Speaker 1:
Dr. Jared Powell:
Welcome back to the Shoulder Physio Podcast. A few weeks ago, I posted a meta-analysis on LinkedIn. It was six randomized control trials that included 355 patients. It was platelet-rich plasma versus saline for tennis elbow. I thought the finding was pretty straightforward. Platelet-Rich plasma or PRP did not outperform saline for pain or function at any time point, not at four weeks or three months or six months, nothing nada, zilch, zero, no effect at any time point.
Dr. Jared Powell:
However, within 48 hours, the post had garnered thousands of impressions and dozens of comments literally from orthopedic surgeons, regenerative medicine physicians, whatever that is, and sports medicine doctors one told me the meta-analysis means nothing. Another said, I clearly hadn't read beyond the abstract. One particular clinician reported a 100% success rate with PRP. Another cited a registry of 2,200 elbows with no control group as evidence of its effectiveness. Someone even compared me to a football fan criticizing a coach because I don't personally inject PRP. So this was a really fun exercise, and I got to see the very best and very worst of medicine. What I found most striking was that every single response was a meta argument. It was either about dose or about my credentials or about my nefarious motives. Nobody all engaged with the actual data. So today I want to walk through the evidence carefully, the meta-analysis that ignited this whole controversy, the paper everyone cited in their rebuttal to my claims, the corticosteroid trajectory that explains the apparent paradox and a framework from Professor John Orchard that ties it all together.
Dr. Jared Powell:
By the end of this episode, I think you'll see why the comparator you choose plays a large role in the conclusion you reach. Okay, so let's start with the paper that I posted on LinkedIn and also on Instagram. It was a paper published in the American Journal of Sports Medicine in 2026. So quite recent, it's a systematic review of meta-analysis with one very important design decision.
Dr. Jared Powell:
It only included RCTs that compared PRP, platelet rich plasma to placebo, and in this case, placebo was defined as a normal saline injection. So there were no comparisons to corticosteroids, no comparisons to autologus blood, no comparison to active co comparators, just PRP versus saline. This is important because previous meta analyses muddied the waters by pooling PRP against different co comparators. When you mix corticosteroid comparisons, autologous blood comparisons, and and saline comparisons into one analysis, you get a blended and confused estimate. That doesn't clearly answer the question. Does PRP actually do anything above and beyond what saline does? So the meta-analysis itself included six RCTs and 355 patients. The results were pretty clear, no significant benefit for PRP over saline at four weeks, at eight weeks at 12 weeks, or 24 to 26 weeks for both pain and function. The effect sizes were essentially zero. Furthermore, the I squared measure for function was 0% at every time.
Dr. Jared Powell:
Point I square is just a measure of the heterogeneity in the findings in the individual studies within the meta-analysis. So this means that when the I squared measure is zero, it means that all the studies within this meta-analysis were consistent in finding nothing. Yes, they were consistent in finding absolutely nothing. But hang on. Before we go too far, we must talk about a key detail. The studies varied considerably in PRP preparation, and I fully accept that. So some studies had leukocyte rich, others had leukocyte poor. In others, it was unspecified volumes of PRP injected varied. There were single versus multiple injections, different techniques, different kit, so on and so forth. So you'd expect that variability to obscure a real effect if one existed. Instead, the null result held regardless of preparation, no version, no variety, no particular method of PRP preparation outperformed simple saline. So the study published in the American Journal of Sports Medicine this year in 2026 is pretty clear.
Dr. Jared Powell:
PRP doesn't beat saline for tennis elbow. However, the pushback centered on a different paper published by ODing and colleagues also published in the American Journal of Sports Medicine this time in 2025. So this is the paper that was consistently and incessantly sent to me by various members of the regenerative medicine community. The argument was, dose matters. The dose is important, why aren't you considering the dose? And they sent me this paper to prove their point. So I went and studied the paper in agonizing detail because I'm weird like that. And let me walk you through exactly what ODing and colleagues did in this particular paper. So they identified 13 RCTs evaluating PRP for tennis elbow and classified them by platelet concentration or the dose of PRP. High dose was defined as greater than three times baseline platelet level versus low dose, which was less than three times baseline platelet level.
