Non-invasive spinal cord stimulation for upper-limb impairments: a mechanistic electrophysiological and neuroimaging study in able-bodied individuals
DOI: Pending DOI
Synopsis
Cervical transcutaneous spinal cord stimulation (tSCS) shows promise for upper-limb neurorehabilitation, yet the determinants of underlying effects remain poorly understood. In fourteen able-bodied participants, we systematically compared evoked muscle responses across four anode placements and two cathode levels, while acquiring structural MRI to characterize brachial plexus trajectories and spinal cord cross-sectional area (CSA). Anode location critically shaped recruitment thresholds and response amplitudes; middle clavicle placement yielded the lowest thresholds while medial placement produced the largest responses. Motor thresholds correlated with CSA at C4–C5, suggesting individual spinal cord anatomy partly drives stimulation response. These findings support integrating neuroimaging into tSCS parameter optimization.
Background
Recovery of arm and hand function following stroke or spinal cord injury (SCI) remains a key priority for those affected and a major challenge in neurorehabilitation. Cervical transcutaneous spinal cord stimulation (tSCS) has demonstrated promising therapeutic effects for improving upper-limb function after SCI (1), yet its underlying mechanisms remain poorly understood. Here, we sought to gain insights into the neural processes engaged by cervical tSCS through an ongoing clinical study investigating different stimulation configurations in able-bodied participants (NCT07334977).
Methods
Electrophysiological and neuroimaging data were collected from fourteen able-bodied participants. Each experimental electrophysiological session consisted of the administration of monophasic tSCS paired pulses delivered every 5 s at an inter-pulse interval of 50 ms with increasing stimulation amplitude. The cathode was placed at either C5/C6 or C7/T1 vertebrae, and the anodes placed bilaterally at one of the following locations: medial, lateral, middle clavicles, or iliac crests. Evoked responses peak-to-peak amplitudes were measured from bilateral upper-limb muscles to obtain muscle- and condition-specific recruitment and paired-pulse depression curves. STIR and T2-weighted MRI sequences were additionally acquired to respectively characterize plexus nerves trajectory and measures of spinal cord cross-sectional areas (CSA) across vertebral levels using the spinal cord toolbox (SCT v7.1) (2). Mixed linear models were used to compare evoked responses across muscles and conditions. Pearson correlations were used to correlate electrophysiological measures and CSA across stimulation configurations.
Results
Consistent differences in response thresholds, recruitment curves, and paired-pulse depression were observed across muscles and participants for all stimulation conditions. Anodes placed over the middle clavicles were associated with the overall lowest response thresholds, while those placed medially resulted in largest-amplitude responses and depression. Muscle recruitment did not differ significantly between C5/C6 and C7/T1 cathode placements; however, C5/C6 resulted in lower motor thresholds, and larger amplitude responses and depression. Other stimulation parameters, including pulse width, also influenced these electrophysiological outcomes. STIR acquisitions enabled 3D reconstruction of brachial plexus nerves trajectories, providing anatomical for interpreting the condition-dependent differences observed. Motor thresholds recorded from wrist extensors and index finger abductors with tSCS using middle and lateral anode placements correlated with CSA measured at the C4 and C5 vertebral levels.
Discussion
These findings demonstrate that anode placement is a critical determinant of upper-limb muscle recruitment during cervical tSCS, with middle clavicle placement minimizing response threshold and medial placement maximizing response amplitude and depression; the later suggesting higher relative recruitment of sensory afferents (3). This distinction likely reflects differential engagement of neural structures, as supported by brachial plexus trajectory reconstructions. The correlation between motor thresholds and spinal cord CSA suggest that anatomical variations may partly account for inter-individual differences in cervical tSCS response. Further neuroimaging outcomes, including diffusion and functional measures, will further be examined in stroke and SCI populations, to better characterize the neural substrates of tSCS effects and guide protocol optimization for upper-limb neurorehabilitation.
Disclosures
N/A
References
1. Moritz C, Field-Fote EC, Tefertiller C, van Nes I, Trumbower R, Kalsi-Ryan S, et al. Non-invasive spinal cord electrical stimulation for arm and hand function in chronic tetraplegia: a safety and efficacy trial. Nat Med. 2024 May 20;1–8. https://doi.org/10.1038/s41591-024-02940-9
2. De Leener, B., Lévy, S., Dupont, S. M., Fonov, V. S., Stikov, N., Louis Collins, D., Callot, V., & Cohen-Adad, J. (2017). SCT: Spinal Cord Toolbox, an open-source software for processing spinal cord MRI data. NeuroImage, 145(Pt A), 24–43. https://doi.org/10.1016/j.neuroimage.2016.10.0093.
3. Alashqar A, Brihmat N, Gemar V, Hu Z, Koh S, Diaz S, de Freitas R, Sasaki A, Milosevic M, Keesey R, Seáñez I, Neufeld E, Hofstoetter U, Minassian K, Wagner F, Rowald A. Virtual prototyping of non-invasive spinal cord electrical stimulation targeting upper limb motor function. bioRxiv 2026. https://doi.org/10.64898/2026.01.22.701010