What Is Shoulder Pain? — A Ball on a Shallow Saucer, the Hand That Holds It and the Bag Around It

"My Arm Won't Go Up"

A 55-year-old office worker reaches for a box on a top shelf, and the outside of their shoulder shoots with pain. For weeks, lying on that shoulder at night makes them wake up in pain. A 52-year-old colleague on the same floor has a different worry. Their shoulder had been aching for months, and now it is hard to reach their hand behind their back to get dressed, and even when family members try to lift their arm, it stops dead at a certain point. Neither person has ever fallen or bumped themselves.

Shoulder pain is common. The proportion of people who experience shoulder pain at least once in their lifetime is estimated to be close to 70% (PMID 40118541). Most of it happens without an injury (PMID 42606850). That pain usually comes from conditions in the soft tissues around the joint that subside on their own, and rarely, a serious cause is hidden behind it (PMID 42606850).

How the Shoulder Is Held Together

A single shoulder handles reaching the arm overhead, rotating it to the side, and wrapping it behind the back. The shoulder is a joint with an extremely wide range of motion, and in exchange, it has less bone support, making it prone to injury (PMID 40118541).

A joint is the meeting place where two bones move together, made of cartilage covering the bone ends and a joint capsule wrapping around them. The two bones that meet in the shoulder joint are the shoulder blade and the upper arm bone. The shoulder blade (scapula) is a wide, triangular bone attached to both upper sides of the back, featuring a shallow dish at its outer end that receives the arm bone. The humeral head is the rounded end at the very top of the upper arm bone, resting in the shallow dish of the shoulder blade to act as the ball of the shoulder joint.

The job of holding the ball in place is shared between the capsule and the muscles. The joint capsule is a tough membranous bag that entirely encloses the ends of the two bones forming a joint. A tendon is a tough cord that attaches the end of a muscle to a bone, transmitting the force generated by the muscle to the bone. The rotator cuff consists of four muscles and their tendons that originate on the shoulder blade and attach as if wrapping around the humeral head, holding the humeral head in place while rotating and lifting the arm.

There is a roof at the very top of the shoulder. The acromion is the bone end where the shoulder blade extends to the top of the shoulder, covering above the humeral head like a roof. The upper part of the rotator cuff tendon passes beneath this roof.

If you draw the shoulder, it is a ball resting on a shallow base. The hand holding and turning the ball is the rotator cuff, and the capsule wrapping the ball and base entirely is the joint capsule. The analogy holds up to this point. Because the hand and capsule are living tissues, their appearance changes with age, and as we will see later, those changes do not necessarily mean pain.

In exchange for moving widely, the shoulder relies on tendons and a capsule rather than bones to stay connected. The two common branches of shoulder pain also stem from here. The holding hand gets damaged, or the wrapping capsule shrinks.

<!-- FIG: Front of the shoulder — Humeral head (ball) on the shallow dish of the shoulder blade, enclosing joint capsule, rotator cuff tendons, and the acromion covering above them -->

When the Hand That Reaches Out Hurts

Like the shoulder of a 55-year-old reaching for a shelf, pain and weakness when lifting the arm or turning it outward are the hallmark signs of a rotator cuff problem (PMID 26390274). Pain originating from the rotator cuff is the most common cause of shoulder pain (PMID 41276811).

There are stages to the changes in the rotator cuff tendon. Tendinopathy is a state where, related to heavy use of the tendon, the tendon hurts and fails to perform its full function. In the rotator cuff, a change where the internal weave of the tendon becomes disordered comes first, which progresses to a partial tear, and then again to a tear where the entire thickness is severed (PMID 38332156). A rotator cuff tear is a tear of the rotator cuff tendon, divided into a partial-thickness tear where only part of the tendon thickness is torn and a full-thickness tear where the entire thickness is severed. Rotator cuff disease is an inclusive term for the changes in the rotator cuff tendon that run from tendinopathy to partial-thickness tears and full-thickness tears.

Some tears happen from an injury, but most rotator cuff problems happen as the tendon changes with age (PMID 38332156).

Torn tendons do not all follow the same path. Not all tears hurt, not all tears grow larger, and among people whose tears grow larger, many do not see their symptoms worsen (PMID 38332156). When looking together at studies following unoperated full-thickness tears for an average of 3 to 4 years, roughly one in three grew by more than 5 mm, and there was no clear difference between tears that had hurt and those that had not (PMID 30472019). The authors of this analysis, which pooled studies of varying quality, viewed that based on a 5 mm threshold, most tears do not grow large over short or intermediate periods (PMID 30472019). On the other side, there is also a review noting that tears that did not hurt have a high likelihood of becoming painful over time (PMID 31246863).

Most rotator cuff tears are changes in the tendon that come with age, and being torn does not mean they will all hurt or grow larger.

The Explanation That It "Pinches and Wears Down"

The idea that shoulder pain heals only when you shave away a bone spur and the edge growing under the acromion came from American orthopedic surgeon Charles Neer in 1972, and surgery to shave the acromion originated here (PMID 30774463). Impingement syndrome is a name derived from the explanation that when you lift your arm, the rotator cuff tendon pinches beneath the acromion and hurts. This explanation went unchallenged for at least 30 years (PMID 30774463).

