- Email: [email protected]
Epilepsy with grand mal on awakening and sleep-waking cycle
Clinical Neurophysiology 111, Suppl. 2 (2000) S103±S110
www.elsevier.com/locate/clinph
Epilepsy with grand mal on awakening and sleep-waking cycle Dieter Janz* Department of Neurology, Virchow Klinikum 13353, Humboldt University of Berlin, Berlin, Germany 10099
Abstract Awakening epilepsy (AE) is an age related syndrome of idiopathic generalized epilepsy (IGE) characterized by generalized tonic clonic seizures (GTCS) occurring predominantly on awakening (independent of the time of day) or at leisure time (almost at evening). The GTCS can be the only symptom or they can be combined with the other subsyndromes of IGE in childhood or adolescence. The EEG shows the characteristics of IGE (generalized spike wave frequent, foca1 abnormalities rare, photosensitivity increased). The common denominator of external seizures precipitating in¯uences is lack of sleep. The sleep habits of patients with AE who could roughly be characterized as late sleepers and late risers may dispose them to a chronic sleep de®cit. Polygraphic studies indicated that their sleep is more unstable and subject to external in¯uences. Microstructural sleep analysis con®rms the presence of a disturbance of sleep stability in patients with IGE. Furthermore, it clearly shows that in the prototype of AE, the juvenile myoclonic epilepsy, the epileptiform activity during non-REM sleep is correlated with the arousal phase of the so called cyclic alternating pattern indicating that even in the smallest sleep-waking oscillations awakening is the most sensitive part. q 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Epilepsy; Sleep; Sleep-waking cycle
1. Historical perspectives A diagnostic feature long considered to differentiate epileptic from non-epileptic seizures is that the former may occur repetitively at speci®c times of the day or night. However, it has been discovered that the relation is not really to times of day or night, but rather to the sleepwaking cycle. It is also a clinical experience that such relations are more clearly seen in major than in minor seizures, and that they are quite stable characteristics in the course of the disease (Janz, 1974). Gowers (1885) ®rst mentioned patients whose attacks occurred only in the early morning and who accounted for 5% of his subjects. Langdon-Down and Brain (1929) registered 2524 major seizures of 66 institutionalized patients during a 6 month period. On their chart (Fig. 1) one can see these peaks: 1. Seizures after awakening, 2. Seizures before awakening, in the second half of the sleep, 3. Seizures after falling asleep, 4. Seizures in the afternoon, in `leisure time'. In 1931, F.L. Patry, a doctor at an American State Hospital who con®rmed the ®ndings of Langdon-Down and Brain,
already raised the question whether the different time peaks of epileptic seizures may be related to different sleep patterns of the patients (Patry, 1931). He wrote: ` Such as we have different types of patients as to their time peaks, we observe different types of sleepers: those whose curve of depth of sleep is deepest within 1 h after falling to sleep and those showing a morning type with the deepest curve occurring just before awaking. It is reasonable to correlate such characteristics of sleep with the occurrence of the epileptic time peaks'. In 1933, Helen Hopkins found that characteristic seizures peaks also in outpatients could be used to differentiate them into two groups: patients with seizures during sleep and patients with seizures during the period of awakenings (de®ned as the ®rst hour after arousal from sleep) (Hopkins, 1933). The original description of `awakening epilepsy' as an epileptic syndrome characteristic of idiopathic epilepsy with generalized tonic-clonic seizures (GTCS) goes back to 1953 (Janz, 1953). There I pointed out that many of these patients had a second seizure peak, mostly in the late afternoon or evening hours of leisure and relaxation (`Feierabend-Peak'), and that there was a small group for whom this evening peak was the main or even the only seizure peak. The temporal distribution of these variants is illustrated in Fig. 2, which shows the seizure peaks in the form of pyramids on a time line, while the incidence of the variants is expressed in numbers. The illustration
* Fax: 149-308035124. 1388-2457/00/$ - see front matter q 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S 1388-245 7(00)00409-0
CLINPH 1039
S104
D. Janz / Clinical Neurophysiology 111, Suppl. 2 (2000) S103±S110
seizures with this biorhythmic peculiarity, sometimes associated with myoclonic seizures or absences, and with generalized epileptic EEG discharges. 2. De®nitions How many seizures are required and how closely should the seizures be associated with characteristic situations to permit a diagnosis of epilepsy with GM on awakening (EGMA)? A diagnosis of epilepsy requires two or more seizures, and repeated seizures are necessary to identify a consistent association of the seizures with a certain situation. For a de®nite diagnosis of EGMA or EGMS (with GM predominantly in sleep) a minimum of 6 GTCS is required. Patients with fewer than 6 seizures may be classi®ed tentatively as having isolated seizures or `oligoepilepsy' (epilepsy with rare seizures), or a tentative diagnosis of EGMA or EGMS may be given. A diagnosis of EGMA is made when patients have had at least 6 GM seizures exclusively or predominantly after awakening (regardless of the time of the day) or while at leisure. This `predominantly' is de®ned as `more than 90% of the time' in the proposal of the Commission of Classi®cation of the International League of Epilepsy (1981). However, some investigators as we do use the general meaning of predominantly, `more than half of the time'. Fig. 1. Daily distribution of 2524 recorded major seizures registered from 66 institutionalized patients in the course of 6 months. The black entries represent the seizures of the `night' group, the paler ones the `day' group (Langdon-Down and Brain, 1929).
shows that epilepsy on awakening is sometimes accompanied by seizures during sleep, which may even predominate during the later clinical course. Because of this complexity one should for classi®catory reasons always refer to the biorhythmicity of the seizures at the beginning of disease. Loiseau (1964), who reviewed the French investigations concluded that seizures of different kinds could be related to the awakening situations, but that one syndrome of awakening epilepsy could still be de®ned by grand mal (GM)
3. Epidemiology Excluding all patients with fewer than 6 GTCS and including all patients with GTCS occurring only or predominantly after awakening, during the evening (leisure time), or both, Janz reported EGMA in 10% of patients having GTCS without and 17% with additional minor seizures (Janz, 1994) (Table 1). With increased attention, exact history taken and meticulous observation, the percentage of pure (i.e. without any absences or myoclonic jerks) EGMA cases has decreased and the percentage of cases with minor generalized seizures has increased, as a comparison between the data of 1969 (Table 1) and the data of 1989 (Schmitz and Wolf, 1989)
Fig. 2. Variation in the clinical course of 98 patients with grand mal epilepsy on awakening. The horizontal line represents the circadian rhythm of 1 day, on the left side the sleeping period, on the right side the waking period. The pyramids indicate the circadian predilection of the seizures. The height of the pyramids denotes the relative frequency of the seizures (Janz, 1962).
D. Janz / Clinical Neurophysiology 111, Suppl. 2 (2000) S103±S110 Table 1 Frequency of epilepsies with grand mal on awakening in epileptic patients (n 4816*) at a University Department of Neurology, data from 1969 (Janz, 1994) a
GMA in combination With CAE, JAE, JME Other GMA pure GMA total
N
%
812
17
508 1320
10 27
N
%
616 196
13 4
a GMA, epilepsy with grand mal on awakening; CAE, childhood absence epilepsy; JAE, juvenile absence epilepsy; JME, juvenile myoclonic epilepsy. *Without isolated and oligoepilepsies (up to 6 GM seizures).
shows (Table 2). According to the experience of Loiseau et al. (1991), cases of epilepsy with GTCS on awakening are more frequently seen in private practice than in the hospital. All subsyndromes of idiopathic generalized epilepsy (IGE) in childhood and adolescence, which are characterized by minor seizures, may combine with GTCS, usually of the awakening type. When GTCS occur, they are of the EGMA type in 96% of cases of childhood absence epilepsy, 91% of cases of juvenile absence epilepsy (Janz, 1994), and 90% of cases of juvenile myoclonic epilepsy (JME) (Janz, 1989). The proportion of pure Grand Mal epilepsy on awakening within all syndromes of IGE epilepsy was of 1:10 in the private practice of Loiseau et al. (1991), and of 1:6 in our hospital (Schmitz and Wolf, 1989). 4. Etiology EGMA as a syndrome clearly belongs to the idiopathic epilepsies, with a strong genetic predisposition. In 90% of 719 patients with EGMA, no external causes could be elicited (Janz, 1953). The same ®gure of 10% with pathogenic clues was reported by Loiseau (1964). The hereditary risk - that is, the percentage of ®rst-degree relatives who also have non-febrile epileptic seizures or epilepsies - is slightly smaller in EGMA than in the other syndromes of IGE. In pure EGMA, it is 3.3% (10/304) (Puruker, 1994); in juvenile absence epilepsy (JME) it is 4.5% (16/353); in childhood absence epilepsy it is 5.3% (17/322) (Janz, 1997); and in
S105
JME it is 5.5% (33/600) (Janz et al., 1989). These differences are not statistically signi®cant. The genetic mode is complex. Molecular genetic studies show hints to gene localization for JME on chromosome 6 p 21.3 (Greenberg et al., 1988; Sander et al., 1997a), 6 p 21 - p 11 (Liu et al., 1995), 15 q 14 (Elmslie et al., 1997; Sander et al., 1997b), and for IGE on chromosome 8 q 24 (Zara et al., 1995; Sander et al., 1998). Greenberg et al. (1995) have found the same gene localization as for JME (6 p 21) in families with pure GTCS on awakening, in contrast to pure random Grand Mal. This ®nding underlines not only the close relation between EGMA and JME, but also the biological differences between epilepsies with pure GM. The collaborative European genome search study for genetic loci of IGE using sib pairs provided signi®cant evidence for a novel IGE susceptibility locus on chromosome 3q26 and suggestive evidence for two IGE loci on chromosome 14q23 and chromosome 2q36 (Sander et al., 2000). 5. Clinical presentation 5.1. Sex distribution Epilepsy with GMA shows a male preponderance, this is more marked in pure EGMA (62.2% according to Tsuboi and Christian (1976) and 65.8% according to Puruker (1994)) than in EGMA combined with minor seizures (53% according to Goosses (1994) and 57% according to Tsuboi and Christian (1976)). 5.2. Age of onset Epilepsy with GMA is one of the age-related epilepsy syndromes. Compared with the other age-related syndromes, its age of onset has the widest range (Fig. 3). It extends from 6 to 35 years, thus encompassing the range of age of manifestation of the other age-related IGEs of childhood and adolescence (Fig. 4). The age of onset is the strongest single indicator of a biologic coherence of EGMA with the other subsyndromes of IGE, which tend to overlap in many ways and may differ only subtly in some respects.
