The Rash Healed and the Burning Did Not: Inside Post-Herpetic Neuralgia

What this video covers

• Why shingles damages the dorsal root ganglion and how that becomes post-herpetic neuralgia
• How ectopic firing and sodium-channel changes are thought to drive burning, shock-like nerve pain
• What happens during the procedure, step by step, under fluoroscopy or ultrasound with contrast
• What the randomized evidence does and does not show about preventing post-herpetic neuralgia
• How long relief typically lasts, why the evidence sits in the first weeks after the rash, and why repeat blocks for long-standing post-herpetic neuralgia are unsupported
• Why injected corticosteroid is used off-label here, and what safety steps lower the risk of neurologic injury
• 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.

The rash healed months ago, and everyone keeps pointing that out. Your skin looks almost normal now, and yet you sleep in a recliner because the weight of a bedsheet across that one band of skin is intolerable. A shirt seam. The draft from a car window. And underneath the burning, jolts — electric and unannounced — while you sit perfectly still. That is not skin that failed to heal. It is a nerve cell body that a virus attacked and that did not fully survive. This article explains where the damage actually is, why it produces exactly the symptoms you have, what a selective nerve root block can and cannot do about it, and what the randomized evidence honestly shows.

The damage is not where the rash was

The varicella-zoster virus you met as a child never left your body. It withdrew into the dorsal root ganglia — clusters of sensory nerve cell bodies sitting in a bony opening beside your spinal cord — and stayed there, silent, for decades.

When cell-mediated immunity thins with age or illness, the virus reactivates inside one ganglion and travels down that ganglion’s axons to the skin. That is why the rash was a band: a band is the skin territory of one ganglion. The eruption is the visible end of the journey. The lesion is at the beginning of it, upstream, next to your spinal cord.

What the virus does there is not gentle. Human autopsy studies show hemorrhagic ganglionitis — bleeding inside the ganglion — and loss of the large myelinated fibers leaving it. Myelin is the insulating sheath that makes a nerve fiber fast; the large, well-insulated fibers are the ones that normally carry ordinary touch and vibration. Those are the fibers that drop out. (Established: human autopsy pathology.)

Animal nerve-injury models then show something further: the surviving pain-sensing fibers, the nociceptors, upregulate sodium channels and begin firing ectopically — discharging on their own with no stimulus reaching them at all. That ectopic firing is inferred to drive sensitization in the dorsal horn of the spinal cord. Be precise about the status of that step: it is animal work, and the inference has not been demonstrated in humans. (Investigational / inferred mechanism.)

Why a bedsheet feels like an injury

Two of your symptoms come directly out of that picture.

The spontaneous burning is damaged neurons firing on their own. Nothing is touching you. The signal is being generated inside a nerve that is no longer able to stay quiet. This is why it does not respect what you are doing, why it arrives while you sit still, and why it is worst at the times of day when there is least to distract you from it.

The allodynia — the light touch of fabric registering as injury — is the harder one to explain, and the honest answer is that we have a leading explanation rather than a proven one. The leading explanation is that losing those large myelinated fibers degrades the circuitry that separates touch from damage. When the fast touch channel is stripped out, the nervous system loses part of what it uses to tell the two apart, and ordinary contact gets read as harm.

That is a leading explanation, not a proven mechanism, and not the only one. Some patients present with the opposite picture — a patch of skin that is numb and painful at the same time — which any complete account has to accommodate. (Contested: mechanism of post-herpetic allodynia is not settled.)

What this does mean is that your symptom is not exaggeration, and it is not psychological. It is the predictable output of a specific, documented lesion in a specific place.

A history of shingles is a clue, not a conclusion

This is where evaluation earns its keep, because several conditions produce a band of pain and only one of them is post-herpetic neuralgia.

• Radicular pain from a disc is also dermatomal — it follows a single nerve root’s skin territory — but it is typically provoked by position and loading, and it often brings a diminished reflex.
• Intercostal neuralgia after a rib fracture follows the same band from an entirely different cause.
• Length-dependent polyneuropathy, the kind that comes from diabetes or chemotherapy, is symmetric and stocking-shaped. It does not respect a single band, and that distinction matters enormously for what follows.
• Zoster reactivation without any rash is a recognized entity, considered when the pain fits but no skin history exists.

Sorting this out is largely tactile rather than technological. A soft brush and a cold metal probe, drawn across the skin, map whether the abnormal zone honors a dermatomal border — the boundary of one nerve root’s territory — or ignores it. Strength and reflexes are tested. Imaging is ordered when the picture stops behaving like one dermatome, not reflexively at the start.

Two failures that put people in the recliner

Standard care around shingles is often good care. Antivirals started inside seventy-two hours are the right call, and most people who get shingles do not end up with lasting pain. But two specific failures recur, and they run in opposite directions.

The first failure is time. The randomized evidence for nerve blocks in shingles lives in the first weeks after the rash. Patients are told, correctly, that this fades in most people — and then nobody sets a date to see them again. Pain severe enough to predict post-herpetic neuralgia gets treated as something to outlast. By the time a referral is written, the window the trials studied has closed. Waiting is not a neutral act.