Afterwards, questions were raised about the role "impingement" plays in shoulder pain and the place of surgery (PMID 30774463). A guideline created collaboratively by the Dutch Orthopaedic Association chose the name "subacromial pain syndrome," stating that the anatomical explanation of a pinched tendon alone cannot cover this disease (PMID 24847788).

One review article writes that physical examination tests meant to sort out this diagnosis come back positive in nearly half of people who do not have impingement syndrome, going so far as to call the impingement theory outdated and this diagnosis close to a "medical myth" (PMID 31890103). The opposing view remains. Most studies showed no difference between surgery and non-surgical treatments, but some found long-term outcomes after surgery were better (though the quality of the studies was generally low), and the view is that breaking down types of subacromial damage can explain the group that responds to surgery (PMID 30774463).

The picture that the tendon pinches beneath the acromion and wears down is a long-ruling explanation, and it is currently under debate.

When the Enclosing Pouch Stiffens

A fifty-two-year-old shoulder that cannot reach behind the back looks different. Even when a family member tries to lift the arm, it is blocked. Two terms separate these differences. Active range of motion is the range in which a joint can move using the strength of one's own muscles. Passive range of motion is the range in which a joint moves when another person moves the relaxed arm. If both ranges are reduced in an uninjured shoulder, adhesive capsulitis or arthritis of the shoulder joint is considered first, and imaging is used to distinguish between the two; if both ranges remain, physical examination helps narrow down other causes (PMID 40118541).

Adhesive capsulitis (frozen shoulder) is a condition in which the shoulder's joint capsule thickens and shrinks, causing the shoulder to hurt and stiffen in all directions. It becomes painful while active and passive ranges of motion gradually decrease together (PMID 21885699), and movement turning the arm outward is particularly prone to reduction (PMID 30811157).

Inflammation is a response in which the body gathers immune cells, swells the area, and reacts against stimuli such as injury, infection, or substances introduced from outside the body. Adhesive capsulitis is suspected to be a disease in which the inflammatory response of the joint capsule progresses toward fiber accumulation and shrinkage, and looking into the tissue reveals an increase in fiber-making cells that pile up tough fibers like thick bands (PMID 40545326). Why it starts is still unknown (PMID 27180923).

To put it in terms of a ball on a socket, a rotator cuff problem is an injury to the hand holding the ball, whereas adhesive capsulitis is a shrinkage of the pouch enclosing the ball and socket. A shoulder with an injured hand usually goes up when someone else lifts it, but a shoulder with a shrunken pouch is blocked even when someone else lifts it.

Adhesive capsulitis affects 2 to 5% of people and is more common in women (PMID 40545326). Diabetes is a condition in which blood glucose levels remain higher than the normal range for a long time. Adhesive capsulitis was 5 times more common in people with diabetes than in the control group (PMID 27331029). Those authors suggested considering diabetes screening for people who present with adhesive capsulitis (PMID 27331029). A more recent analysis also confirmed a strong association, but noted that other unmeasured factors may have been mixed in (PMID 41896299). Thyroid disease is also cited as a risk factor, with reviews pointing to an excess of thyroid hormones coexisting with reviews pointing to a deficiency (PMID 40545326, PMID 30811157).

The course has been described as three stages: freezing, frozen, and thawing (PMID 40545326). This explanation, which originated in the 1940s, was that people go through three stages and recover completely without treatment, but an analysis pooling studies following untreated people found no evidence for that (PMID 27641499). At 1 to 4 years of follow-up for untreated people, movement improved to some extent but did not fully return (low quality of evidence), and most of the improvement came early (PMID 27641499). While some reviews state that symptoms resolve without surgery for the majority (PMID 40545326), evidence is also accumulating that symptoms persist for a long time and do not fully recover in many people (PMID 41581632).

Adhesive capsulitis mostly gets better, but for many people it lasts a long time or does not fully return.

Tears in the Picture and the Pain

When both shoulders of 602 randomly selected adults aged 41 to 76 in Finland were imaged with MRI, something appeared in the rotator cuff of nearly everyone (98.7%) (PMID 41697693). The abnormalities appeared in nearly equal proportions in painful and pain-free shoulders (98% vs 96%). Full-thickness tears alone were more common in painful shoulders, but once other factors like age were adjusted for, that difference nearly vanished (PMID 41697693). The authors concluded that rotator cuff abnormalities after age 40 are present in almost anyone, do not match symptoms well, and in many cases may be normal age-related changes. They also questioned the routine use of imaging for shoulder pain without a history of injury (PMID 41697693). This is a survey conducted in a single country at a single point in time.

The pattern of increasing with age is the same in other surveys. In a 1999 study that examined 411 volunteers with pain-free shoulders using ultrasound, tears not separated into partial- and full-thickness increased from 13% in people in their 50s to 51% in those aged 80 and older (49 people) (PMID 10471998). In a survey of 664 residents from their 20s to their 80s in a Japanese village, full-thickness tears also increased with age, and about two-thirds of them were pain-free tears (PMID 24403741).

