In a recent article in the New York Times, the overuse of MRI technology was exposed. They point out a study performed by James Andrews, MD, probably the best known sports medicine physician in the world, where he looked at MRI results of 31 perfectly healthy, normal shoulders of professional baseball pitchers. What he found may shock the general public, but it is nothing new to healthcare professionals. Studies like this repeatedly prove a simple point: An MRI may show you more than you need to know.
What his study found was that around 90% of healthy, normal shoulders have “abnormal” findings on MRI. What this means is that it is actually normal to have an “abnormal” MRI.
A history of these studies
The first of these studies that I remember reading was an article published in The New England Journal of Medicine in 1994 that looked at MRIs of the lumbar spine of 98 healthy, pain-free individuals. To oversimplify their findings, they found a whole mess of “abnormal” findings. Let me give you a brief description of how this information is useful:
Let’s assume one of these individuals strains his back doing yard work. He gets some really severe muscle spasms due to the strain and goes to see his physician who orders an MRI. The results come back with significant disk protrusion so the physician recommends surgery. Whoops! Surgery was just inappropriately performed on an abnormal finding that had nothing to due with the cause of his pain, a simple muscle strain.
Similar studies, like the recent one conducted by Dr. Andrews, have been performed on shoulders, knees, and hips with similar findings – abnormal MRIs are normal.
How an injury should be diagnosed
Research has shown that the most effective way to evaluate a medical condition is through a systematic approach. This means that there is an ordered hierarchy to one’s “differential diagnosis” (medical term for ruling out or ruling in potential diagnoses). In medicine we have a wonderful and ancient tool called the “clinical exam”. This includes listening to the patient’s chief complaint and performing a series of simple tests in the clinic.
Researchers have been able to validate the accuracy of these findings though hundreds of studies. With the information gathered from a proper clinical exam, a good medical provider will have an effective plan of care for either treatment or further evaluation. Further evaluation may include a referral to a specialist (who will begin with their own differential diagnosis) or radiographic tools such as x-rays, MRIs, and CT scans.
When the MRI is useful and how to use those findings
In order to require an extremely expensive MRI you need two reasons that must both be present:
- The physician needs more evidence to confirm their suspected diagnosis.
- The likely findings will have an immediate effect on the plan of care.
Let’s use an example. A patient dislocated their shoulder which means that there is a high likelihood that they tore their glenoid labrum. An MRI would confirm this diagnosis, fulfilling number 1 above. However, research shows that, as long as the shoulder does not dislocate a second time, there is a good chance that they will do well with physical therapy and not need surgery. This fails number 2 above. In other words, who cares if the labrum is torn, either way we are going to do physical therapy first. Now, if physical therapy fails and the shoulder dislocates a second time, number 2 is now fulfilled (the confirmation of a torn labrum now makes surgery a much more appropriate option in this case) and the MRI is now warranted.
So the main purpose of an MRI is to confirm a highly suspected yet unconfirmed diagnosis that would require more immediate and drastic intervention, usually surgery. When your physician recommends an MRI, they should tell you the exact reason why. For example, “From the clinical exam, I highly suspect that you tore your ACL which will likely require surgery.” IT IS NOT JUST “TO SEE WHAT’S GOING ON IN THERE”! Pain by itself is not a reason to perform an MRI.
When surgery is necessary and an MRI is not required
Some findings during clinical exam are so specific that the diagnosis is confirmed without further evaluation. For example, if you have the history of knee instability and a positive pivot shift (a simple test performed in the clinic), I don’t need an MRI to tell me that you have a torn ACL. Why waste the money? Or, if a previous MRI was inconclusive (there is no such thing as a “negative MRI”) yet you continue to have all of the classic signs and symptoms of a torn meniscus in the knee after conservative management (like physical therapy), then surgery is still warranted.
If you fear that your medical provider is not performing a proper systematic exam, contact us. We have several physicians to whom we can refer you who “do it the right way”.