Abstract

Sleep anomalies are found in a variety of conditions causing dementia. Specific features of the sleep and electroencephalographic impairments vary in some respects, but a common pattern emerges. Sleep is usually more fragmented, both from more awakenings and from a longer duration of time awake; slow-wave sleep is decreased; spindles and K-complexes are less well formed or less numerous, so sleep stages are more difficult to distinguish; and rapid eye movement (REM) sleep may be reduced. Quantitative analyses show a slowing of the electroencephalogram (EEG) during wakefulness and, in Alzheimer’s disease, during REM sleep. These impairments typically worsen with the progression of the disease. Demented patients are also more likely to present with periodic limb movements in sleep and respiratory disturbances. This chapter gives an overview of sleep disturbances, characteristics of sleep architecture and microstructure, and quantitative analyses of wakefulness and sleep EEG in a variety of conditions associated with dementia: Alzheimer’s disease, progressive supranuclear palsy, Parkinson’s disease, dementia with Lewy bodies, vascular dementia, Huntington’s disease, Creutzfeldt-Jakob disease, and frontotemporal dementia.

Alzheimer’s Disease

Alzheimer’s disease (AD) is a progressive neurodegenerative disorder characterized by progressive decline in memory and other cognitive domains. It is considered the primary cause of irreversible dementia in old age. Diagnostic criteria, first established by the National Institute of Neurological and Communicative Disorders and Stroke—Alzheimer’s Disease and Related Disorders Association Work Group, were revised ¹ in light of the discovery of new markers, brain imaging findings, and cerebrospinal fluid analyses of amyloid B and tau proteins. Accumulation of abnormal tau proteins brings dysfunction and neuronal death. Highly affected structures include: entorhinal cortex, hippocampus, amygdala, nucleus basalis of Meynert, suprachiasmatic nucleus (SCN), intralaminar nuclei of the thalamus, locus coeruleus, raphe nuclei, central autonomic regulators, and the cortex (sparing the primary cortical areas for a long time). Pathologically, findings include neurofibrillary tangles, neuritic plaques, and neuronal loss.

Sleep Problems

The prevalence of sleep disturbance in AD has been estimated to be around 25% in mild to moderate cases and around 50% in moderate to severe cases. These sleep problems can take many forms: napping excessively during the daytime and having difficulty falling asleep at night, frequent nocturnal awakenings, and waking up to start the day too early. These sleep problems are of consequence because daytime sleepiness, for instance, was found to be associated with greater functional impairments in AD independently of level of cognitive impairment.²

Perhaps the most noticeable sleep problem of patients with AD is the sundowning phenomenon, a delirium-like state characterized by nocturnal agitation or wandering. This phenomenon can be explained, at least in part, by an alteration in the biological clock: the SCN of the hypothalamus. The secretion rhythm of many hormones is affected in the elderly but it is even more apparent in AD patients.³ In addition, there is an earlier timing of the biological clock as evidenced by two common markers, core body temperature and plasma melatonin.³

Since the turn of the century, a series of pathophysiology findings has shed some light on the circadian rhythm disorder of AD patients. A disrupted melatonin production and rhythm was found,⁴ even in the early preclinical stages of AD.⁵ This might be caused by a dysfunction in the sympathetic regulation by the SCN of the pineal melatonin synthesis.⁴ The SCN itself would be under the modulatory influence of the nucleus basalis of Meynert, which degenerates early in AD. It was proposed that sundowning would be the result of the neocortex being slowly turned off to sleep (due to the defective nucleus basalis) when an arousal signal is still being processed.⁶ Circadian rhythm disturbances are known to be associated with a variety of medical conditions: cardiovascular problems, the metabolic syndrome, physical disabilities, respiratory symptoms, decreased immune system functioning, and mood and cognitive impairment.

Obstructive sleep apnea syndrome (OSAS) has been reported to occur with a greater prevalence in AD patients than in the general population.⁷ A relationship between AD and apolipoprotein ε (APOE), a lipoprotein made in the liver and brain and involved in cholesterol transport and deposition, were first noted in the early 1990s.⁸ It was shown that the risk of developing AD was associated with the APOE4 allele. An association has been found between the APOE4 allele and OSAS.⁹ Finally, the presence of rapid eye movement (REM) sleep behavior disorder (RBD) has been shown in one case of AD with mixed Alzheimer’s and Lewy body dementia (DLB) pathology confirmed on autopsy¹⁰ and in one out of 15 consecutive patients with AD.¹¹ The latter study also showed that three more patients with AD presented with REM sleep without atonia.

