Associations between Quantitative Sleep EEG Data and Subsequent Cognitive Decline in Community-Dwelling Older Women

Abstract

Introduction: Healthy aging is associated with changes in both circadian and homeostatic sleep regulatory processes as well as a higher prevalence of sleep disorders, which in turn can accelerate the aging process and increase the risk for cognitive impairment. EEG power spectrum can provide additional information regarding the memory consolidation processes compared to the traditional sleep scoring. This study examined whether quantitative sleep EEG traits predict development of mild cognitive impairment (MCI) and/or incident dementia. Methods: Participants consisted of a randomly selected subgroup of women (mean age 83 yrs) from the Sleep and Cognition Study, an ancillary study to the longitudinal Study of Osteoporotic Fractures (SOF), who were free of cognitive impairment at the time of a baseline polysomnography study (SOF Visit 8). Cases (n=85) were women who developed mild MCI or dementia by objective testing at an exam 4 years after polysomnography. Controls were those with no MCI/Dementia (n=85). Absolute power for EEG frequency bands were generated using off-line spectral analysis based on a Fast-Fourier Transform (FFT) Routine (Vitascore, TEMEC). Results: At Visit 8, groups did not differ in age, BMI, education, co-morbidities, subjective sleepiness, sleep architecture as well as respiratory disturbance, arousal and periodic limb movement indices. Differences in EEG power density during nonREM sleep were observed over a broad frequency range, with higher values in the Dementia/MCI group particularly in the alpha band (8-12Hz;p=0.003), but also in the sigma (12-15Hz;p=0.04) and theta (4-8Hz;p=0.03) bands, and for slow wave activity (0.75-4.5Hz) during the first 2 hours of sleep (p=0.02) and during hours 4 and 6 (p=0.01). Conclusion: Our study provides evidence for quantitative EEG changes, which precede the onset of cognitive decline and the diagnosis of dementia in elderly women. Whether the observed differences are functionally related to neurodegenerative changes or represent unspecific EEG alterations remains to be investigated. Support (if any) NIH grant # K23HL103850 (ID)