The prevalence of Parkinson’s disease (PD) increases with aging and both processes share similar cellular mechanisms and alterations in the dopaminergic system. Yet it remains to be investigated whether aging can also demonstrate electrophysiological neuronal signatures typically associated with PD. Previous work has shown that phase-amplitude coupling (PAC) between the phase of beta oscillations and the amplitude of gamma oscillations as well as beta bursts features can serve as electrophysiological biomarkers for PD. Here we hypothesize that these metrics are also present in apparently healthy elderly subjects. Using resting state multichannel EEG measurements, we show that PAC between beta oscillation and broadband gamma activity (50-150 Hz) is elevated in a group of elderly (59-77 years) compared to young volunteers (20-35 years) without PD. Importantly, the increase of PAC is statistically significant even after ruling out confounds relating to changes in spectral power and non-sinusoidal shape of beta oscillation. Moreover, a trend for a higher percentage of longer beta bursts (> 0.2 s) along with the increase in their incidence rate is also observed for elderly subjects. Using inverse modeling, we further show that elevated PAC and longer beta bursts are most pronounced in the sensorimotor areas. Moreover, we show that PAC and longer beta bursts might reflect distinct mechanisms, since their spatial patterns only partially overlap and the correlation between them is weak. Taken together, our findings provide novel evidence that electrophysiological biomarkers of PD may already occur in apparently healthy elderly subjects. We hypothesize that PAC and beta bursts characteristics in aging might reflect a pre-clinical state of PD and suggest their predictive value to be tested in prospective longitudinal studies.
Changes in incidence rate of longer duration windows (0.2–0.5 s) (old vs. young).
A. Spatial topography of the difference in incidence rates of bursts with different durations. Electrodes with labels showed significant difference between the two groups (old vs. young) after FDR correction (p < 0.05). For bursts with a duration of 0.2–0.3 s, a significant increase was observed in many scalp sites but most prominently in fronto-central regions. For bursts with a duration between 0.4 and 0.5 s, the most prominent differences were found in centro-parietal areas. B. Spatial difference pattern (old vs. young) in burst incidence rate of long beta bursts (0.2–0.5 s) in source level. C. Normalized histogram of mean incidence rates of longer bursts (0.2–0.5 s) across all channels and subjects for the old (in red) and young (in black) group. Each count represents one channel from one subject. Color bar indicates the test statistic. Positive values indicate stronger bursting incidence in the elderly subjects.