Alzheimer's disease results in degenerative changes to brain stem nuclei and cortical pathways that in all probability contribute to the breakdown in sleep and its diurnal pattern, though the actual pathological processes remain unknown (Vitiello and Borson, 2001). Decreased cholinergic input to the cerebral cortex secondary to neuronal loss in the nucleus basalis may affect both daytime arousal and REM sleep propensity (Chokroverty, 1996). The circadian breakdown in the sleep-wake cycle commonly seen in AD may be due to degeneration of the suprachiasmatic nucleus (or circadian pacemaker or "body clock") (Swaab et al., 1985). The cholinesterase inhibitors commonly prescribed to patients with AD can improve daytime alertness but may also contribute to insomnia (Stahl et al., 2003). High rates of sleep-disordered breathing (obstructive hypopneas and apneas) may also contribute to the daytime sleepiness and agitation in people with dementia (Ancoli-Israel et al., 1991; Gehrman et al., 2003). Gehrman et al. (2003) found that 63% of the dementia subjects had apneas or hypopneas of probable clinical severity; a strikingly high rate that contrasts with the 4% to 8% prevalence rate in non-demented elderly.
Assessment Tools
Assessment of sleep symptoms requires help from family or caregivers. Simple sleep diaries or sleep logs in which the caregiver records sleep times, describes nighttime awakenings and estimates sleep quality can be very helpful to the clinician. Validated rating scales of sleep symptoms can also be helpful, both in the initial assessment as well as follow up for treatment interventions. The Sleep Disorders Inventory (SDI) is an easy-to-use instrument that has been validated in a cohort with AD (Tractenberg et al., 2003). Based on the Neuropsychiatric Inventory (NPI) (Cummings et al., 1994), the SDI evaluates the frequency, severity and caregiver distress of eight sleep-related symptoms. The Epworth Sleepiness Scale (ESS) can be helpful in evaluating daytime sleepiness, particularly in less impaired, community-residing patients (Johns, 1991).
Wrist actigraphs provide nonintrusive technology to provide objective data on the sleep-wake cycle in patients with AD. They are worn like a wristwatch, and although they only measure movement, they are useful for long-term studies and can have excellent reliability and validity against sleep electroencephalogram (Ancoli-Israel et al., 2003; Singer et al., 2003). However, actigraphs are expensive and are generally used only by sleep labs or sleep researchers. Some investigators are experimenting with pressure sensors under the beds of long-term care residents to obtain longitudinal data on time in bed. This technology may be of some assistance to clinicians and researchers in the future.
The gold standard of sleep assessment is polysomnography (PSG). This technology is the only way to obtain diagnostic information critical to the assessment of obstructive sleep apnea, periodic limb movement and parasomnias such as REM behavior disorder, but it is expensive and not always practical for more severely demented patients. Polysomnography should be reserved for those patients with daytime somnolence that cannot be attributed to insomnia, medications or acute illness, and then only if the patient can tolerate the procedure. Some sleep labs will allow a caregiver to remain in the lab with the patient, improving the likelihood of a successful study. In-home studies are available in many areas, and performing the study in the familiar environment of home makes it much easier on the patient with dementia.
Treatment
Initial approach. Diagnosis of the underlying cause of insomnia or daytime sleepiness is a necessary first step in treatment. Although AD itself can cause sleep symptoms, other medical, psychiatric and environmental causes of sleep disturbance in old age need to be considered (Table 1). If the dementia is far advanced, some of these problems (e.g., obstructive sleep apnea) can be approached very conservatively. Pain and comfort issues, of course, would be actively managed, regardless of the stage of dementia. Routine administration of analgesics at bedtime can be very helpful. Restless legs syndrome and periodic leg movements can be treated with benzodiazepines, opioids or dopamine agonists (Boeve et al., 2002). The potential for these agents to affect cognition and behavior has to be considered. Patients in long-term care facilities are especially likely to spend excessive amounts of time in bed during the day, potentially contributing to poor nighttime sleep (Ancoli-Israel et al., 1989). Structured daytime activity has been shown to improve sleep in these patients (Alessi et al., 1999).
It is necessary to define goals before prescribing for sleep disorders. In most cases, the goal should be to improve daytime alertness. There are times, however, when it may be appropriate to target improved nighttime sleep for the purpose of relieving caregiver distress at home or reducing nighttime disruption that endangers the patient or other residents in long-term care facilities.
