Subacromial impingement exercises: if you have arrived here, it is because someone told you that the tendons of your shoulder "rub" under the bone and that the solution is exercise. The second part is true to a large extent. The first is much less solid than it seems, and it is worth explaining why: it changes how you train and the expectations you bring into rehabilitation.
This article puts together what research really supports, what it does not, and a practical programme you can use anyway, because putting an explanation into perspective does not mean there is nothing to do.
⚠ Important: read this before continuing
The exercises and advice in this article are general indications and may not be suitable for your situation. Every person has a unique clinical history: what works for others could be counterproductive for you.
Want a professional's opinion? Start with the questionnaire: you register and see who will follow you, a physiotherapist registered with the Albo chosen for your case, whom you write to in chat. If you would rather hear their voice, the first consultation is free and we look at your situation together.
Before you start: the signs that have nothing to do with tendons
Shoulder pain is not always a tendon problem. Before reading any programme, check this list. If you recognise one of these situations, the priority is a medical assessment, not a band.
- Recent trauma with inability to raise the arm, visible deformity or sudden loss of strength: it may indicate a major tendon injury, a dislocation or a fracture, especially after a fall and after the age of 50.
- Fever, a swollen, hot and red shoulder: an infected joint is an emergency.
- Pain in the left shoulder or arm together with chest pressure, breathlessness, cold sweating or nausea: call the emergency number. The heart can cause pain referred to the shoulder.
- Continuous night pain that does not change with position, associated with weight loss, persistent cough, a history of cancer or a heavy smoking history: it needs investigating and should not be dismissed as tendinitis.
- Tingling, numbness or weakness running down the arm and hand: the problem may originate in the neck.
- Right shoulder pain that appears after meals, with pain in the upper abdomen or a yellowish colour: the gallbladder irritates the diaphragm and the pain reaches the shoulder.
If none of this applies to you, and the pain typically appears when raising the arm or bringing it behind your back, you are in the territory this article covers. If you want to go deeper into the reasoning about causes, we wrote about it in when to worry about shoulder pain.
What subacromial impingement is (and why the name is debated)
The classic explanation is geometric: above the head of the humerus there is the acromion, a bony roof; below it run the rotator cuff tendons and a bursa. When you raise the arm the space narrows, the tendons are compressed, become inflamed and hurt. Hence the name "impingement", or conflitto, if you have read the report in Italian.
The problem is that this story does not stand up well to testing. First crack: the clinical tests that assign the label are not reliable. In a systematic review with meta-analysis of physical examinations of the shoulder, Neer's test had a sensitivity of 72% and a specificity of 60%, Hawkins-Kennedy's test 79% and 59%, the painful arc 53% and 76%; the authors conclude that no single test can be recommended for a certain diagnosis, and that combining them improves accuracy only marginally. The right way to read a "positive test" (and why these tests rule out better than they confirm) is explained in positive test: what it really means.
Second crack, heavier: if the problem really were the space, widening it should work. In the British randomised CSAW study, published in Lancet, 313 patients with subacromial pain for at least 3 months and an intact rotator cuff, all of whom had already gone through a non-surgical pathway that included exercise and at least one cortisone injection without success, and were therefore candidates for surgery, were assigned to arthroscopic decompression (106), arthroscopy alone as a placebo, that is without removing bone or tissue (103), or no treatment (104). At 6 months the Oxford Shoulder Score did not differ between the two surgical groups (32.7 against 34.2 points; mean difference -1.3, 95% confidence interval from -3.9 to 1.3). Both surgical groups did slightly better than no treatment, but the difference was not clinically important.
Put simply: removing bone to make room added nothing compared with going in with a camera and doing nothing. For this reason many clinicians now prefer to speak of rotator cuff related shoulder pain: it describes the symptom without promising a mechanical cause that has not been proven. The phenomenon is real (the pain is there, and function is reduced), but the image of tendons "squeezed like in a vice" is a simplification.
The names you find on the report, and why they are almost all the same
This is the part that causes the most confusion, and it is worth clearing up straight away. If you have a report or a diagnosis in hand, it is likely that one of these words is written on it:
- Impingement or subacromial impingement (conflitto subacromiale): the English and Italian versions of the same label.