There are also surveys suggesting that tears in the picture have meaning. In a cohort of 1,003 British women aged 64 to 87, about half of full-thickness tears were pain-free, but people with a full-thickness tendon tear on even one side were more likely to have symptoms than those with healthy tendons on both sides (PMID 36100305). However, until the tear exceeded 2.5 cm, there was no relationship between the size of the tear and the severity of symptoms (PMID 36100305). Even among 106 people who had never visited a hospital for their shoulders, only full-thickness tears lowered shoulder function scores (PMID 31701233).

Because rotator cuff tears are common with age even in pain-free shoulders, we cannot immediately say that a tear seen on a picture is causing current pain. While there are surveys where full-thickness tears showed an association with symptoms and function (PMID 36100305, PMID 31701233), there are also surveys where that difference nearly disappeared once other factors were adjusted for (PMID 41697693).

Therefore, imaging is not rushed for shoulders without an injury. Overdiagnosis is the process of finding and diagnosing a condition that would have passed without harm if it had not been found. Unless there are suspicious signs such as a major injury, fever, unexplained weight loss, or a history of cancer, early imaging is unnecessary. This is because structural abnormalities often do not match symptoms well, meaning they rarely change treatment and can lead to overdiagnosis and overtreatment (PMID 42606850). Guidelines generally do not recommend MRI for early shoulder pain, and recommendations regarding X-rays are mixed (PMID 37832814).

<!-- FIG: Age and Tears — Bar chart showing tear rates from a 1999 study of 411 pain-free volunteers: 50s 13%, 60s 20%, 70s 31%, 80s and older 51%, with volunteers and a single study specified in the caption -->

The First Line Is Explanation and Exercise, Injections Are Short

The first treatment after ruling out serious causes is mostly similar. It consists of explaining that the condition usually runs a favorable course, reducing pain and modifying pain-aggravating activities if needed, and watchful waiting (PMID 42606850). Most people with subacromial pain recover completely with minimal intervention (PMID 42606850).

Guidelines addressing rotator cuff problems all recommend active treatments like exercise, while positioning pain relievers, nonsteroidal anti-inflammatory drugs, and injections as recommended means for reducing pain (PMID 32007452). One review notes that symptoms steadily improve with progressively increasing exercise, to a degree comparable to the results of surgical trials (PMID 26390274). The principles of that exercise are relative rest by reducing painful activities, starting with exercises that do not provoke pain, and controlled reloading to progress from simple to complex movements (PMID 26390274).

How much exercise adds remains debated. Compared to unexercised controls, the exercise group showed improved pain, but the quality of the evidence was very low (PMID 28630217). In one high-quality trial comparing a common clinical practice of manual therapy combining joint and tissue mobilization with exercise against a sham treatment, there was no clinically important difference (PMID 27283590). In a randomized trial from the UK involving 708 participants, up to six sessions of a progressive exercise program were no better than a single well-structured session of advice from a physiotherapist in terms of pain and function over 12 months (PMID 34265255).

The first line is explanation, activity modification, and exercise, and most people get better, but it remains unclear how much more an exercise program adds compared to well-structured advice.

Injections are placed on top of that. A steroid injection is an injection that uses a needle to deliver inflammation-calming steroid medication close to the painful site. In rotator cuff problems, steroid injections reduced pain compared to no treatment (very low quality of evidence) (PMID 28630217), and when compared to manual and exercise therapy, pain and function were similar, but more people reported treatment success in the injection group up to 11 weeks (low quality of evidence) (PMID 27283590). However, in the prior UK trial, subacromial steroid injections provided no benefit over 12 months (PMID 34265255).

In frozen shoulder, intra-articular steroid injections were the only treatment that clearly outperformed other treatments in short-term pain and function (PMID 33326025). Using them in the early stage within less than a year of onset yielded better results, and adding home exercises could add intermediate-term benefits (PMID 33326025). In a trial compared against combined manual therapy and exercise, the superiority of the injection was seen up around 7 weeks, with no clinically important differences at 6 and 12 months (PMID 25157702). For people with diabetes, the potential impact of steroids on metabolism is also considered (PMID 40545326).

Steroid injections work shortly, and evidence for long-lasting benefit is weak.

Two Surgeries With Different Evidence

A placebo surgery is a fake surgery where everything is done the same in the operating room up to looking inside the joint with a camera, but only the core step believed to produce an effect is omitted. The surgery tested with this method is subacromial decompression. Subacromial decompression is a surgery that uses a camera and instruments inserted through small holes to shave off bone spurs and soft tissue under the acromion, aiming to widen the space through which the tendon passes.

