99% of adults over 40 have shoulder "abnormalities" on an MRI, study finds

Up to a third of people worldwide have shoulder pain; it’s one of the most common musculoskeletal complaints. But medical imaging might not reveal the problem—in fact, it could even cloud it.

In a study published in JAMA Internal Medicine this week, 99 percent of adults over 40 were found to have at least one abnormality in a rotator cuff on magnetic resonance imaging (MRI). The rotator cuff is the group of muscles and tendons in a shoulder joint that keeps the upper arm bone securely in the shoulder socket—and is often blamed for pain and other symptoms. The trouble is, the vast majority of people in the study had no shoulder problems.

The finding calls into question the growing use of MRIs to try to diagnose shoulder pain—and, in turn, the growing problem of overtreatment of rotator cuff (RC) abnormalities, which includes partial- and full-thickness tears as well as signs of tendinopathy (tendon swelling and thickening).

“While we cannot dismiss the possibility that some RC tears may contribute to shoulder symptoms, our findings indicate that we are currently unable to distinguish clinically meaningful MRI abnormalities from incidental findings,” the study authors concluded.

The study was conducted by Finnish researchers, who tapped into a nationally representative sample of adults who had signed up for a public health survey. In the end, 602 participants, aged 41 to 76, completed the study, which included answering questions about shoulder pain and getting MRIs on both shoulders. Of the 602 participants, 492 (82 percent) reported no shoulder symptoms, while 110 (18 percent) did report symptoms.

The MRI revealed that 595 of them (99 percent) had at least one RC abnormality. The most common abnormality was a partial-thickness tear (62 percent), followed by tendinopathy (25 percent), and a full-thickness tear (11 percent). The prevalence of abnormalities was similar between males and females. But the abnormalities showed a clear age-related progression, with no full-thickness tears in participants under 45 and the greatest proportion in people 70 to 76.

Breaking the findings down to shoulders instead of people, of the 1204 shoulders in the study, 1,076 (90 percent) were asymptomatic while 128 (10 percent) were symptomatic. Of the 1,076 asymptomatic shoulders, 96 percent had RC abnormalities (1,039 of 1,076), and of the 128 symptomatic shoulders, 98 percent had abnormalities (126 of 128).

Prevalence of tendinopathy and partial-thickness tears was similar between the symptomatic and asymptomatic groups. It initially looked like full-thickness tears were more common in the symptomatic groups, but when researchers adjusted for other factors, including additional abnormalities spotted in the MRIs, the difference between the symptomatic and asymptomatic groups vanished.

Context

The authors argue that the findings suggest clinicians should rethink MRI findings, changing not just how they’re used, but also how they’re explained to patients. The language in particular should change given that “abnormalities” are ubiquitous—thus normal—and shouldn’t be described in terms that indicate a need for repair, like “tear.”

“While we refer to these findings as abnormalities, many likely represent normal age-related changes rather than clinically relevant structural changes,” the authors write. “Adopting more precise and less value-laden terminology—such as lesion, defect, fraying, disruption, structural alteration, or degeneration—may help reduce patient anxiety and the perceived need to do something or fix something by avoiding language that implies trauma or a requirement for repair.”

In an accompanying editorial, two orthopedic surgeons from the University of California, San Francisco, Edgar Garcia-Lopez and Brian Feeley, agree with the language shift and caution clinicians to proactively put MRI findings in context.

They also address the glaring question of when MRIs should even be used for shoulder pain. They suggest that for pain that’s not related to an injury, clinicians should first try a couple months of watch-and-wait with rest or physical therapy to regain function. If there’s no meaningful improvement, an MRI may be warranted. But they stressed that any further decisions on treatment should be based on the patient’s history, clinical exam, and functional limitations of their shoulder—not just what’s seen on the imaging.

“Of course, the findings of this study are not meant to dissuade clinicians from using MRI when appropriate, but to reinforce that the diagnosis and management of shoulder pain should be guided primarily by functional limitations,” the surgeons write.