CME
Primary Sleep Disorders in People with Epilepsy: What We Know, Don’t Know, and Need to Know
Keywords
• Sleep disorders and epilepsy • Sleep apnea and epilepsy • Sleepwalking and frontal lobe epilepsy • Insomnia and epilepsy
Sleep and epilepsy are common bedfellows. Many sleep disorder symptoms and some primary sleep disorders such as excessive daytime sleepiness (EDS), sleep maintenance insomnia, and obstructive sleep apnea (OSA) are 2 to 3 times more common in people with epilepsy than the general population.1–14 Adults with epilepsy and sleep complaints have significantly lower quality of life than those without sleep problems.11–13,15,16 Sleep problems in children with epilepsy are associated with negative effects on daytime behavior and academic performance.17–19 Recognition of this situation has led to increasing numbers of patients being referred to sleep centers to evaluate whether untreated sleep disorders may be contributing to their seizures. Late-onset or worsening seizure control in older adults may herald OSA.20,21 Identifying and treating sleep disorders in people with epilepsy improves seizure control and quality of life in some cases. This article reviews the recent evidence for this claim.
Questionnaire-based studies of the prevalence of primary sleep disorders in adults with epilepsy
Sleep disorders are 2 to 3 times more common in adults11–14 and children1–10 with epilepsy compared with the general age-matched population, especially when seizures are poorly controlled or complicated by comorbid neurologic conditions.
Most recent studies examining the prevalence of sleep complaints in adults with epilepsy are based on sleep questionnaires, sometimes coupled with structured clinical interviews, neuropsychological testing, or psychiatric evaluation.11–13,21–24 Table 1 summarizes these studies. The most common sleep complaints in adults with epilepsy are sleep maintenance insomnia and EDS. A prospective study found 52% of 100 consecutive Greek adults with epilepsy endorsed symptoms of sleep maintenance insomnia versus 38% of 90 age-matched controls,12 although another study found only 25% of Greek adults with epilepsy reported insomnia.13 Among 201 adults with medically refractory partial epilepsy, 10% were prescribed hypnotics for complaints of insomnia.15
Two questionnaire-based studies have reported that EDS is statistically more common in people with epilepsy versus controls.13,23 Between 18% and 28% of adults with epilepsy complained of EDS (Epworth Sleepiness Scale [ESS] score >10) compared with 12% to 17% of controls.13,23 An international cross-sectional survey of 35,327 adults found 24% reported they did not sleep well and 12% complained of severe or dangerous EDS.25 Although sleepiness in adults with epilepsy is more likely multifactorial, symptoms suggestive of OSA or restless legs syndrome (RLS) were independent predictors of an increased ESS score.22,23,26
Several studies find that EDS in people with epilepsy is more likely to be associated with depression or anxiety. Depression and anxiety are more common among adults with epilepsy than healthy controls.27 Scores on the Beck Depression Inventory suggestive of moderate to severe depression best predicted a complaint of EDS in patients with epilepsy, whereas sleep apnea scores contributed only minor independent effects.14 Thirty-two percent of 201 patients with refractory partial epilepsy were also taking medications to treat depression, 21% for anxiety; those taking psychotropics were more likely to complain of sleep problems than those not taking them.15 A retrospective study found 31 mature adults with partial epilepsy endorsed more symptoms of EDS, depression, anxiety, and awakening short of breath or with a headache than age-matched and gender-matched controls.21 Complaints of EDS reported by 48% of 99 unselected adult patients with epilepsy correlated with anxiety and neck circumference.28
Comorbid neurodevelopmental disorders increased the likelihood for sleep complaints in adults with epilepsy. Thirty-one percent of 35 adults with tuberous sclerosis complex (TSC) complained of insomnia, 71% of whom also had a history of epilepsy.26 Complaints of insomnia were associated with OSA and RLS scores. Daytime sleepiness was associated with depression, antisocial behavior, and psychotropic medications. Patients treated with antiepileptic drugs (AEDs) were more likely to report daytime sleepiness, attention deficits, and anxiety.