A trial involving 313 people across 32 hospitals in the UK divided people who had subacromial pain for more than three months, whose rotator cuff tendon was not torn, and who had already gone through exercise and steroid injections into decompression, placebo surgery, and no treatment (PMID 29169668). At 6 months, there was no difference between decompression and placebo surgery, and both surgery groups were slightly better than no treatment, but the difference fell short of a clinically important magnitude (PMID 29169668). The authors noted that the small difference might be due to a placebo effect or physical therapy after surgery, casting doubt on the value of this surgery (PMID 29169668).

A trial in three Finnish hospitals dividing 210 people aged 35 to 65 into decompression, placebo surgery, and exercise therapy found no advantage of decompression over placebo surgery at 2 years, 5 years, and 10 years (PMID 30026230, PMID 33020137, PMID 41330610). At the 10-year mark, there was no benefit even when compared against exercise therapy (PMID 41330610). A Cochrane review pooling these trials rated the certainty of the evidence as high that subacromial decompression provides no clinically important benefit over placebo surgery in pain, function, and quality of life up to 1 year (PMID 30707445). Serious harms are rare but not zero, with one analysis estimating them at 6 per 1,000 people (moderate certainty) (PMID 30647053).

This evidence concerns decompression for subacromial pain without a full-thickness tear. Bringing together guidelines published up to 2022, 6 recommended or suggested recommending subacromial surgery for chronic rotator cuff tendinopathy, while 4 stated they did not recommend it, leaving discrepancies remaining among guidelines (PMID 37832814).

Repair surgery to stitch up a full-thickness tear has different evidence. A Cochrane review pooling trials comparing repair surgery to nonsurgical treatment in full-thickness tears found that at 1 year, surgery probably reduced pain little to no more (moderate certainty), and not a single trial was compared with a placebo surgery (PMID 31813166). The authors noted that this conclusion was mainly for small tears that developed with age, so it may not apply to tears caused by injury, large tears, and younger people (PMID 31813166).

In a randomized trial dividing 103 people with full-thickness tears of 3 cm or less into repair surgery and physical therapy, shoulder function and pain scores were better on the repair side after 15 years (81% of people remaining), with no difference in quality-of-life scores (PMID 39197154). Among the 51 people in the physical therapy group, 15 later switched to surgery, and the unstitched tears grew by nearly double on average (PMID 39197154). The authors of this single trial on small or medium tears concluded by supporting repair as a first-line treatment, citing the 15-year results (PMID 39197154). The Quebec guidelines stated that surgery is appropriate for some patients with full-thickness tears (PMID 35881707), and the Dutch guidelines stated that there is no reason to operate on rotator cuff tears that do not hurt (PMID 24847788).

Repair surgery is a procedure considered for specific cases among full-thickness tears, and which tears stand to benefit from surgery is still being figured out.

For frozen shoulder, when nonsurgical treatment brings no improvement, methods of moving the arm under anesthesia to break the stiff capsule or surgery to cut and release the capsule inside the joint are considered (PMID 40545326, PMID 21885699). One review noted that manipulation of the arm under anesthesia is associated with joint damage and may be no better than physical therapy, and that high-quality studies on surgery to cut and release the capsule are not yet available (PMID 25107826).

Where It Is Not the Shoulder, and Shoulders That Need Haste

There is a case where the muscle dividing the chest and abdomen was stimulated, yet the shoulder hurt (PMID 15152534). Referred pain is feeling pain in a spot different from where the pain started. This is why, when looking at shoulder pain, serious causes like infection, cancer, and pain carried over from somewhere other than the shoulder are screened out first (PMID 42606850). A problem in the cervical spine can make the shoulder hurt and vice versa, so the two are distinguished by physical examination rather than imaging (PMID 32206519).

Dislocation is when the ends of the bones forming a joint slip out of place. When infection or cancer is suspected, features suggesting fracture, dislocation, loss of function significant enough to prevent using the arm, neurological abnormalities, or systemic inflammatory disease are cited as signals to send a patient for specialist care (PMID 42606850). Shoulder joint infections appear together with pain, reduced movement, swelling, redness, tenderness to touch, and fever in gathered cases that led up to surgical treatment (PMID 29096980). This evidence compiles 27 case reports and case series involving 115 people (PMID 29096980). A disease in which the heart's blood vessels suddenly become blocked makes prompt recognition important for treatment (PMID 32363989). The most common symptom of this disease in both men and women is chest pain, but it can also appear as pain between the shoulder blades, shortness of breath, or nausea, and these presentations were more frequent in women (PMID 32363989). This is the result of gathering what the symptoms of heart disease patients look like, and does not state how much of shoulder pain comes from the heart. An arm that cannot be lifted after a fall is viewed separately from tears that develop with age, and the conclusion from the Cochrane review mentioned earlier that "surgical benefit is small" may not apply to tears caused by injury (PMID 31813166).

When an arm cannot be lifted after a severe injury or the shoulder feels like it has slipped out, when the shoulder swells and turns red along with fever, and when shoulder pain is accompanied by chest pain or shortness of breath, check without waiting. Distinguishing the cause using these signals is the role of physical examination and tests.

