Epileptic Electroencephalographic Abnormalities and Developmental Dysphasias: A Study of 32 Patients Bernard Echenne, MD, Renee Cheminal, MD, Fran~ois Rivier, MD, Christian Negre, MD, Jacques Touchon, MD and Michel Billiard, MD
The relationships between severe developmental dysphasias and epilepsy were analysed in 32 patients with congenital dysphasias. The mean age was 8 years 2 months; 19 of 32 had never had seizures; 9 had had occasional seizures; 4 were epileptic. Twenty-two of 32 had normal repeated standard EEGs, but 10(2 of which never had seizures) showed epileptic interictal discharges. During prolonged EEG after sleep deprivation, epileptic abnormalities were observed in 13 of the 32 cases (4 of which never had seizures). The overall night sleep recordings showed epileptic abnormalities in 30 of the 32 cases (17 of which had never had seizures). The epileptic interictal abnormalities varied considerably in intensity and aspect in the same patient from one examination to another. Developmentally aphasic children show a higher incidence of abnormal EEG than expected, particularly during overall night recordings. In most cases, the physiopathology of the language disturbance might be identical to that in Landau-Kleffner syndrome. Key words: Dysphasia, language disorder, epilepsy, Landau-Kleffner syndrome. Echenne B, Cheminal R, Rivier F, Negre C, Touchon J, Billiard M Epileptic electroencephalographic abnormalities and developmental dysphasias: a study of 32 patients. Brain Dev 1992;14: 216-25
Epileptic seizures in the dominant hemisphere can be associated with intermittent or prolonged language impairment in a number of circumstances. These include complex partial epilepsy, absence status, nonconvulsive status and Landau-Kleffner syndrome [1-7]. The latter is an acquired aphasia, of rapid onset most often, observed in children with already patterned normal language [1,8,9]. In these children, significant EEG abnormalities are in contrast with the absence or rare occurrence of clinical seizures. In preliminary studies [7, lO-12] , we were struck by the frequency of the intericial epileptic EEG abnormalities noted in children with developmental dysphasias. We hypothesized that, in most cases, the mechanism underlying developmental dysphasias is identical to that
From the Sern.ces de Neuropediatrie (BE, RC, FR), et Neuroiogie B (JT, MB), Centre Gui de Chauliac, Montpellier; Service de Neuroiogie, Hopitai Joffre, Perpignan (CN). Received for publication: November 23, 1991. Accepted for publication: April 25, 1992. Correspondence address: Pro B. Echenne, Service de Neuropediatrie, Centre Gui de Chauliac, 34059 Montpellier Cedex, France.
in Landau-Kleffner syndrome, the dysfunction occurring before language begins to become patterned. In order to confirm this hypothesis we have, from 1987, been conducting a prospective study of standard EEGs, prolonged EEGs with sleep deprivation and night polygraphic EEGs in children with developmental dysphasias. SUBJECTS AND METHODS
We investigated children who were referred for consultation to the Neuropediatrics Unit (BE, RC) for severe language impairment. Since 1984, a multidisciplinary team in charge of the analysis and management of language dysfunctions has existed in our unit. This was a prospective study carried out from 1986 to 1990. The children were followed for 1 to 4 years. None of the patients had been specifically referred for epilepsy even though some had already had seizures. Developmental dysphasia was defined as a severe language disorder in which children's speech level was at least 2 years delayed as to their chronological age, and when troubles in comprehension and/or expression were responsible for this delay. Each child was investigated as follows:
