It gets worse before it gets better, and that is the treatment working.

RECOVERY

The single most common reason people conclude PRP did not work is that they judged it at two weeks. This page sets out what actually happens, in what order, and why the early part feels like a mistake.

Why it gets worse before it gets better

PRP is not an anesthetic and not an anti-inflammatory. It is the opposite: a deliberate provocation. The platelet concentrate, plus the needle passing repeatedly through degenerate tissue, creates a controlled acute injury in a structure that had settled into a chronic non-healing state.

The soreness of the first few days is the inflammatory phase doing exactly what it is supposed to do. Neutrophils and macrophages arrive, debris is cleared, and the signaling that recruits repair cells begins. Suppressing that phase with anti-inflammatories undermines the treatment, which is why we ask you to hold NSAIDs.

Days 1 to 4 — the flare

Expect the treated area to hurt more than it did before. In a joint, particularly a knee or hip, this is more noticeable because a closed capsule has nowhere to accommodate swelling. In a tendon it is usually a deep ache with local tenderness.

• Acetaminophen is generally fine. NSAIDs and ice are avoided.
• Move normally within comfort. Complete rest is not the goal and stiffens things.
• A day or two off heavy work is reasonable if your job loads the treated area.
• Contact us for fever, spreading redness, or pain that escalates rather than settles after day three — that pattern is not expected.

Weeks 1 to 2 — the quiet part

The flare settles and very little appears to happen. Function is roughly where it started. This is where most people privately conclude it has failed.

Biologically the proliferative phase is underway: fibroblasts are being recruited, new vessels are forming, and type III collagen — the disorganized, provisional kind — is being laid down. It has no strength yet and no organization. There is nothing to feel because there is nothing yet built.

Weeks 3 to 6 — the first real signal

Most people notice something here. Often it is not less pain at rest but a specific activity that has stopped provoking it — stairs, a grip, the first steps in the morning. That is the more meaningful signal, because it reflects load tolerance rather than mood or weather.

This is also when the loading program matters most. Collagen remodels along lines of mechanical stress. Provisional tissue that is never loaded organizes badly and produces a structure no stronger than the one it replaced. The exercise is not rehabilitation in the recuperative sense; it is instruction to the tissue about how to arrange itself.

Months 2 to 3 — the honest assessment point

Three months is when we know. Type III collagen is being replaced by stronger, better-aligned type I. If nothing at all has changed by now, a second identical injection is unlikely to change that, and the conversation should turn to whether the diagnosis was complete, whether the loading actually happened, and whether the metabolic terrain was ever addressed.

Months 3 to 6 — where most of the gain arrives

Remodeling continues well past the point where most people stop paying attention. Tendon in particular keeps improving to six months and sometimes beyond. People who return at twelve months frequently report they kept improving after they had stopped noticing.

How this differs by tissue

• Knee and hip joints. Larger early flare; first change commonly weeks four to eight; benefit often peaks around three to six months.
• Elbow tendons. Reliable pattern — little at two weeks, clear signal at six, substantial gain at three months, continued to a year.
• Plantar fascia. Slower and less linear. First-step pain is often the last symptom to leave.
• Achilles. The slowest structure on this list. Months, and impact returns last.
• Rotator cuff. Night pain often improves before range does, which people find odd but is typical.

What slows recovery down

Some of this is modifiable and some is not, and it is better to know which is which before you start.

• Poor glycemic control. Glycated collagen is weaker collagen. This is the single largest modifiable factor and the least discussed.
• Smoking and nicotine. Vasoconstriction in tissue whose problem is already poor perfusion.
• Fragmented sleep. Most repair signaling is nocturnal, and sleep debt raises inflammatory tone.
• Returning to full load early. Feeling better precedes being strong by weeks, and that gap is where re-injury happens.
• Skipping the loading program. The most common cause of a disappointing result in people whose diagnosis was correct.
• Anti-inflammatories through the early phase. Directly opposed to the mechanism.

When a second injection makes sense

Partial response at three months is the clearest indication: something started, and more signal is a reasonable bet. No response at all is a weaker case, and repeating an identical treatment without changing anything is not a plan. Full response needs nothing further.

What we ask of you

Three things, none of them complicated. Do the loading program at the doses given rather than the ones that feel productive. Hold the anti-inflammatories through the window we specify. And come back at three months whether it worked or not — the failures are more informative than the successes, and we want to know.

How to tell real progress from a good week

Pain is a noisy measure. It moves with sleep, stress, weather and what you did the day before, which makes it a poor instrument for judging a treatment that works slowly. People who track pain alone tend to conclude either that it worked brilliantly or not at all, depending on which week they were asked.

Load tolerance is the better signal. Pick two or three specific, repeatable things at the start — the number of stairs before it complains, whether you can carry shopping in that hand, how far into the morning the first-step pain lasts — and check them monthly rather than daily. Those move in one direction when tissue is genuinely changing, and they do not swing with a bad night.

Daily monitoring has a second cost worth naming: attending closely to a body part several times a day tends to increase how much it hurts, through entirely ordinary mechanisms of attention and threat appraisal. Checking monthly is better data and, for most people, a better three months.

Where this sits against pain medication

The reason the timeline matters for medication is straightforward: a treatment that takes three months to declare itself is a treatment people abandon at week three, and what usually fills that gap is an increase in whatever they were already taking. Knowing that week two is supposed to feel like nothing is, in practice, one of the more useful pieces of stewardship on this website.

What people ask while waiting

How soon can I go back to work?

Most people work the next day; jobs that heavily load the treated area may need two or three. What to expect from the visit.

Can I take ibuprofen for the flare?

We ask you not to during the early window, because it opposes the mechanism. Acetaminophen is usually fine. Risks and side effects. Blunting the flare works against the mechanism you paid for.

It has been four weeks and nothing has changed. Has it failed?

Almost certainly not. Four weeks is early. Three months is the assessment point. How we assess candidacy. Three things vary, and only one of them is the treatment.

How many injections will I need?

Commonly one to three depending on tissue and response. How PRP works. There is no magic in a series of three.

Related reading

• Injury and work comp
• Platelet lysate recovers differently
• What to expect from the visit
• Risks and side effects
• Am I a candidate?
• How PRP works

How long the change holds is a separate clock from how long it takes to arrive.