Place · Level 3 · Recovery
Tendon Recovery · Load It, Don't Just Rest It
肌腱慢半拍的真相 · -itis是个误称, 慢性肌腱痛是失败的修复不是炎症 · 机械负荷才是修复的开关 (mechanotransduction) · 渐进、离心、重慢负荷胜过休息 · 胶原 + 维 C 时机的诚实版本 · 类固醇注射的长期代价
Story path
- 1Why tendons heal slowly · structure is destinyWhy tendons heal slowly · structure is destiny
- 2'-itis' is a misnomer · chronic tendon pain isn't inflammation'-itis' is a misnomer · chronic tendon pain isn't inflammation
- 3Load is the repair switch · mechanotransductionLoad is the repair switch · mechanotransduction
- 4How to load · eccentric / heavy-slow / isometric (RCT-backed)How to load · eccentric / heavy-slow / isometric (RCT-backed)
- 5How much nutrition helps · the honest collagen + vitamin C storyHow much nutrition helps · the honest collagen + vitamin C story
- 6What to do · decision tree + the steroid cost + atlas loopWhat to do · decision tree + the steroid cost + atlas loop
Chapter 1
Why tendons heal slowly · structure is destiny
Why tendons heal slowly · structure is destiny
A pulled muscle heals in two or three weeks; a 'tennis elbow' or an aching Achilles can drag on for months, sometimes a year. That is not because you are slacking — it is because tendon, as a tissue, is structurally slow. Get this and every later 'why rest alone won't fix it' makes sense.
What tendon is made of
The bulk is dense type-I collagen, laid down in parallel bundles built to transmit muscle force to bone — one of the body's highest tensile-strength connective tissuesIts cells (tenocytes) sit sparsely in the collagen matrix — few in number, low in metabolismVery poor blood supply (relative to muscle): tendon is a low-vascular, low-oxygen, low-metabolic-rate tissue — the first reason it heals slowly
How slow: the carbon-14 nuclear-bomb-pulse experiment (Heinemeier et al. 2013)
A beautiful natural experiment: the 1955-1963 nuclear tests spiked atmospheric ¹⁴C, effectively 'time-stamping' people born or growing in those yearsThey measured ¹⁴C in the core of adult Achilles tendons. The result was striking: it matched the atmospheric level from before that person turned 17 (the height-growth years)Meaning: the collagen in the tendon core is laid down during growth and barely renewed in adulthood — in sharp contrast to muscle, which turns over continuouslyThis explains at the molecular level why tendon 'heals slowly once injured and gets more brittle with age'
The framing in one line: tendon does not 'fail to heal' — it renews extremely slowly. Precisely because it is slow, it is fussy about *how* you stimulate it to repair — and as the next stop shows, that stimulus surprises most people: not rest, but load.
(Articular cartilage is a similar avascular tissue that is 'fed and repaired by load'; for that thread you can dive 到 exercise-as-medicine and its exercise-tissue-adaptation section.)
What tendon is made of
The bulk is dense type-I collagen, laid down in parallel bundles built to transmit muscle force to bone — one of the body's highest tensile-strength connective tissuesIts cells (tenocytes) sit sparsely in the collagen matrix — few in number, low in metabolismVery poor blood supply (relative to muscle): tendon is a low-vascular, low-oxygen, low-metabolic-rate tissue — the first reason it heals slowly
How slow: the carbon-14 nuclear-bomb-pulse experiment (Heinemeier et al. 2013)
A beautiful natural experiment: the 1955-1963 nuclear tests spiked atmospheric ¹⁴C, effectively 'time-stamping' people born or growing in those yearsThey measured ¹⁴C in the core of adult Achilles tendons. The result was striking: it matched the atmospheric level from before that person turned 17 (the height-growth years)Meaning: the collagen in the tendon core is laid down during growth and barely renewed in adulthood — in sharp contrast to muscle, which turns over continuouslyThis explains at the molecular level why tendon 'heals slowly once injured and gets more brittle with age'
The framing in one line: tendon does not 'fail to heal' — it renews extremely slowly. Precisely because it is slow, it is fussy about *how* you stimulate it to repair — and as the next stop shows, that stimulus surprises most people: not rest, but load.
