Pharmacological and non-pharmacological treatment options for sleep disturbances in Alzheimer’s disease

ABSTRACT

Introduction

Alzheimer’s disease (AD) is one of the most common neurodegenerative disorders among the older population. Sleep disruption and circadian rhythm disorders often develop in AD patients, and many experience sleeping difficulties requiring pharmacological and non-pharmacological interventions.

Areas covered

This review appraised the evidence from clinical studies on various pharmacological and non-pharmacological therapies for sleep disturbances in AD patients and proposed an algorithm to manage sleep disturbances in this population of patients.

Expert opinion

Non-pharmacological interventions are generally preferred as the first-line approach to improve sleep-related symptoms in AD due to their favorable safety profile. However, when non-pharmacological interventions alone are insufficient, a range of pharmacological agents can be considered. Trazodone and melatonin are commonly used as adjunctive therapies, while Z-drugs including zopiclone and zolpidem are specifically employed to treat insomnia in patients with late-onset AD. Furthermore, a newer class of agents known as dual orexin receptor antagonists has emerged and gained approval for improving sleep onset and maintenance in AD patients.

KEYWORDS:

• Alzheimer’s disease
• dementia
• sleep disorders
• tau proteins
• melatonin levels
• cholinesterase inhibitors

1. Introduction

Alzheimer’s disease (AD) is the most common type of dementia among the older population [1] About 44 million people were diagnosed with AD or other related dementia globally, as reported by Alzheimer’s Disease International (ADI). They also projected this figure to reach about 135 million by 2050 [2]. AD is characterized by progressive cognitive decline caused by plaques (diffuse and neuritic) and the accumulation of abnormal proteins in the brain, leading to the formation of neurofibrillary tangles. These proteins include extracellular amyloid beta (Aβ) and intracellular hyperphosphorylated tau (p-tau) [3,4,].

Current pharmacological research in AD focuses largely on the development of disease-modifying drugs that can slow down or reverse progression of the disease. These agents target various aspects, including beta-amyloid production, aggregation, and clearance, as well as tau phosphorylation and assembly. Nevertheless, none of these drugs has demonstrated efficacy in phase III trials to date. As a result, there is currently no available treatment option to cure AD to date, and only symptomatic therapies exist. Table 1 provides information on the commonly used drugs for AD, including the indication, side effects, and mechanism of action [5,6].

Table 1. Drugs used for the treatment of AD.

Drug class	Drug name	Indication	Side effects	MoA
Reversible non-competitive cholinesterase inhibitor	Donepezil Hydrochloride Rivastigmine	Mild to moderate dementia in AD	Gastrointestinal discomfort Aggression Headache Hypertension	By irreversibly inhibiting cholinesterase, they prevent the breakdown of ACh, leading to an increased concentration of ACh in the synaptic cleft. This prolonged presence of ACh enhances cholinergic neurotransmission, resulting in prolonged stimulation of cholinergic receptors, including nicotinic and muscarinic receptors.
Reversible cholinesterase inhibitor	Galantamine	Mild to moderately severe dementia in AD	Arrhythmias Diarrhoea Hypertension	It works by inhibiting the breakdown of ACh, leading to enhanced cholinergic neurotransmission. It also acts as a positive allosteric modulator of nicotinic receptors, which can enhance the sensitivity and response of these receptors to acetylcholine.
N-methyl-D-aspartate (NMDA) receptor antagonists	Memantine	Moderate to severe dementia in AD	Constipation Drowsiness Hypertension	By binding to and blocking NMDA receptors, it helps regulate glutamate neurotransmission, protect against excitotoxicity, and modulate synaptic plasticity.
Monoclonal antibody	Aducanumab	Mild dementia in AD	Headache Nausea Confusion	It binds to aggregated forms of beta-amyloid and helps facilitate their clearance from the brain, potentially slowing down the progression of AD.

Sleep disruption and circadian rhythm disorders often develop in AD patients, and a significant number of these patients (45%) have sleeping difficulties [7]. AD patients often experience decreased nighttime sleep duration, frequent nighttime awakenings, increased wakefulness during the night and early morning, as well as excessive daytime sleepiness[8]. In addition to causing significant distress to individuals with AD and their caregivers, sleep disturbances can also exacerbate other symptoms of AD, such as agitation, aggression, and depression. Sleep disturbances can also negatively affect cognitive function, with studies suggesting that poor sleep quality may accelerate the progression of AD [9]. It is recommended to thoroughly explore and implement nonpharmacological approaches for managing sleep disturbances in AD before considering pharmacotherapy, with only a few exceptions. For instance, in cases of restless legs syndrome (RLS) or periodic limb movement disorder (PLMD), a dopamine agonist may be considered, while melatonin or low-dose clonazepam may be used for rapid eye movement (REM) sleep behavior disorder (RBD) accompanied by potentially harmful behaviors.Pharmacological therapies prescribed for sleep disorders in AD patients, include melatonin, cholinesterase inhibitors, benzodiazepines, N-methyl D-aspartate [NMDA] receptor antagonist, and trazodone. However, they normally lose their effectiveness after weeks of continuous use. It should also be considered that improving sleeping behaviour through pharmacological treatments may lead to the worsening of other AD symptoms through the associated side effects, rendering the clinical management of these patients challenging. These shortcomings have led to the evaluation of non-pharmacological and novel treatments for AD-related sleep disorders, which include light therapy, controlled positive air pressure, acupressure, exercise, physical activity, and aromatherapy [10].

