Periodic and Aperiodic Components of Subthalamic Nucleus Activity Reflect Different Aspects of Motor Impairment in Parkinson’s Disease

Background

Parkinson’s disease (PD) is a common neurodegenerative disorder characterized by core symptoms such as bradykinesia, rigidity, and tremor. Although the neurophysiology of PD has been extensively studied, many questions remain unanswered. In particular, identifying specific electrophysiological biomarkers associated with PD motor symptoms remains a key focus of research. These biomarkers not only help elucidate the mechanisms of the disease but also advance the development of deep brain stimulation (DBS) technologies.

Beta oscillations (13-35 Hz) in the subthalamic nucleus (STN) are recognized as the primary electrophysiological biomarkers of PD. However, despite their use as feedback signals in DBS systems, the precise boundaries of the “pathological” oscillatory range remain unclear, especially when patients are in different states (e.g., on or off levodopa). Therefore, optimizing stimulation parameters and identifying optimal biomarkers that can adapt to varying patient conditions remain critical research priorities.

Source of the Paper

This paper was co-authored by Ksenia Sayfulina, Veronika Filyushkina, Svetlana Usova, and others from the Laboratory of Human Cell Neurophysiology at the Federal Research Center for Chemical Physics of the Russian Academy of Sciences, the N.N. Burdenko National Medical Research Center for Neurosurgery, and the Moscow Center for Advanced Studies. Published in 2025 in the European Journal of Neuroscience, the paper is titled Periodic and Aperiodic Components of Subthalamic Nucleus Activity Reflect Different Aspects of Motor Impairment in Parkinson’s Disease.

Research Process and Results

Research Process

1. Study Participants
   The study included 14 PD patients (aged 36-64, mean age 51.9, 10 females). All patients underwent bilateral implantation of STN directional DBS electrodes, and postoperative recordings were made using externalized electrodes.

2. Data Collection
   STN local field potentials (LFPs) were recorded during the resting state on the 1st and 5th days post-surgery. Recordings were divided into two parts: before and after levodopa administration. Before the recordings, patients were withdrawn from levodopa overnight. Afterward, patients received 1.5 times their usual levodopa dose.

3. Data Processing and Analysis
   Data preprocessing included bandpass filtering (1-500 Hz), notch filtering, and visual inspection to exclude artifact segments. Power spectral density (PSD) in the 2-49 Hz range was calculated using Welch’s method. The periodic and aperiodic components were separated using the method proposed by Donoghue et al. (2020). The periodic component was represented as oscillatory peaks, while the aperiodic component followed a 1/fβ spectral law.

4. Cluster Analysis
   To identify functionally distinct sub-bands within the 5-35 Hz range, cluster analysis was performed on oscillatory peaks. Based on peak height, width, and frequency, Ward’s hierarchical clustering method was used to divide the peaks into four clusters corresponding to 5-14 Hz, 14-20 Hz, 20-28 Hz, and 28-35 Hz.

5. Statistical Models
   Linear mixed-effects models (LMEM) were used to analyze the relationship between STN activity and medication state/motor symptoms. Patient variability was included as a random intercept.

Key Findings

1. Effect of Levodopa on STN Activity
   Levodopa administration significantly suppressed oscillatory activity in the 11-32 Hz range and increased the slope of the aperiodic component. Changes in the aperiodic slope correlated with the alleviation of motor symptoms.

2. Relationship Between Oscillatory Activity and Motor Impairment
   In the off-medication state, the amplitude of oscillatory peaks in the 14-20 Hz range was significantly associated with overall motor impairment (including bradykinesia, rigidity, and tremor). In the on-medication state, the amplitude of oscillatory peaks in the 7-11 Hz range was significantly associated with bradykinesia.

3. Relationship Between Aperiodic Component and Symptom Alleviation
   The increase in the aperiodic slope was negatively correlated with the alleviation of bradykinesia and weakly positively correlated with the alleviation of rigidity. This suggests that the aperiodic component may serve as an effective biomarker for evaluating symptom improvement.

Conclusions and Significance

Conclusions

The study demonstrates that the periodic and aperiodic components of STN activity reflect different aspects of motor impairment in PD. In the off-medication state, low-beta oscillations (14-20 Hz) are associated with overall motor impairment, while in the on-medication state, alpha oscillations (7-11 Hz) are linked to bradykinesia. Additionally, changes in the aperiodic slope significantly correlate with the alleviation of motor symptoms, indicating its potential as an effective biomarker for evaluating treatment efficacy.

Scientific and Practical Value

This research provides new insights into the electrophysiological biomarkers of PD, particularly by separating periodic and aperiodic components to reveal the association between different frequency bands and motor impairment. These findings not only enhance our understanding of the neural mechanisms of PD but also offer potential directions for optimizing adaptive DBS systems. For example, feedback signal frequency bands could be adjusted based on the patient’s medication state to improve therapeutic outcomes.

Research Highlights

1. Novel Methodology: Cluster analysis identified functionally distinct sub-bands of STN oscillatory activity, revealing associations between different frequency ranges and motor impairment.
2. Application of Aperiodic Component: The study is the first to report the correlation between changes in the aperiodic slope and motor symptom alleviation, providing evidence for the aperiodic component as a biomarker.
3. Multi-State Analysis: The study considered both on- and off-medication states, uncovering changes in “pathological” frequency ranges across different states.

Additional Valuable Information

The study also proposed an adaptive DBS strategy based on oscillatory activity and the aperiodic component, suggesting the use of low-beta oscillations as feedback signals in the off-medication state and switching to alpha oscillations in the on-medication state. Additionally, the aperiodic slope could be used to evaluate stimulation efficacy and adjust stimulation parameters.

This research provides new directions for neuromodulation therapy in PD and lays an important foundation for future clinical practice and basic research.