Table 2 Syndromes of idiopathic generalized epilepsy: prevalence (%) in general practice, neurological private praxis, and neurological hospital (Janz, 1994)
GMA CAE CAE 1 JAE JAE JME a b c
England (n 592) General practice a
Bordeaux (n 263) Hospital b
Bordeaux (n 570) Private practice b
Berlin (n 699) Hospital c
± ± 2.2 ± 1.5
± 5.3 ± 2.7 5.3
2.1 10.5 ± 3.0 5.4
4.6 7.5 ± 8.5 9.2
Manford et al. (1992). Loiseau et al. (1991). Schmitz and Wolf (1989).
S106
D. Janz / Clinical Neurophysiology 111, Suppl. 2 (2000) S103±S110
purpose, we need to have a closer look at the following 3 clinical situations which correlate closely to the phenomenon of the `awakening seizures': 1. The precipitation of seizures by lack of sleep, 2. The quality or structure of sleep preceding the seizures, 3. The awakening after sleep, i.e. the process of waking up during which the seizures occur.
6.1. Sleep deprivation Fig. 3. Age onset in 88 cases of idiopathic epilepsy with pure grand mal on awakening (Janz, 1994).
5.3. EEG characteristics The EEGs of 150 patients with pure GMA or GMA with minor seizures were compared with the same number of patients with GMS, again with and without minor seizures (Christian, 1960, 1961). The two GMA groups contrasted clearly with the GMS groups: whereas 57% of the pure GMS and 23% of the combined GMS patients had a completely normal resting EEG, this was true only for 18% of patients with pure GMA and for 3% of those of combined GMA. The most frequent ®ndings in GMA where increased slow waves (76%), disorganized background activity with sleep transients (63%), and generalized spike-and-waves activity (41%). In patients with additional minor seizures, spike-and-waves were found more frequently (70%). Focal abnormalities were rare (2.6%). In contrast with other GM syndromes, GMA is positively correlated with photosensitivity (Wolf and Goosses, 1986). The EEG ®ndings, therefore, characterize epilepsies on awakening clearly as generalized epilepsies of idiopathic nature.
6. Pathogenetic considerations How is it to be understood that the seizures of idiopathic generalized epilepsies (IGE) usually occur after awakening from sleep? How does the experience of the patients, the clinicians' observations, and the neurophysiologists' analyses contribute to answering this question? For this
Fig. 4. Schematic manifestation spans and peaks in childhood absence epilepsy (CAE), juvenile absence epilepsy (JAE), juvenile myoclonic epilepsy (JME) and epilepsy with grand mal on awakening (GMA) (Janz, 1994).
Lack of sleep up to deprivation of sleep is a speci®c means to cause seizures in all forms of awakening epilepsy. One of the ®rst authors to describe them, David (1955), speaks directly of `epilepsy of the ®rst white nights' (des nuits blanches) referring as well to the beginning of this form of epilepsy at the age of the ®rst love, during which sleep is neglected. But not only at the beginning, but also during many following years, and for some people during their whole lives, the seizures are conceived as the consequence of nights after celebrations such as birthday parties, wedding parties, new year's parties, carnival days, or of nights of sicknesses or before exams, i.e. nights that are spent, happily or not, without sleep. Also the transcontinental time difference or night spent in trains, buses or cars at the beginning or the end of holidays are again and again mentioned as the causes of seizures. If we take into consideration that a number of patients do not report spontaneously of those ± either because they are ashamed, or they do not take these `diet mistakes' as dramatic, or they fear the consequences ± we can assume that almost every seizure of awakening epilepsy is caused by lack of sleep. This pathophysiological characteristic can be used diagnostically in case of unclear diagnoses, by proceeding with the EEG examination only after the patient has had a night without sleep. Although this has not been analyzed systematically, the effect of provocation by deprivation of sleep is presumably more important on an EEG of awakening epilepsies or of generalized epilepsies than on the one of other forms of epilepsy. In the case of JME, lack of sleep has been proved to be more ef®cient as a means of provocation than others, as for example hyperventilation or even photostimulation (Christian, 1960; Genton et al., 1994). The neurophysiological mechanism set off by deprivation of sleep is not clear, as it is combined with other variables like stress and hormones. Experiments with animals have not yet led to precise results. It is known that sleep-andwake rhythm can be traced back to the hypothalamus, and that the stimulation of the posterior region leads to arousal, while the stimulation of the anterior region leads to sleep. The arousal cells or waking-active neurones have direct projections to the whole neocortical mantle, as well as to the ascending reticular activating system (ARAS), which has itself important terminal in the thalamus and the cortex. One of the neurotransmitters which are produced in the cells
D. Janz / Clinical Neurophysiology 111, Suppl. 2 (2000) S103±S110
of ARAS is norepinephrine (NE). It usually has antiepileptic effects. NE is said to support also attention in the state of being awake. Therefore it can be assumed that a surge of NE release simply concentrates neuronal attention and thus could activate epileptic neocortical cell populations (Shouse and Martins da Silva, 1997). 6.2. Quality or structure of sleep When asking patients with awakening epilepsy about their sleeping habits ± having become attentive to this due to the frequent indication of lack of sleep as a provoking factor ± one often gets to hear about the same sleeping habits: that in the morning they wake up extremely slowly, that they put off getting up as long as possible and that, also when they have got up, they remain drowsy for some time. The time before noon is not one of their best time; during the afternoon and the evening, though, they are perfectly awake and ef®cient. They then tend to stay up later at night and fall asleep often only with dif®culties and after some ceremonies (TV, reading or the like). Subjectively, they get their deepest sleep towards morning. This description ± which is the result of irregularly pursued inquiries (Janz, 1969) ± corresponds to the experience likewise gained by a number of investigators who were interested in the differences between sleep epilepsy and awakening epilepsy (GaÈnshirt and Vetter, 1961; Beyer and Jovanovic, 1966; Krischek, 1962). Yet the sleeping habits or the subjective sleeping conduct of patients with sleep or awakening epilepsy, or with temporal and with generalized epilepsies, have unfortunately never been examined by systematically controlled inquiries and with standardized modes of proceeding. In 1967 Jovanovic showed sleeping curves of patients with sleep epilepsy and of patients with awakening epilepsy in comparison with healthy persons (Jovanovic, 1967a). These curves have been integrated by EEG and acoustic arousing
S107
stimulations. It becomes obvious that patients with awakening epilepsy can be woken up quite easily during the ®rst few hours of their sleep, while patients with sleep epilepsy quickly gain a deep sleep. Towards morning, the situation is rather reversed. The middle curve shows the average curve of a healthy person (Fig. 5). In the 1960s, a number of EEG investigations have been made, partly already with a polygraphic method (Christian, 1960, 1961; Jovanovic, 1967a,b,c). They revealed unanimously that patients with awakening epilepsy fall asleep with a certain delay, that during the whole sleeping time they do not gain the same depth as patients with sleep epilepsy, that the stage of greater depth is reached only after 4 to 5 h, that this stage is relatively short and that the whole sleep cycle shows a de®cit compared to that of a healthy person. These results seemed to con®rm Patry's statement of 1931 previously mentioned: `Such as we have different types of patients as to their time peaks, we observe different types of sleepers¼'. (Christian, 1960, 1961) concludes from his observations that patients with awakening GM suffer from a `chronic de®cit of sleep'. Every additional lack of sleep leads consequently to seizures by using up the `too little of sleep reserves'. This hypothesis has not been pursued, and, in later model concepts of sleep, ideas of quantity and dynamics do not play any role in general. Investigations of the 1980s, which have been made by Wolf and his colleagues, lead, however, rather to opposite results. In comparison to cases not in treatment, patients with awakening GM had a signi®cantly shorter waking time, movement time, and sleeping time in stage 1, but spent signi®cantly more time during stages 3 and 4 (RoÈder-Wanner et al., 1985). Reinvestigations of the same patients under the use of antiepileptic drugs lead, however, to the result that the sleep of patients with awakening epilepsy is clearly more unstable and more easily in¯uenced
Fig. 5. Sleep diagram of patients with sleep grand mal and grand mal on awakening in comparison to healthy persons, integrated from EEG data and acoustic arousing stimuli (Jovanovic, 1967a,b,c).
