Brain-controlled Epidural Spinal Cord Stimulation Facilitates Descending Motor Control After Complete Spinal Cord Injury

Introduction: Brain–computer interface (BCI)–guided epidural spinal cord stimulation (ESCS) builds upon the established role of ESCS in neurosurgery for chronic pain and the growing evidence for its ability to modulate spinal circuits and enhance motor recovery after stroke, spinal cord injury (SCI), and movement disorders. Integrating BCIs with ESCS offers the potential for brain-driven, targeted stimulation, yet few studies have explored this approach for upper-limb motor restoration.

Methods: We report the case of an individual with chronic, motor-complete cervical SCI (AIS A, C4/5) who underwent implantation of an electrocorticography (ECoG)-based BCI to detect motor intent and control a unilateral ESCS device. Percutaneous leads were placed in the right epidural space caudal to the injury, targeting dorsal rootlets from C5–T1 to engage upper-limb sensorimotor circuits. The participant completed a 28-day BCI-ESCS intervention (19 sessions, ~40 min/session), progressing from simple to complex functional tasks, with real-time stimulation triggered upon detection of motor intent. Safety assessments included neurological status, independence, quality of life, and pain. Neurophysiological measures (motor-evoked potentials), three-dimensional kinematic analysis, and muscle synergy analysis were conducted pre- and post-intervention, with a one-month follow-up. Upper-limb motor function was evaluated with the Toronto Rehabilitation Institute Hand Function Test (TRI-HFT) and GRASSP quantitative prehension, and compared with a control cohort receiving conventional therapy.

Results: No adverse effects were observed. Following therapy, corticospinal excitability increased significantly in the extensor carpi radialis and flexor carpi radialis muscles. Kinematic analysis demonstrated smoother reaching movements and persistent changes in initial movement direction, suggesting motor exploration and adaptation. Muscle synergy analysis indicated increased neuromuscular coordination complexity post-therapy. These neurophysiological and biomechanical improvements returned to baseline at one-month follow-up.
Functionally, the targeted right hand improved in the TRI-HFT 10-object subtest from 19 to 28 points post-therapy, with gains retained (30 points) at follow-up. The rectangular blocks task improved from 19 to 36 points, with sustained follow-up performance. GRASSP quantitative prehension improved by 4 points, exceeding the minimal detectable change after SCI. Functional gains exceeded those in the conventional therapy control cohort.

Conclusion: This first-in-human case demonstrates that BCI-triggered ESCS can transiently enhance corticospinal excitability, improve neuromuscular coordination, and yield sustained upper-limb functional gains in an individual with motor-complete SCI at recovery plateau. These results support the safety and potential efficacy of BCI-ESCS and warrant further investigation in larger clinical cohorts.