Sleep hygiene may contribute to sleep/wake complaints in people with epilepsy. A study examining sleep hygiene in 270 adults with epilepsy compared with controls found that among the individuals with epilepsy: (1) 23% smoked at bedtimes; (2) 29% had irregular sleep/wake schedules or varying degrees of sleep deprivation; and (3) 17% engaged in high-concentration/upsetting activities at bedtime.22 Controls had many (if not more) poor sleep habits. However, adults with epilepsy were more likely to drink coffee before bedtime (50% of patients with epilepsy vs 30% controls) and nap after dinner (16% epilepsy vs 6% controls). Another study in 108 adults with epilepsy found that many did not practice healthy lifestyle behaviors (including sleep hygiene) even if they were compliant with AED therapy.29
Several factors not associated with more sleep complaints in adults with epilepsy have been reported. Most studies did not find gender a risk factor for sleep problems in people with epilepsy,12,13,22 except one in which women with refractory partial epilepsy reported more severe sleep problems than men.15 Neither EDS nor insomnia was particularly more common in adults with partial or primary generalized epilepsies.12,13,22,23 Nocturnal seizures were not more likely to be associated with sleep problems,15,23 except in people with nocturnal frontal lobe epilepsy (NFLE), who reported only more midsleep awakenings than controls.24
Is sleep architecture altered in adults with epilepsy?
Several older studies reported abnormalities in sleep architecture in adults with epilepsy but few of these controlled either for seizures or medication.30,31 These studies found: (1) reduced time spent in rapid eye movement (REM) sleep; (2) prolonged REM latency; (3) increased wake after sleep onset (WASO) resulting in reduced total sleep time (TST) and sleep efficiency; and (4) increased number of arousals, awakenings, and stage shifts,30,31 even in the absence of seizures the night of polysomnography (PSG).31
Abnormalities in REM Sleep Often Seen in Adults with Epilepsy
REM sleep may be particularly susceptible to the occurrence of seizures in people with partial epilepsy. One study found that REM sleep time decreased from a mean of 18% to 12% if the patient had a seizure that day and 16% to 7% if the seizure occurred during nighttime sleep.32 Night seizures (but not day seizures) significantly reduced sleep efficiency, increased REM latency, increased stage 1 sleep, reduced stage 2 and 4 sleep, and increased drowsiness on the Maintenance of Wakefulness test.32 The reduced sleep efficiency and prolonged REM latency were even greater if the temporal lobe seizure occurred before the first REM period. Both diurnal and nocturnal seizures prolonged REM sleep latency. Nocturnal, but not diurnal, seizures increased stage nonrapid eye movement (NREM) 1 and decreased deeper NREM 3 sleep.
If a motor convulsion occurs during a night of PSG in a person with epilepsy (regardless of whether it is primary generalized or focal in onset), changes in sleep architecture observed may include a prolonged REM latency, decreased TST and REM sleep time and increased WASO, arousals, and NREM 1 and 2 sleep time.31,33
Sleep Architecture May Be More Disrupted in People with Temporal Lobe Epilepsy
Some studies suggest that sleep architecture is more disturbed in adults with temporal lobe epilepsy (TLE) compared with those whose seizures emanate from the frontal lobes (frontal lobe epilepsy [FLE]) or who have primary generalized epilepsy (PGE).31,33,34 In a study involving 15 patients with mesial TLE and 15 with FLE, patients with TLE had reduced sleep efficiency, increased WASO, and more arousals than those with FLE despite their seizures occurring less often in sleep.34 These differences persisted even after sleep deprivation or AED withdrawal. Fig. 1 shows 2 sleep histograms, one from a patient with FLE, the other TLE.
Abnormalities in NREM Sleep Microarchitecture Identified Using Cyclic Alternating Pattern Analysis in People with Epilepsy
For more than 2 decades, sleep researchers have used cyclic alternating pattern (CAP) analysis of NREM sleep microarchitecture to confirm instability of NREM sleep in a variety of sleep disorders, include different epilepsies.35–45 NREM sleep using CAP analysis can be divided into 2 phases: (A and B. CAPs are periodic cyclic variations in electroencephalographic activity during NREM sleep as the sleeping brain is challenged by the modification of environmental conditions. Close study of several epochs of NREM sleep identified that undisturbed periods of typical NREM sleep (phase B) alternate with phasic events (phase A). The nature of the phasic activity scored as phase A events is further categorized as (1) phase A1: intermittent alpha rhythm in stage NREM; sequences of K-complexes or of delta bursts in NREM 2 or 3; (2) phase A2: periods of desynchronized electroencephalographic activity that follow K-arousals; and (3) phase A3: change in electroencephalogram (EEG) which is scored on arousal following American Academy of Sleep Medicine (AASM) Scoring Manual rules.
Sleep disorders confirmed by PSG in adults with epilepsy
Sleep studies in adults with epilepsy are most often performed for suspected OSA, occasionally to characterize nocturnal spells, rarely for suspected REM sleep behavior disorder (RBD), or unexplained hypersomnia. Some studies suggest that OSA is found in 10% of unselected adult patients with epilepsy and 30% of patients with medically refractory epilepsy.19,20,46–48 If an apnea hypopnea index (AHI) of 5 or greater is found in approximately 24% of men and 9% of women in the general adult population (ages 30–60 years),49,50 then OSA is more prevalent in adults with epilepsy.
