Why Your Leg Still Burns After a Back Surgery That Worked

What this video covers

• Why neuropathic pain keeps firing after the original injury has healed, and what dorsal horn sensitization actually means
• How conventional stimulation recruits large A-beta fibers in the dorsal columns, why the classic gate-control explanation still rests largely on animal models, and how 10 kHz and burst waveforms relieve pain without buzzing paresthesia through different and still-debated mechanisms
• What closed-loop stimulation adds: the device measures the evoked compound action potential and adjusts every pulse as the cord moves with posture and breathing
• What the trials actually measured, including the 83.1 percent versus 61.0 percent twelve-month rates for pain relief of 50 percent or more without increasing pain medication in the blinded Evoke trial, and why the much larger gap in the unblinded SENZA-PDN diabetic neuropathy trial has to be read more cautiously
• Where the evidence is genuinely weak, including a 2023 Cochrane review of chronic low back pain, most of whose trials were not restricted to patients who had never had surgery, and the specialty-society rebuttals to it
• The risks you would be consenting to: infection, with about 2 percent requiring device removal in the diabetic neuropathy trial, plus lead migration, dural puncture, the rare risk of epidural hematoma or nerve injury, loss of effect over time, and the real possibility of revision surgery or explant
• MEDICAL DISCLAIMER: This content is for educational purposes only and is not medical advice. It does not substitute for professional diagnosis or treatment. Always consult a licensed healthcare provider regarding your condition. Viewing this video does not establish a doctor-patient relationship.

Eleven months after the laminectomy, the MRI shows a well-decompressed canal and no recurrent disc. And a line of electric burning still runs from your buttock, down the back of your thigh, into the outside of your foot. Worse at night. Unmoved by gabapentin, or by a higher dose of it. Your leg is numb and it hurts at the same time.

The imaging has acquitted your spine, and nobody has said the sentence that actually explains you: the structure was corrected and the nerve was not. This article explains what is physically generating that line down your leg, why it produces exactly that combination of numbness and burning, what the usual next steps miss, and what an honest evaluation for neuromodulation looks like — including the trial evidence that does not make it into the device brochures.

It is a rewiring problem, not an injury

If there is nothing left in the canal to blame, what generates an electric line down your calf?

Damaged axons do not simply go quiet. They fire spontaneously — ectopic discharge, meaning a signal with no stimulus behind it. That barrage travels up into the dorsal horn, the relay station in the back of your spinal cord where incoming sensory traffic is sorted before it goes to the brain. Under a sustained barrage, that relay changes state:

• Thresholds fall. Second-order neurons that used to require a real injury to fire begin firing for much less.
• Receptive fields widen. The territory a single neuron answers for expands, so pain spreads beyond the original nerve’s strict boundary.
• Descending inhibition loses its grip. The brainstem fibers that normally turn the volume down stop doing it effectively.

You are left with a spinal cord that amplifies instead of filters. That is why a structurally successful operation can leave the pain untouched. Surgery treats anatomy. Nothing in a decompression reaches the state of the circuit above it.

Why the same leg is numb and on fire

The numbness and the burning are not competing descriptions of the same leg. They are two halves of one problem.

Numbness is loss — fibers that have stopped reporting, which is why sensory testing across that territory comes back blunted. The burning is generated somewhere else: by the spontaneous discharge described above, and by the amplification that discharge drives in the dorsal horn. One is a signal that stopped. The other is a signal being manufactured downstream of it. They can coexist in one limb because they are not the same event.

The pattern down the limb is not random either. Pain that follows a nerve’s territory — buttock to posterior thigh to lateral foot, for example — is called radicular or dermatomal, and it points to a nerve problem rather than a joint or muscle problem. Burning, electric quality, in a nerve’s territory, with sensory loss, is what “neuropathic” actually means in a clinic. It is a different thing from mechanical axial back pain, the deep ache across the low back that changes with position and load. That distinction is not academic. It decides whether the treatment discussed below has any business being offered to you at all.

