Home › Blog › MRI of the Shoulder: A Great Tool or a Fool’s Errand?

I’ve blogged many times about our penchant as a medical care system to over-interpret MRI results. Because of the attention physicians place on these tests and the decrease of reliance on physical exam, patients often believe that their MRI report was written by the oracle of Delphi and contains explanations of all of the hidden mysteries of the universe and an explanation for their pain. However, the reality is quite different.

MRI is a Tool

Most physicians who depend on MRIs can’t read one themselves. This is a pity because there are often findings on these studies that are not read by the radiologist and thus not in the report. However, no matter how good you are at reading MRIs, as a physician, you must take what’s on the film and combine that with an extensive history and physical exam for it to have any meaning. However, all too often, physicians substitute the physical exam part for a reading of the MRI report. Why? It’s faster and easier and you don’t have to touch the patient.

Yet Another Study Showing that Over-Reliance on MRI is a Bad Idea

A research study was just published on 123 patients who only had shoulder pain on one side. However, the researchers decided to perform an MRI on both shoulders at once. What did they find? That abnormalities on shoulder MRI were as common in the bad shoulder as the one that didn’t hurt! Only full-thickness tears of the rotator cuff and shoulder arthritis were found more often in the painful shoulder. However, they both only occurred about 10% more often on that symptomatic side. Meaning that even those findings were not a sure thing in identifying pain because they were also commonly found on the asymptomatic side.

This is One of Many Such Studies

We have many studies just like this one that I’ve covered, but focused in other areas:

• Most MRI findings of the knee showing meniscus tears don’t correlate with pain. In fact, finding one on MRI in a patient over age 35 is about as important as finding a wrinkle around the eye.
• MRI of the knee showing cartilage loss is not associated with pain. In fact, only findings in the bone like swelling and cysts correlate with knee pain.
• Pain and disability after a low back herniated disc weren’t associated with structural changes on MRI, but instead inflammatory chemical levels in the blood.
• After shoulder surgery, pain in the shoulder wasn’t correlated with whether the patient’s rotator cuff tear healed on MRI, but was associated with an inflammatory chemical level in the joint.

So How Should You Think of Your MRI Findings?

An MRI is a tool that should be used to confirm what the doctor “sees” on history and exam. That means that you need to spend time with the doctor with his or her hands-on your problem. That should be at least 10-30 minutes of time spent performing an exam, depending on how complex your problem is. What if you never see the doctor and instead only see a physician assistant or nurse practitioner? You’re not in the right place. What if the doctor only spends a few minutes performing a quick exam? You’re not in the right place.

The upshot? Shoulder MRI, like other MRIs, can only be used as part of a puzzle that paints a picture. When combined with an extensive hands-on physical exam and history, it can help understand why you hurt, but trying to use MRI as a substitute for time spent with the patient is always a fool’s errand.

Chris Centeno, MD is a specialist in regenerative medicine and the new field of Interventional Orthopedics. Centeno pioneered orthopedic stem cell procedures in 2005 and is responsible for a large amount of the published research on stem cell use for orthopedic applications. View Profile