Shoulder scans in most people above 40 show rotator cuff abnormalities, pain or not

99% of adults aged 41–76 had at least one RC abnormality on MRI. Credit: www.pexels.com/photo/magnetic-resonance-imaging-machine-13176358/

Shoulder pain is the third most common musculoskeletal complaint seen by doctors, affecting approximately 18–31% of the global population each month. Up to 85% of these cases are due to problems with the rotator cuff (RC)—the shoulder’s built-in support system. Made up of four muscles and their tendons, the RC keeps the upper arm bone securely in the socket while allowing the arm to lift, rotate, and move smoothly.

A Finnish Imaging of Shoulder (FIMAGE) study found that rotator cuff changes on MRI are nearly universal after age 40. Overall, 99% of people in this age group showed some form of abnormality on imaging, whether or not they had shoulder pain. The findings are published in JAMA Internal Medicine.

The results point toward the fact that many MRI findings likely reflect normal age-related changes in RC rather than the true source of pain. The researchers suggest that routine imaging should not be taken as the sole guide for the diagnosis or treatment of shoulder pain.

Overuse of imaging in clinical settings

Imaging is often one of the first steps recommended, with about half of patients receiving a scan at their initial visit to a general practitioner. These scans frequently show rotator cuff changes, including tendon wear, partial tears, or full tears. Once a finding appears on imaging, it often leads to treatment—regardless of whether the person’s symptoms clearly match the scan results.

Bar graph showing the prevalence of rotator cuff abnormalities by age and sex. Credit: JAMA Intern Med. (2026). DOI: 10.1001/jamainternmed.2025.7903

In many countries, rotator cuff surgeries have increased by two- to seven-fold, and researchers question whether the heavy reliance on imaging as a primary diagnostic tool is contributing to this rise.

Previous studies have found that the connection between rotator cuff imaging findings and real shoulder pain remains unclear.

Relation between RC abnormalities and pain

The researchers in this study wanted to determine how common rotator cuff issues are in the general public and to see whether findings from imaging actually explain the shoulder pain people experience. To do this, they selected 602 adults aged 41–76 in Finland at random from a large national health database. They ensured the sample represented a wide range of people, not just those already seeking medical help for shoulder pain.

Participants were asked to complete a questionnaire asking them to explain their shoulder pain or dysfunction status, and, based on their answers, they were categorized as with or without symptoms. Then, both groups underwent an MRI of both shoulders on a high-powered 3-Tesla scanner for detailed imaging of RC tendons.

These images went to radiologists for blinded review—in other words, they had no idea which patients were in pain and which weren’t. Based on their analysis, the researchers categorized the tendon conditions as: normal, wear and tear, partial tears, or full tears.

They found that almost all patients showed some form of RC wear and tear. These findings were present in 96% of people with no shoulder pain and 98% of those with symptoms, a difference that offers very little explanation for why some people hurt, and others do not.

These numbers suggest that rotator cuff abnormalities seen on MRI might just be signs of normal aging, just like graying hair, rather than a clear source of pain.

The study highlights the need for better ways to tell the difference between rotator cuff changes that actually matter and those that are simple incidental findings. Relying on routine imaging alone is not enough to make that distinction.

© 2026 Science X Network

Q&A: Rotator cuff tears—treatment and recovery