Chapter 49: Sleep Disorders

Problems with sleep are frequently encountered in older adults. More than two-thirds of older people with multiple comorbidities report sleep problems. These problems commonly include difficulty falling asleep, difficulty staying asleep, or sleepiness during the day. This chapter will explore the epidemiology, pathophysiology, clinical presentation, diagnosis, and management of sleep disorders commonly encountered in older adults.

Sleep is a reversible behavioral state that is characterized by a decreased responsiveness to the environment and stereotypical changes in the electroencephalogram (EEG). Each sleep period involves a progression of sleep stages on a cycle that averages 90 minutes. There are two major sleep states: rapid eye movement (REM) and nonrapid eye movement (NREM) sleep. REM sleep is characterized by having brain activity that is more similar to wake, REMs, and a complete loss of skeletal muscle tone. When people are awaken from REM sleep they usually report vivid dreaming. NREM sleep is divided into three stages; N1, N2, and N3 (slow wave sleep). As sleep progresses from N1 through N3 patients become more difficult to arouse, hence N1 sleep is considered light sleep, while N3 sleep is considered deep sleep. As people age there is a reduction in sleep efficiency, which is the amount of time actually asleep divided by time spent in bed (Figure 49-1). Furthermore, the amount of N3 sleep decreases with age starting in early adulthood. The reduction in N3 sleep with age is more prominent in men. Whether there are changes in REM sleep with aging is less clear. Objective measures of sleepiness have also shown that, compared to younger adults, older adults are sleepier during the day.

Although it is somewhat controversial, many experts believe that the age-related changes in sleep are due to a decreased ability to sleep rather than a decreased need to sleep. When older poor sleepers and good sleepers are compared, poor sleepers have worse health-related quality of life, increased medication use, and greater health care utilization. Many of the illnesses associated with aging can disrupt sleep, which may inhibit the normal progression through sleep stages. Thus, chronological age itself may not be the greatest predictor of sleep quality. Several lines of evidence have suggested that lower sleep efficiency and higher amount of wake after sleep onset are associated with more cognitive decline compared to those in whom sleep is well preserved. Since N3 sleep is associated with growth hormone secretion, the reduction in N3 sleep with aging may be partly responsible for the decrease in growth hormone in older men. Given the link between insufficient and fragmented sleep on quality of life and health outcomes, there is a need for awareness, evaluation, and treatment of sleep disorders that commonly affect older adults.

Definition

Sleep-disordered breathing (SDB) is characterized by disturbed respiration during sleep arising from repetitive events of complete (ie, apnea) or partial (ie, hypopnea) cessation of airflow lasting at least 10 seconds. This is classified as OSA when the events are due to an obstructed airway, which is determined by the persistence of respiratory effort. On the other hand, if there is concomitant cessation of breathing effort, the disorder is classified as central sleep apnea (CSA) because there is a momentary defect in the central control of breathing (Figure 49-2). OSA is by far the most common sleep-related breathing disorder; however, there is often overlap between the two clinical syndromes. CSA can be associated with several different clinical conditions. Heart failure is the most commonly recognized cause of CSA and is often characterized by Cheyne-Stokes respiration, which is periodic cycling between hypoventilation and hyperventilation (Figure 49-3). Other common causes of CSA include cerebrovascular accidents, opioid use, and hypoventilation syndromes. The severity of SDB is generally determined by the apnea hypopnea index (AHI), which is the number of apneas and hypopneas per hour of sleep. SDB can be diagnosed when the AHI is greater than 15, or when it is greater than 5 with significant symptoms or related comorbidities. An AHI greater than 30 is generally considered to indicate severe SDB.

The consequences of respiratory events during sleep include arousals from sleep, intrathoracic pressure swings, and cyclical drops in the blood oxygen level. This ultimately leads to sleep fragmentation and nocturnal hypoxemia, which may lead to significant health consequences.

Epidemiology

Many of the risk factors for OSA increase with age (Table 49-1). Estimates of the prevalence of OSA have varied widely and are dependent on the populations studied. Prior work estimated the prevalence of moderate to severe SDB to be 4% for women and 9% for men aged 30 to 60. However, with the current rise in obesity more patients are now at risk for OSA. A more recent update from a cohort between 2007 and 2010 has estimated moderate to severe SDB to occur in 6% of women and 13% of men. Evidence suggests that OSA is underdiagnosed in the general population, particularly in women. If the rate of obesity and overweight continues to increase the rates of SDB will also increase. Less is known about the epidemiology of SDB in an older population; however most studies have shown that the risk of OSA increases with increasing age until the age of 70 at which time there is a plateau (Figure 49-4). Furthermore, premenopausal status appears to protect against OSA meaning that the gender discrepancy between men and women narrows in the older population. OSA may be more severe in African-American and Asian populations compared to the Caucasian population.

It is estimated that upwards of 50% of patients with stable heart failure have some form of SDB. The majority of these patients have a form of CSA, however many have OSA or a combination of the two disorders. It has also become increasingly recognized that patients with end-stage renal disease have a very high prevalence of SDB as well. Other conditions associated with CSA include stroke, neurodegenerative diseases, respiratory disorders leading to hypoventilation, and chronic pain requiring opiates. Since the conditions that contribute to CSA are more common in older adults, the prevalence of CSA is much higher in older compared to younger adults.

Pathophysiology

There are three types of SDB events: central, obstructive, and mixed. Central events result from a failure of the respiratory control center to send a signal to breathe. This often occurs because of an abnormal sensitivity of the respiratory controller, which can lead to oscillations between hyperventilation and hypoventilation. Obstructive events occur from anatomic obstruction of the upper airway despite respiratory effort. The major factor that governs the presence of obstructive events is an anatomically narrow airway; however, other mechanisms such as failure of adequate upper airway neuromuscular activation are thought to play a role. Features that govern the size of the upper airway include excess adiposity, craniofacial structure, excessive tonsilar and peritonsilar soft tissue, and lower lung volumes. Mixed events are a combination of central and obstructive components.

The most significant immediate consequence of OSA is excessive daytime sleepiness and the risk of motor vehicle accidents. Significant epidemiologic associations of OSA with adverse cardiovascular and metabolic consequences have been reported. The association between SDB and these comorbidities is likely multifactorial. SDB results in sleep fragmentation, which has been shown to increase sympathetic nervous system activation, increase nocturnal cortisol levels, and interfere with insulin sensitivity. The potential downstream effects of these neurohumoral changes include the development of hypertension and diabetes. Furthermore, the cyclic deoxygenation and reoxygenation that occurs with OSA leads to endothelial cell injury and inflammatory changes, which predispose to coronary and cerebral vascular disease (Figure 49-5). The intrathoracic pressure changes associated with increasing respiratory effort against a closed airway can lead to esophageal reflux as well as potentially deleterious hemodynamic effects in heart failure.

