Do MRI Findings Change Over a Period of up to 1 Year in Patients With Low Back Pain and/or Sciatica?

A Systematic Review

John Panagopoulos, BAppSci. (Physiotherapy), BMedSci; Julia Hush, PhD; Daniel Steffens, PhD; Mark J. Hancock, PhD


Spine. 2017;42(7):504-512. 

In This Article


Statement of Principal Findings

The present study reveals that there is limited evidence regarding changes in most MRI findings during the short term in patients with LBP or sciatica. Of the 12 available studies with a total of 901 participants, sample sizes in each study were generally small (ranging from 11 to 246), and studies reported different MRI findings, with most data on disc herniations.

The proportion of MRI findings that changed varied between the different MRI findings. Between 0% and 41% of herniations disappeared during the first year, whereas 15% to 93% of herniations reduced in size. Disc sequestrations showed higher rates of change, with 0% to 63% completely resolving, whereas 57% to 100% reduced in size. Similar rates were reported for nerve root compression with up to 90% reducing during 1 year; however, only a small number of studies investigated short-term change in this and other MRI findings. Only one included study provided data on the association between change in MRI findings and change in clinical outcomes. That study found no association between change in disc herniations or nerve root compressions with change in clinical outcome in people with sciatica managed nonsurgically.

Strengths and Weaknesses of the Study

The strengths of this systematic review include the use of a prespecified protocol and use of a highly sensitive search strategy. We assessed risk of bias of included studies using an appropriate modification of a previously reported tool for prognosis studies.

The primary limitation of this systematic review is the small number of included studies investigating change over time of most MRI findings. With regards to our secondary aim we were only able to provide data from a single study. Therefore, these findings must be interpreted with caution. We excluded studies in which participants had undergone spinal surgery, so our results do not generalize to this population.

Comparison With Other Studies

To our knowledge this is the first systematic review to report on changes in MRI findings occurring within a year. A recent systematic review by Chiu et al[27] investigated changes in disc herniations during any follow-up time period. They reported similar rates of reduction in disc sequestration (91%) and herniation (41%). They also reported that disappearance of disc herniations occurred in 43% of sequestered discs and 15% of extruded discs, which aligns with proportions reported in our review. This raises the possibility that most changes in disc herniations occur within 1 year. Of the nine studies they included four (44%) had follow-up time periods that were longer than 12 months. In addition, they only reported on disc herniation changes and included studies which used MRI, computerized tomography and myelography. Our review reported on a broader range of MRI findings and did so in a follow-up time period, which we believe is most clinically relevant.

Meaning of the Study

An understanding of how lumbar spine MRI findings change in patients experiencing LBP or sciatica is crucial for researchers and clinicians to better understand the potential clinical importance of MRI findings. If lumbar MRI findings do not change during 1 year after an episode of LBP, then it is hard to understand how they can be important to the clinical course of pain, which usually changes substantially during this period.[28] Our review found that some MRI findings do appear to change substantially (e.g., reduce in size or fully resolve) and therefore may be associated with changes in symptoms. Understanding how MRI findings change, can be used to reassure patients and provide a rationale for why their pain may be expected to reduce in the future. Although this review, provides evidence of MRI findings changing over time, it is unclear how these changes relate to changes in clinical outcomes. In the one study in our review that investigated this, there was no association between change in MRI findings and change in pain. Understanding that some MRI findings typically resolve well, during 1 year, provides clinicians with important information that can be used to reassure patients. For patients to understand that some MRI findings can improve or resolve naturally can be useful information to facilitate patient-centered, informed treatment decisions during the course of an episode of LBP or sciatica. It is important to acknowledge that, in some cases, MRI findings either do not change or worsen over time and this may contribute to the range of pain, disability, and functional loss experienced by different patients.

Recommendations for Further Research

Large high-quality longitudinal studies are required to better understand how different MRI findings change during an episode of LBP and/or sciatica and also to determine whether these changes contribute to changes in clinical symptoms. Future studies should assess changes in MRI findings and clinical outcomes in the period when symptoms change most (3 months), and ensure use of standardized methods to evaluate MRI findings and clinical outcomes. It is important for future studies to investigate the change in different MRI findings (e.g., disc herniation and spinal stenosis) separately as both the proportion of change and relation with back pain are likely to be different. It is also important to investigate the relationship of the MRI findings separately for LBP and radicular pain.