Epidural Injections With, Without Steroids Effective for Lumbar Radiculopathy

Doctor is inserting a fine plastic tube called an epidural catheter into a patient’s back
A systematic review and meta-analysis of RCTs was performed to evaluate the efficacy of various types of epidural steroid injections for lumbar radiculopathy.

Epidural injections of local anesthetic with or without steroids may effectively manage lumbar radiculopathy, with moderate to strong evidence available, according to a systematic review and meta-analysis of randomized controlled trials results published in Pain Physician.1

In the current review, researchers sought to evaluate the efficacy—or lack thereof—of epidural injections with saline, local anesthetic alone, or local anesthetic with steroids, and compare those results with results of a recently published Cochrane Review.2,3

Literature published through January 2021 was eligible for inclusion. Predefined inclusion criteria included fluoroscopic guidance and outcomes reporting to at least 6 months, epidural injections with sodium chloride solution, local anesthetic, or steroids via caudal, interlaminar, or transforaminal approaches.

The predefined outcomes were measurement of pain and function with a description of composite outcomes with significant pain and an improvement in functional status of 50% or more.

A total of 21 trials met inclusion criteria, 7 of which evaluated caudal epidural injections, 10 of which evaluated interlaminar epidural injections a d 12 of which assessed transforaminal epidural injections.

The original Cochrane review included 25 trials, multiple of which did not meet the current predefined inclusion criteria. Among the 21 trials in the current review, 15 were not included in the Cochrane review, representing an overlap of 6 trials.

According to investigators, the current assessment demonstrated the importance of interventional pain management-specific scoring that utilizes Interventional Pain Management Techniques – Quality Appraisal of Reliability and Risk of Bias (IPM-QRB) criteria. These criteria has shown assessment results that are different from Cochrane review-derived data. In 19 of 21 trials, there was agreement between Cochrane review and IPM-QRB scoring.

Among the included studies, there was 1 placebo-controlled trial and 1 study that compared conservative management; 7 studies compared local anesthetic alone with local anesthetic plus steroids, and the remaining studies compared technical aspects or dose responses.

Across 5 studies, inclusive of 527 patients, that compared local anesthetics with steroids vs local anesthetic alone in a dual-arm meta-analysis. Results of this review showed no statistically significant difference between these 2 groups (standardized mean difference, 0.19; 95% CI, -0.49 to 0.87).

Five studies assessed pain score after 6 months using the numeric rating scale (NRS) in patients who received epidural local anesthetic injections. Pooled mean difference of pain scores from baseline through 6 months of follow-up decreased by 3.637 points (95% CI, -3.787 to -3.487).

Another 5 studies assessed pain score at 6 months using the NRS in patients who received epidural steroid injections. The pooled mean difference of pain scores between baseline and 6-month follow-up decreased by 4.105 points (95% CI, -4.2024 to -4.005).

Results of a conventional dual-arm analysis suggested that although there was no significant difference between local anesthetic and local anesthetic plus steroids, there was a “slight advantage” in favor of local anesthetic with steroids (pain decrease, 3.637 vs 4.015 points).

Five studies including 527 patients compared local anesthetics with combined local anesthetics and steroids. There was no statistically significant difference between these groups (standardized mean difference, 0.70; 95% CI, -0.11 to 1.51).

Researchers also conducted a single-arm analysis of 5 studies of local anesthetics, in which functionality score at 6 months was evaluated. Pooled mean difference of pain score decreased by 13.697 points (95% CI, -13.785 to -13.609). Analyses of functionality show statistically similar results in those who received local anesthetic alone vs local anesthetic with steroids, although a trend towards higher improvement with local anesthetic plus steroids was noted.

In a conventional dual-arm analysis, 1-year follow-up showed results with no significant difference between local anesthetic alone and local anesthetic with steroids. No tendency was noted for superiority of either treatment. In a single-arm analysis, however, of local anesthetic alone, difference in pain scores from baseline to 12 months was a 3.773-point decrease.

Changes in functionality at 1 year were also evaluated from 5 studies with 527 patients comparing local anesthetic with steroids vs local anesthetic alone. No statistically significant between-group difference was noted (standardized mean difference, 0.61; 95% CI, -0.14 to 1.35).

“Qualitative and quantitative evidence synthesis showed Level I or strong evidence for the effectiveness of lumbar epidural injections with local anesthetic and steroids and Level II to I or moderate to strong evidence for local anesthetic alone in managing lumbar radiculopathy or sciatica secondary to lumbar disc herniation,” the researchers concluded.

Disclosure: Study authors declared affiliations with the pharmaceutical industry. Please see the original reference for a full list of authors’ disclosures.


  1. Manchikanti L, Knezevic E, Knezevic NN, et al. Epidural injections for lumbar radiculopathy or sciatica: A comparative systematic review and meta-analysis of Cochrane review. Pain Physician. 2021;24(5):E539-E554.
  2. Oliveira CB, Maher CG, Ferreira ML, et al. Epidural corticosteroid injections for lumbosacral radicular pain. Cochrane Database Syst Rev. 2021;4(4):CD013577. doi:10.1002/14651858.CD013577
  3. Oliveira CB, Maher CG, Ferreira ML, et al. Epidural corticosteroid injections for sciatica: An abridged Cochrane systematic review and meta-analysis. Spine. 2021;45(21):E1405-E1415. doi:10.1097/BRS.0000000000003651