Tendinopathy: why tendons hurt, and how they actually heal
A painful tendon is rarely "inflamed" in the classic sense. Most chronic tendon pain is tendinopathy: a failed-healing, partly degenerative change in the tissue, which is why the old word "tendinitis" is usually wrong and why anti-inflammatory rest alone tends to fail. Tendons are living, load-sensing tissue, and the single most reliable way to make them stronger and less painful is the opposite of rest: progressive, patient loading over weeks to months.
Key facts
- Biopsy and imaging studies find little or no classic inflammation in chronic tendinopathy; the dominant picture is disorganised collagen, increased ground substance and failed repair, hence the shift from "tendinitis" to "tendinopathy".1
- Cook and Purdam's continuum model describes three overlapping stages: reactive, dysrepair, and degenerative, with reversibility most likely early and least likely late.2
- In Alfredson's landmark 1998 study, 12 weeks of heavy eccentric calf exercise returned all 15 patients with chronic Achilles tendinosis to running, while the conventionally rested comparison group all ended up in surgery.3
- A single bout of isometric loading cut patellar tendon pain from about 7/10 to near zero for at least 45 minutes in Rio's 2015 crossover trial.5
- Corticosteroid injection often helps for a few weeks but is worse than exercise at 6 to 12 months across several tendons.6
On this page
- Why "tendinitis" is the wrong word
- The continuum: reactive, dysrepair, degenerative
- Where it happens and how it feels
- Why rest alone fails
- The loading evidence: eccentrics, heavy slow resistance, isometrics
- Injections, shockwave and other add-ons
- Realistic timelines and load management
- What to do next
Why "tendinitis" is the wrong word
The suffix "-itis" means inflammation, and for decades a sore Achilles or tennis elbow was called tendinitis on the assumption that inflammatory cells were the problem. When researchers looked at tissue removed from chronically painful tendons, that assumption largely collapsed. Instead of the neutrophils and macrophages of acute inflammation, biopsies showed disorganised collagen, excess water-attracting ground substance, in-growth of new vessels and nerves, and cells under stress, with classic inflammatory cells scarce or absent.1 This is degeneration and failed healing, not a fire to be put out.
That distinction reshapes treatment. If the problem were inflammation, rest and anti-inflammatories would be the cure. Because the problem is a tendon that has failed to remodel under load, the logical fix is to give it the right kind of load so the resident cells rebuild stronger tissue. The modern umbrella term is tendinopathy (pain plus impaired function in a tendon), with "tendinosis" reserved for the degenerative tissue change seen on imaging. The honest nuance: low-grade inflammatory signalling does exist early and around the tendon sheath, so inflammation is not zero, it is simply not the main driver of the chronic, painful, thickened tendon most people present with.2
The continuum: reactive, dysrepair, degenerative
The most useful mental model comes from physiotherapy researchers Jill Cook and Craig Purdam, who in 2009 proposed that tendon pathology sits on a continuum rather than in tidy boxes.2 A tendon can move along it in either direction depending on how it is loaded, which is what makes early intervention worthwhile.
- Reactive tendinopathy. A short-term response to a sudden spike in load (a new sport, a long walk in new shoes, a return after a layoff). The tendon thickens to protect itself and the matrix takes on water, but the collagen is largely intact. Genuinely reversible if load is managed.
- Tendon dysrepair. Failed healing. With continued overload the changes deepen, collagen begins to separate and disorganise, and new vessels grow in. There is more breakdown than the tendon can repair. Some reversal is still possible.
- Degenerative tendinopathy. Areas of cell death and disorganised, mechanically useless tissue, often with a thickened, nodular tendon. This is the classic chronic presentation in older or heavily loaded tendons. The degenerated portion does not reliably reverse, but it usually sits beside healthy tendon that can be strengthened to take over the load.
