Place · Level 3 · Movement
Shoulder Pain
肩膀拿稳定换活动度 · 肩袖痛不是撞击 · 减压手术不优于安慰手术 · 冻结肩会自己好 · 主线是渐进负荷
Story path
Chapter 1
Shoulder · a stability-for-motion trade
Shoulder · a stability-for-motion trade
Understand how the shoulder is built first — only then will you see why it hurts so readily.
The shoulder has the largest range of motion of any joint in the body, and the price is that it gives up most of its stability. The knee and hip are deep socket-and-peg joints, stable by design; the shoulder takes the opposite path — it trades a very shallow socket for a reach that can touch almost any angle of the body. The parts that make this work:
Glenohumeral joint: the true 'shoulder joint'. The humeral head is a large ball; the glenoid on the scapula is a shallow, small dish. It's like a golf ball on a tee — free to point anywhere, but inherently unstable on its ownLabrum: a rim of fibrocartilage set around the shallow dish to deepen it a little, and to anchor the ligaments and the long head of the biceps tendonRotator cuff: the tendons of four small muscles (supraspinatus, infraspinatus, teres minor, subscapularis) wrap the humeral head and actively press the ball into the dish like a cupped hand. It is the shoulder's real stabiliser — dynamic and muscular, not bone-lockedBursa: the subacromial bursa is a fluid-filled friction pad sitting between the cuff and the bone above itAC joint: the small joint where the outer end of the clavicle meets the acromion of the scapula, at the very top of the shoulder
And the most overlooked thing, which explains a large share of shoulder pain: scapular rhythm. When you raise your arm overhead, it isn't only the humerus moving — the scapula glides and rotates across the back of the ribcage at the same time, making room for the humeral head. Arm and scapula cooperate at roughly a 2:1 ratio. Once that rhythm goes off (a slumped posture, rounded shoulders, weak scapular muscles), the space above gets squeezed and the cuff is more easily pinched and abraded.
The first intuition this island builds: the shoulder is not a passively locked hinge but a machine kept stable by muscle, moment to moment. The next scene shows that its pain almost always comes from the after-effects of that stability-for-motion trade.
The shoulder has the largest range of motion of any joint in the body, and the price is that it gives up most of its stability. The knee and hip are deep socket-and-peg joints, stable by design; the shoulder takes the opposite path — it trades a very shallow socket for a reach that can touch almost any angle of the body. The parts that make this work:
Glenohumeral joint: the true 'shoulder joint'. The humeral head is a large ball; the glenoid on the scapula is a shallow, small dish. It's like a golf ball on a tee — free to point anywhere, but inherently unstable on its ownLabrum: a rim of fibrocartilage set around the shallow dish to deepen it a little, and to anchor the ligaments and the long head of the biceps tendonRotator cuff: the tendons of four small muscles (supraspinatus, infraspinatus, teres minor, subscapularis) wrap the humeral head and actively press the ball into the dish like a cupped hand. It is the shoulder's real stabiliser — dynamic and muscular, not bone-lockedBursa: the subacromial bursa is a fluid-filled friction pad sitting between the cuff and the bone above itAC joint: the small joint where the outer end of the clavicle meets the acromion of the scapula, at the very top of the shoulder
And the most overlooked thing, which explains a large share of shoulder pain: scapular rhythm. When you raise your arm overhead, it isn't only the humerus moving — the scapula glides and rotates across the back of the ribcage at the same time, making room for the humeral head. Arm and scapula cooperate at roughly a 2:1 ratio. Once that rhythm goes off (a slumped posture, rounded shoulders, weak scapular muscles), the space above gets squeezed and the cuff is more easily pinched and abraded.
The first intuition this island builds: the shoulder is not a passively locked hinge but a machine kept stable by muscle, moment to moment. The next scene shows that its pain almost always comes from the after-effects of that stability-for-motion trade.
Mechanism: shallow socket + dynamic stability = mobile but fragile
Spell out both sides of this trade fully — every later diagnosis follows from here.The knee island explained that articular cartilage has no nerves and no blood vessels, so cartilage itself cannot hurt. The same holds at the shoulder, but the sources of shoulder pain lean more toward soft tissue — tendons, bursa, joint capsule, all richly innervated and all painful when compressed, inflamed, or overloaded.