The second failure runs the other way. Someone with burning in both feet from diabetes or chemotherapy is offered a single nerve root block. A block treats one root, one side, one strip of skin. A polyneuropathy is failing every long axon in both legs at once. Offered without an anatomic reason, that is not treatment — that is scheduling.

Symptom-control medication has a genuine role here, and so does desensitization work. Neither one repairs a damaged ganglion, and neither one answers the question of which structure is generating your pain. Do not start, stop, or change any medication without consulting your physician.

What a selective nerve root block actually is

The target is the nerve root as it emerges from the foramen, the bony opening in the spine — immediately adjacent to the ganglion the virus damaged.

The distinction that gets blurred in conversation matters. A selective nerve root block places a small volume of medication outside the foramen, on the root itself. A transforaminal injection drives medication through the foramen into the epidural space, where it spreads above and below. Different spread, different risk profile, different literature. What is described here is the extraforaminal, selective version — and in a moment you will read that no randomized trial studied it.

Two imaging tools do two different jobs. In the neck, ultrasound shows soft tissue that fluoroscopy cannot: the anterior and posterior tubercles of the transverse process, and the root sitting in the trough between them. Color Doppler runs the intended needle path before anything moves, because the vertebral artery lives in that neighborhood. But be exact about what Doppler does. It helps avoid a vessel. It cannot tell anyone that a needle tip has entered one. Only contrast, injected under live fluoroscopy, does that — and the cervical level carries the highest rate of unrecognized intravascular injection anywhere in the spine.

So ultrasound maps the anatomy and fluoroscopy confirms the injectate, at every level, with contrast outlining the root sleeve before a single drop of nonparticulate steroid goes in. Rathmell and colleagues published a multidisciplinary consensus in Anesthesiology in 2015, developed with the FDA Safe Use Initiative, precisely because rare but catastrophic neurologic injury — including stroke and spinal cord injury — has occurred with these injections. Nonparticulate steroid, live radiographic guidance, and contrast confirmation are the practices that consensus identifies to reduce that risk. (Established: consensus safety document. It is cited here for risk and technique, not for benefit.)

One more thing you are owed out loud: the corticosteroid is not FDA-approved for injection around a nerve root or in the epidural space. That off-label use is accepted practice, and you deserve to hear it said rather than discover it later. The step-by-step procedural detail — positioning, imaging sequence, and what the day itself looks like — is on the selective nerve root block injection treatment page, alongside the broader range of image-guided nerve blocks used for different pain generators.

What the evidence shows — including what it does not

This is the part of the conversation that most often goes missing.

**Blocks early in shingles reduce the incidence of post-herpetic neuralgia.** Kim and colleagues, in the Korean Journal of Pain in 2017, meta-analyzed nine randomized trials. Nerve blocks applied within three weeks of rash onset shortened the duration of zoster-related pain and lowered the incidence of post-herpetic neuralgia at three, six, and twelve months. (Established, moderate-strength evidence — in acute herpes zoster.)

Now the qualifiers, all of which are load-bearing:

• Incidence, not severity. That result is about preventing the condition from establishing itself, not about reducing pain in someone who already lives with it. If you are already living with post-herpetic neuralgia, that study describes a door you have walked past.
• The benefit came from specific techniques. It came from paravertebral blocks and from continuous or repeated epidural blocks. Stellate ganglion block did not deliver it. A single epidural injection did not deliver it.
• None of those nine trials tested a fluoroscopically guided transforaminal or selective nerve root block. The technique described above was not the technique studied.

One placebo-controlled trial tested a root block directly, and hedged its own conclusion. Zheng and colleagues, in Pain Practice in 2019, randomized 140 patients with acute cervical shingles to ultrasound-guided cervical nerve root block or a matched placebo. The block reduced the thirty-day burden of illness, reduced concomitant analgesic use, and improved quality of life, with no serious adverse events. The authors’ own conclusion was that reducing post-herpetic neuralgia “might be feasible.” Feasible is their word, and it will not be upgraded here to proven. (Established for acute burden of illness; not established for post-herpetic neuralgia prevention.)

And the limits are firm. Anesthetic relief lasts hours. Steroid benefit is typically weeks to a few months — not permanent. Repeat blocks for long-standing post-herpetic neuralgia have no trial support at all. Nothing validates the use of this procedure in diabetic or chemotherapy-induced polyneuropathy.

The strongest evidence in this disease is not a needle

Here is the part that does not sell a procedure.

In the pooled ZOE-50 and ZOE-70 trial data, reported in the New England Journal of Medicine in 2016, the recombinant zoster vaccine cut post-herpetic neuralgia by roughly 89 percent in adults seventy and older. Nothing done with a needle comes near that number.

That is prevention, not treatment — it does not help pain you already have. But if you are over fifty and have not had that vaccine, that conversation matters more than this one, and it matters most of all for people who have not had shingles yet.

Who this procedure is actually for

Two groups, stated plainly.

