- Email: [email protected]
Sleep and sleep deprivation as EEG activating methods
Clinical Neurophysiology 111, Suppl. 2 (2000) S47±S53
www.elsevier.com/locate/clinph
Sleep and sleep deprivation as EEG activating methods R. Marinig a, G. Pauletto a, P. Dolso b, M. Valente a, P. Bergonzi a,* a
Clinica Neurologica, DPMSC UniversitaÁ degli Studi di Udine, Via Colugna 50, 33100 Udine, Italy b U.O. Neurologia-Neuro®siopatologia, A. O. Santa Maria della Misericordia, Udine, Italy
Abstract Objectives: We examined retrospectively 19 patients with a history of clinical seizures, but normal activity or unclear epileptiform abnormalities in wake EEG recordings and obtained preliminary data for a controlled cohort study to evaluate the effects of sleep deprivation (SD) on interictal epileptic activity. Methods: Nineteen patients referred to our EEG department for diagnostic or follow-up purposes were divided in two groups on the basis of the different EEG protocols applied. The ®rst group (n 5) underwent two laboratory polysomnographies during afternoon naps, after SD, but the patients failed to fall asleep in one of the two occasions. The second group (n 14) was submitted to two polysomnographies, the ®rst without SD and the second after SD. Results: The ®rst group of patients demonstrated focal epileptic discharges in 4 patients in which wake after SD appeared to be less activated that sleep after SD. In the second group the results obtained from the waking part of the recordings suggest a lack of activating effect due to SD. Conclusions: SD does not seem to offer greater activation than sleep alone. However, a mild SD may be a convenient activating method for inducing sleep and drowsiness without using any drug. q 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Sleep; Sleep deprivation; EEG activating methods
1. Sleep and epilepsy The Greek philosopher Heraclitus wrote: `the sleeper and the woken are the same, the ®rst changes into the second and so on': this is the original theoretic acquisition of sleep as an active part of life. Centuries later, Gower reported that in some patients epileptic attacks occurred mainly during sleep (Gowers, 1885). The relation between brain excitability, epileptic seizures and sleep/wake rhythm was deeply investigated by several authors (Janz, 1962; Martins da Silva et al., 1984; Declerck, 1986; Burr et al., 1991). They pointed out that sleep in¯uences neuronal excitability and can facilitate the occurrence of seizures as well as interictal epileptic discharges, with the tendency to generalization and polyspiking during NREM sleep, and a better limitation and focalization during REM sleep. Furthermore, in spite of individual variability, sleep/wake rhythm seems to be more important than seasonal and astronomical changes, hormone secretion, metabolic activity and social issues in provoking epileptic discharges. Niedermeyer and Rocca, 1972 observed that epileptic activity * Corresponding author. Tel.: 139-0432/559828; fax: 139-0432/ 989336. E-mail address: [email protected] (P. Bergonzi).
changes in frequency and morphology in relation to sleep stages, with generalized discharges occurring mainly during the ®rst stages of NREM sleep. The effect of sleep on epileptic activity, however, is not univocal, and this ambiguous relationship is complicated by the intrinsic instability of sleep itself. Impairment of sleep stability and continuity is a common feature of epilepsy, usually in partial seizures, but also, although to a lesser degree, in generalized seizures (Montplaisir et al., 1980; Touchon et al., 1991). Frequent stage shifts, increased awakenings and arousals, the occurrence of nocturnal seizures, greater REM latency and the reduction of slow wave sleep (SWS) and REM sleep may constitute facilitatory and self-sustaining factors in epileptic activity (Janz, 1974; Touchon et al., 1991; Terzano et al., 1991). In particular, REM sleep reduction results in a greater neuronal excitability and an increased tendency to diffusion of discharges, with the consequence of lowering the epileptic threshold, both in animal models and in man (Cohen et al., 1970; Naquet et al., 1984). On the basis of common experience, Gibbs and Gibbs (1947) introduced sleep as a method of provocation to reveal epileptic paroxysms in patients with negative standard EEG recordings. They found an increase of epileptic discharge rates from 77 to 98% in absences and generalized
1388-2457/00/$ - see front matter q 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S 1388-245 7(00)00401-6
CLINPH 1031
S48
R. Marinig et al. / Clinical Neurophysiology 111, Suppl. 2 (2000) S47±S53
tonic-clonic seizures combined with absences during sleep recordings. Other authors, like GaÈnshirt and Vetter (1961), Angeleri (1975), Billiard et al. (1981); Broughton (1984), Degen et al. (1987), con®rmed the facilitation of partial and generalized epileptic paroxysms by sleep. Gastaut et al. (1991) found an increase of epileptic interictal activity during short periods of natural sleep, and the maximal yield of epileptiform activity was detected during post-prandial `enforced' naps, in which anticonvulsant therapy was discontinued and a mild sedative administered. Another important ®nding is that epileptic discharges during spontaneous sleep in seizure-free epileptic patients with no evidence on routine wake recordings can predict relapse after drug discontinuation (Degen and Degen, 1984). 2. Sleep deprivation
than either one alone. However, there is also evidence that epileptiform activity is present in the wake EEG after SD, without a gain of additional information by proceeding to sleep recordings (El- Ad et al., 1994). Increased epileptic activity in patients subjected to SD was observed mainly during the waking phase of the recording by Mattson et al. (1965) and Pratt et al. (1968), who claimed for an activating in¯uence of SD on EEG independent from that produced by sleep. How SD may favor the disclosure of paroxysmal EEG activity is still a matter of speculation. Rodin (1991) suggested that SD may increase cerebral excitability, while Klingler et al. (1991) claimed for a threshold lowering effect, due to a change in neuromediators or increased activity of the adrenergic system, induced directly by SD. 3. Sleep and sleep deprivation: an open question
Spontaneous sleep is not easy to obtain in places such as the EEG laboratory or hospital rooms. Even home recordings with portable electroencephalograph are sometimes troublesome for the patients. Therefore, sleep deprivation (SD) was ®rst introduced as a method to facilitate sleep and obtain a sleep EEG tracing. Studies about the effect of SD on brain excitability and epilepsy proliferated since the initial observation of paroxysmal EEG activity in normal subjects after SD, by Rodin et al. (1962) and Bennett (1963). Transient EEG abnormalities during sleep-deprived recordings in both epileptic patients and controls, as well as the occurrence of epileptic paroxysms after SD in patients with previous normal EEGs have been reported (Geller et al., 1969; Scollo Lavizzari et al., 1975; see Table 1). Epileptiform manifestations were found in 50% of non-deprived sleep recordings (Rowan et al., 1982), while the yield was 80% with sleep after SD, and were more frequent during the transition from drowsiness to sleep. It has been therefore suggested that a combination of the two provocation techniques, i.e. SD and sleep, shows a grater activating effect
In spite of almost 50 years of studies and practice, whether sleep or SD is preferable to increase diagnostic information of EEG recordings is still a `vexata quaestio'. Several variables may in¯uence the effects of SD on EEG activity, making rather dif®cult to compare results from different studies: duration and degree of SD, duration of the recording before and after SD, control for the state of sleep, sleep architecture, type of epilepsy and patient characteristics (age of patients, sex, concomitant brain damages, etc). 3.1. Duration and degree of SD The duration of SD is greatly variable among the studies. Most authors recommend 24 h of SD to obtain the best ef®cacy, although partial SD has been reported to be effective as well (Ellingson et al., 1984; Kubicki et al., 1991). Furthermore, the effects of selective deprivation have been analyzed in a small number of studies and it has been found that SWS deprivation had no effects on generalized
Table 1 Percentage of EEG recordings with epileptic activity a Author
SS (%)
DIS (%)
SDS (%)
Routine EEG (%)
SDW (%)
Mattson (1965) Pratt (1968) Geller (1969) Scollo Lavizzari (1977) Degen and Degen (1983) Veldhuizen (1983) Declerck (1986)
± ± ± ± ± ± 84.2 (All night) ± 18.3 (Resting) ± ± ±
± ± ± ± 78 52 ±
34 41 32 57.1 72 53 74.5 (4 h rec.) 38 59.01
± ± ± ± ± ± 35.1
± ± 41 16.7 ± ± ±
±
±
53 40 52
± ± ±
± 9 31
Logothetis (1986) Roth et al. (1986) Degen et al. (1987) El- Ad (1994) Fountain (1998) a
± 52.6 ± ±
SS, spontaneous sleep; DIS, drug-induced sleep; SDS, sleep deprived sleep; SDW, sleep-deprived wake.