Clinical Presentation

The major symptoms of SDB include excessive daytime sleepiness and nocturnal snoring. The patient may present to the clinic complaining of sleepiness or at the request of a loved one with the complaint of snoring. The paradigmatic patient with SDB is an obese male with nocturnal snoring, witnessed apneas, frequent nocturnal awakenings, and excessive daytime sleepiness. However, these associations are less predictive in an older population. Other important symptoms of SDB include nocturia, insomnia, morning headaches, nocturnal confusion, and daytime impairments in mood and cognition. Sleepiness can be assessed with a standardized measure such as the Epworth Sleepiness Scale. It is often helpful to obtain corroborating information from a bed partner regarding snoring, witnessed apneas as well as additional features that may point to a comorbid sleep disorder. Snoring is indicative of a partially collapsed airway and is a useful predictor of the presence of OSA or future development of the condition. The lack of classic symptoms and findings of OSA should not preclude further evaluation for SDB.

Physical features that predict increased risk of OSA include an elevated body mass index (BMI), a large neck circumference, and a crowded upper airway. In older adults, an elevated BMI has less predictive value of the likelihood of OSA. In clinic the airway can be assessed using a modified Mallampati score known as the Friedman classification, with positions III and IV conferring a high risk for OSA (Figure 49-6). In general, the less posterior oropharynx that can be visualized confers a greater risk of SDB. Neck circumference greater than 17 inches in men and 16 inches in women is another predictor for SDB on physical examination. However, the predictive value of neck circumference is not as great in older patients.

The presence of disturbed sleep in a patient with congestive heart failure may be indicative of SDB. A high percentage of patients with heart failure have OSA, CSA with Cheyne-Stokes respiration, or a combination of the two conditions. It can be difficult to tease apart symptoms of heart failure such as orthopnea from those of SDB. Opioid medication use is a risk factor for another form of CSA, which should be considered when patients on opioids present with sleep complaints.

Evaluation

Patients suspected of having SDB should be referred for sleep testing. An attended in-laboratory polysomnography (PSG) is considered the gold standard test for SDB. This involves traveling to a sleep laboratory to spend the night. The test involves the use of EEG, electromyogram (EMG), and electrooculogram (EOG) leads to determine presence and stage of sleep. There is measurement of airflow with a nasal pressure transducer as well as an oronasal thermistor. Respiratory effort is measured using chest and abdominal belts; and occasionally with an esophageal pressure monitor or accessory muscle EMG. An alternative approach to the diagnosis of SDB uses out-of-center testing. The patient takes the sleep testing equipment home and puts it on before bed. These tests are not generally able to measure sleep directly. Most of these tests measure airflow, effort, and pulse oximetry. Out-of-center sleep testing is only recommended for patients with a high pretest probability of having SDB who do not have significant cardiopulmonary comorbidities.

The advantages of an attended full-channel PSG include a higher reliability of the data, the ability to carefully define AHI by measuring sleep directly, and the simultaneous assessment for alternative sleep diagnoses to SDB. The advantages of out-of-center sleep testing include lower cost, the convenience for patients to sleep at home, and increased access to sleep testing. In populations at high risk for OSA, studies have shown that out-of-center sleep testing leads to similar patient outcomes compared to in-laboratory PSG. Due to cost issues third-party payers are increasingly pressuring practitioners to move toward out-of-center sleep testing. However, out-of-sleep center testing may be difficult for some older patients due to the need for the patient or caregiver to independently place the sensors and the contraindication for use when conditions that may lead to CSA are present. It must also be emphasized that out-of-center sleep testing does not rule out SDB, but a positive test confirms SDB. An out-of-center sleep test may be negative simply because the patient did not sleep during the test or due to technically inadequate data.

Management

The first-line treatment for SDB is positive airway pressure therapy. The most basic of these therapies is CPAP, which has been available since the early 1980s. CPAP is a device that delivers a constant positive pressure in order to splint the airway open preventing collapse. To apply this pressure an air-tight interface goes in the nose, over the nose, or over the nose and mouth, and then connects to the CPAP machine via a hose. Traditionally, the amount of air pressure (usually between 5 and 20 cm H₂O) is manually determined in the sleep laboratory with the sleep technologist adjusting the pressure to eliminate respiratory events and snoring. There is evidence that patients who have mild to moderate dementia are able to tolerate CPAP therapy and such a condition is not a contraindication to CPAP initiation.

Autotitrating CPAP (APAP) is now available, in which the machine adjusts the pressure based on events and flow characteristics using internal algorithms that are proprietary to the device manufacturers. APAP can be used to determine a fixed CPAP pressure or can be used as the primary modality of therapy. APAP has small advantages of questionable clinical significance for adherence to therapy in all-comers. Individual patients, however, particularly those with very different CPAP needs depending on body position or sleep stage, may find APAP to be more comfortable than fixed CPAP. The main advantage of APAP is the ability to determine, or provide, a therapeutic pressure without requiring a resource-intensive in-laboratory PSG.

CPAP is usually 100% effective in eliminating obstructive apneas and hypopneas. However, long-term adherence is the major limiting factor to this treatment modality. Estimates are that at the end of a year only 50% of patients are adherent to CPAP therapy. Factors that may improve CPAP adherence include early follow-up, education and setting of expectations, humidification, attention to interface comfort, expiratory pressure relief, possible early use of hypnotics, and desensitization to therapy. Each of these interventions has small but important impacts in individual patient acceptance to therapy. Bilevel airway pressure has sometimes been used for patients who have difficulty tolerating therapy due to difficulty exhaling against pressure.

Positive airway pressure therapy has to be individualized in patients who have CSA. Many patients who have Cheyne-Stokes respiration will respond to a combination of CPAP and/or supplemental oxygen. Likewise, the majority of patients with obesity-hypoventilation syndrome respond to CPAP alone. However, many patients require more advanced ventilatory modalities to eliminate CSA. Regardless of the particular mode, these advanced devices involve applying bilevel pressure with a back-up respiratory rate. The amount of pressure support given during a certain breath can be controlled by sophisticated proprietary algorithms in order to achieve specific desired clinical effects.

Dental devices that move the jaw forward (mandibular advancement or mandibular repositioning devices) are a treatment alternative to positive airway pressure in OSA. The principal behind such therapy is that when the lower jaw moves forward the tongue is also pulled forward, thus opening up the airway. These devices are generally thought to be more effective in mild to moderate OSA rather than severe OSA. However, the severity of OSA may not necessarily predict response to mandibular advancement therapy in all patients. In general, mandibular advancement therapy does not reduce the residual AHI to as great of an extent as positive airway pressure therapy. However, recent comparative effectiveness studies suggest equivalence for major outcomes likely due to the fact that adherence rates tend to be higher for mandibular advancement therapy. Side effects include temporomandibular joint pain and occlusion abnormalities. One of the major limiting factors to mandibular advancement therapy, particularly in older adults, is that teeth are generally required to anchor the device in place. However, more specialized devices can sometimes be fabricated for edentulous patients. Tongue repositioning/retaining devices can be used in patients who lack their native teeth. However, these devices have been less extensively studied and are less likely to be effective.