What the usual next steps miss

There is a well-worn sequence for people in this gap. In practice, all three branches can fail the same person inside the same year — that is a clinical observation from this practice, not a published rate, and individual experience varies.

Re-imaged and handed back. The scan is normal, so the visit ends. But the scan was never going to show a hyperexcitable dorsal horn. A normal MRI does not mean nothing is wrong; it means the thing that is wrong is not a shape.

Operated on anyway. Fusion, then extension of the fusion, each iteration adding hardware and changing nothing about the burning. If the generator is a rewired circuit, a fourth operation cannot find it.

Escalated on medication until you are sedated and still hurting. Medications for neuropathic pain have a legitimate place. They are also symptomatic: the benefit lasts while you take the drug, and it does not change the state of the cord. Do not start, stop, or change any medication without consulting your physician.

And when neuromodulation finally comes, it too often arrives as hardware — a device offered before anyone has established that your pain is genuinely neuropathic, or explained what the evidence for it does and does not show.

What spinal cord stimulation actually does — stated with its hedges

Spinal cord stimulation places electrodes in the posterior epidural space, just outside the covering of the spinal cord, sitting over the dorsal columns — the highways of large-diameter A-beta fibers.

Conventional low-frequency stimulation recruits those large fibers first. The classical explanation is gate control: activity in large fibers closes a gate on pain transmission in the dorsal horn. Note how that has to be phrased. The gate-control model, and the dorsal-horn GABA mechanisms attached to it, derive largely from animal work and remain incompletely confirmed in humans. It is the best available account, not a proven human mechanism.

High-frequency 10 kHz and burst waveforms work below the paresthesia threshold — no buzzing sensation at all. Because they produce no paresthesia, they are not operating by classical gate control, and their mechanisms are genuinely debated. Anyone who tells you the mechanism is settled is reciting marketing.

There is also a physical problem a fixed-output device cannot solve. Your spinal cord floats in cerebrospinal fluid. It drifts toward the lead when you lie on your back, away from it when you stand, and oscillates with every breath — while current density falls steeply with distance. A fixed output therefore overshoots and undershoots all day. Closed-loop stimulation is the engineering answer: the lead records the evoked compound action potential — the actual electrical response the nerve fibers make to each pulse — and corrects the next pulse to hold activation inside a therapeutic window.

Selection is where this gets abused

Four questions come before any device conversation. If the answers have not been established, you are being sold hardware.

1. Is the pain neuropathic, or is it mechanical axial back pain? Burning, electric, in a nerve’s territory, with sensory loss, is one thing. A mechanical ache across the low back is another, and a stimulator has no business treating it.

2. Is there a structural lesion still correctable? Recurrent herniation, foraminal stenosis, pseudarthrosis (a fusion that never solidly healed), adjacent segment disease. MRI answers most of these questions. Pseudarthrosis specifically needs thin-slice CT and flexion-extension films, because a fusion that looks solid on a plain radiograph is not always solid under load. Where imaging and symptoms disagree about which level is talking, a diagnostic selective nerve root block settles which root is generating the pain.

3. Is the pain focal at all? Widespread nociplastic pain — pain arising from altered pain processing rather than from identifiable tissue or nerve damage — is not a target for a lead in the epidural space. Hardware placed into that problem mostly adds scar.

4. Psychological readiness — and the honest version of this is softer than the field usually admits. Untreated depression is the one psychological variable that predicts poor long-term outcome with any consistency across the literature. Catastrophizing and active substance use have predicted outcome in some series and not in others, and there is no good evidence that the screening visit by itself improves anyone’s result. Screening is still worth doing — not because a questionnaire forecasts your future, but because treating a collapsed mood or an untreated substance problem beforehand is better medicine than discovering it with hardware already in your back.

None of those four questions is worth asking until the ordinary work is behind you and documented: a real course of physical therapy, appropriate medication trials, targeted diagnostic and therapeutic injections. This is not a first-line treatment and it is not a shortcut past one.