Polysomnography Findings

In addition to the presence of sleep problems, sleep architecture is modified in patients with AD. Certain sleep changes seem to be an exaggeration of changes that normally appear with aging. Specifically, patients with AD show an increased number and duration of awakenings and, as a result, an increased percentage of stage 1 sleep. Compared to elderly controls, they also show a reduced percentage of slow-wave sleep (SWS).¹² This is the most consistently reported change in patients with mild to moderate AD. All the prior sleep disturbances worsen with increasing severity of AD.¹²

Another change in sleep architecture that suggests accelerated aging in AD is a loss of the specific electroencephalographic (EEG) features of stage 2 sleep. Sleep spindles and K-complexes become poorly formed, of lower amplitude, of different frequency, of shorter duration, and much less numerous.¹³^,¹⁴ With advancing severity of the disease, due to the absence of these characteristic EEG features, it becomes progressively more difficult to separate stage 2 from stage 1 sleep. The proportion of indeterminate non-REM (NREM) sleep increases even further with the disappearance of the true delta waves of SWS.

Conversely, other sleep changes observed in AD do not suggest accelerated aging. Of particular interest is the percentage of REM sleep that remains stable in normal aging but that was reduced in AD patients compared to controls, and this as a result of a decrease in mean REM sleep episode duration.¹³ Other REM sleep variables, such as REM density, number of REM sleep episodes, and REM sleep latency, are usually unchanged,¹³^,¹⁵ as are muscle atonia and phasic electromyographic activity in REM sleep.¹³ In other words, variables pertaining to the initiation of REM sleep and to its characteristic features were unaffected in mild AD. This is probably because these variables are under the control of the mesopontine cholinergic populations, structures that are relatively spared in mild AD. The lower REM sleep percentage, however, could be due to degeneration of the cholinergic nucleus basalis of Meynert. This nucleus normally exerts an inhibitory influence on the nucleus reticularis of the thalamus,¹⁶ the rhythm generator responsible for NREM sleep.

Quantitative Electroencephalography in Alzheimer’s Disease

Wakefulness Electroencephalogram

In AD, waking EEG activity is characterized by more slowing of the dominant occipital rhythm than is found in nondemented old people. There is also an increase in theta and delta activity compared to age-matched controls. Several studies attempted to correlate quantitative waking EEG with clinical severity of AD with variable results. Some demonstrated that relative theta power separated all 4 stages of dementia: none, mild, moderate, and severe dementia. Some reported that decreases of relative power in alpha and beta bands and increases of power in the delta band were correlated with severity of AD.

REM Sleep Electroencephalogram

EEG slowing was found to be more marked during REM sleep than during wakefulness in AD patients.¹⁷^,¹⁸ Moreover, there was a distinctive topographical pattern of REM sleep EEG slowing in AD patients that parallels findings from neuroradiologic¹⁹ and neuropathologic²⁰ studies—a pattern not observed for the waking EEG.²¹ Indeed, REM sleep EEG slowing was greater in the temporoparietal and frontal regions.¹⁸ The quantitative EEG derived from REM sleep (but not from wakefulness) was also correlated with a screening assessment of cognitive functioning in AD (the Mini-Mental State Examination)²² and with a measure of interhemispheric asymmetry of regional cerebral blood flow.²³

REM sleep EEG is superior to wakefulness EEG probably because the cholinergic basal forebrain, which degenerates early in AD, is likely more crucial for EEG activation during REM sleep than during wakefulness. EEG activation during wakefulness is the result of many convergent neuronal and neurotransmitter systems, many of which are not active during REM sleep. The importance of the cholinergic system to cortical activation in REM sleep may also be due to an enhanced activity in the cholinergic system during that state.²⁴

NREM Sleep Electroencephalogram

The presence of increased EEG delta activity over the entire sleep–wake spectrum makes it difficult to distinguish, especially in NREM sleep, the pathologic from the normal physiologic delta waves. Despite the increases observed in delta activity, patients with AD had elicited K-complexes less often in response to auditory stimulation than age-matched controls and produced K-complexes of lower amplitude.²⁵ Therefore, patients with AD have an impaired ability to generate normal physiologic delta activity during NREM sleep. In these patients, the probability of eliciting a K-complex was also correlated negatively with dementia.²⁵

Progressive Supranuclear Palsy

Progressive supranuclear palsy (PSP), also called Steele-Richardson-Olszewski syndrome, is characterized by progressive axial rigidity, postural instability, and supranuclear gaze palsy. The dementia that often evolves in PSP primarily reflects dysfunction in the frontosubcortical neural networks.²⁶ Because the sleep characteristics of PSP are discussed in Chapter 87, only the aspects related to the dementia are reviewed here.