• Initial neurological investigation with a set of neuropsycholinguistic tests, which was then repeated once or twice yearly over a period of one to four years: estimation of the global level of performance as to the Wechsler scale, WPPSI or Mac Carthy scale, according to age [13-15]; analysis of language and speech functions: Chevrie-Muller's tests [16, 17] ,audiophonometric analysis including a study of audiophonetic gnosias and discriminative auditory tests (Chevrie-Muller et al [18], Le Normand et al [19], Rourke et al [20]), tests of word and sentences comprehension (Chevrie-Muller's tests [16-18], Dague and Lege [21], Deltour's test [22]); study of articulatory praxis, phonology, morphosyntax, tests of word repeating and naming tests [16,22] ; study of the child's spontaneous production during free or directed games allowing pragmatics, fluency and prosody to be evaluated. • Audiometry and study of auditory evoked potentials. • Cerebral CT scan and/or MRI. During the subsequent follow-up, each child had several types of EEG investigations: first, a standard EEG approximately every 6 months (diurnal waking state EEG for 20 minutes); then a diurnal prolonged EEG (2 hours) after partial deprivation of nocturnal sleep (waking at 2:00 am or 3:00 am), an average of two for each child. During this investigation, the waking state and a short period of spontaneous sleep were recorded; and finally, a continuous overnight polygraphic recording within three months after the diagnosis of developmental dysphasia, which was repeated once in some children. EEGs were not obtained at fixed intervals for each patient; following the diagnosis, a standard EEG was recorded and then a prolonged one with sleep deprivation. Nocturnal EEGs were performed later (within three months following diagnosis). Children with IQs below 70, ones with severe motor disturbances, like congenital hemiplegia and diplegia, and ones with recurrence of seizures on antiepileptic drugs (AEDs) were excluded from the study. None of the children included in the study had had seizures during the EEG investigations. Thus, no bias has been possibly introduced by ictal language disturbance. The EEGs abnormalities have been analysed as follows: were considered as interictal epileptiform potentials localized or generalized spike-waves, polyspike and waves, and multiple spike-and-slow wave complexes, multifocal independent sharp and slow wave complexes, high voltage spikes with prominent aftercoming slow waves appearing singly or in groups at the central-mid temporal region (C3 and T3 or C4 and T4), spikes of similar morphology to rolandic spikes appearing in other regions, temporal sharp waves or spikes, occipital spikes and slow waves foci [23]. Were not considered as epileptic bursts of 2 to 4 Hz, rhythmic waves and rhythmic occipital delta, generalized polyspikes, rhythmic 10 to 20 Hz waves, or electrodecre-
mental events during sleep [23, 24]. All the children referred to us and not presenting any of the exclusion criteria were evaluated. RESULTS Thirty-two patients were included (18 boys, 14 girls), their ages ranging from 3 years and 8 months to 14 years (mean: 8 years 2 months). The neurological examination was normal in 16 cases. In the others, a moderate disability of motor functions was observed: delay in coordination and/or equilibrium occurred in 2 cases, slight hypotonia in 1 case, macrocrania in 1 case and isolated dysfunctioning of orobuccofacial praxis in 12 patients. Neuroradiologic investigations and audiometric studies were normal in all cases. IQ was above or equal to 90 in 9 patients, ranging from 70 to 90 in the others. Language impairments were classified according to Rapin [25]. We thus diagnosed verbal or auditory agnosia when a child was unable to understand sounds in spite of normal audition and when there was no or very little production. The phonologic-syntactic syndrome was diagnosed when the abnormalities were predominant at the phonologic and syntactic levels, with a comprehension problem varying in intensity. The syntactic-lexical syndrome was diagnosed for those cases with normal comprehension but deviant syntactic production, with major trouble in word evocation. The phonologic programming deficit syndrome was defined as almost nonexistent production, with normal or subnormal comprehension. The syndrome of phonologic production (or verbal apraxia) has rarely been diagnosed, the line between it and a simple delay in phonologic production being not clear. Neuropsychologic analysis allowed the identification of 6 cases of auditory or verbal agnosia, 18 of phonologic syntactic deficit syndrome, 4 of phonologic programming deficit syndrome, 2 of lexical syntactic deficit syndrome and 2
