Circadian Rhythm Disturbance in Agitation of Alzheimer's Disease

Abstract

<h3>Introduction</h3> Neuropsychiatric symptoms (NPS) affect nearly all individuals with AD. Agitation in AD (Agit-AD) is one of the most common NPS, with an estimated prevalence over 40%. Agit-AD is a major cause of distress and burden for AD caregivers and may be driven by disrupted circadian rhythms and sleep. In healthy individuals, the circadian clock that generates these rhythms regulates patterns in mood, wakefulness, and other physiologic processes, such as body temperature and hormone levels. Individuals with Agit-AD may experience disturbance in their circadian rhythms such that aggressive and disruptive behavior is increased. This project aims to examine if agitated AD subjects (AgitAD+) express delayed circadian phases and reduced rhythm strength as compared to non-agitated AD subjects (AgitAD-) and healthy controls. Additionally, we seek to 1) evaluate if circadian phase and rhythm strength vary between subtypes of Agit-AD (affective dysregulation vs. executive dysfunction); and to 2) compare sleep amount, sleep quality, and degree of sleep disordered breathing among AgitAD+ subjects vs. AgitAD- subjects and healthy controls. <h3>Methods</h3> We studied 25 subjects (Table 1): 15 AgitAD+ subjects, 8 AgitAD- subjects, and 2 age-matched healthy controls (HC). Participants completed wrist actigraphy for one week to quantify rest/activity rhythms and sleep; consumed an ingestible, disposable temperature sensor for 1-2 days to quantify core body temperature, a marker of the endogenous circadian rhythm; and completed a single night of home sleep testing (HST; Table 2). Inclusion criteria were diagnosis of AD and age 55-90 years, and exclusion criteria include seizures, significant tremors, delirium, using CPAP for sleep apnea, restless legs syndrome, and circadian rhythm disorders. NPS were measured with the Neuropsychiatric Inventory-Clinician Version (NPI-C), the Cohen Mansfield Agitation Inventory (CMAI), the Neurobehavioral Rating Scale (NBRS), and the Cornell Scale for Depression in Dementia (CSDD). Agit-AD was defined using criteria from the International Psychogeriatric Association. AgitAD+ subjects had clinically significant agitation, defined as NPI-C Agitation or Aggression subtotals ≥ 4. Disease progression was measured using the Clinical Dementia Rating Scale (CDR), and subjects underwent a physical and neurological exam. Cognition was assessed using the Montreal Cognitive Assessment (MOCA), the Hopkins Verbal Learning Test (HVLT), the Controlled Oral Word Association Test (COWAT), Animal Naming (AN), and the Trail-Making Test (parts A & B). Sleep questionnaires included the Pittsburgh Sleep Quality Index (PSQI), the Epworth Sleepiness Scale (ESS), and the Morningness-Eveningness Questionnaire (MEQ). Caregivers were asked to complete a sleep diary for subjects and an agitation log over the week they completed actigraphy. Due to the COVID-19 pandemic, we implemented hybrid visits with telemedicine to limit in-person interactions. <h3>Results</h3> 25 subjects have completed the study to date. There were no statistically significant differences observed in actigraphy-derived sleep parameters between agitated and non-agitated AD subjects (2-group t-test); see Table 1). However, significant differences between the groups have been observed in the rest/activity rhythms: compared to subjects without agitation (AgitAD-), AgitAD+ subjects had lower levels of activity in the morning and greater activity in the afternoon and evening (Figure 1). We also have performed function-on-scalar regression (FOSR) (Goldsmith et al., 2015<i>,</i> Biometrics 71:344-353) using subject-specific activity profiles as outcomes and age and agitation/control status as predictors. Figure 2 shows functional effects of age on rest/activity rhythm (left panel) and the functional effect of agitation status (right panel); the red background shows times of day during which there were statistically significant differences, using the pointwise method of analysis (p<0.05). After adjustment for age, agitation was associated with lower activity in the morning period and a trend towards greater activity in the afternoon and evening period (not statistically significant). This preliminary finding may suggest that AgitAD+ subjects express lower and delayed activity rhythms. Additionally, the HST results show a trend toward less REM sleep and longer initial REM latency in the AgitAD+ group (Table 2). Updated results will be presented at AAGP (estimated 6-10 additional subjects). <h3>Conclusions</h3> The collection of multimodal noninvasive data on sleep, activity, and core body temperature is feasible in agitated AD outpatients. This study could provide evidence for CRD/sleep disturbance as a driver of Agit-AD and could identify Agit-AD subtypes who are most likely to benefit from chronobiotic or sleep interventions for Agit-AD. Furthermore, this study could provide evidence for the use of actigraphy as a surrogate biomarker in proof-of-concept intervention studies targeting Agit-AD. <h3>Funding</h3> National Institute on Aging, R01AG054771