First, what 'shallow' brings:
The glenoid covers only a small part of the humeral head, so bone gives almost no restraint. Stability rests almost entirely on soft tissue — statically on the labrum and ligaments, dynamically on the rotator-cuff muscles tightening in real timeSo the shoulder has two opposite families of trouble: one is can't stay put (dislocation, subluxation, labral tear), common in young people and after trauma; the other is the stabilising tendons being overused into degeneration or pinching (rotator-cuff-related pain), common in older adults and in anyone doing repeated overhead work
Then, why the cuff is a hot spot:
The supraspinatus tendon passes over the top of the humeral head, running through a narrow gap between the head and the acromion and coraco-acromial ligament above it (the subacromial space)When you raise the arm overhead again and again (swimming, painting, lifting, carrying a child), this tendon repeatedly glides through the narrow gap under tension, in a zone with a naturally weaker blood supply — over time it tends to fatigue and degenerateA degenerated tendon weakens, can't hold the humeral head down, the head rides up, the gap narrows further, and a self-reinforcing loop forms
This is why shoulder pain so often shows that characteristic pattern — 'catches at a certain angle, overhead is a struggle, hurts at night when you lie on it'. It's telling you the tendon and bursa in that narrow gap above are the ones complaining, not that 'the joint has worn through'. The next scene takes the real causes of shoulder pain apart one by one.
Chapter 2
Why shoulders hurt
Why shoulders hurt
Shoulder pain is not one disease — it's an umbrella over several different mechanisms. By clinic frequency and age, these five are the most common.
1 · Rotator-cuff-related shoulder pain — the numerical heavyweight and this island's focus. It covers the big category once called 'subacromial impingement': cuff tendinopathy, subacromial bursitis, partial tendon tears. The classic picture is pain through the mid-range arc of lifting (roughly 60-120 degrees), a struggle overhead, and tenderness lying on that side at night. The next page is devoted to it, because the old 'impingement' label misled people for decades.
2 · Frozen shoulder (adhesive capsulitis) — the joint capsule inflames, then gradually thickens and contracts, gripping the humeral head so the joint won't move in any direction, with passive external rotation especially restricted (the key difference from cuff pain: with cuff pain you can't lift the arm yourself but someone else can move it for you; with frozen shoulder, nobody can). What torments people is that it is very painful — but the good news is that it usually runs its own course: freezing → frozen → thawing, over several months to a year or two. Common in people with diabetes and in women aged 40-60.
3 · Instability / dislocation — the price of the shallow socket. In young people, trauma (a fall onto an outstretched hand, contact sport) pops the humeral head out of the glenoid, often tearing the labrum (a Bankart lesion). After one dislocation the chance of re-dislocation rises sharply, especially in the young. This is the can't-stay-put pole, managed on completely different logic from cuff degeneration.
4 · AC joint problems — a very localised pain, right at the small bump on the top of the shoulder. A fall onto the shoulder (driving the outer clavicle down) can sprain or even separate the AC joint; degenerative osteoarthritis of the AC joint is also common in older adults, showing as top-of-shoulder pain when the arm crosses the body (reaching for the opposite shoulder).
5 · Referred pain — the family to be most wary of. Shoulder pain doesn't always come from the shoulder: a compressed cervical nerve root can radiate pain into the shoulder and upper arm; and left shoulder pain + chest tightness / breathlessness / cold sweat, especially worse on exertion, can be the heart sounding an alarm — covered carefully in the 'red flags' scene.
Beyond these five there is also long-head-of-biceps tendinopathy, calcific tendinitis, rheumatoid arthritis, and more — but in clinic the large majority of shoulder pain falls into the five above, and the first (cuff-related) alone accounts for well over half.
1 · Rotator-cuff-related shoulder pain — the numerical heavyweight and this island's focus. It covers the big category once called 'subacromial impingement': cuff tendinopathy, subacromial bursitis, partial tendon tears. The classic picture is pain through the mid-range arc of lifting (roughly 60-120 degrees), a struggle overhead, and tenderness lying on that side at night. The next page is devoted to it, because the old 'impingement' label misled people for decades.
2 · Frozen shoulder (adhesive capsulitis) — the joint capsule inflames, then gradually thickens and contracts, gripping the humeral head so the joint won't move in any direction, with passive external rotation especially restricted (the key difference from cuff pain: with cuff pain you can't lift the arm yourself but someone else can move it for you; with frozen shoulder, nobody can). What torments people is that it is very painful — but the good news is that it usually runs its own course: freezing → frozen → thawing, over several months to a year or two. Common in people with diabetes and in women aged 40-60.
3 · Instability / dislocation — the price of the shallow socket. In young people, trauma (a fall onto an outstretched hand, contact sport) pops the humeral head out of the glenoid, often tearing the labrum (a Bankart lesion). After one dislocation the chance of re-dislocation rises sharply, especially in the young. This is the can't-stay-put pole, managed on completely different logic from cuff degeneration.
