Evaluation of magnetic resonance imaging following neuraxial steroid administration: does epidural injection produce pathologic findings?

BACKGROUND AND OBJECTIVES: Infection or hematoma following epidural anesthesia is a rare but potentially devastating complication unless diagnosed early. In order to diagnose spinal cord involvement, the currently recommended imaging test is magnetic resonance imaging (MRI). Despite this, no previous studies have been performed to define typical MRI findings following uneventful epidural injection. The purpose of this pilot study was to compare magnetic resonance images before and after epidural injection to define the characteristic appearance of MRI following an uneventful epidural steroid injection.
METHODS: Ten patients were prospectively enrolled and received an MRI study before and after an epidural injection of steroids plus local anesthetic for chronic low back pain using a loss of resistance to air technique. The magnetic resonance images were reviewed by 2 neuroradiologists who were blinded as to whether the scan was performed before or after the epidural injection.
RESULTS: Review of 20 MRI studies revealed no fluid collections, hematomas, or mass effect on the thecal sac. A needle track and a small amount of epidural air was seen in the majority of post-epidural scans, but did not cause indentation of the thecal sac in any patient. None of the findings observed by the radiologists in the post-epidural scans were considered clinically significant.
CONCLUSIONS: The results of this pilot study demonstrated that after uneventful epidural injection in otherwise healthy patients, there were no new pathologic findings on MRI scans. Further MRI studies in large populations and in different clinical situations should be performed to confirm these preliminary findings.