R. Marinig et al. / Clinical Neurophysiology 111, Suppl. 2 (2000) S47±S53
discharges, but activated those of partial epilepsy, while REM sleep deprivation increased both generalized and partial discharges (Bergonzi et al., 1972, 1975). 3.2. Duration of recordings before and after SD Since there are wide ¯uctuations in the epileptic activity, prolonged recordings might have a better chance of revealing paroxysms. In early studies of SD effects on EEG, the duration of recordings before and after SD is hardly comparable, when it is mentioned. Sometimes, the patients have been left sleeping until they reached at least the ®rst sleep cycle or the designed stage (Veldhuizen et al., 1983; Degen et al., 1987; El- Ad et al., 1994; Fountain et al., 1998). However, when patients without epileptic activity on spontaneous sleep recordings were studied, a greater rate of discharges was seen after SD (Fountain et al., 1998). The activation was neither in¯uenced by the longer total duration of recording after SD or by the longer duration of sleep stage II in sleep-deprived EEGs (ibidem). 3.3. State of sleep The sleep-wake state may further interfere with the effects of SD. In fact, the activation from SD might be due to increased sleepiness or sleep, rather than to direct activation of discharge frequency in an individual sleepwake state (White et al., 1962; Broeker et al., 1973; Ritter et al., 1977). The rigorous analysis of Veldhuizen et al. (1983), in selected patients with partial and secondary generalized seizures concluded that SD has no direct activating effects, but simply increases drowsiness and light sleep, facilitating the appearance of generalized, but not focal discharges, without an increase in discharge rate in wakefulness after SD. Conversely, El- Ad et al. (1994) found a signi®cantly higher activation in wake EEGs after SD. 3.4. Sleep architecture Sleep phases may exert different activating effects and be differently sensitive to SD. Degen and Degen (1984) and Degen et al. (1987) observed most epileptic paroxysms in the lighter stages of sleep in sleep recordings without SD. On the contrary, SD was related with the appearance of epileptic manifestations in the deeper stages of sleep. Drake et al. (1990) compared patients with generalized seizures and patients with complex partial seizures and found that, although present in both sleep and sleepdeprived recordings, generalized epileptic activity was more likely to occur during afternoon naps than at sleepdeprived recordings. Interictal partial activity, instead, was seen more often or only during prolonged sleep-deprived EEG. This may possibly be explained by the characteristics of sleep in the two different situations: afternoon naps may facilitate the appearance of NREM sleep in which general-
S49
ized activity is abundant, while rebound effect from SD might favor the occurrence of partial paroxysms at the onset of NREM sleep or at the change from NREM to REM sleep. The study of Terzano and colleagues concerning sleep analysis in terms of Cyclic Alternating Pattern (CAP) (1991) demonstrated that phase A induced a marked enhancement of generalized EEG paroxysms and a minor but signi®cant activation of focal lesional bursts. Conversely, phase B exerts an inhibitory modulation to generalized phenomena. This complete evaluation, considering also the microstructure of sleep, provides a better understanding of EEG epileptic activity and of the relation between sleep, vigilance and epilepsy. 3.5. Type of epilepsy Some types of epilepsies have been demonstrated to be more sensitive to the effects of SD: the so-called awakening epilepsies (Degen and Degen, 1980; HalaÂsz, 1984; Niedermeyer, 1984) and the light sensitive epilepsies (Bennett et al., 1969; Scollo Lavizzari and Scollo Lavizzari, 1974; Broughton, 1990). After SD, generalized discharges seem to occur more often than partial activity (Ritter et al., 1977; Rumpl et al., 1977), and these data agree with the greater sensitivity to sleep impairment shown by primary generalized epilepsies. The high activation rates in awakening Grand Mal and Grand Mal combined with absences are due to the frequent occurrence of spike-and-wave complexes in these seizure types, which are activated often by sleep, especially by the lighter stages (Gibbs and Gibbs, 1947; GaÈnshirt and Vetter, 1961; HalaÂsz, 1984; Degen and Degen, 1991). 3.6. Patient characteristics It has been suggested that other factors may in¯uence the effects of SD: age, sex, age at ®rst seizure, duration of epilepsy, anticonvulsant therapy, concomitant brain damage or neurological pathology. There is evidence for greater activation of sleep in children than in adults (Degen and Degen, 1980; Tartara et al., 1980), although negative data are also reported (Fountain et al., 1998). Also a review by Degen and Degen (1991), collecting the whole experience of almost 11 years of studies on sleep and SD, reveals no differences between groups after subdividing the patients according to either clinical criteria (age, sex, age at ®rst seizures, seizure type, etc.) or EEG ®ndings (generalized, partial or generalized with lateral emphasis epileptic activity). Logothetis et al. (1986) found a great activation after SD in patients with a high incidence of underlying abnormalities in neuroimaging, and wondered whether a misdiagnosed cerebral damage could be an important factor in this regard. In conclusion, there is no agreement in literature on SD as an activating method, in comparison with sleep alone.
S50
R. Marinig et al. / Clinical Neurophysiology 111, Suppl. 2 (2000) S47±S53
Nevertheless, there is a reasonable consensus that the management of SD is often dif®cult for the patients, their families and the medical staff. Moreover, the duration of recordings before and after SD should be comparable, to avoid variations in discharge rates due to physiological ¯uctuations of brain activity. Among clinical features, only the type of epilepsy seems to in¯uence the sensitivity to SD, the so-called `awakening epilepsies' being activated more often by SD. 4. Personal clinical experience: observations from a pilot study We examined retrospectively 19 patients and obtained preliminary data in order to design a controlled cohort study to evaluate the effects of sleep deprivation on interictal epileptic activity. Patients (age range 14±71 years) had been referred to the EEG department of Ospedale Santa Maria della Misericordia for diagnostic or follow-up purposes. All of them had a history of clinical seizures, but normal activity or unclear epileptiform abnormalities in wake EEG recordings. The patients recorded for follow up were on anticonvulsant therapy and drugs were not modi®ed or suspended before and during the recordings. None of them had evidence of progressive neurological disease nor focal signs during the neurological examination. Cases of paroxysmal events of uncertain origin or with concomitant sleep disorders were excluded. Two groups were identi®ed on the basis of the different EEG protocols applied. The ®rst group (n 5, all females) underwent two laboratory polysomnographies during afternoon naps, after SD, but the patients failed to fall asleep in one of the two occasions. All cases in this group were of partial epilepsies: one cryptogenetic with simple partial seizures, one idiopathic with simple partial seizures (Benign Childhood Epilepsy with Centro-temporal Spikes), one symptomatic with complex partial seizures, one symptomatic with complex partial seizures and secondary generalization, and one cryptogenetic with complex partial seizures and secondary generalization. The second group (n 14, 7 males) was submitted to two polysomnographies, the ®rst without SD and the second after SD. Six of these patients were affected by cryptogenetic partial epilepsy with complex partial seizures and secondary generalization. Three had symptomatic partial epilepsy (one with complex partial seizures and two with complex partial seizures and secondary generalization) and the remaining were affected by generalized epilepsies, of whom 3 presented with Juvenile myoclonic epilepsy (JME) (Table 2). EEG abnormalities were de®ned as indicative of activation if they appeared as focal or generalized sharp waves, spikes or spike and wave complexes. Discharges were counted and tabulated. We rated the discharge of interictal potentials by calculating the number of focal and general-