(Articular cartilage is a similar avascular tissue that is 'fed and repaired by load'; for that thread you can dive 到 exercise-as-medicine and its exercise-tissue-adaptation section.)
Chapter 2
'-itis' is a misnomer · chronic tendon pain isn't inflammation
'-itis' is a misnomer · chronic tendon pain isn't inflammation
For decades we called it 'tendinitis,' assumed inflammation, and reached first for ice, anti-inflammatories, and rest-until-it-stops-hurting. That logic is fine for an acute strain, but for the kind of chronic tendon pain that has dragged on for months, it has the wrong target from the start.
The histology tells the truth
Cut open a chronically painful tendon and you find few classic inflammatory cells; what you see is disorganised collagen, matrix breakdown, abnormal blood-vessel and nerve ingrowth, and altered tenocytes — this is degeneration, not 'itis'So the more accurate names are tendinopathy or tendinosis, not tendinitisThe core mechanism is a failed healing response: the tendon is repeatedly hit by loads beyond its repair capacity, and the half-repaired collagen is disorganised and weak
The Cook & Purdam continuum model (2009, BJSM)
This much-cited model places tendinopathy on a continuous spectrum, explaining why the same tendon resolves in weeks for one person and becomes stubborn chronic pain for another:
Reactive: a short-term load spike → a reversible stress response in tenocytes and matrix (de-load + wait, and it can step back)Dysrepair: a repair attempt that fails; the matrix starts to disorganiseDegenerative: large patches of disordered collagen, cell apoptosis, abnormal vessels — this stage is largely irreversible, and the goal shifts to 'make the surrounding healthy tissue stronger to compensate'
The more sobering evidence (Heinemeier et al. 2018, carbon-14 again)
Healthy adult tendon barely renews; but in diseased tendon they measured substantial collagen turnover — and the modelling implies this abnormal high turnover had been running for several years before pain appearedMeaning: the pain you feel today is a signal that surfaces only after the tendon has been quietly struggling to repair for a long time — no wonder it does not resolve in days, and no wonder 'it stopped hurting, so it's healed' is so often an illusion
Why this matters (not just semantics)
Since the lead actor is not inflammation, anti-inflammatory-only approaches (long-term anti-inflammatory drugs / repeated ice / steroid injections) are pressing on a fire that isn't there — symptomatic at best, possibly harmful (the long-term cost of steroids gets its own stop below)What actually needs fixing is the failed repair + disorganised collagen — and the tool for that is exactly the subject of the next stop: controlled mechanical load.
The histology tells the truth
Cut open a chronically painful tendon and you find few classic inflammatory cells; what you see is disorganised collagen, matrix breakdown, abnormal blood-vessel and nerve ingrowth, and altered tenocytes — this is degeneration, not 'itis'So the more accurate names are tendinopathy or tendinosis, not tendinitisThe core mechanism is a failed healing response: the tendon is repeatedly hit by loads beyond its repair capacity, and the half-repaired collagen is disorganised and weak
The Cook & Purdam continuum model (2009, BJSM)
This much-cited model places tendinopathy on a continuous spectrum, explaining why the same tendon resolves in weeks for one person and becomes stubborn chronic pain for another:
Reactive: a short-term load spike → a reversible stress response in tenocytes and matrix (de-load + wait, and it can step back)Dysrepair: a repair attempt that fails; the matrix starts to disorganiseDegenerative: large patches of disordered collagen, cell apoptosis, abnormal vessels — this stage is largely irreversible, and the goal shifts to 'make the surrounding healthy tissue stronger to compensate'
The more sobering evidence (Heinemeier et al. 2018, carbon-14 again)
Healthy adult tendon barely renews; but in diseased tendon they measured substantial collagen turnover — and the modelling implies this abnormal high turnover had been running for several years before pain appearedMeaning: the pain you feel today is a signal that surfaces only after the tendon has been quietly struggling to repair for a long time — no wonder it does not resolve in days, and no wonder 'it stopped hurting, so it's healed' is so often an illusion
Why this matters (not just semantics)
Since the lead actor is not inflammation, anti-inflammatory-only approaches (long-term anti-inflammatory drugs / repeated ice / steroid injections) are pressing on a fire that isn't there — symptomatic at best, possibly harmful (the long-term cost of steroids gets its own stop below)What actually needs fixing is the failed repair + disorganised collagen — and the tool for that is exactly the subject of the next stop: controlled mechanical load.