4 · AC joint problems — a very localised pain, right at the small bump on the top of the shoulder. A fall onto the shoulder (driving the outer clavicle down) can sprain or even separate the AC joint; degenerative osteoarthritis of the AC joint is also common in older adults, showing as top-of-shoulder pain when the arm crosses the body (reaching for the opposite shoulder).
5 · Referred pain — the family to be most wary of. Shoulder pain doesn't always come from the shoulder: a compressed cervical nerve root can radiate pain into the shoulder and upper arm; and left shoulder pain + chest tightness / breathlessness / cold sweat, especially worse on exertion, can be the heart sounding an alarm — covered carefully in the 'red flags' scene.
Beyond these five there is also long-head-of-biceps tendinopathy, calcific tendinitis, rheumatoid arthritis, and more — but in clinic the large majority of shoulder pain falls into the five above, and the first (cuff-related) alone accounts for well over half.
Mechanism: from 'impingement' to 'cuff-related pain'
Rotator-cuff-related pain is the biggest slice of shoulder pain, and the slice most distorted by one old word.The old (over-mechanical) model was subacromial impingement: it assumed the narrow gap below the acromion 'bumps' the tendon against bone when you lift, and repeated bumping grinds it down — so logically 'shave off a bit of bone and open the space' should fix it. That was exactly the rationale for the subacromial decompression surgery that later became popular.
The modern understanding leans toward an integrated load-and-degeneration process, which is why the field now uses the more neutral name 'rotator-cuff-related shoulder pain':
Under repeated overhead load the tendon undergoes degeneration (tendinosis) rather than simple inflammation — the same idea as jumper's knee on the knee island: under the microscope you see disorganised collagen and repair lagging behind load, not a crowd of inflammatory cellsAcromion shape and bone spurs do exist, but their causal link to pain has been overstated; plenty of people have the same shape on imaging and no painOff-balance scapular rhythm and weak cuff and scapular muscles let the humeral head ride up and shrink the dynamic space — and that's the part training can change
Why does this label change matter so much? Because it rewrites the treatment logic directly: if the problem is 'bone hits tendon', you should operate to shave bone; if the problem is 'tendon degeneration + faulty muscle control', then exercise rehabilitation becomes the mainline and bone-shaving surgery loses its rationale. The next scene (myths) nails this shift down with a set of placebo-surgery-controlled randomised trials.
One age-related fact in passing: cuff-tendon degeneration with age is very common, and partial and even full-thickness tears on imaging grow steadily more frequent with age in people who have no symptoms — which means 'a tear was imaged' and 'this tear is the cause of your pain' are two entirely different things. That point becomes a key clinical judgment in the 'myths' scene.
Chapter 3
Myths
Myths
The shoulder is one of the densest zones of orthopaedic 'reversals' of the past decade. This scene debunks the myths one by one — each against real evidence.
Myth 1: 'Subacromial impingement / shoulder pain — a decompression operation fixes it'
One of the most important orthopaedic reversals of the last decade; it gets its own page, built on placebo-surgery-controlled trials.
Myth 2: 'An imaged rotator-cuff tear always needs surgical repair'
Not necessarily. First separate two kinds of tear: a young person's acute traumatic tear and an older person's degenerative tear, managed on completely different logic. Degenerative cuff tears are very common in older adults, and many people with a tear feel nothing at all — tears on imaging grow more frequent with age, but symptoms don't necessarily follow. So the right question isn't 'do I have a tear?' but 'are my symptoms caused by this tear in a way surgery can fix?'. For many degenerative, non-traumatic tears, a trial of exercise rehabilitation first is a reasonable opening move (detailed next scene). A young person's large acute tear with clear loss of strength, of course, remains a clear indication for surgical repair.
Myth 3: 'A frozen shoulder must be forced open early / released under anaesthetic'
The most counter-intuitive thing about frozen shoulder is that in most people it runs its own course (freezing → frozen → thawing) over several months to a year or two, and most ultimately recover function quite well. So the mainline is patience + maintaining range + controlling pain, not reaching for the aggressive option first. Forcing the joint open too early and too hard (especially during the still-painful freezing phase) can instead worsen the inflammation and prolong the course. Manipulation under anaesthetic (MUA) and capsular-release surgery are options held in reserve for the few who stay stubbornly unchanged over a long time, not a default starting point.
Myth 4: 'Rehab should be no pain no gain — the more it hurts, the better it works'
This is a common way to make a shoulder worse. Cuff-tendon rehab does allow 'pain within an acceptable range', but more is not better — a separate page covers how to find that sweet spot.