Other therapies have been utilized in OSA. One of these therapies is expiratory positive airway pressure (EPAP) provided through nasal valves. This therapy consists of nasal valves that adhere to the nares via adhesive, which makes it more difficult to breathe out than breathe in. This causes pressure build up that helps stent the airway open, as well as causing an increase in lung volume that may prevent obstructive respiratory events. This therapy is less well validated but may be used as a salvage therapy or in mild cases. Another therapy that is available with limited evidence involves the application of negative pressure to the mouth to reposition the soft palate and tongue out of the airway.

One therapeutic modality that is on the horizon is hypoglossal nerve stimulation. This involves an implanted box with a lead that goes to the respiratory muscles to sense the onset of inspiration and a lead that goes to the hypoglossal nerve that causes tongue contraction to open the airway with inspiration. A recent randomized trial found promising results in a highly selected group of patients with a BMI less than 32 kg/m² who could not tolerate CPAP therapy. In this group there was a significant reduction in AHI. However, there were not many older adults included in this study.

Surgery is also an alternative therapy for SDB. There are multiple surgical techniques that are available to otolaryngologists for the treatment of SDB. The most common of which is the uvulopalatopharyngoplasty (UPPP). This procedure has variable success rates depending on anatomic factors and the degree of obesity. Furthermore, even after successful therapy SDB may reoccur due to growth of scar tissue.

A more aggressive surgical approach is a maxillary mandibular advancement. In this approach both the maxilla and mandible are cut and advanced forward. This surgery requires the patient’s jaw to be wired shut for several weeks. Although this is the most effective surgery for SDB, many patients opt to not undergo such an extensive procedure.

Other behavioral changes that the patients with SDB can implement to help treat the syndrome include weight loss, sleeping on the side, and the avoidance of alcohol and other sedatives. In particular, positional techniques to encourage sleeping on the side may be beneficial when SDB occurs primarily in the supine position.

It is becoming increasingly recognized that while positive airway pressure therapy eliminates SDB more effectively than other interventions it is essentially ineffective in a subset of patients who cannot tolerate its use. The use of multiple modalities of therapy, with each having a modest effect on AHI, may be able to achieve adequate control of SDB in those who cannot tolerate CPAP.

Pharmacologic therapy

There are limited pharmacologic agents available for SDB and none that are approved for this indication. However, occasionally patients who are successfully treated for SDB with CPAP will have residual excessive daytime sleepiness. Modafinil and armodafinil, central nervous system stimulants, have Food and Drug Administration (FDA) approval for the treatment of residual excessive daytime sleepiness in treated OSA. Acetazolamide has an unclear role in some CSA syndromes. Ventilatory stimulants such has progesterone have been tried unsuccessfully in obesity hypoventilation syndrome.

Prevention

There is no specific disease prevention strategy for SDB. Addressing factors that contribute the subsequent development of SDB contributes to prevention. Since elevated BMI is the most potent risk factor for the development of OSA, weight loss and lack of weight gain are the most important preventative measures for OSA. Preventing decompensation of heart failure may be a way to prevent the development of heart failure–related CSA.

Patient Preference

Patient preference is clearly a key factor in determining the therapy of choice. CPAP and mandibular advancement devices can both be difficult to tolerate due to the need for equipment use and potential discomfort. The health benefit of treating minimally symptomatic SDB in older patients is unclear. Therefore, it is reasonable to withhold SDB treatment if the patient declines treatment and they have minimal symptoms. In patients who are sleepy and drive, it is more imperative to encourage SDB therapy. Also, in sleepy older patients with SDB the treatment may have a greater beneficial impact on the prevention of SDB-related comorbidities and improvement in quality of life.

Comorbidity

SDB is associated with significant mental and physical comorbidities. Given the association between obesity and SDB, many of the comorbidities of obesity and SDB overlap. However, there are several lines of evidence suggesting an independent relationship between SDB and the development of hypertension, atrial arrhythmias, heart failure, coronary artery disease, cerebrovascular accidents, and diabetes mellitus. Most of these relationships have been elucidated from observational population studies and basic science investigations. Direct evidence that treatment of SDB can reverse some of these comorbidities is more limited. However, treatment of OSA does result in modest improvement in blood pressure among hypertensive patients. In addition, treatment of OSA after cardioversion for atrial arrhythmias such as atrial fibrillation has been found to increase the likelihood of staying in sinus rhythm. Other investigations have been limited by sample size and poor adherence to positive airway pressure therapy. Treatment of CSA using CPAP was studied almost a decade ago in the CANPAP trial. There was no evidence of a survival advantage to treating heart failure–related CSA with CPAP. However, CPAP did not universally suppress CSA in that study. In fact, subsequent post-hoc analysis of data from the CANPAP trial showed that there was a mortality benefit in the subset of patients in whom the CSA was suppressed. Currently, there are ongoing randomized controlled multicenter trials looking at mortality outcomes in bilevel devices, which are better able to eliminate CSA in heart failure.

Insomnia is an important comorbidity of SDB. In fact, SDB itself may contribute to insomnia by disrupting sleep architecture and even preventing stable sleep onset due to OSA respiratory events. Sometimes insomnia improves with the treatment of SDB. However, insomnia may be exacerbated by the invasiveness of positive airway pressure or mandibular advancement. For a successful treatment outcome both disorders must be addressed. SDB and other mental illnesses such as anxiety, depression, and posttraumatic stress are frequently present in the same patient. Once again, the SDB may exacerbate some of the symptoms of these mental illnesses, and these illnesses may present barriers to acceptance of therapy.

Definition

Restless legs syndrome (RLS) is characterized by the urge to move one’s legs. Most often patients with RLS have uncomfortable sensations in their legs, and sometimes arms, which lead to this urge. The other defining features of RLS are that the symptoms are improved with movement, worsen with relaxation, and occur later in the day. RLS can also affect and may even be limited to the arms. There is no objective test that defines RLS.

Periodic leg movements in sleep (PLMS) are characterized by cyclic movements of the lower limbs during sleep. These consist of stereotyped movements that typically involve the flexion of the big toe with partial flexion of the ankle, knee, and sometimes hip. PLMS last between 0.5 seconds and 10 seconds with a period between movements of 5 to 90 seconds (Figure 49-7). At least four movements in a row must occur to count as PLMS. Periodic limb movement disorder (PLMD) can be diagnosed if there are at least 15 PLMS per hour, the PLMS cause sleep disturbance and the symptoms are not better explained by another sleep disorder or a medical illness. Since PLMS are often seen in RLS (80%–90%), REM behavior disorder (70%), and narcolepsy (45%–65%), PLMD can only be diagnosed in the absence of these sleep disorders. A sleep study is required to define PLMS and thus diagnose PLMD.