ized discharges per 1 min of wake and sleep at each sleep stage. The ®rst group of patients, who underwent diurnal polysomnographies after SD, demonstrated focal epileptic discharges in 4 out of 5 patients (80%). In two of the patients with epileptic EEG abnormalities, the spiking activity was present only during sleep. The other two activated subjects presented focal spikes both during sleep and in the wake part of the recording, making possible a direct comparison between the spiking index during sleep and wake after SD. The focal spiking index (FSI) obtained during wake was lower than that calculated during sleep periods. In conclusion, among all the 4 patients with interictal epileptiform abnormalities, wake after SD appears to be less activated than sleep after SD (especially light sleep). These ®ndings agree with other previous papers where SD with sleep is described to be more activating than SD without falling asleep (Scollo Lavizzari et al., 1977; Rowan et al., 1982). On the contrary, there are also studies reporting that epileptiform activity is more frequent during the wake part of sleep recordings after SD (Mattson et al., 1965; Pratt et al., 1968; Geller et al., 1969). As pointed out by El- Ad et al. (1994) in their review of SD activating effects during wake, the studies involving direct comparison between sleep and wake after SD are few and controversial. We believe that controlled prospective studies are needed to supply this lack. In our second group of patients, in which recordings during spontaneous sleep are compared with recordings during sleep after SD, the results obtained from the waking part of the recordings suggest a lack of activating effect due to SD (Table 2). Epileptic discharges are seen during wake before SD, as well as after SD, without differences. Patients who were not activated by spontaneous sleep did not present epileptic discharges during the sleep recording after SD. From our preliminary results, partial epilepsies seem to be less activated in the sleep recordings with SD than in those without SD. In fact, the FSI is generally lower after SD. The patients affected by JME showed no difference between sleep and sleep after SD concerning the positivity for generalized epileptic abnormalities in the recordings. However, the generalized spiking index (GSI) appeared to be higher in the recordings after SD. These data agree with those described by some authors, who demonstrated that Juvenile Myoclonic Epilepsy, as well as the other so called `awakening epilepsies', is sensitive to the effects of SD (Janz, 1974; Degen and Degen, 1980; HalaÂsz, 1984; Niedermeyer, 1984). Fountain et al. (1998), pointed out that an important methodological limit in their study was the lack of a proper control group. In their work, each patient was his/her own control. The ideal control group, instead, would have required two wake and sleep EEG recordings without SD, to exclude the possibility that the modi®cations in the recordings after SD were due to temporal spontaneous ¯uctuations. The solution to this methodological problem could
Cryptogenetic partial epilepsy
Cryptogenetic partial epilepsy Cryptogenetic partial epilepsy Cryptogenetic partial epilepsy
Cryptogenetic partial epilepsy
Symptomatic partial epilepsy
Symptomatic partial epilepsy
Symptomatic partial epilepsy
Juvenile myoclonic epilepsy
Juvenile myoclonic epilepsy
Juvenile myoclonic epilepsy (phase II) Symptomatic generalized epilepsy Idiopathic generalized epilepsy
2
3 4 5
6
7
8
9
10
11
12
a
14
No
No
Yes
Yes
Yes
No
No
Yes
Yes
No No Yes
Yes
Yes
Paroxysms in wakefulness
No
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No No No
Yes
Yes
Paroxysms in sleep
Without sleep deprivation
±
±
±
15.79 (Wakefulness) 0.69 (phase II) ± ± 0.13 (Wakefulness) 5.40 (phase III) 1.21 (phase III) 0.45 (phase III) 2.00 (phase II) 0.15 (phase II) ±
Maximal focal SI
±
1.01 (phase I) 2.60 (Wakefulness) 4.70 (Wakefulness) ±
-
0.12 (phase III) -
-
± ± -
±
±
Maximal generalized SI
No
No
Yes
Yes
Yes
Yes
Yes
No
Yes
No No No
Yes
Yes
Paroxysms in wakefulness
No
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No No No
Yes
Yes
Paroxysms in sleep
After sleep deprivation
EEG recordings were obtained in two separate sessions, one without sleep deprivation and another one after sleep deprivation.
Cryptogenetic partial epilepsy
1
13
Epilepsy
Patient
Table 2 Occurrence of paroxystic discharges and maximal spike index (SI) in a series of 14 subjects with suspected or con®rmed epilepsy a
±
±
±
2.5 (phase III) 1.10 (phase IV) 0.22 (phase III) 1.00 (phase II) 0.20 (phase I) ±
2.99 (phase III) 0.33 Wakefulness ± ± ±
Maximal focal SI
±
±
1.13 (phase IV) 14.60 (phase II) 4.60
±
±
±
-
0.35 (phase II) 0.43 Wakefulness ± ± ±
Maximal generalized SI
R. Marinig et al. / Clinical Neurophysiology 111, Suppl. 2 (2000) S47±S53 S51
S52
R. Marinig et al. / Clinical Neurophysiology 111, Suppl. 2 (2000) S47±S53
be a prospective study on epileptic patients who fall asleep spontaneously during routine EEG being divided into two groups, the study group being submitted to a second EEG after SD and the control group undergoing a second EEG without SD. We intent to use this design in a future study. 5. Conclusion Sleep deprivation does not seem to offer greater activation than sleep alone, when we compare EEG recordings in sleep without SD and after it. Since SD is a cumbersome procedure for the patients, their families and the staff, and considering that it can provoke epileptic seizures in 3±5% of non-epileptic patients (Degen and Degen, 1991), it would be advisable to restrain SD to patients in whom epilepsy is clinically suspected but EEG recordings are otherwise unremarkable. However, a mild SD, e.g. waking up the patient earlier in the morning, may be a convenient activating method if the aim is just to induce sleep and drowsiness. Combined with other cautions (no exciting drinks, afternoon recordings after a good meal) it can guarantee the presence of suf®cient sleep in the recording, without using any drug to induce drowsiness. References Angeleri F. Partial epilepsies and nocturnal sleep. In: Levin P, Koella WP, editors. Sleep 1974, Basel: S. Karger, 1975. pp. 196±203. Bennett DR. Sleep deprivation and major motor convulsions. Neurology 1963;19:375±377. Bennett DR, Ziter FA, Liske EA. Electroencephalographic study of sleep deprivation in ¯ying personnel. Neurology 1969;19:375±377. Bergonzi P, Chiurulla C, Cianchetti C, Quattrini G. Effetti della REM deprivation negli epilettici (attivitaÁ patologica EEG, fasi e cicli del sonno). Riv Pat Nerv Ment 1972;93:49±59. Bergonzi P, Mazza S, Mennuni G, Zolo P. Selective sleep deprivation (stage IV) in epileptic patients with partial seizures. Arch Psicol Neurol Psichiatry 1975;36:313±318. Billiard M, Echenne B, Besset A, Touchon J, Baldy-Moulinier M, Passouant P. InteÂreÃt de l'enregistrement polygraphique du sommeil de nuit chez l'enfant suspect de crises eÂpileptiques, lorsque les EEG de routine et apreÂs privation de sommeil demeurent normaux. Rev Electroencephalogr Neurophysiol Clin 1981;11(3/4):450±456. Broeker H, Sack G, Muller D, Muller J. Schlaf-EEG-Untersuchungen bei unklaren Anfallszustanden und episodischen Verhaltensstoerungen. Ein vorlau®ger Bericht. Psychiat Neurol Med Psychol (Lpz) 1973;11:656± 660. Broughton RJ. Epilepsy and sleep: a synopsis and prospectus. In: Degen R, Niedermeyer E, editors. Epilepsy, sleep and sleep deprivation, Amsterdam: Elsevier, 1984. pp. 37±56. Broughton RJ. Sleep and sleep deprivation studies in epilepsy. In: Wada JA, Ellington RJ, editors. Clinical neurophysiology of epilepsy. EEG handbook, Vol. 4. Amsterdam: Elsevier Science Publishers BV, 1990. pp. 89±119. Burr W, KoÈrner E, Stefan H. Circadian distribution of genaralized spikewave activity in relation to sleep. In: Degen R, Rodin EA, editors. Epilepsy, sleep and sleep deprivation, 2nd edn (Epilepsy Res, Suppl. 2). Amsterdam: Elsevier Science Publishers BV, 1991. pp. 121±135. Cohen H, Thomas J, Dement WC. Sleep stages, REM deprivation and electroconvulsive threshold in the cat. Brain Res 1970;19:317±321.
