Advanced Studies in Medicine S607

ABSTRACT

Sleep concerns are common in epilepsy.Although epilepsy often contributes to sleep diffi-culties, many epilepsy patients exhibit one of sev-eral different types of sleep difficultiesindependent of their epilepsy. A careful historyhelps to differentiate the causes of such sleepproblems. Comorbidity between epilepsy andsleep disorders has been shown to exist, and inthese cases, resolving the underlying sleep disor-der may in turn affect seizure control. In othercases in which epilepsy impacts sleep, the adjust-ment of antiepileptic treatment may be effective inresolving the abnormal sleep condition.

This article presents and examines sleep con-cerns in general and then considers how thesedisorders apply to epileptic patients. The commoncategories of sleep concerns in epilepsy are dis-cussed. Consideration is given to taking a basicsleep history of the patient with epilepsy to iden-tify the key causes of insomnia and daytimesleepiness that are not necessarily related to theepilepsy. Practical treatment guidelines that mayassist in the resolution of the sleep issues involvedare presented. (Adv Stud Med. 2004;4(7C):S607-S612)

The relationship between sleep and epilep-sy is complex. Sleep disorders occur inpeople with and without epilepsy, and thestudy of the differences in the nature andintensity of sleep disorders in both groups

can lead to a greater understanding of both sleep andepilepsy. Sleep disorders, including insomnia, are com-mon in the general population and no less common inindividuals who have epilepsy. This article thereforeexamines sleep concerns in general and then considershow they specifically apply to patients with epilepsy.

COMMON SLEEP CONCERNS IN PATIENTS WITH EPILEPSY

Most sleep difficulties reported by patients withepilepsy fall into 1 of 3 categories: (1) The patient can-not fall asleep or stay asleep; (2) The patient is sleepyduring the day; and (3) The patient does unusualthings in his or her sleep, such as loud snoring orunusual movement.

DIFFICULTY SLEEPINGThe first concern is one of sleep efficiency.

Although there are studies that have found sleep dis-turbances in epilepsy patients, more studies are need-ed. One study by Touchon et al compared 80 patientswith epilepsy with 17 healthy, age-matched, sex-matched, nonepileptic controls.1 Even in the absenceof seizure activity, sleep in the epilepsy group was char-acterized by a marked instability compared with nor-mal subjects. Overall, the patients with epilepsy in thisstudy had decreased sleep efficiency compared withcontrols without epilepsyie, the ratio of the timethat they were asleep compared with the time that theywere in bed was diminished. They experienced anincrease in sleep-stage shifts, moving in and out ofsleep. Both the number and the duration of awaken-ings increased. These parameters were most affected in

PROCEEDINGS

SLEEP DISORDERS AND EPILEPSY*

Beth A. Malow, MD, MS

*Based on a presentation given by Dr Malow at the2003 Annual Meeting of the American Epilepsy Society.

Associate Professor, Department of Neurology,University of Michigan School of Medicine, Ann Arbor,Michigan; Associate Professor, Department of Neurology,Medical Director, Sleep Disorders Program, VanderbiltUniversity Medical Center, Nashville, Tennessee.

Address correspondence to: Beth A. Malow, MD, MS,University of Michigan School of Medicine, Sleep DisordersCenter, Room 8D, 8702 University Hospital, Ann Arbor, MI48109. E-mail: bmalow@umich.edu.

S608 Vol. 4 (7C) August 2004

PROCEEDINGS

untreated newly diagnosed patients, so these sleep dis-turbances were not likely the result of antiseizure med-ication side effects. It is apparent that epilepsy mayhave a direct adverse effect upon sleep even whenseizures are not taking place.

There has been a tendency to blame antiepilepticdrugs (AEDs) for sleep disturbances, because of sedationand other side effects, although the study by Touchon etal and other studies contradict this idea.1,2 Carefulmatching of the AED profile to the patients needs ispotentially beneficial regarding sleep disturbances. Forexample, a patient with insomnia could be given a mild-ly sedating AED. Other studies that have focused onsleep disturbances in patients with epilepsy describedfindings of increased latency to sleep onset, increasednumber and duration of awakenings, abnormal sleepspindles and K-complexes, increased number of stageshifts, and decreased or fragmented rapid eye movement(REM) sleep, as well as the decreased sleep efficiencydescribed above.1,3-7

Sleep disturbances can be a direct result of epilepsyitself, even in the absence of seizures or medication,with the extent of the sleep abnormality sometimesrelated to the severity of the epilepsy.2,3 However, it isimportant when diagnosing sleep disorders to consid-er the whole person. A variety of other factors mayinfluence sleep concernssome related to the epilep-sy, others completely independent. Among the possi-bilities to consider are stress-related insomnia; adversemedication effects (eg, felbamate, lamotrigine); andcoexisting medical or psychiatric disorders (commonin epilepsy), such as arthritis, fibromyalgia, anxiety, ordepression. Anxiety often results in difficulty fallingasleep or in nocturnal panic. Depression can also causedifficulty in falling asleep, as well as waking during thenight followed by inability to resume sleep.

