Epidural Steroids Generally Are Ineffective for Lumbar Spinal Stenosis

Despite widespread use of epidural steroid injections to treat patients with spinal stenosis, high-quality evidence to support their efficacy is nonexistent. In this multicenter, double-blind US study, researchers randomized 400 patients with symptomatic central lumbar spinal stenosis to receive an epidural injection of either lidocaine alone or lidocaine plus a glucocorticoid. All patients had moderate-to-severe pain (with pain in the buttock or leg greater than that in the back) and functional disability. About 40% of patients in each group requested and received a second injection at 3 weeks.

During 6 weeks of follow-up, mean scores on pain and disability scales (the primary outcomes) improved substantially in both the steroid and no-steroid groups. However, at 6 weeks, no significant differences were observed between groups for either of these primary outcomes. At 3 weeks, small, statistically significant differences favored the steroid group (a 0.6-point difference in pain on a 10-point scale, and a 1.8-point difference in disability on a 24-point scale), but these differences were deemed clinically unimportant. The proportions of patients with at least 30% improvement in pain or disability at 6 weeks were similar in the two groups. Interestingly, 10% of patients in the steroid group (but <1% of those in the no-steroid group) had morning serum cortisol levels <3 μg/dL at 3 weeks, suggesting a surprisingly long duration of adrenal suppression after a single epidural injection in some cases.