Utility of repeat magnetic resonance imaging in surgical patients with lumbar stenosis without disc herniation

Document Type



© 2018 Elsevier Inc. BACKGROUND CONTEXT: Routine use of magnetic resonance imaging (MRI) as a diagnostic tool in lumbar stenosis is becoming more prevalent due to the aging population. Currently, there is no clinical guideline to clarify the utility of repeat MRI in patients with lumbar stenosis, without instability, neurological deficits, or disc herniation. PURPOSE: To evaluate the utility of routine use of MRI as a diagnostic tool in lumbar stenosis, and to help formulate clinical guidelines on the appropriate use of preoperative imaging for lumbar stenosis. STUDY DESIGN/SETTING: Retrospective radiographic analysis. PATIENT SAMPLE: Retrospective chart review was performed to review patients with lumbar stenosis, who underwent lumbar decompression without fusion from 2011 to 2015 at a single institution. OUTCOME MEASURES: Previously established stenosis grading systems were used to measure and compare the initial and the subsequent repeat lumbar MRIs performed preoperatively. If patients were found to have a moderate or severe grade change, and if the surgical plan was altered due to such exacerbated radiographic findings, then their grade changes were considered clinically meaningful. METHODS: We identified patients with lumbar stenosis without radiographic instability or neurological deficits, who had at least two preoperative lumbar MRIs performed and underwent decompressive surgeries. At each pathologic disc level, the absolute value of the change in grade for central and lateral recess stenosis, right foraminal stenosis, and left foraminal stenosis from the first preoperative MRI to the repeated MRI was calculated. These changed data were then used to calculate the mean and median changes in each of the three types of stenosis for each pathologic disc level. Identical calculations were carried out for the subsample of patients who only underwent discectomy or had a discectomy included as part of their surgery. RESULTS: Among the 103 patients who met the inclusion criteria, 37 of those patients had more than one level surgically addressed, and a total of 161 lumbar levels were reviewed. Among the subset of patients that had any grade change, the majority of the grades only had a mild change of 1 (36 out of 42 patients, 85.7%, 95% confidence interval [CI]: 73.1%–94.1%); there was a moderate grade change of 2 in two patients (4.8%, CI: 0.8%–14.0%), and a severe change of 3 in one patient (2.4%, CI: 0.2%–10.1%). There were three patients with decreased grade change (7.1%, CI: 1.8%–17.5%). All clinically meaningful grade changes were from the subset of patients who had only discectomy or discectomy as part of the procedure. Lastly, both patients that had a clinically meaningful grade change had their MRIs performed at an interval of greater than 360 days. CONCLUSIONS: The radiographic evaluation of the utility of routinely repeated MRIs in lumbar stenosis without instability, neurological deficits, or disc herniations demonstrated that there were no significant changes found in the repeated MRI in the preoperative setting, especially if the MRIs were performed less than one year apart. The results of this present study can help to standardize the diagnostic evaluation of lumbar stenosis and to formulate clinical guidelines on the appropriate use of preoperative imaging for lumbar stenosis patients.

Publication Date


Publication Title

Spine Journal









First Page


Last Page


PubMed ID


Digital Object Identifier (DOI)


This document is currently not available here.