Preliminary EEG Connectivity Signatures of Long-Term Spinal Cord Stimulation in Chronic Pain
DOI: Pending DOI
Synopsis
Spinal cord stimulation (SCS) is an established therapy for chronic refractory pain. Prior studies suggest that SCS can modulate cortical and subcortical network activity, including somatosensory, limbic, and default mode networks associated with the sensory and affective processing of pain as well as the descending pain modulatory system [1-3]. However, the field is still sparse. The effects of long-term SCS on brain network organization remain incompletely understood, and a consistent connectivity signature has not been established yet. To address this, we recorded high-density (60-channel) EEG in six chronic pain patients beyond the perioperative period (27.24±17.56 months post-implantation) during SCS OFF and clinically optimized ON conditions (Tonic, high frequency stimulation (HFS), Burst). Functional connectivity was quantified via magnitude-squared coherence within individually defined alpha bands (7-13Hz) [4]. Across subjects, SCS was associated with changes in alpha-band connectivity, most consistently reflected by increased frontal and long-range interactions alongside reductions in a subset of connections. These reductions showed a heterogeneous spatial distribution across subjects, involving posterior, lateral, and in some cases frontal regions. These patterns are broadly consistent with prior neuroimaging work suggesting modulation of large-scale brain networks [5]. Notably, the direction and spatial distribution of connectivity changes varied across stimulation paradigms. While tonic stimulation tended toward reductions in long-range connectivity, HFS and burst showed more localized and heterogeneous effects, with burst cases demonstrating reduced frontal connectivity. Overall, these preliminary findings suggest that long-term SCS may be associated with a redistribution of cortical network organization rather than uniform modulation, though conclusions remain limited by the small sample size and inter-subject variability.
Background
Chronic pain is associated with alterations in cortical network organization, but the long-term effects of spinal cord stimulation (SCS) on these networks remain poorly understood. Most EEG studies have focused on early or perioperative periods following implantation, leaving the impact of sustained SCS largely unexplored. In this study, we examined EEG-based functional connectivity in patients with long-term SCS to characterize how ongoing stimulation is associated with changes in cortical network dynamics.
Methods
High-density 60-channel resting-state EEG was recorded from six chronic pain subjects (4 females, 2 males; age = 61.3±9.0 years) at an average of 27.2±17.6 months post-SCS implantation. Recordings were obtained under both stimulation OFF and clinically optimized ON conditions, including Tonic stimulation at 40Hz (n=1), high-frequency stimulation (HFS) at 10kHz (n=3), and Burst stimulation at 40Hz-500Hz (n=2). Functional connectivity was quantified using magnitude-squared spectral coherence computed between all electrode pairs. Analyses were specifically focused within a wide alpha band (7-13Hz), individually defined as ±1Hz around the peak alpha frequency for each subject. To characterize state-dependent alterations in alpha coherence and brain network organization, we compared balanced SCS OFF and SCS ON epochs. Network connectivity matrices were evaluated to assess waveform-specific changes and to identify global and regional shifts in synchronization.
Results
Our cohort included subjects with failed back surgery syndrome (n=3), chronic neuropathic leg pain (n=2), and complex regional pain syndrome (n=1). During the EEG visit, subjects verbally reported pain relief following SCS, with overall improvements ranging from 50% to 100%. Alpha‑band EEG connectivity demonstrated that switching the SCS from OFF to ON was associated with changes in cortical network organization. Across subjects, the most consistent pattern was increased frontal and long-range connectivity, along with reductions in a subset of connections with heterogeneous spatial distribution. The direction and extent of these changes varied across stimulation paradigms. Tonic stimulation tended toward reductions in long-range connectivity, whereas HFS and burst stimulation showed more localized and variable effects. Notably, burst stimulation was associated with reduced frontal connectivity in a subset of subjects. Although these patterns highlight spatial signatures of cortical engagement for each waveform, no uniform global connectivity pattern was observed across subjects or stimulation conditions.