The trial — and why it is not a free week

Only after that record exists does a trial make sense. Prone under live fluoroscopy, a Tuohy needle enters the epidural space by an interlaminar approach, with loss of resistance confirmed on imaging rather than by feel. Percutaneous leads are advanced to the low thoracic levels and then left externalized through the skin while you keep a log of pain and function. If both do not improve, nothing permanent is implanted and the leads come out at the bedside. The staging, the lead placement and the implant itself are described step by step on our spinal cord stimulation treatment page.

That is a genuine advantage over most surgery — you find out before you commit. It is not, however, a free week, and it should not be described as one. A trial is still an epidural procedure: infection at the exit site, dural puncture, bleeding. It is a small, bounded, reversible risk taken deliberately, instead of a large permanent one taken on faith.

What the evidence shows — including the part that is inconvenient

Three studies carry most of this conversation, and each needs its limitations attached.

Closed-loop versus open-loop (Evoke, 2020). Mekhail and colleagues published a double-blind randomized trial in Lancet Neurology comparing closed-loop against open-loop stimulation in chronic back and leg pain. At twelve months, 83.1% of closed-loop patients (49 of 59) achieved at least 50% pain relief without increasing pain medication, versus 61.0% (36 of 59) on open-loop.¹ Read that for exactly what it is. Both arms had a stimulator running. It tells you that a device measuring the nerve’s response outperforms one that only guesses at it. It does not tell you that either one beats placebo, because there was no placebo arm. The trial was funded by Saluda Medical, the manufacturer of the closed-loop system.

10 kHz stimulation in painful diabetic neuropathy (2021). Petersen and colleagues, in JAMA Neurology, reported that 79% of stimulation patients responded at three months against 5% on medical management alone, and that by six months mean pain in the stimulation arm had fallen from 7.6 to 1.7 cm on a visual analogue scale, while the medical-management arm went from 7.0 to 6.9 — it did not move.² That gap looks spectacular, and you should still discount it: the trial was open-label with no sham arm, and an unblinded comparison against pills overstates what the device itself contributes. It was funded by Nevro Corp, the manufacturer. If painful diabetic neuropathy is your situation rather than post-surgical leg pain, the workup that has to come first is laid out on our diabetes and neuropathy page.

The Cochrane review (2023). Traeger and colleagues reviewed 13 trials and 699 participants with chronic low back pain and concluded that stimulation probably does not deliver sustained clinical benefit that outweighs its costs and risks, and that the data in the review do not support its use for low back pain outside a clinical trial.³ That review is not a footnote, it does not get skipped, and — this is the part that usually gets lost — it is not about somebody else. Its included trials were not restricted to people who had never had spine surgery, so it bears directly on readers in exactly the situation this article describes. Specialty societies have publicly disputed the review. That dispute is worth knowing about and it does not cancel the finding. What the review should change is the standard of proof you demand before anything is implanted: the neuropathic-versus-axial determination made first and documented, and then a reversible trial that has to earn the implant rather than a device offered on the strength of a brochure.

The risk sheet, in full.

• Infection, sometimes requiring removal of the entire system — 2% required explant for infection in the diabetic neuropathy trial.²
• Lead migration, which can cost the therapy its coverage.
• Dural puncture, with the positional headache that can follow for days.
• Epidural hematoma, or direct injury to the spinal cord or a nerve root — rare, and the reason this is a real consent conversation and not a signature. A compressive bleed in the thoracic epidural space can leave permanent weakness, or paralysis, if it is not recognized and decompressed within hours. You are entitled to hear that sentence before a needle goes near your cord, not after.
• Revision surgery. In one low back pain trial inside the Cochrane review, 13 of 42 people (31%) receiving stimulation needed revision surgery within 24 months.³
• Loss of effect over time, with some patients electing explant — and an implanted system places lasting conditions on future MRI scanning, which has to be reviewed with you before a trial.

An honest picture of what “better” looks like

What follows is a composite — a picture assembled from many patients with this condition, not one person’s chart.