Epidemiology

The prevalence of RLS has been estimated to be between 5% and 10%, making it a very commonly encountered disorder in the outpatient setting. Prevalence increases with age in most populations up to 60 to 70 years. RLS is 1.5 to 2 times more common in women than in men. Iron deficiency and family history increase the odds of developing RLS. The prevalence of RLS is two to five times greater in patients with chronic renal failure. Taking antihistamines and most antidepressants are risk factors for the development of RLS. RLS severity correlates with a reduced health-related quality of life when comorbid with other conditions. Large population studies have found associations between RLS and cardiovascular disease.

The exact prevalence of PLMD is not known. PLMS increases in frequency with age. Up to 45% of older people have PLMS, often of unknown clinical consequence. The existence of comorbidities with aging makes a clear-cut diagnosis of PLMD difficult. Having a family history of RLS may make PLMD more likely.

Pathophysiology

RLS and PLMS are often thought of as sharing a similar pathophysiology, with RLS being a sensory manifestation and PLMS a motor manifestation. The exact mechanisms of RLS and PLMS are not exactly known. There appears to be a defect in the homeostatic regulation of iron that then leads to a dysfunction in the dopamine system.

Clinical Presentation

RLS may present with complaints of specific RLS symptoms or simply with a primary complaint of insomnia. The most classic presentation is of a creepy crawly sensation that is predominately in the lower calf and gets better with walking. It is worse at night and with rest. Sometimes RLS symptoms involve the upper extremities as well. By definition the symptoms occur while awake. In patients who are not able to describe symptoms due to significant dementia or aphasia the patients may be seen rubbing or massaging the legs or may present with increased motor activity at night. Since 80% to 90% of patients with RLS have PLMS the observation of kicking during sleep makes the diagnosis of RLS more likely. One of the difficult issues clinically is distinguishing RLS symptoms from peripheral neuropathy. The nighttime onset of symptoms and relief of symptoms with movement help distinguish the two conditions. For patients with both conditions present, determining the contribution of RLS over neuropathy remains challenging.

PLMD may present with a patient or bed partner complaint of kicking at night. The patient may simply present with complaints of disturbed sleep, insomnia, and/or excessive daytime sleepiness. Since PLMS are so common in older adults, significant clinical sleep disturbance and/or daytime sleepiness are required to make the diagnosis. The symptom of waking up kicking has to be distinguished from hypnic jerks or sleep starts—total body jerking often with a sensation of falling that occur at sleep onset. Hypnic jerks are very common and generally not pathologic.

Evaluation

The diagnosis of RLS can be based on clinical criteria alone. A PSG is sometimes helpful to rule out other sleep disorders, particularly SDB which may be comorbid or on the differential diagnosis for the patient’s sleep disturbance. Presence of PLMS on PSG is suggestive of RLS but neither rules in nor rules out RLS.

A sleep study with limb electrode measurement is required for the diagnosis of PLMD because this is the only way for PLMS to be accurately measured. Furthermore, since PLMD is a diagnosis of exclusion other sleep disorders must be ruled out at the same time.

Patients with RLS or PLMD should have a serum ferritin measured to rule out any degree of iron deficiency leading to these conditions.

Management

Pharmacologic management

The treatment options for RLS and PLMD are similar. However, whether to initiate treatment for PLMD is more controversial. For either condition the decision to treat needs to be individualized, based on the severity of symptoms and the impact on quality of life.

Patients should receive iron supplementation if the serum ferritin is less than 50 to 70 ng/dL. This entails a much more aggressive iron replacement strategy than is typically used in patients without RLS.

The first-line specific pharmacologic treatment is an evening dose of a long-acting dopamine agonist (either pramipexole or ropinirole) 1 to 2 hours prior to bedtime (Table 49-2). Carbidopa-levodopa can be used in patients for as-needed dosing when symptoms are infrequent. The major drawback to the use of dopamine agents is a phenomenon called augmentation. Augmentation is a worsening of RLS manifested by: the onset of RLS symptoms earlier in the day, and/or increased intensity of RLS symptoms, and/or spread of symptoms to the upper extremities and trunk when these areas were previously unaffected. Augmentation is much more frequent with carbidopa-levodopa, which is why this agent is reserved for use when symptoms are infrequent. Estimates of augmentation on the long-acting dopamine agonists have ranged between 2% and 15% per year. The risk of augmentation is higher with higher doses and longer duration of therapy. Augmentation can sometimes be managed by giving an additional dose of dopamine agonist earlier in the day, switching to a different dopamine agonist or switching to a different class of medication. Other important side effects of dopamine agonists include daytime sleepiness, sleep attacks, and behavioral disinhibition.

The other agents that have good evidence for RLS therapy are the α2δ calcium channel ligands, which include gabapentin, gabapentin enacarbil, and pregabalin. Only gabapentin enacarbil is FDA approved for RLS treatment. These agents may be particularly useful when the patient also has peripheral neuropathy. A recent randomized controlled trial showed that pregabalin was as efficacious as pramipexole, with a lower rate of augmentation. Major adverse effects from these medications can include drowsiness, dizziness, and suicidal ideation.

Other medications that can be used for the treatment of RLS include benzodiazepines and opiates. Benzodiazepines have a long history of use for intermittent RLS; however, the evidence base for this practice is limited. Low-dose opiates including methadone can be used for treatment of refractory RLS. Clearly, benzodiazepines and opiates can have adverse cognitive effects in older patients and should be used with caution.

Nonpharmacologic therapy

Behavioral strategies that can be used for RLS include moderate exercise, pneumatic compression, massage, hot showers, and leg stretching. Avoidance of possible triggers for RLS should also be recommended. These include sleep deprivation, caffeine, and alcohol use. Also, most antidepressants are known to elicit or provoke RLS symptoms. However, bupropion is less frequently associated with RLS and therefore a switch from a selective serotonin reuptake inhibitor (SSRIs) or serotonin-norepinephrine reuptake inhibitors (SNRIs) to bupropion when RLS symptoms are present could be considered.

Prevention

There is no specific prevention for RLS. Adherence to a healthy lifestyle may be beneficial. This includes moderate exercise, a regular sleep schedule, and limiting caffeine and alcohol intake.

Patient Preference

The focus of RLS management is symptom relief. The medications used for the treatment of RLS may have important side effects. Therefore, therapy is determined by the balance between symptom control and side effects, in addition to patient input and preferences.

Comorbidities

Frequently encountered comorbidities of RLS include Parkinson disease, chronic kidney disease, iron deficiency anemia, and peripheral neuropathy. It is essential to address both the underlying disease process and the RLS symptoms at the same time.

