When “Abnormal” Is Normal: Rethinking Rotator Cuff Tears and MRI Findings

A new research paper is making the rounds, and while the headline numbers may surprise the general public, for clinicians it’s another reminder of something we’ve been seeing for years: age-related changes on imaging are incredibly common.

Ibounig and colleagues performed MRIs on both shoulders of 602 Finnish adults aged 41-76 to examine how common rotator cuff “abnormalities" were - defined as tendinopathy, partial-thickness tears (PTRCTs) or full-thickness tears (FTRCTs). What they found was:

• 592 of the 602 participants (98.8%) had abnormalities

• 25% had tendinopathy, 62% had PTRCTs, 11% had FTRCTs

• Prevalence and severity increased with age, as tendinopathy was more common in those 41-54 and tears more common in those 55 and older

• Abnormalities were present in 96.6% of asymptomatic shoulders and 98% in symptomatic shoulders.

Overall, nearly everyone over 40 had something show up on their MRI, regardless of whether they had pain. This may sound dramatic, but it is consistent with previous research. What this paper reinforces is not that shoulders can “fall apart”, but that structural changes are a normal part of aging.

Takeaway #1: What Does “Abnormal” Mean?
If 98.8% of people have something, can we really call it abnormal? By definition, abnormal means no normal (duh), unusual or deviating from the norm. But when almost everyone over 40 shows these findings, wouldn’t that just be normal? We could argue that a “normal” shoulder should be free of degeneration or tears, but that ignores biology. Just like skin wrinkles and hair turns grey, rotator cuff tissue changes with age. We’ve traditionally labelled these findings as degeneration, defects or lesions, but when changes are this common (and often painless), maybe our language needs updating.

Takeaway #2: Pain is Not the Same as Imaging
One of the most important findings in this paper is the lack of difference between painful and non-painful shoulders, with those two groups having close to the same prevalence of abnormalities. It makes sense to assume that a tear must hurt (especially a FTRCT), but this ignores the complexity of pain. Tissue quality is only one piece of the puzzle; psychological, social and even other biological factors all influence whether someone experiences pain and how intense that pain is. An MRI simply shows structure. It does not show pain, nor does it show when the tear occurred, if it’s relevant, how strong the shoulder is or how confident someone is in using their shoulder. As the authors note, “While we cannot dismiss the possibility that some rotator cuff tears may contribute to shoulder symptoms… we are currently unable to distinguish clinically meaningful MRI abnormalities from incidental findings.” Imaging can support a diagnosis (especially in cases of clear acute trauma or significant strength or functional loss) but it cannot be the sole explanation for pain.

Why This Matters
For clinicians, this paper reinforces the need to be thoughtful with our language. Terms like “degeneration” or “defect” imply something is broken and needs fixing. But many of these findings are common age-related changes, not structural catastrophes. For patients, this can be reassuring. If you’ve been told you have a tear - especially if. you don’t even have pain - this paper suggests your shoulder may be far more normal than you were led to believe. Even if you do have pain, this doesn’t automatically mean your shoulder is fragile damaged or destined for surgery.

Age-related changes are not exclusive to the shoulder. We see similar patterns in the spine and knee: structural changes increase with age, but the link to pain is inconsistent. You are not your imaging. The human body is adaptable and resilient. The goal isn’t to fix every MRI finding, but to build strength, restore capacity and help you reclaim confidence in your movement.

Ibounig T, Jarvinen TL, Raatikainen S, Harkanen T, Sillanpaa N, Bensch F, et al. Incidental rotator cuff abnormalities on magnetic resonance imaging. JAMA Intern Med. 2026.