RightRep EN · PT Open the app
BLOG

Does Exercise Help Tendonitis? The Tendon Audit: What Heavy Loading Actually Does to the Tissue You Were Told to Rest, the 44.6%-of-Volleyball-Players File Next to the 0%-of-Cyclists File, the Founding Trial Where All 15 Eccentrically-Loaded Athletes Went Back to Running While Every Conventionally Rested Comparison Ended Up in Surgery, the Injection That Wins Week 4 and Loses Year One 54%-to-12%, the Biopsy Arm Where Twelve Weeks of Heavy Slow Resistance Rebuilt Fibril Density 70% Toward Normal, the Isometric Pain Switch at 7.0 to 0.17, the RCT Where Six Weeks of Rest Bought Nothing, and the Five-Year Honesty Where Only 39.7% Were Fully Pain-Free

By RightRep AI Assistant · · 21 min read

Every tendon pain arrives with the same prescription, issued with total confidence by someone who has never read a tendon trial: rest it, ice it, take the anti-inflammatories, and wait for the inflammation to settle. The tendon, in this picture, is a rope that frayed — and the only sane policy for a fraying rope is to stop pulling on it. The gym version adds a moral layer: you did this to yourself with bad form or too much volume, so penance means time off. It all sounds like physics. It was also, for most of the twentieth century, the actual medical advice. And it is the one thing the evidence file has dismantled most completely — because the painful tendon is almost never inflamed in the way the prescription assumes, rest measurably deconditions the tissue, and the intervention that beats everything else in the trials is the one the folklore forbids: loading it, heavily, slowly, and progressively. This library has already audited what training does for the back, the joints and the bones — the connective-tissue tour now reaches its third and final tissue: the tendon, the one whose injury folklore is not just wrong but backwards. Does exercise help tendonitis — and can you keep training through it?

The evidence answers in three acts. Act one is the verdict: in the founding 1998 trial, twelve weeks of heavy-load eccentric calf training returned all fifteen patients to pre-injury running — pain down, calf strength normalized — while the comparison group treated conventionally (rest, NSAIDs, orthoses, physiotherapy) printed the sentence that ends the rest argument: "in no case was the conventional treatment successful, and all patients were ultimately treated surgically" [5]. Act two is the inversion: the corticosteroid injection that feels like a miracle at week four is the worst performer at year one — 83% vs 96% success against a placebo injection, with 54% vs 12% recurrence in the tennis-elbow factorial trial [8] — while loading programs held their gains at six months, a year, and five years [7][9][12]. Act three is the honest ceiling: this is not a fast fix — jumper's knee symptoms in elite sport run a 32-month average duration [1], only 39.7% of Achilles patients were completely pain-free five years after the famous heel-drop program [12], and the pain is managed through loading, not deleted by it. Every number below comes from an abstract that was read directly and checked against its claim.

The short answer: yes — load is the treatment, and rest is the risk

Strip the folklore away and the claim "exercise helps tendonitis" has a checkable shape. If it's true, randomized groups assigned to load the tendon should do better than groups assigned to protect it; the effect should survive across tendons (Achilles, patellar, elbow) and programs (eccentric, heavy slow resistance); the wins should show up not just in pain scores but in the tissue itself; and the alternatives — rest and anti-inflammatory injections — should be measurable against it in head-to-heads.

Every one of those predictions is in the file. Loading groups beat and matched everything they were compared against: eccentric training against conventional care [5], continued sport against active rest [6], heavy slow resistance against corticosteroid injection [7]. The wins replicated across body parts and decades [5][7][9]. The tissue itself changed under load — tendon thickness and neovascularization fell [9], collagen turnover rose [7], and biopsy-measured fibril morphology moved toward healthy architecture [10]. And the alternatives lost on the clock: rest bought six weeks of nothing [6], and the injection's week-four win inverted into a year-one loss [7][8]. The audit below walks each pillar, then the complications the folklore never mentions: the vocabulary error underneath the whole prescription, the stages where the model changes, the in-season pain tool, and the five-year honesty about what loading does not fix.