Table I Incidence of seizures in the 32 patients with developmental dysphasias Occasional Number No of cases seizures seizures Epilepsy «3)
Verbal auditory agnosia Phonologic-syntactic deficit syndrome Lexical-syntactic deficit syndrome Phonologic programming deficit syndrome Verbal apraxia without programming deficit Total
6
4
18
9
6
3
2
2
0
0
4
3
1
0
2
1
1
0
32
19
9
4
Echenne et al: EEG and developmental dysphasias 217
of verbal apraxia. Nineteen of the 32 children had no history of seizures or epilepsy. Nine of the thirty-two had had occasional seizures (less than 3): non-recurrent febrile convulsions (I case), and generalized clonic seizures without fever once (3 cases) or twice (5 cases). Four of the thirty-two (Table 1) presented primary generalized epilepsy well controlled by AEDs. In these patients, the last seizures occurred at least 3 months before the investigations. A systematic investigation concerning the familial in· cidence of developmental dysphasia was not determined for the siblings because of the overworking it would have presented for our team. Certainly, family histories have evidenced the notion of language delay in some of our patients' brothers and sisters. However, in the absence of complete neuropsychologic investigations, these cases have not been systematically taken into account. Two of the 32 patients were sisters. The results of standard EEG are summarized in Table 2.
Four EEGs on average were recorded for each child (range: 2 to 9). Repeated standard EEGs were consistently normal in 22 children (Table 2). In 7 cases, both normal and abnormal EEGs were recorded without a relationship with the language impairment. In 3 cases, EEGs constantly showed epileptic discharges. In all cases, the epileptic discharges accounted for less than 10% of the graph. Ten of the 32 children presented interictal epileptic discharges on standard EEGs, which were inconstantly noted in 7 cases. Eight of these 10 children had had occasional ~eizures or were epileptic. Two of them had never had seizures. The results of prolonged EEGs after nocturnal sleep partial deprivation were normal in 18 of the 32 children only (Table 3). In 6 cases, they are inconstantly disturbed (auditory verbal agnosia: 1, phonologic syntactic deficit syndrome: 4, verbal apraxia: 1). In 8 cases, the recordings were constantly abnormal. The epileptic abnormalities were polymorphic: there were generalized spike-waves
Table 2 Results of standard EEGs (4 on average for each child) Number of cases Verbal auditory agnosia Phonologic-syntactic deficit syndrome Lexical-syntactic deficit syndrome Phonologic programming deficit syndrome Verbal apraxia Total
Either normal or abnormal
Always normal
Always abnormal
6
2
3
18
12
4
2
2 4 2
2 4 2
0 0 0
0 0 0
32
22
7
3
Generalized
Focal S-Wor S
2 (1)
2 (rolandic or temporal)
s-w
Focal + generalized S·W
5* (2) (rolandic 2, temporal 3)
* In 2 patients, the abnormal signs have not been identical at each control (both with normal EEGs or with generalized S-W, or with focal + generalized signs). ( ) number of cases with occasionally normal EEGs. S: spike, S-W: spike-wave.
Table 3 Results of prolonged EEGs after sleep deprivation (1 to 3 for each child) Number Always Either Always Generalized of cases normal normal or abnorl1U1[ S·W abnormal Verbal auditory agnosia Phonologic-syntactic deficit syndrome Lexical-syntactic deficit syndrome Phonologic programming deficit syndrome Verbal apraxia Total
6
3
18
9
2 4 2
1 4 1
32
18
4
Focal S-WorPS
2
2*
1 (right rolandic PS)
5
5**
1 (bitemporal)
Sleep Focal + generalized continuous signs S-W
2*
1* (right ro1andic PS) 6
8
* One of these patients had a normal recording on the second investigation, * * 3 of these patients had a normal recording on the second investigation. PS: polyspikes.
