- Design
- Systematic review and meta-analysis of 21 randomised trials
- Population
- 1,790 patients with knee osteoarthritis
- Primary outcome
- Pain and function scores (VAS, WOMAC, KOOS, IKDC)
- Effect
- VAS pain at 12 months SMD 0.76 favouring PRP; no difference at 3 and 6 months
A meta-analysis pooled 21 randomised trials (1,790 patients) comparing intra-articular platelet-rich plasma with hyaluronic acid for knee osteoarthritis, published in September 2026.
PRP showed no pain advantage on visual analogue scales at three or six months. At 12 months it reduced pain more (standardised mean difference 0.76). WOMAC scores favoured PRP at every time point. KOOS and IKDC scores did not differ. High-concentration, leukocyte-rich and activated preparations were linked to better 12-month results, and platelet concentration predicted effect on meta-regression.
Heterogeneity and risk of bias were substantial, and PRP preparations vary widely. The authors themselves describe the advantage as small and urge caution.
This does not establish either injection as effective against placebo, and neither changes the foundation of exercise, weight management and analgesia. Where an injection is offered, the data lean modestly towards PRP for longer-term pain.
- Keep exercise and weight management as the basis of knee osteoarthritis care.
- If offering an injection, explain that any PRP advantage over hyaluronic acid is small and later.
- Record the PRP preparation used, since concentration and leukocyte content varied in effect.
- Set expectations on cost, as benefit is modest and preparations are not standardised.
Why it matters
Patients increasingly ask for PRP, and this sets a realistic, modest expectation.
Don't overread it
Benefit was absent at 3 and 6 months on pain scores, and trial quality was mixed.
The statistics, in plain English
A standardised mean difference of 0.76 is moderate on paper, but it appeared only at one time point for one pain scale, and other scores showed no difference. Results that change with the scale and time chosen are less reliable than a consistent effect.
Read the rest in the app
You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.
Scan to keep reading on your phone. No account needed to start.
Tomorrow morning, before your first patient
One edition a day for orthopaedics, written by the desk, every claim tied to its paper. Six minutes.
Get the app — free