What to Do When Epidural Steroid Injections Don’t Work

An epidural steroid injection that fails to relieve your pain is frustrating, but it is not a dead end. It usually means one of two things: either the injection targeted the right spot and the inflammation simply did not respond well enough, or the true source of your pain lies somewhere the steroid never reached. Either way, a range of next steps exists, from revisiting your diagnosis and adjusting medications to procedures like radiofrequency ablation, spinal cord stimulation, structured rehabilitation programs, and surgery. The path forward depends on figuring out why the injection fell short.

Revisiting the Diagnosis

The most important question after a failed epidural steroid injection is whether the injection was aimed at the right target. Spine pain is notoriously tricky to localize. A herniated disc pressing on a nerve root, an arthritic facet joint, a degenerated disc, and an inflamed sacroiliac joint can all produce overlapping patterns of low back and leg pain. If your injection was directed at a nerve root but the real pain generator is a facet joint, no amount of steroid in the epidural space will help.

One useful diagnostic tool is a selective nerve block, where a small amount of local anesthetic (without steroid) is injected around a single spinal nerve. If that specific block eliminates your pain temporarily, it confirms which nerve is responsible. If it does not, the pain is likely coming from somewhere else entirely.¹ This kind of targeted detective work often changes the treatment plan dramatically. A patient who was headed toward a second or third epidural might instead be redirected toward a facet joint procedure or a sacroiliac joint injection, where results could be far better.

When a Repeat Injection Makes Sense

Not every partial response means the injection failed. Some patients get meaningful but incomplete relief from a first injection, and a second round given a few weeks later can build on that partial response. A study of patients with cervical disc herniations found that those who achieved partial relief from an initial epidural and then received a short series of follow-up injections (averaging about two to three sessions at two- to three-week intervals) maintained satisfactory pain control for roughly ten months out of the following year. By contrast, patients who got no meaningful relief from the first injection needed more repeat injections, experienced shorter gaps between them, and spent less of the year in a comfortable pain range.²

The practical takeaway: if your first injection gave you some relief that wore off, a repeat injection within a few weeks has a reasonable chance of extending your comfortable window. If the first injection did essentially nothing, repeating the same procedure at the same site is less likely to help, and your doctor should be looking at alternative explanations for your pain or different treatment strategies altogether.

Adjusting Your Medications

Even when an epidural steroid injection provides some relief, residual nerve pain often lingers. Gabapentinoids, a class of medications that calm overactive nerve signaling, are commonly prescribed after injections to manage whatever pain remains. A trial comparing pregabalin and gabapentin in patients who had already undergone a transforaminal epidural injection found that both drugs reduced pain scores over time, and neither was clearly superior to the other.³ If you are already on one of these and still hurting, switching to the other is unlikely to make a big difference. But if you have not tried either, adding one to your regimen after a partial injection response can help close the gap.

Beyond gabapentinoids, your doctor might consider anti-inflammatory medications, muscle relaxants, or low-dose antidepressants that have pain-modulating effects. The goal of medication optimization is not to replace a failed injection with pills forever, but to bridge your pain control while pursuing a longer-term fix.

Facet Joint Pain and Radiofrequency Ablation

If your pain stems from the small joints at the back of the spine (facet joints) rather than from a compressed nerve root, an epidural steroid injection was never going to solve it. Facet joint pain tends to be worse with extension and twisting, and it usually stays in the back rather than shooting down the leg, though overlap exists. Diagnostic blocks of the tiny nerves that supply these joints can confirm the source.

Once facet joints are identified as the culprit, radiofrequency ablation (also called radiofrequency neurotomy) uses heat delivered through a needle to disable the small medial branch nerves that transmit pain signals from those joints. A randomized trial comparing conventional radiofrequency neurotomy with an endoscopic version found that the conventional approach produced significant pain relief at three weeks, six months, and one year. By two years, the benefit had faded, which is expected because the nerves eventually regenerate.⁴ When pain returns, the procedure can be repeated. For patients whose epidural failed specifically because their pain was facet-driven, this is often the most direct solution.

Sacroiliac Joint Interventions

The sacroiliac joint, where the base of the spine meets the pelvis, is another common source of pain that mimics lumbar disc problems. Sacroiliac joint pain often radiates into the buttock and upper thigh, and physical examination maneuvers can raise suspicion, though they are not always definitive. A diagnostic injection of anesthetic directly into the sacroiliac joint helps clarify whether it is the true pain generator.

