Abstract
As spinal surgeons with clinical and research experience in spinal cord injury (SCI) surgery and neuromodulation, we read with great interest the case series by Wee et al 1 describing the RESTORES protocol, which integrates epidural spinal cord stimulation (SCS) with robotic-assisted gait training and comprehensive neurorehabilitation in individuals with chronic thoracic AIS A SCI. The authors are to be commended for providing a detailed and clinically valuable description of post-implant rehabilitation, an area that remains underreported despite its central importance to outcomes following neuromodulation.
We wish to focus on a single scientific issue that is fundamental to interpretation of the reported neurological outcomes: the performance of AIS/ISNCSCI examinations with the spinal cord stimulator switched on.
In their methods, Wee et al 1 state that outcome measures, including the AIS motor and sensory examination, were conducted with the spinal cord stimulator activated, and volitional lower-limb control is described explicitly in the context of active stimulation. While this approach is appropriate for evaluating stimulation-enabled functional capacity, it introduces significant ambiguity when changes in AIS motor scores are interpreted as evidence of neurological recovery.
The International Standards for Neurological Classification of Spinal Cord Injury (ISNCSCI), including the AIS grading system, represent the internationally accepted gold standard for classifying neurological impairment following SCI.2,3 These standards are designed to provide a technology-independent, standardized neurological assessment, enabling reliable longitudinal tracking, prognostication, and comparison across studies. AIS examination results are therefore intended to reflect intrinsic neurological status rather than performance augmented by assistive or neuromodulatory technologies.
By contrast, a substantial body of high-quality evidence demonstrates that epidural spinal cord stimulation can acutely enable voluntary movement, standing, and stepping when paired with task-specific training, even in individuals classified as motor complete.4 -8 These effects are widely understood to arise from neuromodulation of spinal sensorimotor networks, increasing excitability and facilitating transmission through residual descending pathways that may be anatomically preserved but functionally silent. Importantly, such effects are often state-dependent, present during active stimulation and reduced or absent when stimulation is withdrawn.
From both neurological and surgical perspectives, this distinction is critical. Improvements observed during stimulation may reflect:
stimulation-facilitated voluntary motor output,
training-induced plasticity that remains stimulation-dependent,
true neurological recovery that persists independent of stimulation, or
a combination of these mechanisms.
When AIS motor scores are obtained exclusively with stimulation ON, it is not possible to differentiate among these possibilities. This limitation is particularly relevant in the present report, where improvements in selected lower-extremity motor scores are observed without corresponding changes in sensory scores. 1 Without OFF-stimulation AIS assessments, such findings cannot be confidently attributed to neurological recovery as defined by ISNCSCI standards.
For spine surgeons involved in patient selection, surgical implantation, and long-term follow-up, this distinction has direct clinical implications. AIS classification informs prognosis, rehabilitation planning, patient counseling regarding expected recovery, and medico-legal documentation. Reporting AIS changes obtained under active neuromodulation without explicit differentiation risks conflating device-enabled performance with neurological restitution, potentially leading to misinterpretation by clinicians, patients, and the broader scientific community.
We emphasize that this issue does not diminish the importance of the authors’ findings. Demonstrating that epidural stimulation, when combined with intensive rehabilitation, can enable meaningful function years after severe SCI is of high clinical relevance. However, scientific clarity requires that neurological recovery and stimulation-dependent function be reported as distinct but complementary outcomes.
We respectfully suggest that future RESTORES publications explicitly separate:
ISNCSCI/AIS examinations performed with stimulation OFF, serving as the primary endpoint for neurological recovery; and
functional performance measures obtained with stimulation ON, reflecting stimulation-enabled capacity.
Where feasible, inclusion of within-session ON-OFF comparisons for standardized assessments would further strengthen mechanistic interpretation. This approach has precedent in prior neuromodulation studies and allows readers to appreciate both the enabling effects of stimulation and any durable neurological changes that persist in its absence.4,5,7
As the field of spinal neuromodulation advances toward broader clinical adoption, clear distinction between recovery and enablement will be essential for scientific rigor, responsible patient counseling, and comparability across studies. We believe such clarification will further strengthen the impact of the RESTORES framework and contribute meaningfully to the maturation of this promising therapeutic domain.