S108
D. Janz / Clinical Neurophysiology 111, Suppl. 2 (2000) S103±S110
by external circumstances than the sleep of patients with sleep epilepsy. It was found that, in patients with generalized seizures and awakening epilepsies, the early REM cycles seemed to be more prone to change than in patients with focal or sleep epilepsies; with phenobarbital, the increase of deep sleep in the ®rst cycle was much about average, de®nitely involving stage 4 rather than stage 3, and the duration of stage 2 was even diminished. With phenytoin, the two ®rst cycles became shorter owing to decrease in the duration of sleep stage 2 (Wolf et al., 1984). Both under phenobarbital (Pb) and under phenytoin, the early sleep cycles seemed to be more modi®able in the case of awakening epilepsies than of sleep epilepsies. The investigators put forward that Pb ± with which the patients fall quicker asleep, move less and wake up less often ± has an unusually positive effect on the sleep of patients with awakening epilepsy (Wolf et al., 1984). They explain the fact that they could not con®rm the macrostructural pro®le of the earlier investigators by methodological mistakes of the previous investigators (deductions of the results shortly after stopping the medication). Yet, it could also be that there is an important difference between sleep under laboratory conditions and sleep in natural circumstances, and that the process of adaptation entailed by the procedure of examinations are taken too little into consideration. The discrepancy between the self-experience of the patients and the result of the technical examination is too striking to be ignored. I will get back to the changes which are revealed by an analysis of the microstructural changes after having viewed the third biological situation, namely the process of waking up. 6.3. Becoming awake after waking up The seizures always occur after waking up, irrespectively of the patient still being in bed or already having got up. Information like `always in the bathroom' or `usually during breakfast' or `never after I have left the house' is characteristic. Often the patients remember that before the seizure, they had `not yet been fully awake', `were still dozing ' or `had been admonished by others to hurry up, to ®nally get dressed and ready'. Some remember absences or minor jerks of the sort of impulsive petit mal, the myoclonic terk in JME. Some have had the experience that they are spared from seizures when they go back to bed and fall asleep again. Others try to prevent seizures by taking their time from getting up until leaving the house in order to avoid being rushed. Their preventive device is to become awake steadily and slow. Accordingly, the contrary, i.e. to be woken up suddenly, means the certainty of a high risk of seizures. Thus, it is the transition from sleep to the state of wakefulness, the actual becoming awake, which for the awakening epilepsies is an especially critical process. The crisis becomes manifest when the process is accelerated, and it
is avoided when the process is slowed down or even repeated again. The crisis of the circadian function of awakening can especially be provoked by disturbances of the preceding sleep or by ultradiane in¯uences such as menstruation or stress, which are brought about by hormones. The bridge, however, which leads from the state of sleep to the state of being awake is unstable in patients with this form of epilepsy. Strangely enough, compared to the processes of waking up during sleep, the becoming awake after sleep has been paid little attention to by the neurophysiological side. Touchon (1982), however, was able to ascertain that the condition of the central nervous system during the transfer from sleeping to the waking state ± and conversely, although to a lesser extent ± is favourable to the production and spread of epileptic activity in awakening epilepsy. In JME, polyspike-waves occurred more frequently after night-time awakening than after morning awakening, less frequently during the relaxation period before sleep and during phase 1 sleep, seldom during REM sleep, hardly even during phase 2 sleep, and never in deep non-REM sleep. Provoked awakening results in a higher rate of discharge than spontaneous awakening. The effect was more noticeable when sleep was interrupted during the unstable phases (such as REM sleep during the early part of the night or in phase 2 at the end of the night). Induction of seizures by awakening is ± according to his ®ndings ± favoured by an unstable initial state or by a sudden transition from one to the other. A much more differentiated instrument for the analysis of transitional states than the method of abrupt arousal seems to be the observation of the cyclic alternating pattern (CAP) as it has been described by Terzano et al. (1985). The sleep analysis of patients with JME has been executed by Gigli et al. (1992). CAP consists of periodic micro-arousals of EEG desynchronization occurring on a background of prolonged homogeneous EEG synchronization (non-cyclic alternating pattern (NCAP) in all stages of NREM sleep. Gigli et al. (1992) have found that the epileptiform abnormalities are less associated with the traditional sleep stages, than with microstructural sleep conditions, i.e. the CAP, with their initial phase (A) corresponding to the peak level of EEG desynchronization, and almost never in the second phase (B), which re¯ects a return to the synchronization ± EEG NCAP state. The observation that epileptiform discharges during sleep appear mainly during the `dynamic' periods (i.e. during conditions of transition and instability) is ± according to the authors ± not incompatible with the high degree of activation that occurs after nocturnal and ®nal morning awakenings, when tendency to new sleep onset still exists. In this context, they quote the hypothesis of HalaÂsz (1984), who postulated that bi-directional ¯uctuations of arousals play an important role in the appearance of generalized EEG epileptiform discharges. With regard to the interaction of physiological and pathological processes, one could discuss also in this context the
D. Janz / Clinical Neurophysiology 111, Suppl. 2 (2000) S103±S110
possible initiation of generalized EEG discharges by Vertex-spikes and K-complexes, which are considered as the components of arousals. Furthermore, the relation of such phenomena as photosensitivity and the effect of eyelid-closure to sleep or awakening should also be taken into consideration, especially as these phenomena are clinically closely related to generalized epilepsies. And from the clinical point of view, the manifold sensorial and motor phenomena can be mentioned which can occur on the transition from being awake to sleep and from sleep to waking up, which, yet, do not have an initiating effect for epileptic seizures. At the end, the question remains: in which way can the disturbance of sleep, de®ned as the instability during the transition from sleep to arousal, be related to the genetic tendency towards generalized epilepsy? Are both the expression of a common genetic disturbance, or is it a matter of comorbidity of two genetic disturbances, the coincidence of which promote the manifestation of epilepsy? Wolf (1986) seems to favour the latter when discussing the opinion that disturbance of sleep only promotes the manifestation of the epileptic disposition which would otherwise remain latent. I do not have a clear opinion about this. Further investigations should test the habits of sleeping and awakening in standardized form, and examine the structure of sleep and waking telemetrically in natural circumstances, in order to verify ®rst the general importance of the disturbances of sleep-waking cycle as described. Then, the micro and macrostructural characteristics of the circadian cycle, also in comparison between persons with subclinical discharges and persons with de®ned types of sleeping and waking behaviour, should be examined. Even then, the answer to the question if it is nature or nurture or its interdependence will still remain on the waiting list. References Beyer I, Jovanovic UJ. Elektroencephalographische und klinische Korrelate bei Aufwachepileptikern mit besonderer BeruÈcksichtigung der therapeutischen Probleme. Nervenarzt 1966;37:333±336. Christian W. Bioelektrische Charakteristik tagesperiodisch gebundener Verlaufsformen epileptischer Erkrankungen. J Neurol 1960;181:413± 444. Christian W. Schlaf-Wach-Periodik bei Schlaf- und Aufwachepilepsien. Nervenarzt 1961;32:266±275. Commission on Classi®cation and Terminology of the ILAE. Proposal or a revised classi®cation of epilepsies and epileptic syndromes. Epilepsia 1981;30:389±399. David J. L'eÂpilepsie du reÂveil (aÁ propos de 100 observations). TheÁse, Lyon, 1955. Elmslie FV, Rees M, Williamson MP, et al. Genetic mapping of a major susceptibility locus for juvenile myoclonic epilepsy on chromosome 15q. Hum Mol Genet 1997;6:1329±1334. GaÈnshirt H, Vetter K. Schlafelektroencephalogramm und Schlaf- WachPeriodik bei Epilepsien. Nervenarzt 1961;32:275±279. Genton P, Salas Puig X, Tunon A, Lahoz CH, Del Socorro M. Juvenile myoclonic epilepsy and related syndromes: clinical and neurophysiological aspects. In: Malafosse A, Genton P, Hirsch E, Marescaux C,