2. Scope of the review

This review aims to identify and appraise evidence from clinical studies investigating various pharmacological and non-pharmacological therapies for sleep disturbances in AD. We conducted a comprehensive serach of electronic databases, including PubMed, Scopus, and COCHRANE central, up until 12^th August 2022. The inclusion criteria involved full-text articles in English language that reported any outcome of pharmacological and non-pharmacological therapies for AD related to the sleep-wake cycle.

Our literature search resulted in 835 articles. Nineteen duplicates were removed, after which 817 articles were screened against the inclusion and exclusion criteria, and 663 were excluded by evaluation of abstracts (see Figure S1). Subsequently, 153 reports were sought for retrieval of full texts. Among these, 42 articles met the criteria for inclusion in this review, including 28 randomized controlled trials,, three open clinical trials, two case-control studies, two case reports, two retrospective studies, two crossover studies, one cross-sectional study, one exploratory study, and one prospective study. Table 2 provides a summary of the data extracted from these 42 studies, including their effects on the sleep-wake cycle in AD patients.

Table 2. Summary of 42 studies with the effect of the intervention on sleep quality and sleep disturbances.

Author	Sample Size	Intervention	Summary of findings
Randomized controlled trial
Chong et al. [11]	n=39	Sham continuous positive airway pressure(CPAP) for 3 weeks followed by therapeutic CPAP (tCPAP) for 3 weeks or tCPAP for 6 weeks	In tCPAP group, Epworth Sleepiness Scale after 3 weeks went from 8.89 at baseline to 6.56 and after 6 weeks further reduced to 5.53. In sham CPAP group followed by 3 weeks of tCPAP, there was no significant difference after 3 weeks of sham CPAP but a significant decrease of Epworth Sleepiness Scale from 7.68 to 6.47 after 3 weeks of tCPAP.
Cooke et al. [12]	n=52	Placebo CPAP for 3 weeks followed by tCPAP for 3 weeks or 3 weeks of tCPAP	tCPAP for 3 weeks led to Change in wakefulness after sleep onset (WASO) scores from 126.0 ± 79.5 to 92.3 ± 44.6 Change in percentage Stage 1 sleep from 23.5 ± 14.5 to 16.3 ± 10.1 Change in percentage Stage 3 sleep from 1.9 ± 3.1 to 4.0 ± 6.2
Nascimento et al. [13]	n=42	Multimodal exercise program: three 1-h sessions every week for 6 months	Improvement of Mini-Sleep Questionnaire (MSQ) score by 18.7%
Figueiro et al. [14]	n=46	Two 4-week periods of active intervention of 350 lx of 6000 K light or control of 100 lx of 2700 K light.	Mean Pittsburgh Sleep Quality Index (PSQI) scores after active intervention went from a baseline of 10.30 ± 0.40 to 6.67 ± 0.48 and from 9.80 ± 0.44 to 8.41 ± 0.47 after control intervention.
Dowling et al. [15]	n=50	Experimental group received 10-week active intervention with 2500 lx light exposure for 1 hour along with 5 mg melatonin (LM group) or placebo (LP group) in the evening while control group received exposure to usual indoor light	Night sleep time in minutes In LP group went from 493 ± 105 to 521 ± 108 In LM group went from 459 ± 109 to 489 ± 105. In LM group, daytime sleep-in minutes went from 315 ± 129 to 249 ± 103 showing significant improvement in daytime somnolence.
Gfigueiro et al. [16]	n=14	Low-level “bluish-white” lighting was administered at daytime for 4 weeks	Sleep efficiency changed from 80 ± 5% at baseline to 84 ± 4% at the end of the intervention. Total sleep time (in minutes) changed from 431 ± 37 at baseline to 460 ± 25 at the end of the intervention.
McCurry et al. [17]	n=132	Subjects assigned to one of the three active treatments (walking, light, combination treatment) or contact control	Changes in total wake time per night (min) after 2 months for control, walking, light, and combination treatment were 7.4, −25.8, −31.6, and 32.4, respectively
Fontana Gasio et al. [18]	n=13	Dawn–dusk simulation (DDS) light therapy for 3 weeks in which signal was curtailed at 0.001 lux and at 70% of maximum 400 lux or placebo dim red light (>5 lux)	Sleep efficiency (%) for DDS was 77.2 ± 4.4 at baseline, 80.4± 2.1 during treatment, and 76.5 ± 1.0 during follow-up
Burns et al. [19]	n=48	Exposure to bright light therapy (BLT) of 10,000 lx (full spectrum) or standard fluorescent tube light at 100 lx for two hours	Nocturnal sleep (mean duration) went from 8.3 hours at baseline to 8.6 hours at the end of the intervention with no significant difference
Wade et al. [20]	n=80	2 mg of prolonged-release melatonin (PRM) or placebo for 24 weeks	Significant decrease of −5.5 in the PSQI global score in the PRM group but not in the placebo group.
Singer et al. [21]	n=157	Treatment with placebo, 2.5-mg slow-release (SR) melatonin (ML), or 10-mg melatonin.	Nocturnal total sleep time (NTST) increased by 16 ± 54 minutes in the ML 2.5SR group and 13 ± 44 minutes in the ML 10 group compared to 3 ± 39 minutes for placebo, indicating a weak trend for more nighttime sleep in the melatonin groups.
Serfaty et al. [22]	n=44	6 mg PRM for 2 weeks followed by a week of the washout phase Participants in group 1 were administered melatonin in the first treatment period and received a placebo in the second treatment period and vice versa for group 2	Melatonin had no effect on median total time asleep, number of awakenings, and sleep efficiency.
Fisman et al. [23]	n=27	100 mg t.i.d. of 2-dimethylaminoethanol for 2 days which was increased every 2 days till a maximum 1800 mg/day or placebo for 2 months	No significant improvement was observed
Moraes et al. [24]	n=23	5 mg of donepezil or placebo	Percentage REM sleep changed significantly from 7.2 ± 4.3 at baseline to 15.9 ± 11.1 at the end of treatment (p < 0.05)
Moline et al. [25]	n=62	Lemborexant (LEM 2.5 mg, 5 mg, 10 mg or 15 mg) or placebo for 4 weeks	Mean number of sleep bouts during the daytime In placebo group, went from baseline of 4.76 ± 3,12 to 4.44 ± 3.10 at Week 4 In LEM2.5 group, went from baseline of 5.90 ± 3.05 to 6.07 ± 3.60 at Week 4 In LEM5 group, went from baseline of 4.97 ± 2.97 to 4.21 ± 2.58 at Week 4 In LEM10 group went from baseline of 6.24 ± 3.23 to 7.08 ± 4.21 at Week 4 In LEM15 group, went from baseline baseline of 4.75 ± 2.55 to 4.93 ± 2.22 at Week 4
Stahl et al. [26]	n= 1698	Galantamine: 8 mg BID galantamine (GAL16) or 12 mg BID galantamine(GAL24) or placebo	The rates of sleep problems for GAL 24, GAL 16, and placebo arms were 2.6%, 1.1%, and 2.2%, respectively, and for nightmares were 1.1%, 0.4%, and 0.1%, respectively.
Grippe et al. [27]	n=47	Trazodone 50 mg or placebo for 14 days	Relative rhythm amplitude for the trazodone arm went from 0.68 at baseline to 0.75 post-treatment Mean daily activity levels for the trazodone arm at baseline and post-treatment were 171.481 and 211.341, respectively.
