Low-Frequency Deep Brain Stimulation in Non-Rapid Eye Movement Sleep Modifies Memory Retention in Parkinson's Disease

Abstract

Background and objective: Memory impairment is a frequent and debilitating symptom in neurodegenerative disorders. The objective of this study was to provide proof-of-principle that deep brain stimulation during sleep can modify memory consolidation in people with Parkinson's disease depending on the stimulation frequency that is applied.

Methods: Twenty-four patients with Parkinson's disease who were treated with deep brain stimulation of the subthalamic nucleus were included in this single-blind pilot study. Six patients had to be excluded because of insomnia on the night of testing. Patients were randomized (1:1 ratio) to receiving either low frequency deep brain stimulation (4 Hz) or clinically used high frequency deep brain stimulation (130 Hz) during early non-rapid eye movement (NREM) sleep. The main outcome measure was overnight memory retention as measured by a validated declarative memory task.

Results: Patients receiving low frequency deep brain stimulation during early NREM sleep (n = 9, 4 females, mean age 61.1 ± 4.3 years) showed improved overnight memory retention (z = 2.549, P = 0.011). Patients receiving clinically used high frequency deep brain stimulation (n = 9, 2 females, mean age 62.2 ± 7.1) did not show any improvement (z = 1.023, P = 0.306) leading to a significant difference between groups (z = 2.214, P = 0.027). Stronger improvement in memory function was correlated with increased cortical low frequency activity after low frequency deep brain stimulation as measured by electroencephalography (ρ = 0.711, P = 0.037).

Conclusion: These results provide proof-of-principle that memory can be modulated by frequency-specific deep brain stimulation during sleep. © 2024 The Author(s). Movement Disorders published by Wiley Periodicals LLC on behalf of International Parkinson and Movement Disorder Society.

Keywords: adaptive brain stimulation; cognition; memory retention; neurodegeneration; prefrontal cortex; subthalamic nucleus.

FIG. 1

Study design and main results. (A) Patients performed a validated declarative memory task before and after night sleep. Frequency‐specific DBS was given during the first stage‐2 NREM sleep of the night. One group received clinically used high frequency (130 Hz) DBS, whereas the other group received low frequency (4 Hz) DBS in five 5‐minute blocks. Sleep stages were defined by visual inspection of the online EEG. (B) Patients who received low frequency DBS, but not patients who received high frequency DBS, showed an increase in recalled words. (C) Patients who received low frequency DBS showed higher prefrontal LFO (2–8 Hz) activity in their EEG (recorded in four 1‐minute OFF‐stimulation recordings) compared to patients who received high frequency DBS. (D) Stronger expression of prefrontal LFO was related to better cognitive performance in patients treated with low frequency DBS. Shaded area represents 95% confidence interval. DBS, deep brain stimulation; EEG, electroencephalography; LFO, low frequency oscillations; NREM, non‐rapid eye movement. [Color figure can be viewed at wileyonlinelibrary.com]