The burden file: 44.6% of volleyball players, 0% of cyclists

First, the size of the thing — and the exposure signature that explains it. The cross-sectional study of 613 Norwegian elite athletes across nine sports found current jumper's knee (patellar tendinopathy) in 14.2% overall (87 of 613), with a range across sports from zero to 45%: highest in volleyball at 44.6% (±6.6%) and basketball at 31.9% (±6.8%), and zero cases in cycling and orienteering [1]. Men carried it at 13.5% versus women's 5.6% [1]. Read the gradient like an engineer: the sports that load the tendon with maximal, explosive, repetitive tensile spikes — jumping and sprinting — top the table, and the sports that shield it behind a pedal circle don't even appear on it.

Two more numbers from the same study do the long-term framing. The athletes with current jumper's knee had been symptomatic for 32 (±25) months on average — years, not weeks — with a function score of 64 (±19) on the 0–100 VISA scale [1]. This is not a minor nuisance that a weekend of ice solves; it is one of the most persistent overuse injuries in sport, and its persistence is the first clue that the resting prescription misunderstands the tissue. A condition that lasts 32 months under standard care is a condition whose standard care isn't working.

The vocabulary correction: it isn't "-itis" — and the prescription was written for a diagnosis the tissue doesn't carry

The word "tendonitis" is itself the first finding. The International Scientific Tendinopathy Symposium Consensus — the field's formal terminology agreement, 25 specialists from nine countries — recommends "tendinopathy" as the clinical umbrella term precisely because the "-itis" suffix claims inflammation, and inflammation is not what the microscope shows in the chronic painful tendon [2]. The consensus keeps "-itis" labels for genuine inflammatory presentations and histology, not for the clinic — which retires, in one vocabulary move, the entire rest-ice-NSAID rationale: you cannot anti-inflame a condition that is not primarily inflamed [2][4].

What the microscope shows instead is a loading-history story. The continuum model that organized the modern field describes three stages along one pathology: a reactive tendon (acute overload response, swelling, pain, but reversible — the classic too-much-too-soon presentation), tendon dysrepair (failed healing, matrix breakdown with some collagen disorganization), and degenerative tendinopathy (cell death, matrix disorganization, vessels and nerves growing where they don't belong — the 32-month veteran of the burden file) [3]. The model's point is not academic: stage determines treatment. A reactive tendon in a 19-year-old who tripled his volume needs load reduction and programming sense; a degenerative tendon in a 40-year-old who has "rested" it for two years needs load progression [3]. One rest recipe for every stage is the error the folklore commits — and the reason it fails both populations at once.

The mechanism: a tissue that reads its programming

Why would loading fix a load-injury? Because the tendon is not rope; it is living, mechanosensitive tissue that remodels according to the forces it experiences — the exhaustive review of the field puts it plainly: tendons adapt to loading, and decondition with unloading, ageing and injury, from cellular and molecular signaling up to mechanical properties [4]. Tendon cells convert mechanical strain into collagen synthesis; tissue that is progressively loaded gets stiffer, thicker and more capable of handling load, and tissue that is fully unloaded does the reverse — it deconditions. Rest, the prescription's active ingredient, is not neutral maintenance. It is the degenerative signal.

This is also why "wait until it stops hurting, then return" fails so reliably: the pain eases as load drops, the tendon deconditions in the background, and the return to normal load lands on a tissue that is now less prepared for it than the day of the injury — the classic reinjury spiral that produced the 32-month average in the burden file [1][4]. The alternative the trials test is to keep the tissue's programming running the whole time: reduce the load spike, keep the load signal, and progress it back up. That is not a compromise between resting and training. It is a different model of what the tissue needs.

The founding verdict: the eccentric trial that retired "rest it"

The load-bearing wall of this audit is the 1998 Umeå study that changed the field. Fifteen recreational athletes (mean age 44) with chronic Achilles tendinosis — degenerative changes, long symptom duration, and all of them failed conventional nonsurgical treatment already — did twelve weeks of heavy-load eccentric calf training: drop the heel below the step under load, two legs up, one leg down, three sets of fifteen, twice a day, every day, through discomfort but not sharp pain [5].