Excessive daytime somnolence is a common occurrence in PSP. Hypocretin 1 (orexin A) levels were found to be low in PSP, and these levels were inversely correlated with the duration of PSP morbidity.²⁷ Perhaps even more striking is the absence or drastic reduction in REM sleep in PSP patients, especially with the progression of the disease.²⁸^,²⁹ A reduction in REM sleep has been generally correlated with a decline in cognitive functioning. Despite this reduction in REM sleep, studies have reported some cases of RBD and REM sleep without atonia in patients with PSP.³⁰^,³¹ RBD is, however, much more prevalent (78%) in a Guadeloupean form of PSP probably due to ingestion of sour sop, a tropical fruit containing mitochondrial poisons.³²

Cognitive decline is, in turn, reflected by a slowing of the EEG. One study has quantitatively assessed the EEG from both REM sleep and wakefulness in patients with PSP.²⁸ For the REM sleep EEG, there were no significant between-group differences for any of the 16 regions studied. During wakefulness, a slowing of the EEG was found mainly for the frontal regions, compared to control subjects. The frontal EEG slowing during wakefulness is consistent with the results of numerous neuropsychological studies that show deficits to be related to frontal lobe functions. The fact that no EEG slowing was found in REM sleep suggests that the slowing observed for wakefulness was not likely due to a cholinergic deficit. This is consistent with findings that normal neocortical and hippocampal choline acetyltransferase activity was found in some patients with PSP.³³ Dopamine levels are, however, severely reduced in the caudate, putamen, and substantia nigra in PSP patients.³³ A frontal deafferentation from the striatopallidal complex is thought to be responsible for the impairment because there are extensive fiber connections between these nuclei and the prefrontal region. Indeed, the positive correlations between degree of impairment on frontal tasks and EEG slowing observed in our PSP patients suggest that both impairments could be at least partially the result of a dopaminergic deficiency.²⁸

Parkinson’s Disease

PD is a progressive neurologic disorder characterized by rigidity, resting tremor, bradykinesia, and an impairment of postural reflexes and gait and caused in part by the degeneration of the dopaminergic cells in the substantia nigra. The sleep modifications experienced by patients with PD are discussed in Chapter 87; therefore, only information relevant to dementia associated with PD is reviewed here.

The incidence of overt dementia in PD is relatively high; in a population-based study of dementia in Parkinson’s disease, approximately 80% of nondemented PD patients developed dementia within 8 years.³⁴ Risk factors include advanced age at onset of symptoms, severe motor symptoms (particularly bradykinesia), levodopa-related confusion or hallucinations, presence of speech and axial involvement, presence of depression, and atypical neurologic features, such as modest response to dopaminergic agents or early autonomic dysfunction.³⁵

Demented patients with PD often experience hallucinations. One study found that patients with REM sleep anomalies have more hallucinations than patients without such anomalies.³⁶ It was proposed that sleep reduction, particularly REM sleep reduction, would trigger hallucinations by enabling the emergence of REM sleep during wakefulness. Hallucinations have been significantly correlated with the presence of RBD independently of age, gender, disease duration, or Unified Parkinson’s Disease Rating Scale score but related to the amount of dopaminergic medication.³⁷ There is growing evidence that RBD is an early manifestation of a neurodegenerative disorder, particularly one of the synucleinopathies (e.g., DLB, PD, and multiple system atrophy).³⁸^,³⁹ The occurrence of RBD in PD was estimated at 15% with a structured questionnaire⁴⁰ but at 33% using polysomnographic recordings⁴¹; only half of these had been detected at the clinical interview. The phenomenon of REM sleep without atonia would explain the reduction in REM sleep reported in patients with PD when the sleep staging has been performed according to the standard criteria.⁴²

Previous studies had shown that approximately a third of patients with PD presented with EEG slowing regardless of the presence of dementia.³⁷ Even when only nondemented patients with PD were studied, a slowing of the EEG in the temporo-occipital and the frontal regions has been found in some patients.⁴³ It was demonstrated that only patients with PD who also had RBD had a slowing of the EEG and of the dominant occipital frequency.⁴⁴ A higher theta power was found during wakefulness in frontal, temporal, parietal, and occipital regions in patients with PD and RBD compared with patients who had PD without RBD and control subjects. The EEG slowing found only in patients with PD and RBD might not be related to an evolutional stage of PD but rather to the presence of RBD itself. In support of this hypothesis, a higher theta power in the frontal, temporal, and occipital region during wakefulness has also been observed in patients who have idiopathic RBD without PD.⁴⁵

Similarly, patients with PD and concomitant RBD showed a significantly poorer performance on standardized tests measuring episodic verbal memory, executive functions, and visuospatial and visuoperceptual processing compared to both patients with PD without RBD and control subjects.⁴⁶ Interestingly, patients with idiopathic RBD had a lower performance (compared to control subjects) on the same neuropsychological tests (executive functions and verbal memory)⁴⁷ than did the patients with PD and RBD.