Myth 5: 'Shoulder pain is always a shoulder problem'
Not necessarily. The neck, heart, gallbladder, and diaphragm can all refer pain to the shoulder. In particular, left shoulder pain that does not change with shoulder movement but worsens with exertion or walking should prompt ruling out the heart first — covered carefully in the 'red flags' scene.
Myth 1: 'Subacromial impingement / shoulder pain — a decompression operation fixes it'
One of the most important orthopaedic reversals of the last decade; it gets its own page, built on placebo-surgery-controlled trials.
Myth 2: 'An imaged rotator-cuff tear always needs surgical repair'
Not necessarily. First separate two kinds of tear: a young person's acute traumatic tear and an older person's degenerative tear, managed on completely different logic. Degenerative cuff tears are very common in older adults, and many people with a tear feel nothing at all — tears on imaging grow more frequent with age, but symptoms don't necessarily follow. So the right question isn't 'do I have a tear?' but 'are my symptoms caused by this tear in a way surgery can fix?'. For many degenerative, non-traumatic tears, a trial of exercise rehabilitation first is a reasonable opening move (detailed next scene). A young person's large acute tear with clear loss of strength, of course, remains a clear indication for surgical repair.
Myth 3: 'A frozen shoulder must be forced open early / released under anaesthetic'
The most counter-intuitive thing about frozen shoulder is that in most people it runs its own course (freezing → frozen → thawing) over several months to a year or two, and most ultimately recover function quite well. So the mainline is patience + maintaining range + controlling pain, not reaching for the aggressive option first. Forcing the joint open too early and too hard (especially during the still-painful freezing phase) can instead worsen the inflammation and prolong the course. Manipulation under anaesthetic (MUA) and capsular-release surgery are options held in reserve for the few who stay stubbornly unchanged over a long time, not a default starting point.
Myth 4: 'Rehab should be no pain no gain — the more it hurts, the better it works'
This is a common way to make a shoulder worse. Cuff-tendon rehab does allow 'pain within an acceptable range', but more is not better — a separate page covers how to find that sweet spot.
Myth 5: 'Shoulder pain is always a shoulder problem'
Not necessarily. The neck, heart, gallbladder, and diaphragm can all refer pain to the shoulder. In particular, left shoulder pain that does not change with shoulder movement but worsens with exertion or walking should prompt ruling out the heart first — covered carefully in the 'red flags' scene.
Myth 1: decompression no better than placebo surgery
'Shoulder pain = shave bone and decompress' was once one of the most commonly performed shoulder operations in the world — until a set of randomised controlled trials overturned it.CSAW trial (Beard 2018, Lancet): patients with cuff-related shoulder pain (no full-thickness tear) were randomised into three groups — real arthroscopic subacromial decompression, a placebo operation (arthroscopy to look but no decompression), and no surgery with follow-up only. Result: decompression and placebo surgery showed no clinically meaningful difference in pain and function; both surgical groups were marginally better than 'do nothing', but that gap was too small to reach the clinical-importance threshold and very likely came from having had surgery at all (placebo effect + the rehabilitation that inevitably follows)Ketola series of trials: compared subacromial decompression + exercise against exercise alone; over medium-to-long follow-up (including 5 years), adding decompression brought no extra benefit — exercise was the part doing the work
The implication is direct: for the large majority of cuff-related pain without a full-thickness tear and without mechanical catching, the decompression step itself contributes almost no extra improvement. The improvement comes mostly from time, rehabilitation, and the placebo effect, not from the sliver of bone the surgery removes. This is direct confirmation of the label shift from the previous scene: if the problem isn't 'bone hitting tendon', shaving bone naturally misses the point.
Draw the boundary clearly, to avoid over-correcting: this conclusion is about cuff-related pain (no full-thickness tear, no clear mechanical problem). It does not negate genuinely indicated shoulder surgery — repair of a young person's large acute tear, stabilisation for recurrent dislocation, capsular release for a stubborn frozen shoulder. The right order is: build exercise rehabilitation properly first, and reserve surgery for those with a clear structural indication who have genuinely failed conservative care.
Myth: 'a tear on a scan = must repair' & the imaging-symptom mismatch
Spell out the relationship between imaging and symptoms on its own, because at the shoulder it's even starker than at the knee.Age-related cuff-tendon degeneration is the norm. Ultrasound and MRI studies repeatedly find one thing: among ordinary people with no shoulder pain at all, the rate of partial and even full-thickness cuff tears rises steadily with age — by the sixties and seventies, a sizeable share of people have 'a tear but no pain'. This shows that a tear, as an imaging finding, is in many people just a normal mark of ageing, not necessarily the culprit behind pain.