Obstructive sleep apnea and restless leg syndromeare 2 primary sleep disorders that should be consideredin patients with epilepsy who have sleep difficulties.2

Obstructive sleep apnea, a medical problem involvingrepetitive cycles of snoring, airway collapse, andarousal, can wake people from sleep. Restless leg syn-drome can either keep people from falling asleep ini-tially or, when they awaken, it can keep them fromgoing back to sleep because of the creepy, crawling,annoying sensation in the legs.

When taking a history from a patient with epilep-sy, it is important to ask about each of these possibili-ties and not to automatically attribute the sleep

disruption to the disease process or its treatments.Important considerations include other medical disor-ders, psychiatric disorders, and primary sleep disorders.Does the patient snore loudly? Has the spouse or bedpartner witnessed apneas? Does the patient wake up witha dry mouth that could suggest obstructive apnea? Doeshe or she experience creepy, crawling sensations in theirlegs that are worse in bed at night? What about his or herhabits? Both caffeine and alcohol can contribute to diffi-culty sleeping. Alcohol is often thought of as a sedative,but it can actually disrupt sleep, particularly in the sec-ond part of the night. Posing such questions to thepatient with epilepsy who suffers from a sleep distur-bance might yield insight into underlying causes that arenot related to the epileptic condition.

Insomnia is very common in the general popula-tion, with approximately 10% of adults reportingchronic insomnia.8 For the patient with epilepsy, theapproach should be the same as that for the generalpopulation, that is, to identify and treat the underly-ing causes. In the general population, medication isgenerally avoided for those with chronic insomnia,although those with acute insomnia or anxiety disor-ders may receive pharmacologic treatment, as maythose undergoing behavioral treatment. For patientswith epilepsy, although specific drugs are typically notadded for chronic insomnia, the AED therapy can betailored to assist with the problem.2 Polysomnographyis not indicated unless there is a concern about thepresence of a primary sleep disorder (eg, obstructivesleep apnea) causing insomnia.

It is important to be able to recognize psychophys-iologic insomnia, the most common type of insomnia,which involves a disorder of initiating and maintainingsleep.9 Patients with this disorder may initially experi-ence a few nights of insomnia, perhaps because of amajor stressor (eg, a move or a death in the family). Asa result of this, the patient actually learns behavior thatprevents sleep. This often takes the form of overcon-cern with the inability to sleep, with consequent exag-gerated remedial behaviors that actually worsen thesituation. For example, an individual becomes con-cerned that he/she will oversleep as a result of theinsomnia, so he/she purchases a large, noisy alarmclock with neon lights, which further distracts him/herfrom sleep. As a result, this person has more difficultygetting to sleep, wakes more frequently, and sooncomes to associate the bedroom with not sleeping(conditioned arousal). The individual has actually con-

Advanced Studies in Medicine S609

PROCEEDINGS

ditioned himself/herself to not sleep and thus beginsusing stimulants to stay awake during the day andalcohol to go to sleep at night. This person has devel-oped inadequate sleep hygiene.

Figure 1 illustrates the model that Spielman et alproposed in the 1980s for insomnia: a precipitatingfactor (eg, a new job, a medical or psychiatric condi-tion, or a medication)if it occurs together with anunderlying predisposition (eg, personality, age)maycause acute insomnia.10 At this point, the individualmay rapidly become conditioned to the situation andperpetuate it by abusing alcohol or caffeine or by tak-ing naps during the day to overcome daytime sleepi-ness and not sleeping well at night. It is possible thatthe sleep disturbances caused by epilepsy could func-tion as a predisposing factor in this model. Furtherstudy is needed for clarification of this theory.

INTERVENTIONSWhen counseling patients with epilepsy who have

psychophysiologic insomnia, the same approach canbe taken as with the general population. Consider thewhole patient; ask about sleep-related habits. Stimuluscontrol helps to remove poor conditioning so that thepatient can learn once again to associate the bedroomand the bed with sleep (see Sidebar).