218 Brain & Development, Vol 14,No 4, 1992
(8 cases), localized spike-waves or polyspikes (right rolandic: 2 cases) or bitemporal (1 case), or continuous spike-waves during sleep (3 cases). In 6 cases, the abnormalities had already occurred during the waking state
recording; they were clearly increased during sleep in a total of 14 cases. The epileptic discharges accounted for less than 10% in 10 cases; they ranged from 10 to 20% in 1 case; they were above 85% in 3 others.
Table 4 Results of overnight polygraphic EEGs
Abnormal Continuous generalized SoW activity during sleep Generalized SoW Generalized SoW + (temporal, rolandic or occipital) Left S or SoW Right S or SoW Rolandic spikes Bilateral Bitemporal activity
Verbal agnosia
Phonologicsyntactic deficit syndrome
Syntacticlexical syndrome
Phonologic programming syndrome
Verbal apraxia
6/6 2 2
17/18 2 5
1/2
4/4
2/2
6* 3*
2
2
2 1***
*A second investigation showed the same signs in the same patient (1 patient). ** A second investigation showed generalized spike-waves with a left temporal focus in one of these 3 patients_ ***This was observed twice in the same patient. A third control showed continuous generalized spike-wave activity during sleep.
Fig 1 Variations of the abnormalities observed in the same patient (only one brief afebrile generalized seizure when he was 4 years old;
no recurrence on follow-up for 7 years; severe developmental dysphasia of the phonologic syntactic type. A) (3/1984) Very active right temporo-occipital epileptic activity with some asynchronous occipital spikes and spike-waves. B) (1/1985) Generalized polyspikes and spike-waves_ C) (2/1986) Left fronto-temporal spikes and spike-waves.
Echenne et al; EEG and developmental dysphasias 219
Therefore, after sleep deprivation 14 of the 32 patients presented interictal epileptic ,discharges in waking state or diurnal sleep EEGs. The pathologic signs were inconstantly noted in 6 cases. Of these 14 children,S had had occasional seizures, 4 were epileptic and 5 had never had seizures. The results of night polygraphic recordings are summarized in Table 4 (1 investigation per child at least, 2 investigations in 4 of them and 3 in another one). Epileptiform activity was recorded in 30 of the 32 patients, in all types of dysphasias at any age. In 4 cases, epileptic signs accounted for 85% or more of the graph, and in 6 cases, they accounted for 50 to 85% of it. In 10 patients, they accounted for 10 to 50% of the recording, and in 10 patients, for less than 10%. The results varied greatly from one case to another: generalized continuous spike-waves (4 cases), discharges of generalized spike-waves (10 cases), generalized spike-waves associated with local discharges of temporal, occipital or rolandic spike-waves or polyspikes, either right (3 cases) or left (10 cases), discharges of rhythmic bilateral rolandis spikes and polyspikes (2 cases), bitemporal polyspikes and spikewaves (1 case). When a second overall night recording was obtained in the same patient (5 cases), the features were
found to be identical in 3 cases. In a fourth patient, a left temporal focus was observed instead of the previously right temporal focus and in the fifth patient, generalized continuous spike-wave activity was recorded (bitemporal spike-waves previously). On the whole, thirty of the 32 children with developmental dysphasias exhibited clear interictal epileptic features during sleep. The morphology of interictal signs varied greatly from one case to another. Generalized discharges of 2 or 2.5 Hz associated or not with focal discharges were mainly observed. In the same patient, the frequency and localization of the epileptiform changes varied considerably according to time (Fig 1). These abnormalities clearly increased during slow sleep (Fig 2) and disappeared during REM sleep (except in one patient in which they persisted). In 4 cases only, generalized continuous spike-waves were observed. Among these 30 children, standard EEGs were consistently normal in 21 cases, and 17 of these children had never had seizures. Moreover, the overall night sleep recordings showed epileptic features in 16 children, in whom the afternoon sleep recordings were normal.