Two direct corollaries:
'Scan first, then decide' often backfires at the shoulder: a scan will very likely find something (degeneration, a tear, a spur), and that finding then pulls both patient and clinician's attention onto it, leading to over-treatment and even unnecessary surgery. So for shoulder pain with no red flags and modest symptoms, guidelines usually advise a period of conservative care first, without rushing to imaging'I have a tear, so it must be repaired' is a skipped-step argument: what actually needs judging is whether this tear is the source of the symptoms and the type surgery can fix. Degenerative, non-traumatic tears usually start with exercise rehabilitation; a young person's acute traumatic large tear with clear loss of strength is the clear surgical-repair population
This is the same logic as the degenerative meniscal tear on the knee island: scans find them everywhere and many are symptomless; the point isn't whether one exists but whether it's the cause of your pain and the kind surgery can fix. Holding onto this helps you stay un-frightened in front of a report full of jargon and resist being pushed onto an unnecessary operating table.
Chapter 4
What actually works
What actually works
With the myths cleared, here's what genuinely improves shoulder pain. From strongest evidence down — and the first line is almost entirely non-surgical — sorted by the type of shoulder pain.
Cuff-related pain · first line = exercise therapy (progressive loading)
This is the core treatment for most cuff-related shoulder pain, not an add-on. Same underlying logic as tendinopathy on the knee island: a degenerated tendon doesn't like rest, it likes to be loaded just right, and progressive tension stimulates it to reorganise collagen and grow stronger and thicker.
Cuff strength: train external rotation, internal rotation, and abduction within a pain-free or acceptable-pain range (commonly with a resistance band or light dumbbells)Scapular stability: train the serratus anterior and the mid/lower trapezius to restore scapular rhythm and make room above for the humeral headControlled load, progressively increased: several randomised trials show structured exercise rehabilitation matches surgery (decompression) over the medium-to-long term — which is exactly the active ingredient behind the CSAW / Ketola reversal
Cuff-related pain · two things done alongside
Relative rest, not complete immobility: temporarily cut the overhead load that aggravates pain (don't paint the ceiling or heave things overhead for a while), but keep moving through the rest of the range. Full immobilisation only makes the shoulder stiffer and the tendon weakerPatient education: understanding 'this is a tendon-degeneration and load-management problem, not a worn-through joint that must be operated on' itself lowers fear and raises willingness to stick with rehab — that's part of the effect, not empty words
Frozen shoulder · first line = reassurance + maintaining range + pain control + time
The mainline for frozen shoulder differs from cuff pain, because it usually runs its own course:
In the very painful early phase, the focus is on controlling pain and doing gentle range-of-motion work within a non-severe range, without forcing itAn early intra-articular steroid injection can shorten the most painful stage and help you start range work sooner — one of the few interventions with reasonable evidenceStress patience repeatedly: most people improve markedly over months to a year or two, and simply knowing this lightens the anxiety
Instability / dislocation
After a first traumatic dislocation, rehabilitation (restoring cuff and scapular control) leads; but for young, highly active, repeatedly dislocating people, the indication for stabilisation surgery is far clearer than in older adults
Medication (symptomatic, doesn't change the course)
A short course of oral or topical NSAID buys 'a window in which you can start training' during an acute painful phase — it relieves symptoms but doesn't repair the tendon; mind GI / cardiac / renal risks
Order matters: build the foundation of progressive loading and patience first; reserve injections and surgery for those with a clear indication who have genuinely failed conservative care. Reverse the order — reaching first for a needle or a scalpel — and you usually spend money while missing the most effective window.
Cuff-related pain · first line = exercise therapy (progressive loading)
This is the core treatment for most cuff-related shoulder pain, not an add-on. Same underlying logic as tendinopathy on the knee island: a degenerated tendon doesn't like rest, it likes to be loaded just right, and progressive tension stimulates it to reorganise collagen and grow stronger and thicker.