S109
Broglin D, Bernasconi R, editors. Idiopathic generalized epilepsies: clinical, experimental and genetic aspects, London: John Libbey, 1994. pp. 253±265. Gigli GL, Calia MG, Mazza S, Mennuni G, Diomedi M, Terzano MG, Janz D. Sleep microstructure and EEG epileptiform activity in patients with juvenile myoclonic epilepsy. Epilepsia 1992;33:799±804. Goosses R. Die Beziehung der FotosensibilitaÈt zu den verschiedenen epileptischen Syndromen (Thesis), Berlin: Free University of Berlin, 1994. Gowers WR. Epilepsy and other chronic convulsive diseases: their causes, symptoms and treatment, London: Churchill, 1885 (New York: Dover; 1966 ± repri. Greenberg DA, Delgado-Escueta AV, Widelitz H, Sparkes RS, Treiman L, Maldonado HM, Park MS, Terasaki PI. Juvenile myoclonic epilepsy may be linked to BF and HLA loci on human chromosome 6. Am J Med Genet 1988;31:185±192. Greenberg DA, Durner M, Resor S, Rosenbaum D, Shinnar S. The genetics of idiopatic generalized epilepsies of adolescent onset: differences between juvenile myoclonic epilepsy and epilepsy with random grand mal and with awakening grand mal. Neurology 1995;45:92±946. HalaÂsz P. Sleep, arousal and electroclinical manifestations of generalized epilepsy with spike and wave pattern. In: Degen R, Niedermeyer E, editors. Epilepsy, sleep and sleep deprivation, Amsterdam: Elsevier, 1984. pp. 97±197. Hopkins H. The time of appearance of epileptic seizures in relation to age, duration and type of syndrome. J Nerv Ment Dis 1933;77:153±162. Janz D. `Aufwach'- Epilepsien (als Ausdruck einer den `Nacht- oder Schlaf'- Epilepsien gegenuÈberstehenden Verlaufsform epileptischer Erkrankungen). Arch Psychiatr Nervenkr 1953;191:73±98. Janz D. The grand mal epilepsies and the sleeping-waking cycle. Epilepsia 1962;3:69±109. Janz D. Die Epilepsien. Spezielle Pathologie und Therapie, Stuttgart: Thieme, 1969, 2nd ed. 1998. Janz D. Epilepsy and the sleeping-waking cycle. In: Vinken PJ, Bruyn GW, editors. Handbook of clinical neurology, Vol. XV. Amsterdam: North Holland, 1974. pp. 457±490. Janz D. Juvenile myoclonic epilepsy; epilepsy with impulsive petit mal. Cleveland Clin J Med 1989;56(Suppl 1):S23±S33. Janz D. Pitfalls in the diagnosis of grand mal on awakening. In: Wolf P, editor. Epileptic seizures and syndromes, London: John Libbey, 1994. pp. 213±220. Janz D. The idiopathic generalized epilepsies of childhood and adolescence. Epilepsia 1997;38:4±11. Janz D, Durner M, Beck-Mannagetta G, Pantazis G. Family studies on genetics of juvenile myoclonic epilepsy (epilepsy with impulsive petit mal). In: Beck-Mannagetta G, Anderson VE, Doose H, Janz D, editors. Genetics of the epilepsies, Berlin: Springer, 1989. pp. 43±52. Jovanovic UJ. Schlaf und Epilepsie. Therapiewoche 1967a;17:95. Jovanovic UJ. Das Schlafverhalten der Epileptiker. I. Schlafdauer, Schlaftiefe und Besonderheiten der Schlafperiodik. Dtsch Z Nervenheilk. J Neurol 1967b;190:159±198. Jovanovic UJ. Das Schlafverhalten der Epileptiker. II. Elemente des EEG, Vegetativum und Motorik. Dtsch Z Nervenheilk. J Neurol 1967c;191:257±290. Krischek J. Die Schlaf-wach- Periodik der grand mal-Epilepsie und die sich daraus ergebenden Konsequenzon. Dtsch Med Wschr 1962;87:49. Langdon-Down M, Brain WR. Time of day in relation to convulsions in epilepsy. Lancet 1929;I:1029±1032. Liu AW, Delgado-Escueta AV, Serratosa JM, Alonso ME, Medina MT, Gee MN, Cordova S, Zhao HZ, Spellmann JM, Ramos Peek JR, RubioDonnadieu F, Sparkers RS. Juvenile myoclonic epilepsy locus in chromosome 6 p21.2-p11: linkage to convulsion and electroencephalographie trait. Am J Hum Genet 1995;57:368±381. Loiseau P. Crises eÂpileptiques survenant au reÂveil et eÂpilepsie du reÂveil. Sud MeÂd Chirurgical 1964;99:11492±11502. Loiseau P, Duchet B, Loiseau J. Classi®cation of epilepsies and epileptic
S110
D. Janz / Clinical Neurophysiology 111, Suppl. 2 (2000) S103±S110
syndromes in two different samples of patients. Epilepsia 1991;32:303± 309. Manford H, Harz YM, Sander JW, Shorvon SD. The national general practice study of epilepsy: the syndromic classi®cation of the International League Against Epilepsy applied to epilepsy in a general population. Arch Neurol 1992;49:801±808. Patry FL. The relation of time of day, sleep and other factors to the incidence of epileptic seizures. Am J Psychiatry 1931;10:789±813. Puruker H. Klinik und Klinische Genetik der Epilepsie mit reinem Aufwach- Grand mal (Thesis), Berlin: Free University of Berlin, 1994. RoÈder-Wanner UU, Wolf P, Danninger T. Are sleep patterns in epileptic patients correlated with their type of epilepsy? In: Martins da Silva A, Binnie CD, Meinhardt H, editors. Biorhythms and epilepsy, New York: Raven Press, 1985. pp. 109±119. Sander T, Bockenkamp B, Hildmann T, Blasczyk R, Kretz R, Wienker TF, Voltz A, Schmitz B, Beck-Mannagetta G, Riess O, Epplen JT, Janz D, Ziegler A. Re®ned mapping of the epilepsy susceptibility locus EJM 1 on chromosome 6. Neurology 1997a;49:842±847. Sander T, Kretz R, Williamson MP, Elmslie FV, Rees M, Hildmann T, Bianchi A, Bauer G, Sailer U, Scaramelli A, Schmitz B, Gardiner RM, Janz D, Beck-Mannagetta G. Linkage analysis between idiopathic generalized epilepsies and the GABA a receptor a5, b3 and g3 subunit gene cluster on chromosome 15. Acta Neurol Scand 1997b;96:1±7. Sander T, Kretz R, Schultz H, Sailer U, Bauer G, Scaramelli A, Epplen JT, Riess O, Janz D. Replication analysis of a putative susceptibility locus (EGI) for idiopathic generalized epilepsies on chromosome 8q 24. Epilepsia 1998;39:715±720. Sander T, Schultz H, Saar K, Gennaro E, Riggio MC, Bianchi A, Zara F, Luna D, Bulteau C, Kamminska A, Ville D, Cienta C, Picard F,
Prud'homme J-F, Bate L, Sundquist A, Gardiner RM, Janssen GAMA, de Haan G-J, Kasteleijn-Nolst-Trenite DGA, Bader A, Lindhout D, Riess O, Wienker TF, Janz D, Reis A. Genome search for susceptibility loci of common idiopathic generalised epilepsies. Hum Mol Genet 2000;9:1465±1472. Schmitz B, Wolf P. Epilepsies and epileptic syndromes in a neurological seizure clinic. In: Beck-Mannagetta G, Anderson VE, Doose H, Janz D, editors. Genetics of epilepsies, New York: Springer, 1989. pp. 19±22. Shouse MN, Martins da Silva A. Chronobiology. In: Engel J, Pedly TA, editors. Epilepsy. A comprehensive textbook, Vol. 2. Philadelphia, PA: Lippincott-Raven, 1997. pp. 1917±1928. Terzano MG, Mancia D, Salati MR, Constani G, Decembrino A, Parrino L. The cyclic alternating pattern as a physiologic component of normal NREM sleep. Sleep 1985;8:137±145. Touchon J. Effect on awakening on epileptic activity in primary generalized myoclonic epilepsy. In: Sterman MB, Shouse MN, Passouant P, editors. Sleep and epilepsy, New York: Academic Press, 1982. pp. 239±248. Tsuboi T, Christian W. Epilepsy: a clinical electroencephalographie and statistical study of 466 patients, New York: Springer, 1976. Wolf P. Relations of epilepsy and the sleep and waking cycle. J Auton Nerv Syst 1986:603±610. Wolf P, Goosses R. Relation of photosensitivity to epileptic syndromes. J Neurol Neurosurg Psychiatry 1986;49:1386±1391. Wolf P, RoÈder-Wanner UU, Brede M. In¯uence of therapeutic phenobarbital and phenytoin medication on the polygraphic sleep of patients with epilepsy. Epilepsia 1984;25:467±475. Zara F, Bianchi A, Avanzini G, Di Donato S, Castellotti B, Patel PI, Pandolfo M. Mapping of genes predisposing to idiopathic generalized epilepsy. Hum Mol Genet 1995;4:1201±1207.
