By Dianne M. Augelli, MD
Synopsis: This study of dementia-free adults found that greater variability of the 24-hour activity rhythm (greater during the daytime as well as during sleep time) with fragmentation of sleep predicted increased deposition of amyloid-β (Aβ), especially in apolipoprotein E 4 (APOE4) carriers.
Source: Nguyen Ho PT, Hoepel SJW, Rodriguez-Ayllon M, et al. Sleep, 24-hour activity rhythms, and subsequent amyloid-β pathology. JAMA Neurol. 2024;81(8):824-834.
The authors of this study sought to assess whether the disruption of sleep and the 24-hour activity rhythm resulted in the increased accumulation of amyloid-β (Aβ), which is known to be associated with Alzheimer’s dementia (AD). The subjects of this prospective observational study were 319 dementia-free adults (mean age 61.5 years; 47% women) from the larger Rotterdam Study, a prospective population-based study started in the Netherlands in 1990.
During baseline visits, which occurred from either 2004 to 2006 or from 2012 to 2014, the study’s participants shared self-reported sleep measures, underwent genetic testing for the apolipoprotein E 4 (APOE4) allele, and provided plasma samples measuring baseline biomarkers associated with AD, including the Aβ 42/40 ratio and phosphorylated tau (p-tau181 and p-tau217). They also completed seven days of wrist-worn actigraphy, which provided objective measures of sleep, including total sleep time, sleep onset latency, sleep efficiency, and wake after sleep onset (WASO).
The participants also derived 24-hour activity rhythms, which included intraday variability (the frequency of activity and rest transitions within a 24-hour period) and interday stability (the consistency of activity patterns from day to day). They also determined the five-hour lowest activity period during wakefulness. Participants then were followed for an average of 7.8 years and subsequently underwent an assessment of Aβ deposition using positron emission tomography (PET) from 2018 to 2021.
The primary outcome was the measurement of cortical Aβ deposition using the standardized uptake value ratio (SUVR), which is the ratio of tracer uptake in cortical areas relative to a reference region. Linear regression models were performed to assess associations between baseline sleep, 24-hour activity rhythm measures, and Aβ burden at follow-up. The authors adjusted for age, sex, APOE4 allele presence, actigraphy device type, time between assessments, sleep medication use, education level, possible sleep apnea, body mass index, hypertension, diabetes, smoking, physical activity level, depression, and employment status.
Since no baseline Aβ burden via PET was available, the authors attempted to control for the possibility of baseline pathology by excluding 15% of the most abnormal baseline plasma biomarkers and applying an absolute p-tau217 cutoff point of 0.63 pg/mL. The 15% exclusion was used by the authors because the prevalence rate of positive Aβ PET among study participants also was 15%.
The study’s most significant finding was that greater intraday variability at baseline was associated with a statistically significant increased Aβ PET burden at follow-up (β, 0.15; bootstrapped 95% confidence interval [CI], 0.04 to 0.26; bootstrapped P = 0.02). This finding was more pronounced in APOE4 carriers, which showed a higher rate of Aβ burden (β, 0.38; bootstrapped 95% CI, 0.05 to 0.64; bootstrapped P = 0.03) compared to noncarriers (β, 0.07; bootstrapped 95% CI, -0.04 to 0.18; bootstrapped P = 0.19).
Subjective measures of sleep duration, sleep quality, and daytime sleepiness, as well as objective actigraphy measures of total sleep time, sleep efficiency, WASO, and interday stability, were not predictive of increased Aβ burden.
The study’s authors postulated that, because of the removal of the most abnormal biomarkers of AD, disease pathology at baseline could be determined, thus leading to their conclusion that intraday variation preceded the onset of pathological Aβ accumulations. This finding was most striking for APOE4 carriers.
Commentary
The authors theorized that disruptions to intraday rhythms could lead to increased Aβ accumulation and AD pathology via decreased glymphatic clearance or increased wake time. Their findings were limited in scope as the result of the absence of polysomnography and baseline Aβ PET data.
Polysomnography could have identified known contributors to sleep fragmentation, such as periodic limb movements and, more importantly, sleep apnea. The presence of apnea relied on self-reported symptoms, which likely underestimated the number of participants with sleep-disordered breathing. Therefore, the effect of untreated sleep apnea on sleep fragmentation or intraday variability and dementia burden in this study cannot be fully elucidated.
The absence of baseline Aβ PET data for comparison limits the ability to rule out possible underlying pathology at baseline. The use of plasma biomarkers as proxy baseline measures to exclude patients with potential dementia may not be sufficient to accurately determine dementia pathology in the same capacity as Aβ PET data. Given this uncertainty, it is unclear whether intraday dysfunction truly is a risk factor for Aβ accumulation or a symptom of the disease pathology itself. It is notable that intraday variability may be detectable prior to significant changes in plasma biomarkers or Aβ PET burden in patients who eventually develop AD.
Given the findings of this study, the presence of intraday variability by actigraphy may be an inexpensive and noninvasive means of evaluation for patients at risk for dementia, especially in APOE4 carriers. This early warning sign may be a target for future testing and/or intervention.
Future studies that include longitudinal Aβ PET scans and more effectively exclude underlying sleep disorders likely will help clarify possible causal factors leading to intraday variability and the association with Aβ accumulation and AD pathology.
Dianne Augelli, MD, is Assistant Professor of Clinical Medicine, Weill Cornell Medical Center.