Cuff strength: train external rotation, internal rotation, and abduction within a pain-free or acceptable-pain range (commonly with a resistance band or light dumbbells)Scapular stability: train the serratus anterior and the mid/lower trapezius to restore scapular rhythm and make room above for the humeral headControlled load, progressively increased: several randomised trials show structured exercise rehabilitation matches surgery (decompression) over the medium-to-long term — which is exactly the active ingredient behind the CSAW / Ketola reversal
Cuff-related pain · two things done alongside
Relative rest, not complete immobility: temporarily cut the overhead load that aggravates pain (don't paint the ceiling or heave things overhead for a while), but keep moving through the rest of the range. Full immobilisation only makes the shoulder stiffer and the tendon weakerPatient education: understanding 'this is a tendon-degeneration and load-management problem, not a worn-through joint that must be operated on' itself lowers fear and raises willingness to stick with rehab — that's part of the effect, not empty words
Frozen shoulder · first line = reassurance + maintaining range + pain control + time
The mainline for frozen shoulder differs from cuff pain, because it usually runs its own course:
In the very painful early phase, the focus is on controlling pain and doing gentle range-of-motion work within a non-severe range, without forcing itAn early intra-articular steroid injection can shorten the most painful stage and help you start range work sooner — one of the few interventions with reasonable evidenceStress patience repeatedly: most people improve markedly over months to a year or two, and simply knowing this lightens the anxiety
Instability / dislocation
After a first traumatic dislocation, rehabilitation (restoring cuff and scapular control) leads; but for young, highly active, repeatedly dislocating people, the indication for stabilisation surgery is far clearer than in older adults
Medication (symptomatic, doesn't change the course)
A short course of oral or topical NSAID buys 'a window in which you can start training' during an acute painful phase — it relieves symptoms but doesn't repair the tendon; mind GI / cardiac / renal risks
Order matters: build the foundation of progressive loading and patience first; reserve injections and surgery for those with a clear indication who have genuinely failed conservative care. Reverse the order — reaching first for a needle or a scalpel — and you usually spend money while missing the most effective window.
Clinical: how to train the cuff + how to handle frozen shoulder + who gets surgery
Make 'you should exercise' concrete and executable.How to train (cuff-related pain)
Cuff: band external/internal rotation (elbow tucked to the side), lateral raises to a pain-free height, Y/T/W patterns. Start at pain-free or mild acceptable pain and progress resistance graduallyScapula (often neglected): serratus anterior (wall push-up plus, protraction at the top of a press), mid/lower trapezius (prone Y/T) to restore the upward-rotation rhythm of the scapula as the arm liftsOverhead work: avoid the aggravating high-arc load until pain settles, then progressively add overhead movements backFrequency and timeline: 2-3×/week, sustained for weeks to months before stable improvement shows — the cuff tendon is a long game, like the knee; don't expect results in days
How to gauge 'pain' in rehab (answering Myth 4)
Progressive loading allows 'pain within an acceptable range': a common rule is ≤ 3-5 on a 0-10 scale during training, and not worse within 24 hours'A bit achy/sore but not worsening' is fine to train through; 'more painful each session, worse the next day' is the cue to reduce. Pain is a signal, not a prohibition — and not better in larger doses
How to handle a frozen shoulder
The mainline is patience + maintaining range within tolerance + pain control, not forcing it openAn early steroid injection can shorten the most painful stage; manipulation under anaesthetic / capsular release is only for the few who stay stubbornly unchanged over a long time
Who gets surgery (last resort)
Cuff repair: a young person's acute traumatic (large) tear with clear loss of strength — a clear indication. Degenerative, non-traumatic tears mostly start with rehabStabilisation surgery: young, repeatedly dislocating people with a labral (Bankart) lesionSubacromial decompression: after CSAW / Ketola, no longer a routine recommendation for plain cuff-related painFrozen-shoulder release (MUA / capsular release): only the rare person who stays stubbornly frozen far beyond the typical course
One line to close this island's treatment logic: what truly changes the disease course is progressive loading and time; injections and surgery manage symptoms, specific structural problems, and the end stage. It's the same theme as the knee island — the shoulder is almost always a story of 'move, but move smart', not a story of 'spare it, don't move it'.
Chapter 5
Tendon & load
Tendon & load
Cuff pain and jumper's knee from the knee island are the same underlying story, and this mechanism deserves spelling out on its own, because it directly decides how you should treat your own shoulder.
The key shift: tendinitis → tendinosis
Cuff pain used to be called 'rotator-cuff tendinitis', assuming inflammation, so the logic was 'anti-inflammatories + rest'. Modern histology found that the vast majority of chronic cuff-tendon pain is actually degeneration (tendinosis): under the microscope you see disorganised collagen and repair lagging behind load, with very few inflammatory cells.
This understanding rewrites the treatment directly: since it isn't inflammation, anti-inflammatories and complete rest won't cure it.
