Value of the magnetic resonance imaging in patients with painful lumbar spinal stenosis (LSS) undergoing lumbar epidural steroid injections

Objectives: Purpose of this study is to examine the relationship between the magnetic resonance imaging (MRI) findings, pain scores, and opiates use in patients with lumbar spinal stenosis (LSS) undergoing lumbar epidural steroid (LES) injections by retrospective review of 719 patients' electronic medical records. Methods: Reviewed were Visual Analog Scale (VAS) pain scores and opioid use before and 8 to 12 weeks after series of LES injections. The stenosis pain index (SPI) was produced by adding an assigned numerical value of severity (1 = mild, 2 = moderate, 3 = severe) to the number of lumbar vertebral levels affected by LSS on MRI (lateral or central). Results: The average age of patients was 68.4 years. There was no relationship between the pretreatment age, sex, or number of vertebral levels affected on MRI with pretreatment VAS pain scores or opioid use. The degree of LSS present on MRI, categorized as a mild, moderate, or severe, correlated clearly with initial VAS pain scores (P = 0.017). The improvement in VAS pain scores after LES injections correlated well with number of levels affected (P = 0.003) and the severity of stenosis (P = 0.12). Positive correlation was observed between change in VAS pain score 8 to 12 weeks after the series of LES injections and the SPI (P = 0.001). There were no differences found in opioid use. Discussion: The improvement in VAS pain scores after LES injections correlated well with the changes in the SPI except in those patients classified on MRI as severe LSS and more than 3 lumbar levels affected. That patient group is unlikely to benefit from LES injections.