Sleep restriction is also an effective, common-senseapproach to reversing poor sleep hygiene. Many patientsgo to bed earlier and earlier in an effort to ensure sleep,but only lie awake for several hours as a result.Instructing the patient to delay going to bed until per-haps midnight will break this cycle of behavior. Once thepatient relearns falling to sleep soon after retiring, bed-time can be gradually adjusted to suit individual needs.

Other components of good sleep hygiene includeparticipating in physical exercise (not too soon beforebed to avoid stimulation); avoiding alcohol, cigarettes,and caffeine; ensuring comfortable temperatures andsound levels in the bedroom; and avoiding large mealsjust before bed (a light snack is permissible).11,12 A veryimportant itemalthough perhaps not the easiest toimplementis to avoid taking worry-provoking prob-lems to bed.

DAYTIME DROWSINESSDaytime drowsiness is common among patients

with epilepsy and may be a direct result of the disease.A recent study examined vagus nerve stimulation(VNS) and its effect on alertness (Figure 2).13 The

baseline column of the Figure shows sleep latency forpatients in a multiple sleep latency test. The test is per-formed the night after an overnight polysomnogram asa method of documenting an individuals degree ofsleepiness. Subjects are given 4 chances to nap duringthe day, and the time taken to fall asleep is measuredby electroencephalography (EEG). Sleep latency fornormal adults under these circumstances is 10 minutes

Figure 1. Factors Affecting the Development ofInsomnia

Reprinted with permission from Spielman et al. A behavioral perspective oninsomnia treatment. Psychiatr Clin North Am. 1987;10(4):541-553.10

Steps of Stimulus Control

Go to bed only when sleepy.

Do not use the bed for any activities other than sleep (orsex).

Do not read, watch television, or eat in bed.

If you do not fall asleep in about 15 to 20 minutes, leave thebedroom. Return to bed when you are sleepy. Repeat thisstep as necessary.

Get up at the same time each day regardless of how muchyou slept.

Do not nap during the day or sleep in locations other thanyour bed.

S610 Vol. 4 (7C) August 2004

or more. If sleep latency drops to 8 minutes or less, thepatient is considered to be excessively sleepy. Iflatency is 5 minutes or less, sleepiness is consideredto be severe. In this study, the subjects were all med-ically refractory epilepsy patients. Most of thepatients fell asleep within 8 minutes, and somepatients fell asleep within 5 minutes, an indicationthat these patients were experiencing undue sleepi-ness. The second set of data compares the sleeplatency of the same patients after treatment withVNS. Sleep latency improved considerably, with acorresponding improvement in alertness. This isanother demonstration of the concept thatantiepileptic treatmentspharmacologic and non-pharmacologiccan sometimes benefit patientswith coexisting sleep disorders.

Once again, however, in the case of daytimesleepiness, it is important to consider the wholepatient. Even if epilepsy or its treatments are playinga direct role in daytime drowsiness, a careful historymay reveal other contributing factors. Whether apatient does or does not have epilepsy, a systematicapproach can be revealing. The primary cause forsleepiness in the daytime isunsurprisinglyinsuf-ficient sleep at night. Modern life often leaves peo-ple with insufficient time for sleep, and chronic sleepdeprivation is widespread. The possibility of chronicsleep deprivation should be thoroughly explored.What time does the patient want to go to sleep?What time does the patient wake up? Is the patientsleeping throughout the night? How many naps doesthe patient take during the day? Is the patient sleep-ing late on weekends (which also suggests sleepdeprivation)? Even if the patient is taking enoughtime for sleep, sleep may be disrupted by seizures ora sleep disorder, such as apnea, periodic limb move-ments of sleep, or frequent awakening due to a med-ical disorder. If the amount and quality of sleep seemadequate, the patient could have an intrinsic sleepdisorder, such as narcolepsy or idiopathic hypersom-nolence. Epilepsy may also be a cause of the intrin-sic sleep disorder. The possibility of adverse effects ofmedications must be kept in mindmany AEDs areassociated with sedation. Sometimes, the symptomsare those of fatigue rather than sleepiness per se; thiscould be a sign of depression.

The following case history illustrates the impor-tance of taking a holistic approach when investigatingdaytime drowsiness.

