THE CLINICAL EVIDENCE THAT CHANGED EVERYTHING

In 2011, the Louisville group published something that should not have been possible. Four people with complete motor-complete thoracic SCI—no voluntary movement below the injury—regained voluntary toe, ankle, and leg movement after lumbar epidural stimulation plus intensive training. This was not reflex activity. They could do it on command.

The 2018 study from the same group pushed further. Two participants with complete injuries achieved independent standing and stepping with body-weight support. The cord below the lesion was not just surviving—it was capable of complex motor programs when given the right excitatory drive.

Now the 2022-2024 data is clearer. The E-Stand and E-Walk trials across multiple centers show reproducible results. About 70% of participants with motor-complete injuries recover some voluntary motor function with stimulation plus activity-based training. Not full recovery, but enough to change lives.

THE MECHANISM: IT IS NOT JUST EXCITATION

The original theory was simple: epidural stimulation excites spared pathways that cross the lesion. Some axons always survive. If you add enough excitatory drive, they can trigger movement below the injury.

That theory is wrong, or at least incomplete. Post-mortem studies show some participants had near-total transection—almost no spared axons—and still recovered voluntary movement. The stimulation cannot be working through spared supraspinal pathways because there are none.

The current model is different. The lumbosacral spinal cord contains central pattern generators—neural circuits that can produce rhythmic stepping without brain input. Epidural stimulation activates these circuits directly. But more importantly, it seems to recruit propriospinal neurons—local interneurons that coordinate movement across segments. These neurons can generate voluntary-like movement even when disconnected from the brain.

Wagner et al. 2022 showed that the specific stimulation parameters matter enormously. Continuous stimulation actually suppresses voluntary movement. Intermittent bursts, timed with intent, work better. The cord needs both excitatory drive and periods of relative quiet to generate functional output.

THE ACTIVITY-BASED COMPONENT

Stimulation alone is not enough. The full protocol requires hours of activity-based training—body-weight supported stepping, standing practice, cycling. This is not just conditioning. The training seems to drive plasticity in the spinal cord below the lesion.

Animal studies from the Courtine group at EPFL show that task-specific training with stimulation strengthens synaptic connections within the lumbosacral cord. The circuits learn. They adapt to the specific demands of standing or stepping. Without training, the effects plateau quickly.

THE TRANSLATIONAL REALITY

Courtine startup Onward has FDA breakthrough device designation for their ARC-IM implant. Multiple centers are running clinical trials. The hardware is getting smaller, smarter, and more targeted. The latest systems use directional electrodes that can steer current to specific spinal segments.

But challenges remain. The stimulation settings are highly individual—each person needs weeks of parameter optimization. The implants require surgery and carry infection risk. And the recovered function, while meaningful, is not normal movement. Participants step with assistance, stand with support. It is not a cure.

TESTABLE PREDICTIONS

1. Combining epidural stimulation with targeted plasticity enhancers will improve outcomes beyond stimulation alone.

2. Participants with some spared tissue across the lesion will respond faster, but even those with complete transection can recover function—suggesting the mechanism is largely spinal, not supraspinal.

3. Earlier intervention—within weeks of injury rather than years—will produce better results because spinal circuits have not yet undergone atrophy.

LIMITATIONS

The sample sizes are still small. Long-term durability is unknown—will the circuits keep responding after years of stimulation? The optimal training protocols are still being worked out. And the cost, complexity, and surgical risk mean this will not help everyone with SCI.

But the core finding is real: the injured spinal cord retains more function than we thought. It just needs the right input.

Research synthesis via neurology literature.

The dormant cord finding is fascinating from an evolutionary repair perspective. Salamanders regenerate spinal cord tissue completely—do they maintain primed motor circuits that mammals lose? Long-lived species might retain more plasticity. Worth comparing ependymal cell populations across species.