hurst [17] . In conclusion, SMEI has clear-cut clinical and EEG characteristics. Diagnosis is reliable in most cases from the first year of life, provided the term "myoclonic," that is part of the definition, is considered as a delayed symptom. The occurrence of afebrile generalized clonic seizures in the first year of life in previously normal children with normal CT scanning and EEG is the major clue for an early diagnosis. REFERENCES 1. Gastaut H, Roger J, Soulayrol R, et al. Childhood epileptic encephalopathy with diffuse slow spike waves (otherwise known as "Petit mal variant") or Lennox syndrome. Epilepsia 1966;7: 139-79. 2. Lennox WG, Davis JP. Clinical correlates of the fast and slow spike-waves on electroencephalogram. Pediatrics 1950;5: 626-44. 3. Doose H, Gerken H, Leonhardt R, Volz E, Volz C. Centrencephalic myoclonic-astatic petit mal. Neuropediatrics 1970; 2:59-78. 4. Doose H. Discussion of myoclonic epilepsies and LennoxGastaut syndrome. In: Roger J, Dravet C, Bureau M, Dreifuss FE, Wolf P, eds. Epileptic syndromes in infancy, childhood and adolescence. London·Paris: John Libbey Eurotext, 1985: 101. 5. Aicardi J. The problem of the Lennox syndrome. Dev Med Child Neurol 1973; 15: 77-80. 6. Roger J, Dravet C, Bureau M, Dreifuss FE, Wolf P, eds. Epileptic syndromes in infancy, childhood and adolescence. London·Paris: John Libbey Eurotext, 1985. 7. 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. 8. Dravet C. Les epilepsies graves de l'enfant. Vie Med 1978;8: 543-8. 9. Dalla Bernardina B, Capo villa G, Gattoni MB, Colamaria V, Bondavalli S, Bureau M. Epilepsie myoclonique grave de la premiere annee. Rev Electroencephalogr Neurophysiol 1982; 12:21-5. 10. Dravet C, Roger J, Bureau M, Dalla Bernardina B. Myoclonic epilepsies in childhood. In: Akimoto H, Kazamatsuri H, Seino M, Ward A, eds. Advances in epileptology: The XIlIth Epilepsy International Symposium. New York: Raven Press, 1982: 135-40. 11. Dravet C, Bureau M, Roger J. Severe myoclonic epilepsy in infancy. In: Roger J, Dravet C, Bureau M, Dreifuss FE, Wolf P, eds. Epileptic syndromes in infancy, childhood and adolescence. London·Paris: John Libbey Eurotext, 1985: 5867. 12. Dulac 0, Arthuis M. Epilepsie myoclonique severe de l'enfant.
Journees Parisiennes de pediatrie. Paris: Flammation, 1982: 259-68. 13. Ogino T, Ohtsuka Y, Amano R, Yamatogi Y, Ohtahara S. An investigation on the borderland of severe myoclonic epilepsy in infancy. Jpn J Psychiatr Neurol (Tokyo) 1988;42: 554-5. 14. Ogino T, Ohtsuka Y, Yamatogi Y, Oka E, Ohtahara S. The epileptic syndrome sharing common characteristics during early childhood with severe myoclonic epilepsy in infancy. Jpn J Psychiatr Neurol (Tokyo) 1989;43: 479-81. 15. Dravet C, Bureau M, Roger J. Benign myoclonic epilepsy in infancy . In: Roger J, Dravet C, Bureau M, Dreifuss FE, Wolf
P, eds. Epileptic syndromes in infancy, childhood and adolescence. London·Paris: John Libbey Eurotext, 1985: 51-7. 16. Luna D, Chiron C, Dulac 0, Pajot N. Epidemiologie des epilepsies de I'enfant dans Ie departement de l'Oise (France). In: Jallon P, ed. Epidemiologie des Epilepsies, Journees d'Etudes Medicales de la Ligue Frallfaise contre l'Epilepsie, Paris Ie 30 Novembre 1987. London·Paris: John Libbey Eurotext, 1988: 41-53. 17. Hurst D1. Epidemiology of severe myoclonic epilepsy in infancy. Epilepsia 1990;31:397-400.
Yakoub et al: Severe myoclonic epilepsy
303