1. Acute or subacute shingles pain confined to one dermatome, still within weeks of the rash — where the randomized data lives and the clock is still running.
2. Single-level radicular pain where symptoms, examination, and imaging all name the same root — where the block functions as a diagnostic instrument, confirming the pain generator before a larger decision is made. Note honestly that the trials cited above were run in shingles, not in radiculopathy; this diagnostic use reflects clinical practice, not those trial results.

Everything else falls outside. A whole-body neuropathy, a pain that does not honor a dermatomal border, a picture nobody has mapped — those are not candidates for this procedure, and the right answer in the room is to say so.

Where metabolic disease is degrading nerve repair, glycemic control gets addressed alongside anything procedural, on the reasoning that a nerve trying to heal in hyperglycemia is working against itself. That is a practice approach rather than a trial result for this condition. Individual results vary.

The risks, in full

This is a needle procedure next to a spinal nerve root, and it carries real risk.

The most serious, though rare, is injection into a small artery, which can cause spinal cord infarction or stroke; at cervical levels the vertebral artery is the specific concern. Other risks include nerve injury, dural puncture with a spinal headache, epidural hematoma, infection or epidural abscess, temporary weakness or numbness, local anesthetic toxicity, and pneumothorax — a collapsed lung — at cervical and thoracic levels.

The common ones are less dramatic and more likely: several days of elevated blood sugar, and a brief dip in adrenal function. A post-injection flare in the first day or two is common.

Patients taking anticoagulant or antiplatelet medication require a periprocedural plan agreed in advance. Do not start, stop, or change any medication without consulting your physician.

An honest picture of a course of treatment

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

Someone arrives five weeks after a thoracic rash healed: a burning band across the left ribs, clothing intolerable, sleep only upright. Five weeks sits past the window where the data was generated, and they were told that proceeding was reasonable, not proven. An extraforaminal block was performed under live imaging with contrast, nonparticulate steroid, anticoagulation held per plan.

The first two days were worse — a post-injection flare — and glucose climbed for a week. By the second week the allodynia eased enough to tolerate a shirt and lie flat: the first real sleep in two months. A neuropathic medication was titrated over six weeks, with the daytime dullness that costs, and desensitization work was added — deliberately brushing skin that hates being touched. A second block a month later gave less. That was a judgment call, named as one, because nothing supports repeating a block in established post-herpetic neuralgia. There was no third. A diminishing response is information.

At six months the pain was a fraction of where it started. Not absent.

And now the obligatory part: it is not possible to say how much of that was the block, how much was the medication and desensitization, and how much was six months passing. Post-herpetic neuralgia improves on its own in many people, and no honest account of a single course separates those threads. Individual results vary.

Frequently asked questions

Key takeaways

• Post-herpetic neuralgia is not unhealed skin. It is damage to the dorsal root ganglion beside the spinal cord — hemorrhagic ganglionitis and loss of large myelinated fibers on human autopsy — with the ectopic-firing step inferred from animal models rather than demonstrated in humans.
• Allodynia, where a bedsheet reads as injury, is best explained by the loss of those large touch-carrying fibers degrading the circuitry that separates touch from damage. Leading explanation, not proven mechanism.
• The randomized evidence for nerve blocks sits in the first three weeks after the rash and concerns incidence, not severity in established disease; it came from paravertebral and repeated or continuous epidural blocks, and no trial in that review tested a selective nerve root block.
• A single nerve root block treats one root on one side. It is not a treatment for symmetric diabetic or chemotherapy-related neuropathy, and repeat blocks for long-standing post-herpetic neuralgia have no trial support.
• The strongest evidence in this disease is preventive: the recombinant zoster vaccine cut post-herpetic neuralgia by roughly 89 percent in adults seventy and older. If a rash has healed and the pain has not, the clock matters — waiting is not a neutral act.

Medically reviewed by Gurpreet Singh Padda, MD — 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. To have a band of burning nerve pain properly mapped and localized before any procedure is scheduled, call (314) 481-5000 or text (314) 886-5902.

References

1. Kim HJ, Ahn HS, Lee JY, Choi SS, Cheong YS, Kwon K, Yoon SH, Leem JG. Effects of applying nerve blocks to prevent postherpetic neuralgia in patients with acute herpes zoster: a systematic review and meta-analysis. Korean J Pain. 2017 Jan;30(1):3-17. PMID 2811976710334420173013. PubMed
2. Zheng S, Li X, Yang X, He L, Xue Y, Yang Z. Ultrasound-Guided Cervical Nerve Root Block for the Treatment of Acute Cervical Herpes Zoster: A Randomized Controlled Clinical Study. Pain Pract. 2019 Jun;19(5):500-509. PMID 3073447610111112770. PubMed
3. Rathmell JP, Benzon HT, Dreyfuss P, Huntoon M, Wallace M, Baker R, et al. Safeguards to prevent neurologic complications after epidural steroid injections: consensus opinions from a multidisciplinary working group and national organizations. Anesthesiology. 2015 May;122(5):974-84. PMID 256684111010970000000000000614. PubMed

Dr. Gurpreet Singh Padda, MD, MBA, MHP