Definition

REM sleep behavior disorder (RBD) is defined as recurrent complex movements or vocalizations during sleep characterized by dream enactment. Normally during REM sleep, there is complete loss of skeletal muscle tone (ie, atonia). However, in RBD there is REM sleep without atonia, which is required for the diagnosis of RBD, in addition to dream enactment behavior.

Epidemiology

RBD is a male-predominant syndrome, which usually occurs after the age of 50. The exact prevalence of RBD is not known; however, it is estimated to be from 0.4% to 0.8%. RBD is more common in certain neurologic disorders such as Parkinson disease, multiple system atrophy, and Lewy body dementia.

Pathophysiology

There is an association between neurodegenerative disorders and RBD, particularly synucleinopathies. The synucleinopathies result from a pathologic lesion of aggregates of insoluble α-synuclein protein. These aggregates are linked to degradation of specific brain regions and clinical symptoms. The most important clinical diagnoses of synucleinopathies are Parkinson disease, dementia with Lewy bodies, and multiple system atrophy. The α-synuclein protein aggregates can be found in the brain areas responsible for producing active atonia during REM sleep. Up to 90% of patients who present with RBD develop another synucleinopathy disorder over the course of a 10- to 15-year follow-up (Figure 49-8).

RBD is also strongly linked with narcolepsy, which may be a different form of REM sleep motor-behavior control. The RBD may be worsened by the use of SSRIs to treat cataplexy. Other neurologic disorders have been associated with RBD including cerebrovascular disease, multiple sclerosis, Guillain-Barre syndrome, normal pressure hydrocephalus, Tourette syndrome, and autism.

Clinical Presentation

RBD will often present with sleep-related injury that results from the enactment of unpleasant violent dreams. The dreams that are acted out are often more violent and thus the behaviors are violent. It is rare for patients with RBD to walk. Patients are usually not able to leave the room they are in because their eyes are closed and thus are inattentive to the environment. Because RBD occurs during REM sleep, it usually appears at least 90 minutes after sleep onset and is more likely to occur later in the sleep period. Often RBD is the first manifestation of a Parkinson disease or dementia with Lewy bodies, but may also present later in patients with these disorders.

Evaluation

The evaluation of RBD includes a neurologic and sleep history. The history should focus on common associated conditions such as Parkinson disease and associated dementia. The presence of vivid dream enactment makes RBD more likely than other parasomnias. A predilection for events to occur later in the sleep period also makes RBD more likely. A detailed neurologic examination can elucidate a comorbid neurodegenerative disorder. It may be difficult to distinguish RBD symptoms from those of other parasomnias, nocturnal seizures, and nightmares of posttraumatic stress disorder (PTSD). SDB can cause RBD-like symptoms, which typically resolve with effective SDB therapy.

An attended in-laboratory PSG is the gold standard for diagnosing RBD. Actual acting out of dreams in REM sleep during the sleep study is fairly convincing for RBD. However, often patients will not have RBD behavior events during PSG. In this case, a 30-second epoch of REM sleep during the sleep study in which over 50% of the epoch has increased muscle tone is consistent with RBD. The majority of patients with RBD will also have PLMS on their PSG.

Managegement

Pharmacologic management

There are no currently FDA-approved therapies for RBD. Clonazepam at low doses has been successfully used to treat RBD symptoms. As with any benzodiazepines, caution needs to be exercised in older patients with this drug. The other therapy that has been used with success is melatonin at 3- to 6-mg doses, which is generally well tolerated in older people. It is possible that since REM sleep occurs later in the night that the extended-release melatonin may be more effective. The major drawback to melatonin is that it is not an FDA-regulated pharmaceutical. Therefore, quality and strength may not be as tightly regulated. Also, insurance will usually not cover the expense of this therapy.

Nonpharmacologic therapy

Ensuring that the patient sleeps in a safe environment and that others are not in danger is of paramount importance in RBD. The patient should consider sleeping separately from their bed partner for that person’s safety. Sharp furniture and other sharp objects should be removed from the patient’s bedroom. Likewise any weapons should not be freely accessible. If falling out of bed has been an issue, moving the mattress to the floor can be considered. Some patients have gone so far as to sleep in sleeping bags to prevent injuring themselves.

Well-recognized precipitating factors for RBD symptoms include the use of SSRIs, venlafaxine, mirtazapine, and other antidepressants except for bupropion. Hence, tapering off these medications or switching to bupropion should be considered. Any sort of comorbid SDB disorder should be aggressively treated, which may help with RBD symptoms.

Prevention

There are no known preventative measures for RBD.

Patient Preference

Since the treatment of RBD is symptom based, patient preferences often dictate the intensity and type of therapy. Many patients may opt for melatonin therapy given that its side effect profile appears to be favorable to clonazepam. The desire for the patient to know the ramifications of an RBD diagnosis as a predictor of development of a future neurodegenerative disorder needs to be carefully elucidated.

Comorbidities

As discussed previously presence of RBD strongly predicts the subsequent development of a synucleinopathy such as Parkinson disease, multiple system atrophy, and dementia with Lewy bodies. Often RBD is the first manifestation of one of these disorders. Therefore, RBD will eventually coexist with one of these neurodegenerative disorders in most patients. RBD can also present in patients with narcolepsy. There are some cases in which RBD and NREM parasomnias coexist as well.

Other parasomnias occur in NREM sleep, which include sleep walking, confusional arousals, and sleep terrors. These disorders are much more common in children and young adults, and generally do not first present in old age. The distinguishing features of these conditions are that they predominantly occur in the first part of the night and the patients are difficult to arouse and do not have recollection of the event or associated dream imagery. One aspect of these NREM parasomnias relevant to older adults is that nonbenzodiazepine hypnotics can be associated with very complex sleep behaviors such as sleep cooking, sleep eating, and sleep driving. This is a rare but possibly dangerous side effect of these medications.

Definition

Circadian rhythms refer to the 24-hour biological rhythms that control functions such as hormone secretion, core body temperature, and the sleep-wake cycle. The endogenous circadian rhythms originate in the suprachiasmatic nucleus (SCN), which is the internal circadian pacemaker. The SCN synchronizes itself using other internal rhythms from hormone secretion patterns and external signals, the most potent of which is light. The sleep-wake cycle is driven by melatonin secretion and body temperature, which are both governed by the SCN. However, external light can disrupt this cycle.

Circadian rhythm disorders are a chronic or recurrent pattern of sleep-wake rhythm disruption due to a misalignment between the patient’s endogenous circadian rhythm and their desired or required sleep-wake cycle. For this to be considered a disorder it must cause bothersome insomnia symptoms and/or excessive sleepiness. For instance, a retired person who is not bothered by going to bed at 8 PM and waking up at 4 AM does not have a disorder.

Delayed sleep-wake phase disorder is a delay in the phase of the endogenous major sleep episode with respect to the desired major sleep episode such that the patient complains of an inability to fall asleep and difficulty waking up at the desired time.

