THANK YOU FOR VISITING SWEATSCIENCE.COM!
As of September 2017, new Sweat Science columns are being published at www.outsideonline.com/sweatscience. Check out my bestselling new book on the science of endurance, ENDURE: Mind, Body, and the Curiously Elastic Limits of Human Performance, published in February 2018 with a foreword by Malcolm Gladwell.
- Alex Hutchinson (@sweatscience)
***
Platelet-rich plasma therapy was a super-hot topic a couple of years ago (see earlier blog posts about it), in part because of reports that top athletes like Tiger Woods were using it to speed their recovery from injuries. These days, the fuss has died down a bit. The novelty is gone, subsequent studies haven’t produced the “miracle” results promised by initial case reports, and maybe no one wants to emulate Tiger Woods anymore.
Anyway, studies of PRP continue to trickle in, but the picture isn’t necessarily getting much clearer. Two new studies have just been posted online at the American Journal of Sports Medicine, one of which is a randomized trial of PRP for tennis elbow by researchers in Greece. The design seems pretty good in theory: 28 patients were split into two groups; one group received an injection of PRP (their own blood, spun to produce plasma with elevated levels of healing-enhancing platelets), while the other group received an identical injection of their own unenriched blood. This should eliminate the problem of placebo effects (which are very big in invasive procedures that involve lots of needles), and test only whether the platelets themselves make any difference.
But there’s a problem:
This is a single-blind study. Patients were aware of the treatment because it was practically difficult to mask the process.
I don’t understand this. Maybe there’s something I’m missing — if you know why it would be “practically difficult” to mask the process, please let me know. It seems to me that all you have to do is put the blood-spinning machine in the room next door, and you’re in business with a double-blind study. But that’s not what they did — and to me, that’s an enormous problem, given how much publicity PRP has received over the past few years.
Anyway, the results: they measured subjective pain and perceived elbow function at various points over the next six months. There was only one case where the two groups showed statistically significant differences: pain was lower in the PRP group after six weeks, though the difference was no longer significant at the next measurement (3 months). On the other hand, if you ignore “statistical significance,” the trend was that the PRP patients did better in every measurement.
So how do you interpret these results? It’s pretty clear that the authors of the paper are big boosters of the technique:
[T]here is enough proof to support the superiority of PRP treatment over autologous blood, regarding pain, in the short term…
More studies on this topic could further enlighten aspects of this promising treatment…
In conclusion, we showed that PRP led to pain relief earlier than autologous whole blood, and we believe its application will be increasingly widened in the near future…
Really, I don’t think they showed any such thing. They found results that were statistically insignificant in five of their six outcomes, using two measurements that are largely subjective, in an experimental design that does nothing to eliminate placebo effect for one of the most heavily hyped sports medicine treatments of the past decade. To justify the cost and extra effort required for PRP therapy, they’re going to need more definitive results than that.
Agreed, and there is more: why did they test six outcomes unless they thought each was important? And if they are willing to trumpet significance in one particular outcome as support for the technique, would they not have done the same if any of the six outcomes had happened to be individually significant?
In that case, the correct test of significance should be against the joint null hypotheses over all outcomes, not the individual significance of each outcome. That is a much higher hurdle, and without knowing the details of the data or statistical analysis it is entirely possible that their experiment has no significant results whatsoever.
I continue to love the way you call it like it is — honest, on target.
Wait a minute, do I understand correctly that the health care professional administering the injection was blind to what he/she was injecting, but it was obvious to the patient?
The only practicality that could bring that about is very strict rules for informed consent.
Another possibility is that the patients were randomly selected after they had been treated by their own health care professional with either PHP or blood, and that only the guys doing he questionnaires were blind to the treatment condition, though that can hardly be called a randomized controled trial.
Agree with RH – what does single-blind mean if the patients were aware of the treatment?
Not sure if this answers your question, but after spinning the blood in a centrifuge, it’s the yellow blood plasma liquid that you take off the top and re-inject for PRP. Of course if you just re-inject the blood, it’ll be red. Maybe that’s the problem… although I’m not sure why you couldn’t hide the colour of the liquid that’s being re-injected.
Plasma (yellow) looks different from blood (red), so the practical difficulty may simply have been a lack of opaque bags and injection equipment certified for medical use, in which case both subject and clinician were aware of what was injected. I’d overlooked the “single-blind” claim; “reverse single-blind” seems more accurate in RH’s interpretation.
Thanks for the comments, folks — didn’t realize that PRP was yellow. That’s at least a plausible reason for saying it was “practically difficult” to mask the treatment. Though I still think it would have been worthwhile to make that extra little bit of effort to mask it, either by hiding the colour of the liquid — or even just by putting a blindfold on the patient!
As for what “single-blind” means in this context, it means the clinicians who assessed the progress of the injury before treatment and again after six weeks, three months and six months, were unaware of which treatment the patients received. Since the outcomes essentially involved asking the patients how much their elbow hurt and measuring range of motion, clinician expectations can bias the result, so that blinding is important (though certainly not sufficient if the patients themselves weren’t blinded).
I have been treating hundreds of patients with PRP almost 10 years. This study is seriously flawed for a number of reasons. First of all if PRP is done correctly it is far more successful than cortisone, or failed conservative treatment. For PRP treatments to work it should be done once every 4-8 weeks for 3-5 sessions or more if the person is progressing. Since PRP stimulates healing and new collagen growth stopping treatment afer one session can assure failure over the long haul or undertreatment. Second of all using whole blood is not a fair placebo since it also contains all of the platelets including red blood cells. Furthurmore the injections must be done under image guidance in either case to be certain the platelets or blood is placed in the scarred or torn tissue areas. There is another treatment called percutaneous tenotomy where under ultrasound guidance the tendon is pierced with a 22 guage needle upto 50 times. This study was done by Dr Nazarian from Thomas jefferson in Phila. He showed 75% improvement in one treatment. What this demonstrates is that if you can make the injured tissue bleed than platelets will be released which will than attract stem cells to heal the problem. This is a technique frequently used in Orthopedics. But with PRP we can now coerce and stimulate healing without damaging the tissue. Bottomline this was a poorly designed study done by peolple who do not have experience in doing this treatment properly. Edward Magaziner. MD from New Jersey, DREMagaziner.com