Chapter 3
Load is the repair switch · mechanotransduction
Load is the repair switch · mechanotransduction
Here is the most counterintuitive and most important line on the whole map: the tendon's repair signal comes primarily from mechanical load itself. For chronic tendinopathy, 'rest completely and let it heal on its own' is often counterproductive.
Mechanotransduction (Khan & Scott 2009, BJSM)
Tenocytes sense stretch: mechanical load is 'translated' into a biochemical signal through the cytoskeleton and mechanosensitive channelsThat signal switches on collagen synthesis and matrix remodelling — in other words, tension is what tells the cell 'repair, build, line up straight'The framework is named mechanotherapy: when a physiotherapist prescribes exercise, they are using controlled force to drive the tissue's own repair
The reverse: use it or lose it
Immobilisation / prolonged unloading makes a tendon synthesise less collagen, grow thinner and weaker, lose stiffness — 'resting it' is not repair; it is letting it degenerateThis is also why the acute-injury consensus has evolved from 'complete immobilisation (RICE)' to early controlled loading (that thread is detailed in training-injuries; you can dive 到 training-injuries)
So the correct version of 'rest vs load'
The first days of acute severe pain: relative rest + remove the excess load that set it off (de-load) — not lie flat and motionlessAfter that: gradual, progressive reloading — using 'mildly tolerable discomfort that is no worse the next day' as the dose ceilingThe goal is not 'wait until it stops hurting' but to actively rebuild collagen alignment and strength
Two clinically vivid side-supports (both link to how exercise remodels tissue)
Growth hormone can up-regulate tendon collagen synthesis several-fold without touching myofibrillar protein — showing the tendon's collagen machinery can be mobilised, just with a different key (this is not a cue to inject GH; it is evidence that collagen is 'alive and stimulable')Cartilage works the same way: moderate load feeds it, prolonged unloading degrades it — 'use it or lose it' is the general rule of connective tissue (dive 到 exercise-as-medicine)
In one line: tendon is slow because it renews slowly; but it can be repaired, and the key is the right amount of tension, not the bed. The next stop is how to use that key — with real RCTs behind it.
Mechanotransduction (Khan & Scott 2009, BJSM)
Tenocytes sense stretch: mechanical load is 'translated' into a biochemical signal through the cytoskeleton and mechanosensitive channelsThat signal switches on collagen synthesis and matrix remodelling — in other words, tension is what tells the cell 'repair, build, line up straight'The framework is named mechanotherapy: when a physiotherapist prescribes exercise, they are using controlled force to drive the tissue's own repair
The reverse: use it or lose it
Immobilisation / prolonged unloading makes a tendon synthesise less collagen, grow thinner and weaker, lose stiffness — 'resting it' is not repair; it is letting it degenerateThis is also why the acute-injury consensus has evolved from 'complete immobilisation (RICE)' to early controlled loading (that thread is detailed in training-injuries; you can dive 到 training-injuries)
So the correct version of 'rest vs load'
The first days of acute severe pain: relative rest + remove the excess load that set it off (de-load) — not lie flat and motionlessAfter that: gradual, progressive reloading — using 'mildly tolerable discomfort that is no worse the next day' as the dose ceilingThe goal is not 'wait until it stops hurting' but to actively rebuild collagen alignment and strength
Two clinically vivid side-supports (both link to how exercise remodels tissue)
Growth hormone can up-regulate tendon collagen synthesis several-fold without touching myofibrillar protein — showing the tendon's collagen machinery can be mobilised, just with a different key (this is not a cue to inject GH; it is evidence that collagen is 'alive and stimulable')Cartilage works the same way: moderate load feeds it, prolonged unloading degrades it — 'use it or lose it' is the general rule of connective tissue (dive 到 exercise-as-medicine)
In one line: tendon is slow because it renews slowly; but it can be repaired, and the key is the right amount of tension, not the bed. The next stop is how to use that key — with real RCTs behind it.