That last point is the practical heart of rehabilitation. You are usually not trying to "heal a hole" in a degenerative tendon, but to train the surrounding good tissue to be strong and stiff enough that the whole structure tolerates load again, even if the degenerated patch persists on a scan. It also explains a confusing fact: ultrasound and MRI findings correlate poorly with pain. Plenty of people have thickened, ugly-looking tendons and no symptoms, so imaging is for ruling out other problems, not for grading how much it should hurt.2
Evidence strength, plainly. The continuum model is a well-cited framework, supported by histology and imaging, not a law of nature, and researchers have since debated and refined it. The treatment that flows from it (progressive loading) is the part backed by randomised trials. Treat the staging as a useful map, and the loading evidence as the destination.2
Where it happens and how it feels
Tendinopathy clusters at the tendons that take repeated, high tensile or compressive load. The classic pattern is load-related pain: stiff and sore first thing in the morning, warms up with activity, then aches afterwards and the next day. Pinpoint, localised tenderness on the tendon is common; widespread or burning pain points elsewhere.
| Site | Common name | Typical sufferer | First-line loading idea |
|---|---|---|---|
| Achilles (mid-portion) | Achilles tendinopathy | Runners, midlife "weekend" athletes | Heavy calf raises (eccentric or heavy slow resistance) |
| Patellar | Jumper's knee | Volleyball, basketball, jumping sports | Heavy slow resistance squats; isometrics for pain |
| Rotator cuff (shoulder) | Cuff-related shoulder pain | Overhead athletes, manual workers, older adults | Progressive cuff and scapular loading |
| Gluteal (hip) | Greater trochanteric pain | Women over 40, runners | Hip abductor loading; avoid prolonged compression |
| Lateral elbow | Tennis elbow | Manual workers, racquet sports | Progressive wrist extensor loading |
| Medial elbow | Golfer's elbow | Throwers, golfers, gym lifters | Progressive wrist flexor loading |
A recurring theme in the insertional tendons (gluteal, insertional Achilles, proximal hamstring) is compression. Where a tendon wraps over bone, pressing it against that bone (crossing your legs, lying on the painful hip, deep end-range stretching) aggravates it. That is why advice for gluteal tendinopathy includes avoiding provocative positions as well as loading, a point the LEAP trial built into its education programme.8
Why rest alone fails
Rest is intuitive: it hurts, so stop using it. The trouble is that tendons adapt to the load they receive, and the absence of load is itself a signal. Take load away completely and the tendon does not heal stronger, it de-conditions: it loses stiffness and load tolerance, so when you return to your old activity the now-weaker tendon is overwhelmed faster than before. This is the boom-and-bust cycle that traps people for years.
The biology reinforces this. Mechanical loading is the primary stimulus that tells tenocytes to make and organise new collagen, and collagen synthesis rises for one to three days after a meaningful bout.4 Remove the stimulus and you remove the repair signal. Tendon collagen also turns over slowly, so the tissue adapts over months, not days, which is why a short rest does nothing useful and a sudden return re-injures it.4 The aim is not no load, it is the right load. For the underlying tissue chemistry, our companion piece on collagen, tendons and joints goes deeper on why loading, not a supplement, does the heavy lifting.
The loading evidence: eccentrics, heavy slow resistance, isometrics
This is where tendinopathy has unusually good evidence for a musculoskeletal condition. Three loading strategies dominate, and they are complementary rather than competing.
Eccentric exercise: the original breakthrough
The modern era began with Hakan Alfredson, a Swedish orthopaedic surgeon with his own Achilles tendinopathy. On a surgery waiting list, he loaded the tendon hard with eccentric heel drops (lowering slowly under load), expecting it to rupture so he could have his operation sooner. It got better instead. His 1998 study in the American Journal of Sports Medicine formalised the protocol: 3 sets of 15 repetitions, twice daily, for 12 weeks, performed into some discomfort. All 15 chronic patients returned to full running, while the conventionally treated comparison group of 15 all ended up needing surgery.3 Small and uncontrolled by modern standards, but the effect was dramatic and reproducible, and it overturned the rest-and-inject orthodoxy.
Heavy slow resistance: simpler, equally effective
The twice-daily Alfredson programme is demanding and adherence suffers. Danish researchers led by Mads Kongsgaard and Rikke Beyer tested heavy slow resistance (HSR): heavy weights moved slowly through both the lifting and lowering phase, done only about three times a week. In patellar tendinopathy, HSR matched eccentric training on outcomes but produced greater patient satisfaction.7 A 2015 Achilles trial by Beyer and colleagues found HSR and eccentric training gave similar results at 12 weeks, again with better satisfaction for HSR.9 The practical message: a heavy, slow, three-times-weekly gym programme is as good as the classic eccentric protocol and easier to stick to.
