If Steroid Injections Don’t Work, What’s Next?

When steroid injections stop providing relief, a range of alternatives exists, from physical therapy and weight management to newer injection-based treatments and nerve-targeting procedures. The right next step depends on the joint involved, how long you’ve been dealing with pain, and why the steroids didn’t work in the first place. Sometimes the answer is a different type of injection; sometimes it’s moving away from injections entirely and addressing the structural or neurological root of the problem. And sometimes, the steroid injection’s failure is itself useful information that reshapes the diagnosis.

Why Steroid Injections Fail

Steroid injections aren’t designed to fix anything. They suppress inflammation temporarily, and when they work, the relief typically lasts weeks to a few months. When they don’t work, it can mean one of several things: the pain isn’t primarily inflammatory, the injection didn’t reach the right spot, or the structural damage has progressed beyond what inflammation control can mask. In the case of knee osteoarthritis, there’s also growing concern that repeated steroid injections may accelerate cartilage loss over time. A systematic review and meta-analysis found that steroid use in the knee may contribute to imaging features of cartilage deterioration.¹ And basic science research has shown that at higher cumulative doses, corticosteroids are associated with direct cartilage damage.²

So a steroid injection that stops working isn’t just a nuisance. It’s a signal that something else needs attention. The question is what.

Physical Therapy Often Outperforms Injections Over Time

If you haven’t tried structured physical therapy, that’s frequently the single best next move. The evidence here is surprisingly strong. A randomized trial published in the New England Journal of Medicine compared physical therapy head-to-head with a steroid injection for knee osteoarthritis. At one year, patients who received physical therapy had better pain and function scores than those who got the injection.³ The injection group got faster initial relief, but the physical therapy group pulled ahead over the long haul.

Shoulder impingement tells a similar story. A pragmatic trial comparing steroid injection to manual physical therapy for shoulder impingement syndrome found both groups improved by about half over a year, with no significant difference between them. But here’s the kicker: patients who got the steroid injection were far more likely to need additional medical visits and follow-up injections. About 38% of the injection group needed more steroid shots, compared to 20% of the physical therapy group, and nearly a fifth of the injection patients eventually needed physical therapy anyway.⁴

Physical therapy works differently from injections. Instead of dampening inflammation, it strengthens the muscles supporting a joint, improves range of motion, and can correct movement patterns that contribute to ongoing damage. The effects tend to be cumulative and lasting rather than temporary.

Hyaluronic Acid Injections

Viscosupplementation, where hyaluronic acid is injected directly into the joint, is one of the most common pivot points when steroids fail in knee osteoarthritis. The idea is straightforward: osteoarthritis degrades the natural hyaluronic acid in your joint fluid, reducing its ability to lubricate and cushion. Injecting a synthetic or animal-derived version aims to restore some of that lost protection. Hyaluronic acid improves joint biomechanics by increasing synovial fluid viscosity, reducing mechanical friction, and enhancing shock absorption during weight-bearing activity.⁵

The clinical evidence is mixed but leans positive. An umbrella review pooling results from multiple systematic reviews and meta-analyses of randomized placebo-controlled trials concluded that most reviews, and all the high-quality ones, support the symptomatic benefit of hyaluronic acid in knee osteoarthritis.⁶ The benefit is modest, but for people who’ve stopped responding to steroids, modest and reliable beats nothing. Beyond simple lubrication, hyaluronic acid may also help through anti-inflammatory action and by supporting cartilage repair processes.⁷

Hyaluronic acid injections are typically given as a series of three to five shots spaced a week apart, or as a single injection using a higher-molecular-weight formulation. Relief tends to build gradually and can last several months. They’re not a cure, and they work best in mild to moderate arthritis rather than advanced disease. But if steroids are no longer an option or have stopped helping, viscosupplementation is a reasonable next discussion to have with your doctor.

Platelet-Rich Plasma

PRP therapy involves drawing your own blood, concentrating the platelets in a centrifuge, and injecting that concentrate into the affected area. The appeal is biological: platelets are packed with growth factors that promote tissue repair, reduce inflammation, stimulate cell growth, and support new blood vessel formation.⁸ In theory, rather than masking symptoms the way steroids do, PRP nudges the body toward healing.

Clinical results are encouraging but inconsistent. PRP appears most promising for tendon injuries, where preclinical research shows it improves tissue remodeling, collagen alignment, and biomechanical strength. For knee osteoarthritis, the evidence is building but not yet definitive. Some studies suggest PRP outperforms both hyaluronic acid and placebo for pain relief, though the variability across studies is substantial. The lack of standardization is a real problem: different clinics use different preparation methods, platelet concentrations, and injection protocols, which makes it hard to know if the PRP you’re getting matches what was tested in the study your doctor is citing.