Declerck AC. Interaction sleep and epilepsy. Eur Neurol 1986;25(Suppl 2):117±127. Degen R, Degen HE. The diagnostic signi®cance of sleep EEG after sleep deprivation in patients with psycomotor epilepsy receiving anticonvulsant therapy. Electroenceph Clin Neurophysiol 1980;50:135. Degen R, Degen HE. The diagnostic value of the sleep EEG with and without sleep deprivation in patients with atypical absences. Epilepsia 1983;24(5):557±566. Degen R, Degen HE. Sleep and sleep deprivation in epileptology. In: Degen R, Niedermeyer E, editors. Epilepsy, sleep and sleep deprivation, Amsterdam: Elsevier, 1984. pp. 273±286. Degen R, Degen HE. Sleep and sleep deprivation in epileptology. In: Degen R, Rodin EA, editors. Epilepsy, sleep and sleep deprivation, 2nd edn. (Epilepsy Res, Suppl. 2). Amsterdam: Elsevier Science Publishers BV, 1991. pp. 235±260. Degen R, Degen HE, Reker M. Sleep EEG with or without sleep deprivation? Does sleep deprivation activate more epileptic activity in patients suffering from different types of epilepsy? Eur Neurol 1987;26(1):51± 59. Drake ME, Pakalnis A, Phillis BB, Denio LS. Sleep and sleep deprived EEG in partial and generalized epilepsy. Acta Neurol Belg 1990;90:11± 19. El- Ad B, Neufeld MY, Korczyn AD. Should sleep EEG record always be performed after sleep deprivation? Electroenceph clin Neurophysiol 1994;90(4):313±315. Ellingson RJ, Wilkne K, Bennett DR. Ef®cacy o sleep deprivation as an activation procedure in epilepsy patients. J Clin Neurophysiol 1984;1:83±101. Fountain NB, Sung Kim J, Ik Lee S. Sleep deprivation activates epileptiform discharges independent of the activating effects of sleep. J Clin Neurophysiol 1998;15(1):69±75. GaÈnshirt H, Vetter K. Schlafelektroencephalogramm und Schlaf- WachPeriodik bei Epilepsien. Nervenarzt 1961;32:275±279. Gastaut H, Gomez-Almanzar M, Taury M. enforced nap: a simple effective method of inducing sleep activation in epileptics. In: Degen R, Rodin EA, editors. Epilepsy, sleep and sleep deprivation, 2nd edn. (Epilepsy Res, Suppl. 2). Amsterdam: Elsevier Science Publishers BV, 1991. pp. 31±36. Geller M, Gourdij N, Christoff N, Fox E. The effects of sleep deprivation on the EEGs of epileptic children. Dev Med Child Neurol 1969;11:771± 776. Gibbs EL, Gibbs FA. Diagnostic and localizating value of electroencephalographic studies in sleep. Res Publs Ass Res Nerv Ment Dis 1947;26:366±376. Gowers WR. Epilepsy and other chronic convulsive diseases: their causes, symptoms and treatment, London: Churchill, 1885 (New York: Dover; 1966 reprint). 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. Klingler D, TraÈgner H, Deisenhammer E. The nature of the in¯uence of sleep deprivation on the EEG. In: Degen R, Rodin EA, editors. Epilepsy, sleep and sleep deprivation, 2nd edn. (Epilepsy Res, Suppl. 2). Amsterdam: Elsevier Science Publishers BV, 1991. pp. 231±234. Kubicki W, Scheuler S, Wittenbecher H. Short- term sleep EEG recordings after partial sleep deprivation as a routine procedure in order to uncover epileptic phenomena: an evaluation of 719 EEG recordings. In: Degen R, Rodin EA, editors. Epilepsy, sleep and sleep deprivation, 2nd edn. (Epilepsy Res, Suppl. 2). Amsterdam: Elsevier Science Publishers BV, 1991. pp. 217±230. Janz D. The grand mal epilepsies and the sleeping-waking cycle. Epilepsia 1962;3:69±109. Janz D. Epilepsy and the sleeping- waking cycle. In: Vinken PJ, Bruyn GW, editors. Handbook of clinical neurology, Vol. 15. The epilepsies, Amsterdam: Elsevier, 1974. pp. 457±490.
R. Marinig et al. / Clinical Neurophysiology 111, Suppl. 2 (2000) S47±S53 Logothetis J, Milonas I, Bostantzopoulou S. Sleep deprivation as a method of EEG activation. Eur Neurol 1986;25(Suppl 2):134±140. Martins da Silva A, Aarts JHP, Binnie CD, Laxminarayan R, Lopes da Silva FH, Meijer WA, Nagelkerke N. circadian distribution of interictal epileptiform EEG activity. Electroenceph clin Neurophysiol 1984;58:1±13. Mattson RH, Pratt KL, Calverley JR. Electroencephalograms of epileptics following sleep deprivation. Arch Neurol 1965;13:310±315. Montplaisir J, Laverdiere M, Walsh J, Saint-Hilaire JM, Bouvier G. Epilpesie et sommeil. Union Med Can 1980;109:1±6. Naquet R, Tanaka T, Cepeda C. Epileptic manifestations and in¯uence on sleep in the boboon papio- papio. In: Degen R, Niedermeyer E, editors. Epilepsy, sleep and sleep deprivation, Amsterdam: Elsevier, 1984. pp. 47±57. Niedermeyer E. Awakening epilepsy revisited 30 years later. In: Degen R, Niedermeyer E, editors. Epilepsy, sleep and sleep deprivation, Amsterdam: Elsevier, 1984. pp. 85±96. Niedermeyer E, Rocca M. The diagnostic signi®cance of sleep electroencephalogram in temporal lobe epilepsy: a comparison of scalp and depth tracings. Eur Neurol 1972;7:119±139. Pratt KL, Mattson RH, Weikers NJ, Williams R. EEG activation of epileptics following sleep deprivation. A prospective study of 114 cases. Electroenceph clin Neurophysiol 1968;24:11±15. Ritter B, Becker A, Duensing F. Zum diagnostichen Wert des EEGs nach Schlafentzug. Nervenarzt 1977;58:365±368. Rodin EA. Sleep deprivation and epileptological implications. In: Degen R, Rodin EA, editors. Epilepsy, sleep and sleep deprivation, 2nd edn. (Epilepsy Res, Suppl. 2). Amerstam: Elsevier Science Publishers BV, 1991. pp. 265±273. Rodin EA, Luby ED, Gottlieb JS. The encephalogram during prolonged experimental sleep deprivation. Electroenceph clin Neurophysiol 1962;14:544±551.
S53
Roth B, Nevsimalova S, Rothova N. Activation of EEG recordings by graded sleep deprivation. Schweiz Arch Neurol Psychiatr 1986;137(3): 17±38. Rowan AJ, Veldhuizen RJ, Negelkerke NJD. Comparative evaluation of sleep deprivation and sedated sleep EEGs as diagnostic aids in epilepsy. Electroenceph clin Neurophysiol 1982;54:357±364. Rumpl E, Lorenzi E, Bauer G, Hengl W. The value of EEG after sleep deprivation. Z. EEG-EMG 1977;8:205±209. Scollo Lavizzari G, Scollo Lavizzari GP. Sleep, sleep deprivation, photosensitivity and epilepsy. Eur Neurol 1974;11:1±21. Scollo Lavizzari G, Pralle W, de la Cruz N. Activation effects of sleep deprivation and sleep in seizure patients. An electroencephalographic study. Eur Neurol 1975;13:1±5. Scollo Lavizzari G, Pralle W, Radue EW. Comparative study of ef®cacy of waking and sleep recordings following sleep deprivation as an activation method in the diagnosis of epilepsy. Eur Neurol 1977;15:121±123. Tartara A, Moglia A, Manni R, Corbellini C. EEG ®ndings and sleep deprivation. Eur Neurol 1980;19:330±334. Terzano MG, Parrino L, Garofalo PG, Durisotti C, Filati-Roso C. Activation of partial seizures with motor signs during cyclic alternating pattern in human sleep. Epilepsy Res 1991;10:166±173. Touchon J, Baldy-Moulinier M, Billiard M, Besset A, Cadhilac J. Sleep organization and epilepsy. In: Degen R, Rodin EA, editors. Epilepsy, sleep and sleep deprivation, 2nd edn. (Epilepsy Res, Suppl. 2). Amsterdam: Elsevier Science Publishers BV, 1991. pp. 73±81. Veldhuizen R, Binnie CD, Beintema DJ. The effect of sleep deprivation on the EEG in epilepsy. Electroenceph clin Neurophysiol 1983;55:505± 512. White P, Dyken M, Grant P, Jackson L. Electroencephalographic abnormalities during sleep as related to the temporal distribution of seizures. Epilepsia 1962;3:167±174.