Camargos et al. [28]	n=30	50 mg of trazodone or placebo for 2 weeks	Subjects in the trazodone treatment group had 42.5 minutes more sleep at night compared to the placebo group and had their nighttime percent sleep increased by 8.5% points with respect to baseline.
Kouzuki et al. [29]	n=35	Aroma bath salt with concentrations 0.1%, 0.5% or 1% for a duration of ≥10 min for 24 weeks was preceded and succeeded by an observation period for 4 weeks each.	Changes in the PSQI-J scores from pre-intervention at week 4 to post-intervention at week 28: Ffor 0.1% group = −0.2 ±0.5 Ffor 0.5% group = −0.4 ±0.2 Ffor 1% group = 0.1 ±0
Rose et al. [30]	n=38	CES device provided electrical stimulation (CES) at an intensity of 100 μA for 60 minutes/day for 4 weeks or sham CES	A trend toward statistical significance (p =.09) in PSQI scores for daily disturbances was found between the intervention group (2.34) and sham control group (2.65).
Markowitz et al. [31]	n=136	Galantamine 24 mg or placebo	Mean changes in PSQI score for galantamine and placebo were 0.01 and −0.17, respectively.
Scoralick et al. [32]	n=24	15-mg mirtazapine or placebo for 14 days	Mean NTST (min) for the mirtazapine group went from 383.4 at baseline to 414.0 post-treatment.
Meguro et al. [33]	n=34	1 mg of risperidone or placebo for 1 month	Patients treated with risperidone had a decrease of 1.2 hours in the mean daytime sleep while an increase of 3.8 hours in mean nighttime sleep.
Herring et al. [34]	n= 285	Suvorexant 10 mg or placebo for 4 weeks	Total sleep time (mins) in the suvorexant group went from 277.7 ± 76.9 at baseline to 349.4 ± 71.9 at the end of treatment. The model-based least squares mean improvement-from-baseline in TST was 73 minutes for suvorexant group.
Savaskan et al. [35]	n=30	Quetiapine (25–200 mg) or haloperidol (0.5–4 mg)	Subjects administered with quetiapine had shorter wake bouts at the end of intervention as compared to baseline Subjects administered with haloperidol had fewer immobile phases but with increased duration compared with baseline
Ito et al. [36]	n=28	BLT (3000 lx, 0.900–11.00 h) for 8 weeks and for the latter 4 weeks, half were treated with BLT + Vitamin B12 (1.5 mg/day for the fifth-sixth week, 3.0 mg for the seventh-eighth week)	Percentage sleep during the daytime after BLT for 4 weeks was 27.9 ± 13.2, and after BLT for 8 weeks + Vitamin B12 treatment was 25.5 ± 12.5.
Cooke et al [37].	n=76	Donepezil galantamine, rivastigmine, or no AChEI	Significantly lower percentage of stage 1 sleep in the donepezil group than patients in the galantamine group, but there was no significant difference between the donepezil group and the rivastigmine or no AChEI groups Significantly higher percentage of stage 2 sleep in the donepezil group than patients in the no AChEI group, but there was no significant difference between the donepezil group and either the galantamine group or the rivastigmine group
Exploratory study
Cooke et al. [38]	n=10	6-week of CPAP, after which follow-up of 5 patients who continued CPAP (CPAP+), and 5 patients who discontinued CPAP (CPAP-)	CPAP+ group showed a significant clinical and statistical improvement in PSQI
Open clinical trial
Skjerve et al. [39]	n=11	BLT (5000–8000 lx at a distance of 0.3–0.5 m) 45 mins daily for 4 weeks daily	Sleep–wake disturbances in dementia scores before treatment, during treatment, and after treatment were 13.10 ± 2.08, 13.20 ± 2.67, 12.10 ± 3.25, respectively, with no significant difference.
Yamadera et al. [40]	n=27	BLT (3000 lx) 2 h daily for 4 weeks.	Percentage of sleep time at night went from 49.8 ± 15.8 to 55.0 ± 17.9 before and after therapy, respectively. Percentage of nap time during the day went from 30.6 ± 15.7 to 23.8 ± 13.2 before and after therapy, respectively.
Mizuno et al. [41]	n=12	Donepezil 3 mg daily for 7 days, followed by 5 mg daily for 6 weeks.	Total sleep time in minutes changed from 332.0 ± 118.2 at baseline to 398.6 ± 98.5 after 6 weeks.
Case control study
Mishima et al. [42]	n=14	BLT (3000–5000 lx) 2 h daily for 4 weeks.	Total sleep time and nocturnal sleep time before the light therapy were 7.1 ± 1.5 h and 5.7 ± 1.3 h, respectively, and during the light therapy were 7.6 ± 1.1 h and 6.7 ± 1.2 h, respectively.
Sekiguchi et al. [43]	n=17	BLT (5000 lx) 1 h daily for 2 weeks.	Improvement in sleep quality in four out of seventeen AD patients.
Retrospective study
La et al.[44]	n=25	Trazodone with a median dosage of 50 mg per day (25th percentile 50 mg; 75th percentile 100–125 mg)	Participants treated with trazodone with concomitant baseline sleep complaints, especially those who reported improvement in sleep quality over time, had delayed cognitive decline, compared to patients with sleep disruption that were not on trazodone.
Brusco et al. [45]	n=14	Melatonin 9 mg at bedtime daily for variable periods, ranging from 22 to 35 months	Coefficient of variation of bed time values (indirect estimation of agitated behavior at the beginning of the night) for days 0–2 and days 19–21 for patients with sleep disturbances alone were 32.6 ± 14.9 and 31.5 ± 10.8, respectively, while for patients with sleep disturbances and depression were 27.1 ± 14.1 and 24.5 ± 13.5, respectively, and for patients with sleep disturbances and dementia were 58.0 ± 24.7 and 41.5 ± 20.9, respectively.
Bromundt et al. [46]	n=20	Individually timed DDS over the subjects’ beds for 7–8 weeks.	The sleep efficiency [%] for the participants undergoing DDS went from 79.23 ± 11.21 to 82.64 ± 10.84.
Katsunuma et al. [47]	n=9	Zopiclone 7.5 mg/day for 20 days with concomitant administration of aniracetam	In 7/9 cases (78%), there was a 50% prolongation of sleeping time, while in the remaining two cases (22%), there was no significant improvement.
Cross-sectional study
Shih et al. [48]	n=184	Observation of patients who were able to walk with or without a walking aid (e.g. walker or cane) as well as participation in a regular program of physical activity	Significant improvement in sleep quality by having an accompanying walker Improvement in sleep was also seen in those doing weekly walking
Case report
Niitsu et al. [49]	-	Donepezil 5 mg/day with concomitant tiapride (50 mg daily) and yokukansan (7.5 g per day) which was later switched to keishi-ka-ryukotsu-borei-to (KRBT) at 7.5 g per day	Improvement in Neuropsychiatric Inventory (NPI) total score from 76 to 35.
Anderson et al. [50]	-	Melatonin 5 mg nocte followed by 10 mg nocte after 1 month.	At 5 mg dose, there was partial mitigation of the symptoms of REM sleep behavior disorder, while 10 mg dose led to a complete resolution of these symptoms. Every 2–3 days, nocturnal jerks were observed initially, but after melatonin treatment, there were no such episodes observed.
Prospective study
Simoncini et al. [51]	n= 129	Device H7 Insomnia Control was used to apply manual acupressure locally all night (from 7 pm to 7 am) for 2 months.	Change in PSQI scores from 12.59 ± 8.56 at baseline to 8.56 ± 3.27 after 2 months and to 9.2 ± 3.6 after 4 months.