-, ,
Cz· Pz i""-v/Jw.;'fI~.~v.¥",~-#M"""~","","V'''''I'\:I''"-J'''''''-'-''''' '~IIVn\N
FP'·C.:Vw;~r~
T. · Fp'
, ,,,,
The relationships between severe developmental dysphasias and epilepsy were analysed in 32 patients with congenital dysphasias. The mean age was 8 years 2 months; 19 of 32 had never had seizures; 9 had had occasional seizures; 4 were epileptic. Twenty-two of 32 had normal repeated standard EEGs, but 10(2 of which never had seizures) showed epileptic interictal discharges. During prolonged EEG after sleep deprivation, epileptic abnormalities were observed in 13 of the 32 cases (4 of which never had seizures). The overall night sleep recordings showed epileptic abnormalities in 30 of the 32 cases (17 of which had never had seizures). The epileptic interictal abnormalities varied considerably in intensity and aspect in the same patient from one examination to another. Developmentally aphasic children show a higher incidence of abnormal EEG than expected, particularly during overall night recordings. In most cases, the physiopathology of the language disturbance might be identical to that in Landau-Kleffner syndrome. Key words: Dysphasia, language disorder, epilepsy, Landau-Kleffner syndrome. Echenne B, Cheminal R, Rivier F, Negre C, Touchon J, Billiard M Epileptic electroencephalographic abnormalities and developmental dysphasias: a study of 32 patients. Brain Dev 1992;14: 216-25
Epileptic seizures in the dominant hemisphere can be associated with intermittent or prolonged language impairment in a number of circumstances. These include complex partial epilepsy, absence status, nonconvulsive status and Landau-Kleffner syndrome [1-7]. The latter is an acquired aphasia, of rapid onset most often, observed in children with already patterned normal language [1,8,9]. In these children, significant EEG abnormalities are in contrast with the absence or rare occurrence of clinical seizures. In preliminary studies [7, lO-12] , we were struck by the frequency of the intericial epileptic EEG abnormalities noted in children with developmental dysphasias. We hypothesized that, in most cases, the mechanism underlying developmental dysphasias is identical to that
From the Sern.ces de Neuropediatrie (BE, RC, FR), et Neuroiogie B (JT, MB), Centre Gui de Chauliac, Montpellier; Service de Neuroiogie, Hopitai Joffre, Perpignan (CN). Received for publication: November 23, 1991. Accepted for publication: April 25, 1992. Correspondence address: Pro B. Echenne, Service de Neuropediatrie, Centre Gui de Chauliac, 34059 Montpellier Cedex, France.
in Landau-Kleffner syndrome, the dysfunction occurring before language begins to become patterned. In order to confirm this hypothesis we have, from 1987, been conducting a prospective study of standard EEGs, prolonged EEGs with sleep deprivation and night polygraphic EEGs in children with developmental dysphasias. SUBJECTS AND METHODS
We investigated children who were referred for consultation to the Neuropediatrics Unit (BE, RC) for severe language impairment. Since 1984, a multidisciplinary team in charge of the analysis and management of language dysfunctions has existed in our unit. This was a prospective study carried out from 1986 to 1990. The children were followed for 1 to 4 years. None of the patients had been specifically referred for epilepsy even though some had already had seizures. Developmental dysphasia was defined as a severe language disorder in which children's speech level was at least 2 years delayed as to their chronological age, and when troubles in comprehension and/or expression were responsible for this delay. Each child was investigated as follows:
• Initial neurological investigation with a set of neuropsycholinguistic tests, which was then repeated once or twice yearly over a period of one to four years: estimation of the global level of performance as to the Wechsler scale, WPPSI or Mac Carthy scale, according to age [13-15]; analysis of language and speech functions: Chevrie-Muller's tests [16, 17] ,audiophonometric analysis including a study of audiophonetic gnosias and discriminative auditory tests (Chevrie-Muller et al [18], Le Normand et al [19], Rourke et al [20]), tests of word and sentences comprehension (Chevrie-Muller's tests [16-18], Dague and Lege [21], Deltour's test [22]); study of articulatory praxis, phonology, morphosyntax, tests of word repeating and naming tests [16,22] ; study of the child's spontaneous production during free or directed games allowing pragmatics, fluency and prosody to be evaluated. • Audiometry and study of auditory evoked potentials. • Cerebral CT scan and/or MRI. During the subsequent follow-up, each child had several types of EEG investigations: first, a standard EEG approximately every 6 months (diurnal waking state EEG for 20 minutes); then a diurnal prolonged EEG (2 hours) after partial deprivation of nocturnal sleep (waking at 