www.elsevier.com/locate/clinph
Sleep and sleep deprivation as EEG activating methods R. Marinig a, G. Pauletto a, P. Dolso b, M. Valente a, P. Bergonzi a,* a
Clinica Neurologica, DPMSC UniversitaÁ degli Studi di Udine, Via Colugna 50, 33100 Udine, Italy b U.O. Neurologia-Neuro®siopatologia, A. O. Santa Maria della Misericordia, Udine, Italy
Abstract Objectives: We examined retrospectively 19 patients with a history of clinical seizures, but normal activity or unclear epileptiform abnormalities in wake EEG recordings and obtained preliminary data for a controlled cohort study to evaluate the effects of sleep deprivation (SD) on interictal epileptic activity. Methods: Nineteen patients referred to our EEG department for diagnostic or follow-up purposes were divided in two groups on the basis of the different EEG protocols applied. The ®rst group (n 5) underwent two laboratory polysomnographies during afternoon naps, after SD, but the patients failed to fall asleep in one of the two occasions. The second group (n 14) was submitted to two polysomnographies, the ®rst without SD and the second after SD. Results: The ®rst group of patients demonstrated focal epileptic discharges in 4 patients in which wake after SD appeared to be less activated that sleep after SD. In the second group the results obtained from the waking part of the recordings suggest a lack of activating effect due to SD. Conclusions: SD does not seem to offer greater activation than sleep alone. However, a mild SD may be a convenient activating method for inducing sleep and drowsiness without using any drug. q 2000 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Sleep; Sleep deprivation; EEG activating methods
1. Sleep and epilepsy The Greek philosopher Heraclitus wrote: `the sleeper and the woken are the same, the ®rst changes into the second and so on': this is the original theoretic acquisition of sleep as an active part of life. Centuries later, Gower reported that in some patients epileptic attacks occurred mainly during sleep (Gowers, 1885). The relation between brain excitability, epileptic seizures and sleep/wake rhythm was deeply investigated by several authors (Janz, 1962; Martins da Silva et al., 1984; Declerck, 1986; Burr et al., 1991). They pointed out that sleep in¯uences neuronal excitability and can facilitate the occurrence of seizures as well as interictal epileptic discharges, with the tendency to generalization and polyspiking during NREM sleep, and a better limitation and focalization during REM sleep. Furthermore, in spite of individual variability, sleep/wake rhythm seems to be more important than seasonal and astronomical changes, hormone secretion, metabolic activity and social issues in provoking epileptic discharges. Niedermeyer and Rocca, 1972 observed that epileptic activity * Corresponding author. Tel.: 139-0432/559828; fax: 139-0432/ 989336. E-mail address: [email protected] (P. Bergonzi).
changes in frequency and morphology in relation to sleep stages, with generalized discharges occurring mainly during the ®rst stages of NREM sleep. The effect of sleep on epileptic activity, however, is not univocal, and this ambiguous relationship is complicated by the intrinsic instability of sleep itself. Impairment of sleep stability and continuity is a common feature of epilepsy, usually in partial seizures, but also, although to a lesser degree, in generalized seizures (Montplaisir et al., 1980; Touchon et al., 1991). Frequent stage shifts, increased awakenings and arousals, the occurrence of nocturnal seizures, greater REM latency and the reduction of slow wave sleep (SWS) and REM sleep may constitute facilitatory and self-sustaining factors in epileptic activity (Janz, 1974; Touchon et al., 1991; Terzano et al., 1991). In particular, REM sleep reduction results in a greater neuronal excitability and an increased tendency to diffusion of discharges, with the consequence of lowering the epileptic threshold, both in animal models and in man (Cohen et al., 1970; Naquet et al., 1984). On the basis of common experience, Gibbs and Gibbs (1947) introduced sleep as a method of provocation to reveal epileptic paroxysms in patients with negative standard EEG recordings. They found an increase of epileptic discharge rates from 77 to 98% in absences and generalized
1388-2457/00/$ - see front matter q 2000 Elsevier Science Ireland Ltd. All rights reserved. PII: S 1388-245 7(00)00401-6
CLINPH 1031
S48
R. Marinig et al. / Clinical Neurophysiology 111, Suppl. 2 (2000) S47±S53
tonic-clonic seizures combined with absences during sleep recordings. Other authors, like GaÈnshirt and Vetter (1961), Angeleri (1975), Billiard et al. (1981); Broughton (1984), Degen et al. (1987), con®rmed the facilitation of partial and generalized epileptic paroxysms by sleep. Gastaut et al. (1991) found an increase of epileptic interictal activity during short periods of natural sleep, and the maximal yield of epileptiform activity was detected during post-prandial `enforced' naps, in which anticonvulsant therapy was discontinued and a mild sedative administered. Another important ®nding is that epileptic discharges during spontaneous sleep in seizure-free epileptic patients with no evidence on routine wake recordings can predict relapse after drug discontinuation (Degen and Degen, 1984). 2. Sleep deprivation
than either one alone. However, there is also evidence that epileptiform activity is present in the wake EEG after SD, without a gain of additional information by proceeding to sleep recordings (El- Ad et al., 1994). Increased epileptic activity in patients subjected to SD was observed mainly during the waking phase of the recording by Mattson et al. (1965) and Pratt et al. (1968), who claimed for an activating in¯uence of SD on EEG independent from that produced by sleep. How SD may favor the disclosure of paroxysmal EEG activity is still a matter of speculation. Rodin (1991) suggested that SD may increase cerebral excitability, while Klingler et al. (1991) claimed for a threshold lowering effect, due to a change in neuromediators or increased activity of the adrenergic system, induced directly by SD. 3. Sleep and sleep deprivation: an open question
Spontaneous sleep is not easy to obtain in places such as the EEG laboratory or hospital rooms. Even home recordings with portable electroencephalograph are sometimes troublesome for the patients. Therefore, sleep deprivation (SD) was ®rst introduced as a method to facilitate sleep and obtain a sleep EEG tracing. Studies about the effect of SD on brain excitability and epilepsy proliferated since the initial observation of paroxysmal EEG activity in normal subjects after SD, by Rodin et al. (1962) and Bennett (1963). Transient EEG abnormalities during sleep-deprived recordings in both epileptic patients and controls, as well as the occurrence of epileptic paroxysms after SD in patients with previous normal EEGs have been reported (Geller et al., 1969; Scollo Lavizzari et al., 1975; see Table 1). Epileptiform manifestations were found in 50% of non-deprived sleep recordings (Rowan et al., 1982), while the yield was 80% with sleep after SD, and were more frequent during the transition from drowsiness to sleep. It has been therefore suggested that a combination of the two provocation techniques, i.e. SD and sleep, shows a grater activating effect
In spite of almost 50 years of studies and practice, whether sleep or SD is preferable to increase diagnostic information of EEG recordings is still a `vexata quaestio'. Several variables may in¯uence the effects of SD on EEG activity, making rather dif®cult to compare results from different studies: duration and degree of SD, duration of the recording before and after SD, control for the state of sleep, sleep architecture, type of epilepsy and patient characteristics (age of patients, sex, concomitant brain damages, etc). 3.1. Duration and degree of SD The duration of SD is greatly variable among the studies. Most authors recommend 24 h of SD to obtain the best ef®cacy, although partial SD has been reported to be effective as well (Ellingson et al., 1984; Kubicki et al., 1991). Furthermore, the effects of selective deprivation have been analyzed in a small number of studies and it has been found that SWS deprivation had no effects on generalized
Table 1 Percentage of EEG recordings with epileptic activity a Author
SS (%)
DIS (%)
SDS (%)
Routine EEG (%)
SDW (%)
Mattson (1965) Pratt (1968) Geller (1969) Scollo Lavizzari (1977) Degen and Degen (1983) Veldhuizen (1983) Declerck (1986)
± ± ± ± ± ± 84.2 (All night) ± 18.3 (Resting) ± ± ±
± ± ± ± 78 52 ±
34 41 32 57.1 72 53 74.5 (4 h rec.) 38 59.01
± ± ± ± ± ± 35.1
± ± 41 16.7 ± ± ±
±
±
53 40 52
± ± ±
± 9 31
Logothetis (1986) Roth et al. (1986) Degen et al. (1987) El- Ad (1994) Fountain (1998) a
± 52.6 ± ±
SS, spontaneous sleep; DIS, drug-induced sleep; SDS, sleep deprived sleep; SDW, sleep-deprived wake.
