P 75 Movement-induced γ oscillations in the subthalamic nucleus are increased by dopamine and scaled by velocity in patients with Parkinson’s disease

Abstract

Parkinson’s disease (PD) is a basal ganglia disorder that results in general slowness of movement and an effective treatment option consists in subthalamic deep brain stimulation (DBS). Postoperatively, DBS-electrodes are externalized which gives the unique opportunity of recording local field potentials (LFP). Previous studies have shown that there are disease specific oscillatory patterns at rest, most prominently an increased β synchronization (Kuhn, 2006) that correlates with clinical symptoms of PD such as rigidity and bradykinesia (Neumann, 2016). During movement, subthalamic oscillations show an increase of γ power that increases after L-Dopa intake (Androulidakis, 2007). In the motor cortex of healthy subjects (Cheyne, 2013) as well as in the internal pallidum of dystonia patients (Brucke, 2012) γ power is moreover scaled by kinematic parameters such as movement speed or force. Here we investigated whether γ oscillations in the human STN are related to scaling of movement and to what extent this modulation is influenced by dopamine and relates to symptom severity as assessed by the UPDRS-III motor score. To shed light on these questions, 16 PD patients (4 female, mean age 58.4 years ±2.8 SEM) undergoing electrode implantation in the STN for DBS were included in the study. Bipolar LFPs from 3 contact pairs were recorded while patients on their usual dopaminergic medication performed forearm pronation movements of three amplitudes (small, medium, large). 7 patients additionally performed the task after withdrawal of the dopaminergic replacement therapy, in the so called medical OFF state (>30% increase in UPDRS score). Signals were taken to the frequency domain using wavelet analysis, averaged and normalized to a pre-trial baseline period. Normalized time frequency plots were averaged across hemispheres and patients. Statistical analysis was conducted using FDR-corrected Monte Carlo permutation tests and correlations were calculated using Spearman’s Rho. Averaged Θ (2–8 Hz), β (13–30 Hz) and γ (40–90 Hz) power were compared across hemispheres, movement amplitudes and single trial velocities and correlated with the clinical symptom severity as measured by the UPDRS-ON score. We were able to reveal a neural signature of movement in the STN consisting in a synchronization in the Θ, a desynchronization in the β and a synchronization in the γ band compared to baseline. Our findings could be of special interest in the field of adaptive stimulation in order to monitor dopamine dependent subthalamic movement related activity in relation to current motor demand.