Research News

Mismatch between spinal MRI and symptoms of sciatica

BMJ 2013; 346 doi: (Published 20 March 2013) Cite this as: BMJ 2013;346:f1716


Magnetic resonance imaging (MRI) of the spine is a poor diagnostic test for people with sciatica that persists or recurs after treatment. Scan results bore little relation to symptoms in one study from the Netherlands, and a linked editorial suggests doctors avoid imaging in this situation (p 1056). Spinal scans are already overused and notorious for finding abnormalities of dubious relevance that trigger a cascade of further tests and treatments, it says.

The latest study compared scan results with symptoms in 283 trial participants one year after surgery or conservative treatment for sciatica and a herniated lumbar disc (both treatments worked equally well). A third of participants with persistent or recurrent symptoms had a herniated disc on the follow-up scan (14/43), but so did a third of those with no symptoms (79/224). In addition, 85% of those with a herniated disc had no symptoms at one year (79/93), as did 83% of those without a herniated disc (145/174).

Traditional tests of diagnostic accuracy confirmed that MRI scans of the spine do not distinguish between patients with and without symptoms one year after treatment (area under the receiver operating characteristic curve 0.48, 95% CI 0.39 to 0.58). The scans identified significantly more herniated discs and nerve root compression one year after conservative treatment than one year after surgery, but neither abnormality was associated with poor outcome.


Cite this as: BMJ 2013;346:f1716