R. Marinig et al. / Clinical Neurophysiology 111, Suppl. 2 (2000) S47±S53
discharges, but activated those of partial epilepsy, while REM sleep deprivation increased both generalized and partial discharges (Bergonzi et al., 1972, 1975). 3.2. Duration of recordings before and after SD Since there are wide ¯uctuations in the epileptic activity, prolonged recordings might have a better chance of revealing paroxysms. In early studies of SD effects on EEG, the duration of recordings before and after SD is hardly comparable, when it is mentioned. Sometimes, the patients have been left sleeping until they reached at least the ®rst sleep cycle or the designed stage (Veldhuizen et al., 1983; Degen et al., 1987; El- Ad et al., 1994; Fountain et al., 1998). However, when patients without epileptic activity on spontaneous sleep recordings were studied, a greater rate of discharges was seen after SD (Fountain et al., 1998). The activation was neither in¯uenced by the longer total duration of recording after SD or by the longer duration of sleep stage II in sleep-deprived EEGs (ibidem). 3.3. State of sleep The sleep-wake state may further interfere with the effects of SD. In fact, the activation from SD might be due to increased sleepiness or sleep, rather than to direct activation of discharge frequency in an individual sleepwake state (White et al., 1962; Broeker et al., 1973; Ritter et al., 1977). The rigorous analysis of Veldhuizen et al. (1983), in selected patients with partial and secondary generalized seizures concluded that SD has no direct activating effects, but simply increases drowsiness and light sleep, facilitating the appearance of generalized, but not focal discharges, without an increase in discharge rate in wakefulness after SD. Conversely, El- Ad et al. (1994) found a signi®cantly higher activation in wake EEGs after SD. 3.4. Sleep architecture Sleep phases may exert different activating effects and be differently sensitive to SD. Degen and Degen (1984) and Degen et al. (1987) observed most epileptic paroxysms in the lighter stages of sleep in sleep recordings without SD. On the contrary, SD was related with the appearance of epileptic manifestations in the deeper stages of sleep. Drake et al. (1990) compared patients with generalized seizures and patients with complex partial seizures and found that, although present in both sleep and sleepdeprived recordings, generalized epileptic activity was more likely to occur during afternoon naps than at sleepdeprived recordings. Interictal partial activity, instead, was seen more often or only during prolonged sleep-deprived EEG. This may possibly be explained by the characteristics of sleep in the two different situations: afternoon naps may facilitate the appearance of NREM sleep in which general-
S49
ized activity is abundant, while rebound effect from SD might favor the occurrence of partial paroxysms at the onset of NREM sleep or at the change from NREM to REM sleep. The study of Terzano and colleagues concerning sleep analysis in terms of Cyclic Alternating Pattern (CAP) (1991) demonstrated that phase A induced a marked enhancement of generalized EEG paroxysms and a minor but signi®cant activation of focal lesional bursts. Conversely, phase B exerts an inhibitory modulation to generalized phenomena. This complete evaluation, considering also the microstructure of sleep, provides a better understanding of EEG epileptic activity and of the relation between sleep, vigilance and epilepsy. 3.5. Type of epilepsy Some types of epilepsies have been demonstrated to be more sensitive to the effects of SD: the so-called awakening epilepsies (Degen and Degen, 1980; HalaÂsz, 1984; Niedermeyer, 1984) and the light sensitive epilepsies (Bennett et al., 1969; Scollo Lavizzari and Scollo Lavizzari, 1974; Broughton, 1990). After SD, generalized discharges seem to occur more often than partial activity (Ritter et al., 1977; Rumpl et al., 1977), and these data agree with the greater sensitivity to sleep impairment shown by primary generalized epilepsies. The high activation rates in awakening Grand Mal and Grand Mal combined with absences are due to the frequent occurrence of spike-and-wave complexes in these seizure types, which are activated often by sleep, especially by the lighter stages (Gibbs and Gibbs, 1947; GaÈnshirt and Vetter, 1961; HalaÂsz, 1984; Degen and Degen, 1991). 3.6. Patient characteristics It has been suggested that other factors may in¯uence the effects of SD: age, sex, age at ®rst seizure, duration of epilepsy, anticonvulsant therapy, concomitant brain damage or neurological pathology. There is evidence for greater activation of sleep in children than in adults (Degen and Degen, 1980; Tartara et al., 1980), although negative data are also reported (Fountain et al., 1998). Also a review by Degen and Degen (1991), collecting the whole experience of almost 11 years of studies on sleep and SD, reveals no differences between groups after subdividing the patients according to either clinical criteria (age, sex, age at ®rst seizures, seizure type, etc.) or EEG ®ndings (generalized, partial or generalized with lateral emphasis epileptic activity). Logothetis et al. (1986) found a great activation after SD in patients with a high incidence of underlying abnormalities in neuroimaging, and wondered whether a misdiagnosed cerebral damage could be an important factor in this regard. In conclusion, there is no agreement in literature on SD as an activating method, in comparison with sleep alone.
S50
R. Marinig et al. / Clinical Neurophysiology 111, Suppl. 2 (2000) S47±S53
Nevertheless, there is a reasonable consensus that the management of SD is often dif®cult for the patients, their families and the medical staff. Moreover, the duration of recordings before and after SD should be comparable, to avoid variations in discharge rates due to physiological ¯uctuations of brain activity. Among clinical features, only the type of epilepsy seems to in¯uence the sensitivity to SD, the so-called `awakening epilepsies' being activated more often by SD. 4. Personal clinical experience: observations from a pilot study We examined retrospectively 19 patients and obtained preliminary data in order to design a controlled cohort study to evaluate the effects of sleep deprivation on interictal epileptic activity. Patients (age range 14±71 years) had been referred to the EEG department of Ospedale Santa Maria della Misericordia for diagnostic or follow-up purposes. All of them had a history of clinical seizures, but normal activity or unclear epileptiform abnormalities in wake EEG recordings. The patients recorded for follow up were on anticonvulsant therapy and drugs were not modi®ed or suspended before and during the recordings. None of them had evidence of progressive neurological disease nor focal signs during the neurological examination. Cases of paroxysmal events of uncertain origin or with concomitant sleep disorders were excluded. Two groups were identi®ed on the basis of the different EEG protocols applied. The ®rst group (n 5, all females) underwent two laboratory polysomnographies during afternoon naps, after SD, but the patients failed to fall asleep in one of the two occasions. All cases in this group were of partial epilepsies: one cryptogenetic with simple partial seizures, one idiopathic with simple partial seizures (Benign Childhood Epilepsy with Centro-temporal Spikes), one symptomatic with complex partial seizures, one symptomatic with complex partial seizures and secondary generalization, and one cryptogenetic with complex partial seizures and secondary generalization. The second group (n 14, 7 males) was submitted to two polysomnographies, the ®rst without SD and the second after SD. Six of these patients were affected by cryptogenetic partial epilepsy with complex partial seizures and secondary generalization. Three had symptomatic partial epilepsy (one with complex partial seizures and two with complex partial seizures and secondary generalization) and the remaining were affected by generalized epilepsies, of whom 3 presented with Juvenile myoclonic epilepsy (JME) (Table 2). EEG abnormalities were de®ned as indicative of activation if they appeared as focal or generalized sharp waves, spikes or spike and wave complexes. Discharges were counted and tabulated. We rated the discharge of interictal potentials by calculating the number of focal and general-
