Steroid Injections for Knee Arthritis: How Long They Last and What They Cost You

A corticosteroid injection into an arthritic knee gives real but short-lived relief: the best pooled evidence shows the benefit is largest in the first 1 to 2 weeks, smaller by 6 weeks, and largely gone by 3 to 6 months. It is a tool for getting through a flare, not a way to hold off surgery for years.

I had two cortisone injections in my right knee in the years before it was replaced. The first one felt like a small miracle. The second, about a year later, gave me a few good weeks and then nothing much. At the time I assumed I had done something wrong. I had not; I had simply met the limits of what these injections can do. This article sets out what the research says about how well they work, how long they last, what they can cost you, and how to fit them sensibly into the rest of your knee care.

What a knee steroid injection is

A knee steroid injection puts a corticosteroid, a synthetic version of the body’s own anti-inflammatory hormone, directly into the knee joint. Common preparations include triamcinolone and methylprednisolone, often mixed with a local anaesthetic. It is usually done in a clinic in a few minutes, sometimes with ultrasound guidance, and you walk out afterward.

The aim is to calm the inflammation in the joint lining that flares up in osteoarthritis. That is why injections tend to work best in a knee that is swollen, warm, and angry, and less well in one that simply aches from worn cartilage.

Guidance from OARSI, the international osteoarthritis research society, places injections alongside, not ahead of, the core treatments of exercise, weight management, and education ¹. The wider picture of what helps before surgery is in alternatives before knee surgery.

How well do they work?

On average, modestly. The Cochrane review of intra-articular corticosteroid for knee osteoarthritis, by Jüni and colleagues in 2015, pooled the randomised trials and found injections reduced pain more than sham or no treatment by roughly 1 point on a 10 point scale ². Put another way, about 8 people need to be treated for one extra person to get a meaningful benefit.

That average hides a wide spread. Some women get excellent relief, some get almost none, and it is hard to predict in advance who will be in which group. The review also rated the quality of the evidence as low across all its outcomes, because the trials were inconsistent and many had design weaknesses ². The authors said it remains unclear whether the benefit in the first weeks is clinically important for most people.

How long the relief lasts

This is the question most women actually want answered, and the trials give a fairly clear shape ²:

Time after the injection	Size of the benefit compared with no injection
1 to 2 weeks	Moderate
4 to 6 weeks	Small to moderate
About 13 weeks (3 months)	Small, and may be no real effect
26 weeks (6 months)	No measurable benefit

So the honest answer is weeks, not months. If yours lasted longer, you are not imagining it; individual responses vary. But if you are planning around an injection, plan around the first month or so.

The case against frequent injections

Repeated injections did not keep working in the best trial we have, and they came with a cost. In a randomised trial published in JAMA in 2017, 140 people with knee osteoarthritis and signs of joint inflammation received either triamcinolone or saline into the knee every 12 weeks for 2 years ³. At the end, the steroid group had no significant difference in pain compared with the saline group. They did have significantly greater loss of cartilage thickness: an average of 0.21 mm against 0.10 mm.

That cartilage difference is small, and the trial was not large. But it points in the same direction as clinical caution: an occasional injection for a flare is reasonable, while a steady schedule of injections every few months is hard to justify on the evidence. Many clinicians keep to a few a year at most, spaced at least around 3 months apart, and will ask what else is being done for the knee before offering another.

Side effects to know about

Most injections pass without trouble. The things to be aware of:

• A short flare. Some women have more pain for a day or two after the injection before the benefit starts. Rest and ice usually settle it.
• Blood sugar. If you have diabetes, a steroid injection can raise your blood sugar for a few days. Tell the person injecting you, and keep a closer eye on your readings.
• Flushing. A warm, red face for a day or so is common and harmless.
• Skin changes. Occasionally the skin at the injection site thins or becomes lighter, which shows more on darker skin.
• Infection. Rare but serious. A knee that becomes increasingly hot, red, swollen, and painful in the days after an injection, especially with a fever, needs urgent medical attention, not another dose of painkillers.

For the red flags that apply to any painful knee, see knee pain: when to see a doctor.

Injections and a future knee replacement

If a knee replacement is on the horizon, tell your surgeon about every injection, with the date. Many surgeons prefer a gap of several months between a steroid injection and surgery on the same knee, because of concern that a recent injection may raise the risk of infection in the new joint. Practice varies, so the interval is your surgeon’s decision, and it is far easier to ask before booking an injection than to find out later that it has pushed back your operation date.

The opposite question comes up too: whether an injection can delay surgery. It can buy time through a bad patch, but it does not change the course of the arthritis. If you find yourself needing injections more and more often for less and less relief, that is useful information in itself, and part of what signs it’s time for a knee replacement describes.

How steroid injections compare with the others

Corticosteroid has the clearest evidence of the knee injections, and even that is for short-term relief. Hyaluronic acid has mixed and limited evidence and is conditionally recommended at best. Platelet-rich plasma (PRP) and stem-cell injections are not established and are not routinely recommended ¹, however they are marketed. None of the injections rebuilds cartilage.

Compare that with exercise. Strengthening and aerobic exercise reliably reduce pain and improve function in knee osteoarthritis, with benefits comparable in size to many drugs and with fewer harms ⁴. The best use of an injection is often to make that exercise possible: a few weeks of calmer knee in which to build the thigh muscles that will still be working long after the steroid has worn off. The guide to staying active with arthritic knees has low-impact options that suit that window.

Questions to ask before an injection

• What do you expect this injection to do for me, and for how long?
• Will you use ultrasound guidance, and does it matter for my knee?
• How many have I had, and how many would you be comfortable giving in a year?
• I have diabetes (or take blood thinners): does that change anything?
• Is there any chance I will need a knee replacement in the next several months, and if so, should I wait?
• What should I be doing for my knee during the weeks it feels better?

What I would tell my younger self

That first injection was worth having. It got me through a family wedding I would otherwise have spent sitting down. What I wish I had done differently is use the good weeks. I treated the relief as the treatment, and when it wore off I was back where I started, with weaker legs. The second time round I spent those weeks at a physiotherapist’s strengthening class, and that, more than the needle, is what made the last stretch before my operation bearable.

This article is general information, not medical advice. Whether an injection is right for your knee, which one, and how often, are decisions for you and a clinician who can examine you and knows your medical history.