Multi-night naturalistic cortico-basal recordings reveal mechanisms of NREM slow wave suppression and spontaneous awakenings in Parkinson’s disease

Abstract

Background: Sleep disturbance is a prevalent and highly disabling comorbidity in individuals with Parkinson's disease (PD) that leads to worsening of daytime symptoms, accelerated disease progression and reduced quality of life. Objectives: We aimed to investigate changes in sleep neurophysiology in PD particularly during non-rapid eye movement (NREM) sleep, both in the presence and absence of deep brain stimulation (DBS). Methods: Multi-night (n=58) intracranial recordings were performed at-home, from chronic electrocorticography and subcortical electrodes, with sensing-enabled DBS pulse generators, paired with portable polysomnography. Four people with PD and one person with cervical dystonia were evaluated to determine the neural structures, signals and connections modulated during NREM sleep and prior to spontaneous awakenings. Recordings were performed both ON and OFF DBS in the presence of conventional dopaminergic replacement medications. Results: We demonstrate an increase in cortico-basal slow wave activity in delta (1-4 Hz) and a decrease in beta (13-31 Hz) during NREM (N2 and N3) versus wakefulness in PD. Cortical-subcortical coherence was also found to be higher in the delta range and lower in the beta range during NREM versus wakefulness. DBS stimulation resulted in a further elevation in cortical delta and a decrease in alpha (8-13 Hz) and low beta (13-15 Hz) power compared to the OFF stimulation state. During NREM sleep, we observed a strong inverse interaction between subcortical beta and cortical slow wave activity and found that subcortical beta increases prior to spontaneous awakenings. Conclusions: Chronic, multi-night recordings in PD reveal opposing sleep stage specific modulations of cortico-basal slow wave activity in delta and subcortical beta power and connectivity in NREM, effects that are enhanced in the presence of DBS. Within NREM specifically, subcortical beta and cortical delta are strongly inversely correlated and subcortical beta power is found to increase prior to and predict spontaneous awakenings. We find that DBS therapy appears to improve sleep in PD partially through direct modulation of cortico-basal beta and delta oscillations. Our findings help elucidate a contributory mechanism responsible for sleep disturbances in PD and highlight potential biomarkers for future precision neuromodulation therapies targeting sleep and spontaneous awakenings.

Generalized sleep decoding with basal ganglia signals in multiple movement disorders

“…Failure to identify a patient's current sleep state and persistently using the wakeful beta activity level as a threshold may result in either insufficient or excessive stimulation during sleep 32 . With BGOOSE, beta thresholds used for informing stimulation can be adjusted in pace with sleep stage transitions (e.g., adjusting the beta threshold to a lower level during NREM sleep so that pathological beta in PD could be better suppressed through stimulation 32,45 ). In addition, obtaining the patient's sleep stage enables the possibility of enhancing specific beneficial waveforms relevant to different sleep stages, such as spindle and slow waves in NREM sleep 46 and sawtooth waves in REM sleep 47 .…”

“…One potential reason could be that the reduced proportion of N3 sleep due to sleep disturbances in a considerable portion of movement disorder patients complicates training and generalization. Accumulating more deep sleep data from chronic recordings for model training 45 or identifying N3 sleep using ECoG electrodes 12 are potential solutions. Fourth, BGOOSE relies on feature extraction and tree-based algorithms for sleep staging recognition, whereas deep learning algorithms have shown significant potential in the sleep classification field 31,52,53 .…”

“…Including more chronically recorded data with all types of environmental artifacts in training will enhance BGOOSE's ability to generalize. Sixth, BGOOSE is currently trained in the off-stimulation condition, and its performance during on-stimulation states for sleep staging remains unknown, which is a crucial consideration for continuous monitoring and stimulation 45 . Seventh, BGOOSE is trained solely for nighttime sleep, and its applicability for accurate daytime napping sleep staging is uncertain.…”

“…The first dataset, referred to as the Tsinghua dataset, comprised 12 PD patients who underwent STN-DBS surgery 16 . The second dataset, the UCSF dataset, included two PD patients who received STN-DBS and one PD patient who received GPi-DBS 45 . All 15 patients underwent overnight synchronous PSG and basal ganglia signal recording in the off-medication & off-stimulation state.…”

Proceedings of the 11th Annual Deep Brain Stimulation Think Tank: pushing the forefront of neuromodulation with functional network mapping, biomarkers for adaptive DBS, bioethical dilemmas, AI-guided neuromodulation, and translational advancements

“…This demonstrated a reduction in dyskinesia and improved movement speeds in a natural typing task. Finally, they demonstrated multi-night at-home sleep recordings and negative interactions between beta and slow waves, in addition to a fully data-driven aDBS algorithm for sleep stage modification targeting NREM sleep (Anjum et al, 2023 ; Smyth et al, 2023 ). Their work shows that personalized nighttime stimulation adjustments were well tolerated with initial evidence that this technique might be able to optimize NREM slow waves linked to disease progression.…”