ized discharges per 1 min of wake and sleep at each sleep stage. The ®rst group of patients, who underwent diurnal polysomnographies after SD, demonstrated focal epileptic discharges in 4 out of 5 patients (80%). In two of the patients with epileptic EEG abnormalities, the spiking activity was present only during sleep. The other two activated subjects presented focal spikes both during sleep and in the wake part of the recording, making possible a direct comparison between the spiking index during sleep and wake after SD. The focal spiking index (FSI) obtained during wake was lower than that calculated during sleep periods. In conclusion, among all the 4 patients with interictal epileptiform abnormalities, wake after SD appears to be less activated than sleep after SD (especially light sleep). These ®ndings agree with other previous papers where SD with sleep is described to be more activating than SD without falling asleep (Scollo Lavizzari et al., 1977; Rowan et al., 1982). On the contrary, there are also studies reporting that epileptiform activity is more frequent during the wake part of sleep recordings after SD (Mattson et al., 1965; Pratt et al., 1968; Geller et al., 1969). As pointed out by El- Ad et al. (1994) in their review of SD activating effects during wake, the studies involving direct comparison between sleep and wake after SD are few and controversial. We believe that controlled prospective studies are needed to supply this lack. In our second group of patients, in which recordings during spontaneous sleep are compared with recordings during sleep after SD, the results obtained from the waking part of the recordings suggest a lack of activating effect due to SD (Table 2). Epileptic discharges are seen during wake before SD, as well as after SD, without differences. Patients who were not activated by spontaneous sleep did not present epileptic discharges during the sleep recording after SD. From our preliminary results, partial epilepsies seem to be less activated in the sleep recordings with SD than in those without SD. In fact, the FSI is generally lower after SD. The patients affected by JME showed no difference between sleep and sleep after SD concerning the positivity for generalized epileptic abnormalities in the recordings. However, the generalized spiking index (GSI) appeared to be higher in the recordings after SD. These data agree with those described by some authors, who demonstrated that Juvenile Myoclonic Epilepsy, as well as the other so called `awakening epilepsies', is sensitive to the effects of SD (Janz, 1974; Degen and Degen, 1980; HalaÂsz, 1984; Niedermeyer, 1984). Fountain et al. (1998), pointed out that an important methodological limit in their study was the lack of a proper control group. In their work, each patient was his/her own control. The ideal control group, instead, would have required two wake and sleep EEG recordings without SD, to exclude the possibility that the modi®cations in the recordings after SD were due to temporal spontaneous ¯uctuations. The solution to this methodological problem could
Cryptogenetic partial epilepsy
Cryptogenetic partial epilepsy Cryptogenetic partial epilepsy Cryptogenetic partial epilepsy
Cryptogenetic partial epilepsy
Symptomatic partial epilepsy
Symptomatic partial epilepsy
Symptomatic partial epilepsy
Juvenile myoclonic epilepsy
Juvenile myoclonic epilepsy
Juvenile myoclonic epilepsy (phase II) Symptomatic generalized epilepsy Idiopathic generalized epilepsy
2
3 4 5
6
7
8
9
10
11
12
a
14
No
No
Yes
Yes
Yes
No
No
Yes
Yes
No No Yes
Yes
Yes
Paroxysms in wakefulness
No
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No No No
Yes
Yes
Paroxysms in sleep
Without sleep deprivation
±
±
±
15.79 (Wakefulness) 0.69 (phase II) ± ± 0.13 (Wakefulness) 5.40 (phase III) 1.21 (phase III) 0.45 (phase III) 2.00 (phase II) 0.15 (phase II) ±
Maximal focal SI
±
1.01 (phase I) 2.60 (Wakefulness) 4.70 (Wakefulness) ±
-
0.12 (phase III) -
-
± ± -
±
±
Maximal generalized SI
No
No
Yes
Yes
Yes
Yes
Yes
No
Yes
No No No
Yes
Yes
Paroxysms in wakefulness
No
No
Yes
Yes
Yes
Yes
Yes
Yes
Yes
No No No
Yes
Yes
Paroxysms in sleep
After sleep deprivation
EEG recordings were obtained in two separate sessions, one without sleep deprivation and another one after sleep deprivation.
Cryptogenetic partial epilepsy
1
13
Epilepsy
Patient
Table 2 Occurrence of paroxystic discharges and maximal spike index (SI) in a series of 14 subjects with suspected or con®rmed epilepsy a
±
±
±
2.5 (phase III) 1.10 (phase IV) 0.22 (phase III) 1.00 (phase II) 0.20 (phase I) ±
2.99 (phase III) 0.33 Wakefulness ± ± ±
Maximal focal SI
±
±
1.13 (phase IV) 14.60 (phase II) 4.60
±
±
±
-
0.35 (phase II) 0.43 Wakefulness ± ± ±
Maximal generalized SI
R. Marinig et al. / Clinical Neurophysiology 111, Suppl. 2 (2000) S47±S53 S51
S52
R. Marinig et al. / Clinical Neurophysiology 111, Suppl. 2 (2000) S47±S53
be a prospective study on epileptic patients who fall asleep spontaneously during routine EEG being divided into two groups, the study group being submitted to a second EEG after SD and the control group undergoing a second EEG without SD. We intent to use this design in a future study. 5. Conclusion Sleep deprivation does not seem to offer greater activation than sleep alone, when we compare EEG recordings in sleep without SD and after it. Since SD is a cumbersome procedure for the patients, their families and the staff, and considering that it can provoke epileptic seizures in 3±5% of non-epileptic patients (Degen and Degen, 1991), it would be advisable to restrain SD to patients in whom epilepsy is clinically suspected but EEG recordings are otherwise unremarkable. However, a mild SD, e.g. waking up the patient earlier in the morning, may be a convenient activating method if the aim is just to induce sleep and drowsiness. Combined with other cautions (no exciting drinks, afternoon recordings after a good meal) it can guarantee the presence of suf®cient sleep in the recording, without using any drug to induce drowsiness. References Angeleri F. Partial epilepsies and nocturnal sleep. In: Levin P, Koella WP, editors. Sleep 1974, Basel: S. Karger, 1975. pp. 196±203. Bennett DR. Sleep deprivation and major motor convulsions. Neurology 1963;19:375±377. Bennett DR, Ziter FA, Liske EA. Electroencephalographic study of sleep deprivation in ¯ying personnel. Neurology 1969;19:375±377. Bergonzi P, Chiurulla C, Cianchetti C, Quattrini G. Effetti della REM deprivation negli epilettici (attivitaÁ patologica EEG, fasi e cicli del sonno). Riv Pat Nerv Ment 1972;93:49±59. Bergonzi P, Mazza S, Mennuni G, Zolo P. Selective sleep deprivation (stage IV) in epileptic patients with partial seizures. Arch Psicol Neurol Psichiatry 1975;36:313±318. Billiard M, Echenne B, Besset A, Touchon J, Baldy-Moulinier M, Passouant P. InteÂreÃt de l'enregistrement polygraphique du sommeil de nuit chez l'enfant suspect de crises eÂpileptiques, lorsque les EEG de routine et apreÂs privation de sommeil demeurent normaux. Rev Electroencephalogr Neurophysiol Clin 1981;11(3/4):450±456. Broeker H, Sack G, Muller D, Muller J. Schlaf-EEG-Untersuchungen bei unklaren Anfallszustanden und episodischen Verhaltensstoerungen. Ein vorlau®ger Bericht. Psychiat Neurol Med Psychol (Lpz) 1973;11:656± 660. Broughton RJ. Epilepsy and sleep: a synopsis and prospectus. In: Degen R, Niedermeyer E, editors. Epilepsy, sleep and sleep deprivation, Amsterdam: Elsevier, 1984. pp. 37±56. Broughton RJ. Sleep and sleep deprivation studies in epilepsy. In: Wada JA, Ellington RJ, editors. Clinical neurophysiology of epilepsy. EEG handbook, Vol. 4. Amsterdam: Elsevier Science Publishers BV, 1990. pp. 89±119. Burr W, KoÈrner E, Stefan H. Circadian distribution of genaralized spikewave activity in relation to sleep. In: Degen R, Rodin EA, editors. Epilepsy, sleep and sleep deprivation, 2nd edn (Epilepsy Res, Suppl. 2). Amsterdam: Elsevier Science Publishers BV, 1991. pp. 121±135. Cohen H, Thomas J, Dement WC. Sleep stages, REM deprivation and electroconvulsive threshold in the cat. Brain Res 1970;19:317±321.