Dr. Jared Powell:
So to put that in concrete terms, normal blood has roughly 150,000 to 450,000 platelets per microliter. A three times concentration means you're aiming for at least 450,000 to 1.3 million. This is SRA physiological levels. You achieve this by drawing more blood and centrifuging it down into a smaller volume. That's how you prepare platelet-rich plasma. So you draw more blood and then you centrifuge it or spit it down into a smaller volume in this creates this high dose PRP. So in this study, they ran a meta-analysis of meta regression and found that high dose PRP produced a mean positive difference of 1.3 points on a 10 point visual analog scale compared to all alternative treatments, while low dose PRP showed essentially nothing. So let me repeat that. In this meta-analysis, they found that high dose PRP produced a positive benefit of 1.3 points for pain on a 10 point visual analog scale compared to all other alternative treatments.
Dr. Jared Powell:
So it made a difference of 1.3 points outta 10. Okay, sounds compelling, but let's look at the numbers and let's look at the stats. That's what this episode is all about. Of the eight high dose studies included four of these compared PRP to corticosteroids. Those four papers happened to produce the biggest effects. They, they had mean differences ranging from 1.1 to 2.4 points outta ten three papers compared PRP to autologous blood. These had smaller effects. None were individually significant and one, only one compared high dose PRP to saline. That study showed a mean difference of 0.170. Let's call this nothing. So the pooled estimate of 1.3 points that's being cited as proof high dose PRP works is being pulled there almost entirely by corticosteroid comparisons. Now, you may be thinking to yourself, why is this the case? And I'm glad you're thinking this because this is where we're going to be.
Dr. Jared Powell:
This is where we're going to go next. So I think this is where the argument becomes crystal clear and the whole thing just kind of unravels for the high dose PRP proponents. Because corticosteroids are not a neutral comparator. They're an actively deteriorating one. In 2010, Brooke and colleagues published a paper in the Lancet, one of the most prestigious medical journals in the world. It included 41 RCTs and 2,600 patients, and they looked at injection therapies for tendinopathy, for tennis elbows. Specifically, they found that corticosteroid injection was effective in the short term. We kind of know this, and that's okay. There was, that's no surprise. But at six months and at 12 months, corticosteroids performed worse than no treatment at all. Yes, worse, not equal, worse, in my opinion or my estimation. This knowledge should change the interpretation of every PRP versus corticosteroid trial. If corticosteroids are declining below or back towards baseline, while PRP is following the natural trajectory of a condition that self resolves in roughly 80% of cases, the lines will cross.
Dr. Jared Powell:
The lines will intersect. At some point. PRP will end up looking superior because it's not a detrimental intervention. But PRP is not beating a stable control. It's beating a control that's failing. Saline would win this comparison. Two as wait and see already. Did we actually see this in the ODing paper that was published in 2025? So what I, what I did is I went through and pulled each high dose PRP versus corticosteroid paper and looked at the results in every single one of these four papers. Like literal clockwork, the corticosteroid groups improved quickly over six to eight weeks and then deteriorated. Conversely, the PRP groups improved gradually. The lines intersect at around three months. The long-term PRP superiority is a corticosteroid rebound effect, not evidence that PRP heals tendons. Despite this, the conclusion of the study tells clinicians they should ensure a S supra physiological platelet concentration if they're going to do a PRP injection.
Dr. Jared Powell:
This is actually a clinical directive built on a single pain outcome, driven by the wrong comparator with no placebo controlled support and significant industry conflicts. It's simply hypothesis dressed as recommendation. So where does this leave us? I hear you screaming at me through the airwaves. Have what looks like a paradox. So PRP beats corticosteroids at longer term corticosteroids beat placebo, at least in the short term. A PRP doesn't seem to beat placebo. It seems impossible. So this is where I'm gonna talk about Professor John Orchard, who is a sports medicine physician and public health researcher, the University of Sydney, and he published a paper in 2025, names this pattern and explains it beautifully. Calls it the Rock Paper Scissors phenomenon. Founded across four common musculoskeletal conditions. S elbow, knee osteoarthritis, rotator cuff tendinopathy, implanter fas osteopathy pattern, he says is quite consistent. PRP beats corticosteroids in the medium to long term corticosteroids beat placebo in the short term SIBO ties or beats PRP.
