Objectives:

Low back pain is common and injections with corticosteroids are a frequently used treatment option. This report reviews the current evidence on effectiveness and harms of epidural, facet joint, and sacroiliac corticosteroid injections for low back pain conditions.

Data Sources:

A prior systematic review (searches through July 2008), electronic databases (Ovid MEDLINE, Scopus, and the Cochrane Libraries from January 2008 through October 2014), reference lists, and clinical trials registries.

Review Methods:

Using predefined criteria, we selected randomized trials of patients with lumbosacral radiculopathy, spinal stenosis, nonradicular back pain, or chronic postsurgical back pain that compared effectiveness or harms of epidural, facet joint, or sacroiliac corticosteroid injections versus placebo or other interventions. We also included randomized trials that compared different injection techniques and large (sample sizes >1000) observational studies of back injections that reported harms. The quality of included studies was assessed, data were extracted, and results were summarized qualitatively and using meta-analysis on outcomes stratified by immediate- (1 week to ≤2 weeks), short- (2 weeks to ≤3 months), intermediate- (3 months to <1 year), and long-term (>1 year) followup.

Results:

Seventy-eight randomized trials of epidural injections, 13 trials of facet joint injections, and one trial of sacroiliac injections were included. For epidural corticosteroid injections versus placebo interventions for radiculopathy, the only statistically significant effects were on mean improvement in pain at immediate-term followup (weighted mean difference [WMD] −7.55 on a 0 to 100 scale, 95% CI −11.4 to −3.74) (strength of evidence [SOE]: moderate), mean improvement in function at immediate-term followup when an outlier trial was excluded (standardized mean difference [SMD] −0.33, 95% CI −0.56 to −0.09) (SOE: low), and risk of surgery at short-term followup (relative risk [RR] 0.62, 95% CI 0.41 to 0.92) (SOE: low). The magnitude of effects on pain and function was small, did not meet predefined thresholds for minimum clinically important differences, and there were no differences on outcomes at longer-term followup. Evidence on effects of different injection techniques, patient characteristics, or comparator interventions estimates was limited and did not show clear effects. Trials of epidural corticosteroid injections for radiculopathy versus nonplacebo interventions did not clearly demonstrate effectiveness (SOE: insufficient to low).

Evidence was limited for epidural corticosteroid injections versus placebo interventions for spinal stenosis (SOE: low to moderate) or nonradicular back pain (SOE: low), but showed no differences in pain, function, or likelihood of surgery.

Studies found no clear differences between various facet joint corticosteroid injections (intra-articular, extra-articular [peri-capsular], or medial branch) and placebo interventions (SOE: low to moderate). There was insufficient evidence from one very small trial to determine effects of peri-articular sacroiliac joint corticosteroid injections injection (SOE: insufficient).

Serious harms from injections were rare in randomized trials and observational studies, but harms reporting was suboptimal (SOE: low).

Conclusions:

Epidural corticosteroid injections for radiculopathy were associated with immediate improvements in pain and might be associated with immediate improvements in function, but benefits were small and not sustained, and there was no effect on long-term risk of surgery. Evidence did not suggest that effectiveness varies based on injection technique, corticosteroid, dose, or comparator. Limited evidence suggested that epidural corticosteroid injections are not effective for spinal stenosis or nonradicular back pain and that facet joint corticosteroid injections are not effective for presumed facet joint pain. There was insufficient evidence to evaluate effectiveness of sacroiliac joint corticosteroid injections.