Diurnal modulation of subthalamic beta oscillatory power in Parkinson’s disease patients during deep brain stimulation

Abstract

Background: Beta band activity in the subthalamic local field potential (LFP) is correlated with Parkinson's disease (PD) symptom severity, and is the therapeutic target and feedback signal for adaptive deep brain stimulation (aDBS). While clinically relevant beta fluctuations in PD patients are well characterised on shorter timescales and in the clinic, it is not known how beta activity evolves around the diurnal cycle, outside a clinical setting. Objective: To characterise diurnal fluctuations in beta amplitude in PD patients receiving continuous, high frequency DBS. Methods: We obtained chronic recordings (34+/-13 days) of subthalamic beta power in PD patients implanted with the Percept DBS device during high-frequency DBS, and analysed its diurnal properties. To investigate the influence of non-frequency-specific effects and artifacts, we compared beta to contralateral theta amplitude and recorded LFPs during various movements. Results: Beta power had strong 24-hour periodicity, and time of day explained 41+/-9% of the variance in all long-term beta power recordings (p<0.001 in all patients). For all patients, beta activity was high during the day and reduced at night. Beta activity was not fully explained by theta activity and could show independent diurnal modulation. Movement artifacts affected the recorded LFPs, influenced band power estimates, and could have contributed to diurnal patterns in some patients. Conclusions: Diurnal fluctuations in beta amplitude will need to be accommodated in aDBS to prevent suboptimal stimulation, particularly at night. Careful screening and/or mitigation of movement artifacts is needed to ensure that the signal is suitable for adaptive stimulation or neurophysiological investigation.

When do bursts matter in the primary motor cortex? Investigating changes in the intermittencies of beta rhythms associated with movement states

“…Notably, we were able to discriminate between periods of rest and movement preparation, despite similar beta SNR observed across these states. These burst features are good candidates for control signals in closed loop neuromodulation, as they can be readily computed from narrowband data such as that available on current sensing/stimulation devices such as Percept (Medtronic) ( Van Rheede et al 2022 ) and they are known to be modulated by deep brain stimulation ( Pauls et al 2022 ). Additionally, motor state discrimination was enhanced compared to linear surrogates, with the degree of nonlinearity being largest during rest and movement preparation ( figure 3 ).…”

When do bursts matter in the motor cortex? Investigating changes in the intermittencies of beta rhythms associated with movement states

“…Notably, we were able to discriminate between periods of rest and movement preparation, despite similar beta SNR observed across these states. These burst features are good candidates for control signals in closed loop neuromodulation, as they can be readily computed from narrowband data such as that available on current sensing devices such as Percept (Van Rheede et al 2022) and they are known to be modulated by deep brain stimulation (Pauls et al 2022). Additionally, motor state discrimination was enhanced compared to linear surrogates, with the degree of nonlinearity largest during rest and movement preparation (figure 3).…”

Artifact characterization and mitigation techniques during concurrent sensing and stimulation using bidirectional deep brain stimulation platforms

“…When interpreting neural data or designing adaptive algorithms, it is important to better understand all potential sources of artifact that contaminate neural signals onboard bidirectional DBS platforms. Beyond the stimulation artifact distortion and modulation artifacts described in this work, it is important for researchers to also be aware of additional sources of artifact, including stimulation ( Dastin-van Rijn et al, 2021 ; Chen et al, 2022 ), electrocardiogram ( Neumann et al, 2021 ), and body movement ( Thenaisie et al, 2021 ; van Rheede et al, 2022 ). While some of these artifacts may be generalizable, many may only appear in certain contexts that are device, patient, or target specific.…”