How Epidural Steroid Injections Treat Sciatica

Epidural steroid injections for sciatica provide real but modest short-term relief of leg pain and disability, with effects that tend to fade over weeks to months. A large meta-analysis pooling results from multiple randomized trials found a statistically significant but small reduction in leg pain and disability at short-term follow-up compared to placebo, with long-term effects that were not statistically significant.¹ That gap between “statistically significant” and “life-changing” sits at the center of a procedure that remains one of the most commonly performed spinal interventions in the world, and understanding where injections genuinely help, where they fall short, and who benefits most can save you time, money, and unnecessary procedures.

Why Sciatica Is Not Just a Pinched Nerve

For decades, sciatica was explained as a purely mechanical problem: a bulging disc presses on a nerve root, and the pressure causes pain shooting down the leg. That picture is incomplete. Clinical observations have long challenged it: disc surgery does not always relieve pain, large herniations are sometimes painless, severe symptoms can occur with no visible nerve compression on imaging, and the size of a herniation correlates poorly with how bad the pain actually is.² These observations pointed researchers toward a chemical component: the inner core of the disc releases inflammatory substances when it comes in contact with nerve tissue.

The current understanding is that both factors matter and that they amplify each other. Animal research has shown that combining mechanical compression with chemical irritation produces significantly more pain sensitivity than either insult alone.³ A separate study concluded the same: the combination of a herniated disc’s physical mass and the resulting inflammation around the nerve root causes more nerve dysfunction than either factor by itself.⁴ This dual mechanism is why epidural steroids, which target inflammation, can help even when the disc is still bulging. It is also why they do not always help: if the dominant problem is mechanical compression rather than chemical irritation, dampening inflammation may not be enough.

How Well Do the Injections Actually Work

The honest answer is that the average benefit is small and short-lived. A pooled analysis of trials comparing epidural steroids to conservative treatment found that injections were better for pain relief at short- and intermediate-term follow-up, but the advantage disappeared at long-term follow-up. There were no significant differences in functional improvement at any time point.⁵ When compared head-to-head with gabapentin, a common nerve-pain medication, the injections produced similar average pain relief at one and three months. The injection group did report greater reductions in worst leg pain at one month, but by three months there were no significant differences between the two approaches.⁶

These averages, though, conceal a wide range of individual responses. Some people get dramatic relief and others get none. In one randomized trial of caudal epidural injections for sciatica, about two-thirds of the steroid group met the success threshold at four weeks, compared to only about one in six receiving placebo. But by 12 weeks, the placebo group had caught up significantly and the difference was no longer meaningful.⁷ That pattern, a clear early advantage that narrows with time, appears consistently across studies. It suggests the injection buys a window of relief during which the underlying condition may improve on its own, rather than fixing the underlying problem.

Can the Injection Prevent Surgery

This is one of the more compelling arguments for epidural steroid injections even if their direct pain relief is modest. In a retrospective case series of patients with herniated discs who had been referred for microdiscectomy, about 56% of those who received transforaminal epidural steroid injections reported enough improvement that they voluntarily declined surgery. Both single and repeat injections were effective at avoiding surgery, and the benefit was not merely delaying an inevitable operation.⁸ A large UK trial found that surgery for sciatica from disc herniation had an incremental cost-effectiveness ratio of about £38,737 per quality-adjusted life-year gained compared to transforaminal injection, making surgery relatively expensive per unit of benefit.⁹ For people hoping to avoid an operation, trying injections first is a reasonable step.

Three Ways to Deliver the Injection

Not all epidural injections are the same procedure. There are three main routes, each with different trade-offs.

Transforaminal: The needle enters through the bony opening where the nerve root exits the spine, delivering the steroid directly to the affected nerve. This is the most targeted approach and is often preferred when one specific nerve root is the problem.¹⁰
Interlaminar: The needle goes between two vertebral arches into the dorsal epidural space, spreading the medication more broadly. This can be useful when multiple spinal levels are involved or the pain source is not well localized.
Caudal: The needle enters through the sacral hiatus at the base of the spine. It is technically the easiest approach and has the lowest risk of accidental dural puncture, but the medication has to travel farther and spreads more diffusely.

