The short version

Corticosteroid is the most predictable — fastest onset, best-characterised decay, cheapest, and covered.
Hyaluronic acid is slower and more contested, with a small inconsistent effect and a coverage problem.
PRP ranks highest in pooled analyses and failed the best single trial. It is also the most expensive and rarely covered.
None of them change the joint. The strongest-evidence intervention for an arthritic knee is still exercise.

This page exists because the three options are usually presented one at a time, by whoever offers that one. Seeing them next to each other changes the decision — not because one wins, but because the trade-offs become visible.

The three, side by side

Knee osteoarthritis only. Each column links to the full article on that injectable, where the evidence and its limits are set out properly. Costs vary widely by region and practice.
	Corticosteroid	Hyaluronic acid	PRP
How fast	Days. Strongest at 1–2 weeks.	Slower — weeks rather than days.	Slower still, where it works at all.
How long	Decays steadily; no detectable effect by 26 weeks.	Signal at 2–8 weeks on some measures, absent on others.	Disputed. The largest trial found no difference at 12 months.
Strength of evidence	Consistent short-term benefit; Cochrane graded the overall quality low.	Small and inconsistent — significant on some outcomes, not others.	Pooled analyses rank it first; the best single trial was negative.
Cost and coverage	Cheapest, and routinely covered.	More expensive, usually a course, coverage tightening.	Most expensive, typically out of pocket.
Main drawback	Repeated use has a real question over it.	You may be paying for a small effect.	You may be paying a lot for no effect.
Read in full	Corticosteroid, explained	Hyaluronic acid	PRP and the evidence

Why do the rankings disagree?

Network meta-analyses comparing injectables for the knee tend to place PRP at or near the top, hyaluronic acid in the middle, and corticosteroid lower — because steroid's effect fades fastest and those analyses often measure at three to six months.

That ordering inverts if you rank by quality of evidence rather than by effect size. Corticosteroid's modest benefit comes from a Cochrane review of 27 trials with a clearly described time course. PRP's higher ranking comes from pooling many small trials, and the one large rigorous trial found nothing.

Both orderings are defensible. They answer different questions — which had the biggest measured effect versus which finding do I trust most — and knowing which question a ranking answered is the difference between using it and being used by it.

How I actually decide

Inflamed knee, needs relief now, cost matters: corticosteroid. Responds to steroid but wants fewer of them, or difficult diabetes: hyaluronic acid is reasonable. Wants PRP, can afford it, and understands the RESTORE result: their call, made with open eyes.

That is a decision heuristic rather than a finding. What it is not is a ranking — the right answer depends on how inflamed the joint is, what it costs the person in front of me, and what else is being done about the knee.

What outranks all three

Every injection on this page is a window, not a repair. None of them change the structure of the joint, and none of them are what the guidelines put first.

Both the OARSI and ESCEO guidelines position education, structured exercise and weight management as core treatments, with intra-articular injection for pain that persists despite them. An injection that buys twelve weeks of tolerable pain is genuinely valuable if those twelve weeks are spent strengthening. It is a postponement if they are not.

That is also the honest framing for cost. Money spent on PRP is money not spent on supervised strengthening, which has better evidence and more durable effects.

This article is general education comparing injectable options for knee osteoarthritis. It is not medical advice, does not create a patient–provider relationship, and cannot account for an individual situation. Which option is appropriate depends on the stage of arthritis, other medical conditions, cost and coverage, and what else has been tried. For a specific person, that is a conversation with their own clinician. See the editorial policy for how this content is written and reviewed.

Sources

Jüni P, Hari R, Rutjes AWS, et al. Intra-articular corticosteroid for knee osteoarthritis. Cochrane Database of Systematic Reviews 2015;10:CD005328. Retrieved via PubMed.
Bennell KL, Paterson KL, Metcalf BR, et al. Effect of intra-articular platelet-rich plasma vs placebo injection on pain and medial tibial cartilage volume: the RESTORE randomized clinical trial. JAMA 2021;326(20):2021–2030. Retrieved via PubMed.
Migliorini F, Maffulli N, Schäfer L, et al. Less pain with intra-articular hyaluronic acid injections for knee osteoarthritis compared to placebo. Pharmaceuticals 2024;17(11):1557. Retrieved via PubMed.
Qiao X, Yan L, Feng Y, et al. Efficacy and safety of corticosteroids, hyaluronic acid, and PRP and combination therapy for knee osteoarthritis: a network meta-analysis. BMC Musculoskeletal Disorders 2023;24(1):926. Retrieved via PubMed.
Anil U, Markus DH, Hurley ET, et al. The efficacy of intra-articular injections in the treatment of knee osteoarthritis: a network meta-analysis. The Knee 2021;32:173–182. Retrieved via PubMed.
Arden NK, Perry TA, Bannuru RR, et al. Non-surgical management of knee osteoarthritis: comparison of ESCEO and OARSI 2019 guidelines. Nature Reviews Rheumatology 2021;17(1):59–66. Retrieved via PubMed.