PRP & Exosomes

PRP for Knee Osteoarthritis Did Not Beat Saline in the Largest Blinded Trial

A blinded 288-patient trial found no clear gain over saline at 12 months, and guideline groups disagree.

The largest blinded trial of PRP for knee osteoarthritis does not favor the treatment. In the RESTORE trial, platelet-rich plasma (PRP) injections did no better than saline placebo injections on knee pain or cartilage volume at 12 months.

Guideline groups in Australia and Europe disagree about whether PRP has any role in knee osteoarthritis (OA). The U.S. Food and Drug Administration (FDA) has not approved PRP for knee OA, and insurers generally do not pay for it.

What the Evidence Shows on PRP for Knee Osteoarthritis

PRP for knee osteoarthritis showed no clear benefit over saline in RESTORE, the largest blinded placebo-controlled trial on the question. At Medical Frontier, every guide we write starts from the published trial and guideline text, and for knee OA those sources point in different directions.

The RESTORE trial, led by Bennell, Paterson, Hunter and colleagues, was published in JAMA (Journal of the American Medical Association) in November 2021. It enrolled 288 adults aged 50 or older with painful, mild to moderate osteoarthritis on the inner (medial) side of the knee. Participants, the clinicians giving injections, and the assessors measuring results were all blinded, so nobody knew who received PRP.

Each group had 144 people, and 269 participants (93%) finished the study. Everyone received three ultrasound-guided injections into the knee joint, one week apart. The PRP group received fresh leukocyte-poor PRP, meaning low in white blood cells, made with a commercial kit using a single spin at 1500g for 5 minutes, while the control group received saline.

OutcomePRPSaline placeboReported difference
Change in knee pain at 12 months (0 to 10 scale)-2.1-1.8-0.4 (95% CI -0.9 to 0.2), P = .17
Yearly change in medial tibial cartilage volume-1.4%-1.2%-0.2% (95% CI -1.9% to 1.5%), P = .81
Reported overall improvement at 2 months48.2%36.2%P = .02
3 or more areas of cartilage thinning at 12 months17.1%6.8%P = .02

The 95% confidence interval (CI) for pain runs from a 0.9-point edge for PRP to a 0.2-point edge for saline. Because that range includes zero, RESTORE could not rule out that PRP and saline perform the same. The cartilage result followed the same pattern.

Of 31 secondary outcomes, 29 showed no difference between groups. The two exceptions pulled in opposite directions: more PRP participants felt better overall at 2 months, and more PRP participants had three or more areas of cartilage thinning at 12 months. The authors concluded that their findings do not support using PRP for knee OA.

RESTORE has limits that matter for your decision. The authors noted that PRP preparations vary from one system to another and that their sample had mild to moderate disease only. Commentators added that the trial tested one commercial, non-standardized preparation.

How the Guideline Positions Compare

Medical guideline groups that reviewed PRP for knee osteoarthritis reached conflicting positions, and that conflict is the main takeaway for a patient. No single body speaks for the field.

SourcePosition on PRP for knee OAWho it applies to
RESTORE authors (JAMA, 2021)Findings do not support PRP useAdults 50+ with KL grade 2 or 3 medial knee OA
RACGP guidelineUnable to recommend for or againstKnee and hip OA
ESSKA-ICRS consensusAppropriate after failed conservative treatmentAge 80 or younger, KL grade 0 to III
FDANot approved as a treatmentAll knee OA patients

The Royal Australian College of General Practitioners (RACGP) states it is unable to recommend for or against PRP injection for knee or hip OA. The ESSKA-ICRS consensus is a European consensus. It considered PRP appropriate for patients aged 80 or younger with Kellgren and Lawrence (KL) grade 0 to III knee OA after conservative non-injection or injection treatments have failed.

ESSKA-ICRS did not consider PRP appropriate as a first treatment. Our read of these sources is that they give no reason to pay for PRP before trying other care. If conservative treatment has failed and your knee falls within KL grade 0 to III, the European consensus leaves room for PRP, but the largest blinded trial still found no gain over saline.

Some readers should hold off on PRP for now:

  • People who have not yet tried conservative treatment, since ESSKA-ICRS did not consider PRP appropriate as a first step.
  • People whose imaging shows KL grade IV knee OA, where ESSKA-ICRS did not consider PRP appropriate.
  • People older than 80, who fall outside the ESSKA-ICRS age range.

Our assessment would change if a large blinded trial of a different, standardized PRP preparation showed a clear gain over saline. FDA approval of a PRP product for knee OA, or guideline groups moving toward the same position, would also change it.

Bone-on-Bone Knees and Cartilage Loss

The studies in this guide do not support PRP for the most severe knee arthritis. The ESSKA-ICRS consensus did not consider PRP appropriate in KL grade IV.

RESTORE enrolled only knees graded KL 2 or 3. That design means the trial tells you nothing direct about grade IV knees, in either direction. Ask your clinician which KL grade your X-ray report shows, because the ESSKA-ICRS position depends on it.