3. Sleep dysfunction in AD

Circadian rhythm disruption is a key factor that contributes to sleep disturbances in AD. The circadian rhythm is regulated by the brain’s suprachiasmatic nucleus (SCN), which responds to light and dark signals to coordinate the sleep-wake cycle. In AD, changes in the functioning of the SCN and other brain regions involved in sleep regulation can lead to disruption of the sleep-wake cycle [52]. This disruption can manifest in several ways, including fragmented sleep, daytime sleepiness, and nocturnal awakenings. Additionally, individuals with AD often experience a reversal of their sleep-wake cycle, with increased daytime sleepiness and nighttime agitation.

Neurotransmitter imbalances are another factor that can contribute to sleep disturbances in AD. Several neurotransmitters are involved in regulating sleep, including acetylcholine and serotonin. In AD, these neurotransmitters are affected by changes in the brain, and imbalances in these systems can contribute to sleep disturbances [53]. For instance, reductions in the acetylcholine levels in AD patients due to degeneration of basal forebrain cholinergic nuclei can contribute to the development of insomnia [54]. In addition, a large body of literature also describes serotonergic deficits in AD, manifesting as degeneration of neurons in the raphe nucleus and locus coeruleus and reduced cortical levels of serotonin [55]. Since serotonin is involved in regulating the sleep-wake cycle, serotonin imbalance might assist in developing sleep disturbance in AD patients.

