Osteoarthritis Clinic · Comparing options
Three injections get asked about most for knee osteoarthritis. They work differently, over different timescales, and the guideline groups do not rate them equally. This page sets out what each one is, what the evidence supports, and where the experts disagree, so the choice is an informed one.
| Corticosteroid (Depo-Medrol) | Hyaluronic acid (Durolane) | Platelet-rich plasma (PRP) | |
|---|---|---|---|
| What it is | A long-acting steroid injected into the joint, which calms the immune response that drives swelling and pain | A gel-like injection of hyaluronic acid, meant to lubricate and cushion the joint | A sample of your own blood, spun to concentrate the platelets, then injected into the joint |
| How soon | Often within 24 to 48 hours | Builds over the following weeks rather than straight away | See the guide, in preparation |
| How long | Usually short-term, commonly fading over a few weeks to a couple of months | Can last up to about six months in those who respond, though the average benefit over a placebo injection is small | See the guide, in preparation |
| American College of Rheumatology, 2019 | Strongly recommended for knee and hip osteoarthritis | Conditionally recommended against for the knee, and strongly recommended against for the hip | Strongly recommended against for the knee and hip |
| OARSI, 2019 | Conditionally recommended for acute (1 to 2 weeks) and short-term (4 to 6 weeks) pain relief in the knee | Conditionally recommended in the knee for longer-term effect, with symptom improvement beyond 12 weeks and a favourable safety profile | Not covered here |
This disagreement is real, and it is worth understanding rather than glossing over. The American College of Rheumatology reviewed the trials and found that the apparent benefit was restricted to studies at higher risk of bias. When the analysis was limited to trials at low risk of bias, the effect compared with a saline injection approached zero. That is what drove their recommendation against it. They also said something that often gets left out: the conditional recommendation against is still consistent with using hyaluronic acid through shared decision-making when other options have been exhausted or have not given satisfactory benefit, and it was not intended to influence insurance coverage decisions.
OARSI read the evidence differently and conditionally recommends hyaluronic acid for the knee, on the basis of symptom improvement beyond 12 weeks and a safety profile they considered favourable, including compared with repeated steroid injections.
Both agree on one thing. Neither supports hyaluronic acid for hip osteoarthritis.
If you need relief quickly, or the joint is flaring, a corticosteroid injection is the option with the strongest guideline support and the fastest onset, and it can help you take part more fully in physiotherapy. If you have already tried the basics and a steroid injection has not given you enough, hyaluronic acid is a reasonable thing to discuss, with the caveat above about what the best trials show. Neither replaces exercise and weight management, which remain the strongest treatments for osteoarthritis.
PRP uses a small sample of your own blood, spun to concentrate the platelets, which is then injected into the joint. The American College of Rheumatology strongly recommends against it for knee and hip osteoarthritis, and their stated concern is the variation between available preparations and techniques, which makes it hard to know exactly what is being injected. Our full PRP guide is in preparation.
Read the full guides for the hyaluronic acid injection and the corticosteroid injection, including the risks in numbers for each. The cost and coverage guide covers the fee and insurance. If you are not sure your knee pain is osteoarthritis at all, start with osteoarthritis or inflammatory arthritis.
A corticosteroid injection usually works fastest, often within 24 to 48 hours. Hyaluronic acid builds over the following weeks rather than straight away.
Relief from a corticosteroid injection is usually short-term, commonly fading over a few weeks to a couple of months. Some people get relief from a single hyaluronic acid injection that can last up to about six months, though across high-quality reviews the average benefit over a placebo injection is small and not always meaningful.
Not on all of them. They agree that corticosteroid injections have a place in knee osteoarthritis. They disagree about hyaluronic acid: the American College of Rheumatology recommends against it for the knee, while OARSI conditionally recommends it. On platelet-rich plasma they are clearer, and the American College of Rheumatology strongly recommends against it for the knee and hip.
The Durolane product is not covered by OHIP, so there is an out-of-pocket cost, and many extended health plans cover part or all of it. The clinic confirms the current fee with you before you book. The cost and coverage guide has the detail.
Sometimes, at different times. Injections are spaced out, and which one suits you depends on your joint, how quickly you need relief, and what you have already tried. That is a conversation for your assessment.