OSA in Adults with Epilepsy Often Mild
Using clinical assessment, OSA (AHI ≥5) was found in 10% of 283 unselected adults with epilepsy; however, the degree of sleep-disordered breathing (SDB) was mild (AHI 5 to <14) in 67%, moderate (15 to <30) in 22%, and severe (≥30) in only 11%.46 Using overnight PSG and comparing it with the Sleep Apnea-Sleep Disorders Questionnaire (SA-SDQ), another study found OSA (AHI >5) in 45% of 125 unselected adults with epilepsy.51 These investigators validated the use of the SA-SDQ: a score of more than 29 on it provided a sensitivity of 75% and a specificity of 65% in men, and 80% and 67%, respectively, in women with epilepsy.
OSA More Likely to be Found in Adults with Medically Refractory, Late-onset, or Worsening Epilepsy
Three recent studies suggest that OSA is more likely to be found in adults with epilepsy who are older, heavier, male, or have late-onset, medically refractory, or worsening epilepsy.20,46,47 In the first of these groups, OSA was more likely to be found on PSG in adults with epilepsy who were male (15.4% men, 5.4% women), older (46 ± 15 vs 33 ± 12 years), sleepier (23% vs 9%), heavier (28.5 ± 3.6 vs 23.3 ± 3.7 kg/m2), and had experienced their first seizure at an older age (32 vs 19 years).46
A recent, prospective study compared the prevalence of OSA by PSG in 11 adults with late-onset or worsening seizures with 10 who were seizure free or had improving seizure control at or after the age of 50 years.20 The group with late-onset or worsening seizures had higher AHI and higher scores on the SA-SDQ and ESS than the group with better-controlled epilepsy. The 2 groups were similar in age, body mass index (BMI, calculated as weight in kilograms divided by the square of height in meters), neck circumference, number of prescribed AEDs, and frequency of nocturnal seizures. The investigators concluded that OSA in older adults is associated with seizure exacerbation in some cases. A retrospective chart review found that the appearance of OSA symptoms in 21 of 29 older adults (median age 56 years, 86% men) coincided with a clear increase in seizure frequency or the first episode of status epilepticus.47
OSA (respiratory disturbance index >5) was found in 33% of 39 unselected consecutive adults with medically refractory epilepsy (59% of the men, 19% of the women).52 Individuals with OSA were more likely to be older, male, have a higher SA-SDQ score, and more likely to have seizures during sleep than those without OSA.52 A prospective pilot study found OSA (AHI >10) in 46% of 13 adults with refractory epilepsy.48 Larger prospective studies are needed to confirm these findings.
PLMS and RLS in Adults with Epilepsy
PLMS are common and often nonspecific in adults who do not endorse symptoms of RLS, RBD, narcolepsy with cataplexy, or take psychotropic medications. A few studies have reported PLMS in PSG in adults or children with epilepsy. Sleep studies were most often performed for sleep complaints suggestive of OSA and less often for limb jerking or RLS. In 1 study, complaints of RLS were not more prevalent in adults with epilepsy than healthy controls (18% vs 12%) but RLS was ascertained using a single question.12 In a second study involving 158 adults with epilepsy, 35% endorsed symptoms of RLS, but so did 29% of controls with other neurologic disorders.23 These investigators recorded PSG in 27 of the 42 adults with epilepsy who complained of RLS and found PLMS arousal indexes greater than 10 in only 15%. Another retrospective study by the same group recorded PSG in 63 adults with epilepsy with sleep complaints and found a PLMS index (PLMI) 20 or greater in 17% (45% of whom had PLMI >30).53 However, most of the PLMS did not cause arousal or need treatment.
RBD Occasionally Found in Older Adults with Epilepsy
Two case series have reported RBD coexisting with epilepsy in older adults. The first described 2 men (ages 60 and 75 years) who developed late-onset sleep-related motor convulsions and who also had symptoms and PSG findings of RBD. RBD preceded the onset of epilepsy by 5 to 10 years.54 A prospective study found RBD in 10 (12.5%) of 80 older adults (mean age 71 ± 7 years, 47 men) with epilepsy.55 RBD episodes preceded seizure onset by 4.5 years in 6 individuals and followed it by 9.7 years in 4.