Advanced sleep-wake phase disorder is an advance in the phase of the endogenous major sleep episode with respect to the desired major sleep episode such that the patient complains of difficulty staying awake until the desired bedtime, with an inability to remain asleep until the desired awakening time.

Irregular sleep-wake rhythm disorder is characterized by a lack of a clearly defined circadian rhythm characterized by insomnia at night and excessive sleep during the day. This disorder is often associated with neurodegenerative diseases and is likely common in the nursing home setting (Figure 49-9).

Non–24-hour sleep-wake rhythm disorder usually occurs from a lack of the ability to entrain their circadian rhythm to light due to complete blindness. The natural circadian rhythm is typically longer than 24 hours. Thus each subsequent day, the patient will go to bed and wake up a little later. This leads to episodes of insomnia and hypersomnia (Figure 49-10).

Epidemiology

Delayed sleep-wake phase disorder is more common among adolescents and young adults; however familial patterns and psychiatric disorders are risk factors for development of delayed sleep-wake phase disorder in later life.

The prevalence of advanced sleep-wake disorder is not fully known and is thought to comprise 1% of a middle-aged population. Advanced age appears to be a risk factor, given that aging is associated with the development of morningness tendencies. However, older patients may be more tolerant of these changes in endogenous circadian rhythm.

Irregular sleep-wake rhythm disorder is commonly seen in older adults with neurodegenerative disorders such as Alzheimer disease, Parkinson disease, and Huntington disease.

Pathophysiology

Possible causes of advanced sleep-wake phase disorder include a loss of the ability for the circadian clock to phase delay, a tendency to abnormally phase advance to entraining stimuli or a shorter endogenous circadian pacemaker. The latter of these has been shown to underlie the pathophysiology in select familial patients with circadian clock gene mutations.

There are several factors that may cause irregular circadian rhythms with aging. These may include (1) degeneration of the SCN with age, (2) a blunting of melatonin secretion at night, (3) decreased sensitivity to entraining cues, and (4) less availability to cues such as bright light. Furthermore, neurodegenerative disorders may result in physical disruption of circadian rhythm pathways. Irregular sleep patterns prevent the normal sequence of progression through sleep stages.

Clinical Presentation

Patients with delayed sleep phase disorder present with sleepiness in the morning and an inability to sleep in the evening. Patients with advanced sleep phase disorder present with sleepiness in the evening, and an inability to sleep in the early morning. These patients will essentially present with insomnia complaints. However, an ability to sleep when the patient aligns their sleep-wake cycle with their endogenous circadian rhythm will be indicative of a circadian rhythm disorder rather than insomnia. Patients with advanced sleep phase syndrome may develop a self-imposed sleep deprivation from trying to adhere to a “normal schedule” but will wake up too early. A strong familial pattern may also suggest a circadian disorder.

Patients with irregular sleep-wake rhythm disorder will present with poorly consolidated sleep. Caregivers of patients with dementia may complain of nighttime wandering and agitation. In fact, sundowning may be a phenotype of a circadian rhythm disturbance. Patients will generally sleep for less than 4 hours at a time, and will have many short periods of sleep and wake throughout the day and night.

Evaluation

The key component in evaluating patients with suspected circadian rhythm disturbances is a comprehensive sleep history. Sleep logs for 1 to 2 weeks are helpful in establishing the diagnosis of a circadian disorder. Wrist actigraphy (a device with an accelerometer) can also be useful, particularly in patients who cannot fill out sleep logs, or for those in whom the accuracy of the sleep logs is doubtful. Newer actigraphy monitors that have a light sensor can also give the clinician a sense of the patient’s light exposure. Polysomnography is only indicated when other sleep disorders are suspected.

Management

The primary management for circadian rhythm disorders involves the use of appropriately timed bright light therapy. Appropriately timed melatonin can be used as an alternative to, or in conjunction with, bright light therapy. However, there is less evidence for this approach. There are commercially available light boxes, or where feasible, natural sunlight can be used. In general, exposure to bright light in the morning will advance the circadian rhythm, while exposure in the evening will cause a delay in rhythm. Patients with advanced sleep-wake phase disorder should wear sunglasses in the morning and spend late afternoons outdoors without sunglasses.

For irregular sleep-wake rhythm disorder a strong routine with natural light and increased activity during the day and darkness with decreased noise at night may be helpful. Tasimelteon is a newly released melatonin agonist for the MT1 and MT2 receptors with greater affinity for the MT2 receptor. This agent is only approved for non–24-hour sleep-wake rhythm disorder.

Definition

Insomnia is defined as the inability to fall asleep, the inability to stay asleep, or waking up earlier than desired. In order to have a clinical diagnosis of insomnia the patient must have an adequate sleep opportunity and adequate sleep environment. The sleep disturbance must also have an impact on their quality of life such as fatigue, impaired cognitive performance, mood disturbance, daytime sleepiness, behavioral problems, reduced motivation, proneness for errors, or concerns about sleep. Insomnia is categorized as chronic if it persists more than 3 months and short term if it has lasted fewer than 3 months. The most recent International Classification of Sleep Disorders no longer emphasizes previously distinguished insomnia subtypes or insomnia due to mental or medical disorders. This is because it is increasingly recognized that even when insomnia is related to another condition, treatment of the comorbid condition often does not cure the insomnia. Furthermore, individual patients often have an overlap across insomnia subtypes.

Epidemiology

Insomnia is a highly prevalent sleep disorder, affecting up to 10% of young adults; however it increases to about 30% in those greater than 65 years. Transient insomnia may affect up to 30% to 40% of the population. The prevalence of insomnia is much higher in older adults, which is likely due to age-related reductions in sleep efficiency and the accrual of comorbidities that are associated with insomnia. A recent meta-analysis of epidemiologic studies shows that the relative risk of insomnia in women compared with men increases with age.

Comorbid psychiatric conditions increase the likelihood of developing chronic insomnia. Depression is perhaps the most common and strongly associated mental illness with insomnia. Anxiety is also a risk factor for developing insomnia; as is being a caregiver for an ill family member, with up to 40% of caregivers reporting taking sleep aids for insomnia.

A wide variety of medical problems are associated with insomnia. Epidemiologic evidence shows a greater prevalence of insomnia in hypertension, heart disease, arthritis, lung disease, gastrointestinal reflux, stroke, and neurodegenerative disorders, to only name a few. Symptoms of medical illnesses that can disrupt sleep include pain, paresthesias, cough, dyspnea, reflux, and nocturia.

Many medications can impair sleep or change sleep architecture. If stimulating medications (eg, caffeine, sympathomimetics, bronchodilators, activating psychiatric medications) are taken too near to bedtime, sleep can be disturbed (Table 49-3).