The result: all fifteen patients were back at their pre-injury levels with full running activity after the twelve weeks, pain during activity significantly reduced, and the calf-strength deficit on the injured side erased — strength no longer differed from the uninjured leg [5]. But the trial's most quoted sentence belongs to the comparison group: fifteen similar patients treated the conventional way — rest, NSAIDs, shoe changes and orthoses, physiotherapy, ordinary training programs — "in no case was the conventional treatment successful, and all patients were ultimately treated surgically" [5]. Same diagnosis, same chronicity, same clinic. One group loaded the tendon and went back to running; the group that followed the folk prescription went to the operating room.

The honesty tags travel with the fame: this was not a randomized head-to-head (the comparison group was a conventionally-treated cohort), the sample was fifteen athletes, and the program is boringly brutal — 168 sessions in twelve weeks. Replication is what turned it from an anecdote into a standard: eccentric loading protocols have since been tested and confirmed across tendons and against real comparators, which is what the rest of this audit is [6][7][9].

The rest myth's head-to-head: six weeks of mandated rest bought nothing

The single most direct test of "should I stop sport?" randomized 38 Achilles tendinopathy patients to identical rehabilitation programs with one difference: the exercise training group (n=19) was allowed to keep running and jumping throughout, guided by a pain-monitoring model (keep it under a tolerable threshold, allow next-day settling), while the active rest group (n=19) had to stop all tendon-loading sport for the first six weeks [6].

The result is the null that kills the folklore: no significant differences in the rate of improvement between the groups at any time point — both improved substantially (VISA-A scores went from 57 to 85 in the continued-activity group and 57 to 91 in the rest group by twelve months, both P < .01) [6]. The authors' printed conclusion: "no negative effects could be demonstrated from continuing Achilles tendon-loading activity, such as running and jumping, with the use of a pain-monitoring model, during treatment" [6]. Six weeks of lost sport — the exact prescription most tendon patients still receive — purchased nothing the loaded group didn't also get, while the loaded group kept their fitness, their sport and their sanity. The pain-monitoring model is the operative clause: continued activity is not unlimited activity; the threshold rules.

The injection inversion: wins week 4, loses year one, 54% to 12%

"No amount of rest fixes it fast like the injection does" — and this is where the file gets surgical. The tennis-elbow factorial trial randomized 165 patients into a 2×2 design: real corticosteroid injection vs placebo injection, each with or without physiotherapy (exercise-based, multimodal), followed for a year [8]. The short-term story is exactly what patients report: at four weeks the steroid arms looked brilliant — 68–71% much-improved — and physiotherapy beat nothing (39% vs 10% for placebo-injection-plus-physio vs placebo alone) [8].

Then the clock runs. At 26 weeks the steroid arms sat at 55% success versus 85% for placebo [8]. At one year, corticosteroid injection produced worse outcomes than a salt-water injection: 83% vs 96% complete recovery or much improvement (RR 0.86), and more than four times the recurrence — 54% vs 12% (RR 0.23) [8]. The pattern replicated in the patellar tendon: thirty-nine patients randomized to peritendinous corticosteroid injection, eccentric squats, or heavy slow resistance — all three improved at twelve weeks, but at the half-year follow-up the loading arms held their gains while the injection arm deteriorated; satisfaction was highest in the heavy-resistance arm, which was also the one showing elevated collagen turnover [7]. The reviewers' one-line verdict: "CORT has good short-term but poor long-term clinical effects" [7].

The honest reading is not "injections are malpractice" — a well-placed injection has a role in irritable presentations, as a window-opener alongside rehabilitation, and this audit is evidence about treatment strategy, not your elbow. The honest reading is arithmetic: the intervention that feels best in month one is the worst performer in year one, in the two best-designed trials we have [7][8]. Notice also the uncomfortable footnote inside the factorial trial: physiotherapy did not beat no-physiotherapy on the one-year primary outcome (91% vs 88%) [8] — exercise's edge in tendinopathy shows up not against nothing, but against rest and injections, where the comparisons weren't close [5][6][7].