CASE STUDY 1

The patient is a 61-year-old man who experiencedhis first seizure during sleep at age 33 in a setting ofviral encephalitis.14 His seizures were well controlledwith medication for 20 years but then increased in fre-quency to several per week. The patient reported day-time sleepiness and a 20-lb weight gain. His wife notedthat he snored heavily and occasionally stopped breath-ing during sleep. Overnight polysomnography showedan apnea/hypopnea index of 15.9 events per hour, witha minimum oxygen saturation of 86%. A multiple sleeplatency test the next day revealed mild daytime sleepi-ness. He was prescribed continuous positive airway pres-sure (CPAP), a pneumatic splint that keeps the airwayopen by putting pressure into the back of the throat. HisAED dosages were kept constant. This patient becameseizure free; he reported that his seizure control returnedto the way it had been for many years and that he wasexperiencing improved daytime alertness.

Other studies have reported success in controllingseizures by treating sleep apnea:

Tracheostomy diminished generalized seizures in1 patient.15

CPAP or other therapy improved seizure control

PROCEEDINGS

Figure 2. Differences in Mean Sleep Latency with VNSStimulation

VNS = Vagus nerve stimulation. Reprinted with permission from Malow et al. Vagus nerve stimulation reduces day-time sleepiness in epilepsy patients. Neurology. 2001;57(5):879-884.13

Advanced Studies in Medicine S611

and daytime alertness in 6 of 7 patients with par-tial seizures.16

CPAP or positional therapy improved seizurecontrol in 7 of 10 patients with seizures andobstructive sleep apnea. Three had antiepilepticmedications optimized.17

CPAP improved seizure control or daytimesleepiness in 7 of 9 patients.14

Pediatric studies have also found similar benefits ofsleep apnea treatment to the symptoms of epilepsy.The precise reasons for the relationship between sleepapnea treatment and reduction in daytime drowsinessand seizure activity are unknown; possibilities includethe reduction of sleep deprivation or improvements inconsolidation of sleep at night.

When treating daytime sleepiness, it is importantto identify the cause. Polysomnography can be a use-ful diagnostic tool. Once the etiology has been deter-mined, treatment choices become clear. If thesleepiness is seizure related, then an increase in AEDsmay be required, or alternatives may be considered,(eg, surgery for frequent seizures). Conversely, if thesleepiness is caused by side effects of AED therapy, theaim may be to reduce polytherapy, determine the low-est effective dose, space out the patients dosing, orswitch medications. If the drowsiness is due to a coex-isting sleep disorder (eg, sleep apnea, periodic limbmovements of sleep, narcolepsy), treatment should bedirected at the specific disorder.

SLEEP ATTACKS

Sleep attacks can confuse the clinical picture. Theyare not necessarily associated with narcolepsy; narcolep-sy can occur without sleep attacks and vice versa. Sleepattacks can be associated with other sleep disorders, suchas restless leg syndrome and sleep apnea, and can be dis-abling to the patient. The patient falls irresistibly asleep,and the sudden loss of consciousness can be mistaken forseizures or pseudoseizures.18 The following case studydemonstrates the importance of correctly diagnosingsleep attacks and the associated challenges.

CASE STUDY 2

An 18-year-old woman described brief spells ofimpairment of consciousness preceded by a stereotyp-ical sensation of a head rush and consisting of 5 to

10 seconds of lost awareness, with eyes rolled back andfluttering, followed by eye closure and unresponsive-ness.19 She stated that she could not focus and that shefelt in a daze during these episodes.

The patient experienced no staring, tonic-clonic activ-ity, automatisms, oral trauma, or bladder or bowel incon-tinence during these episodes. She regainedresponsiveness very quickly, without postevent confusion,but was not aware of what was happening during thespell. She was involved in a motor vehicle accident dur-ing one of these episodes, which prompted her to seekmedical attention. She had no symptoms of narcolepsy:cataplexy, sleep paralysis, or hypnagogic hallucinations.The events occurred regardless of sitting or standing posi-tion, but she never fell during an event. They occurred 4to 10 times per day and increased in frequency when shewas tired or emotionally upset. She denied a syncopeprodrome (feeling warm or weak). Video EEG monitor-ing was inconclusive for epileptic seizures. She was pre-scribed an empiric trial of carbamazepine and reportedincreased daytime sleepiness. Polysomnography with amultiple sleep latency test was performed after a 2-weekperiod of discontinuing carbamazepine and achieving 8hours of sleep per night. The overnight sleep study wasnormal except for some increased arousals. The multiplesleep latency tests did show that she fell asleep on averagein 3.7 minutes, and she also went into REM sleep on 3of the 5 naps, which is diagnostic of narcolepsy. Shebegan treatment with modafinil for narcolepsy and herepisodes of loss of awareness resolved completely, as didher excessive daytime sleepiness.

