Ninety people with bone-on-bone knee arthritis. All of them already on the waiting list for a knee replacement. All of them told, in effect, that they had run out of options.
Researchers randomised them anyway — two platelet-rich plasma injections, or a cortisone injection, or an anti-inflammatory tablet — and then watched what happened.
Nobody in the PRP group dropped out. Six people in the other two groups went on to have their knee replaced earlier than planned.
That is the finding worth sitting with, because it speaks to something no imaging study can tell you: whether being on the list means your knee is finished, or only that the standard sequence ran out before anyone looked further.
What the trial actually measured
The study is Lacko M and colleagues, Journal of Orthopaedic Surgery and Research, in press (accepted 26 May 2026). The patients had Kellgren–Lawrence grade 3–4 osteoarthritis — the advanced end, the stage where “bone on bone” gets said out loud.
In the PRP group, pain on the visual analogue scale fell from 6.2 to 3.3 and held at 3.7 at final follow-up. Function measured by WOMAC improved from 49.0 to 32.5. Opioid requirement fell. Blood markers moved coherently with all of that — COMP, MMP-3, IL-6, IL-18, TNF-α and CGRP down; sTREM2, sRAGE and TGF-β1 up.
The anti-inflammatory arm did not merely underperform. It went backwards — WOMAC worsened from 49.0 to 57.9. The people taking the tablet were functionally worse at the end of the study than when they started.
A second, independent trial points the same direction. Sánchez Santiuste and colleagues, publishing in the Journal of Clinical Medicine in November 2025, studied 86 patients with grade III–IV disease and found intraosseous PRGF outperformed intraosseous saline, with clinically important improvement reached by 66.7% versus 46.3%.
What this does not mean
This is where this field routinely overreaches, so let us be exact.
Cartilage did not grow back. The authors make no such claim, and neither do we. Their own words for what PRP did here are “safe bridge therapy prior to arthroplasty.”
A bridge. Not a cure. Not regeneration. Not a permanent alternative to replacement.
What it means is narrower and more useful: in people with advanced arthritis who were already scheduled for surgery, a non-surgical injection produced real pain and function improvement, lowered opioid use, and postponed an operation for most of them.
Why postponement is worth something on its own
A knee replacement cannot be undone.
If replacement is still the right answer for you in eighteen months, it will still be available in eighteen months. The reverse is not true — there is no route back to your own knee once it has been removed. That asymmetry is the entire argument for sequencing carefully, and it is rarely part of the conversation.
There is a second reason, specific to younger patients: implants have a finite service life. Having the first replacement earlier makes a revision later more likely, and revision is a harder operation than the original. Delay is not merely avoidance — for some people it changes how many operations they will have in a lifetime.
The step that usually gets skipped
The standard pathway runs: anti-inflammatories, then physical therapy, then cortisone, then replacement. When the cortisone stops working, the pathway is exhausted and surgery is what remains.
But cortisone losing its effect is not evidence that the joint is beyond help. It tells you the inflammatory driver was being suppressed rather than addressed — and repeated cortisone has its own trial data on cartilage worth reading before the next one.
Nothing in that sequence asks why this joint is degrading faster than the other one. Insulin resistance, systemic inflammation and poor tissue perfusion do not show up on an X-ray, and they do not resolve because a joint is replaced. A knee can be replaced in a body that is still actively degrading the next joint.
Five questions before you accept the date
Am I on this list because the knee is unsalvageable, or because the standard sequence ran out? Has anyone assessed the metabolic and inflammatory drivers, or only the structure on imaging? What is my realistic expected outcome — not the average, but for someone my age, weight and metabolic status? If I delay six to twelve months and try a biological approach first, what specifically do I lose? And if I am not one of the people who does well, what are my options afterwards?
A surgeon who welcomes those questions is a good sign. One who treats them as obstruction has told you something too.
Where we stand
We are not against knee replacement. It is the right operation for a great many people, and for a knee that is genuinely finished nothing else restores that function.
We are against taking the irreversible step before the reversible ones have been honestly tried — particularly now that there is randomised evidence, in exactly this population, that a meaningful proportion get enough relief to wait.
If you are on a waiting list and nobody has looked past the X-ray, that is worth a second opinion before the date arrives.
Frequently asked questions
Can anything be done if I am already on the knee replacement waiting list?
That is precisely the population this trial studied. Ninety patients with grade 3–4 osteoarthritis, all already scheduled for replacement, were randomised to PRP, cortisone or an anti-inflammatory. In the PRP group nobody dropped out of the study, while six patients across the two comparison groups proceeded to earlier replacement. Being on the list is not the same as having no remaining options — but what applies to your knee specifically requires examination and evaluation.
Does PRP regrow cartilage?
No, and you should be skeptical of anyone who says otherwise. The trial authors describe PRP here as a “safe bridge therapy prior to arthroplasty” — it improved pain, function and opioid requirement, and postponed surgery. It did not rebuild the joint. The honest answer on cartilage regrowth is worth reading alongside this.
Is there a risk in waiting?
Replacement remains available later, which is the central point — the operation is irreversible, the delay is not. For younger patients there is an additional consideration in the other direction: implants have a finite service life, so an earlier first replacement makes a later revision more likely. That is a conversation to have with both your surgeon and a physician assessing the non-surgical options.
Why did the anti-inflammatory group do worse?
In this trial the aceclofenac arm’s WOMAC score moved from 49.0 to 57.9 — worse at the end than at the start. An anti-inflammatory suppresses a signal without addressing what is generating it, and the long-term data on daily anti-inflammatory use is its own subject.
I have been told I am too young for a replacement. What does that mean?
It usually means the surgeon is weighing implant lifespan against your remaining years, and would rather you have one operation later than two or three starting now. That window is exactly when the biological options are worth examining — we have written separately about not wasting it.
Key takeaways
- Ninety patients with grade 3–4 knee osteoarthritis, all already on the replacement waiting list, were randomised to PRP, cortisone or an anti-inflammatory tablet.
- Nobody in the PRP group dropped out of the study; six patients across the two comparison groups went on to replacement earlier than planned.
- The authors describe PRP here as a “safe bridge therapy prior to arthroplasty” — postponement, not cartilage regrowth.
- The anti-inflammatory arm went backwards: WOMAC worsened from 49.0 to 57.9, worse at the end than at the start.
- Replacement stays available later. The operation is irreversible; the delay is not.
Medically reviewed by Gurpreet Singh Padda, MD, MBA, MHP — Board Certified in Anesthesiology, Pain Medicine, Interventional Pain Management, Addiction Medicine, and Obesity Medicine. Last reviewed August 2026.
This article is educational and is not a substitute for evaluation, diagnosis, or treatment by a physician. Individual results vary, and not every patient is a candidate for the therapies described. Do not start, stop, or change any treatment without consulting your physician. Orthobiologic therapies including PRP are not FDA-approved for this indication and are provided as part of physician-directed care. The Lacko trial is cited as in press.
Find out whether your knee is actually finished
Regen.MD begins with a physician-led Clinical Evaluation — a review of your history, imaging, and metabolic data, and a written terrain roadmap. Evaluation is contingent upon review of your data.
Questions? Call (314) 295-3000 or text (314) 886-5902.