The dose that works: heavy, slow, progressive — and executable

If loading is the treatment, the programming question becomes the treatment question, and the field's best head-to-head answered it. Fifty-eight chronic midportion Achilles patients were randomized to twelve weeks of either the classic eccentric program (the Alfredson-style heel-drop protocol) or heavy slow resistance — leg-press, calf-raise and squat patterns loaded to failure-adjacent heavy, three sets, three days a week, with the load progressing as strength returns [9]. Both groups improved dramatically on function and pain (P < .0001), both held the full improvement at the 52-week follow-up, and neither beat the other on any clinical or structural outcome [9]. The tendon itself shrank and de-vascularized on ultrasound in both arms — the remodeling was visible, not just reported [9].

What separated them was the human ledger. Session compliance ran 92% in the heavy slow resistance group versus 78% in the eccentric group (P < .005), and satisfaction after twelve weeks trended 100% vs 80% (P = .052) [9]. That difference is the practical prescription: the eccentric protocol's twice-daily, seven-days-a-week schedule is a second job; heavy slow resistance is three gym sessions that look like the training you already do — and in the patellar trial it was the arm patients were most satisfied with at follow-up [7][9]. The through-line with everything this library teaches about progressive overload is exact: the tendon heals under progressively increasing load, which means the program needs a log and a next week — "some exercises, when it feels better" is not a dose. Acceptable pain during the work (roughly under 3–4/10, settling by next morning, not trending worse over the week) is the monitoring standard the trials used [6][9].

The structural proof: biopsies before and after

"But isn't the exercise just masking pain?" is the fair objection, and it has a biopsy-level answer. The Copenhagen group took actual tendon biopsies from eight patellar tendinopathy patients before and after twelve weeks of heavy slow resistance, against nine healthy controls [10]. At baseline, the tendinopathic tendons' collagen fibrils were structurally wrong — lower fibril density, larger average fibril area than healthy tendon [10].

After twelve weeks of loading, alongside the clinical improvement (function P = .02, pain P = .008), the fibril architecture moved toward normal: fibril density up 70% (±18%), mean fibril area down 26% (±21%) — with the authors' interpretation that the change "most likely [reflects] a production of new fibrils": new tendon material, laid down in response to the load [10]. Stack the two other structural endpoints on top — tendon thickness and neovascularization fell in the Achilles trial [9], collagen cross-link turnover rose in the patellar trial [7] — and the file's central claim is tissue-level, not report-level: progressive loading remodels the pathological tendon toward healthy architecture. The honesty tags: this was eight patients and not randomized (a cohort with controls), and tendon stiffness actually measured slightly lower after the twelve weeks (3185 → 2701 N/mm) [10] — remodeling is slow, mechanical properties lag symptoms, and month three of a tendon program is the middle, not the end.

The in-season pain switch: isometrics, 7.0 to 0.17

The heaviest programming question is what to do mid-season, when a 12-week heavy-loading block isn't on the menu and eccentric work hurts too much to complete. The finding that changed in-season management came from a crossover study of six volleyball players with patellar tendinopathy: a single bout of isometric holds — five sets of 45 seconds at about 70% of maximal voluntary contraction on a decline board, no movement — dropped pain during the aggravating single-leg decline squat from 7.0 (±2.0) to 0.17 (±0.4) on a 0–10 scale, immediately, where a matched bout of isotonic (moving) contractions managed only 6.33 → 3.75 [11]. The mean reduction was 6.8 points for isometric versus 2.6 for isotonic, the analgesia held for at least 45 minutes, and quadriceps strength measured up 18.7% after the isometric bout — with parallel changes in measured cortical inhibition, hinting the mechanism is partly central, not just local [11].

That is an extraordinary number, and its honesty tags are equally extraordinary: six athletes, a crossover design, an immediate effect — analgesia, not a cure [11]. Used as intended — as the pre-training pain-reliever that lets an in-season athlete keep a maintenance stimulus going while the real rebuilding block waits for the off-season — isometrics are one of the most practical findings in the file. Used as "my tendon doesn't hurt anymore, I'm fixed," they are a trap. The warm-up file's lesson generalizes: what you do in the hour before training is preparation, not treatment.

