Heart rate variability during sleep stages in older adults with mild cognitive impairment
Abstract
Background: Several dementias are associated to cardiovascular disease risk factors. Mild cognitive impairment (MCI) can lead to dementia. One of the less known possible cardiovascular risk factor comprises a lower index of arousals per hour during Rapid Eye Movement (REM) sleep. Therefore, the aim of this pilot study was to compare heart rate variability (HRV) during sleep stages in people with MCI and those without it considering two methodologies: spectral and Markov chain analysis. Method: Seven older adults with MCI (68.6 ± 2 years old) and the same number of older adults without it (64.6 ± 4 years old) underwent a one‐night polysomnography. Sleep stages were classified and Interbeat Intervals (IBI) were obtained from the electrocardiogram during each sleep stage. From the spectral analysis, two frequencies were considered, low (LF) and high frequency (HF). A Markov chain analysis was calculated between probabilities of transitions from arrhythmic to rhythmic states during each sleep stage. Result: Between both groups, there were not statistically significant differences in LF and HF spectral analyses within sleep stages. Instead, in the Markov chain analyses, probabilities of transitions from arrhythmic to rhythmic states during N2 and REM sleep were significant between groups. Conclusion: Further studies are needed to disentangle the relationships between transitions of the HRV and the lower index of arousals per hour during REM sleep in MCI. Older adults with MCI having fewer arousals per hour in REM sleep also could result to be more arrhythmic than controls.
The paper
Secretaría de Ciencia Tecnología e Innovación; Universidad Autónoma del Estado de Hidalgo
AAIC 2025 Abstract, 24 Dec 2025, Open access, CC BY
Presented at AAIC 2025