Isometrics: a tool for pain
Holding a contraction without movement (isometric loading) can give immediate pain relief. In Ebonie Rio's 2015 crossover trial, co-authored with Cook and Purdam, a single bout of isometric quad contractions dropped patellar tendon pain from about 7/10 to near zero, and the relief lasted at least 45 minutes, alongside reduced cortical inhibition.5 That trial was tiny (6 athletes) and later work shows the effect is real but variable and not always superior to other loading. So isometrics are best seen as a useful in-season pain tool, for example before a game or on a flare day, not a stand-alone cure. The structural rebuilding still comes from progressive heavy loading.
"Should it hurt?" Some discomfort during tendon loading is acceptable and does not mean harm. A widely used rule is to keep pain at or below about 3 to 4 out of 10 during the exercise, and to check that it settles back to baseline by the next morning. Pain that climbs above that, or that is clearly worse the following day for several days, means the load was too much: reduce it, do not stop entirely. Sudden, sharp pain, a snapping sensation, or sudden loss of strength needs urgent assessment to rule out a tendon tear.
Injections, shockwave and other add-ons
People understandably want a faster fix than months of loading. The evidence on the popular add-ons is sobering and worth knowing before you pay for any of them.
Corticosteroid injection. The most important finding in tendinopathy treatment may be that steroid injections trade short-term gain for long-term harm. Brooke Coombes, Bill Vicenzino and colleagues pooled the randomised trials in a 2010 Lancet systematic review: corticosteroid reduced pain well in the short term (weeks), but at intermediate and long-term follow-up it was worse than no injection or than exercise, with higher recurrence.6 The classic example is tennis elbow: in Bisset's 2006 BMJ trial of 198 people, the steroid group did best at 6 weeks but had a 72% recurrence and the worst 1-year outcomes, while physiotherapy and even "wait and see" overtook it.10 A steroid shot can be a reasonable bridge for someone who cannot function, but it is not a cure and may slow true recovery.
Platelet-rich plasma (PRP). Injecting a concentrate of your own platelets is heavily marketed and the evidence is genuinely mixed: some trials find no benefit over placebo or saline, while several recent meta-analyses in lateral elbow tendinopathy suggest PRP may beat corticosteroid for long-term pain and function despite doing worse early.11 Wide variation in preparation methods and high risk of bias mean it cannot yet be recommended as routine. It is expensive, rarely NHS-funded for tendinopathy, and should be considered experimental.
Extracorporeal shockwave therapy (ESWT). Focused acoustic pulses to the tendon. The evidence is best for calcific rotator cuff tendinopathy, where high-energy shockwave can reduce pain and help break down calcium deposits, and is reasonable for some chronic cases (plantar fasciopathy, sometimes patellar or insertional Achilles).12 It is much weaker for non-calcific cuff problems, and NICE supports its use for several tendinopathies under normal consent and audit arrangements. Treat it as an adjunct to loading, not a replacement.
The consistent thread: in head-to-head trials, structured exercise wins over time. The LEAP trial randomised 204 people with gluteal tendinopathy to education plus exercise, a single steroid injection, or wait-and-see; education plus exercise gave the best outcome and pain at both 8 weeks and 1 year.8 To sanity-check any injectable or supplement against the rest of your routine, our stack builder can help.
Realistic timelines and load management
Honesty about time is the kindest thing anyone can offer here. Tendinopathy is slow, and unrealistic expectations are what drive people into the rest-and-relapse trap or onto an injection that backfires. Rough timelines, with wide individual variation:
- First 2 to 6 weeks: pain often starts to settle and morning stiffness eases, but the tendon is not yet structurally stronger. Isometrics can take the edge off pain in this window.
- 6 to 12 weeks: the timeframe of the main loading trials. Meaningful gains in pain and function for many people, with the tendon beginning to remodel.
- 3 to 6 months and beyond: realistic horizon for a stubborn or degenerative tendon, especially the Achilles, patellar and gluteal tendons. Full return to high-level sport can take longer still.