Two Shoulders in Front of the Shelf, Again

Let us return to our first two people. The 55-year-old office worker has no history of injury, hurts when lifting the arm and at night, and can raise the arm all the way when someone else lifts it for him. Such a shoulder narrows down its causes through clinical examination (PMID 40118541), among which the most common is pain originating from the rotator cuff (PMID 41276811).

Suppose this person had an MRI that showed a partial-thickness tear. Because such findings are common even in the painless shoulders of people that age, we cannot immediately say that this tear is causing the current pain (PMID 41697693). If there was no injury and no suspicious warning signs, the initial treatment is an explanation that the condition generally runs a favorable course, activity modification, and watchful waiting (PMID 42606850). Guidelines recommend active treatments such as exercise (PMID 32007452), but it remains unclear how much a structured exercise program adds beyond a single well-tailored consultation (PMID 34265255). As for steroid injections placed on top of that, more people reported feeling better for a few weeks (PMID 27283590), but there was no benefit at 12 months (PMID 34265255). If subacromial pain remains even after going through exercise and injections, decompression surgery in such shoulders without a full-thickness tear showed no difference from a sham surgery (PMID 29169668, PMID 30707445).

However, if this person had been unable to lift the arm ever since a fall, the story is different from the start (PMID 42606850). If imaging shows a full-thickness tear, we consider whether it is a case for which repair should be evaluated (PMID 35881707). Yet for small tears that develop with age without injury like this person's, there is an analysis suggesting surgery would have reduced pain little more at 1 year (PMID 31813166), alongside a trial showing that the surgical repair group fared better at 15 years (PMID 39197154).

The 52-year-old coworker has no history of injury, but the arm is blocked even when someone else lifts it. Because both active and passive ranges of motion are reduced together, frozen shoulder or arthritis comes to mind first, and imaging is used to distinguish between the two (PMID 40118541). If this person has diabetes, the association between diabetes and frozen shoulder is strong (PMID 27331029). Intra-articular steroid injections at an early stage within 1 year of onset improved short-term (within 12 weeks) pain and function (PMID 33326025), but at 6 and 12 months, the difference compared to manual therapy and exercise was not distinct (PMID 25157702). Most people improve without surgery, but many also have symptoms that last a long time (PMID 41581632).

Although the pain of the two people looks similar, one likely has a problem with the grabbing hand, while the other has a problem with the enveloping pouch. How to apply this sequence is something to decide with a doctor in the clinic room.

Where to Look Further

There is no consensus yet on which exercise is best (PMID 41276811, PMID 38848304). To know which rotator cuff tears benefit from repair surgery, longer-term trials compared with sham surgery are still needed (PMID 31813166). We also do not yet know why frozen shoulder occurs (PMID 27180923).

The mismatch between what is seen in images and pain is revisited in other contexts in What Is Knee Osteoarthritis? — The Door Frame That Changes With the Hinge and What Is Low Back Pain? — Wrinkles in the Spine, Pain Without a Name. The idea that what a test result tells you depends on pretest probability is covered in What a Test Actually Tells You — An Introduction to Sensitivity, Specificity, and Pretest Probability. Diabetes associated with frozen shoulder begins in What Is Diabetes? — The Letter That Never Comes, the Letter That Isn't Read, and excess or deficient thyroid hormones in What Does the Thyroid Do? — The Letter That Sets the Body's Pace, the Reminder That Grows When It Runs Short. The principles by which exercise becomes a treatment and the restoration of function continue in Rehabilitation — Starting to Plan a Livable Home as the Fire Is Brought Under Control.

What a tendon is and how it becomes painful continue in What Is Tendinopathy? — A Rope That Follows Slowly, a Burden That Arrived Early, and pain that does not heal for months in What Is Chronic Pain? — The Microphone and the Amplifier, a Sound Ringing for Over Three Months.