Given the prevalence of OSA in late-onset or worsening epilepsy, RBD needs to be distinguished from pseudo-RBD caused by severe OSA. The term pseudo-RBD was coined by investigators of a study involving 16 adults with severe OSA (mean AHI 68 ± 19) who were believed likely to have RBD because they complained of dream-enacting behaviors and unpleasant dreams.56 However, skeletal atonia was preserved during REM sleep and continuous positive airway pressure (CPAP) therapy eliminated the abnormal behaviors, unpleasant dreams, daytime sleepiness, and snoring.
Questionnaire-based studies on the prevalence of sleep disorders in children with epilepsy
Like adults with epilepsy, children with it are more likely to have sleep problems than the general pediatric population.1–10 Sleep disruption in children with epilepsy is more likely multifactorial, including varying combinations of epilepsy per se, frequent nocturnal seizures disrupting nocturnal sleep organization, effects of AEDs on daytime alertness and nighttime sleep, and treatable primary sleep disorders. Comorbidities such as physical disability,8 intellectual disability,2,57,58 neurodevelopmental syndromes,59,60 autism spectrum disorder,61 and behavioral disorders1,7–9,62 may add to the likelihood of sleep disorders in a child with epilepsy. Sleep complaints in children with epilepsy are rarely reported by patients and caretakers and often misdiagnosed.63 A case-control parental report study of 43 children with idiopathic benign rolandic epilepsy (ages 6–16 years) found that those with epilepsy had significantly shorter sleep duration and more frequent parasomnias and daytime sleepiness than the controls.5
In a prospective study evaluating the prevalence of sleep problems in children with epilepsy, sleep and daytime behavior problems were more common in children with epilepsy than their siblings or healthy age-matched controls.1 Using post hoc comparisons, 89 children with idiopathic partial or generalized epilepsy had significantly more parasomnias, bedtime difficulties, sleep fragmentation, and daytime drowsiness than their 49 siblings or 321 healthy controls. Using multiple regression analysis, sleep complaints, longer sleep latencies, and shorter sleep times were more likely in children with poor seizure control. Daytime seizures and high nighttime IED discharge rates predicted daytime drowsiness. The presence of behavior problems (inattention, hyperactivity, impulsivity, oppositional defiant disorder) greatly increased the likelihood that sleep problems would be reported in children with epilepsy. Three variables significantly associated with greater sleep problems in these otherwise normal children with epilepsy were length of freedom from seizure, age, and higher rates of IEDs during sleep. The investigators concluded that the presence of epilepsy in a highly selected sample of children (without other comorbidities and whose epilepsy was more often well controlled) was still associated with sleep, behavior, and adjustment problems beyond those seen in their siblings or healthy controls.
A case-control study found that children with epilepsy (n = 79, mean age 10.1 ± 3.1 years) had a mean of 4 ± 3 sleep problems compared with 2 ± 2 in 73 controls matched for age and gender (P<.001).7 Reports of frequent unsound sleep, snoring, daytime hyperactivity, sudden daytime sleep attacks, limb movements during sleep, and bedtime refusal were more common in children with epilepsy. Mean scores for SDB symptoms were 2 times higher among the children with epilepsy compared with controls (10.5 vs 5.0). Other questionnaire-based studies have found that: (1) symptoms of OSA were 15 times more likely to be reported by the parents of 26 children with epilepsy (mean age 14.6 years) than among an equal number of healthy controls (65% vs 4%)6; (2) children with epilepsy compared with control individuals had more daytime sleepiness, less on-task behavior, and less attention58; and (3) children with benign rolandic epilepsy had significantly shorter sleep duration, more frequent parasomnias, and daytime sleepiness than a reference sample of children.5
Poor sleep hygiene may also contribute to sleep problems in children with epilepsy. In a prospective study evaluating sleep habits, 121 children with epilepsy were compared with a similar number of healthy Brazilian schoolchildren.2 Compared with controls, children with epilepsy were more likely to need to be put to bed by their parents, have an afternoon nap, wake during the night, take more than 30 minutes to fall asleep, express fear of the dark, awake with a distressing dream or worry, call out for the parent during the night, or visit the parental bed. Poor seizure control was associated with poorer sleep habits. Compared with the children whose seizures usually occurred when awake, those whose seizures were primarily nocturnal (47%) had significantly more sleep problems. Parental fear and anxiety about seizure recurrence often result in a return to cosleeping in families of children with epilepsy. One study found that 22% of 179 children with epilepsy changed to less independent sleep arrangements after epilepsy onset compared with 8% of 155 children with juvenile diabetes.10 Cosleeping reduces risk for sudden unexpected death in epilepsy.64
Stay updated, free articles. Join our Telegram channel
Full access? Get Clinical Tree
