MRI Findings are Wildly Inconsistent, at Best

This post is going to be a quick review of a recent article that I came across that I think most people need to understand. As always you should read the original research article. The reference is at the bottom of the page.

Why They Did It

The ever cliche “A picture is worth a thousand words” unfortunately is true in the case of medical imaging. Magnetic resonance imaging or MRI for short is a wildly popular and ever increasing technology used in healthcare today.

While MRI’s are a fantastic technology their purpose and more importantly interpretation are not as perfect as the general public think. People often feel that because an MRI is a picture that it makes no difference how, when, by who, or why they are getting an MRI.

Most everyone is told they need an MRI and then go select a provider based on price and convenience because “all MRI’s are the same”. With this assumption you would expect that no matter where you go and which radiologist reviews your images you will get the same diagnosis. It’s a good thing a group of doctors recently but MRI’s and radiologists to the test to see if your assumptions are correct.

What They Did

A 63 year old woman who had low back pain and pain down her right leg was sent for MRI’s from 10 different locations. Before having her MRI’s she had a physical examination performed by the researchers. The physical examination ruled out any serious problem.

At the 10 centers she received routine MRI examinations. Between the 10 centers there were three different types of MRI machines used. The 10 locations were in New York City to make travel as convenient as possible for the participant. If she was asked for any additional information the volunteer was given a script to say/write.

All of the findings from the “body” or “impression” sections of the MRI report were organized into a list so they could be compared easily compered.

What They Found

There were “49 distinct findings” between the 10 different MRI reports. Of the “49 distinct findings” not one was finding was the same throughout all 10 reports. Yes you read that right, the 10 different MRI’s did not have one finding in common. One finding of the 49 was reported in 9 out of the 10 MRI reports. You can see the complete breakdown in the graph below.


The findings ranged from disc damage, to bony alignment, to degeneration. If you have had an MRI you have most likely seen some of these words. Finding like disc herniation, stenosis, or nerve root involvement are common. The authors took disc herniation’s as an example to show the difference between the 10 reports. Below is a graph of what they found in the 10 reports about disc herniation’s at various levels of the spine.


The Bottom Line

If you have gotten this far then hopefully you are thinking a little differently about MRI’s. Like most everything in healthcare MRI’s and imaging rely on human interpretation. Imaging should used to rule out serious pathology. MRI’s and other imaging should only be done when specific “red flag” symptoms are present.

Pain, especially spine pain, is common and can be helped with conservative treatment. It has already been shown that the connection between “abnormal” imaging and pain is not good. Everyone is going to have some sort of “abnormality” when looked at through the eyes of an MRI or x-ray.

What This is NOT Saying

This is NOT saying that all MRI’s are bad and that you should never get one. It is pointing out the fact that MRI findings are NOT as consistent and guaranteed as most people think. I would not look into MRI finding too much unless a serious problem, such as a fracture, was found.

The Article

Herzog, Richard, et al. “Variability in diagnostic error rates of 10 MRI centers performing lumbar spine MRI examinations on the same patient within a 3-week period.” The Spine Journal (2016).


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s