www.elsevier.com/locate/clinph
Epilepsy with grand mal on awakening and sleep-waking cycle Dieter Janz* Department of Neurology, Virchow Klinikum 13353, Humboldt University of Berlin, Berlin, Germany 10099
Abstract Awakening epilepsy (AE) is an age related syndrome of idiopathic generalized epilepsy (IGE) characterized by generalized tonic clonic seizures (GTCS) occurring predominantly on awakening (independent of the time of day) or at leisure time (almost at evening). The GTCS can be the only symptom or they can be combined with the other subsyndromes of IGE in childhood or adolescence. The EEG shows the characteristics of IGE (generalized spike wave frequent, foca1 abnormalities rare, photosensitivity increased). The common denominator of external seizures precipitating in¯uences is lack of sleep. The sleep habits of patients with AE who could roughly be characterized as late sleepers and late risers may dispose them to a chronic sleep de®cit. Polygraphic studies indicated that their sleep is more unstable and subject to external in¯uences. Microstructural sleep analysis con®rms the presence of a disturbance of sleep stability in patients with IGE. Furthermore, it clearly shows that in the prototype of AE, the juvenile myoclonic epilepsy, the epileptiform activity during non-REM sleep is correlated with the arousal phase of the so called cyclic alternating pattern indicating that even in the smallest sleep-waking oscillations awakening is the most sensitive part. q 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Epilepsy; Sleep; Sleep-waking cycle
1. Historical perspectives A diagnostic feature long considered to differentiate epileptic from non-epileptic seizures is that the former may occur repetitively at speci®c times of the day or night. However, it has been discovered that the relation is not really to times of day or night, but rather to the sleepwaking cycle. It is also a clinical experience that such relations are more clearly seen in major than in minor seizures, and that they are quite stable characteristics in the course of the disease (Janz, 1974). Gowers (1885) ®rst mentioned patients whose attacks occurred only in the early morning and who accounted for 5% of his subjects. Langdon-Down and Brain (1929) registered 2524 major seizures of 66 institutionalized patients during a 6 month period. On their chart (Fig. 1) one can see these peaks: 1. Seizures after awakening, 2. Seizures before awakening, in the second half of the sleep, 3. Seizures after falling asleep, 4. Seizures in the afternoon, in `leisure time'. In 1931, F.L. Patry, a doctor at an American State Hospital who con®rmed the ®ndings of Langdon-Down and Brain,
already raised the question whether the different time peaks of epileptic seizures may be related to different sleep patterns of the patients (Patry, 1931). He wrote: ` Such as we have different types of patients as to their time peaks, we observe different types of sleepers: those whose curve of depth of sleep is deepest within 1 h after falling to sleep and those showing a morning type with the deepest curve occurring just before awaking. It is reasonable to correlate such characteristics of sleep with the occurrence of the epileptic time peaks'. In 1933, Helen Hopkins found that characteristic seizures peaks also in outpatients could be used to differentiate them into two groups: patients with seizures during sleep and patients with seizures during the period of awakenings (de®ned as the ®rst hour after arousal from sleep) (Hopkins, 1933). The original description of `awakening epilepsy' as an epileptic syndrome characteristic of idiopathic epilepsy with generalized tonic-clonic seizures (GTCS) goes back to 1953 (Janz, 1953). There I pointed out that many of these patients had a second seizure peak, mostly in the late afternoon or evening hours of leisure and relaxation (`Feierabend-Peak'), and that there was a small group for whom this evening peak was the main or even the only seizure peak. The temporal distribution of these variants is illustrated in Fig. 2, which shows the seizure peaks in the form of pyramids on a time line, while the incidence of the variants is expressed in numbers. The illustration
* Fax: 149-308035124. 1388-2457/00/$ - see front matter q 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S 1388-245 7(00)00409-0
CLINPH 1039
S104
D. Janz / Clinical Neurophysiology 111, Suppl. 2 (2000) S103±S110
seizures with this biorhythmic peculiarity, sometimes associated with myoclonic seizures or absences, and with generalized epileptic EEG discharges. 2. De®nitions How many seizures are required and how closely should the seizures be associated with characteristic situations to permit a diagnosis of epilepsy with GM on awakening (EGMA)? A diagnosis of epilepsy requires two or more seizures, and repeated seizures are necessary to identify a consistent association of the seizures with a certain situation. For a de®nite diagnosis of EGMA or EGMS (with GM predominantly in sleep) a minimum of 6 GTCS is required. Patients with fewer than 6 seizures may be classi®ed tentatively as having isolated seizures or `oligoepilepsy' (epilepsy with rare seizures), or a tentative diagnosis of EGMA or EGMS may be given. A diagnosis of EGMA is made when patients have had at least 6 GM seizures exclusively or predominantly after awakening (regardless of the time of the day) or while at leisure. This `predominantly' is de®ned as `more than 90% of the time' in the proposal of the Commission of Classi®cation of the International League of Epilepsy (1981). However, some investigators as we do use the general meaning of predominantly, `more than half of the time'. Fig. 1. Daily distribution of 2524 recorded major seizures registered from 66 institutionalized patients in the course of 6 months. The black entries represent the seizures of the `night' group, the paler ones the `day' group (Langdon-Down and Brain, 1929).
shows that epilepsy on awakening is sometimes accompanied by seizures during sleep, which may even predominate during the later clinical course. Because of this complexity one should for classi®catory reasons always refer to the biorhythmicity of the seizures at the beginning of disease. Loiseau (1964), who reviewed the French investigations concluded that seizures of different kinds could be related to the awakening situations, but that one syndrome of awakening epilepsy could still be de®ned by grand mal (GM)
3. Epidemiology Excluding all patients with fewer than 6 GTCS and including all patients with GTCS occurring only or predominantly after awakening, during the evening (leisure time), or both, Janz reported EGMA in 10% of patients having GTCS without and 17% with additional minor seizures (Janz, 1994) (Table 1). With increased attention, exact history taken and meticulous observation, the percentage of pure (i.e. without any absences or myoclonic jerks) EGMA cases has decreased and the percentage of cases with minor generalized seizures has increased, as a comparison between the data of 1969 (Table 1) and the data of 1989 (Schmitz and Wolf, 1989)
Fig. 2. Variation in the clinical course of 98 patients with grand mal epilepsy on awakening. The horizontal line represents the circadian rhythm of 1 day, on the left side the sleeping period, on the right side the waking period. The pyramids indicate the circadian predilection of the seizures. The height of the pyramids denotes the relative frequency of the seizures (Janz, 1962).
D. Janz / Clinical Neurophysiology 111, Suppl. 2 (2000) S103±S110 Table 1 Frequency of epilepsies with grand mal on awakening in epileptic patients (n 4816*) at a University Department of Neurology, data from 1969 (Janz, 1994) a
GMA in combination With CAE, JAE, JME Other GMA pure GMA total
N
%
812
17
508 1320
10 27
N
%
616 196
13 4
a GMA, epilepsy with grand mal on awakening; CAE, childhood absence epilepsy; JAE, juvenile absence epilepsy; JME, juvenile myoclonic epilepsy. *Without isolated and oligoepilepsies (up to 6 GM seizures).
shows (Table 2). According to the experience of Loiseau et al. (1991), cases of epilepsy with GTCS on awakening are more frequently seen in private practice than in the hospital. All subsyndromes of idiopathic generalized epilepsy (IGE) in childhood and adolescence, which are characterized by minor seizures, may combine with GTCS, usually of the awakening type. When GTCS occur, they are of the EGMA type in 96% of cases of childhood absence epilepsy, 91% of cases of juvenile absence epilepsy (Janz, 1994), and 90% of cases of juvenile myoclonic epilepsy (JME) (Janz, 1989). The proportion of pure Grand Mal epilepsy on awakening within all syndromes of IGE epilepsy was of 1:10 in the private practice of Loiseau et al. (1991), and of 1:6 in our hospital (Schmitz and Wolf, 1989). 4. Etiology EGMA as a syndrome clearly belongs to the idiopathic epilepsies, with a strong genetic predisposition. In 90% of 719 patients with EGMA, no external causes could be elicited (Janz, 1953). The same ®gure of 10% with pathogenic clues was reported by Loiseau (1964). The hereditary risk - that is, the percentage of ®rst-degree relatives who also have non-febrile epileptic seizures or epilepsies - is slightly smaller in EGMA than in the other syndromes of IGE. In pure EGMA, it is 3.3% (10/304) (Puruker, 1994); in juvenile absence epilepsy (JME) it is 4.5% (16/353); in childhood absence epilepsy it is 5.3% (17/322) (Janz, 1997); and in
S105
JME it is 5.5% (33/600) (Janz et al., 1989). These differences are not statistically signi®cant. The genetic mode is complex. Molecular genetic studies show hints to gene localization for JME on chromosome 6 p 21.3 (Greenberg et al., 1988; Sander et al., 1997a), 6 p 21 - p 11 (Liu et al., 1995), 15 q 14 (Elmslie et al., 1997; Sander et al., 1997b), and for IGE on chromosome 8 q 24 (Zara et al., 1995; Sander et al., 1998). Greenberg et al. (1995) have found the same gene localization as for JME (6 p 21) in families with pure GTCS on awakening, in contrast to pure random Grand Mal. This ®nding underlines not only the close relation between EGMA and JME, but also the biological differences between epilepsies with pure GM. The collaborative European genome search study for genetic loci of IGE using sib pairs provided signi®cant evidence for a novel IGE susceptibility locus on chromosome 3q26 and suggestive evidence for two IGE loci on chromosome 14q23 and chromosome 2q36 (Sander et al., 2000). 5. Clinical presentation 5.1. Sex distribution Epilepsy with GMA shows a male preponderance, this is more marked in pure EGMA (62.2% according to Tsuboi and Christian (1976) and 65.8% according to Puruker (1994)) than in EGMA combined with minor seizures (53% according to Goosses (1994) and 57% according to Tsuboi and Christian (1976)). 5.2. Age of onset Epilepsy with GMA is one of the age-related epilepsy syndromes. Compared with the other age-related syndromes, its age of onset has the widest range (Fig. 3). It extends from 6 to 35 years, thus encompassing the range of age of manifestation of the other age-related IGEs of childhood and adolescence (Fig. 4). The age of onset is the strongest single indicator of a biologic coherence of EGMA with the other subsyndromes of IGE, which tend to overlap in many ways and may differ only subtly in some respects.
