Follow-up MRIs can't distinguish between one-year outcomes after low-back procedures

Magnetic resonance imaging (MRI) performed as a one-year follow-up in patients who had been treated for sciatica and lumbar-disk herniation did not distinguish patients with persistent or recurrent symptoms of sciatica from asymptomatic patients, a study found.

Researchers studied 267 patients from the Sciatica Trial, a nine-center, randomized trial of patients with a history of six to 12 weeks of sciatica. In the study, patients were randomized to either surgery or prolonged conservative therapy, with 44% of the study group ultimately being treated surgically. Outcomes were evaluated by intention-to-treat analysis.

All patients underwent a follow-up MRI one year after randomization, which was interpreted for evidence of disc herniation based on standard criteria by experienced neuroradiologists blinded to the clinical outcome. Patient outcomes were assessed using a 7-point Likert scale for global perceived recovery, with a favorable clinical outcome defined as complete or nearly complete disappearance of symptoms at one year.

Results appeared in the March 14 New England Journal of Medicine.

At one year, disk herniation was visible in 35% of the patients with a favorable outcome (no symptoms) and in 33% of those with an unfavorable outcome (95% CI for difference in proportion, −18.8 to 12.6; P=0.70). Nerve-root compression was present in 24% of the patients with a favorable outcome and in 26% of the patients with an unfavorable outcome. Similar results were seen in patients with and without persistent leg and back pain at one year, as well as those who would be considered more or less disabled based on responses to the Roland Disability Questionnaire.

The absence of disk herniation on follow-up MRI was not associated with a favorable outcome at one year (odds ratio, 0.82; 95% CI, 0.40 to 1.71; P=0.60), nor was the absence of MRI-assessed nerve-root compression (odds ratio, 1.03; 95% CI, 0.48 to 2.25; P=0.93). The size of the disk herniation (odds ratio, 1.48; 95% CI, 0.43 to 5.01; P=0.53) or the herniation form (protrusion vs. extrusion) (odds ratio, 0.88; 95% CI, 0.25 to 3.16; P=0.85) on follow-up MRI also had no association with a favorable clinical outcome.

The researchers wrote, “[P]atients asking for reimaging because of persistent or recurrent symptoms should be informed about the difficulty in MRI interpretation after a first episode of acute sciatica. A recent systematic review concluded that even in the acute setting of sciatica, evidence for the diagnostic accuracy of MRI is not conclusive.”