Why Two Radiologists Can Read the Same MRI Differently - Spinal DISC Center | Kris Radcliff, MD | New Jersey

wp-singular,page-template-default,page,page-id-17222,wp-theme-bridge,bridge-core-3.3.5,wp-schema-pro-2.12.4,qode-optimizer-1.2.3,qode-page-transition-enabled,ajax_fade,page_not_loaded,,qode-theme-ver-30.9.1,qode-theme-bridge,disabled_footer_bottom,qode_advanced_footer_responsive_1000,wpb-js-composer js-comp-ver-9.0.1,vc_responsive,elementor-default,elementor-kit-16783

Why Two Radiologists Can Read the Same MRI Differently

Many patients are surprised to learn that two radiologists can look at the same spine MRI and write different reports. This is not a secret, and it is not about any one doctor. It is a well-studied problem in spine imaging, and knowing about it can help you get the right care.

One patient, ten MRI centers, ten different reports

In a study published in The Spine Journal in 2017, one patient with back pain and sciatica had a lumbar MRI at 10 different MRI centers within 3 weeks. Her spine did not change, but the reports did:

The 10 reports listed 49 different findings between them.
Not one finding appeared in all 10 reports.
About one-third of the findings appeared in only one report.
On average, each report missed about 44% of the findings and listed about 1 or 2 findings that were not there.

(Herzog et al., The Spine Journal, 2017)

“Mild,” “moderate” and “severe” are often a judgment call

Most reports grade narrowing (stenosis) as mild, moderate or severe. Unless the report names a published grading system, those words reflect the reader’s impression.

In a 2026 study of 37 physicians from six specialties, stenosis was graded correctly only about 55% of the time in routine practice. Accuracy improved with training on a defined grading system. (Walsh et al., Academic Radiology, 2026)
In images from a large national spine study, readers agreed well on central stenosis but only moderately on narrowing where the nerves exit and along the sides of the canal. (Lurie et al., Spine, 2008)
Agreement on facet arthritis was poor enough that researchers called it a common cause of misunderstanding between radiologists and clinicians. (Winklhofer et al., European Spine Journal, 2017)

There is no national standard reporting system for lumbar stenosis.

Words can make a real problem sound minor

Words like “bulge,” “mild,” “shallow,” “abutment” and “degenerative changes” are common and can sound reassuring. But some of these words have no agreed definition, and a herniation can sometimes be described as a “bulge.” Research has shown that the wording of an MRI report changes how doctors and patients think about the problem. See our MRI report glossary for which words are standard and which are not.

Your symptoms make the reading more accurate

Radiologists usually read the MRI without meeting you. That matters:

When readers knew the patient’s symptoms, their agreement on the likely source of pain went from fair to almost perfect. (Balza et al., American Journal of Roentgenology, 2021)
When radiologists spoke with patients for about 3 minutes, 42.6% of reports changed, and about a third of those changes mattered for treatment. (Gutzeit et al., European Radiology, 2021)
The information sent with the MRI order often leaves out important history. (Obara et al., Journal of Digital Imaging, 2015)

What this means for you

An MRI report is a starting point, not the final word.
Bring the actual images, not just the written report.
See a doctor who looks at the images and examines you, and who matches what is on the MRI to where you hurt.
If your report and your symptoms do not match, ask about it. A second look at the images is reasonable.

Dr. Radcliff reviews every patient’s images himself and matches them to the exam. Send your MRI images for a review or call 609-800-3472.

This page is for general education and is not medical advice.