Introduction
Alzheimer’s disease (AD) is a neurodegenerative disorder exhibiting gradual disease progression [1]. Cognitive decline represents the most prominent clinical manifestation of AD, it is often accompanied by sleep disturbances and varied psychobehavioral symptoms, including apathetic and depressive behavior as well as agitated behavior characterized by exaggerated motor activity and verbal and/or physical aggression [2,3]. Approximately 90% of patients develop one or more psychobehavioral symptoms during the AD course [4,5]. Moreover, patients in advanced AD stages may exhibit severe cognitive impairment, behavioral disturbances, and high caregiver dependence, possibly reducing the quality of life of these patients, worsening the difficulty of caregiving, and increasing the healthcare burden [6].
Sleep disorders are common in AD, with 70% of patients experiencing sleep disruption in the early stages [7]. The suprachiasmatic nucleus (SCN), a sleep regulator in the brain’s hypothalamus [8], synchronizes the circadian rhythm by regulating the light-dark and sleep-wake cycles [9]. In patients with AD, reduced light input due to reduced outdoor exposure and poor sensitivity to light stimuli owing to neuropathic damage can lead to decreased circadian rhythm stability [10]. Additionally, reduced social interaction and age-related eye defects (e.g., macular degeneration and cataracts) may also affect light transmission in the eyes, leading to circadian rhythm and sleep disorders [11].
The primary treatment for AD involves pharmacological therapy, including cholinesterase inhibitors, NMDA receptor antagonists, Aβ inhibitors, and intestinal flora modulators [12]. Although medication may alleviate AD-induced cognitive and memory impairments to some extent, it cannot stop the progression or cure this neurodegenerative disease. Moreover, some patients with AD may develop medication-related side effects, such as poor appetite, diarrhea, and hallucinations [13–15]. Therefore, a highly efficacious treatment with no side effects is currently a hot spot and challenging research area.
Photobiomodulation (PBM) is a non-pharmacological therapy employing light energy to modulate biological function and promote therapeutic effects. PBM stimulates the SCN with varying light intensities and durations to regulate melatonin secretion and facilitate the communication between the hypothalamus and cortex [16], ultimately modulating circadian rhythms. Light therapy can also shorten sleep latency, reduce nocturnal insomnia, increase total sleep time, and improve sleep quality in patients with AD [17,18]. Recently, research on applying PBM in neurology has risen rapidly, particularly concerning the treatment and regulation of neurodegenerative diseases [19]. Studies have shown that PBM can improve cognitive function, enhance the quality of life, and reduce caregiver burden in patients with AD by reducing neuronal damage and inflammatory response [20].
Despite light therapy receiving increasing attention as a potential non-pharmacological intervention for AD, a systematic evaluation of the efficacy and safety of this treatment for patients with AD is still unavailable. To fill this knowledge gap, a comprehensive systematic review and meta-analysis is required to more rigorously evaluate the effectiveness of light therapy in ameliorating sleep disorders and psychobehavioral symptoms among patients with AD.
A meta-analysis by Van Maanen et al. [21] used broad study inclusion criteria and included those involving patients with reported or diagnosed sleep disorders, showing that light therapy was effective in treating sleep problems. Similarly, Roccaro et al. [22] conducted a systematic review of research examining the effect of light therapy in alleviating sleep and rhythm disorders in patients with AD, concluding that light therapy ameliorates circadian rhythm disturbances and sleep efficiency (SE) in patients with AD. Furthermore, previous studies have indicated that light therapy can reduce cognitive decompensation and depression behavior in patients with AD [23,24], however, further research is warranted to add to the scarce evidence on its usefulness. Compared with prior meta-analyses, our meta-analysis had a relatively larger number of included articles and outcome metrics, expanded the sample size, and only included randomized controlled trials (RCTs) to reduce sampling error. Therefore, our meta-analysis may strengthen the evidence on the effectiveness of light therapy in treating sleep disorders and psychobehavioral symptoms in patients with AD.
Materials and methods
The systematic review protocol used in this research was according to the Preferred Reporting for Items for Systematic Reviews and Meta-Analyses 2020 (PRISMA) criteria to guide the reporting of results [25]. The PRISMA checklist is shown in S1 Table. The protocol for this systematic review was registered in the PROSPERO database (registration number: CRD42023406390).
Literature search strategy
To conduct this meta-analysis, we performed electronic searches in various databases until December 2022 under the guidance of a library search specialist to identify all RCTs that were related to light therapy intervention for AD or dementia. The subject terms used in the search included “Alzheimer Disease,” “Phototherapy,” “Sleep Disorders,” and “Cognition Disorders,” while free terms comprised “Dementia,” “AD,” “Light Therapy,” “Sleep Wake Disorder,” “Disorder,” and “Cognition.” For a complete list of the used keywords, see S2 Table. The searched databases encompassed Embase, the Clinical Trials Registry, Web of Science, PubMed, and the Cochrane Library. We searched for the keywords described above in each database with separate title, keywords, and abstract filters for each search string. The reference lists of all relevant studies in previous reviews were examined to avoid potential omissions.