Late-life insomnia is often a long-lasting problem. One study showed that a third of older patients had persistent severe insomnia symptoms at 4-year follow-up. Among women older than 85 years, more than 80% reported sleeping difficulties, with many using over-the-counter (OTC) sleeping medications. Several epidemiologic studies have linked sleep disturbances with worse health-related quality of life, nursing home placement, and even death.

Pathophysiology

The causes of insomnia are often multifaceted. One popular theoretical model for insomnia posits that there are predisposing factors, precipitating factors, and perpetuating factors. Predisposing factors are a vulnerability to insomnia and may include anxiety, depression, or hyperarousal. Precipitating factors are triggers for insomnia such as loss of a spouse, retirement, moving to a new home, or any other sort of stressor. Perpetuating factors are maladaptive habits or beliefs that the patient has acquired to deal with the insomnia such as spending long periods in bed or taking naps.

As noted earlier, insomnia in older adults is often comorbid with medical or psychiatric illnesses. The symptoms such as pain, dyspnea, or nocturia may be so significant that these become the primary driving factors behind the sleep disturbance.

Clinical Presentation

Patients with insomnia may present to their primary care provider with specific complaints about either not being able to fall asleep or stay asleep. However, many patients do not talk to their doctors about their sleep complaints. Instead, the presence of insomnia may be revealed by eliciting self-medication with OTC or alternative sedatives.

A careful sleep history and evaluation should be conducted in patients with insomnia in order to assess whether there are comorbid sleep disorders, such as SDB, underlying the sleep disturbance.

Evaluation

PSG and other sleep studies are not necessary in the evaluation of insomnia; however, they should be obtained if a comorbid sleep condition is suspected. Sleep diaries with daily entries over 1 to 2 weeks can be very helpful in determining the severity of the insomnia as well as identifying possible perpetuating factors such as irregular bedtimes or late-night caffeine (Table 49-4). This may help the patient gain more insight into their sleep problem. Wrist actigraphy in conjunction with a sleep diary can be used to obtain a more objective measure of the patient’s overall sleep-wake pattern. This device is particularly helpful when accuracy of the sleep diary is questionable or when sleep-wake misperception is suspected.

Management

Behavioral and nonpharmacologic interventions appear to have better long-term efficacy with fewer side effects than pharmacologic interventions. Therefore, behavioral interventions are preferred for the treatment of insomnia, particularly in older patients. However, access to highly skilled professionals trained in such interventions may be limited in some health care systems.

Pharmacologic management

Most pharmacotherapy for insomnia is designed and approved for the treatment of transient sleep disturbances. Pharmacotherapy can be considered when insomnia is triggered by an acute event or when chronic insomnia persists despite behavioral insomnia treatment. Sedative-hypnotics have been associated with adverse side effects in older adults, such as falls, cognitive slowing, and fractures. Therefore, if a sedative-hypnotic is used in older individuals, the smallest dose of an agent with the least risk of adverse events should be chosen for the shortest duration necessary. In general, short-acting drugs should be used for patients who have trouble falling asleep while intermediate-acting drugs should be used when patients have trouble staying asleep (Table 49-5).

Benzodiazepines have had a long history of being used for insomnia. These drugs bind nonselectively to the γ-aminobutyric acid benzodiazepine (GABA-A) receptor subunits. The overall effects of these drugs are to induce anxiolysis, sedation, and amnesia. Temazepam, lorazepam, and estazolam are intermediate-acting agents that are most commonly used for insomnia, while triazolam is a shorter-acting agent that is sometimes used for insomnia. Due to next-day effects, the longer-acting agents, flurazepam and quazepam, should not be used in older adults. Benzodiazepines do decrease the time to fall asleep by about 10 minutes on average and the number of awakenings, thereby increasing the total sleep time by 30 to 60 minutes during the nighttime. The side effects associated with this class of medications include confusion, falls, rebound insomnia, tolerance, and withdrawal symptoms on discontinuation. These agents are all Beer criteria–listed medications as potentially inappropriate for use in geriatric patients.

Nonbenzodiazepine-benzodiazepine receptor agonists (NBRAs [ie, nonbenzodiazepines such as eszopiclone, zolpidem, and zaleplon]) are structurally unrelated to benzodiazepines but bind selectively to the GABA-A receptors. They generally produce sedation and amnestic effects without the anxiolytic properties. These agents likely have similar efficacy to benzodiazepines but with a somewhat better side effect profile, in part due to their relatively short duration of action. In healthy older adults without comorbidities, NBRAs are relatively well tolerated. Zolpidem and zaleplon should only be taken immediately before bed because of their rapid onset of action. Eszopiclone has a longer duration of action than the other NRBAs, and is better for sleep maintenance but may cause drowsiness in the morning. There is also an extended-release form of zolpidem that can be used for sleep maintenance insomnia. Guidelines only recommend these medications for use in short-term insomnia. Concerns remain regarding the risks of falls, confusion, and fracture in older adults, particularly in those with frailty. The emergence of complex sleep-related behaviors such as sleep driving and sleep eating has been seen with zolpidem, which may be class side effect with these agents.

Melatonin receptor agonists (eg, ramelteon) are approved for insomnia. Rather than activating GABA receptors, ramelteon is selective for the MT1/MT2 melatonin receptors. Ramelteon has been shown to reduce total sleep latency and increase sleep time in older adults, with fewer side effects. Somnolence, dizziness, headache, and fatigue can be side effects of this agent. The major downside of ramelteon is that its lack of amnestic properties makes it somewhat less effective for subjective improvement in sleep.

Other agents are available for treatment of insomnia. The tricyclic antidepressant doxepin has recently been developed for insomnia in a low-dose formulation (3–6 mg). For comparison, the usual antidepressant dose of doxepin is 100 to 300 mg per day. At these low dosages doxepin selectively antagonizes H₁ receptors, which is believed to have sleep-promoting effects. The duration of action for doxepin makes it more attractive for problems with sleep maintenance rather than sleep onset. The use of low doses of other sedating antidepressants such as trazodone and mirtazapine at bedtime is common. However, there is limited evidence to support this practice. In fact, a study in the use of trazodone for nondepressed patients showed that there was short-term benefit, which dissipated after 2 weeks. Thus, the benefits may be very small compared to the potential for side effects. However, if there is comorbid depression it should be treated, and sedating antidepressants may be a good choice as primary or adjunctive therapy.

Although sedating antipsychotics are sometimes used to treat insomnia, there is scant evidence demonstrating their efficacy, and these medications can have significant adverse events, including increased mortality particularly in older adults with dementia. Guidelines suggest that sedating antipsychotics should not routinely be used in the management of insomnia in older adults without coexisting serious psychiatric conditions that warrant use of these agents.

