Neurobiology of Aging Volume 80, August 2019, pp. 71-82. Free full-text
DOI: 10.1016/j.neurobiolaging.2019.04.002
Péter P Ujma1,2, Péter Simor3, Axel Steiger4, Martin Dresler5*, Róbert Bódizs1,2*
1 Semmelweis University, Institute of Behavioural Science, Budapest
2 National Institute of Clinical Neuroscience, Budapest
3 Institute of Psychology, ELTE, Eötvos Loránd University, Budapest
4 Max Planck Institute of Psychiatry, Munich
5 Donders Institute of Brain, Cognition and Behaviour, Radboud University Medical Centre, Nijmegen
* equal contribution
Abstract
Slow wave activity is a hallmark of deep NREM sleep. Scalp slow wave morphology is stereotypical, it is highly correlated with the synchronized onset and cessation of cortical neuronal firing measured from the surface or depth of the cortex. It is also strongly affected by ageing, and these changes are causally associated with age-related cognitive decline. We investigated how normal ageing affects the individual morphology of the slow wave, and whether these changes are captured by the summary slow wave parameters generally used in the literature. We recorded full-night polysomnography in 176 participants (age 17-69 years) and automatically detected slow waves. We established individual slow morphologies using average amplitude at 501 data points for each participant and also calculated the individual average slow wave amplitude, average ascending and descending slope steepness, halfwave duration and the total number of slow waves (gross parameters). Using LASSO penalized regression we found that SW gross parameters explain up to 60% of age variance, but using fine morphology up to 80% of age variance can be accounted for. This predictive power was greatest when data from multiple channels was averaged, in midline derivations and in the first quarter of the night. Young participants had faster slow wave polarity reversals, suggesting a more efficient initiation and termination of slow wave down- and up-states. Our results demonstrate the superiority of the high-resolution slow wave morphology as a biomarker of ageing, and highlights downstate-upstate transitions as promising targets of restorative stimulation-based interventions.
Keywords: slow wave, sleep, NREM, ageing, LASSO regression