Dr. Jared Powell:
It does seem paradoxical until we remember the corticosteroid trajectory. Corticosteroids have a biphasic response. So this means it has an excellent short-term relief followed by deterioration below or near placebo levels or baseline levels. The apparent superiority of PRP over corticosteroids is primarily a corticosteroid deterioration effect, not a PRP treatment effect. Ord makes another observation, and I really agree with this one, and it's sort of contrary to what we've been talking about, plays out the four possible positions a clinical guideline could take. So a clinical guideline could say, recommend both PRP and corticosteroids could say, recommend PRP but not corticosteroids. It could say or recommend corticosteroids but not PRP or recommend neither. He identifies which positions are logically coherent. The one position he calls implausible, the one most society guidelines currently take, recommend corticosteroids recommend against PRP. So his logic is, if PRP is an elaborate placebo, fine, recommend it.
Dr. Jared Powell:
If you don't recommend PRP, you must also not recommend corticosteroids because corticosteroids lose to the elaborate placebo in inverted commas I'm doing here, just in case you're wondering, head to head trials. So you cannot logically endorse the loser, which is corticosteroids project, the winner, which is PRP of the same comparison. Here is where I tend to agree with proponents of PRP. I think we should in most cases, for, for many musculoskeletal conditions, recommend PRP above and beyond corticosteroid injections. Corticosteroid injections don't do well long term. PRP alternatively is not harmful. So now I wanna spend a few minutes on the LinkedIn thread itself, and also happened on Instagram as well, because I think the reaction as always, these posts reveal something important about how evidence gets processed or doesn't in clinical communities. Responses fell into a few predictable categories. Most common was the dose defense.
Dr. Jared Powell:
The argument that PRP works at the right dose. These studies just use too little. As we've seen the highest dose, the American Journal of Sports Medicine study in 2026 showed an effect size of 0.02. The ODing papers dose response finding is confounded by comparator type dose. Hypothesis is testable and it is legitimate, and I think we should do the work. But as it currently stands, hasn't been tested against the right comparator. Several commenters accused me of not reading the paper, also making claims the numbers directly contradict made a false claim that the highest dose study, the American Journal of Sports Medicine study in 2026 was the only paper that showed PRP worked. This in fact was completely false. That study that used the highest dose did not work. Funnily enough, this comment was then copy and pasted and commented by many others without checking.
Dr. Jared Powell:
The data turned into a, to a full on orchestrated attack where everybody was comment copy and pasting the exact same comment. That was completely wrong. It was actually quite fun to watch and observe. One clinician reported a 100% success rate over three years using PRP combined with shock wave therapy. A 100% success rate should raise alarm bells and certainly not instill confidence. No treatment in the history of medicine as far as I'm aware and tell me if I'm wrong, works 100% of the time on every individual it's tested on. And remember that if you are combining two interventions, PRP and shockwave, you can't attribute outcomes to either one. Another commenter posted a registry of 2,200 elbows and they, and they found a 93% satisfaction with PRP. There was no control group. Tennis elbow resolves in about 80% of cases without a comparison or a control group.
Dr. Jared Powell:
These numbers are uninterpretable. It doesn't mean anything. It just means that 93% of people end up being satisfied down the line after receiving a PRP injection. It doesn't tell us that PRP caused the effect. Interestingly, this post that ended with a promotion for an orthobiologics conference, effectively this post was pure marketing and sadly not anything close to compelling evidence, but my favorite, favorite response was actually about my profession. Several commenters pointed out that I'm just a lowly physiotherapist. I'm not an interventionalist, and therefore I can't meaningfully comment on the evidence for PRP one compared me to a football fan criticizing a coach's decision, which I, I found quite funny. This is a genetic fallacy and what a genetic fallacy is where you dismiss an argument based on its source rather than its content. So it's not an ad hom attack, it's different to that. It's a genetic fallacy.