As discussed beforehand, the accumulation of Aβ plaques in the brain is a hallmark of AD, and these plaques are also thought to play a role in sleep disturbances. Beta-amyloid plaques can disrupt the normal functioning of brain regions involved in sleep regulation, such as the thalamus and hippocampus [56]. This disruption can lead to altered sleep patterns and sleep disturbances, including insomnia and daytime sleepiness. Yet, medications used to treat AD and its associated symptoms can also contribute to sleep dysfunction. For example, cholinesterase inhibitors can cause vivid dreams and nightmares, while antipsychotic medications, used to treat agitation and psychosis, can cause sedation and impaired motor coordination.

Besides, sleep disturbances can also be associated with various medical comorbidities in AD patients. Conditions such as sleep apnea and restless leg syndrome, common in this patient population, can lead to sleep quality and quantity disruptions. These conditions can also exacerbate other sleep-related symptoms in individuals with AD. Moreover, the neuropsychiatric comorbidities frequently implicated in AD, including depression, anxiety, and agitation, can also lead to sleep dysfunction. To illustrate, depression is associated with reduced sleep quality and quantity, while anxiety and agitation can interfere with sleep onset and maintenance.

Sleep disturbances can negatively influence AD progression. Figure 1 demonstrates how sleep dysfunction is linked with the accumulation of toxic metabolites in AD patients. Small metabolites and neurotoxic peptides like tau and beta-amyloid peptide within the CSF are transported to interstitial space. This transportation is important for eliminating toxic waste products of cellular activity and neural metabolism [57]. The pineal gland secretes melatonin in the human body, where its main function is to regulate circadian rhythms and prevent the accumulation of Aβ. In AD patients, it has been found that melatonin level in CSF is decreased even in the preclinical stages [58].

Figure 1. Link between aggregation of toxic metabolites of AD with sleep dysfunction.

Another essential feature of the brain is the glymphatic system responsible for eliminating Aβ. This system occurs more proficiently during anesthesia and physiological sleep than during wakefulness [59]. The lymphatic system in the central nervous system consists of vessels accompanied by astrocyte podocytes that accommodate aquaporin-4. These are water channels, and it has been found that mice lacking these aquaporin channels have decreased CSF influx through the para-vascular system and diminished interstitial Aβ clearance [60].

4. Pharmacological options

Numerous clinical trials have been published to investigate the efficacy and safety of drugs used for the management of sleep disturbances in AD. For example, Grippe et al. [27] conducted a trial of two phases involving 30 patients. The initial phase, 7–9 days long, involved setting baseline parameters. In the final phase, patients were randomized and equally distributed for treatment with 50 mg trazodone or placebo. Actigraphy was utilized to assess and monitor the sleeping pattern of the subjects. The study showed that patients in the active treatment arm had an increase in mean daily activity (MDA), mean activity in the most active ten h (M10), and a decrease in inter-daily stability mean (ISM), and mean activity in the least active five consecutive hours (L5). However, it was not statistically different compared to the patients in the placebo arm. Nevertheless, compared to the placebo arm, a statistically significant increase was found in relative rhythm amplitude (RRA) in patients treated with trazodone. The findings suggest that trazodone not only improves sleep parameters but leads to an overall improvement in circadian rhythms [27]. A similar double-blind, randomized controlled trial was conducted by Camargos et al. [28]. It consisted of an initial phase of 7–9 days to set baseline parameters and a final phase of 2 weeks long in which subjects were distributed in 1:1 ratio for treatment with 50 mg of trazodone once daily and placebo. Actigraphy was used to monitor the effects of the intervention on sleep patterns. Subjects in the active treatment arm post-treatment showed an increase in nighttime percent sleep by 8.5% points and an increase in total sleep duration during the nocturnal period (NTST) by 10.8% points. There was a decrease in awakenings by 18.9% points, a decrease in nighttime waking after sleep onset by 18.3% points, and a decrease in naps by 3.5% points. There was no induction of significant daytime sleepiness or naps in both active treatment or placebo arms [28].

Singer et al. [21] conducted a randomized, placebo-controlled clinical trial involving 157 subjects who were randomly allocated to 1 of 3 treatment arms: placebo, 2.5-mg slow-release melatonin (ML 2.5SR), or 10-mg melatonin (ML 10). It was found that the melatonin treatment arms showed a slight trend for increased nighttime sleep as there was an increase in NTST by 16 ± 54 minutes in the ML 2.5SR group and 13 ± 44 minutes in the ML 10 group compared to 3 ± 39 minutes for placebo. The proportion of subjects gaining at least 30 minutes of NTST was highest in ML 10 group [21]. In addition, Wade et al. [20] conducted a randomized, double-blind trial involving 80 outpatients diagnosed with mild to moderate AD between the ages of 50 and 85. The initial phase consisted of treatment with a placebo for two weeks. Subjects were then distributed in a 1:1 ratio for treatment with 2 mg of prolonged-release melatonin (PRM) or placebo nightly for 24 weeks, followed by a placebo for two weeks. Sleep efficiency was assessed through the Pittsburgh Sleep Quality Index (PSQI), while cognitive performance was measured through Instrumental Activities of Daily Living (IADL) and Mini-Mental State Examination (MMSE). Not only was the PSQI score significantly better with PRM (p-value: 0.017, PSQI = 4), but also subjects in the PRM treatment arm had a considerably better cognitive performance compared to those in the placebo treatment arm assessed through IADL and MMSE (P-values: of 0.004 and 0.044, respectively; IADL = 4.06 ± 2.34, MMSE = 21.9 ± 3.8) [20]. It was found through a randomized controlled trial by Serfaty et al. [22] that there was no overall clinical benefit or adverse outcomes from exogenous melatonin (regarding the number of awakenings, the efficiency of sleep, median total time asleep, or sleep efficiency). There was the absence of any carry-over effects after the administration of melatonin [22]. On the other hand, a retrospective study conducted by Brusco et al. [45] found that melatonin administration improved sleep quality in patients with or without depression symptoms. It also showed that daily alertness and morning freshness were enhanced only in patients with sleep disturbances without other major symptoms. Patients with sleep disturbances administered with exogenous melatonin had their bedtime changed from 32.6 ± 14.9 to 31.5 ± 10.8 days [45].