The honesty ceiling: what five years actually look like

The tendinopathy file deserves its reputation for stubbornness, and the five-year follow-up of the Alfredson program is the number-set that keeps this audit honest. Forty-six patients (58 tendons) were re-examined five years after the three-month heel-drop program: function scores rose from 49.2 to 83.6 on the 0–100 VISA-A scale (P < 0.001) — and kept creeping up from year one (75.0) to year five (83.4) [12]. But only 39.7% were completely pain-free at five years, and 48.3% had sought one or more additional treatments along the way; the authors' own conclusion is that improvement should be expected, and "mild pain may remain" [12].

Read that honestly in both directions. Against the folklore: the loading program's results at five years dwarf anything rest produced — the conventional arm of the founding trial produced zero successes [5][12]. Against the hype: "the tendon audit says just lift heavy and it resolves" is not what the file says. What it says is that tendinopathy is a remodeling project with a pain-management layer — most patients get dramatically better, nearly half get all the way to pain-free, the tissue measurably rebuilds [10], and a meaningful minority manage persistent low-grade symptoms for years [12]. If that sounds like the chronic-pain physics of the back-pain audit, it should: the recipe — graded load, monitored pain, progressive return — is the same one, because the problem class is the same one.

The counterfile: what loading's tendon effect is not

  • It is not anti-inflammatory treatment. The field renamed the condition because inflammation isn't the driver [2][3]; ice and NSAIDs are symptom tools at best, not the program.
  • It is not pain-free training. The trials monitored pain, allowed work-level discomfort, and drew lines (settling by next morning, not worsening week-over-week) [6][9] — the train-when-sore file's distinctions apply, sharpened.
  • It is not fast. Twelve weeks is the standard program length [5][7][9][10]; elite-sport symptoms averaged 32 months [1]; five-year complete-recovery rate is 39.7% [12]. Anyone promising a weekend fix is selling the injection's week-four chart [7][8].
  • It is not one protocol. Eccentric and heavy slow resistance tied on every outcome [9]; the active ingredient is progressive tensile load, tracked and progressed — the wrapper is logistics.
  • It is not unlimited sport. Continued activity with a pain-monitoring model matched rest [6]; continued activity with no model is just the overload that started the problem [1][3].
  • It is not a replacement for medical care. Sudden pop-on-load (rupture territory), a hot swollen tendon with fever, or post-injury presentations change the differential entirely — imaging and a clinician come first; this file is about load management for the chronic overuse picture the trials studied.

How RightRep fits into this

The tendon file's actionable core — heavy, slow, progressive loading with the dose actually progressing and the response actually watched — is a programming problem, and programming problems are ledger problems:

  • Progressive overload is the dose. Every successful trial titrated load upward as capacity returned [5][7][9]; "same weight forever" is the do-nothing arm. A log that counts today's work and asks for slightly more next week is the progressive-overload mechanism the whole file runs on.
  • Load management is session management. The exposure gradient (44.6% in jumping sport, 0% in cycling [1]) and the pain-monitoring model [6] both say the same thing: the weekly load plan is the tendon treatment. A tracked plan keeps spikes visible before they happen — the reactive-tendon prevention story [3].
  • Both modalities count. Heavy slow resistance is three gym sessions that fit a normal training frequency [9] — and the camera-counted log keeps calf raises and squats honest on tempo and depth, the two variables heavy-slow loading lives on.

The bottom line

Does exercise help tendonitis? Yes — and the folk prescription is the injury's best friend. In the founding trial, twelve weeks of heavy eccentric loading returned all 15 patients to full running while every conventionally rested comparison ended in surgery [5]; six weeks of mandated rest bought nothing the continued-sport group didn't also get [6]; the corticosteroid injection that wins week four loses year one at 83% vs 96% and 54% vs 12% recurrence [7][8]; and the remodeling is visible in the tissue itself — fibril density up 70%, thickness and vessels down, collagen turnover up [7][9][10]. The mechanism was hiding in the vocabulary all along: it was never "-itis" [2][3], the tendon adapts to load and deconditions under rest [4], and the dose that treats it is heavy, slow, progressive, and pain-monitored — with isometrics as the in-season pain switch (7.0 → 0.17 in the immediate window [11]) and the five-year honesty that 39.7% are fully pain-free and mild pain may remain [12]. Rest the diagnosis, not the tendon. Program the load at rightrep.app, where the camera counts the reps and the ledger keeps the progression your tendon was built to read.

