Why an Inflamed Nerve Root Burns — and What an Epidural Steroid Injection Actually Does

What this video covers

• Why a disc herniation causes burning nerve pain through chemical irritation, not just pressure
• How corticosteroid in the epidural space is thought to reduce inflammation, swelling, and ectopic nerve firing
• Why fluoroscopy and contrast are used to confirm the medication reaches the intended nerve root
• How a selective nerve root block is used to test whether one specific level is responsible
• What the evidence actually shows about how much relief to expect, for how long, and why spinal stenosis is different
• The real risks, including a temporary rise in blood sugar, a short-lived pain flare, and the rare but serious neurologic complications that the FDA-convened safeguards were written to prevent
• 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.

It is two in the morning and you are standing in your own kitchen, because lying down made it worse. The pain is not really in your back. It is a hot wire running from your buttock, past your knee, along the outside of your calf, into your big toe. Coughing sets it off. Sitting — the car, the desk, the couch — is the worst thing you can do, and walking a little actually takes the edge off. Or it is your arm, and the only position that lets you sleep is with your hand resting on top of your head. That posture has a name: the shoulder abduction relief sign. What this article explains is why that pain behaves the way it does, what an epidural steroid injection is actually doing when it helps, and what the published evidence says it can and cannot deliver.

What is actually happening at the nerve root

At the center of every spinal disc sits the nucleus pulposus — a gel-like core that is avascular and walled off, tissue your immune system has very little ongoing acquaintance with. That unfamiliarity is thought to be part of why its escape provokes such a strong reaction.

When the outer ring of the disc, the annulus, tears and nuclear material extrudes into the epidural space, two separate injuries begin.

The first is mechanical. Displaced material presses on the nerve root, and pressure tends to produce the symptoms of a conduction block: sensory loss mapped to one dermatome, a reflex that drops out, a weak muscle.

The second is chemical, and the evidence suggests it is the chemistry, more than the pressure, that accounts for the burning. Escaped nucleus pulposus exposes the nerve root — and critically the dorsal root ganglion, the cluster of sensory cell bodies that sits just outside the spinal canal — to phospholipase A2, tumor necrosis factor alpha, and interleukins. That is chemical radiculitis.

The ganglion is the vulnerability. It is less protected than the rest of the root. Inflamed, it discharges ectopically: signal with no stimulus at all.

This is also why some people have dramatic herniations on film and no symptoms whatsoever. The image shows geometry. It does not show chemistry.

Why that produces the pain you actually feel

Once you know the ganglion can fire on its own, the strangest feature of radicular pain stops being strange.

Your leg screams while you are lying perfectly still. There is nothing pressing harder at 2 a.m. than there was at noon. An inflamed ganglion does not need a mechanical trigger to fire, so the pain does not require you to be doing anything.

The rest of the pattern is recognizable even where the mechanism is not the point. Coughing sets it off. Sitting — the car, the desk, the couch — is often the worst thing you can do, while standing up and walking a little takes the edge off. In the arm, the only position that lets you sleep is a hand resting on top of the head: the shoulder abduction relief sign. None of these positions prove anything on their own, and the reviews cited below do not test them. What they do is point at a nerve root rather than a muscle, which is where the evaluation has to start.

Numbness and weakness feel like a different problem than the burning. In a sense they are. Numbness, a lost reflex, and weakness track the mechanical injury. The burning tracks the chemical one. That split matters, because the mechanical injury is the part a scan can show and the chemical one is the part it cannot.

What the standard approach often misses

Two failures are common, and neither is subtle.

The series of three, booked before the diagnosis. Patients are handed three epidural steroid injections, scheduled six weeks apart and booked in advance, before anyone has decided which nerve root is generating the pain. The number three is not a biological finding. It is a scheduling convention. A treatment plan that exists before a diagnosis does is not a treatment plan.

