Shoulder Pain

Rotator cuff related shoulder pain is an umbrella term covering subacromial pain (impingement) syndrome, rotator cuff tendinopathy, and both partial and full thickness tears. If you have been told you have shoulder impingement, that sits inside this umbrella rather than beside it — a different name for the same clinical problem, not a different problem.^[1]

Changes in the rotator cuff turn up constantly in shoulders that do not hurt. Pooling 30 studies and 6,112 shoulders, abnormalities rose with age from 9.7% in people aged 20 years and younger to 62% in those 80 and older — enough for those authors to describe cuff degeneration as a common aspect of normal human aging. That is not an argument for ignoring a scan. It is an argument against assuming that the finding on the report is why your shoulder hurts, or that the word “tear” means an operation.^[2]

“Impingement” began as a mechanical explanation — the idea that a tendon was being pinched under a piece of bone. That explanation has not held up, and the naming in the literature moved on more than a decade ago.

Subacromial pain syndrome is the umbrella term, and the Dutch guideline’s 2026 update defines it descriptively rather than mechanically: non-traumatic, non-rheumatologic shoulder complaints that are particularly painful during arm elevation, including supraspinatus tendinosis, calcific tendinitis and degenerative tears. The rotator cuff literature arrives at the same place from the other side, and Cochrane does not separate them either — its review of decompression surgery describes the people in it as having rotator cuff disease manifest as painful shoulder impingement.^{[4, 1, 5]}

There is also no single test that settles it. The diagnosis is made from a cluster of clinical tests rather than from any one of them, which is a large part of why an examination takes as long as it does.^{[3, 4]}

You were probably handed one of these terms by a clinician. We will not present impingement and rotator cuff problems as two different things treated two different ways — they are one problem.

Start with what the surgical trials found, because it sets the frame. Keyhole “decompression” surgery has been tested against placebo surgery: in the CSAW trial, shoulder scores at six months did not differ between decompression and arthroscopy alone, and the authors wrote that the findings question the value of the operation.^[6]

High certainty is the strongest rating Cochrane awards, and an international guideline panel with patients on it then made a strong recommendation against the operation. For a torn cuff the picture is less settled but still not automatic: Cochrane says repair may provide little or no clinically important benefit for pain, function or quality of life compared with non-operative treatment.^{[7, 8]}

Conservative care is where the guidelines put their weight. The Dutch Orthopaedic Association’s 2026 guideline opens by asking clinicians to encourage gradual exercise within sport and work, and suggests considering a subacromial corticosteroid injection where pain is severe enough to stop someone exercising — the injection there to open a door to the exercise, not to stand in for it. The UK commissioning guide adds a checkpoint: do not consider further physiotherapy unless there is improvement during the first six weeks.^{[4, 9]}

Measured against surgery, a masterclass review reports that published high quality research suggests a graduated, well-constructed exercise approach confers at least equivalent benefit — with the caveat that belongs beside it, that outcomes may be incomplete.^[1]