Epidural Electrical Stimulation Within an Integrated Rehabilitation Pathway for Traumatic Cervical Spinal Cord Injury: Safety and 12-Month Functional Outcomes

Abstract

Background/Objectives: Epidural electrical stimulation (EES) is an investigational, off-label neuromodulation strategy used adjunctively with rehabilitation in chronic spinal cord injury (SCI), but safety and longitudinal functional data specific to traumatic cervical SCI remain limited. This is a descriptive safety-and-feasibility study, not an efficacy study; no efficacy or causal claim is made. Methods: In this single-centre observational cohort, 32 consecutive patients with traumatic cervical SCI underwent EES implantation and were included in baseline and safety analyses, while longitudinal analyses comprised the 26 participants with complete Spinal Cord Independence Measure (SCIM) III data at baseline and 3, 6, 9 and 12 months. Rehabilitation dose was quantified (approximately 1087 prescribed therapy hours over 12 months) and adverse events were classified using ISO 14155:2020 definitions with WHO–UMC causality grading. Prespecified outcomes were device- and procedure-related adverse events and total and domain-specific SCIM III scores; analyses were descriptive. Results: Patients were aged 33.8 ± 10.3 years, 28 (87.5%) were male, and baseline AIS grade was A in 31 and B in 1. Mean total SCIM III increased from 14.81 ± 8.92 at baseline to 21.77 ± 14.60, 25.12 ± 15.90, 28.23 ± 17.68 and 32.42 ± 19.45 at 3, 6, 9 and 12 months (all p < 0.001), with gains across self-care, respiration and sphincter management, and mobility; 10 of 32 patients (31.3%) improved by at least one AIS grade. Adverse events were infrequent: sterile hardware-site inflammation (2, 6.3%), explantation (1, 3.1%) and a first-ever seizure (1, 3.1%), without stimulation-related neurological deterioration. Conclusions: The cohort was highly selected (31/32 baseline AIS A; 87.5% male; injury-to-implant interval 10–339 months), no comparator group was available, stimulation programming was individualised rather than protocolised, and 6 of 32 patients lacked complete 12-month follow-up; these features limit reproducibility and generalisability and preclude any inference about effectiveness. EES with rehabilitation was well tolerated and accompanied by progressive functional gains, although the uncontrolled design precludes causal inference. The contribution of this report is therefore transparent safety, dose and programming data, together with a prespecified protocol framework intended to make a controlled, propensity-matched evaluation of EES in traumatic cervical SCI feasible and reproducible.