2:00 am or 3:00 am), an average of two for each child. During this investigation, the waking state and a short period of spontaneous sleep were recorded; and finally, a continuous overnight polygraphic recording within three months after the diagnosis of developmental dysphasia, which was repeated once in some children. EEGs were not obtained at fixed intervals for each patient; following the diagnosis, a standard EEG was recorded and then a prolonged one with sleep deprivation. Nocturnal EEGs were performed later (within three months following diagnosis). Children with IQs below 70, ones with severe motor disturbances, like congenital hemiplegia and diplegia, and ones with recurrence of seizures on antiepileptic drugs (AEDs) were excluded from the study. None of the children included in the study had had seizures during the EEG investigations. Thus, no bias has been possibly introduced by ictal language disturbance. The EEGs abnormalities have been analysed as follows: were considered as interictal epileptiform potentials localized or generalized spike-waves, polyspike and waves, and multiple spike-and-slow wave complexes, multifocal independent sharp and slow wave complexes, high voltage spikes with prominent aftercoming slow waves appearing singly or in groups at the central-mid temporal region (C3 and T3 or C4 and T4), spikes of similar morphology to rolandic spikes appearing in other regions, temporal sharp waves or spikes, occipital spikes and slow waves foci [23]. Were not considered as epileptic bursts of 2 to 4 Hz, rhythmic waves and rhythmic occipital delta, generalized polyspikes, rhythmic 10 to 20 Hz waves, or electrodecre-
mental events during sleep [23, 24]. All the children referred to us and not presenting any of the exclusion criteria were evaluated. RESULTS Thirty-two patients were included (18 boys, 14 girls), their ages ranging from 3 years and 8 months to 14 years (mean: 8 years 2 months). The neurological examination was normal in 16 cases. In the others, a moderate disability of motor functions was observed: delay in coordination and/or equilibrium occurred in 2 cases, slight hypotonia in 1 case, macrocrania in 1 case and isolated dysfunctioning of orobuccofacial praxis in 12 patients. Neuroradiologic investigations and audiometric studies were normal in all cases. IQ was above or equal to 90 in 9 patients, ranging from 70 to 90 in the others. Language impairments were classified according to Rapin [25]. We thus diagnosed verbal or auditory agnosia when a child was unable to understand sounds in spite of normal audition and when there was no or very little production. The phonologic-syntactic syndrome was diagnosed when the abnormalities were predominant at the phonologic and syntactic levels, with a comprehension problem varying in intensity. The syntactic-lexical syndrome was diagnosed for those cases with normal comprehension but deviant syntactic production, with major trouble in word evocation. The phonologic programming deficit syndrome was defined as almost nonexistent production, with normal or subnormal comprehension. The syndrome of phonologic production (or verbal apraxia) has rarely been diagnosed, the line between it and a simple delay in phonologic production being not clear. Neuropsychologic analysis allowed the identification of 6 cases of auditory or verbal agnosia, 18 of phonologic syntactic deficit syndrome, 4 of phonologic programming deficit syndrome, 2 of lexical syntactic deficit syndrome and 2
Table I Incidence of seizures in the 32 patients with developmental dysphasias Occasional Number No of cases seizures seizures Epilepsy «3)
Verbal auditory agnosia Phonologic-syntactic deficit syndrome Lexical-syntactic deficit syndrome Phonologic programming deficit syndrome Verbal apraxia without programming deficit Total
6
4
18
9
6
3
2
2
0
0
4
3
1
0
2
1
1
0
32
19
9
4
Echenne et al: EEG and developmental dysphasias 217
of verbal apraxia. Nineteen of the 32 children had no history of seizures or epilepsy. Nine of the thirty-two had had occasional seizures (less than 3): non-recurrent febrile convulsions (I case), and generalized clonic seizures without fever once (3 cases) or twice (5 cases). Four of the thirty-two (Table 1) presented primary generalized epilepsy well controlled by AEDs. In these patients, the last seizures occurred at least 3 months before the investigations. A systematic investigation concerning the familial in· cidence of developmental dysphasia was not determined for the siblings because of the overworking it would have presented for our team. Certainly, family histories have evidenced the notion of language delay in some of our patients' brothers and sisters. However, in the absence of complete neuropsychologic investigations, these cases have not been systematically taken into account. Two of the 32 patients were sisters. The results of standard EEG are summarized in Table 2.