An anti-inflammatory drug has little to act on at an 'inflammation' target that barely exists — short-term pain relief, yes; a cure, noComplete rest further lowers the tendon's collagen synthesis and mechanical strength, effectively dismantling its load capacity — the more a shoulder rests, the weaker and stiffer it gets
So what makes a tendon stronger
Controlled, progressive tension. Loading the tendon just right stimulates it to reorganise collagen and gain strength. This is the same toolkit validated on the knee tendinopathy island, transferred to the shoulder:
Eccentric training: Alfredson 1998's classic Achilles eccentric protocol established the 'progressive load repairs tendon' paradigm, and the principle applies to the cuff tooHeavy slow resistance (HSR): Kongsgaard 2009 showed on the patellar tendon that at 12 weeks it matches eccentric work with better adherence — an efficient way to load a tendonIsometric contractions: Rio 2015 showed isometrics give immediate analgesia in the acute phase, letting you start loading safely even while it's quite painful
How to find the sweet spot
The cuff tendon doesn't like rest — it likes to be loaded just right. Dial load into the 'challenging but not aggravating' window:
Pain during training/daily movement within an acceptable range (commonly ≤ 3-5 on a 0-10 scale), and not worse within 24 hours — keep loading inside that windowTreat pain as a signal, not a prohibition: 'a bit sore but not worsening' is fine to continue; 'more painful each session, worse the next day' is the cue to reduce
In one line: the shoulder is almost always a story of 'move, but move smart'. Internalise this mechanism and you'll stop being led by the 'rest it and take anti-inflammatories' intuition — the very intuition that, in many people, lets the shoulder grow stiffer and weaker the longer it's spared.
The key shift: tendinitis → tendinosis
Cuff pain used to be called 'rotator-cuff tendinitis', assuming inflammation, so the logic was 'anti-inflammatories + rest'. Modern histology found that the vast majority of chronic cuff-tendon pain is actually degeneration (tendinosis): under the microscope you see disorganised collagen and repair lagging behind load, with very few inflammatory cells.
This understanding rewrites the treatment directly: since it isn't inflammation, anti-inflammatories and complete rest won't cure it.
An anti-inflammatory drug has little to act on at an 'inflammation' target that barely exists — short-term pain relief, yes; a cure, noComplete rest further lowers the tendon's collagen synthesis and mechanical strength, effectively dismantling its load capacity — the more a shoulder rests, the weaker and stiffer it gets
So what makes a tendon stronger
Controlled, progressive tension. Loading the tendon just right stimulates it to reorganise collagen and gain strength. This is the same toolkit validated on the knee tendinopathy island, transferred to the shoulder:
Eccentric training: Alfredson 1998's classic Achilles eccentric protocol established the 'progressive load repairs tendon' paradigm, and the principle applies to the cuff tooHeavy slow resistance (HSR): Kongsgaard 2009 showed on the patellar tendon that at 12 weeks it matches eccentric work with better adherence — an efficient way to load a tendonIsometric contractions: Rio 2015 showed isometrics give immediate analgesia in the acute phase, letting you start loading safely even while it's quite painful
How to find the sweet spot
The cuff tendon doesn't like rest — it likes to be loaded just right. Dial load into the 'challenging but not aggravating' window:
Pain during training/daily movement within an acceptable range (commonly ≤ 3-5 on a 0-10 scale), and not worse within 24 hours — keep loading inside that windowTreat pain as a signal, not a prohibition: 'a bit sore but not worsening' is fine to continue; 'more painful each session, worse the next day' is the cue to reduce
In one line: the shoulder is almost always a story of 'move, but move smart'. Internalise this mechanism and you'll stop being led by the 'rest it and take anti-inflammatories' intuition — the very intuition that, in many people, lets the shoulder grow stiffer and weaker the longer it's spared.
Myth: 'shoulder pain = anti-inflammatory + rest' often doesn't work
Pull this one out on its own, because it harms the most people, and is especially typical at the shoulder.The common approach (often ineffective): at the first sign of shoulder pain, 'put it in a sling / immobilise it / anti-inflammatories + complete rest'. Short-term it may not hurt (because there's no load), but the moment activity resumes it recurs, on and off for years; worse, prolonged immobility can also invite capsular contracture, dragging a cuff pain into a near-frozen stiffness.
Why it's wrong:
Chronic cuff pain is fundamentally degeneration with few inflammatory cells, so an anti-inflammatory drug has little to act on at a target that barely existsThe shoulder is especially sensitive to immobilisation: it relies on muscle for moment-to-moment stability, so once it stops moving, cuff and scapular strength drop quickly and the capsule tightens — a vicious cycleThis also clashes with modern acute-injury care: even for acute soft-tissue injury, the POLICE principle (Bleakley 2012) has replaced the old 'Rest' with 'Optimal Loading' — total immobility only slows healing
The right approach:
Not stopping load, but dialling it into the sweet spot (the ≤ 3-5 on 0-10, not-worse-in-24-hours window from the previous page), and keep loading progressively within itTemporarily avoid the aggravating high-arc overhead movements (relative rest), but keep moving through the rest of the range — don't 'vault' the whole armUse eccentric / heavy slow resistance / isometric to give the tendon progressive tension, while training scapular stability to restore the rhythm
This is also why this island repeatedly stresses: the shoulder (cuff pain or the thawing phase of a frozen shoulder alike) is almost always a story of 'move, but move smart', not a story of 'sling it, don't touch it'. The only things that genuinely need immobilising or prompt medical care first are the few red flags in the next scene.