Research on sacroiliac joint injections with both anesthetic and steroid shows an interesting pattern. Among patients who experienced complete initial pain relief from the anesthetic component, about half maintained at least 50% relief at follow-up two to four weeks later. More telling was the negative side: every single patient who failed to get initial relief from the anesthetic also failed to achieve meaningful relief at follow-up.⁵ That finding underscores an important principle in spine care: if a targeted injection does not relieve your pain even briefly, the problem likely is not where you think it is. For confirmed sacroiliac joint pain, treatment options beyond injections include radiofrequency ablation of the sacroiliac nerve supply and, in severe cases, sacroiliac joint fusion.

Multidisciplinary Pain Rehabilitation

When injections and procedures have not delivered lasting relief, one of the strongest evidence-based options is a structured multidisciplinary rehabilitation program. These programs combine physical therapy, psychological pain management (including cognitive behavioral therapy), education about pain neuroscience, and sometimes aquatic therapy or relaxation training into an intensive package, typically over several weeks.

A systematic review of trials on intensive multidisciplinary rehabilitation for chronic low back pain found strong evidence that these programs improve physical function compared with non-multidisciplinary treatments, and moderate evidence that they reduce pain compared with standard care.⁶ A controlled trial of an outpatient program found that patients improved in pain intensity, posture, and well-being both immediately after treatment and at six months, with strength and endurance gains continuing to emerge even after the program ended.⁷

One nuance worth knowing: a study of patients in a three-week program found that how helpful patients perceived each component of the program to be correlated with how much their pain actually improved.⁸ That does not mean the treatments only work through placebo. It suggests that engagement and buy-in matter. Walking into a rehabilitation program expecting it to fail makes the outcome worse. This is one of those areas where the evidence really is strong, but the format (group exercise, pain education, psychological therapy) can feel like a letdown after hoping a single injection would fix everything. Patients who commit to the process tend to get real, measurable benefits.

How Psychological Factors Shape Your Outcomes

This is not about pain being “in your head.” It is about the well-documented reality that the brain’s pain-processing systems are influenced by mood, catastrophizing, fear-avoidance, and resilience. A prospective study found that patients with higher levels of negative affect before an epidural steroid injection reported roughly one point more pain on a standard ten-point scale at twelve weeks than those with lower negative affect, even after controlling for baseline pain severity and demographics. On the other side, patients who scored higher on cognitive resilience reported about one point less pain at six months post-injection.⁹

A one-point difference on a pain scale might not sound like much, but across real life it can be the difference between coping and not coping. If your injections have repeatedly underperformed, addressing psychological contributors through cognitive behavioral therapy, mindfulness-based stress reduction, or working with a pain psychologist is not a consolation prize. It can genuinely change your pain experience at a neurological level.

Spinal Cord Stimulation

For patients with persistent nerve-related leg pain that has not responded to injections, medications, or rehabilitation, spinal cord stimulation is a well-established option. The device delivers mild electrical pulses to the spinal cord through implanted leads, interfering with pain signals before they reach the brain. Before permanent implantation, patients undergo a trial period with temporary leads to see whether the stimulation provides meaningful relief.

The technology has advanced considerably, with newer devices offering different waveform patterns that work better for different pain types. While much of the published literature involves patients with failed back surgery syndrome, spinal cord stimulation is also used for people who never had surgery but whose radicular pain proved refractory to conservative measures. One documented case involved a patient with severe radiculopathy that left her bedridden and unresponsive to escalating opioids and caudal epidural steroids. After a spinal cord stimulation trial, she experienced greater than 80% pain relief, her opioid requirement dropped substantially, and she went from being bedridden to walking over 500 feet without pain.¹⁰ That is a single case, and results vary widely, but it illustrates the dramatic potential for the right candidate. Spinal cord stimulation works best when the pain is neuropathic (nerve-generated) and well-localized, and when the trial produces clear relief.