- Tendinitis or supraspinatus tendinopathy: the supraspinatus is the rotator cuff tendon most often involved, because of its position. "Tendinopathy" is the more correct term, because there is almost never inflammation but a tendon that has suffered the load.
- Shoulder tendinitis: the same thing, without specifying which tendon.
- Subacromial bursitis: the bursa under the acromion, inflamed. In practice it almost always coexists with tendon suffering, and telling them apart on ultrasound does not change what is done.
- Periarthritis: a container that holds together this and other very different conditions: the distinctions are in shoulder periarthritis.
These are not five different diseases: they are five ways of describing the same picture, which is rotator cuff related shoulder pain, with pain when lifting the arm and often at night lying on the side.
Why it matters: if you were given a different word from the one you read here, you are not reading the wrong page. And above all, the treatment does not change according to the label. It changes according to how much the shoulder tolerates load today, which is what is measured in the assessment.
One thing the distinction does do: if there is a tear of the tendon, that is, a rupture, the pathway is different, and you can find it in rotator cuff tear.
Why this helps you. Because it changes the goal: you are not "creating space", you are making the shoulder more tolerant of load. It is a less spectacular goal but a much more trainable one, and it saves you from chasing perfect posture for months.
How much exercise really helps with subacromial impingement
Here we need to be precise, because this is where most articles exaggerate.
The guidelines are on the side of exercise. The clinical practice guideline published in 2022 in the Journal of Orthopaedic & Sports Physical Therapy on rotator cuff disorders (73 recommendations) advises prescribing an active, task-oriented rehabilitation programme, exercise and education, to reduce pain and disability, reserves imaging for selected circumstances, and does not recommend subacromial decompression for rotator cuff tendinopathy. Be careful, however, not to read it as "never operate": the same guideline considers surgery appropriate in selected patients with a full-thickness rotator cuff tear, which is a different condition from tendinopathy, and which was in fact a reason for exclusion from the CSAW study.
But the comparison with a placebo is less brilliant than it is made out to be. The Cochrane review on manual therapy and exercise in rotator cuff disease included 60 studies and 3,620 participants, and found only one study comparing the combination actually used in clinic with a placebo (inactive ultrasound), in 120 people with chronic disorder. At 22 weeks the average improvement in pain was 17.3 points out of 100 with placebo and 24.8 with manual therapy and exercise (adjusted difference 6.8 points, 95% CI from -0.70 to 14.30: the interval includes zero, so "no difference" remains compatible with the data); on function, 15.6 against 22.4 points, with an adjusted difference of 7.1 points (95% CI from 0.30 to 13.90). Treatment success was reported by 57% (31 out of 54) with manual therapy and exercise against 41% (24 out of 58) with placebo, a difference that is also inconclusive. The authors, who judged the study to be of high quality, conclude that there are no clinically important differences between the groups on any outcome. It should also be said that there is an unpleasant aspect: adverse events were more frequent with manual therapy and exercise (31%, 17 out of 55) than with placebo (8%, 5 out of 61), although they were mild: short-lived pain after treatment.
One finding, however, is encouraging, and it concerns the type of exercise. A Swedish randomised study published in the BMJ enrolled 102 people with persistent subacromial impingement lasting more than six months, in whom previous conservative treatment had already failed, all referred through orthopaedic specialists. One group followed a specific strategy: eccentric strengthening for the cuff, concentric/eccentric work for the shoulder blade stabilisers, plus manual mobilisation; the other did generic movement exercises for the neck and shoulder. Both: five or six guided sessions over 12 weeks, plus home exercises once or twice a day. The Constant-Murley score improved by 24 points in the specific group against 9 in the control group (mean difference 15 points, 95% CI 8.5-20.6). 69% (35 out of 51) reported a positive outcome against 24% (11 out of 46). And above all: 20% (10 out of 51) of the specific group then chose to have surgery, against 63% (29 out of 46) of the other.
