If the cortisone shots that used to calm your knee have stopped doing much, take a breath: this is one of the most common and most misunderstood moments in knee arthritis. Fading relief usually doesn't mean your knee is beyond help, and it definitely doesn't automatically mean surgery is next. It means it's time to reassess — and there's more between "another shot" and "replace the joint" than most people are told.
Why injections stop working — it's by design
A steroid (cortisone) injection was never built to be permanent. It's an anti-inflammatory that quiets a flare, and the relief is meant to be temporary. Pooled evidence puts the typical benefit in the range of a few months — meta-analysis and Cochrane reviews describe meaningful symptom improvement over roughly the first weeks to a few months, after which it tapers off (meta-analysis of corticosteroid injections for knee OA, PMC).
So when a shot "stops working," it's often doing exactly what it was always going to do. On top of that, as arthritis advances, each injection may deliver a little less than the last. Fading relief is a signal about the tool, not a verdict on you.
The case for not just re-injecting forever
The natural instinct is to ask for another shot. Sometimes that's reasonable. But quietly stacking up steroid injections in a knee that no longer responds well isn't a good long-term plan, and the evidence gives a specific reason for caution.
In a randomized controlled trial, patients who received a knee steroid injection every three months for two years showed greater cartilage loss on MRI than those who got a placebo injection — with no better pain relief to show for it (Cochrane summary in Clinical Orthopaedics and Related Research). The effect on cartilage was small, but the direction is the wrong one. There's also large-scale data examining steroid injections in relation to later knee replacement outcomes, underscoring that these shots are a symptom tool, not a disease fix (national cohort study of steroid injections and subsequent knee replacement, PMC).
The honest read: an occasional injection is fine, but repeatedly re-injecting a knee that has stopped responding trades diminishing relief for a small structural downside. That's a reason to look at what else is available.
What "injections stopped working" does and doesn't mean
Here's the reframe that helps most people. Failed injections mean the simplest symptom tool has been exhausted — not that you've reached the end of the ladder. Before anyone jumps to surgery, two things are worth confirming:
- Have the proven basics really been finished? Structured strengthening of the muscles that support the knee and weight management remain the most consistently effective non-surgical treatments at every stage, and many people were offered injections before completing this work.
- What's the character of the pain? A deep, mechanical ache behaves differently from a burning, nerve-type pain — and that distinction points toward different next options.
What it should not mean is an automatic slide toward a total knee replacement, which is major, irreversible surgery. There's meaningful middle ground now.
The next options — and how they differ from a shot
When injections have run their course and your knee still limits your life, two outpatient, minimally invasive procedures sit between injections and replacement. Both are done under light sedation with same-day discharge — no open incision, no general anesthesia — and each attacks the pain in a fundamentally different way than a steroid does.
Genicular artery embolization (GAE) goes after a root driver rather than bathing the joint in medication that wears off. An arthritic knee grows abnormal, overactive small blood vessels that feed its inflamed lining; GAE reduces that excess blood flow through a pinhole catheter, calming the inflammation at its supply. That's a different mechanism than a shot — and part of why its benefit can last far longer. The evidence is emerging but encouraging: the Society of Interventional Radiology position statement supports GAE for symptomatic knee osteoarthritis in people who've failed conservative care and want to delay or avoid replacement, sham-controlled trials show meaningful short-term relief with an excellent safety profile, and some data show benefit sustained to at least two years.
Peripheral nerve stimulation (PNS) works even further downstream — it interrupts the pain signal itself before it reaches the brain. A thin lead is placed near the genicular nerves; you trial the relief in your normal life for several days, and only proceed further if it works. If it doesn't help, the lead comes out — nothing is made permanent. Early studies of stimulating the genicular and saphenous nerves report significant relief for people with severe knee pain, including some who hadn't responded to other treatments (peripheral stimulation of the genicular and saphenous nerves for chronic knee pain, PMC). PNS is often the right tool when the pain has a burning, nerve-related quality — the kind an anti-inflammatory shot was never well suited to touch.
Fading injections aren't a dead end. They're a prompt to look at options that address the pain more durably while keeping a future knee replacement fully available. If you want a plain-language read on which path fits your knee, start with our knee assessment, and explore the knee treatments we offer.