Despite these differences in precision, a head-to-head comparison of transforaminal and interlaminar injections for lumbar disc disease found no significant difference in pain scores, disability scores, or functional outcomes at one, three, or six months.¹¹ Where the approaches may diverge is in spinal stenosis: one study found that transforaminal injections produced pain relief in 90% of stenosis patients at six months, compared to about 55% with caudal injections.¹² The caudal approach still has value: a pilot study of 40 patients reported an average drop in pain scores from about 8.6 to 2.9 after the procedure, with roughly 89% of patients experiencing at least 50% relief.¹³ Your doctor’s choice of approach typically depends on where the problem is, how many levels are involved, and your anatomy.

Why Image Guidance Matters

Epidural injections used to be performed by feel alone, with the doctor relying on anatomical landmarks and tactile feedback to guide the needle. That approach is still sometimes used, but research has made a strong case for using fluoroscopy (live X-ray) or ultrasound. Without imaging, the rate of incorrect needle placement can reach 50% for caudal injections, about 30% for lumbar interlaminar injections, and over 50% for cervical interlaminar injections.¹⁴ Those are startling numbers. If the needle is not in the right place, you get less benefit and potentially more risk.

A matched-pair study comparing the same patients’ outcomes with and without fluoroscopy for interlaminar injections found slightly better pain reduction and slightly fewer complications in the fluoroscopy-guided group, though the differences did not reach statistical significance in that particular study.¹⁵ Even so, with misplacement rates that high without imaging, most spine societies now recommend image guidance as standard practice, particularly for transforaminal injections where the needle passes near arteries that supply the spinal cord.

Particulate Versus Non-Particulate Steroids

The steroid injected into the epidural space comes in two broad categories. Particulate steroids (like triamcinolone or methylprednisolone) are suspensions of tiny crystite-like particles that linger at the injection site. Non-particulate steroids (like dexamethasone) dissolve in solution and are cleared more quickly. For years, the particulate formulations dominated because clinicians believed the longer local residence translated into better outcomes.

Multiple systematic reviews and meta-analyses have concluded otherwise. A 2016 meta-analysis found that particulate steroids were not demonstrably better at relieving pain compared to non-particulate alternatives.¹⁶ A separate systematic review reached the same conclusion and explicitly recommended non-particulate steroids as first-line agents, given that all reported catastrophic complications (like strokes from accidental injection into a spinal artery) had involved particulate formulations.¹⁷ A 2024 meta-analysis confirmed no significant differences in pain or disability between the two types for lumbar radicular pain.¹⁸ If you are being offered a transforaminal injection and the clinician plans to use a particulate steroid, it is worth asking why, given the safety data.

Does the Steroid Even Matter

Here is a twist that surprises many people. When researchers have compared epidural injections containing steroid plus local anesthetic to injections of local anesthetic alone, the results are often remarkably similar. A systematic review of randomized trials found the two approaches equally effective for most conditions, with the exception of disc herniation, where adding the steroid did appear to provide an edge over anesthetic alone.¹⁹ For chronic low back pain from spinal stenosis, a meta-analysis of 13 trials found that both injections with steroids and injections with local anesthetic alone provided significant pain relief and functional improvement, with no advantage from adding the steroid.²⁰ Another meta-analysis of ten trials looking at chronic low back and leg pain found the same thing.²¹

What to make of this? It is possible that the physical act of injecting fluid into the epidural space has its own therapeutic effect, perhaps by diluting inflammatory chemicals, stretching adhesions, or simply flushing the area around the nerve root. The mechanism by which steroids help epidurally remains officially unproven, and some benefit may come through pathways unrelated to inflammation.²² If you have disc herniation with radiculopathy, the steroid component likely adds something. For stenosis-related pain, the evidence for adding steroid is weaker.

Who Responds Best

Since the average effect is modest, identifying who is most likely to benefit matters a lot. Several studies have looked for predictors, and a reasonably consistent picture has emerged. People with a shorter duration of symptoms, lower-grade nerve root compression, and herniations located toward the side of the spinal canal (foraminal or extraforaminal) tend to respond better to transforaminal injections.²³

A retrospective study identified three clinical and imaging features that strongly predicted a good response at three months: leg pain following a clear nerve-root pattern (dermatomal distribution), prior responsiveness to anti-neuropathic medications, and moderate disc degeneration on MRI. The model using these three predictors correctly identified the outcome about 85% of the time.²⁴ Patients with mild central canal stenosis had nearly five times the odds of a favorable response compared to those with severe stenosis.²⁵

In contrast, psychological distress before the injection does not appear to reliably predict how well you will respond. One study using validated screening tools found that pain improvement and the likelihood of reaching a meaningful clinical threshold were largely independent of pre-treatment psychological distress levels.²⁶ This is reassuring: even if you are anxious or depressed about your pain, the injection is not less likely to work on that basis alone.

