A rotator cuff tear is not always the problem. MRI scans are very sensitive, and they frequently pick up age-related tendon changes even in shoulders that feel completely normal. A scan is one part of the story. It does not automatically identify the source of your pain, and it does not automatically mean that an operation is required. In this article I explain what the evidence actually shows, and why, for many rotator cuff tears, structured rehabilitation is a sensible place to start.
Your symptoms, strength, function, injury history and personal goals often matter more than the scan alone. For many gradual-onset and partial-thickness rotator cuff tears, structured rehabilitation is a sensible first treatment. Prompt specialist assessment becomes more important when there has been a significant injury followed by sudden weakness or loss of function.
A rotator cuff tear is not always the problem
When a scan report mentions a rotator cuff tear, it is natural to assume that this must be the cause of the pain, and that it needs to be fixed. But tendons change as we age, in much the same way that hair greys or skin wrinkles, and modern imaging is extremely good at detecting those changes. The important question is not simply "is there a tear?", but "is this tear actually responsible for your symptoms, and what is the best way to treat them?".
Over the years I have personally scanned many thousands of rotator cuffs, both in my own clinic and working in radiology. When I scan a painful shoulder, I will often compare it with the person's pain-free side. Time and again, the shoulder that does not hurt can actually look worse on the scan than the one that does. Seeing this repeated so many times is a powerful, hands-on reminder that what shows up on imaging and what is driving someone's pain are simply not the same thing.
Rotator cuff abnormalities are common on MRI
This is not just my personal impression, it is exactly what the research shows. In a 2026 population study, Thomas Ibounig and colleagues performed MRI scans of both shoulders in 602 adults aged 41 to 76, examining 1,204 shoulders in total.
- 99% of participants had at least one rotator cuff abnormality.
- 96% of pain-free shoulders had an abnormality.
- 98% of painful shoulders had an abnormality.
- 25% showed rotator cuff tendinopathy.
- 62% showed a partial-thickness tear.
- 11% showed a full-thickness tear.
- Only 1% were reported as completely normal.
Full-thickness tears were initially more common in painful shoulders, but once other factors were taken into account, the difference was small and statistically uncertain. This does not mean that rotator cuff tears never cause pain. It means that MRI findings have to be interpreted alongside your symptoms, strength, function and clinical examination, not in isolation.
Ultrasound lets me do exactly this in real time, in the clinic. I can look at the tendon, measure a tear, watch how the shoulder moves and compare sides, then link what I see to what you actually feel.
Can my tear be fixed?
This is one of the questions I am asked most often, and it is completely understandable: when something is described as torn, it is natural to want it repaired. The honest answer is that many rotator cuff tears do not need to be, and often cannot simply be, stitched back to exactly how they were, and yet the shoulder can still become pain-free and strong. With degenerative and partial-thickness tears in particular, the goal is not to close the tear on a scan, it is to calm the tendon, rebuild strength and restore function. Surgery can repair certain tears, and for a recent traumatic tear in an active person it may well be the right choice, but a repaired tendon is not automatically a stronger or less painful one. So a more useful question than "can it be fixed?" is often "what will help this shoulder work and feel better?", and for many people the answer starts with rehabilitation.
Why surgery is not usually the first treatment
Most gradual-onset or atraumatic rotator cuff tears are initially treated without an operation. A progressive exercise programme can improve pain, movement, strength and confidence, while avoiding the risks and the lengthy recovery that come with surgery.
Non-surgical treatment may include:
- Temporarily modifying the activities that sharply aggravate the shoulder
- Maintaining comfortable shoulder movement
- Progressive rotator cuff and shoulder-blade strengthening
- Gradually rebuilding your work, lifting or sporting capacity
- Appropriate short-term pain relief when it is needed
- Improving sleep, general activity and confidence in using the arm
Surgery is more likely to be considered when there is:
- A recent traumatic tear
- Substantial or worsening weakness
- Sudden loss of the ability to lift the arm
- A repairable tear thought to be at risk of getting worse
- Persistent, disabling symptoms despite appropriate rehabilitation
- A need to restore demanding occupational or sporting function
Age, activity level, occupation, tendon quality, tear pattern and personal goals should all feed into the decision. There is no single right answer that applies to everyone.
Rehabilitation compared with keyhole repair
Khanna and colleagues followed 50 patients with symptomatic partial-thickness supraspinatus tears. Twenty-five chose supervised rehabilitation, and 25 had arthroscopic (keyhole) repair. At six months, both groups had improved substantially, and there was no meaningful difference between them.
Rehabilitation group: 75.7 · Surgery group: 75.0 · No statistically significant difference (p=0.723). Rehabilitation gave faster early recovery at one month. Surgery was linked to slightly less pain at six months, but the difference was only 0.4 points, below the study's own threshold for a meaningful change.
This study was small, non-randomised and only followed people for six months, so it cannot prove the two treatments are equivalent for everyone. What it does do is support rehabilitation as a reasonable first option for many partial-thickness tears.
What usually helps
Keep moving. Temporarily reduce the movements that sharply aggravate the shoulder, but avoid complete rest. Comfortable movement helps preserve confidence and capacity.
Follow a progressive exercise programme. Exercise should gradually address shoulder movement, rotator cuff strength, shoulder-blade control and the physical demands of your everyday life, work or sport.
Use pain relief strategically. When it is medically appropriate, pain relief can make sleep, normal movement and rehabilitation easier. It should support your recovery, not simply mask repeated overload.