Numbers borrowed from a different condition. If your imaging shows lumbar spinal stenosis — and your legs go heavy and dead after two blocks of walking, then recover the moment you sit or lean on a cart — you may be quoted success figures that were generated in patients with disc herniation and true radicular pain. The 2025 American Academy of Neurology systematic review found that in lumbar spinal stenosis, epidural steroid injections did not significantly reduce short-term pain (success rate difference −3.5%, 95% CI −12.6 to 5.6). There was possible improvement in disability (short-term success rate difference −26.2%, 95% CI −52.4 to 3.6 — an interval that still includes no benefit at all), which is a genuinely different promise from pain relief. For cervical spinal stenosis, the same review found the evidence insufficient to determine effectiveness. An earlier meta-analysis in Annals of Internal Medicine likewise found no clear effect of epidural corticosteroid injections for spinal stenosis.

There is also a technical failure worth naming: an injection performed without live imaging, or with imaging but no contrast, cannot be proven to have reached the space it was aimed at. That is precisely why the FDA-convened multidisciplinary safety group placed radiographic guidance near the center of its recommendations.

What an epidural steroid injection can and cannot do

Where the medication goes. A transforaminal injection places medication in the ventral epidural space and directly onto the sleeve of the inflamed root — the location where the chemistry is happening. An interlaminar approach covers a broader posterior field. A caudal approach suits altered postsurgical anatomy. The full procedural detail, including preparation and recovery, is on the epidural steroid injection treatment page.

What the steroid does once it is there is less settled than you may have been told. Corticosteroid is thought to damp the inflammatory cascade at the root — but that mechanism is inferred from animal work, not demonstrated in patients. We know the injection helps some people for a while. We do not fully know why. Any clinician who tells you otherwise is overstating the science.

The numbers, plainly. Armon and colleagues, writing in Neurology in 2025 for the American Academy of Neurology, reviewed ninety randomized trials. For cervical and lumbar radicular pain, epidural steroid injections probably reduce short-term pain: a success rate difference of 24.0% (95% CI 12.6% to 34.9%), number needed to treat 4. The authors were explicit that the heterogeneity of outcome measures precluded presenting an integrated effect size — so precision beyond that is not available. For longer-term pain relief, the same review found the evidence insufficient. Chou and colleagues, in Annals of Internal Medicine in 2015, found immediate improvement averaging 7.55 points on a 100-point scale (95% CI 3.74 to 11.4) — small, and not sustained.

“Short term” means three months or less in these reviews. Repeat injections are limited by cumulative steroid exposure. What an injection buys is a window — the interval in which rehabilitation becomes possible — not a cure.

What you are consenting to. Expect a rise in blood sugar for several days, flushing, insomnia, and possibly a pain flare in the first 48 hours. The needle can breach the dura and leave a positional headache lasting days, and any needle crossing skin carries a real, if small, risk of infection. Catastrophic neurologic injury, including stroke and spinal cord injury, is rare — but it has occurred. Rare is not zero. That is why Rathmell and colleagues published seventeen clinical considerations in Anesthesiology in 2015 for preventing neurologic complications, why particulate steroid is not used for cervical transforaminal injections, and why a nonparticulate steroid such as dexamethasone is recommended for an initial lumbar transforaminal injection.

What this means for your evaluation

First, prove it is a nerve root at all. In the leg, radicular pain is imitated by hip osteoarthritis (groin pain, worse on internal rotation); by greater trochanteric pain syndrome (lying on that side becomes impossible); by facet and sacroiliac referral, which usually stops above the knee; by diabetic lumbosacral radiculoplexus neuropathy; and by polyneuropathy, which is symmetric and stocking-shaped rather than dermatomal. In the arm: rotator cuff disease, cubital tunnel, carpal tunnel, and thoracic outlet compression.

Resolution requires concordance. Pattern, examination, sensory and reflex findings, and imaging should all point at one level, with electrodiagnostic testing when the picture stays muddy.