Chapter 6
Red flags
Red flags
Everything this island says about 'move, don't panic' rests on one premise: your shoulder pain has a common, benign cause. But several situations are genuine red flags — they don't need a rehab plan, they need prompt medical care, and a few are emergencies. Please read the following carefully.
Call emergency services / go to the ED immediately (possibly the heart)
Left shoulder or left arm pain together with chest tightness, chest pain, a pressure sensation, breathlessness, cold sweat, or nausea — especially if it worsens with exertion or activity and isn't relieved by rest — may be angina or a heart attack referring to the shoulder. This is an emergency: call for emergency help immediately, do not rub it away as a 'crick' or 'frozen shoulder'. One key distinguishing point: cardiac pain usually does not change when you rotate or press on the shoulder.
Needs immediate emergency evaluation (possible septic arthritis)
A shoulder that is red, swollen, and hot, together with fever, chills, and pain so severe you can barely move it — this is a joint-space infection, a surgical emergency, and bacteria can destroy the joint within hours. Do not apply heat, do not push through it, do not wait until tomorrow — go to the emergency department immediately.
Needs immediate medical care (acute trauma)
Visible deformity + inability to move after trauma: after a fall or impact, the shoulder's shape has changed, the arm can't be raised, and there's severe pain — possibly a dislocation or fracture needing prompt reduction/treatment; do not force it back yourself
Needs prompt medical care (not self-rehab)
Progressive loss of strength / numbness: a sudden inability to power certain arm movements, a numb hand, or muscle wasting — suggests a larger tendon tear or nerve involvementNight pain + systemic symptoms: persistent pain at rest and at night, with unexplained weight loss, persistent low-grade fever, night sweats, or a palpable lump in the shoulder / above the collarbone — needs work-up for infection, inflammatory arthritis, or tumour (an apical lung tumour can present as shoulder pain)Keeps worsening despite training, no improvement over weeks: if proper relative rest + progressive rehab over several weeks makes it worse rather than better, it's worth going back for review to rule out an overlooked structural problem
This site provides general education and advice — it does not replace a physician's diagnosis and treatment. For any shoulder pain that is persistent, worsening, or carries the red flags above, see an orthopaedic, sports-medicine, or rheumatology doctor in person for assessment; if the heart is suspected, go straight to the emergency department.
Call emergency services / go to the ED immediately (possibly the heart)
Left shoulder or left arm pain together with chest tightness, chest pain, a pressure sensation, breathlessness, cold sweat, or nausea — especially if it worsens with exertion or activity and isn't relieved by rest — may be angina or a heart attack referring to the shoulder. This is an emergency: call for emergency help immediately, do not rub it away as a 'crick' or 'frozen shoulder'. One key distinguishing point: cardiac pain usually does not change when you rotate or press on the shoulder.
Needs immediate emergency evaluation (possible septic arthritis)
A shoulder that is red, swollen, and hot, together with fever, chills, and pain so severe you can barely move it — this is a joint-space infection, a surgical emergency, and bacteria can destroy the joint within hours. Do not apply heat, do not push through it, do not wait until tomorrow — go to the emergency department immediately.
Needs immediate medical care (acute trauma)
Visible deformity + inability to move after trauma: after a fall or impact, the shoulder's shape has changed, the arm can't be raised, and there's severe pain — possibly a dislocation or fracture needing prompt reduction/treatment; do not force it back yourself
Needs prompt medical care (not self-rehab)
Progressive loss of strength / numbness: a sudden inability to power certain arm movements, a numb hand, or muscle wasting — suggests a larger tendon tear or nerve involvementNight pain + systemic symptoms: persistent pain at rest and at night, with unexplained weight loss, persistent low-grade fever, night sweats, or a palpable lump in the shoulder / above the collarbone — needs work-up for infection, inflammatory arthritis, or tumour (an apical lung tumour can present as shoulder pain)Keeps worsening despite training, no improvement over weeks: if proper relative rest + progressive rehab over several weeks makes it worse rather than better, it's worth going back for review to rule out an overlooked structural problem
This site provides general education and advice — it does not replace a physician's diagnosis and treatment. For any shoulder pain that is persistent, worsening, or carries the red flags above, see an orthopaedic, sports-medicine, or rheumatology doctor in person for assessment; if the heart is suspected, go straight to the emergency department.