The honest picture is this: exercises for subacromial impingement are not a guaranteed cure, and their superiority over a credible placebo has not been clearly demonstrated; but they are the recommended first line, they do not cause serious harm, and which exercises you do seems to matter: with the right load on the cuff and the shoulder blade, many people stop considering surgery. With a surgical alternative that does not beat placebo, starting with exercise remains the rational choice.
The programme, phase by phase
Below you will find six exercises, organised in progression. They are not a protocol to follow to the letter: they are the categories of work that appear in the studies cited above, and that you find in almost all subacromial impingement programmes.
1. Codman's pendulum (irritable phase)
Standing, trunk bent forward, free hand resting on a table. Let the painful arm hang and draw small circles using the movement of the body, not the shoulder. It serves to move without loading, on the days when raising the arm is unbearable. We described how to do it in detail in the guide to Codman's pendulum exercise.
2. Posterior capsule stretch
Lying on the painful side, arm at 90°, gently push the forearm downwards with the other hand (sleeper stretch). Alternatively, standing, take the straightened arm across the chest. Useful if you feel the shoulder is stiff when you bring your hand to the opposite shoulder or behind your back.
3. External rotation with a band
The cornerstone exercise. Elbow at the side, bent at 90°, a rolled towel under the armpit; rotate the forearm outwards against the band and return slowly (counting to three is a practical way to stop the arm from snapping back). It is the return phase, the eccentric one, that is the heart of the Swedish programme. If you want to see the variations, we discuss them in the shoulder external rotation exercises.
4. Internal rotation with a band
Same position, resistance from the opposite side, you rotate towards the belly. It completes the work on the rotators: training only external rotation creates a useless imbalance.
5. Prone Y and T
Lying face down on a bed or an inclined bench, arms lift to form a Y and then a T. They work the lower, middle trapezius and the rhomboids, that is the muscles that orient the shoulder blade when the arm rises. They start bodyweight: the weight of the arm is already a load.
6. Wall slide
Back and forearms against the wall, slide the arms upwards keeping contact, without arching the back and without pulling the shoulders towards the ears. It is the exercise that trains coordination between the shoulder blade and the humerus in the movement that bothers you.
An important tip: before trying these exercises, make sure you have received a correct diagnosis. People often think they have one problem, but the real cause of the pain is another. Doing the wrong exercises can make the situation worse instead of better.
How to progress without invented numbers
I will not give you sets and repetitions: the studies used individualised and supervised programmes, and copying a dose from an article is not what made them effective. The rules of conduct, however, can be transferred.
Start from the lowest load you find challenging, often just the weight of the arm, and increase one element at a time: first the quality of the movement, then the repetitions, then the resistance, then the range above the head. In the Swedish programme the guided sessions were few and spread over three months, while the real work was done at home almost every day: consistency matters more than the intensity of any single session.
On pain: the rule "pain = stop" is too rigid for a tendinopathy. Moderate discomfort during the exercise, which fades within a few hours and does not leave the shoulder worse the next morning, is acceptable. Pain that grows during the set, or that wakes you the following night, tells you that you have put on too much load. If it helps you frame the timescales, we wrote about why the number of sessions varies so much from person to person.
Posture and shoulder blades: what to expect (and what not to)
"You have rounded shoulders, that is why it hurts" is the most repeated sentence on this subject, and it deserves a clarification.
That scapular control has something to do with it is plausible: a systematic review with meta-analysis of 419 asymptomatic athletes found that those with scapular dyskinesis had a 43% higher risk of developing shoulder pain in the follow-up of 9-24 months: pain appeared in 35% (56 out of 160) of those with dyskinesis against 25% (65 out of 259) of those without. It is a signal, not a destiny: the majority of those with dyskinesis remained free of pain, and a risk factor in healthy athletes does not prove that correcting the shoulder blade cures a pain that is already present.
So yes, work on scapular control: it is within the programme that reduced the need for surgery. But treat it as training, not as postural repair. There is no posture to reach in order to be allowed to feel well, and spending months "keeping the shoulder blades back" all day is not a sensible therapeutic goal.