When Surgery Becomes the Right Call

Surgery is not a last resort in the way people often think of it. For certain conditions, particularly a clearly herniated disc compressing a specific nerve root, surgery has some of the best outcomes in all of spine care. A lumbar microdiscectomy, where the small fragment of disc pressing on the nerve is removed through a minimally invasive approach, is highly effective even in patients who previously received epidural steroid injections without lasting benefit.¹¹ The fact that an injection failed does not predict surgical failure. They work through entirely different mechanisms.

For patients with spinal stenosis (narrowing of the spinal canal) causing leg symptoms, a laminotomy or laminectomy to decompress the nerves can provide substantial relief. Newer combinations of techniques are also being explored. A case report described a patient with both chronic low back pain and radiculopathy who underwent basivertebral nerve ablation combined with a lumbar laminotomy, a combination not previously reported. He experienced progressive improvement in pain and mobility over the following two years.¹² Basivertebral nerve ablation targets pain originating from the vertebral body itself, a source that epidural steroids do not address at all.

The decision to pursue surgery typically depends on clear imaging findings that match your symptoms, a failure of conservative treatments over a reasonable period (usually at least six to twelve weeks), and the absence of red flags that might indicate a more urgent problem. Progressive weakness, loss of bowel or bladder control, or rapidly worsening neurological symptoms warrant urgent surgical evaluation regardless of how many injections you have or have not tried.

Platelet-Rich Plasma and Regenerative Approaches

Platelet-rich plasma (PRP) injections have generated considerable interest as an alternative to steroids, particularly for disc-related pain. PRP is made from your own blood, concentrated to boost the growth factors that platelets contain, and then injected into the painful area. The idea is to promote healing rather than simply dampening inflammation.

A prospective trial of intradiscal PRP for disc-related low back pain found that higher platelet concentrations in the prepared sample correlated with better outcomes, supporting the plausibility of the biological mechanism.¹³ Another study of PRP injected into discs with specific bone-marrow changes (Modic changes) on MRI reported that average pain scores dropped from 7 out of 10 to 3 out of 10 at twelve months, with improvements in disability scores and even changes visible on follow-up MRI.¹⁴

The honest picture, though, is that the evidence base for spinal PRP is still thin. Most of the robust PRP research has focused on knee arthritis and tendon conditions, with spinal applications having far less data behind them.¹⁵ PRP for spine pain is promising but not yet proven to the standard that would make it a routine recommendation. It is also rarely covered by insurance, which brings up a broader issue.

Insurance Barriers to Next-Step Treatments

One of the most maddening realities patients face when epidural steroids fail is that many effective next-step treatments are increasingly difficult to get covered. Despite the growth of interventional pain techniques over the past two decades, insurers have been narrowing their willingness to reimburse procedures that were previously covered. This creates a perverse situation where clinicians may be pushed toward prescribing medications, including opioids, not because the evidence favors them over procedures, but because the insurer will pay for pills and not for the intervention that might actually resolve the problem.¹⁶

If you are facing a coverage denial for a recommended procedure, a few practical steps can help. Ask your pain specialist to submit a peer-to-peer review, where your doctor speaks directly with the insurer’s medical reviewer to argue the clinical case. Request the specific denial criteria in writing so you know exactly what standard the insurer is applying. Appeals succeed more often than patients expect, especially when supported by documentation of failed conservative treatments and clear diagnostic evidence. Patient advocacy organizations focused on chronic pain can sometimes provide guidance on navigating these disputes.

Alternative Injectates Beyond Steroids

Some patients cannot receive corticosteroid injections due to uncontrolled diabetes, osteoporosis risk, or side effects from repeated steroid exposure. For these individuals, alternative substances injected into the epidural space are being explored. A small case series evaluated a plant-based injectable anti-inflammatory preparation in five patients with chronic low back pain who could not use corticosteroids. All five reported pain reduction averaging about 50% and improvements in function, with no side effects.¹⁷ Five patients is far too few to draw firm conclusions, but it reflects an active area of investigation. Other clinicians are exploring epidural injections with local anesthetic alone, etanercept (a biologic anti-inflammatory), or combinations aimed at reducing nerve inflammation through non-steroidal pathways.

These alternatives are most relevant for patients who responded to the procedural approach of an epidural injection (the medication reached the right spot and provided brief relief from the anesthetic) but whose pain returned quickly once the steroid effect was expected to take over. In that scenario, the delivery route works, but the drug does not, which makes trying a different drug through the same route a logical next step.