When two adjacent levels are both plausible, a selective nerve root block becomes a test, not a treatment. Local anesthetic is placed on one root, and your response over the next few hours is the readout. Its honest limit: anesthetic spreads and does not respect the level it was aimed at, and the evidence supporting the test is modest. It is used to break a tie, not to manufacture certainty. For the same reason, a diagnostic block should not be performed under deep sedation — you need to be awake enough to report what you feel.

Every injection under live fluoroscopy with contrast. Contrast should outline the root before any steroid is delivered. If contrast enters a vessel, the needle moves. Digital subtraction is used to check the vascular field.

Metabolic status is part of the plan, not a substitute for it. If you are insulin resistant, the steroid will push your glucose harder and push it longer, so that is managed as part of the procedure. Nothing on a nutrition plan will decompress a root or clear an extrusion. Omega-3-derived resolution mediators and near-infrared photobiomodulation remain adjunctive and still emerging — they are not why a nerve root calms down.

What a realistic course looks like. The following is a composite — a picture assembled from many patients with this condition, not one person’s chart, and not a trial result. Someone presents after eleven weeks of pain running into the sole and outer edge of the foot, a diminished ankle reflex, and an MRI showing a disc extrusion at L5-S1 contacting the traversing S1 root. Everything concords, so a transforaminal injection is performed at S1 under live fluoroscopy with contrast. The first two days are worse, not better — a flare, expected, and still frightening at three in the morning. Blood sugar runs high for four days. By day ten the pain has dropped enough to sleep flat, and that is when the real work starts: nerve mobilization, hip and trunk strengthening, and rebuilding the confidence to bend. Around the third month some leg pain creeps back, though not to where it began, and a second injection is performed. Function holds better than pain does. That pattern is common; it is not what every patient experiences. Individual results vary.

Ask which root. Ask to see the contrast. Insist on a diagnosis before a series.

Frequently asked questions

Key takeaways

• Radicular pain has two injuries in it: mechanical pressure, which produces numbness, reflex loss, and weakness, and chemical inflammation at the dorsal root ganglion, which produces the burning.
• For cervical and lumbar radicular pain, a 2025 review of ninety randomized trials found epidural steroid injections probably reduce short-term pain — a 24.0% success rate difference (95% CI 12.6% to 34.9%), NNT 4 — with outcome measures too heterogeneous to present an integrated effect size, and insufficient evidence for longer-term pain relief.
• Those figures do not transfer to spinal stenosis: in lumbar stenosis the same review found no significant short-term pain benefit, only possible disability improvement, and in cervical stenosis the evidence was insufficient.
• The injection buys a window for rehabilitation, not a cure; “short term” means three months or less, and repeat injections are limited by cumulative steroid exposure.
• Insist on a diagnosis before a series: which root, confirmed by concordant pattern, exam, and imaging, and injected under live fluoroscopy with contrast.

Medically reviewed by 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. To be evaluated at Padda Institute Center for Interventional Pain Management, call (314) 481-5000 or text (314) 886-5902.

References

1. Armon C, Narayanaswami P, Potrebic S, Gronseth G, Bačkonja MM, et al. Epidural Steroids for Cervical and Lumbar Radicular Pain and Spinal Stenosis Systematic Review Summary: Report of the AAN Guidelines Subcommittee. Neurology. 2025;104(5):e213361. PMID 399380001012120000000000213361. PubMed
2. Chou R, Hashimoto R, Friedly J, Fu R, Bougatsos C, Dana T, Sullivan SD, Jarvik J. Epidural Corticosteroid Injections for Radiculopathy and Spinal Stenosis: A Systematic Review and Meta-analysis. Ann Intern Med. 2015;163(5):373-381. PMID 26302454107326150934. PubMed
3. Rathmell JP, Benzon HT, Dreyfuss P, Huntoon M, Wallace M, et al. Safeguards to prevent neurologic complications after epidural steroid injections: consensus opinions from a multidisciplinary working group and national organizations. Anesthesiology. 2015;122(5):974-984. PMID 256684111010970000000000000614. PubMed

Dr. Gurpreet Singh Padda, MD, MBA, MHP