Chapter 4
How to load · eccentric / heavy-slow / isometric (RCT-backed)
How to load · eccentric / heavy-slow / isometric (RCT-backed)
'Load is the key' is not a slogan — several generations of RCTs have turned it into concrete protocols. Below are the most evidence-backed, each marked with its honest weight (no overselling).
Eccentric training · Alfredson's 'heel drop' (1998, AJSM)
A classic small study (n=15) had chronic Achilles-tendinopathy patients do daily heavy-load eccentric calf raises (stand on a step, slowly 'lower' the heel with the affected leg); at 12 weeks all returned to running'Eccentric' = the muscle generating force while lengthening, a particularly effective remodelling stimulus for tendonCaveat: small sample, pioneering rather than definitive — but it ignited the entire 'treat tendon with load' paradigm
Heavy Slow Resistance (HSR) (Beyer et al. 2015, AJSM; Kongsgaard et al. 2009)
Heavier weight, slow tempo (e.g. 3 s down, 3 s up), fewer repsAn Achilles RCT showed HSR and eccentric training are equally effective at 12 months, but HSR had better adherence and higher patient satisfaction (simpler movement, fewer sessions per week)In patellar-tendinopathy comparisons, both HSR and eccentric beat steroid injection on long-term outcomesPractical takeaway: eccentric and heavy-slow loading both work — pick the one you'll actually stick to; adherence is itself part of the effect
Isometrics · for 'too painful to train' (Rio et al. 2015, BJSM)
A small crossover study showed that in patellar tendinopathy, isometric holds (joint still, sustained effort, e.g. a wall sit) gave immediate analgesia (~45 minutes in the paper), letting people get into the subsequent trainingHonest flag: later replications were mixed — don't treat it as magic; but as a 'stepping stone when it's too painful to load' it's a reasonable option
Dose principles running through every protocol
Pain-monitored loading: pain during and in the 24 hours after training, kept within 'tolerable, no worse the next day,' is acceptable — tendon rehab allows training with a little pain, unlike a muscle tearProgressive overload: nudge weight / sets up every 1-2 weeks to keep the remodelling signal comingSlow: tendons improve over weeks-to-months, not days. A common mistake is 'it's a bit better, so ramp hard / return to sport' → falling back into the reactive stageProfessional guidance: midportion Achilles tendinopathy has a formal physiotherapy clinical practice guideline (JOSPT 2018) that lists progressive loading as a grade-A recommendation — 'load beats rest' codified into guidelines
The next stop sorts out the nutrition piece (the honest version of collagen + vitamin C), then gives a decision tree for 'when to see a clinician.'
Eccentric training · Alfredson's 'heel drop' (1998, AJSM)
A classic small study (n=15) had chronic Achilles-tendinopathy patients do daily heavy-load eccentric calf raises (stand on a step, slowly 'lower' the heel with the affected leg); at 12 weeks all returned to running'Eccentric' = the muscle generating force while lengthening, a particularly effective remodelling stimulus for tendonCaveat: small sample, pioneering rather than definitive — but it ignited the entire 'treat tendon with load' paradigm
Heavy Slow Resistance (HSR) (Beyer et al. 2015, AJSM; Kongsgaard et al. 2009)
Heavier weight, slow tempo (e.g. 3 s down, 3 s up), fewer repsAn Achilles RCT showed HSR and eccentric training are equally effective at 12 months, but HSR had better adherence and higher patient satisfaction (simpler movement, fewer sessions per week)In patellar-tendinopathy comparisons, both HSR and eccentric beat steroid injection on long-term outcomesPractical takeaway: eccentric and heavy-slow loading both work — pick the one you'll actually stick to; adherence is itself part of the effect
Isometrics · for 'too painful to train' (Rio et al. 2015, BJSM)
A small crossover study showed that in patellar tendinopathy, isometric holds (joint still, sustained effort, e.g. a wall sit) gave immediate analgesia (~45 minutes in the paper), letting people get into the subsequent trainingHonest flag: later replications were mixed — don't treat it as magic; but as a 'stepping stone when it's too painful to load' it's a reasonable option
Dose principles running through every protocol
Pain-monitored loading: pain during and in the 24 hours after training, kept within 'tolerable, no worse the next day,' is acceptable — tendon rehab allows training with a little pain, unlike a muscle tearProgressive overload: nudge weight / sets up every 1-2 weeks to keep the remodelling signal comingSlow: tendons improve over weeks-to-months, not days. A common mistake is 'it's a bit better, so ramp hard / return to sport' → falling back into the reactive stageProfessional guidance: midportion Achilles tendinopathy has a formal physiotherapy clinical practice guideline (JOSPT 2018) that lists progressive loading as a grade-A recommendation — 'load beats rest' codified into guidelines
The next stop sorts out the nutrition piece (the honest version of collagen + vitamin C), then gives a decision tree for 'when to see a clinician.'