Four EEGs on average were recorded for each child (range: 2 to 9). Repeated standard EEGs were consistently normal in 22 children (Table 2). In 7 cases, both normal and abnormal EEGs were recorded without a relationship with the language impairment. In 3 cases, EEGs constantly showed epileptic discharges. In all cases, the epileptic discharges accounted for less than 10% of the graph. Ten of the 32 children presented interictal epileptic discharges on standard EEGs, which were inconstantly noted in 7 cases. Eight of these 10 children had had occasional ~eizures or were epileptic. Two of them had never had seizures. The results of prolonged EEGs after nocturnal sleep partial deprivation were normal in 18 of the 32 children only (Table 3). In 6 cases, they are inconstantly disturbed (auditory verbal agnosia: 1, phonologic syntactic deficit syndrome: 4, verbal apraxia: 1). In 8 cases, the recordings were constantly abnormal. The epileptic abnormalities were polymorphic: there were generalized spike-waves
Table 2 Results of standard EEGs (4 on average for each child) Number of cases Verbal auditory agnosia Phonologic-syntactic deficit syndrome Lexical-syntactic deficit syndrome Phonologic programming deficit syndrome Verbal apraxia Total
Either normal or abnormal
Always normal
Always abnormal
6
2
3
18
12
4
2
2 4 2
2 4 2
0 0 0
0 0 0
32
22
7
3
Generalized
Focal S-Wor S
2 (1)
2 (rolandic or temporal)
s-w
Focal + generalized S·W
5* (2) (rolandic 2, temporal 3)
* In 2 patients, the abnormal signs have not been identical at each control (both with normal EEGs or with generalized S-W, or with focal + generalized signs). ( ) number of cases with occasionally normal EEGs. S: spike, S-W: spike-wave.
Table 3 Results of prolonged EEGs after sleep deprivation (1 to 3 for each child) Number Always Either Always Generalized of cases normal normal or abnorl1U1[ S·W abnormal Verbal auditory agnosia Phonologic-syntactic deficit syndrome Lexical-syntactic deficit syndrome Phonologic programming deficit syndrome Verbal apraxia Total
6
3
18
9
2 4 2
1 4 1
32
18
4
Focal S-WorPS
2
2*
1 (right rolandic PS)
5
5**
1 (bitemporal)
Sleep Focal + generalized continuous signs S-W
2*
1* (right ro1andic PS) 6
8
* One of these patients had a normal recording on the second investigation, * * 3 of these patients had a normal recording on the second investigation. PS: polyspikes.
218 Brain & Development, Vol 14,No 4, 1992
(8 cases), localized spike-waves or polyspikes (right rolandic: 2 cases) or bitemporal (1 case), or continuous spike-waves during sleep (3 cases). In 6 cases, the abnormalities had already occurred during the waking state
recording; they were clearly increased during sleep in a total of 14 cases. The epileptic discharges accounted for less than 10% in 10 cases; they ranged from 10 to 20% in 1 case; they were above 85% in 3 others.
