What a cortisone injection does to a painful knee

Injected directly into the knee joint space, a corticosteroid (cortisone) solution works by suppressing local inflammation inside the joint capsule — the fibrous sleeve that encloses the joint and is often the dominant source of pain, particularly in earlier-stage osteoarthritis. Reducing that capsular inflammation quickly lowers the chemical signals driving pain and swelling.

The response is fast. Most people notice improvement within 24–72 hours, with the benefit typically peaking over the following few weeks before gradually fading. NHS guidance puts the average duration at around two months.

Cortisone does not repair cartilage, slow disease progression, or address any structural change inside the joint. No injectable corticosteroid has been shown to modify the underlying course of osteoarthritis. Its value is entirely symptomatic — bringing inflammation down so the knee becomes manageable again.

That palliative role is genuinely useful in specific situations: settling an acute flare, buying a window of comfort ahead of a planned event, or lowering pain enough to allow meaningful engagement with physiotherapy and exercise. Because the benefit is palliative rather than structural, getting the most from a cortisone injection usually depends on having a plan for what happens during that two-month window — not treating the injection itself as the endpoint.

How long the relief lasts — what the evidence says

The clearest picture of how long relief lasts comes from a 2024 systematic review and meta-analysis pooling 11 randomised controlled trials and 842 patients. Its finding is precise: intra-articular corticosteroid injections reach the minimal clinically important difference (MCID) — the threshold at which patients actually notice a meaningful change — on both VAS pain and WOMAC function scales only at short-term follow-up, defined as six weeks or less. Beyond that point, the evidence becomes less favourable. At mid-term follow-up, statistical improvements are still detectable, but they fall below the MCID threshold, meaning the average patient would not perceive the difference compared to baseline. By six months or more, corticosteroids performed no better than placebo in the same evidence base.

This is not a new discovery. Cochrane-level analysis by Arroll and Goodyear-Smith in 2004 had already identified a reliable benefit window as short as two weeks, suggesting the time-limited nature of the effect has been understood in the clinical literature for over two decades.

Knowing the shape of that curve is practically useful rather than discouraging. The evidence does not say cortisone fails — it says cortisone works within a defined window. For someone planning physiotherapy, a return to activity, or a functional goal, a realistic six-week benefit window is a planning tool: it tells you roughly how long you have to make the most of reduced pain before the injection's contribution fades. What happens in that window — whether it is used actively or not — matters considerably more than the injection alone.

Why OA stage changes what cortisone can achieve

The stage of osteoarthritis matters more than most patients realise when weighing whether an injection is likely to help.

In early OA, the joint capsule — the fibrous sleeve around the joint — tends to be the dominant pain generator. Cortisone is designed precisely for this: it bathes the capsule in anti-inflammatory medication, suppresses the chemical signals driving symptoms, and typically produces a reliable period of relief. The injection and the primary problem are well matched.

Advanced OA changes that picture. Once the cartilage has worn away sufficiently to create bone-on-bone contact, structural change rather than capsular inflammation becomes the main source of pain. Cortisone can still calm the capsule, and that provides some symptomatic relief — but the structural problem remains entirely untouched. Because the underlying cause of pain reasserts itself more quickly, the benefit window in advanced OA tends to be shorter and less predictable than in earlier disease.

Putting it plainly: the injection works on inflammation, not on bone loss. The more bone-on-bone the joint is, the narrower its window of benefit becomes.

This is why knowing your OA stage — ideally confirmed with weight-bearing X-ray or MRI — is a practical planning tool before committing to an injection. It also guides clinicians on when repeating the same injection is likely to offer diminishing returns and when a different treatment pathway deserves consideration.

Risks of repeated cortisone injections

Frequency limits exist for a reason. NHS guidance caps cortisone injections at no more than four per year, spaced at least three months apart.

The primary concern behind that cap is cartilage. Repeated exposure to high-dose corticosteroid carries a risk of accelerating cartilage degradation within the joint — a mechanism supported by the evidence base, though the exact dose-response relationship remains imprecisely characterised across studies. This is an honest reflection of what the literature shows: the risk is real enough to drive clinical guidelines, but precise harm thresholds cannot be stated with confidence. Cortisone addresses symptoms rather than the underlying structural disease — it neither slows nor modifies OA progression — which means each injection is a short-term measure, not a cumulative treatment strategy.

A secondary consideration applies to people with diabetes: intra-articular corticosteroid can produce a transient rise in blood glucose, typically resolving within a few days. For most patients this is manageable with awareness, but it should be discussed with the treating team before injection.

Taken together, these constraints shape a practical planning question. If the benefit window runs to around six weeks and the minimum gap before the next injection is three months, the interval between injections is substantially longer than the period of relief. That gap — rather than being treated as a waiting period — is the natural window for physiotherapy, supervised exercise, or assessment of whether a longer-horizon treatment would be better placed as the next step.

Cortisone as a bridge to physiotherapy and exercise

Used strategically, a cortisone injection is less a treatment destination and more an opening — a period of reduced pain in which exercise and physiotherapy become practically possible.