Cooke et al. conducted a study involving 76 mild to moderate AD patients. The percentage of each sleep stage was determined for the subjects through thorough medication history screening and polysomnography. Based on the type of AChEI used, participants were distributed into various groups and administered donepezil, galantamine, rivastigmine, or no AChEI. The percentage of stage 1 and stage 2 sleep was significantly affected through AChEI therapy with a p-value of 0.01 and 0.03, respectively. The percentage of stage 1 sleep was significantly reduced in patients administered with donepezil compared to those in the galantamine group (mean = 17.3%, SD = 11.7 vs 29.2%, SD = 15.0, respectively). At the same time, it did not significantly differ between the rivastigmine (mean = 25.0%, SD = 12.3) or no AChEI groups (mean = 27.6%, SD = 17.7). The percentage of stage 2 sleep was significantly higher in the donepezil group compared to the no AChEI group (mean = 63.6%, SD = 14.4 vs. 51.4%, SD = 16.9, respectively; p = 0.04). No substantial and major differences were observed regarding the percentage of REM sleep or other sleep parameters among different groups [37]. It can be concluded that donepezil, galantamine, and rivastigmine can treat sleep disturbances in patients with mild to moderate AD.

Zopiclone and Zolpidem are Z-drugs frequently prescribed for insomnia in patients with late-onset Alzheimer’s disease. Louzada et al. [61], in a randomized triple-blind placebo-controlled trial, showed that zopiclone (81-minute decrease in main nocturnal sleep duration (MNSD); 26-minute drop in awake time after sleep onset (WASO)) and zolpidem (no significant difference in MNSD; 22-minute drop in WASO) had clinical benefit when used short-term even though there was a need for further evaluation of safety. In a prospective study by Huo et al. [62], eszopiclone improved sleep quality and cognition in elderly patients with dementia and insomnia. A usual daily oral dose of Zolpidem is 5–10 mg, and the usual daily maximum dose is 10 mg. The usual dosage of Zopiclone is 2–3 mg, with a usual daily maximum of 3 mg [5].

Patients with Alzheimer’s disease may find relief from their psychotic symptoms by using antipsychotic drugs like risperidone and quetiapine. However, because of the possibility of major adverse effects such as extrapyramidal symptoms, sedation, and an increased risk of death, especially in older patients with dementia, its usage should be restricted. Antipsychotics should only be used when other treatments have failed, and behavioral problems are severe. Alzheimer’s patients who experience anxious and depressive symptoms might benefit from antidepressant drugs such as selective serotonin reuptake inhibitors (SSRIs). However, their usage should also be carefully examined because of possible negative effects and medication interactions. For example, SSRIs may interact with other drugs, such as monoamine oxidase inhibitors (MAOIs), and raise the risk of falls in older individuals.

5. Non-pharmacological options

Non-pharmacological interventions are investigated in clinical trials used for the management of sleep disturbances in AD. For example, Cooke et al. conducted a randomized placebo-controlled trial that involved fifty-two participants (mean age = 77.8 years, SD = 7.3) suffering from AD and obstructive sleep apnea (OSA). It consisted of 3 weeks of therapeutic Continuous Positive Airway Pressure (tCPAP) vs. three weeks of placebo CPAP (pCPAP), followed by three weeks of tCPAP. It was found that after one treatment night, compared to the patients in pCPAP group, the patients in the tCPAP group had considerably less % Stage 1 (p = 0.04) and more % Stage 2 sleep (p = 0.02). The paired analysis showed that 3 weeks of tCPAP significantly decreased the % Stage 1 (p = 0.001), WASO (p = 0.005), arousals (p = 0.005), and also led to an increase in % Stage 3 (p = 0.006). The study concluded that in patients with OSA with mild to moderate AD, administration with tCPAP led to a deeper sleep after just one night, and the improvements in sleep could be sustained for three weeks [12]. Cooke et al. conducted another study involving follow-up of patients with mild to moderate AD who were a subset of participants from a 6-week randomized clinical trial (RCT) of CPAP, which consisted of sustained use of CPAP for a mean duration of 13.3 months with SD of 5.2 months. The study consisted of 5 patients who continued CPAP (CPAP+) and five patients who discontinued CPAP (CPAP-) after completion of the RCT. In the follow-up of these patients, it was observed that even with a small sample size, patients in CPAP+ showed moderate-to-large effect sizes. In addition, it was observed that patients in the CPAP+ group had decreased cognitive decline, stabilized depressive symptoms, and diminished daytime somnolence compared to the CPAP- group. Sustained CPAP uses also significantly improved subjective sleep quality and psychopathological behavior. It was also reported that the caregivers of the patients in the CPAP+ group also reported that their sleep was enhanced compared with the final RCT visit of the initial study and their patients [38].