RESTORE measured medial tibial cartilage volume and found yearly loss of 1.4% with PRP and 1.2% with saline, with no clear difference. If a clinic claims cartilage regrowth, ask which published study supports that claim, using the steps in our guide on how to vet a clinic. More background on knee conditions sits in our knee guide.

What a PRP Knee Session Involves and Its Risks

A PRP knee injection uses your own blood, spun in a centrifuge and injected back into the knee joint.

In RESTORE, the injections were guided by ultrasound and given three times at weekly intervals. Ask a clinic whether it uses imaging guidance and how many injections its protocol includes, since RESTORE’s protocol may differ from what a clinic offers.

Adverse events in RESTORE were mostly minor and short-lived, and the trial reported no serious events related to treatment. More PRP participants than saline participants reported knee pain, swelling and stiffness after injection. Our guide to PRP injection recovery and side effects covers what to expect after a session.

Cost and Insurance for PRP for Knee Osteoarthritis

Insurers generally do not cover PRP for knee osteoarthritis. Read our guide on whether insurance covers PRP, then ask your plan for a written coverage decision.

Prices vary by clinic; see our PRP injection cost guide.

PRP is not FDA-approved for knee OA, as our FDA status overview describes. Use the cost estimator to compare a written clinic quote against your budget.

Questions to Ask a Clinic before Booking

A clinic should answer five questions in writing before you pay for PRP knee injections:

  1. Is your preparation leukocyte-poor or leukocyte-rich, and which system makes it? RESTORE tested a leukocyte-poor preparation from one commercial kit.
  2. How many injections does your protocol include, how far apart are they, and are they guided by ultrasound?
  3. What is the total written cost for the full course, including follow-up visits?
  4. Which pain and function scores do you record before and after treatment to track my progress?
  5. Which conservative treatments should I try first, given that ESSKA-ICRS did not consider PRP appropriate as a first treatment?

A clinic that cannot answer the first question cannot tell you how its product relates to the trial data. See our guide on how to vet a clinic. To compare PRP with other injections, see PRP vs cortisone, PRP vs hyaluronic acid, and PRP vs stem cell.

Before scheduling PRP for knee osteoarthritis, ask your clinician for your KL grade and get a written quote for the full injection course.

This is general information, not medical advice; talk to a licensed clinician.

Frequently asked questions

Does PRP work for knees?

In the largest blinded trial, PRP did no better than a saline placebo. The RESTORE trial followed 288 adults with mild to moderate knee osteoarthritis (OA). Knee pain fell by 2.1 points on a 0 to 10 scale with PRP and by 1.8 points with saline at 12 months, and the researchers could not separate the two groups. Guideline groups split on what this means: one Australian guideline is unable to recommend for or against PRP, while a European consensus considers it appropriate for some patients after other treatments fail. For other injection options, see our guides on PRP vs hyaluronic acid and PRP vs cortisone.

Is PRP for knees FDA approved?

No. The U.S. Food and Drug Administration (FDA) has not approved platelet-rich plasma as a treatment for knee osteoarthritis. See the agency's consumer alert on regenerative medicine therapies. Our FDA status overview explains what approval does and does not mean for a clinic offering PRP.

How long does PRP for knees last?

No duration has been established by the studies in this guide. In the RESTORE trial, more PRP participants reported overall improvement at 2 months (48.2% vs 36.2% with saline). By 12 months, knee pain scores in the PRP group were no different from the saline group. RESTORE reported results at 12 months, so treat any clinic promise of a set number of pain-free months as unsupported unless the clinic shows you a published study.

Does insurance or Medicare cover PRP for knees?

Coverage is generally not provided. Read our guide on whether insurance covers PRP, then ask your plan, including Medicare if you are enrolled, for a written answer before treatment. Our PRP injection cost guide breaks down what clinics charge.

Is PRP good for bone on bone knees?

The evidence in this guide does not support PRP for the most severe knee arthritis. The ESSKA-ICRS consensus did not consider PRP appropriate for knees at Kellgren and Lawrence (KL) grade IV. The RESTORE trial enrolled only grade 2 or 3 knees, so it says nothing direct about grade IV. Our knee conditions guide covers other treatments people weigh for knee pain.

Is PRP or stem cell therapy better for knees?

Guideline positions on the two differ. The Royal Australian College of General Practitioners (RACGP) is unable to recommend for or against PRP for knee osteoarthritis, and it does not recommend offering stem cell therapy for knee or hip OA. Compare the two in our PRP vs stem cell guide and review pricing in our knee stem cell cost guide.

Medical disclaimer. Medical Frontier is an independent educational resource. This page is for general information only and is not medical advice, diagnosis, or treatment, and does not create a doctor–patient relationship. Most regenerative therapies discussed here are not FDA-approved for the uses described and may be offered under limited exemptions or in clinical trials only. Always consult a licensed physician before making any treatment decision.