The one place where the twelve-month data goes against the steroid.

PRP FOR LATERAL EPICONDYLITIS

Lateral epicondylitis is misnamed twice over. Most people who have it have never played tennis, and by the time it has lasted more than a few weeks it is not an -itis at all.

It is not inflammation, which changes everything

Tissue taken from chronic lateral elbow pain does not show inflammatory cells. It shows angiofibroblastic tendinosis: disorganized collagen, an abnormal ingrowth of small vessels and nerve fibers, and an absence of the tidy repair architecture you would expect. The tendon has not been inflamed for months. It has been failing to finish a repair.

This single fact explains why the standard sequence disappoints so many people. Rest, ice and anti-inflammatories are aimed at inflammation that is no longer present. A corticosteroid injection reliably helps for six to eight weeks and, in the longer follow-up studies, leaves people worse at a year than doing nothing at all — because suppressing an already-absent inflammatory signal removes what little repair drive remains.

Why the repair stalled

The common extensor origin is a small, poorly vascularized footprint asked to transmit large repetitive loads. That is the mechanical setup. Two biological factors decide whether it recovers: local microvascular supply, which is measurably worse in smokers and in people with insulin resistance, and collagen quality, which degrades with glycation and with the chronic low-grade inflammatory state that accompanies metabolic disease. A tendon repairing with poorly cross-linked collagen in a poorly perfused footprint is a tendon that will take the long route.

The third factor is occupational and nobody controls it clinically. Electricians, hairdressers, line cooks, plumbers, anyone gripping and rotating all day — they cannot rest the arm, because resting the arm means not working. The tendon is re-injured before it consolidates, over and over, for reasons that have nothing to do with compliance.

What PRP does here

This is one of the better indications for PRP, precisely because the problem is a stalled repair rather than an inflamed one. Delivering a concentrated growth-factor signal into a degenerate tendon is a direct attempt to restart the cascade that stopped: recruiting fibroblasts, driving angiogenesis into a hypovascular zone, and prompting collagen deposition that then needs months of graded loading to organize.

The needle itself contributes. Fenestration — deliberately passing the needle through the degenerate tissue several times — creates a small controlled injury, which is part of the point. You are converting a chronic non-healing lesion into an acute one that the body still knows how to handle.

What the evidence says

Lateral epicondylitis is where PRP has some of its most durable comparative data. Multiple randomized comparisons against corticosteroid show the same crossover pattern: steroid ahead at four to eight weeks, PRP ahead from around three months, and the gap widening at one and two years. That shape is consistent with what each treatment is doing biologically, which is a reasonable reason to trust it.

It is not unanimous. Some trials find PRP no better than a saline injection, and saline injections themselves are not inert when the needle is fenestrating tissue. The honest summary is that PRP performs at least as well as the alternatives at a year and better than steroid, and that a share of the benefit may come from the procedure as much as the preparation.

What treatment looks like

Usually one injection, sometimes a second at six weeks. Ultrasound guidance is used to find the degenerate portion of the tendon rather than the point that is most tender, which is not always the same place. Expect the elbow to be sore for three to five days afterward.

The loading program is not optional and is not a handout. Eccentric and heavy slow resistance work through the extensor mass is what organizes the collagen the injection prompted. Without it you have prompted a repair and given it no instruction about how to arrange itself. A counterforce brace can offload the origin while you work, and a grip assessment often finds the actual culprit — frequently a wrist and shoulder that stop contributing, leaving the elbow to absorb everything.

What else causes pain on the outside of the elbow

Roughly one in five people sent for a lateral elbow injection has something other than simple tendinosis driving the pain, and injecting the wrong structure is how a treatment gets blamed for failing at a job it was never given.

• Radial tunnel syndrome — compression of the posterior interosseous nerve, tender three to four centimeters distal to the epicondyle rather than on it, often a deeper ache that disturbs sleep
• Posterolateral rotatory instability — usually follows a previous injury or repeated steroid injection, and presents with apprehension rather than point tenderness
• Radiocapitellar joint disease — arthritic or a plica, producing catching and pain through rotation arcs
• Cervical referral — a C6 radiculopathy that presents laterally, with a neck that has been quietly stiff for years

Examination and ultrasound sort most of this out in the room. Where it stays ambiguous we say so rather than inject hopefully.

The loading program, in more detail

The injection prompts a repair. Load tells that repair how to organize, and a tendon healing without mechanical instruction lays down disorganized collagen — which is the problem you started with.

• Weeks 0–2 Relative rest from provocative gripping. Pain-free range and light isometrics, which have a genuine analgesic effect on tendon pain.
• Weeks 2–6 Eccentric wrist extension, slow, into mild discomfort but not through sharp pain. Volume matters more than intensity here.
• Weeks 6–12 Heavy slow resistance. This is the phase most people skip and the one that does the structural work.
• Beyond 12 weeks Return to full grip loading, with the upstream chain addressed — shoulder external rotation and scapular control, because an elbow often fails as the last link in a sequence that stopped working further up.

Repeating a treatment that failed is not a plan

A share of people do not respond. When that happens the first question is whether the diagnosis was right, the second is whether the loading actually happened, and the third is whether the metabolic terrain was ever addressed. Only after those does it make sense to discuss a second injection, a different preparation, or a surgical opinion. Repeating an identical treatment that did not work the first time, without changing any of the inputs, is not a plan.

What people ask after a steroid injection wore off

How is this different from a cortisone shot?

Cortisone suppresses; PRP attempts to restart. Cortisone wins the first two months and loses the year. The comparison in full.

Do I have to stop working?

Usually not, but the grip and rotation load has to be modified for a period, and we will be specific about what that means for your actual job. What to expect.

What if it is the inside of my elbow?

That is medial epicondylitis, the same pathology at the flexor-pronator origin, and it is treated on the same principles. Golfer’s elbow.

How long until I know?

Six weeks for the first signal, three months for the real answer. Recovery timeline.

Related reading

• Golfer’s elbow
• PRP compared with cortisone
• How PRP works
• Recovery timeline

One honest gap worth naming: the systemic evidence at the elbow is thinner than at the Achilles or the rotator cuff, where we can quote effect sizes. The general finding that tendinopathy is markedly more common in people with diabetes is set out on the tendinopathy page, and it is where the numbers live rather than being repeated here as though the elbow had its own.