Table 4 Results of overnight polygraphic EEGs
Abnormal Continuous generalized SoW activity during sleep Generalized SoW Generalized SoW + (temporal, rolandic or occipital) Left S or SoW Right S or SoW Rolandic spikes Bilateral Bitemporal activity
Verbal agnosia
Phonologicsyntactic deficit syndrome
Syntacticlexical syndrome
Phonologic programming syndrome
Verbal apraxia
6/6 2 2
17/18 2 5
1/2
4/4
2/2
6* 3*
2
2
2 1***
*A second investigation showed the same signs in the same patient (1 patient). ** A second investigation showed generalized spike-waves with a left temporal focus in one of these 3 patients_ ***This was observed twice in the same patient. A third control showed continuous generalized spike-wave activity during sleep.
Fig 1 Variations of the abnormalities observed in the same patient (only one brief afebrile generalized seizure when he was 4 years old;
no recurrence on follow-up for 7 years; severe developmental dysphasia of the phonologic syntactic type. A) (3/1984) Very active right temporo-occipital epileptic activity with some asynchronous occipital spikes and spike-waves. B) (1/1985) Generalized polyspikes and spike-waves_ C) (2/1986) Left fronto-temporal spikes and spike-waves.
Echenne et al; EEG and developmental dysphasias 219
Therefore, after sleep deprivation 14 of the 32 patients presented interictal epileptic ,discharges in waking state or diurnal sleep EEGs. The pathologic signs were inconstantly noted in 6 cases. Of these 14 children,S had had occasional seizures, 4 were epileptic and 5 had never had seizures. The results of night polygraphic recordings are summarized in Table 4 (1 investigation per child at least, 2 investigations in 4 of them and 3 in another one). Epileptiform activity was recorded in 30 of the 32 patients, in all types of dysphasias at any age. In 4 cases, epileptic signs accounted for 85% or more of the graph, and in 6 cases, they accounted for 50 to 85% of it. In 10 patients, they accounted for 10 to 50% of the recording, and in 10 patients, for less than 10%. The results varied greatly from one case to another: generalized continuous spike-waves (4 cases), discharges of generalized spike-waves (10 cases), generalized spike-waves associated with local discharges of temporal, occipital or rolandic spike-waves or polyspikes, either right (3 cases) or left (10 cases), discharges of rhythmic bilateral rolandis spikes and polyspikes (2 cases), bitemporal polyspikes and spikewaves (1 case). When a second overall night recording was obtained in the same patient (5 cases), the features were
found to be identical in 3 cases. In a fourth patient, a left temporal focus was observed instead of the previously right temporal focus and in the fifth patient, generalized continuous spike-wave activity was recorded (bitemporal spike-waves previously). On the whole, thirty of the 32 children with developmental dysphasias exhibited clear interictal epileptic features during sleep. The morphology of interictal signs varied greatly from one case to another. Generalized discharges of 2 or 2.5 Hz associated or not with focal discharges were mainly observed. In the same patient, the frequency and localization of the epileptiform changes varied considerably according to time (Fig 1). These abnormalities clearly increased during slow sleep (Fig 2) and disappeared during REM sleep (except in one patient in which they persisted). In 4 cases only, generalized continuous spike-waves were observed. Among these 30 children, standard EEGs were consistently normal in 21 cases, and 17 of these children had never had seizures. Moreover, the overall night sleep recordings showed epileptic features in 16 children, in whom the afternoon sleep recordings were normal.
-, ,
Cz· Pz i""-v/Jw.;'fI~.~v.¥",~-#M"""~","","V'''''I'\:I''"-J'''''''-'-''''' '~IIVn\N
FP'·C.:Vw;~r~
T. · Fp'
, ,,,,