SUMMARY

Sleep complaints can be categorized into 1 of 3 cat-egories: difficulty falling or staying asleep, difficulty stay-ing awake during the day (which includes daytime sleepattacks), and unusual behaviors during sleep. Patientsmay simultaneously experience difficulties in 1, 2, or all3 categories. Epilepsy and/or its treatments can be adirect cause of sleep difficulties, or conversely, may beunconnected. The physician should not assume a con-nection. A sleep history can help to determine cause of asleep disorder. Once diagnosed, any sleep disorders thatare not associated with the epileptic condition are readi-ly treatable and can impact favorably on patients overallhealth and quality of life. Where epilepsy is a factor,adjustment of the epilepsy treatment regimen may besufficient to ameliorate the problem.

PROCEEDINGS

S612 Vol. 4 (7C) August 2004

REFERENCES

1. Touchon J, Baldy-Moulinier M, Billiard M, Besset A,Cadilhac J. Sleep instability in temporal lobe epilepsy. In:Wolf P, Dam M, Janz D, Dreyfuss FE, eds. Advances inEpileptology. New York: Raven Press; 1987:709-711.

2. Mendez M, Radtke RA. Interactions between sleep andepilepsy. J Clin Neurophysiol. 2001;18(2):106-127.

3. Baldy-Moulinier M. Temporal lobe epilepsy and sleep orga-nization. In: Sterman MB, Passouant P, eds. Sleep andEpilepsy. New York: Academic Press; 1982:347-359.

4. Besset A. Influence of generalized seizures on sleep organi-zation. In: Sterman MB, Shouse MN, Passouant P, eds.Sleep and Epilepsy. New York: Academic Press;1982:339-346.

5. Halasz P. Runs of rapid spikes in sleep: a characteristicEEG expression of generalized malignant epilepticencephalopathies. In: Degen R, Rodin EA, eds. Epilepsy,Sleep and Sleep Deprivation. 2nd ed. Amsterdam: Elsevier;1991:49-71.

6. Hrachovy RA, Frost JD, Kellaway P. Sleep characteristics ininfantile spasms. Neurology. 1981;31(6):668-693.

7. Touchon J, Baldy-Moulinier M, Billiard M, et al. Sleep orga-nization and epilepsy. In: Degen R, Rodin EA, eds.Epilepsy, Sleep and Sleep Deprivation. Amsterdam:Elsevier; 1991:73-81.

8. Zorick FJ, Walsh JK. Evaluation and management of insom-nia: an overview. In: Kryger M, Roth T, Dement W, eds.Principles and Practice of Sleep Medicine. 3rd ed.Philadelphia, Pa: WB Saunders; 2000.

9. American Sleep Disorders Association. InternationalClassification of Sleep Disorders: Diagnostic and CodingManual. Rochester, Minn: American Sleep DisordersAssociation; 1990.

10. Spielman AJ, Caruso LS, Glovinsky PB. A behavioral per-spective on insomnia treatment. Psychiatr Clin North Am.1987;10(4):541-553.

11. Lannon SL, Vaughn BV. Sleep hygiene in people withepilepsy [abstract]. Epilepsia. 1997;38:227.

12. Manni R, Politine L, Ratti MT, Sartori I, Galimberti CA.Sleep hygiene in epilepsy patients: a questionnaire-basedsurvey in 270 epileptic patients of adult age [abstract].Sleep. 1998;21:175.

13. Malow BA, Edwards J, Marzec M, Sagher O, Ross D,Fromes G. Vagus nerve stimulation reduces daytime sleepinessin epilepsy patients. Neurology. 2001;57(5):879-884.

14. Malow BA, Fromes GA, Aldrich MS. Usefulness ofpolysomnography in epilepsy patients. Neurology.1997;48(5):1389-1394.

15. Wyler AR, Weymuller EA Jr. Epilepsy complicated by sleepapnea. Ann Neurol. 1981;9(4):403-404.

16. Devinsky O, Ehrenberg B, Barthlen GM, Abramson HS,Luciano D. Epilepsy and sleep apnea syndrome.Neurology. 1994;44(11):2060-2064.

17. Vaughn BV, DCruz OF, Beach R, Messenheimer JA.Improvement of epileptic seizure control with treatment ofobstructive sleep apnoea. Seizure. 1996;5(1):73-78.

18. Malow BA, Fromes GA, Selwa LM. Sleep attacks mimick-ing epileptic seizures and pseudoseizures. J Epilepsy.1997;10:232-235.

19. Mihaescu M, Malow BA. Sleep disorders: a sometimes for-gotten cause of nonepileptic spells. Epilepsy Behav.2003;4(6):784-787.

PROCEEDINGS