Chapter 5
How much nutrition helps · the honest collagen + vitamin C story
How much nutrition helps · the honest collagen + vitamin C story
Tendon is made of collagen, so 'eat collagen / drink bone broth to fix tendons' sounds self-evident. But there is an honest version — neither dismiss it outright nor let marketing mythologise it.
The foundation: total protein + vitamin C (this part is solid)
Collagen synthesis needs adequate amino-acid raw material (especially glycine and proline) — and that comes first from enough total daily protein, not a magic powder (the full protein picture: dive 到 protein)Vitamin C is an essential cofactor for collagen synthesis: prolyl and lysyl hydroxylases depend on it; without it you can't make hydroxyproline and the collagen triple helix is unstable — which is exactly why scurvy causes wounds to split and connective tissue to fall apart (full vitamin-C mechanism: dive 到 vitamin-c)Meaning: enough protein + no vitamin-C deficiency is the nutritional floor for tendon repair — established, cheap, universal
An emerging but honest addition: the 'timing window' of collagen/gelatin + vitamin C (Shaw et al. 2017)
This small crossover study had subjects take 15 g of vitamin-C-enriched gelatin 1 hour before exercise; afterward the blood marker of type-I collagen synthesis (the propeptide) doubled, and engineered ligaments cultured in the subjects' serum improved in collagen content and mechanicsProposed mechanism: providing raw material + vitamin C before loading catches the window when exercise stimulates collagen synthesisHonest flag: this is mechanistic / surrogate-endpoint evidence (blood markers, in-vitro ligaments), not a clinical endpoint of 'tendon pain heals faster'; the sample is small and bigger trials are needed. It is mildly promising, far from 'must take'
Drawing a line against the 'bone broth / collagen myth'
Most 'bone broth heals joints and tendons' and 'collagen reverses ageing' claims run well past the evidence — bone broth's collagen content and amino-acid profile are inconsistent, a long way from Shaw's 'measured gelatin + vitamin C + load timing' (this debunk: dive 到 collagen-bone-broth)The key distinction: collagen you eat is digested into amino acids; it does not 'travel intact and patch onto the tendon.' Its value is raw material + a possible timing effect, not 'eat-tendon-to-fix-tendon'
A pragmatic protocol (if you want to try it)
Build the foundation first: enough daily protein (e.g. 1.2-1.6 g/kg) + no vitamin-C deficiency (food first; most people need no megadose)To layer on Shaw's timing strategy: take ~15 g collagen peptide/gelatin + some vitamin C about 30-60 minutes before a rehab session — as icing, not a substitute for the trainingDon't expect a supplement to work alone: no nutrient replaces 'progressive loading' as the primary treatment — nutrition is the support role, load is the treatment.
