Electroencephalography and Local Field Potential fusion to characterize effect of Deep Brain Stimulation in Parkinson's disease

Background and Objectives: Deep brain stimulation (DBS) is a well-established therapy for Parkinson’s disease (PD), yet its neurophysiological mechanisms are not fully understood. This study investigated DBS and medication effects through multimodal EEG and local field potential (LFP) recordings in PD patients implanted with sensing-enabled devices targeting the subthalamic nucleus (STN) or globus pallidus internus (GPi). Methods: Twenty PD patients (12 STN-DBS, 8 GPi-DBS) underwent simultaneous 64-channel EEG and LFP recordings across four therapeutic conditions: medication-off/stimulation-off (M0S0), medication-off/stimulation-on (M0S1), medication-on/stimulation-off (M1S0), and medication-on/stimulation-on (M1S1). Following comparison of four ECG artifact removal methods on LFP signals, spectral power analysis and cortico- subcortical connectivity assessment using imaginary coherence were performed across alpha and beta frequency bands. Results: SVD-based algorithms demonstrated superior performance for ECG removal. In STN-DBS patients, medication predominantly suppressed low beta power while stimulation induced desynchronization of pathological cortico-subcortical coupling, particu- larly in the low beta band. GPi-DBS exhibited distinct characteristics, with medication showing extensive cortical and connectivity effects primarily in beta bands, whereas stimulation provided complementary subcortical modulation, suggesting an output pathway substitution mechanism. Discussion and Conclusions: These findings suggest that DBS acts through target-specific network reorganization beyond simple beta suppression. STN-DBS appears to modulate input pathways with complementary medication and stimulation mechanisms, while GPi-DBS operates through output pathway substitution with medication-dominant therapeutic effects. The complementary nature of medication and stimulation effects supports the development of personalized DBS programming strategies based on individual neurophysiological profiles. Keywords: Deep brain stimulation, Parkinson’s disease, beta oscillations, imaginary coherence, subthalamic nucleus, globus pallidus internus, multimodal electrophysiology.