The association of RBD with dementia in PD was more directly demonstrated by one study.⁴⁸ Of the 65 patients with PD who completed the study, 24 met the clinical diagnosis of RBD. The incidence of RBD was significantly higher in the demented PD group compared to the nondemented PD group (77% versus 27%). Patients who had PD but not RBD had a lower incidence of dementia (7.3%) compared to patients with PD and RBD (42%).

The topography of EEG slowing observed in PD with RBD is similar to that of the hypoperfusion and hypometabolism seen in DLB,⁴⁹^,⁵⁰ the profile of cognitive impairments noted in PD with RBD resembles that of DLB,⁵¹ and many RBD patients later develop DLB.⁵²^,⁵³ Thus there is reason to believe that the presence of RBD in patients with PD may be an early sign of an evolution toward dementia. Some evidence suggests that cortical Lewy body–type degeneration is the main source of dementia in PD. One study reported diffuse cortical Lewy bodies only in demented patients with PD, whereas nondemented patients had only brainstem Lewy bodies.⁵⁴ Two other studies have suggested that α-synuclein–positive cortical (especially frontal) Lewy bodies were associated with cognitive impairment, independent of or more specifically than an AD-type pathologic process.⁵⁵^,⁵⁶ More studies are necessary to determine the pathologic and neurochemical underpinnings of dementia in PD.

Dementia With Lewy Bodies

Dementia with Lewy bodies represents the second most common neurodegenerative cause of dementia in old age. Until recently, it was a much underdiagnosed form of dementia. The core clinical features of DLB are progressive cognitive decline, spontaneous parkinsonism, recurrent visual hallucinations, and fluctuating cognition and vigilance.⁵⁷ Autonomic dysfunction is often present.⁵⁷ It is pathologically characterized by the presence of Lewy bodies in limbic or neocortical structures, or both. Three subtypes of DLB have been put forward: a brainstem subtype, a limbic subtype, and a diffuse neortical subtype.⁵⁷

A questionnaire study showed that patients with DLB had more overall sleep disturbances, more movement disorders while asleep, and more daytime sleepiness than patients with AD.⁵⁸ Based on results from the Epworth Sleepiness Scale, 50% of patients with DLB experience excessive daytime sleepiness.⁵⁹ However, normal hypocretin-1 levels were found in the cerebrospinal fluid of patients with DLB and daytime sleepiness, suggesting that sleepiness in DLB is not related primarily to a dysfunction of hypocretin neurotransmission.²⁷^,⁶⁰ A polysomnographic study⁶¹ found that 73% of patients with DLB had a sleep efficiency less than 80%. High proportions of these patients also had pathologic indexes of respiratory disturbances (88%) or periodic leg movements during sleep (PLMS) with arousal (74%).⁶¹

A number of studies or review papers have reported that RBD is a common finding in patients with DLB.⁵²^,⁵³ Twelve of 15 patients with RBD and a neurodegenerative disease had limbic or neocortical Lewy body disease at autopsy (the other three had multiple system atrophy), which indicates a synucleinopathy.⁶² It had even been suggested that the inclusion of RBD in the list of core criteria for DLB would improve the sensitivity and specificity of the diagnosis.^51,^53,⁶² RBD now figures as a suggestive feature for the diagnosis of DLB in the third report of the DLB consortium⁵⁷; these criteria have just been validated.⁶³ As in PD, restless legs syndrome (RLS) and PLMS are also common in DLB and can play a part in sleep-onset insomnia and nocturnal arousals or awakenings, respectively.⁶⁴

A few quantitative EEG studies have reported a slowing of the awake EEG in DLB, expressed as a loss of the alpha rhythm during wakefulness combined with a slowing of both dominant and nondominant rhythms,⁶⁵ or expressed as an increase in theta activity,⁶⁶^,⁶⁷ which correlates with the degree of dementia.⁶⁶ Frontal intermittent rhythmic delta activity has also been reported.⁶⁷ Fluctuating cognition, one of the core features of DLB, was shown to be reflected in cortical activation by the variability of the mean EEG power in DLB patients compared to both AD patients and elderly control subjects.⁶⁸