The foundation: total protein + vitamin C (this part is solid)
Collagen synthesis needs adequate amino-acid raw material (especially glycine and proline) — and that comes first from enough total daily protein, not a magic powder (the full protein picture: dive 到 protein)Vitamin C is an essential cofactor for collagen synthesis: prolyl and lysyl hydroxylases depend on it; without it you can't make hydroxyproline and the collagen triple helix is unstable — which is exactly why scurvy causes wounds to split and connective tissue to fall apart (full vitamin-C mechanism: dive 到 vitamin-c)Meaning: enough protein + no vitamin-C deficiency is the nutritional floor for tendon repair — established, cheap, universal
An emerging but honest addition: the 'timing window' of collagen/gelatin + vitamin C (Shaw et al. 2017)
This small crossover study had subjects take 15 g of vitamin-C-enriched gelatin 1 hour before exercise; afterward the blood marker of type-I collagen synthesis (the propeptide) doubled, and engineered ligaments cultured in the subjects' serum improved in collagen content and mechanicsProposed mechanism: providing raw material + vitamin C before loading catches the window when exercise stimulates collagen synthesisHonest flag: this is mechanistic / surrogate-endpoint evidence (blood markers, in-vitro ligaments), not a clinical endpoint of 'tendon pain heals faster'; the sample is small and bigger trials are needed. It is mildly promising, far from 'must take'
Drawing a line against the 'bone broth / collagen myth'
Most 'bone broth heals joints and tendons' and 'collagen reverses ageing' claims run well past the evidence — bone broth's collagen content and amino-acid profile are inconsistent, a long way from Shaw's 'measured gelatin + vitamin C + load timing' (this debunk: dive 到 collagen-bone-broth)The key distinction: collagen you eat is digested into amino acids; it does not 'travel intact and patch onto the tendon.' Its value is raw material + a possible timing effect, not 'eat-tendon-to-fix-tendon'
A pragmatic protocol (if you want to try it)
Build the foundation first: enough daily protein (e.g. 1.2-1.6 g/kg) + no vitamin-C deficiency (food first; most people need no megadose)To layer on Shaw's timing strategy: take ~15 g collagen peptide/gelatin + some vitamin C about 30-60 minutes before a rehab session — as icing, not a substitute for the trainingDon't expect a supplement to work alone: no nutrient replaces 'progressive loading' as the primary treatment — nutrition is the support role, load is the treatment.
Chapter 6
What to do · decision tree + the steroid cost + atlas loop
What to do · decision tree + the steroid cost + atlas loop
Let's gather the mechanism into something usable. The core line: for chronic tendon pain, the primary treatment is progressive load + patience, not rest and anti-inflammatories.
A general rehab route
1. First days of acute severe pain: relative de-load (drop the offending movement) but don't fully immobilise — keep pain-free-range activity
2. Enter progressive loading: eccentric or heavy-slow resistance (HSR), whichever you'll stick to; if it's too painful to load, start with isometric holds for analgesia
3. Pain-monitored dose: keep going if pain during/24 h after is 'tolerable, no worse next day'; back off a notch if it worsens
4. Progress over weeks-to-months — don't ramp hard or rush back to sport just because it's 'a bit better'
5. Nutrition support role: enough protein + no vitamin-C deficiency; optionally try ~15 g collagen peptide + vitamin C 30-60 min before a session
An honest word on steroid injections (Coombes et al. 2010, Lancet)
This systematic review's conclusion is famous: corticosteroid injection for tendinopathy does relieve pain well in the short term (weeks) — but long-term (6-12 months) outcomes are worse, with higher recurrenceIt makes mechanistic sense: steroids suppress the 'inflammation' that wasn't the lead actor anyway, while potentially interfering with collagen repair and weakening the tendonMeaning: not an absolute no (short-term relief is useful in some situations), but it should not be the routine treatment for tendinopathy, and certainly not repeated. This is the real-world cost of 'thinking it's -itis when it's degeneration'
Signals to see a clinician / physiotherapist (this site is education, not a substitute for a clinician / 医师)
Pain persisting beyond 6-8 weeks, affecting daily life or training, not improving with self-directed progressive load → see a physiotherapist for an individualised loading planTo confirm it's tendinopathy (and not something else) → clinical assessment
Red flags (seek care promptly; do not self-load)
Sudden severe pain + a 'pop' + inability to push off / rise on toes / a palpable gap → suspect a tendon rupture (e.g. Achilles rupture) — this is an emergencyObvious red, hot, swollen joint + fever → rule out infection / inflammatory arthropathy, not ordinary tendinopathyNight pain that wakes you, unexplained weight loss, persistent pain unrelated to load → rule out other causes
Atlas loop
protein — the master valve for collagen raw material; the nutritional floor of repairvitamin-c — the essential cofactor for collagen synthesis (scurvy is the cautionary tale)collagen-bone-broth — separating the 'broth heals tendons' myth from Shaw's real evidenceexercise-as-medicine — the 'use it or lose it' bus: how load remodels tendon, cartilage, bonetraining-injuries — how not to overload the tendon in the first place (volume + PEACE & LOVE)sarcopenia — the same 'load stimulates synthesis' logic applied to muscle
Bottom line: tendon heals slowly because it renews slowly and is poorly vascularised; chronic pain is failed repair, not 'itis.' So the fix is not lying down to wait, nor repeated anti-inflammatories, but well-judged, progressive, patient load — pressing the 'tension → collagen remodelling' switch back on. Know this mechanism and you'll stop oscillating between 'rest completely' and 'push through the pain,' and stop paying repeatedly for a needle that only treats the symptom.