Source Papers

• "A Simplified Approach to Evaluate and Manage Shoulder Pain." (2024) — PMID 40118541 — Lifetime prevalence of shoulder pain is close to 70%; it is prone to injury due to a wide range of motion and low bony stability; when active and passive ranges of motion decrease together, imaging points to adhesive capsulitis or arthritis, and if both are preserved, special tests narrow down the diagnosis.
• "Management of Shoulder Pain in Primary Care: A Review." (2026) — PMID 42606850 — Third most common musculoskeletal complaint in primary care; mostly non-traumatic, periarticular soft-tissue, and benign self-limiting with rare serious causes; the term subacromial pain is recommended; exclude infection, cancer, and referred pain from outside the shoulder; first-line treatment is education on a good natural course, symptom relief, activity modification, and watchful waiting; early imaging is unnecessary in the absence of trauma, fever, unexplained weight loss, or history of cancer (overdiagnosis and overtreatment); refer to specialists for features of infection, cancer, fracture, dislocation, distinct functional or neurological deficits, and systemic inflammatory diseases; subacromial pain yields no surgical benefit (high certainty), and most recover completely with minimal intervention.
• "Rotator Cuff Tendinopathy: Navigating the Diagnosis-Management Conundrum." (2015) — PMID 26390274 — Rotator cuff tendinopathy is characterized by pain and weakness during external rotation and elevation; correlation between structural damage and symptoms is low; graded exercise achieves improvements comparable to surgery trials, fewer days off work, and faster return; principles are relative rest, modulating painful activities, initially avoiding exercises that worsen pain, controlled reloading, and progressing from simple to complex.
• "Effects of seven types of exercise in the treatment of rotator cuff-related shoulder pain (RCRSP): a systematic review and Bayesian network meta-analysis." (2025) — PMID 41276811 — Rotator cuff-related shoulder pain is the most common cause of shoulder pain; there is no consensus on the optimal exercise.
• "Rotator cuff tears." (2024) — PMID 38332156 — A spectrum ranging from tendinopathy to full-thickness tears accompanied by arthritic changes; some are traumatic but most are degenerative; not all tears cause symptoms or progress, and many people do not get worse even as tears grow; starts with matrix disorganization and inflammatory changes within the tendon, progressing from partial-thickness to full-thickness; muscle size and fatty infiltration affect symptoms, progression, and surgical outcomes; non-surgical treatment is sufficient for most.
• "Full-Thickness Rotator Cuff Tears: What Is the Rate of Tear Progression? A Systematic Review." (2019) — PMID 30472019 — Out of 411 non-surgically treated full-thickness tears, progression by 5mm or more showed no difference between asymptomatic (40.6% at mean 46.8 months) and symptomatic patients (34.1% at 37.8 months); most do not progress significantly in the short-to-medium term (level IV evidence).
• "Asymptomatic Rotator Cuff Tears." (2019) — PMID 31246863 — Asymptomatic tears are common and age-related; they are likely to become symptomatic over time; functional findings vary across studies.
• "Subacromial impingement syndrome: management challenges." (2018) — PMID 30774463 — In 1972, Neer proposed removing bony spurs under the acromion (acromioplasty), which dominated for over 30 years; subsequently, questions arose regarding the role of impingement and the place of surgery; most studies show no difference between surgery and conservative treatment, though some find surgery better long-term; the authors' opinion is that many studies are of low quality and tear types must be considered.
• "Guideline for diagnosis and treatment of subacromial pain syndrome: a multidisciplinary review by the Dutch Orthopaedic Association." (2014) — PMID 24847788 — The anatomical explanation of impingement is insufficient, making "subacromial pain syndrome" more appropriate; prioritize non-surgical treatment; exclude tears with imaging (ultrasound) after 6 weeks; no persuasive evidence shows surgery is better than conservative treatment; asymptomatic tears are not an indication for surgery.
• "Subacromial Impingement Syndrome of the Shoulder: A Musculoskeletal Disorder or a Medical Myth?" (2019) — PMID 31890103 — Diagnostic tests have a specificity averaging about 50%; surgery is no better than conservative treatment; the author's view is that the impingement theory is outdated and a "medical myth."
• "Adhesive capsulitis of the shoulder." (2011) — PMID 21885699 — Fibrosis and contracture of the joint capsule lead to a painful, gradual decrease in active and passive motion; the natural course is debated; start with gentle, progressive stretching, and most improve with non-surgical treatment; consider surgery if treatment fails for 6 months or more.
• "Adhesive Capsulitis: Diagnosis and Management." (2019) — PMID 30811157 — External rotation in particular is reduced; more common in diabetes and hypothyroidism; the traditional view of complete recovery within 1 to 2 years is challenged by evidence of persistent functional limitation; if there is little improvement after 6 to 12 weeks of non-surgical treatment, perform manipulation under anesthesia or arthroscopic capsular release.
• "Frozen shoulder: Diagnosis and treatment of adhesive capsulitis." (2026) — PMID 41581632 — A common cause in middle age; once thought to be self-limiting, many patients experience long-lasting symptoms and incomplete recovery; diagnosis is clinical (generalized restriction of passive motion) and imaging is used to rule out other diseases; steroid injections and physical therapy provide meaningful benefits; surgery is reserved for refractory cases.
