Associations between sleep and Alzheimer’s disease biomarkers within the EPAD cohort

Abstract

AbstractBackgroundChanges in sleep quality are common in Alzheimer’s Disease (AD) and may even play a role in the onset and accumulation of disease. However, in preclinical stages, it is unclear whether sleep quality or sleep disturbance relate to disease pathology after controlling for known covariates and AD risk factors. This study aimed to determine if self‐reported sleep quality is associated with AD biomarkers after accounting for such factors.MethodData were obtained from the European Prevention of Alzheimer’s Disease (EPAD) Longitudinal Cohort Study (LCS) data set (v1500.0). Cerebrospinal fluid (CSF) samples were collected for measurement of b‐amyloid (Ab42) and phosphorylated tau (p‐tau). Self‐reported sleep quality was assessed by the Pittsburgh Sleep Quality Index (PSQI). Participants with Clinical Dementia Rating (CDR) scale scores greater than 0.5, indicative of dementia, were excluded. Linear regression models were used to determine whether the p‐tau/Ab42 ratio was associated with PSQI component scores when controlling for multiple potential confounders such as demographics, ApoE4 status, depressive symptoms, body mass index, vascular risks, smoking status, use of psychotropics, white matter lesions, and hippocampal volume. PSQI component scores of 2 or 3 were combined due to small numbers of participants with component scores of 3.ResultFrom the 1500 participants in the EPAD‐LCS v1500.0 data, a total of 1239 participants were included in the regression analysis (mean age=65.30 years, SD=7.11; mean PSQI total score=5.31, SD=3.38). After adjustment for all covariates, higher p‐tau/Ab42 ratio was found to be associated with longer self‐reported sleep latency (component score of 1: b=0.16, p=0.007; component score of 2/3: b=0.12, p=0.134; R2=0.22) and better self‐reported sleep efficiency (component score of 1: b=‐0.22, p=0.04; component score of 2/3: b=‐0.31, p=0.009; R2=0.22)ConclusionWithin the EPAD cohort, longer sleep latency and greater efficiency are associated with greater AD pathology after controlling for covariates. These findings contribute to growing evidence suggesting sleep is an important early marker of underlying neurodegeneration. Longitudinal assessment of EPAD‐LCS participants will allow for the evaluation of self‐reported sleep as a predictive marker of neurodegeneration.