A general rehab route
1. First days of acute severe pain: relative de-load (drop the offending movement) but don't fully immobilise — keep pain-free-range activity
2. Enter progressive loading: eccentric or heavy-slow resistance (HSR), whichever you'll stick to; if it's too painful to load, start with isometric holds for analgesia
3. Pain-monitored dose: keep going if pain during/24 h after is 'tolerable, no worse next day'; back off a notch if it worsens
4. Progress over weeks-to-months — don't ramp hard or rush back to sport just because it's 'a bit better'
5. Nutrition support role: enough protein + no vitamin-C deficiency; optionally try ~15 g collagen peptide + vitamin C 30-60 min before a session
An honest word on steroid injections (Coombes et al. 2010, Lancet)
This systematic review's conclusion is famous: corticosteroid injection for tendinopathy does relieve pain well in the short term (weeks) — but long-term (6-12 months) outcomes are worse, with higher recurrenceIt makes mechanistic sense: steroids suppress the 'inflammation' that wasn't the lead actor anyway, while potentially interfering with collagen repair and weakening the tendonMeaning: not an absolute no (short-term relief is useful in some situations), but it should not be the routine treatment for tendinopathy, and certainly not repeated. This is the real-world cost of 'thinking it's -itis when it's degeneration'
Signals to see a clinician / physiotherapist (this site is education, not a substitute for a clinician / 医师)
Pain persisting beyond 6-8 weeks, affecting daily life or training, not improving with self-directed progressive load → see a physiotherapist for an individualised loading planTo confirm it's tendinopathy (and not something else) → clinical assessment
Red flags (seek care promptly; do not self-load)
Sudden severe pain + a 'pop' + inability to push off / rise on toes / a palpable gap → suspect a tendon rupture (e.g. Achilles rupture) — this is an emergencyObvious red, hot, swollen joint + fever → rule out infection / inflammatory arthropathy, not ordinary tendinopathyNight pain that wakes you, unexplained weight loss, persistent pain unrelated to load → rule out other causes
Atlas loop
protein — the master valve for collagen raw material; the nutritional floor of repairvitamin-c — the essential cofactor for collagen synthesis (scurvy is the cautionary tale)collagen-bone-broth — separating the 'broth heals tendons' myth from Shaw's real evidenceexercise-as-medicine — the 'use it or lose it' bus: how load remodels tendon, cartilage, bonetraining-injuries — how not to overload the tendon in the first place (volume + PEACE & LOVE)sarcopenia — the same 'load stimulates synthesis' logic applied to muscle
Bottom line: tendon heals slowly because it renews slowly and is poorly vascularised; chronic pain is failed repair, not 'itis.' So the fix is not lying down to wait, nor repeated anti-inflammatories, but well-judged, progressive, patient load — pressing the 'tension → collagen remodelling' switch back on. Know this mechanism and you'll stop oscillating between 'rest completely' and 'push through the pain,' and stop paying repeatedly for a needle that only treats the symptom.