Insurance coverage for PRP is limited in most countries, and out-of-pocket costs can run several hundred to over a thousand dollars per injection. If you’re considering it, ask what preparation method the clinic uses and whether the formulation includes or excludes white blood cells, because that distinction appears to affect outcomes.

Weight Loss Changes the Equation for Weight-Bearing Joints

For knee and hip osteoarthritis specifically, weight management is one of the most powerful interventions available, and it’s the one most likely to be underdiscussed in an orthopedic office focused on procedural options. The IDEA trial, a randomized clinical trial of overweight and obese adults with knee osteoarthritis, found that participants who lost 10% or more of their body weight improved function and reduced knee compressive force, systemic inflammation, and pain more than those who lost less.⁹

Follow-up research showed a dose-response relationship: the more weight lost, the better the outcomes across the board. The group that lost 20% or more of their body weight had roughly 25% less pain and better function than even the group that lost 10%.¹⁰ Weight loss helps through two mechanisms. The obvious one is biomechanical: less body mass means less force pounding through the joint with every step. The less obvious one is metabolic: fat tissue produces inflammatory chemicals, and weight loss lowers those levels. Research has shown that weight loss in obese adults with knee osteoarthritis reduces IL-6, a key inflammatory marker, suggesting anti-inflammatory benefits alongside the mechanical ones.¹¹

None of this means weight loss is easy or that it replaces other treatments. But if you’re carrying extra weight and your knee injections have stopped working, losing even a moderate amount can produce the kind of pain relief that another round of injections probably won’t.

Nerve-Targeted Approaches

When the pain itself becomes the primary problem, rather than something an injection into the joint can fix, treatments that target the nerves carrying pain signals are worth considering.

Radiofrequency Ablation

Radiofrequency ablation uses heat to disable the specific sensory nerves that carry pain signals from a joint. It doesn’t heal the joint, but it can dramatically reduce what you feel. A comprehensive review of radiofrequency techniques for chronic knee pain found significant benefit for both pain reduction and functional improvement lasting between three and twelve months.¹² The procedure is performed under imaging guidance, is minimally invasive, and carries relatively low risk. It’s most commonly used for knee and spine pain.

Peripheral Nerve Stimulation

Peripheral nerve stimulation takes a different approach: rather than destroying nerve tissue, it delivers electrical impulses to the nerve to change how pain signals are processed. A prospective case series on percutaneous peripheral nerve stimulation for chronic low back pain showed sustained relief lasting a year after a short-term implant was removed.¹³ Researchers believe this works partly by reversing central sensitization, a state where the nervous system amplifies pain signals even after the original tissue injury has resolved. By flooding the nervous system with non-pain signals, stimulation may help reset those overactive circuits, and the effects can persist even after the device is removed.¹⁴

Nerve Hydrodissection

For pain caused by a nerve being compressed or trapped by surrounding tissue, ultrasound-guided nerve hydrodissection is a relatively newer technique. A physician injects fluid around the nerve to physically separate it from the structures pressing on it.¹⁵ A literature review found that this approach is associated with meaningful improvements in both pain and function across a variety of entrapment neuropathies.¹⁶ It’s particularly relevant when steroid injections for conditions like carpal tunnel syndrome stop working, because it addresses the physical trapping rather than just the inflammation around it.

Prolotherapy

Prolotherapy involves injecting a concentrated sugar solution (usually dextrose) into damaged tendons or ligaments. The mild irritation is thought to kick-start a healing response. A double-blind randomized controlled trial of hypertonic dextrose injection for chronic supraspinatus tendon disease found improvements in tendon thickness and structural characteristics at both six and twelve weeks compared to baseline.¹⁷ Prolotherapy is less studied than PRP, and insurance rarely covers it, but the cost per injection is usually lower. It tends to attract interest from people dealing with tendon-related pain who’ve exhausted steroid options.

Medications That Aren’t Opioids

When injections of any kind stop being adequate, oral or topical medications can fill part of the gap. Beyond standard anti-inflammatories, duloxetine is one option that’s gained traction for chronic musculoskeletal pain. It works on the brain’s descending pain-control pathways rather than at the joint itself. Three separate randomized, double-blind, placebo-controlled trials showed that duloxetine produces a meaningful decrease in pain by about four weeks, improves physical functioning, and is safe and effective even when used alongside anti-inflammatory drugs.¹⁸ Duloxetine is especially worth considering when the pain has a centralized component, meaning the nervous system has become hypersensitive beyond what the joint damage alone would explain.