Almost half of all older adults report the use of nonprescription OTC sleeping agents. Common nonprescription agents include sedating antihistamines, acetaminophen, alcohol, melatonin, and herbal products. The sedating antihistamines (eg, diphenhydramine) are the most common ingredients in these OTC drugs marketed for sleep. Diphenhydramine is sedating through its potent anticholinergic effect, and tolerance to its sedating effect develops rapidly. The long half-life of diphenhydramine may result in next-day sedation. Patients may also experience dry mouth, urinary retention, delirium, decreased cognition, constipation, and increased ocular pressure. For these reasons, diphenhydramine should not be used as a primary treatment for insomnia in older patients. Many patients may use alcohol to self-medicate for insomnia; however, it can interfere with sleep in the later evening and worsen sleep difficulties.

Other supplements are available OTC, such as melatonin and valerian. One drawback of these supplements is the lack of standardization and oversight of traditional pharmaceuticals. There is evidence that melatonin improves time to fall asleep and sleep efficiency in older adults. However, the results of studies have been mixed. Valerian is a herbal product marketed for insomnia due to its mild sedative properties. Some valerian preparations contain multiple botanicals, which may increase the chance of side effects. A recent systematic review on the efficacy for valerian for insomnia was inconclusive. Kava, another herbal product marketed for insomnia, has a significant risk of adverse events, including hepatotoxicity, and is not recommended.

Behavioral and other nonpharmacologic interventions

Behavioral treatment of insomnia is the safest, and perhaps, the most effective therapy for insomnia in older adults. This modality of therapy has also proven effective in patients with comorbid conditions. Effective behavioral interventions for insomnia go beyond sleep hygiene, which is not generally effective when used alone for chronic insomnia. Several randomized trials and systematic reviews provide strong evidence for cognitive behavioral therapy for insomnia (CBT-I). CBT-I usually combines sleep hygiene, stimulus control, sleep restriction, and cognitive therapy, each of which will be discussed in greater detail below. CBT-I has reliably been shown to produce improved sleep efficiency, decreased nighttime wakefulness, and greater satisfaction with sleep. In at least two randomized trials with older adults comparing CBT-I with a prescription sedative-hypnotic agent, participants reported better improvement in sleep and more satisfaction with the CBT-I therapy. One feature that permeates studies comparing CBT-I to pharmacologic therapy is that the improvements with CBT-I are more sustained. One of the major downsides to CBT-I is the limited access to practitioners who are trained in this specific therapy, but efforts are underway in some health care systems to increase access. Also, CBT-I requires significant buy-in from the patient. New delivery models that involve the use of the internet, nonspecialist providers, and telehealth-based CBT-I have yielded promising preliminary results.

Sleep hygiene is education on a list of general guidelines to maintain a healthy sleep-wake routine. When a sleep history is performed behaviors that contribute to disruptive sleep should be ascertained. Examples of these behaviors include sleeping with the television on, or excessive caffeine consumption. The patient should be educated on some of their behaviors that constitute poor sleep hygiene (Table 49-6).

Stimulus-control therapy is designed to break the negative associations patients have with their sleep environment, which have come about from maladaptive behaviors. Patients are instructed to not have any other in-bed activities aside from sleep or sex, and to only go to bed when tired enough to fall asleep. They should unwind prior to going to bed and should not watch an alarm clock. If the patient cannot fall asleep within approximately 20 minutes they should get out of bed and do a relaxing activity and only return to bed when able to fall asleep. For a regular pattern to develop the patient needs to avoid napping and should get out of bed the same time each day. Daytime sleepiness may increase early in this therapy (Table 49-7).

Sleep-restriction therapy was developed from the observation that many patients with insomnia spend a large amount of time in bed unsuccessfully attempting to sleep. This unsuccessful attempt at sleep causes anxiety and frustration which then can lead to the negative associations around the sleep environment. Sleep-restriction therapy is guided by the patient’s sleep diary. The amount of time the patient actually sleeps is calculated and the patient is only allowed to spend that amount of time plus around 15 minutes in bed each night for the following week. Since the amount of sleep will be less this will lead to some sleep deprivation, which then may allow the patient to subsequently fall asleep more quickly and have more consolidated sleep. Once the patient is spending at least 85% of their time asleep the time in bed is increased gradually until the patient is getting adequate sleep (Table 49-8).

Cognitive therapy in CBT-I addresses the maladaptive thoughts or dysfunctional beliefs patients have about their sleep. This is essential to be addressed to ensure adherence to the behavioral aspects of the therapy. Other components of CBT-I may include various relaxation techniques and scheduled worry time. Image rehearsal therapy may be helpful with patients who have trouble sleeping from nightmares related to PTSD.

There are several small studies that have found a beneficial effect of bright light, either from natural sunlight or light boxes, on the sleep of older adults. The reported effects of light on insomnia are more variable than those reported for circadian rhythm disorders. Evening light exposure may be useful for those who go to sleep and wake up too early (ie, advanced sleep phase), while morning exposure may be useful in patients who stay up and sleep in late (ie, delayed sleep phase).

There is less evidence to support other behavioral interventions, but individual patients may find them useful. For example, bathing before sleep can enhance sleep quality in some older individuals, which is likely related to changes in body temperature. A moderate exercise program can improve sleep in healthy sedentary older adults. However, strenuous exercise should not be done close to bedtime because this may interfere with sleep. Some studies have shown a beneficial effect of Tai Chi on symptoms of insomnia.

Sleep in older adults living in nursing homes is well known to be disturbed particularly in patients with neurodegenerative disorders. The prevalence of disturbed sleep in neurodegenerative disorders may be due to damaged brain structures responsible for regulation of sleep and circadian rhythms. Furthermore, a lack of exposure to light and alerting activities during the day can contribute to poor sleep at night. At the extremes, some patients may not spend a full hour fully awake or fully asleep in a 24-hour period. This lack of consolidated sleep prevents the normal sequence through various sleep stages and results in light nonrestorative sleep. In noninstitutionalized patients with significant dementia, progressive disturbances in the sleep-wake cycle often reach a point where nighttime behaviors become a significant stressor for caregivers. In fact, frequent nocturnal wandering and confusion is a leading cause of institutionalization.

The presence of dementia and other neurodegenerative disorders may impact therapy for sleep disorders. Severe dementia may impair the ability of patients to use CPAP for SDB, however studies have shown that CPAP can be successfully used in mild to moderate dementia. Treatment of SDB in these patients can actually improve cognitive function. Furthermore, dementia can impact appropriateness of CBT for insomnia, where patients may not be able to participate. Insomnia behavioral treatment programs that involve the caregiver have been used in patients with dementia. Most pharmacologic therapies for insomnia are relatively contraindicated in those with significant dementia. However, there are behavioral interventions that may be helpful in nursing home patients (Table 49-9). There is mixed evidence for use of melatonin for insomnia or circadian rhythm disturbances in dementia, however bright light may be beneficial. However, the most beneficial timing of bright light therapy in dementia is not clearly known.