Someone two years past a lumbar decompression and fusion. Leg pain electric and unbearable, and nothing left to decompress by two surgical opinions. Examination showed a dermatomal sensory deficit with allodynia — ordinary touch read as injury. Screening flagged collapsed sleep and demoralization, and those were treated first.

The trial week is where a composite stops being tidy: dressings, no showering, wires taped along the back, and the strange work of judging a treatment while wearing it. Relief was partial — leg pain much better, the axial ache unchanged, which is the usual pattern and needs saying beforehand rather than afterward. Then programming visits across two months and a winter of physical therapy. Glycemic control and anti-inflammatory nutrition ran alongside all of it, because an exit site and a fresh implant pocket heal better in a body that is not hyperglycemic. Those measures supported the procedure; they did not replace it, and they should not be sold as if they had.

At a year, the leg pain was manageable and the opioid dose was lower. Not absent. Not cured. That is what success looks like after a decompression that corrected the anatomy and did not repair the nerve — a leg you can stand on, not a leg you stop noticing. Some patients never reach even that, and it belongs in the conversation before the trial, not after. Individual results vary.

Frequently asked questions

Key takeaways

• Persistent burning leg pain after a technically successful decompression is usually a rewiring problem: injured axons discharge spontaneously and the dorsal horn becomes hyperexcitable, so the cord amplifies instead of filtering.
• Numbness and burning in the same leg are not contradictory — one is a signal that stopped, the other is a signal manufactured downstream of it, and sensory loss plus burning in a nerve’s territory is exactly what “neuropathic” means.
• Spinal cord stimulation targets that circuit through leads over the dorsal columns; the classical gate-control explanation derives largely from animal work and is incompletely confirmed in humans, and the mechanisms of 10 kHz and burst waveforms remain debated.
• Closed-loop devices measure the evoked compound action potential and correct each pulse as the cord moves in cerebrospinal fluid — and the trial supporting them compared closed-loop against open-loop, not against placebo.
• A 2023 Cochrane review of 13 trials and 699 participants found stimulation probably does not deliver sustained benefit outweighing its costs and risks in chronic low back pain, and its trials were not restricted to surgery-naive patients, so that conclusion applies to post-surgical readers too; societies dispute it, and the honest response is stricter selection and a trial that has to earn the implant, not a device sold on a brochure.
• This is not first-line care: a documented course of conservative treatment, a neuropathic-versus-axial determination, exclusion of a still-correctable structural lesion, psychological screening, and a reversible externalized trial all come before any permanent implant.

Medically reviewed by Dr. Gurpreet Singh Padda, MD, MBA, MHP — Board Certified in Anesthesiology, Pain Medicine, Interventional Pain Management, Addiction Medicine, and Obesity Medicine. Last reviewed July 2026.

This article is educational and is not a substitute for evaluation, diagnosis, or treatment by a physician. Individual results vary. Do not start, stop, or change any medication without consulting your physician. If the surgery is behind you, the imaging is unremarkable, and the leg is still burning, call (314) 481-5000 or text (314) 886-5902 to arrange an evaluation of whether your pain is genuinely neuropathic.

References

1. Mekhail N, Levy RM, Deer TR, et al. Long-term safety and efficacy of closed-loop spinal cord stimulation to treat chronic back and leg pain (Evoke): a double-blind, randomised, controlled trial. Lancet Neurol. 2020;19(2):123-134. PMID 318707661010161474442219304144. PubMed
2. Petersen EA, Stauss TG, Scowcroft JA, et al. Effect of high-frequency (10-kHz) spinal cord stimulation in patients with painful diabetic neuropathy: a randomized clinical trial. JAMA Neurol. 2021;78(6):687-698. PMID 3381860010100120210538. PubMed
3. Traeger AC, Gilbert SE, Harris IA, Maher CG. Spinal cord stimulation for low back pain. Cochrane Database Syst Rev. 2023;3(3):CD014789. PMID 36878313101002146518580147892. PubMed

Dr. Gurpreet Singh Padda, MD, MBA, MHP