Subacromial impingement: exercises to avoid, and the most common mistakes
First, the list people look for. With an irritated shoulder, these three things are set aside until the arm rises without compensation: lateral raises with dumbbells above shoulder line, bench press and overhead pressing with the loads you used before, and wide-grip pull-ups on the bar. Not because those movements hurt in themselves: it is the load on a pattern that cannot yet handle it, and it is the difference between removing an exercise for good and postponing it by a few weeks.
What you should not avoid is raising the arm. Stopping lifting it overhead for weeks is the mistake that costs the most, and it opens the list that follows.
These mistakes recur, and almost all of them come from the wrong idea that subacromial impingement is a problem of space to protect rather than load to rebuild.
- Stopping at the first discomfort. Tendinopathy improves with progressive load, not with prolonged rest. Avoiding movements above the head altogether for weeks makes the shoulder less tolerant, not more.
- Loading too much and too soon, typically going back to bench press or lateral raises with the weights you used before.
- Shrugging the shoulders towards the ears during exercises: if the upper trapezius does all the work, the rest of the chain is not trained.
- Training only what you can see. Deltoids and pectorals yes, cuff and scapular stabilisers no: that imbalance is what the programme should correct.
- Expecting healing in two weeks. The studies that showed results measured at three months, and on people who trained almost every day.
- Confusing stiffness with impingement. If the arm does not rise even when someone else is moving it, the problem may be the capsule: it is approached differently, as in rehabilitation for frozen shoulder.
If after three months nothing has changed
This is the most frustrating scenario, and it is also the one in which the subacromial impingement label needs to be questioned again. Reassess, do not insist on the same thing. The options are not "exercises forever" or "the operating theatre": you can change the type of load, review the diagnosis (the review of clinical tests explains why the initial label may be wrong), or discuss with the specialist injections or imaging, which the guidelines reserve for selected circumstances. On surgery, the most useful answer we can give you remains that of CSAW: in the comparison with a surgical placebo, decompression showed no advantage.
In summary
Subacromial impingement is a convenient label for a real pain, built on a mechanism that the evidence has scaled down. Exercises remain the first thing to do: recommended by the guidelines, with a good signal that programmes targeting the cuff and the shoulder blade reduce the need for surgery. But they are not magic, and anyone who presents them to you as a guaranteed recovery is telling you more than we know. The physiotherapists at Ri-Hub can help you build and correct the programme over time, by video call, adapting it to how your shoulder responds.
Sources
- Arthroscopic subacromial decompression for subacromial shoulder pain (CSAW): a multicentre, pragmatic, parallel group, placebo-controlled, three-group, randomised surgical trial, Lancet, 2018
- Manual therapy and exercise for rotator cuff disease, Cochrane Database of Systematic Reviews, 2016
- Effect of specific exercise strategy on need for surgery in patients with subacromial impingement syndrome: randomised controlled study, BMJ, 2012
- Diagnosing, Managing, and Supporting Return to Work of Adults With Rotator Cuff Disorders: A Clinical Practice Guideline, Journal of Orthopaedic & Sports Physical Therapy, 2022
- Scapular dyskinesis increases the risk of future shoulder pain by 43% in asymptomatic athletes: a systematic review and meta-analysis, British Journal of Sports Medicine, 2018
- Which physical examination tests provide clinicians with the most value when examining the shoulder? Update of a systematic review with meta-analysis of individual tests, British Journal of Sports Medicine, 2012
Last source check: August 2026.
You might also be interested in:
Neurological Rehabilitation: A Complete Guide to Recovery After Brain Injury Neck and Headaches: The Connection, the Causes and How to Get Rid of Them Muscle Strain: What to Do, First Aid and Recovery Times
Stay in touch with us Join our dedicated Facebook group: we answer all patients, give practical advice and clear up the most common doubts.
Need a consultation?
Start with the questionnaire on Ri-Hub, a few questions about your case, you register, and see who will follow you: a physiotherapist registered with the Albo, whom you write to in chat whenever you like. If you prefer to talk it over by voice, the first consultation is free, lasts 20 minutes and commits you to nothing: they arrive having already read what you wrote, and together you decide whether and how to go on.