Dr. Jared Powell:
They're dismissing my argument because I don't do PRP injections. They're dismissing it because I'm communicating the findings. The ability to critically appraise a meta-analysis or any piece of evidence doesn't require you to perform the intervention being studied. Guys, statisticians don't do surgeries. Reviewers, the Cochrane database don't administer every treatment they evaluate. The entire framework of evidence-based medicine is built on the premise that evidence appraisal is a transferable and teachable skill that we should teach to most healthcare workers. So we can avoid all of this controversy of misinterpreting evidence. You could also argue the opposite. Someone who doesn't charge patients a thousand dollars for A PRP injection is probably better positioned to follow the data wherever it leads in an unbiased way. Okay, so before I wrap up, it's worth pausing on something that I think was visible throughout the entire thread, but that nobody really named explicitly almost every comment a defending PRP had in inverted regenerative medicine in their bio.
Dr. Jared Powell:
They also had a link to book an appointment for PRP in their bio, or a disclosed affiliation with an orthobiologics company or conference. One post defending PRP ended with a promotion for a paid conference. As I already mentioned, another came from a clinician at a clinic whose business model is built on ortho biologic injections. The paper most cited in P'S defense, the high dose paper from 2025, has authors declaring consulting fees and royalties from arthrex, one of the largest PRP kit manufacturers in the world. Now, it would be remiss to think that incentives don't shape perception. And when you charge patients 500 bucks to 1500 bucks out of pocket for a PRP injection, you have a structural motivation to interpret ambiguous evidence favorably. It's exactly why journals require conflict of interest declarations. It's why we insist on open science and pre-registered reports in adequate control groups.
Dr. Jared Powell:
These tools exist because smart, well-meaning clinicians can be pulled towards conclusions that serve their practice and do not correspond to the facts. Also, none of this is to blame or disparage patients. I understand completely the allure of healing your own tissue with your own platelets rather than a six month progressive loading program, especially when this is sold to you by industry experts with fancy titles, fancy offices, white lab coats, and who are running this elaborate injection therapy. So if I was a patient and I was uneducated or a layman when it comes to this stuff, as most patients are, they don't often see what goes on behind the curtain and they don't understand the science of it. I would be completely attracted to getting a PRP injection for my tennis elbow, for my rotator cuff tendinopathy, for my plantar faciopathy. So again, this is not to blame patients.
Dr. Jared Powell:
This is to point the finger at the clinical experts who are overpromising and over-delivering treatments that have little to no effect compared to a placebo. So let me be clear about what I am not saying. I do not want this argument to be misinterpreted, although I'm sure some deliberately will. That comes with the territory. I'm not saying PRP is useless for everything. I'm not saying the PRP dose hypothesis is wrong. I'm simply saying the current evidence for PRP in tennis elbow does not support its use over placebo. And the paper most often cited to argue otherwise is built on corticosteroid comparisons. John Orchard's rock paper, scissors framework is the simplest way to hold all of this together. PRP beats corticosteroids because corticosteroids get worse. Corticosteroids beat placebo because they work short term. Placebo ties PRP because PRP doesn't add much beyond the injection itself. Once you see these trajectories, the paradox dissolves.
Dr. Jared Powell:
The dose hypothesis deserves a proper trial. Let me be very clear about that. High dose PRP against saline, adequately powered pre-registered with functional outcomes. If that trial indeed shows benefit, I'll update my position in a heartbeat. That's how evidence works. I hope to be proven wrong because that's progress. That means we're getting somewhere. But until that trial exists, the claim that high dose PRP works for Tennis Elbow is a hypothesis dressed as a conclusion sustained by corticosteroid comparisons, registry data without control groups, and the financial incentives of the people making the claim. Alright, thanks for listening. I'll link all the papers in the show notes. I'll chat to you soon. The Shoulder Physio Podcast would like to acknowledge that this episode was recorded from the lands of the Ang people. I also acknowledge the traditional custodians of the lands on which each of you are living, learning, and working from every day. I pay my respects to elders past, present, and emerging, and celebrate the diversity of Aboriginal and Torres Strait Islander peoples and their ongoing cultures and connections to the lands and waters of Australia.