Table 2 Syndromes of idiopathic generalized epilepsy: prevalence (%) in general practice, neurological private praxis, and neurological hospital (Janz, 1994)
GMA CAE CAE 1 JAE JAE JME a b c
England (n 592) General practice a
Bordeaux (n 263) Hospital b
Bordeaux (n 570) Private practice b
Berlin (n 699) Hospital c
± ± 2.2 ± 1.5
± 5.3 ± 2.7 5.3
2.1 10.5 ± 3.0 5.4
4.6 7.5 ± 8.5 9.2
Manford et al. (1992). Loiseau et al. (1991). Schmitz and Wolf (1989).
S106
D. Janz / Clinical Neurophysiology 111, Suppl. 2 (2000) S103±S110
purpose, we need to have a closer look at the following 3 clinical situations which correlate closely to the phenomenon of the `awakening seizures': 1. The precipitation of seizures by lack of sleep, 2. The quality or structure of sleep preceding the seizures, 3. The awakening after sleep, i.e. the process of waking up during which the seizures occur.
6.1. Sleep deprivation Fig. 3. Age onset in 88 cases of idiopathic epilepsy with pure grand mal on awakening (Janz, 1994).
5.3. EEG characteristics The EEGs of 150 patients with pure GMA or GMA with minor seizures were compared with the same number of patients with GMS, again with and without minor seizures (Christian, 1960, 1961). The two GMA groups contrasted clearly with the GMS groups: whereas 57% of the pure GMS and 23% of the combined GMS patients had a completely normal resting EEG, this was true only for 18% of patients with pure GMA and for 3% of those of combined GMA. The most frequent ®ndings in GMA where increased slow waves (76%), disorganized background activity with sleep transients (63%), and generalized spike-and-waves activity (41%). In patients with additional minor seizures, spike-and-waves were found more frequently (70%). Focal abnormalities were rare (2.6%). In contrast with other GM syndromes, GMA is positively correlated with photosensitivity (Wolf and Goosses, 1986). The EEG ®ndings, therefore, characterize epilepsies on awakening clearly as generalized epilepsies of idiopathic nature.
6. Pathogenetic considerations How is it to be understood that the seizures of idiopathic generalized epilepsies (IGE) usually occur after awakening from sleep? How does the experience of the patients, the clinicians' observations, and the neurophysiologists' analyses contribute to answering this question? For this
Fig. 4. Schematic manifestation spans and peaks in childhood absence epilepsy (CAE), juvenile absence epilepsy (JAE), juvenile myoclonic epilepsy (JME) and epilepsy with grand mal on awakening (GMA) (Janz, 1994).
Lack of sleep up to deprivation of sleep is a speci®c means to cause seizures in all forms of awakening epilepsy. One of the ®rst authors to describe them, David (1955), speaks directly of `epilepsy of the ®rst white nights' (des nuits blanches) referring as well to the beginning of this form of epilepsy at the age of the ®rst love, during which sleep is neglected. But not only at the beginning, but also during many following years, and for some people during their whole lives, the seizures are conceived as the consequence of nights after celebrations such as birthday parties, wedding parties, new year's parties, carnival days, or of nights of sicknesses or before exams, i.e. nights that are spent, happily or not, without sleep. Also the transcontinental time difference or night spent in trains, buses or cars at the beginning or the end of holidays are again and again mentioned as the causes of seizures. If we take into consideration that a number of patients do not report spontaneously of those ± either because they are ashamed, or they do not take these `diet mistakes' as dramatic, or they fear the consequences ± we can assume that almost every seizure of awakening epilepsy is caused by lack of sleep. This pathophysiological characteristic can be used diagnostically in case of unclear diagnoses, by proceeding with the EEG examination only after the patient has had a night without sleep. Although this has not been analyzed systematically, the effect of provocation by deprivation of sleep is presumably more important on an EEG of awakening epilepsies or of generalized epilepsies than on the one of other forms of epilepsy. In the case of JME, lack of sleep has been proved to be more ef®cient as a means of provocation than others, as for example hyperventilation or even photostimulation (Christian, 1960; Genton et al., 1994). The neurophysiological mechanism set off by deprivation of sleep is not clear, as it is combined with other variables like stress and hormones. Experiments with animals have not yet led to precise results. It is known that sleep-andwake rhythm can be traced back to the hypothalamus, and that the stimulation of the posterior region leads to arousal, while the stimulation of the anterior region leads to sleep. The arousal cells or waking-active neurones have direct projections to the whole neocortical mantle, as well as to the ascending reticular activating system (ARAS), which has itself important terminal in the thalamus and the cortex. One of the neurotransmitters which are produced in the cells
D. Janz / Clinical Neurophysiology 111, Suppl. 2 (2000) S103±S110
of ARAS is norepinephrine (NE). It usually has antiepileptic effects. NE is said to support also attention in the state of being awake. Therefore it can be assumed that a surge of NE release simply concentrates neuronal attention and thus could activate epileptic neocortical cell populations (Shouse and Martins da Silva, 1997). 6.2. Quality or structure of sleep When asking patients with awakening epilepsy about their sleeping habits ± having become attentive to this due to the frequent indication of lack of sleep as a provoking factor ± one often gets to hear about the same sleeping habits: that in the morning they wake up extremely slowly, that they put off getting up as long as possible and that, also when they have got up, they remain drowsy for some time. The time before noon is not one of their best time; during the afternoon and the evening, though, they are perfectly awake and ef®cient. They then tend to stay up later at night and fall asleep often only with dif®culties and after some ceremonies (TV, reading or the like). Subjectively, they get their deepest sleep towards morning. This description ± which is the result of irregularly pursued inquiries (Janz, 1969) ± corresponds to the experience likewise gained by a number of investigators who were interested in the differences between sleep epilepsy and awakening epilepsy (GaÈnshirt and Vetter, 1961; Beyer and Jovanovic, 1966; Krischek, 1962). Yet the sleeping habits or the subjective sleeping conduct of patients with sleep or awakening epilepsy, or with temporal and with generalized epilepsies, have unfortunately never been examined by systematically controlled inquiries and with standardized modes of proceeding. In 1967 Jovanovic showed sleeping curves of patients with sleep epilepsy and of patients with awakening epilepsy in comparison with healthy persons (Jovanovic, 1967a). These curves have been integrated by EEG and acoustic arousing
S107
stimulations. It becomes obvious that patients with awakening epilepsy can be woken up quite easily during the ®rst few hours of their sleep, while patients with sleep epilepsy quickly gain a deep sleep. Towards morning, the situation is rather reversed. The middle curve shows the average curve of a healthy person (Fig. 5). In the 1960s, a number of EEG investigations have been made, partly already with a polygraphic method (Christian, 1960, 1961; Jovanovic, 1967a,b,c). They revealed unanimously that patients with awakening epilepsy fall asleep with a certain delay, that during the whole sleeping time they do not gain the same depth as patients with sleep epilepsy, that the stage of greater depth is reached only after 4 to 5 h, that this stage is relatively short and that the whole sleep cycle shows a de®cit compared to that of a healthy person. These results seemed to con®rm Patry's statement of 1931 previously mentioned: `Such as we have different types of patients as to their time peaks, we observe different types of sleepers¼'. (Christian, 1960, 1961) concludes from his observations that patients with awakening GM suffer from a `chronic de®cit of sleep'. Every additional lack of sleep leads consequently to seizures by using up the `too little of sleep reserves'. This hypothesis has not been pursued, and, in later model concepts of sleep, ideas of quantity and dynamics do not play any role in general. Investigations of the 1980s, which have been made by Wolf and his colleagues, lead, however, rather to opposite results. In comparison to cases not in treatment, patients with awakening GM had a signi®cantly shorter waking time, movement time, and sleeping time in stage 1, but spent signi®cantly more time during stages 3 and 4 (RoÈder-Wanner et al., 1985). Reinvestigations of the same patients under the use of antiepileptic drugs lead, however, to the result that the sleep of patients with awakening epilepsy is clearly more unstable and more easily in¯uenced
Fig. 5. Sleep diagram of patients with sleep grand mal and grand mal on awakening in comparison to healthy persons, integrated from EEG data and acoustic arousing stimuli (Jovanovic, 1967a,b,c).