The periarticular muscles around the knee — quadriceps, hamstrings, and hip stabilisers — play a significant role in offloading the joint, and strengthening them produces functional benefit that outlasts any injection. The obstacle is that significant pain makes meaningful exercise difficult. By temporarily suppressing the inflammatory component, cortisone can lower that barrier enough for patients to engage properly with a supervised rehabilitation programme. The injection creates the window; the physiotherapy fills it.

Patients who rely on repeat injections purely to suppress symptoms — without layering in rehabilitation or considering longer-horizon options — are unlikely to accumulate lasting benefit. Planning the next step before the relief window closes is therefore the more useful frame than counting injections.

One question that sometimes shapes that planning is whether cortisone injections affect the eventual path to knee replacement. A large English national cohort study (CPRD GOLD, patients from 2005 to 2019) found contradictory results: instrumental variable analysis suggested that a single injection was associated with a lower risk of knee replacement at five years, while propensity score analyses pointed the opposite way. Neither signal should be read as settled. The most plausible explanation is residual confounding in the data rather than a genuine causal effect in either direction. What the same study found more clearly is that prior injection use did not worsen outcomes after knee replacement — complication rates at one year were comparable regardless of injection history.

Longer-horizon options and where cortisone sits in the pathway

Cortisone occupies a specific position in the injection menu — fast to act, short in duration — and understanding where the alternatives sit helps patients and clinicians think through sequencing rather than substitution.

PRP (platelet-rich plasma) takes longer to reach full effect, typically 8–12 weeks, which makes it poorly suited for acute flare control. Where it tends to pull ahead is at the 6–12 month mark: several trials and a 2025 systematic review suggest PRP may offer more sustained benefit over that window than corticosteroid, though no consensus on clear superiority has yet emerged.

Stromal vascular fraction (SVF), a cell-based preparation derived from adipose tissue, shows a similar temporal pattern. A 2025 meta-analysis of nine RCTs found SVF inferior to corticosteroid at three months, comparable at six months, and marginally superior at twelve — positioning it as a slower-building option for patients willing to accept a delayed-onset trade-off.

Amniotic tissue injections produced broadly similar outcomes to triamcinolone (a standard corticosteroid) at six weeks in a 2025 double-blind RCT of 81 patients with severe knee OA (KL grade 3–4). The key divergence was at one year: the amniotic tissue group maintained clinically meaningful gains across pain and function scores; the corticosteroid group did not.

Hyaluronic acid sits in a different category — primarily a viscosupplement — and is generally considered appropriate where medium-term symptom management is the goal without the cartilage-risk concerns associated with repeat steroid use.

Looking further ahead, the international MOTION study is running a 24-month RCT comparing genicular artery embolisation (GAE) against corticosteroid injection in patients across KL grades 1–4. Results are awaited; the trial's existence reflects the field's appetite for more durable minimally invasive options rather than endorsing GAE ahead of the data.

The right approach depends on OA stage, symptom pattern, and what the patient is trying to achieve — factors that a specialist assessment can map against the available pathways. A suitability check, available through this site, is a practical first step for patients who want to explore which option fits their situation.

1. [1] Amniotic Tissue Injections Are an Effective Alternative to Corticosteroid Injections for Pain Relief and Function in Patients With Severe Knee Osteoarthritis: A Double-Blind, Randomized, Prospective Study. (2025). https://doi.org/10.5435/JAAOSGlobal-D-23-00276 https://doi.org/10.5435/JAAOSGlobal-D-23-00276
2. [2] A Review of Recent Innovations in Cartilage Regeneration Strategies for the Treatment of Primary Osteoarthritis of the Knee: Intra-articular Injections. (2023). https://doi.org/10.1177/23259671231155950 https://doi.org/10.1177/23259671231155950
3. [3] Intra-articular corticosteroid injections provide a clinically relevant benefit compared to placebo only at short-term follow-up in patients with knee osteoarthritis: A systematic review and meta-analysis. (2024). https://doi.org/10.1002/ksa.12057 https://doi.org/10.1002/ksa.12057
4. [4] Efficacy and Safety of Intra-articular Platelet-Rich Plasma (PRP) Versus Corticosteroid Injections in the Treatment of Knee Osteoarthritis: A Systematic Review of Randomized Clinical Trials. (2025). https://doi.org/10.7759/cureus.80948 https://doi.org/10.7759/cureus.80948
5. [5] Intra-Articular Stromal Vascular Fraction and Mesenchymal Stem Cell Injections Show Variable Efficacy and Higher Potential Complications Compared to Corticosteroid and Hyaluronic Acid in Treatment of Knee Osteoarthritis: A Meta-Analysis of RCTs. (2025). https://doi.org/10.1016/j.arthro.2025.01.050 https://doi.org/10.1016/j.arthro.2025.01.050
6. [6] MOTION Study Protocol: Genicular Artery Embolization vs Corticosteroid Injections for Symptomatic Knee OA (RCT). (2025). https://doi.org/10.1007/s00270-025-03994-z https://doi.org/10.1007/s00270-025-03994-z