Knee Osteoarthritis in an Active 54-Year-Old: Platelet-Rich Plasma Against a Rigorous Null Trial
The best single trial ever run on PRP for knee osteoarthritis found nothing beyond saline. An active 54-year-old who has researched it himself wants it anyway, and is prepared to pay for it.
Miguel R., 54, played varsity tennis in college and never really stopped — three matches a week now in a competitive league, coaching his daughter's high school team on the side, the kind of schedule most men his age gave up years ago and one he has no intention of scaling back voluntarily. A grade 2 right knee osteoarthritis has been slowing him down for about eighteen months, not enough to threaten daily life but enough that he's noticed his level of play declining match by match, which bothers him considerably more than the pain itself does. An oral NSAID gets him through ordinary days without trouble; it doesn't get him through a three-set match the way it used to, and lately he's started skipping the third set rather than push through it.
He has read extensively about platelet-rich plasma, has a specific clinic in mind, and is prepared to pay for it himself since it isn't covered by his insurance. The single most rigorous trial ever conducted on the question — Bennell's 2021 RESTORE trial, 288 patients, blinded to participant, injector and assessor alike against a genuine saline placebo, followed for a full year — found no significant difference between PRP and saline on either pain or the rate of cartilage thinning on MRI. The uncomfortable part, for him, is where he sits relative to it: RESTORE enrolled patients aged fifty and over with symptomatic medial knee osteoarthritis at Kellgren-Lawrence grade 2 or 3, and at fifty-four with grade 2 disease Miguel meets every one of those criteria. He is not an extrapolation from its result; he is the population it recruited, and what distinguishes him from the average participant is the level of function he wants back, not his eligibility. That result sits against a broader, more heterogeneous literature that includes some positive trials, most of them smaller, often compared against hyaluronic acid rather than a true placebo, and rarely using the same PRP preparation as any other study, which makes comparing them to each other almost as difficult as comparing any one of them to RESTORE. Miguel isn't asking the group to resolve that broader literature for him; he's asking whether, given his own specific goals and his own willingness to accept the cost and the uncertainty, PRP is a reasonable thing for a man in his position to do.
An elective procedure, a null trial, and a patient who has already decided
Miguel is inside the trial population on paper, but what he's asking of his knee isn't what that population was asked about. He's fifty-four, otherwise healthy, and trying to preserve a specific, high-demand level of function for what could be another twenty or thirty years of playing — an outcome no trial here measured, because the endpoints were average knee pain and cartilage volume, not whether a man can hold his level through a third set.
Weighed against that timeline, repeated corticosteroid carries its own real cartilage-loss signal, and daily NSAID use across decades isn't free either. PRP has a comparatively benign physical safety profile. If he wants to spend his own money on an uncertain-but-plausible benefit rather than accept a slow decline in his level of play, I don't think that's an unreasonable choice for him specifically to make.
I'd be more comfortable with that framing if RESTORE hadn't been run, and I want to be precise about what it found, because overstating it would be its own kind of dishonesty. It did not find zero. Pain fell 2.1 points on PRP against 1.8 on saline, and medial tibial cartilage volume fell 1.4% against 1.2% — PRP numerically ahead on one, numerically behind on the other, neither difference anywhere near significance, and the pain gap of 0.3 points against a minimal clinically important difference of 1.8. What it found is a difference too small to matter, measured well enough that we can say so.
And Miguel isn't outside that population looking in. He meets its entry criteria on age and on radiographic grade. Whatever we tell him, we can't tell him the trial studied someone else.
The positive studies people cite are smaller, often compared against an unproven comparator rather than a real placebo, and use PRP preparations that vary enough between studies that 'PRP works' isn't really one claim being tested twice. Recommending it because he wants it to work isn't the same as recommending it because the evidence supports it, and I think we owe him the difference stated plainly.
I don't think those two positions are actually as far apart as they sound. Nobody here is going to tell Miguel PRP is proven to work — RESTORE settles that question honestly, and he deserves to hear it exactly the way it was just stated.
He should also hear the guideline language exactly as it reads, not softened: the 2019 ACR guideline doesn't merely decline to endorse PRP for knee osteoarthritis, it strongly recommends against it, and its stated reason is the same one just raised — that preparations vary so much between studies that there isn't a standardized product to recommend. "Strongly recommended against" is a heavier thing to walk past than "unproven," and he's owed the actual phrase.
But once he's heard that clearly and still wants to proceed with his own money, on a procedure with a low physical risk, that's a legitimate choice for an informed adult to make about his own knee and his own goals. Our job here isn't to gatekeep a low-risk elective procedure from someone who's done more homework on it than most patients ever do — it's to make sure the homework he's done includes what we just told him.
Not resolved in the usual sense: the group agreed on what to tell Miguel — RESTORE found no benefit over saline in a population he belongs to rather than resembles, the broader supportive literature is smaller and methodologically inconsistent, and the ACR guideline strongly recommends against PRP for knee osteoarthritis — and agreed that if he chose to proceed after hearing that clearly, he should do so through a reputable practitioner rather than being refused outright.
Miguel, after hearing the full picture, said he wanted to proceed anyway. Not agreed among the three physicians: whether that outcome represents shared decision-making working as intended, or a case where a patient's own strong prior belief made the honest-disclosure conversation close to symbolic. The rheumatologist remained uneasy with the outcome even while agreeing the process itself was handled honestly.