S108
D. Janz / Clinical Neurophysiology 111, Suppl. 2 (2000) S103±S110
by external circumstances than the sleep of patients with sleep epilepsy. It was found that, in patients with generalized seizures and awakening epilepsies, the early REM cycles seemed to be more prone to change than in patients with focal or sleep epilepsies; with phenobarbital, the increase of deep sleep in the ®rst cycle was much about average, de®nitely involving stage 4 rather than stage 3, and the duration of stage 2 was even diminished. With phenytoin, the two ®rst cycles became shorter owing to decrease in the duration of sleep stage 2 (Wolf et al., 1984). Both under phenobarbital (Pb) and under phenytoin, the early sleep cycles seemed to be more modi®able in the case of awakening epilepsies than of sleep epilepsies. The investigators put forward that Pb ± with which the patients fall quicker asleep, move less and wake up less often ± has an unusually positive effect on the sleep of patients with awakening epilepsy (Wolf et al., 1984). They explain the fact that they could not con®rm the macrostructural pro®le of the earlier investigators by methodological mistakes of the previous investigators (deductions of the results shortly after stopping the medication). Yet, it could also be that there is an important difference between sleep under laboratory conditions and sleep in natural circumstances, and that the process of adaptation entailed by the procedure of examinations are taken too little into consideration. The discrepancy between the self-experience of the patients and the result of the technical examination is too striking to be ignored. I will get back to the changes which are revealed by an analysis of the microstructural changes after having viewed the third biological situation, namely the process of waking up. 6.3. Becoming awake after waking up The seizures always occur after waking up, irrespectively of the patient still being in bed or already having got up. Information like `always in the bathroom' or `usually during breakfast' or `never after I have left the house' is characteristic. Often the patients remember that before the seizure, they had `not yet been fully awake', `were still dozing ' or `had been admonished by others to hurry up, to ®nally get dressed and ready'. Some remember absences or minor jerks of the sort of impulsive petit mal, the myoclonic terk in JME. Some have had the experience that they are spared from seizures when they go back to bed and fall asleep again. Others try to prevent seizures by taking their time from getting up until leaving the house in order to avoid being rushed. Their preventive device is to become awake steadily and slow. Accordingly, the contrary, i.e. to be woken up suddenly, means the certainty of a high risk of seizures. Thus, it is the transition from sleep to the state of wakefulness, the actual becoming awake, which for the awakening epilepsies is an especially critical process. The crisis becomes manifest when the process is accelerated, and it
is avoided when the process is slowed down or even repeated again. The crisis of the circadian function of awakening can especially be provoked by disturbances of the preceding sleep or by ultradiane in¯uences such as menstruation or stress, which are brought about by hormones. The bridge, however, which leads from the state of sleep to the state of being awake is unstable in patients with this form of epilepsy. Strangely enough, compared to the processes of waking up during sleep, the becoming awake after sleep has been paid little attention to by the neurophysiological side. Touchon (1982), however, was able to ascertain that the condition of the central nervous system during the transfer from sleeping to the waking state ± and conversely, although to a lesser extent ± is favourable to the production and spread of epileptic activity in awakening epilepsy. In JME, polyspike-waves occurred more frequently after night-time awakening than after morning awakening, less frequently during the relaxation period before sleep and during phase 1 sleep, seldom during REM sleep, hardly even during phase 2 sleep, and never in deep non-REM sleep. Provoked awakening results in a higher rate of discharge than spontaneous awakening. The effect was more noticeable when sleep was interrupted during the unstable phases (such as REM sleep during the early part of the night or in phase 2 at the end of the night). Induction of seizures by awakening is ± according to his ®ndings ± favoured by an unstable initial state or by a sudden transition from one to the other. A much more differentiated instrument for the analysis of transitional states than the method of abrupt arousal seems to be the observation of the cyclic alternating pattern (CAP) as it has been described by Terzano et al. (1985). The sleep analysis of patients with JME has been executed by Gigli et al. (1992). CAP consists of periodic micro-arousals of EEG desynchronization occurring on a background of prolonged homogeneous EEG synchronization (non-cyclic alternating pattern (NCAP) in all stages of NREM sleep. Gigli et al. (1992) have found that the epileptiform abnormalities are less associated with the traditional sleep stages, than with microstructural sleep conditions, i.e. the CAP, with their initial phase (A) corresponding to the peak level of EEG desynchronization, and almost never in the second phase (B), which re¯ects a return to the synchronization ± EEG NCAP state. The observation that epileptiform discharges during sleep appear mainly during the `dynamic' periods (i.e. during conditions of transition and instability) is ± according to the authors ± not incompatible with the high degree of activation that occurs after nocturnal and ®nal morning awakenings, when tendency to new sleep onset still exists. In this context, they quote the hypothesis of HalaÂsz (1984), who postulated that bi-directional ¯uctuations of arousals play an important role in the appearance of generalized EEG epileptiform discharges. With regard to the interaction of physiological and pathological processes, one could discuss also in this context the
D. Janz / Clinical Neurophysiology 111, Suppl. 2 (2000) S103±S110
possible initiation of generalized EEG discharges by Vertex-spikes and K-complexes, which are considered as the components of arousals. Furthermore, the relation of such phenomena as photosensitivity and the effect of eyelid-closure to sleep or awakening should also be taken into consideration, especially as these phenomena are clinically closely related to generalized epilepsies. And from the clinical point of view, the manifold sensorial and motor phenomena can be mentioned which can occur on the transition from being awake to sleep and from sleep to waking up, which, yet, do not have an initiating effect for epileptic seizures. At the end, the question remains: in which way can the disturbance of sleep, de®ned as the instability during the transition from sleep to arousal, be related to the genetic tendency towards generalized epilepsy? Are both the expression of a common genetic disturbance, or is it a matter of comorbidity of two genetic disturbances, the coincidence of which promote the manifestation of epilepsy? Wolf (1986) seems to favour the latter when discussing the opinion that disturbance of sleep only promotes the manifestation of the epileptic disposition which would otherwise remain latent. I do not have a clear opinion about this. Further investigations should test the habits of sleeping and awakening in standardized form, and examine the structure of sleep and waking telemetrically in natural circumstances, in order to verify ®rst the general importance of the disturbances of sleep-waking cycle as described. Then, the micro and macrostructural characteristics of the circadian cycle, also in comparison between persons with subclinical discharges and persons with de®ned types of sleeping and waking behaviour, should be examined. Even then, the answer to the question if it is nature or nurture or its interdependence will still remain on the waiting list. References Beyer I, Jovanovic UJ. Elektroencephalographische und klinische Korrelate bei Aufwachepileptikern mit besonderer BeruÈcksichtigung der therapeutischen Probleme. Nervenarzt 1966;37:333±336. Christian W. Bioelektrische Charakteristik tagesperiodisch gebundener Verlaufsformen epileptischer Erkrankungen. J Neurol 1960;181:413± 444. Christian W. Schlaf-Wach-Periodik bei Schlaf- und Aufwachepilepsien. Nervenarzt 1961;32:266±275. Commission on Classi®cation and Terminology of the ILAE. Proposal or a revised classi®cation of epilepsies and epileptic syndromes. Epilepsia 1981;30:389±399. David J. L'eÂpilepsie du reÂveil (aÁ propos de 100 observations). TheÁse, Lyon, 1955. Elmslie FV, Rees M, Williamson MP, et al. Genetic mapping of a major susceptibility locus for juvenile myoclonic epilepsy on chromosome 15q. Hum Mol Genet 1997;6:1329±1334. GaÈnshirt H, Vetter K. Schlafelektroencephalogramm und Schlaf- WachPeriodik bei Epilepsien. Nervenarzt 1961;32:275±279. Genton P, Salas Puig X, Tunon A, Lahoz CH, Del Socorro M. Juvenile myoclonic epilepsy and related syndromes: clinical and neurophysiological aspects. In: Malafosse A, Genton P, Hirsch E, Marescaux C,
