PRP is the oldest modern regenerative biologic and still the most heterogeneous product in the field. Two vials labelled 'PRP' can differ by an order of magnitude in platelet count, leukocyte content, and growth factor concentration. This is the science of what actually drives the outcome.
What PRP actually is
PRP is autologous plasma concentrated to contain a platelet count several multiples above baseline peripheral blood. That is the definition — everything else (leukocyte content, red cell content, activation status, matrix composition) is a design choice made by the preparation system.
The classification you should be using
- P-PRP (leukocyte-poor PRP): low WBC content, gentler inflammatory profile. Favored for facial aesthetics, hair loss, and intra-articular use.
- L-PRP (leukocyte-rich PRP): higher WBC content, pro-inflammatory cytokine load. Favored for chronic tendinopathies and wound bed applications.
- P-PRF / L-PRF (fibrin matrix): after activation, the fibrin scaffold traps platelets and slowly releases growth factors over 7–14 days. Ideal for scaffolded delivery in scar remodeling, bone grafting, and wound care.
- PRP-PRFM: a hybrid preparation used widely in aesthetics — combines liquid PRP with a fibrin matrix component for both immediate release and sustained delivery.
The growth factors that drive the response
- PDGF-AA / BB: fibroblast and mesenchymal cell proliferation, matrix synthesis.
- TGF-β1 / β2: ECM production, chemotaxis of osteoblasts and fibroblasts.
- VEGF: angiogenesis, endothelial migration.
- EGF: keratinocyte and endothelial proliferation.
- IGF-1: anabolic signaling in muscle, tendon, cartilage.
- bFGF (FGF-2): broad mitogen for mesenchymal cell lineages.
The concentration and ratio of these factors — not just the platelet count — determine the biological effect. Two preparations with identical platelet counts can behave very differently if the activation status or matrix content differs.
Activation: exogenous or in situ?
In situ activation on contact with type-I collagen at the injection site releases growth factors gradually over the first 60 minutes. Exogenous activation with CaCl₂ or thrombin gives a bolus release. Which you choose depends on the target — bolus for aesthetic delivery and scar work; in situ for intra-articular and tendon injections.
Concentration: more is not always better
The dose–response curve for PRP is not linear. Below 3× baseline, effects are subtherapeutic. Between 4× and 7× is the operational sweet spot for most indications. Above 8× the effect plateaus and, in some tissue (cartilage, hair follicle), may even reverse due to inflammatory cytokine load.
Preparation systems affect biology
- Single-spin systems: higher red cell content, lower purity, moderate concentration. Cost-effective and fast.
- Double-spin buffy-coat systems: higher purity, tighter concentration control, moderate throughput. The most common workhorse in mature clinics.
- Automated closed systems: reproducible concentration, sterility, and leukocyte control. Higher consumable cost but tighter clinical results.
Evidence base at a glance
- Knee osteoarthritis: Level 1 evidence for symptomatic and functional improvement out to 12 months, superior to hyaluronic acid in most meta-analyses.
- Chronic tendinopathies: lateral epicondylitis, patellar tendinopathy — strong evidence, especially L-PRP preparations.
- Androgenic alopecia: multiple meta-analyses confirm improved hair density and count.
- Aesthetic skin quality: reliable improvement in texture and tone, particularly with microneedling or dermal-infusion delivery.
What PRP does not do
- Does not regenerate full-thickness cartilage defects.
- Does not reverse advanced hair follicle miniaturization or scarring alopecia.
- Does not treat structural volume loss — filler still does what PRP cannot.
How to read someone else's PRP claims
Ask three questions of any PRP protocol: What was the platelet concentration multiple above baseline? Was the preparation leukocyte-rich or leukocyte-poor? Was activation exogenous or in situ? If any of these cannot be answered, the outcome data attached to the protocol is uninterpretable.