Inclusion and exclusion criteria
Inclusion criteria for selected studies:
- Patients were older adults (aged 60–85 years) diagnosed with AD by health care institutions and had Mini-Mental State Examination (MMSE) scores between 6 and 26.
- Study design was RCT with complete data and results.
- Patients in the intervention group had undergone light therapy, whereas those in the control group had received dim light or usual care.
- Patients who were not sensitive to light.
- Studies had at least one of the following outcome indicators: SE, interdaily stability (IS), intradaily variability (IV), Pittsburgh sleep quality index (PSQI), relative amplitude (RA), or wake after sleep onset (WASO) or at least one of the following instruments: the Alzheimer’s Disease Assessment Scale-Cognitive subscale (ADAS-cog), Cohen-Mansfield agitation inventory (CMAI), Cornell Scale for Depression in Dementia (CSDD), Zarit Caregiver Burden Interview (ZBI), Neuropsychiatric Inventory (NPI), or MMSE.
Exclusion criteria for selected studies:
- Studies including patients who regularly use valium, antidepressants, or sleeping pills.
- Studies involving patients with eye diseases (e.g., glaucoma, blindness, or cataracts) or other neurological disorders (e.g., Parkinson’s disease or intellectual disability).
Study selection
All references were stored and managed in EndNote X9. The included studies were screened by two independent reviewers according to the title and abstract. The extracted results were then cross-checked for accuracy and consistency, and all eligible studies were retained for full-text evaluation. Next, two researchers independently extracted and stored the data in the Review Manager software. In the case of controversial literature, the study inclusion was decided by the research team after discussion. To ensure the completeness of our systematic review, only articles for which the full text was available were included. Furthermore, we collected the following data (if available) from each study: first author’s name, publication year, publication country, sample size, gender, the type of light therapy, pre-intervention MMSE score, and the data of the outcome indicators. To avoid potential errors, two researchers assessed the data before analysis. Finally, all relevant information was extracted to the standard template by two reviewers independently.
Assessment of quality and risk of bias
Two researchers used the risk bias assessment tool in the Cochrane Handbook of Systematic Reviews of Interventions (version 5.1.0) [26] to evaluate the included literature in terms of high, unclear, and low risk of bias based on random sequence generation, allocation concealment, blinding of participants and performers, outcome evaluator, study data integrity, selective outcome reporting, and other sources of bias. Each study was evaluated using “yes (low bias),” “unclear (lack of relevant information or uncertainty about bias),” or “no (high bias)” responses for the seven items mentioned above. The risk of bias was determined by two researchers independently. Moreover, if any disagreement arose, a third researcher was asked to help resolve the debate.
Outcome measures
A total of 12 outcome indicators were included in this meta-analysis: SE, IS, IV, PSQI, RA, WASO, ADAS-cog, CMAI, CSDD, ZBI, NPI, and MMSE.
Statistical analysis
We used Review Manager (RevMan) version 5.3 software (The Nordic Cochrane Center, The Cochrane Collaboration, Copenhagen, Denmark) to combine the results of the various studies and assess the overall effect of light therapy. Mean difference (MD) was used as a summary statistic in the trials that assessed the same outcome and measure using a consistent method [27]. Furthermore, MD and 95% confidence intervals (CIs) were employed to represent the outcomes because all outcomes were continuous, wherein MD was calculated as treatment-control. The heterogeneity was evaluated by I2 statistics and expressed as the percentage of variation across studies. In this test, an I2 value of <50% indicates no heterogeneity, while an I2 value of >50% demonstrates a high degree of heterogeneity. P < 0.05 was considered statistically significant. Furthermore, we performed subgroup analysis to explore the sources of heterogeneity across the included studies. Finally, funnel plots were used to evaluate the included literature for publication bias.
Results
The study selection process is illustrated in Fig 1. A total of 318 articles were filtered via the database search, along with three relevant studies that were included after applying the snowballing method. Further, 257 articles remained after the removal of duplicate articles. According to the screening process of the title and abstract of the articles, 218 articles were excluded due to the following reasons: 38 articles were not RCTs, three were meta-analyses, 70 were animal experiments, 46 did not use eligible interventions, 10 did not involve eligible populations, seven did not have full-text availability, and 44 had irrelevant research objectives. Furthermore, we screened the full text of the remaining 39 articles, after which 10 were excluded because they did not have the required outcome indicators and 14 were removed due to incomplete data. Finally, a meta-analysis of 15 studies (encompassing 598 patients with AD) was conducted [28–42] to investigate the effects of light therapy versus usual care on sleep disorders and psychobehavioral symptoms in patients with AD.