Platelet-Rich Plasma (PRP): Dosing, Evidence and Results

Platelet-rich plasma, or PRP, is an injection made from your own blood. A sample is drawn in the office, spun in a centrifuge to separate and concentrate the platelets, and the concentrate is placed into the injured joint, tendon or ligament, usually under ultrasound guidance. That much is fairly standard from clinic to clinic. What is not standard is what ends up in the syringe. Two practices can both call the product PRP and deliver platelet doses that differ by more than tenfold. This hub collects what Don Buford, MD has written about PRP across more than a decade of performing these procedures, measuring every one of them, and publishing on the dosing question.

Dose is the variable that decides whether PRP works

Platelet dose is volume multiplied by concentration. A 3cc injection of a highly concentrated product and a 6cc injection of a weak one are not the same treatment, which is why any study or marketing claim that describes PRP by volume alone tells you almost nothing. A rotator cuff meta-analysis published in Arthroscopy defined dose that way, and its negative conclusion is misleading for exactly that reason.

For tendon, ligament and muscle problems, our published review found that 14 of 14 studies delivering under 3.5 billion platelets in a single injection did not help patients, while 9 of 11 studies delivering above that threshold reported a positive response. In knee osteoarthritis, a well powered 2016 level 1 study of more than 100 patients used a product concentrated to roughly 1.7 times the baseline platelet count, which is a very different exposure from a high dose protocol. When PRP does not work, an under-dose is one of the first explanations worth ruling out.

We measure every dose we give

Our office runs its own lab, with multiple centrifuges and a hematology analyzer, and every PRP sample is quantified before it is injected. A recent knee patient's printout makes the point: a 53cc blood draw yielded 11.7 billion platelets in a 6cc injection, about eight times baseline concentration. We classify what we make using the DEPA system, because unlike older classifications it accounts for dose.

Measuring also reveals how much things move. In one patient, on one day, with the same technician and the same centrifuge, two draws thirty minutes apart produced a 12.43X product and then a 7.29X product. Even a validated hematology analyzer can report roughly a 10 percent swing on the same sample. Fewer than 5 percent of physicians offering PRP measure what they inject, so asking your doctor what your platelet dose was and how they know it is a reasonable question.

The kit matters, and so does the metric you compare kits on

We have bench tested about eight commercially available PRP systems using a standardized blood draw so their output can be compared directly. The number we watch is platelet recovery percentage, the share of the platelets in the drawn blood that actually end up in the injection. Unlike a concentration multiple, it cannot be inflated by squeezing the product into a tiny final volume or by treating a patient who happened to start with a high platelet count. If two protocols deliver the same platelet dose, the one that needs less of your blood to do it is the better protocol.

Where the evidence is strong, and where it is not

Knee osteoarthritis is the best supported use. More than 30 level 1 studies have compared PRP against corticosteroid, hyaluronic acid or placebo, with benefit reported out to a year. A 2015 analysis pooling 24 studies and 2,315 patients found pain and function improved through 6 months and were still better than before the injection at 12 months. In a 90 patient randomized trial of advanced bone-on-bone arthritis, three monthly PRP injections moved average pain scores from 8.0 to 5.7 at 6 months, while saline moved from 7.9 to 6.6.

Elsewhere the picture is mixed, and this practice says so. A 2023 level 1 study following 64 patients with lateral elbow tendinopathy for two years favored PRP over a steroid injection. A randomized trial in AJSM that gave 19 patellar tendinopathy patients a single leukocyte-rich injection found no strong evidence of benefit. PRP does not reattach a type 2 SLAP lesion in the shoulder and it will not stabilize a meniscus tear that is moving too much to heal, and in those situations the honest answer is an operation, sometimes with PRP added during the repair. More of the study-by-study reading sits in the research hub.

What PRP is not

There are no meaningful numbers of stem cells in circulating blood, and therefore none in platelet-rich plasma, whatever a podcast or a nearby clinic may claim. Cells come from bone marrow, which is a different procedure with different indications. PRP is also not classified as a drug or an HCT/P, a point that causes a lot of confusion about what physicians are permitted to offer and advertise, and the FDA and safety hub covers that ground. Responses vary from patient to patient, and no one should tell you otherwise.