• "Adhesive Capsulitis of the Shoulder." (2025) — PMID 40545326 — Inflammatory response is suspected to progress to fibrous contracture; fibroblast proliferation and thick band-like collagen; three stages (freezing, frozen, thawing); risk factors include diabetes, hyperthyroidism, prior fractures, shoulder and neck surgery, and radiation therapy; prevalence is 2–5%, more common in women; early steroid injections are associated with a shorter duration of symptoms; consider metabolic complications of steroids in diabetes; it is self-limiting and mostly resolves without surgery; arthroscopic release or manipulation under anesthesia after 9–12 months of failed non-surgical treatment.
• "Shoulder Stiffness: Current Concepts and Concerns." (2016) — PMID 27180923 — Frozen shoulder is a stiff shoulder of unknown cause; pathology involves joint capsule fibrosis and inflammation, but the exact cause remains unknown; conservative treatment is first-line.
• "Frozen shoulder contracture syndrome - Aetiology, diagnosis and management." (2015) — PMID 25107826 — Evidence for joint capsule adhesions has been refuted, with claims that the term "adhesive capsulitis" should be discarded; manipulation under anesthesia is associated with joint damage and may not be superior to physical therapy; capsular release is supported only by expert opinion and case reports.
• "Adhesive capsulitis of the shoulder and diabetes: a meta-analysis of prevalence." (2016) — PMID 27331029 — Patients with diabetes have 5 times the risk of frozen shoulder compared to control groups; prevalence of frozen shoulder in diabetes patients is 13.4%; prevalence of diabetes in frozen shoulder patients is 30%; consider screening for diabetes in patients with frozen shoulder.
• "The diabetic shoulder: association between diabetes mellitus and adhesive capsulitis - a systematic review and meta-analysis." (2026) — PMID 41896299 — Strong association between diabetes and frozen shoulder (odds ratio 3.69); potential for unmeasured confounding.
• "Natural history of frozen shoulder: fact or fiction? A systematic review." (2017) — PMID 27641499 — No evidence supports the 1940s theory of three stages and complete recovery without treatment; without treatment, only partial improvement occurs in 1 to 4 years (low quality); improvement mostly occurs in the early period (moderate quality).
• "Incidental Rotator Cuff Abnormalities on Magnetic Resonance Imaging." (2026) — PMID 41697693 — A random sample of 602 Finns aged 41–76; rotator cuff abnormalities in 98.7%; 96% in asymptomatic shoulders and 98% in symptomatic ones; only full-thickness tears were more frequent in symptomatic shoulders (14.6% vs 6.5%), but the difference diminished after adjustment; nearly universal after age 40 and poorly correlated with symptoms; may represent normal age-related changes, casting doubt on routine imaging for non-traumatic shoulder pain.
• "Age-related prevalence of rotator cuff tears in asymptomatic shoulders." (1999) — PMID 10471998 — Ultrasound of 411 asymptomatic individuals: 13% in their 50s, 20% in their 60s, 31% in their 70s, and 51% in those 80 and older.
• "Prevalence of symptomatic and asymptomatic rotator cuff tears in the general population: From mass-screening in one village." (2013) — PMID 24403741 — In 664 villagers, full-thickness tears were found in 22.1%, with 65.3% being asymptomatic.
• "Prevalence of rotator cuff tendon tears and symptoms in a Chingford general population cohort, and the resultant impact on UK health services: a cross-sectional observational study." (2022) — PMID 36100305 — Women aged 64–87; full-thickness tears in 22.2%, of whom 48.4% were asymptomatic; presence of a full-thickness tear increases the likelihood of symptoms; tear size and symptom severity are unrelated until exceeding 2.5cm.
• "High prevalence of rotator cuff tears in a population who never sought for shoulder problems: a clinical, ultrasonographic and radiographic screening study." (2020) — PMID 31701233 — In 106 individuals who had never visited a hospital, only full-thickness tears affected functional scores, while partial-thickness tears can be viewed as age-related changes.
• "A Systematic Review of Clinical Practice Guidelines on the Diagnosis and Management of Various Shoulder Disorders." (2024) — PMID 37832814 — 26 guidelines up to August 2022; MRI is generally not recommended for early shoulder pain, while X-ray findings are mixed; pain relievers, anti-inflammatory drugs, and exercise are recommended; acromioplasty for chronic tendinopathy is recommended in 6 guidelines and not recommended in 4.
• "Shoulder Rotator Cuff Disorders: A Systematic Review of Clinical Practice Guidelines and Semantic Analyses of Recommendations." (2020) — PMID 32007452 — All guidelines recommend active treatments like exercise; pain relievers, anti-inflammatory drugs, and steroid injections can be recommended for pain reduction; opinions on imaging and surgery are mixed.
• "Diagnosing, Managing, and Supporting Return to Work of Adults With Rotator Cuff Disorders: A Clinical Practice Guideline." (2022) — PMID 35881707 — Quebec guidelines; imaging only in select cases; pain relievers and injections may be useful for short-term pain reduction; prescription of active rehabilitation (exercise and education); decompression surgery not recommended for tendinopathy; surgery is appropriate for some patients with full-thickness tears.
• "Effectiveness of conservative interventions including exercise, manual therapy and medical management in adults with shoulder impingement: a systematic review and meta-analysis of RCTs." (2017) — PMID 28630217 — Exercise is better than non-exercise controls for pain; steroid injections are better than no treatment; the quality of evidence is very low.
• "Manual therapy and exercise for rotator cuff disease." (2016) — PMID 27283590 — Cochrane review; one high-quality trial comparing manual therapy plus exercise to placebo showed no clinically important difference; generally similar to steroid injections, though treatment success up to 11 weeks was higher with injections (low quality).