Declerck AC. Interaction sleep and epilepsy. Eur Neurol 1986;25(Suppl 2):117±127. Degen R, Degen HE. The diagnostic signi®cance of sleep EEG after sleep deprivation in patients with psycomotor epilepsy receiving anticonvulsant therapy. Electroenceph Clin Neurophysiol 1980;50:135. Degen R, Degen HE. The diagnostic value of the sleep EEG with and without sleep deprivation in patients with atypical absences. Epilepsia 1983;24(5):557±566. Degen R, Degen HE. Sleep and sleep deprivation in epileptology. In: Degen R, Niedermeyer E, editors. Epilepsy, sleep and sleep deprivation, Amsterdam: Elsevier, 1984. pp. 273±286. Degen R, Degen HE. Sleep and sleep deprivation in epileptology. In: Degen R, Rodin EA, editors. Epilepsy, sleep and sleep deprivation, 2nd edn. (Epilepsy Res, Suppl. 2). Amsterdam: Elsevier Science Publishers BV, 1991. pp. 235±260. Degen R, Degen HE, Reker M. Sleep EEG with or without sleep deprivation? Does sleep deprivation activate more epileptic activity in patients suffering from different types of epilepsy? Eur Neurol 1987;26(1):51± 59. Drake ME, Pakalnis A, Phillis BB, Denio LS. Sleep and sleep deprived EEG in partial and generalized epilepsy. Acta Neurol Belg 1990;90:11± 19. El- Ad B, Neufeld MY, Korczyn AD. Should sleep EEG record always be performed after sleep deprivation? Electroenceph clin Neurophysiol 1994;90(4):313±315. Ellingson RJ, Wilkne K, Bennett DR. Ef®cacy o sleep deprivation as an activation procedure in epilepsy patients. J Clin Neurophysiol 1984;1:83±101. Fountain NB, Sung Kim J, Ik Lee S. Sleep deprivation activates epileptiform discharges independent of the activating effects of sleep. J Clin Neurophysiol 1998;15(1):69±75. GaÈnshirt H, Vetter K. Schlafelektroencephalogramm und Schlaf- WachPeriodik bei Epilepsien. Nervenarzt 1961;32:275±279. Gastaut H, Gomez-Almanzar M, Taury M. enforced nap: a simple effective method of inducing sleep activation in epileptics. In: Degen R, Rodin EA, editors. Epilepsy, sleep and sleep deprivation, 2nd edn. (Epilepsy Res, Suppl. 2). Amsterdam: Elsevier Science Publishers BV, 1991. pp. 31±36. Geller M, Gourdij N, Christoff N, Fox E. The effects of sleep deprivation on the EEGs of epileptic children. Dev Med Child Neurol 1969;11:771± 776. Gibbs EL, Gibbs FA. Diagnostic and localizating value of electroencephalographic studies in sleep. Res Publs Ass Res Nerv Ment Dis 1947;26:366±376. Gowers WR. Epilepsy and other chronic convulsive diseases: their causes, symptoms and treatment, London: Churchill, 1885 (New York: Dover; 1966 reprint). 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. Klingler D, TraÈgner H, Deisenhammer E. The nature of the in¯uence of sleep deprivation on the EEG. In: Degen R, Rodin EA, editors. Epilepsy, sleep and sleep deprivation, 2nd edn. (Epilepsy Res, Suppl. 2). Amsterdam: Elsevier Science Publishers BV, 1991. pp. 231±234. Kubicki W, Scheuler S, Wittenbecher H. Short- term sleep EEG recordings after partial sleep deprivation as a routine procedure in order to uncover epileptic phenomena: an evaluation of 719 EEG recordings. In: Degen R, Rodin EA, editors. Epilepsy, sleep and sleep deprivation, 2nd edn. (Epilepsy Res, Suppl. 2). Amsterdam: Elsevier Science Publishers BV, 1991. pp. 217±230. Janz D. The grand mal epilepsies and the sleeping-waking cycle. Epilepsia 1962;3:69±109. Janz D. Epilepsy and the sleeping- waking cycle. In: Vinken PJ, Bruyn GW, editors. Handbook of clinical neurology, Vol. 15. The epilepsies, Amsterdam: Elsevier, 1974. pp. 457±490.
R. Marinig et al. / Clinical Neurophysiology 111, Suppl. 2 (2000) S47±S53 Logothetis J, Milonas I, Bostantzopoulou S. Sleep deprivation as a method of EEG activation. Eur Neurol 1986;25(Suppl 2):134±140. Martins da Silva A, Aarts JHP, Binnie CD, Laxminarayan R, Lopes da Silva FH, Meijer WA, Nagelkerke N. circadian distribution of interictal epileptiform EEG activity. Electroenceph clin Neurophysiol 1984;58:1±13. Mattson RH, Pratt KL, Calverley JR. Electroencephalograms of epileptics following sleep deprivation. Arch Neurol 1965;13:310±315. Montplaisir J, Laverdiere M, Walsh J, Saint-Hilaire JM, Bouvier G. Epilpesie et sommeil. Union Med Can 1980;109:1±6. Naquet R, Tanaka T, Cepeda C. Epileptic manifestations and in¯uence on sleep in the boboon papio- papio. In: Degen R, Niedermeyer E, editors. Epilepsy, sleep and sleep deprivation, Amsterdam: Elsevier, 1984. pp. 47±57. Niedermeyer E. Awakening epilepsy revisited 30 years later. In: Degen R, Niedermeyer E, editors. Epilepsy, sleep and sleep deprivation, Amsterdam: Elsevier, 1984. pp. 85±96. Niedermeyer E, Rocca M. The diagnostic signi®cance of sleep electroencephalogram in temporal lobe epilepsy: a comparison of scalp and depth tracings. Eur Neurol 1972;7:119±139. Pratt KL, Mattson RH, Weikers NJ, Williams R. EEG activation of epileptics following sleep deprivation. A prospective study of 114 cases. Electroenceph clin Neurophysiol 1968;24:11±15. Ritter B, Becker A, Duensing F. Zum diagnostichen Wert des EEGs nach Schlafentzug. Nervenarzt 1977;58:365±368. Rodin EA. Sleep deprivation and epileptological implications. In: Degen R, Rodin EA, editors. Epilepsy, sleep and sleep deprivation, 2nd edn. (Epilepsy Res, Suppl. 2). Amerstam: Elsevier Science Publishers BV, 1991. pp. 265±273. Rodin EA, Luby ED, Gottlieb JS. The encephalogram during prolonged experimental sleep deprivation. Electroenceph clin Neurophysiol 1962;14:544±551.
S53
Roth B, Nevsimalova S, Rothova N. Activation of EEG recordings by graded sleep deprivation. Schweiz Arch Neurol Psychiatr 1986;137(3): 17±38. Rowan AJ, Veldhuizen RJ, Negelkerke NJD. Comparative evaluation of sleep deprivation and sedated sleep EEGs as diagnostic aids in epilepsy. Electroenceph clin Neurophysiol 1982;54:357±364. Rumpl E, Lorenzi E, Bauer G, Hengl W. The value of EEG after sleep deprivation. Z. EEG-EMG 1977;8:205±209. Scollo Lavizzari G, Scollo Lavizzari GP. Sleep, sleep deprivation, photosensitivity and epilepsy. Eur Neurol 1974;11:1±21. Scollo Lavizzari G, Pralle W, de la Cruz N. Activation effects of sleep deprivation and sleep in seizure patients. An electroencephalographic study. Eur Neurol 1975;13:1±5. Scollo Lavizzari G, Pralle W, Radue EW. Comparative study of ef®cacy of waking and sleep recordings following sleep deprivation as an activation method in the diagnosis of epilepsy. Eur Neurol 1977;15:121±123. Tartara A, Moglia A, Manni R, Corbellini C. EEG ®ndings and sleep deprivation. Eur Neurol 1980;19:330±334. Terzano MG, Parrino L, Garofalo PG, Durisotti C, Filati-Roso C. Activation of partial seizures with motor signs during cyclic alternating pattern in human sleep. Epilepsy Res 1991;10:166±173. Touchon J, Baldy-Moulinier M, Billiard M, Besset A, Cadhilac J. Sleep organization and epilepsy. In: Degen R, Rodin EA, editors. Epilepsy, sleep and sleep deprivation, 2nd edn. (Epilepsy Res, Suppl. 2). Amsterdam: Elsevier Science Publishers BV, 1991. pp. 73±81. Veldhuizen R, Binnie CD, Beintema DJ. The effect of sleep deprivation on the EEG in epilepsy. Electroenceph clin Neurophysiol 1983;55:505± 512. White P, Dyken M, Grant P, Jackson L. Electroencephalographic abnormalities during sleep as related to the temporal distribution of seizures. Epilepsia 1962;3:167±174.