The Placebo Effect in Injection Therapies Is Enormous

This is the part of the conversation that rarely happens in a clinic room: for knee osteoarthritis, up to three-quarters of the pain reduction from injections of biological agents can be attributed to the placebo effect.¹⁹ A separate meta-analysis of placebo responses to joint injections found that roughly half of patients receiving sham injections qualified as treatment responders at one month, three months, and six months.²⁰

This doesn’t mean your pain isn’t real, or that injection therapies are useless. But it does mean that when you’re evaluating a new injection-based treatment, the bar for what counts as “working” is higher than you might think. The ritual of an injection, the expectation of relief, the attention from a provider, and the act of doing something all contribute to pain improvement independent of what’s in the syringe. Keeping this in mind is especially important when you’re being pitched expensive, unproven alternatives.

Bone Marrow Aspirate Concentrate and the Stem Cell Question

BMAC involves drawing bone marrow, concentrating it to enrich for stem cells and growth factors, and injecting it into the affected joint. The premise is regenerative: rather than just managing symptoms, BMAC aims to slow or reverse cartilage damage. Early results have been encouraging, with some studies showing outcomes superior to hyaluronic acid and PRP in the short to medium term.²¹ However, the regenerative potential of BMAC remains inadequately demonstrated, and the existing published studies use varied methodologies, patient profiles, and disease severity criteria. Larger-scale, longer-term studies are needed to establish when and for whom BMAC actually works.²²

A word of caution about the broader “stem cell” market: over 2,700 clinics in the United States offer stem cell-related injections, with over 60% focused on musculoskeletal conditions.²³ Many of these clinics market aggressively and charge thousands of dollars for treatments that lack rigorous evidence. The difference between a carefully performed BMAC injection by an experienced orthopedic specialist and what a strip-mall regenerative medicine clinic sells as “stem cell therapy” is substantial. If someone is promising joint regeneration with a money-back guarantee, be skeptical.

Rethinking the Pain Itself

When multiple treatments fail, it’s worth asking whether the problem is being framed correctly. Chronic pain, especially pain that persists beyond what imaging or physical findings can explain, often has a significant central nervous system component. The brain’s pain-processing circuits can become amplified, maintaining pain even after the original tissue injury has partially or fully healed. Pain neuroscience education, a structured approach to helping patients understand what’s happening in their nervous system, has been shown to reduce catastrophizing and fear of movement, both of which amplify pain.²⁴

This is not the same as saying your pain is “in your head.” It’s saying that your nervous system is part of the pain equation, and treating only the joint while ignoring the nervous system is like fixing a fire alarm while ignoring the short circuit that keeps tripping it. Cognitive-behavioral approaches, mindfulness-based stress reduction, and graded exercise programs all have evidence behind them for chronic musculoskeletal pain. They work best not as stand-alone treatments but alongside physical therapies and, when appropriate, medical management.

When Surgery Enters the Conversation

Surgery is generally considered after a sustained period of conservative management has failed to provide adequate relief. For some conditions this threshold is formalized: for example, thumb-base joint replacement is typically indicated after at least six months of nonsurgical treatment hasn’t controlled pain that interferes with daily activities.²⁵ For larger joints like the knee or hip, total joint replacement is one of the most successful operations in modern medicine, but it’s a permanent change and the timing matters.

Surgery isn’t always a joint replacement. Depending on your condition, options might include arthroscopic procedures, tendon repair, osteotomy to realign joint surfaces, or nerve decompression. The key question to ask any surgeon is whether the structural problem visible on imaging matches the pain you’re experiencing. If it doesn’t, a procedure to fix the structural problem may not fix the pain.

Disease-Modifying Drugs on the Horizon

One reason the current treatment landscape for conditions like osteoarthritis feels inadequate is that nothing currently available changes the underlying disease process. Everything discussed in this article either manages symptoms (pain relief, improved function) or attempts regeneration with incomplete evidence. Pharmaceutical agents that could arrest structural joint changes while also alleviating symptoms, called disease-modifying osteoarthritis drugs, do not yet exist in approved form. However, research into highly selective inhibitors that could slow cartilage breakdown is actively underway.²⁶ Until those arrive, the best strategy after steroid injections fail is usually a combination of approaches: physical therapy to strengthen and protect the joint, weight management if applicable, a targeted injection or nerve procedure for symptom control, and honest attention to how your nervous system processes pain.