Two principles make or break recovery. First, progression: the load has to keep increasing as the tendon adapts, or improvement stalls. Second, load management: tendinopathy is fundamentally load exceeding capacity, so alongside building capacity you have to control the spikes by avoiding sudden jumps in volume, building back gradually after time off, and respecting the 24-hour pain response. The general-health picture matters too: type 2 diabetes and high cholesterol are linked to worse tendon health.4
What to ask your GP or physiotherapist
- Is this genuinely a tendinopathy, or could it be a tear, referred pain, or an inflammatory arthritis that needs different treatment?
- Which specific loading programme fits my tendon and my life: eccentric, heavy slow resistance, or a mix, and at what dose?
- If a steroid injection is offered, what does the evidence say about my 6 to 12 month outcome, not just the next month?
- What pain level is acceptable during my exercises, and what would tell me I am overdoing it?
- Can I be referred to NHS physiotherapy, and is shockwave available locally if loading plateaus?
What to do next
References
- Khan KM, Cook JL, Kannus P, Maffulli N, Bonar SF. Time to abandon the "tendinitis" myth. BMJ. PMC1122271, 2002.
- Cook JL, Purdam CR. Is tendon pathology a continuum? A pathology model to explain the clinical presentation of load-induced tendinopathy. Br J Sports Med. PMID 18812414, 2009.
- Alfredson H, Pietila T, Jonsson P, Lorentzon R. Heavy-load eccentric calf muscle training for the treatment of chronic Achilles tendinosis. Am J Sports Med. PMID 9617396, 1998.
- Magnusson SP, Langberg H, Kjaer M. The pathogenesis of tendinopathy: balancing the response to loading. Nat Rev Rheumatol. PMID 20697404, 2010.
- Rio E, Kidgell D, Purdam C, Gaida J, Moseley GL, Pearce AJ, Cook J. Isometric exercise induces analgesia and reduces inhibition in patellar tendinopathy. Br J Sports Med. PMID 25979840, 2015.
- Coombes BK, Bisset L, Vicenzino B. Efficacy and safety of corticosteroid injections and other injections for management of tendinopathy: a systematic review of randomised controlled trials. Lancet. PMID 20970844, 2010.
- Kongsgaard M, Kovanen V, Aagaard P, et al. Corticosteroid injections, eccentric decline squat training and heavy slow resistance training in patellar tendinopathy. Scand J Med Sci Sports. PMID 19793213, 2009.
- Mellor R, Bennell K, Grimaldi A, et al. Education plus exercise versus corticosteroid injection use versus a wait and see approach on global outcome and pain from gluteal tendinopathy: prospective, single blinded, randomised clinical trial (LEAP). BMJ. PMID 29759865, 2018.
- Beyer R, Kongsgaard M, Hougs Kjaer B, Ohlenschlaeger T, Kjaer M, Magnusson SP. Heavy slow resistance versus eccentric training as treatment for Achilles tendinopathy: a randomized controlled trial. Am J Sports Med. PMID 25979856, 2015.
- Bisset L, Beller E, Jull G, Brooks P, Darnell R, Vicenzino B. Mobilisation with movement and exercise, corticosteroid injection, or wait and see for tennis elbow: randomised trial. BMJ. PMC1633771, 2006.
- Niemiec P, Szyluk K, Balcerzyk-Matic A, et al. Effectiveness and safety of platelet-rich plasma for chronic lateral epicondylitis: meta-analysis of randomised controlled trials. (see also: PRP for tendinopathy systematic review). PMID 38357713, 2024.
- National Institute for Health and Care Excellence (NICE). Extracorporeal shockwave therapy for calcific tendinopathy of the shoulder (IPG742) and related guidance. NICE IPG742.
This article is educational and does not constitute medical advice, diagnosis, or a treatment recommendation. Medication uses described as “off-label” are not licensed for that purpose in the UK and should only be considered under qualified clinical supervision. Always speak to your GP, pharmacist, or a registered specialist before starting, stopping, or changing any treatment. If you have severe or alarm symptoms - unintentional weight loss, blood in your stool, difficulty swallowing, persistent vomiting, a fever, or severe pain - seek urgent medical care.