Match the treatment to the diagnosis. Treatments work best when they are chosen for the actual clinical problem. For example, shockwave therapy may help some people with calcific rotator cuff disease, but the evidence is much less convincing for non-calcific rotator cuff pain. If the main problem is stiffness rather than a tear, that points towards a different pathway, such as frozen shoulder.
When a corticosteroid injection can help
A corticosteroid injection into the subacromial–subdeltoid bursa, the fluid-filled cushion above the rotator cuff, can reduce pain and disability in the short term. That window of relief can be genuinely useful: it may let you sleep more comfortably, restore movement and restart your rehabilitation.
The injection does not repair the torn tendon, and its benefit commonly fades. It works best when it is combined with a clear plan to restart or progress your exercises, rather than used on its own.
Could a corticosteroid injection increase the risk of a tear?
The honest answer is that the evidence suggests caution, but it does not prove that a single, appropriately performed corticosteroid injection causes a rotator cuff tear.
Laboratory research shows that corticosteroids can temporarily affect tendon-cell health, collagen production and mechanical strength, and these effects appear to depend on the dose and how often they are repeated. In one small prospective study, 9 of 53 patients developed a full-thickness tear within 12 weeks of a subacromial injection, and most of those already had a partial-thickness tear. However, that study had no untreated comparison group, so it could not tell whether the tears were caused by the injection or simply reflected the natural progression of pre-existing tendon disease.
Observational research in patients who went on to have rotator cuff repair found that a single pre-operative injection was not linked to a higher revision rate, whereas two or more injections in the preceding year were linked to a higher likelihood of later revision surgery. This is an association, not proof of cause and effect, because people who receive repeated injections may simply have had more severe or persistent shoulder problems to begin with.
There is no strong clinical evidence that one appropriately performed corticosteroid injection causes a rotator cuff tear. Repeated injections should still be used cautiously, particularly when a repairable tear is already present, when tendon surgery is being considered, when previous injections have given only brief relief, or when injections are being used instead of progressive rehabilitation. The possible short-term benefit should always be weighed against your individual diagnosis, treatment plan and longer-term goals.
What about PRP?
Platelet-rich plasma, or PRP, is prepared from a small sample of your own blood, though preparation methods and injection protocols vary considerably. Some reviews suggest that corticosteroid injections may give better early improvement, while PRP may offer some later advantages, but the average differences have often not reached the thresholds usually accepted as clinically meaningful. PRP is not currently a proven replacement for progressive rehabilitation.
To be clear, I do not currently offer PRP at Insight MSK. My focus for rotator cuff problems is accurate ultrasound diagnosis, structured rehabilitation, and a well-targeted corticosteroid injection when it genuinely fits the plan.
When to seek a prompt assessment
Most rotator cuff pain is not dangerous and settles with the right plan. As general safety advice, however, it is sensible to arrange a timely clinical assessment if you experience any of the following:
- Sudden inability to lift the arm after a significant injury
- Clear new or worsening weakness
- Shoulder deformity or a suspected dislocation
- New numbness or other neurological symptoms
- Fever, redness or feeling generally unwell
- Severe, unremitting pain
- Persistent loss of function despite appropriate treatment
If any of these come on suddenly after a significant injury, such as a fall or a sudden wrench on the arm, do not wait for a routine appointment. Go straight to your nearest emergency department (A&E). In an emergency, or if you feel very unwell with a hot, swollen shoulder, call 999.
The key message
A rotator cuff tear on MRI is not automatically the cause of shoulder pain, and it does not automatically require surgery. The best treatment decision comes from combining the scan with your history, clinical examination, strength, function and priorities. For many gradual-onset and partial-thickness tears, rehabilitation is a safe and evidence-supported place to start, and an accurate ultrasound assessment helps make sure the plan fits the real problem.
References
- Ibounig T, Jarvinen TLN, Raatikainen S, et al. Incidental Rotator Cuff Abnormalities on Magnetic Resonance Imaging. JAMA Internal Medicine. 2026;186(4):406–414.
- Khanna P, Rehncy J, Singh D, et al. Conservative Management Versus Arthroscopic Repair for Symptomatic Partial-Thickness Supraspinatus Tears. Cureus. 2026;18(1).
- Ramirez J, Pomes I, Cabrera S, et al. Incidence of Full-Thickness Rotator Cuff Tear After Subacromial Corticosteroid Injection: A 12-Week Prospective Study. Modern Rheumatology. 2014;24(4):667–670.
- Puzzitiello RN, Patel BH, Forlenza EM, et al. Adverse Impact of Corticosteroids on Rotator Cuff Tendon Health and Repair: A Systematic Review of Basic Science Studies. Arthroscopy, Sports Medicine, and Rehabilitation. 2020;2(2):e161–e169.
- Desai VS, Camp CL, Boddapati V, et al. Increasing Numbers of Shoulder Corticosteroid Injections Within a Year Preoperatively May Be Associated With a Higher Rate of Subsequent Revision Rotator Cuff Surgery. Arthroscopy. 2019;35(1):45–50.
- Bhatia M, Singh B, Nicolaou N, Ravikumar KJ. Correlation Between Rotator Cuff Tears and Repeated Subacromial Steroid Injections: A Case-Controlled Study. Annals of the Royal College of Surgeons of England. 2009;91(5):414–416.
This article provides general information only. It cannot diagnose an individual shoulder or replace assessment and advice from a qualified healthcare professional.