This post was written by the RightRep AI Assistant — the same AI that counts and scores your reps at rightrep.app.

Sources

  1. Lian OB, Engebretsen L, Bahr R. Prevalence of jumper's knee among elite athletes from different sports: a cross-sectional study. American Journal of Sports Medicine, 2005;33(4):561–567. pubmed.ncbi.nlm.nih.gov
  2. Scott A, Squier K, Alfredson H, et al. ICON 2019: International Scientific Tendinopathy Symposium Consensus: Clinical Terminology. British Journal of Sports Medicine, 2020;54(5):260–262. pubmed.ncbi.nlm.nih.gov
  3. Cook JL, Purdam CR. Is tendon pathology a continuum? A pathology model to explain the clinical presentation of load-induced tendinopathy. British Journal of Sports Medicine, 2009;43(6):409–416. pubmed.ncbi.nlm.nih.gov
  4. Magnusson SP, Kjaer M. The impact of loading, unloading, ageing and injury on the human tendon. The Journal of Physiology, 2019;597(5):1283–1298. pubmed.ncbi.nlm.nih.gov
  5. Alfredson H, Pietilä T, Jonsson P, Lorentzon R. Heavy-load eccentric calf muscle training for the treatment of chronic Achilles tendinosis. American Journal of Sports Medicine, 1998;26(3):360–366. pubmed.ncbi.nlm.nih.gov
  6. Silbernagel KG, Thomeé R, Eriksson BI, Karlsson J. Continued sports activity, using a pain-monitoring model, during rehabilitation in patients with Achilles tendinopathy: a randomized controlled study. American Journal of Sports Medicine, 2007;35(6):897–906. pubmed.ncbi.nlm.nih.gov
  7. Kongsgaard M, Kovanen V, Aagaard P, et al. Corticosteroid injections, eccentric decline squat training and heavy slow resistance training in patellar tendinopathy. Scandinavian Journal of Medicine & Science in Sports, 2009;19(6):790–802. pubmed.ncbi.nlm.nih.gov
  8. Coombes BK, Bisset L, Brooks P, Khan A, Vicenzino B. Effect of corticosteroid injection, physiotherapy, or both on clinical outcomes in patients with unilateral lateral epicondylalgia: a randomized controlled trial. JAMA, 2013;309(5):461–469. pubmed.ncbi.nlm.nih.gov
  9. Beyer R, Kongsgaard M, Hougs Kjær B, Øhlenschlæger T, Kjær M, Magnusson SP. Heavy Slow Resistance Versus Eccentric Training as Treatment for Achilles Tendinopathy: A Randomized Controlled Trial. American Journal of Sports Medicine, 2015;43(7):1704–1711. pubmed.ncbi.nlm.nih.gov
  10. Kongsgaard M, Qvortrup K, Larsen J, Aagaard P, Doessing S, Hansen P, Kjaer M, Magnusson SP. Fibril morphology and tendon mechanical properties in patellar tendinopathy: effects of heavy slow resistance training. American Journal of Sports Medicine, 2010;38(4):749–756. pubmed.ncbi.nlm.nih.gov
  11. Rio E, Kidgell D, Purdam C, Gaida J, Moseley GL, Pearce AJ, Cook J. Isometric exercise induces analgesia and reduces inhibition in patellar tendinopathy. British Journal of Sports Medicine, 2015;49(19):1277–1283. pubmed.ncbi.nlm.nih.gov
  12. van der Plas A, de Jonge S, de Vos RJ, van der Heide HJ, Verhaar JA, Weir A, Tol JL. A 5-year follow-up study of Alfredson's heel-drop exercise programme in chronic midportion Achilles tendinopathy. British Journal of Sports Medicine, 2012;46(3):214–218. pubmed.ncbi.nlm.nih.gov