Figure 1. Reorganization of cortical alpha-band connectivity across spinal cord stimulation (SCS) waveforms from OFF to ON conditions. (A) Connectivity heatmaps: Adjacency matrices displaying the coherence landscape across all electrode pairs, highlighting connectivity reorganization as OFF-only (loss, dark gray), ON-only (gain, dark blue), and persistent (modulated, dark gold). (B) Power spectral density (PSD): Frequency-domain representations of alpha-band power distribution, showing peak frequencies and waveform-specific differences in spectral intensity. (C) Top 5 connectivity shifts: The largest changes in coherence (Shift = ON − OFF). Red bars represent increased synchronization, with the strongest effects observed in parietal/posterior hubs for high-frequency stimulation (HFS) and frontal/anterior hubs for burst stimulation. Blue bars indicate decreased synchronization, reflecting more global desynchronization, which is most prominent during tonic stimulation.
Discussion
These findings suggest that SCS may influence large-scale brain network organization rather than producing uniform changes in cortical activity. The observed patterns are consistent with a redistribution of connectivity toward more distributed and integrative network dynamics, which may reflect engagement of higher-order cognitive-affective processes associated with the medial pain pathway, rather than purely sensory-discriminative processing [5, 6]. Variability across stimulation paradigms further indicates that different waveforms may interact with cortical networks in distinct ways, potentially aligning with prior reports of differential modulation of sensory vs affective components of pain [1, 5, 6]. Given the small sample size and inter-subject heterogeneity, these observations remain preliminary but support the feasibility of using EEG-based connectivity to study network-level effects of long-term SCS.
Conclusion: In this cohort, long-term SCS was associated with changes in cortical functional connectivity, reflected in changes in alpha-band coherence during stimulation. The observed patterns suggest a redistribution of network organization rather than uniform modulation, potentially involving both sensory and higher-order cognitive-affective circuits. While waveform-specific trends were noted, the small sample size and inter-subject variability limit definitive interpretation. These preliminary findings support further investigation of EEG-based connectivity as a potential marker of network-level effects of sustained SCS in chronic pain.
Disclosures
Dr. Telkes received grant support from NIH R00NS119672, University of Arizona Technology Research Initiative Fund (TRIF), College of Medicine-Tucson Spark Grant, and Arizona Department of Health Services RFGA2024-022-016.
References
[1]. Deogaonkar M, Sharma M, Oluigbo C, Nielson DM, Yang X, Vera-Portocarrero L, Molnar GF, Abduljalil A, Sederberg PB, Knopp M, Rezai AR. Spinal Cord Stimulation (SCS) and Functional Magnetic Resonance Imaging (fMRI): Modulation of Cortical Connectivity With Therapeutic SCS. Neuromodulation. 2016 Feb;19(2):142-53. doi: 10.1111/ner.12346. Epub 2015 Sep 16. PMID: 26373920.
[2]. Witjes B, Ottenheym LA, Huygen FJPM, de Vos CC. A Review of Effects of Spinal Cord Stimulation on Spectral Features in Resting-State Electroencephalography. Neuromodulation. 2023 Jan;26(1):35-42. doi: 10.1016/j.neurom.2022.04.036. Epub 2022 May 10. PMID: 35551867.
[3]. De Groote S, Goudman L, Peeters R, Linderoth B, Van Schuerbeek P, Sunaert S, De Jaeger M, De Smedt A, De Andrés J, Moens M. The influence of High Dose Spinal Cord Stimulation on the descending pain modulatory system in patients with failed back surgery syndrome. Neuroimage Clin. 2019; 24:1020872019; 24:102087. doi: 10.1016/j.nicl.2019.102087. Epub 2019 Nov 12. PMID: 31795057; PMCID: PMC6978217.
[4]. Bowyer SM. Coherence a measure of the brain networks: past and present. Neuropsychiatr Electrophysiol 2, 1 (2016). https://doi.org/10.1186/s40810-015-0015-7
[5]. Gartner KE, Rustioni S, Vohra A, Almosawi M, Hill N, Stewart T, van Wouwe NC, Zemmar A. A comprehensive review of the supraspinal mechanisms of spinal cord stimulation on chronic pain and cognition. Front Pain Res (Lausanne). 2025 Jul 28; 6:1589723. doi: 10.3389/fpain.2025.1589723. PMID: 40792126; PMCID: PMC12336175.
[6]. De Ridder D, Adhia D, Vanneste S. The anatomy of pain and suffering in the brain and its clinical implications. Neurosci Biobehav Rev. 2021 Nov; 130:125-146. doi: 10.1016/j.neubiorev.2021.08.013. Epub 2021 Aug 16. PMID: 34411559.