Side Effects and Risks

Epidural steroid injections are generally safe, but “generally safe” is not the same as “risk-free.” The most common side effects are transient and include a temporary spike in pain at the injection site, mild headache, and facial flushing. Infection rates for epidural and transforaminal injections range from about 1% to 2%, with serious infections occurring in roughly 0.1% of cases.²⁷

If you have diabetes, the blood sugar spike after injection is worth planning for. One study of 30 diabetic patients found an average glucose increase of about 126 mg/dL the day after injection, with levels returning to their normal range within roughly two days.²⁸ The magnitude of the spike depends on how well your diabetes is controlled. Patients with higher baseline HbA1c levels experienced bigger jumps: those with HbA1c above 7% saw glucose increases averaging about 99 mg/dL, compared to about 48 mg/dL in better-controlled patients.²⁹ In all cases studied, glucose returned to baseline within about two days. Insulin-dependent patients saw the largest spikes. If you have diabetes and are getting an injection, plan to monitor your glucose more closely for a couple of days afterward.

The injection also temporarily suppresses your body’s own cortisol production. A study using dexamethasone found that both cortisol and its controlling hormone dropped to less than 10% of baseline levels the day after injection but returned to normal within a week.³⁰ This is why most guidelines advise against stacking injections too frequently. The idea that a standard “series of three” high-dose injections can be safely repeated multiple times a year is a clinical myth that is not supported by effectiveness data or safety profiles of epidural glucocorticoids.³¹

Rare but serious complications deserve mention, especially for cervical (neck) injections. Epidural hematomas, new neurological deficits from injections into the spinal cord itself, and strokes caused by accidental injection into spinal arteries have all been documented in the cervical region. These events are often underreported.³² The risk of catastrophic vascular injury is the primary reason that non-particulate steroids and image guidance are now strongly recommended, especially for cervical and transforaminal injections.

Cost-Effectiveness

Whether epidural steroid injections are a good financial investment depends on the context. For acute sciatica (symptoms lasting less than eight weeks), a Dutch randomized trial found that adding transforaminal injection to usual care was not cost-effective from a societal perspective.³³ The reasoning is straightforward: most acute sciatica improves on its own, so the injection adds cost without adding much benefit over waiting it out. A separate study comparing injection plus medical management to medical management alone for disc-related radiculopathy found no significant difference in total costs or cost-per-quality-adjusted-life-year at three and six months, with medical management alone coming in cheaper, though the difference was not statistically significant.³⁴

Where injections may offer the best value is as a bridge to avoid surgery. Given that surgery costs substantially more than injections and carries its own risks and recovery time, using an injection to test whether your symptoms can be managed non-surgically makes economic sense for the right patient. The UK NERVES trial showed that surgery had to cost nearly £39,000 per quality-adjusted life-year gained when compared against injection, making it an expensive incremental step.³⁵ The calculus changes, though, if repeated injections are needed every few months, because costs accumulate while the underlying problem persists.

Physical Therapy and the Bigger Treatment Picture

Epidural steroid injections do not exist in a vacuum. A head-to-head study comparing epidural steroids to physical therapy for lumbar spinal stenosis found that both groups improved significantly, with no meaningful difference between them.³⁶ Physical therapy has no systemic side effects, no blood sugar spikes, and no limit on how many sessions you can have per year. The injection’s advantage is speed: when pain is severe enough to prevent you from sleeping, working, or participating in rehabilitation exercises, a short burst of relief can break the cycle and let you engage with a physical therapy program. Clinicians increasingly view the injection not as a standalone treatment but as a tool to create a window of reduced pain during which active rehabilitation can gain traction.

The evidence also hints that the first epidural injection of a series might deliver the most important information. If the initial injection produces meaningful relief, that is both a diagnostic signal (confirming which nerve root is the source) and a prognostic one (suggesting the problem is inflammation-driven and may continue to respond). If the first injection does nothing, repeating it rarely changes the picture, and it may be time to explore other options, whether that means surgery, a different medication strategy, or a more aggressive physical therapy approach.