Utility of Follow-Up MRI for Sciatica

Abstract & Commentary

By Michael Rubin, MD, Professor of Clinical Neurology, Weill Cornell Medical College. Dr. Rubin reports no financial relationships relevant to this field of study.

Synopsis: After successful treatment of sciatica, routine follow-up MRI of the lumbar spine provides no useful information. Recurrent or persistent symptoms mandate additional evaluation based on clinical symptoms and signs.

Source: el Barzouhi A, et al. Magnetic resonance imaging in follow-up assessment of sciatica. N Engl J Med 2013;368:999-1007.

Afflicting 80% of Americans at some point in their lives, and 50% in any given year, low back pain is the second leading cause for physician visits in this country, the third leading cause for surgical procedures, the fifth leading cause for hospitalizations, and accounts for more than $24 billion a year in direct medical costs. Best practices that might positively impact these statistics would be welcome, and the present study, examining the value of obtaining follow-up magnetic resonance imaging (MRI) at 1 year in patients with persistent sciatica and known to have lumbar-disc herniation, is a first step in this direction.

Patients in this study had participated in a multicenter, randomized trial comparing conservative management to surgery for lumbar disc herniation with sciatica, having been eligible if they experienced 6-12 weeks of sciatica and demonstrated disc herniation on MRI, with pain in a dermatomal distribution associated with neurologic findings corresponding to the nerve root affected on imaging. MRI was performed at baseline and repeated at 1 year. Complete or near complete resolution of patient-reported symptoms at 1 year, using the 7-point Likert self-rating scale of global perceived recovery, was defined as a favorable outcome, and was measured at baseline and at 2, 4, 8, 12, 26, 38, and 52 weeks. Patients were blinded to the results of their MRI findings and prior assessments. Statistical analysis included student’s t-test and logistic regression models, and model-based multiple imputation was used to account for missing data.

Among 599 patients screened for the trial, 283 were randomized, and after 1 year, 267 were available for a second MRI (94.3%). Of these 267 patients, 131 had been randomized to surgery, of which 15 recovered before surgery, and 136 were randomized to receive conservative care, of which 54 underwent surgery within the first year. At 1 year, a favorable outcome was achieved in 84%, based on the 7-point Likert scale. In the as-treated analysis, a herniated disc was present at 1 year in 21% of patients treated surgically and 60% of those treated conservatively. In the intention-to-treat analysis, a herniated disc was present in 22% of the former and 47% of the latter. Among those with a favorable outcome, 35% demonstrated disc herniation on MRI at 1 year, compared to 33% among those with an unfavorable outcome, with nerve root compression visible on imaging in 24% and 26%, respectively. Of 170 patients who had surgery, 88% (150) demonstrated a visible scar on MRI, of which 86% reported a favorable outcome, compared to 75% with no scar (P = 0.19). Among those with a definite, probable, or possible disc herniation, favorable outcome was reported in 85% at 1 year, compared to 83% among those with no disc herniation. Follow-up MRI at 1 year did not distinguish between those who improved and those who did not, and MRI is not warranted in this clinical setting as a routine follow-up examination.


How does the utility of electrodiagnostic studies (EMG) compare with MRI in patients with radicular leg pain? Among 152 subjects with sciatica for at least 6 weeks’ duration and no prior history of surgery, malignancy, autoimmune disease, or trauma, MRI and EMG were abnormal in 104 (68.4%), whereas both were normal in 10 (6.5%). MRI alone was abnormal in 30 patients (19.7%) while EMG was the sole abnormal test in 21 (13.8%). EMG can be of value in MRI-negative sciatica patients.1


1. Hasankhani EG, Omidi-Kashani F. Magnetic resonance imaging versus electrophysiologic tests in clinical diagnosis of lower extremity radicular pain. ISRN Neuroscience 2013; Available at: http://dx.doi.org/10.1155/2013/952570). Accessed April 16, 2013.