Nascimento et al. conducted a 6-month-long study involving 42 patients with Parkinson’s disease and 35 demented patients with Alzheimer’s disease to examine the role of physical activity on sleep disturbances [13]. Participants were divided into four groups based on age, sex, disease duration, severity, and medication dosage. Participants were asked to carry out three 1-h sessions per week of a multimodal exercise program of moderate intensity spread over a long duration designed to stimulate aerobic metabolism. The effect of physical activity on sleep disturbances (SD) was assessed using the Mini-Sleep Questionnaire (MSQ), and the performance of instrumental activities daily living (IADL) was assessed through The Pfeffer Questionnaire for Instrumental Activities (PIAQ). The results showed that AD patients had an improvement of 13.1% in PIAQ scores and 18.7% in MSQ scores. The study concluded that when a multimodal physical exercise program was carried out for 6 months, it significantly improved IADL performance and decreased SD in elderly patients suffering from PD or AD.

A study by Skjerve et al. demonstrated that bright light of around 5000–8000 lux given for 45 minutes in the morning for four weeks improved behavioral symptoms. However, no deviations were found in the sleep-wake cycle [39]. Similarly study done by Yamadera et al. showed light therapy had led to an improvement in circadian rhythm disturbances. However, no significant effects were seen in the severity of the disease process of dementia [40]. Light therapy administered daily in the morning for 28 days was one way to re-synchronize and adapt the disrupted cycle in demented people with sleep disturbances [42]. Figueiro et al. had a clinical trial with active light intervention and a controlled intervention of light to provide high and low circadian stimulus, respectively, which can improve depressive symptoms, quality of sleep, and agitated nature of patients with Alzheimer’s disease [14]. In a randomized controlled trial by McCurry et al., participants were intervened with a walking program, light exposure, or combination and demonstrated that it could improve the sleep cycle in patients with the disease with good adherence to treatment [17]. Another trial by Burns et al. demonstrated that bright light therapy could improve sleep disturbance and agitation and reduce the requirements of medicines [19]. However, a study by Dowling et al. outlined that morning light therapy for 1 hour may not be sufficient to improve the sleep-wake cycle. Still, if combined with melatonin, then daytime wakefulness can be improved. Not all patients treated with bright light therapy demonstrated overall improvement, but only those whose rest-activity has been disturbed showed a positive response to light intervention [63].

Although recent evidence has been mixed, alternative therapies, acupuncture, and acupressure have shown benefits in managing insomnia. In a systematic review from Sarris & Byrne [64], the Pittsburgh sleep quality index (PSQI) score showed that the positive effects of acupuncture were seen as compared to sham acupuncture or with no treatment. Subsequently, no serious side effects were observed [65,66]. A study was conducted by Chen et al. [66] to demonstrate that HT7 point (Shenmen) acupressure administered to elderly institutionalized patients improved sleep quality. Sleep quality was assessed through PSQI, which was shown to improve in those administered real acupressure. Subjects in the two control groups who were either treated with false acupressure or those who weren’t treated showed no improvement in sleep quality. It was also observed that the time of permanence in bed was increased, and there was a decrease in frequency and number of nocturnal awakenings in the subjects treated with real HT7 point (Shenmen) acupressure. Specifically in patients with Alzheimer’s, Simoncini et al. [51] conducted a study involving 129 patients with insomnia in which acupressure was administered daily for eight weeks. The pressure was administered using manual HT7 point from 7 pm to 7 am all night long. After two months, sleep quality was assessed using PSQI, and it was observed that sleep disturbances were reduced, and effective slept hours were substantially increased. It was also observed that the administration of acupressure resulted in much faster sleep induction and better global sleep quality [65,66].

Transcranial magnetic stimulation (TENS) is an alternative noninvasive treatment modality for patients with neuropsychiatric ailments, including Alzheimer’s disease. For primary insomnia associated with dementia, repetitive TENS (rTENS) was associated with improvement in symptoms following a four-week intervention (PQSI = 9.4 ± 0.97, p = 0.001). There was also an improvement in cognitive performance, a well-known benefit of TENS. The result was based on a double-blind, randomized, and sham-controlled pilot study conducted by Zhou et al. [67].

6. Expert opinion

As the advancement of sleep disturbances have found to cause a more rapid progression of illness and neurodegenerative symptoms in patients with AD; hence the management of sleep disturbances in AD patients is deplorable.

Through the various trials conducted on trazodone, it has been shown that it can increase the nocturnal level of melatonin without affecting its daytime level. All three randomized controlled trials of trazodone included in the review showed improved sleep quality after its administration. Trazodone is a relatively safe and convenient option for treating disturbed circadian rhythm and sleep disruption in AD patients. It is considered a good alternative as it can cause a significant improvement in the fragmentation and duration of sleep while simultaneously addressing the relative amplitude of activities day- and night-times [27,44]. Trazadone attains its efficacy by causing a reduction in the aggregation of beta-amyloid plaques and decreasing the pathologic tau burden, which leads to an improvement in memory consolidation mediated by undisrupted sleep.

Out of the three included randomized controlled trials for melatonin, two showed improved sleep quality and efficiency. The main parameter of sleep affected in patients with AD is sleep efficiency, and PRM mediates its therapeutic effect on sleep by improving sleep efficiency. Besides sleep, melatonin significantly enhances cognition by affecting hippocampal networks through its high levels at night, making the hippocampus more sensitive to the hormone [20].