• "Progressive exercise compared with best practice advice, with or without corticosteroid injection, for the treatment of patients with rotator cuff disorders (GRASP): a multicentre, pragmatic, 2 × 2 factorial, randomised controlled trial." (2021) — PMID 34265255 — 708 patients in the UK; progressive exercise (up to 6 sessions) was no better than a single session of best practice advice for 12-month pain and function; subacromial steroid injections yielded no long-term benefit.
• "Comparison of Treatments for Frozen Shoulder: A Systematic Review and Meta-analysis." (2020) — PMID 33326025 — Intra-articular steroids alone are statistically and clinically superior for short-term pain and function; earlier use within less than 1 year is associated with better outcomes; adding home exercises may provide additional mid-term benefits.
• "Manual therapy and exercise for adhesive capsulitis (frozen shoulder)." (2014) — PMID 25157702 — Cochrane review; at 7 weeks, steroid injections are better than manual therapy plus exercise; at 6 and 12 months, there are no clinically important differences.
• "Arthroscopic subacromial decompression for subacromial shoulder pain (CSAW): a multicentre, pragmatic, parallel group, placebo-controlled, three-group, randomised surgical trial." (2018) — PMID 29169668 — 313 patients across 32 hospitals in the UK (with subacromial pain lasting over 3 months who had intact tendons and tried exercise and steroid injections); at 6 months, there was no difference between decompression and arthroscopy alone (placebo); both surgeries were slightly better than no treatment but not clinically important; this may be due to a placebo effect or postoperative physical therapy; questions the value of surgery, which should be shared with patients.
• "Subacromial decompression versus diagnostic arthroscopy for shoulder impingement: randomised, placebo surgery controlled clinical trial." (2018) — PMID 30026230 — 210 patients in Finland; no difference between decompression and diagnostic arthroscopy at 24 months; while there was a statistical difference compared to exercise therapy, it fell short of the minimal clinically important difference, with a potential bias toward decompression.
• "Subacromial decompression versus diagnostic arthroscopy for shoulder impingement: a 5-year follow-up of a randomised, placebo surgery controlled clinical trial." (2021) — PMID 33020137 — At 5 years, decompression offers no benefit over diagnostic arthroscopy (or exercise therapy).
• "Arthroscopic subacromial decompression versus placebo surgery for subacromial pain syndrome: 10 year follow-up of the FIMPACT randomised, placebo surgery controlled trial." (2025) — PMID 41330610 — 210 participants aged 35–65; at 10 years, decompression offers no benefit over placebo surgery or exercise therapy.
• "Subacromial decompression surgery for rotator cuff disease." (2019) — PMID 30707445 — Cochrane review; excluding full-thickness tears, no clinically important benefit over placebo in pain, function, or quality of life (high certainty); risk of serious adverse events estimated at under 1%.
• "Subacromial decompression surgery for adults with shoulder pain: a systematic review with meta-analysis." (2020) — PMID 30647053 — No important benefit compared to placebo surgery or exercise therapy; 6 serious harms per 1,000 patients (moderate certainty).
• "Surgery for rotator cuff tears." (2019) — PMID 31813166 — Cochrane review; 9 trials with 1,007 participants with full-thickness tears; at 1 year, surgery likely reduced pain little or not at all (moderate certainty); no placebo-controlled trials; mostly small, degenerative supraspinatus tears, so results may not apply to traumatic, large, subscapularis tears, or younger individuals; placebo-controlled and long-term trials are needed.
• "Fifteen-Year Results of a Comparative Analysis of Tendon Repair Versus Physiotherapy for Small-to-Medium-Sized Rotator Cuff Tears: A Concise Follow-up of Previous Reports." (2024) — PMID 39197154 — 103 randomized patients with full-thickness tears of 3cm or less; at 15 years, the repair group was superior in function and pain; 15 out of 51 physical therapy patients crossed over to surgery; 26 unrepaired tears increased in size from 16.2mm to 31.6mm.
• "[Referred shoulder pain]." (2004) — PMID 15152534 — Referred pain is feeling pain in a location different from its source; a case where diaphragmatic irritation caused shoulder pain.
• "The Efficacy of Exercise Therapy for Rotator Cuff-Related Shoulder Pain According to the FITT Principle: A Systematic Review With Meta-analyses." (2024) — PMID 38848304 — Movement control exercise may have been slightly better than general exercise (moderate certainty); it remains unclear whether this was due to the type of exercise or other features like progression and individualization.
• "Arthroscopic Management of Septic Arthritis of the Native Shoulder: A Systematic Review." (2018) — PMID 29096980 — 19 case reports and 8 case series comprising 115 patients; clues indicating arthroscopic indication include pain, restricted motion, swelling, redness, tenderness, fever, and inflammatory markers; high reoperation rate (level IV evidence).
• "Sex Differences in Symptom Presentation in Acute Coronary Syndromes: A Systematic Review and Meta-analysis." (2020) — PMID 32363989 — 27 studies; the most common symptom in both men and women is chest pain (men 79%, women 74%); women more frequently experience interscapular back pain, nausea, and shortness of breath; due to substantial overlap, the conclusion is that the term "atypical" should be abandoned.
• "Overlapping, Masquerading, and Causative Cervical Spine and Shoulder Pathology: A Systematic Review." (2020) — PMID 32206519 — Cervical spine pathology can cause shoulder pain, and shoulder pathology can cause neck pain; imaging can be misleading, making a differential examination crucial.