S109
Broglin D, Bernasconi R, editors. Idiopathic generalized epilepsies: clinical, experimental and genetic aspects, London: John Libbey, 1994. pp. 253±265. Gigli GL, Calia MG, Mazza S, Mennuni G, Diomedi M, Terzano MG, Janz D. Sleep microstructure and EEG epileptiform activity in patients with juvenile myoclonic epilepsy. Epilepsia 1992;33:799±804. Goosses R. Die Beziehung der FotosensibilitaÈt zu den verschiedenen epileptischen Syndromen (Thesis), Berlin: Free University of Berlin, 1994. Gowers WR. Epilepsy and other chronic convulsive diseases: their causes, symptoms and treatment, London: Churchill, 1885 (New York: Dover; 1966 ± repri. Greenberg DA, Delgado-Escueta AV, Widelitz H, Sparkes RS, Treiman L, Maldonado HM, Park MS, Terasaki PI. Juvenile myoclonic epilepsy may be linked to BF and HLA loci on human chromosome 6. Am J Med Genet 1988;31:185±192. Greenberg DA, Durner M, Resor S, Rosenbaum D, Shinnar S. The genetics of idiopatic generalized epilepsies of adolescent onset: differences between juvenile myoclonic epilepsy and epilepsy with random grand mal and with awakening grand mal. Neurology 1995;45:92±946. HalaÂsz P. Sleep, arousal and electroclinical manifestations of generalized epilepsy with spike and wave pattern. In: Degen R, Niedermeyer E, editors. Epilepsy, sleep and sleep deprivation, Amsterdam: Elsevier, 1984. pp. 97±197. Hopkins H. The time of appearance of epileptic seizures in relation to age, duration and type of syndrome. J Nerv Ment Dis 1933;77:153±162. Janz D. `Aufwach'- Epilepsien (als Ausdruck einer den `Nacht- oder Schlaf'- Epilepsien gegenuÈberstehenden Verlaufsform epileptischer Erkrankungen). Arch Psychiatr Nervenkr 1953;191:73±98. Janz D. The grand mal epilepsies and the sleeping-waking cycle. Epilepsia 1962;3:69±109. Janz D. Die Epilepsien. Spezielle Pathologie und Therapie, Stuttgart: Thieme, 1969, 2nd ed. 1998. Janz D. Epilepsy and the sleeping-waking cycle. In: Vinken PJ, Bruyn GW, editors. Handbook of clinical neurology, Vol. XV. Amsterdam: North Holland, 1974. pp. 457±490. Janz D. Juvenile myoclonic epilepsy; epilepsy with impulsive petit mal. Cleveland Clin J Med 1989;56(Suppl 1):S23±S33. Janz D. Pitfalls in the diagnosis of grand mal on awakening. In: Wolf P, editor. Epileptic seizures and syndromes, London: John Libbey, 1994. pp. 213±220. Janz D. The idiopathic generalized epilepsies of childhood and adolescence. Epilepsia 1997;38:4±11. Janz D, Durner M, Beck-Mannagetta G, Pantazis G. Family studies on genetics of juvenile myoclonic epilepsy (epilepsy with impulsive petit mal). In: Beck-Mannagetta G, Anderson VE, Doose H, Janz D, editors. Genetics of the epilepsies, Berlin: Springer, 1989. pp. 43±52. Jovanovic UJ. Schlaf und Epilepsie. Therapiewoche 1967a;17:95. Jovanovic UJ. Das Schlafverhalten der Epileptiker. I. Schlafdauer, Schlaftiefe und Besonderheiten der Schlafperiodik. Dtsch Z Nervenheilk. J Neurol 1967b;190:159±198. Jovanovic UJ. Das Schlafverhalten der Epileptiker. II. Elemente des EEG, Vegetativum und Motorik. Dtsch Z Nervenheilk. J Neurol 1967c;191:257±290. Krischek J. Die Schlaf-wach- Periodik der grand mal-Epilepsie und die sich daraus ergebenden Konsequenzon. Dtsch Med Wschr 1962;87:49. Langdon-Down M, Brain WR. Time of day in relation to convulsions in epilepsy. Lancet 1929;I:1029±1032. Liu AW, Delgado-Escueta AV, Serratosa JM, Alonso ME, Medina MT, Gee MN, Cordova S, Zhao HZ, Spellmann JM, Ramos Peek JR, RubioDonnadieu F, Sparkers RS. Juvenile myoclonic epilepsy locus in chromosome 6 p21.2-p11: linkage to convulsion and electroencephalographie trait. Am J Hum Genet 1995;57:368±381. Loiseau P. Crises eÂpileptiques survenant au reÂveil et eÂpilepsie du reÂveil. Sud MeÂd Chirurgical 1964;99:11492±11502. Loiseau P, Duchet B, Loiseau J. Classi®cation of epilepsies and epileptic
S110
D. Janz / Clinical Neurophysiology 111, Suppl. 2 (2000) S103±S110
syndromes in two different samples of patients. Epilepsia 1991;32:303± 309. Manford H, Harz YM, Sander JW, Shorvon SD. The national general practice study of epilepsy: the syndromic classi®cation of the International League Against Epilepsy applied to epilepsy in a general population. Arch Neurol 1992;49:801±808. Patry FL. The relation of time of day, sleep and other factors to the incidence of epileptic seizures. Am J Psychiatry 1931;10:789±813. Puruker H. Klinik und Klinische Genetik der Epilepsie mit reinem Aufwach- Grand mal (Thesis), Berlin: Free University of Berlin, 1994. RoÈder-Wanner UU, Wolf P, Danninger T. Are sleep patterns in epileptic patients correlated with their type of epilepsy? In: Martins da Silva A, Binnie CD, Meinhardt H, editors. Biorhythms and epilepsy, New York: Raven Press, 1985. pp. 109±119. Sander T, Bockenkamp B, Hildmann T, Blasczyk R, Kretz R, Wienker TF, Voltz A, Schmitz B, Beck-Mannagetta G, Riess O, Epplen JT, Janz D, Ziegler A. Re®ned mapping of the epilepsy susceptibility locus EJM 1 on chromosome 6. Neurology 1997a;49:842±847. Sander T, Kretz R, Williamson MP, Elmslie FV, Rees M, Hildmann T, Bianchi A, Bauer G, Sailer U, Scaramelli A, Schmitz B, Gardiner RM, Janz D, Beck-Mannagetta G. Linkage analysis between idiopathic generalized epilepsies and the GABA a receptor a5, b3 and g3 subunit gene cluster on chromosome 15. Acta Neurol Scand 1997b;96:1±7. Sander T, Kretz R, Schultz H, Sailer U, Bauer G, Scaramelli A, Epplen JT, Riess O, Janz D. Replication analysis of a putative susceptibility locus (EGI) for idiopathic generalized epilepsies on chromosome 8q 24. Epilepsia 1998;39:715±720. Sander T, Schultz H, Saar K, Gennaro E, Riggio MC, Bianchi A, Zara F, Luna D, Bulteau C, Kamminska A, Ville D, Cienta C, Picard F,
Prud'homme J-F, Bate L, Sundquist A, Gardiner RM, Janssen GAMA, de Haan G-J, Kasteleijn-Nolst-Trenite DGA, Bader A, Lindhout D, Riess O, Wienker TF, Janz D, Reis A. Genome search for susceptibility loci of common idiopathic generalised epilepsies. Hum Mol Genet 2000;9:1465±1472. Schmitz B, Wolf P. Epilepsies and epileptic syndromes in a neurological seizure clinic. In: Beck-Mannagetta G, Anderson VE, Doose H, Janz D, editors. Genetics of epilepsies, New York: Springer, 1989. pp. 19±22. Shouse MN, Martins da Silva A. Chronobiology. In: Engel J, Pedly TA, editors. Epilepsy. A comprehensive textbook, Vol. 2. Philadelphia, PA: Lippincott-Raven, 1997. pp. 1917±1928. Terzano MG, Mancia D, Salati MR, Constani G, Decembrino A, Parrino L. The cyclic alternating pattern as a physiologic component of normal NREM sleep. Sleep 1985;8:137±145. Touchon J. Effect on awakening on epileptic activity in primary generalized myoclonic epilepsy. In: Sterman MB, Shouse MN, Passouant P, editors. Sleep and epilepsy, New York: Academic Press, 1982. pp. 239±248. Tsuboi T, Christian W. Epilepsy: a clinical electroencephalographie and statistical study of 466 patients, New York: Springer, 1976. Wolf P. Relations of epilepsy and the sleep and waking cycle. J Auton Nerv Syst 1986:603±610. Wolf P, Goosses R. Relation of photosensitivity to epileptic syndromes. J Neurol Neurosurg Psychiatry 1986;49:1386±1391. Wolf P, RoÈder-Wanner UU, Brede M. In¯uence of therapeutic phenobarbital and phenytoin medication on the polygraphic sleep of patients with epilepsy. Epilepsia 1984;25:467±475. Zara F, Bianchi A, Avanzini G, Di Donato S, Castellotti B, Patel PI, Pandolfo M. Mapping of genes predisposing to idiopathic generalized epilepsy. Hum Mol Genet 1995;4:1201±1207.