For treating symptoms in patients with mild to moderate AD, acetylcholinesterase inhibitors (AChEIs) such as donepezil, galantamine, and rivastigmine are considered the first line of therapy [68]. The randomized controlled trials included in this review about administering AChEIs have shown that AChEIs have greater efficacy than placebo. However, although the efficacy of AChEIs on patients with AD included in randomized controlled trials has been widely accepted, their effectiveness in the real world is controversial [68]. Although the different AChEIs belong to the same class, they have a differential effect on sleep due to the variations in their mechanism of action. The varied effect can also be due to their diverse selectivity for different cholinesterase as well as the different schedules for the administration of the medications.

CPAP is usually administered to patients suffering from obstructive sleep apnea [69], but the results of studies included in the review have shown that CPAP treatment for a long duration in AD patients can lead to significant improvement in sleep and mood as well as a decrease in the rate of cognitive deterioration.

The studies included in this review have shown that light therapy in AD patients with sleep disturbances significantly improved sleep-wake and the rest-activity cycle rhythm. It has been shown that light directly impacts the circadian rhythm and hence can be considered an effective therapy for the treatment of sleep disruptions in AD patients [42,43]. A randomized trial consisting of residents from a nursing home for AD patients involving light therapy administration through exposure with BLT((≥ 2500 lx) for one hour daily for ten weeks led to improved appetite and depression as agitation [63]. BLT treatment considerably affects total sleep time and time required for sleep onset. It also improved the quality and efficiency but did not significantly affect the number of early morning awakenings and wake afterward sleep onset [70].

The current evidence based on the studies involving DORAs has propagated optimism, but further prospective studies must be conducted to explore its use and determine and monitor the risk-benefit ratio [71].

The studies in this review involve patients largely suffering from moderate to severe dementia with a diverse range of sleep problems primarily encountered in clinical practice. Therefore, the results from these studies are widely applicable but cannot be applied to patients with sleep problems in the early stage of AD as pathology associated with circadian rhythm is less severe and hence more sensitive and produces a better response to drug intervention. However, some of the randomized controlled trials included are limited in the capacity to explore the adverse effects of therapy. Untreated sleep disturbances lead to an accelerated rate of amyloid plaque depositions; hence, clinical intervention can improve a patient’s quality of life. There is a need for conducting trials on commonly used drugs for their efficacy and safety with their adverse effects on sleep quality and efficiency. Patients who made no response to non-pharmacological therapy can be included in pragmatic trials based on clinical settings. Utilizing sequential and well-structured strategies following clinical guidelines can produce valuable results. Patient and caregiver-specific outcomes will greatly benefit sleep research in dementia. Focusing more on the consistency and structure of trial design and actigraphy techniques can further aid the synthesis and interpretation of research in this area.

Overall, sleep disruption in patients with AD can have a significant role in advancing AD symptoms. Therefore, patients suffering from AD not only have a deteriorated life quality but also suffer from potential health consequences. It has been projected that the prevalence of AD is anticipated to increase; hence the collection of evidence regarding the treatment modalities is of paramount importance. Effective management of sleep disturbances in Alzheimer’s typically involves a multimodal approach that addresses the underlying causes of sleep disturbances and the associated symptoms. Nevertheless, managing sleep disturbances in individuals with AD can be challenging, as many of the medications commonly used to treat sleep disorders are not recommended in this population of patients due to their potential side effects. Therefore, non-pharmacological interventions which have shown some promise in managing sleep disturbances in individuals with AD, as discussed beforehand, should be considered. Additionally, careful monitoring and management of comorbid medical conditions that may contribute to sleep disturbances, such as sleep apnea and restless leg syndrome, are important for optimal management of sleep disturbances in individuals with AD. A stepwise algorithm for the management of sleep disturbances in AD is proposed (Figure 2), which complements the above discussion.

Figure 2. A proposed stepwise algorithm for the management of sleep disturbances in AD.

Article highlights

• Alzheimer’s disease (AD) is the most common type of dementia among the older population.

• Sleep disruption in AD patients can have a significant role in the progression of AD symptoms.

• Melatonin was the most widely used treatment of choice, despite its limited efficacy.

• Non-pharmacological interventions such as CPAP treatment in AD patients can lead to significant improvement in sleeping disturbances.

Abbreviations

CPAP	=	Continuous positive airway pressure
ESS	=	Epworth Sleepiness Scale
AD	=	Alzheimer’s disease
OSA	=	Obstructive sleep apnea
tCPAP	=	therapeutic CPAP
pCPAP	=	placebo CPAP
PD	=	Parkinson’s disease patients
MSQ	=	Mini-Sleep Questionnaire
PIAQ	=	Pfeffer Instrumental Activity Questionnaire
BLT	=	Bright light treatment
PSQI	=	Pittsburgh Sleep Quality Index
LP	=	Light and Placebo
LM	=	Light and Melatonin
SEM	=	standard error of the mean
DDS	=	Dawn – dusk simulation
NITE-AD	=	Nighttime Insomnia Treatment and Education in Alzheimer’s Disease;
TST	=	Total sleep time
NST	=	nocturnal sleep time
MDA	=	mean daily activity
LPRS	=	London Psychogeriatric Rating Scale
CES	=	Cranial electrical stimulation
NPI	=	Neuropsychiatric Inventory
PRM	=	Prolonged release melatonin
SWD	=	Short scale for sleep – wake disturbances in dementia

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Acknowledgments

The authors would like to thank Ezekwesiri Nwanosike for his support in completing this review.

Additional information

Funding

This research was not funded.