A systematic review to assess comparative effectiveness studies in epidural steroid injections for lumbar spinal stenosis and to estimate reimbursement amounts.

OBJECTIVE: To systematically appraise published comparative effectiveness evidence (clinical and economic) of epidural steroid injections (ESI) for lumbar spinal stenosis and to estimate Medicare reimbursement amounts for ESI procedures. TYPE: Systematic review.

LITERATURE SURVEY: PubMed, Embase, and CINAHL were searched through August 2012 for key words that pertain to low back pain, spinal stenosis or sciatica, and epidural steroid injection. We used institutional and Medicare reimbursement amounts for our cost estimation. Articles published in English that assessed ESIs for adults with lumbar spinal stenosis versus a comparison intervention were included. Our search identified 146 unique articles, and 138 were excluded due to noncomparative study design, not having a study population with lumbar spinal stenosis, not having an appropriate outcome, or not being in English. We fully summarized 6 randomized controlled trials and 2 large observational studies.

METHODOLOGY: Randomized controlled trial articles were reviewed, and the study population, sample size, treatment groups, ESI dosage, ESI approaches, concomitant interventions, outcomes, and follow-up time were reported. Descriptive resource use estimates for ESIs were calculated with use of data from our institution during 2010 and Medicare-based reimbursement amounts.

SYNTHESIS: ESIs or anesthetic injections alone resulted in better short-term improvement in walking distance compared with control injections. However, there were no longer-term differences. No differences between ESIs versus anesthetic in self-reported improvement in pain were reported. Transforaminal approaches had better improvement in pain scores (≤4 months) compared with interlaminar injections. Two observational studies indicated increased rates of lumbar ESI in Medicare beneficiaries. Our sample included 279 patients who received at least 1 ESI during 2010, with an estimated mean total outpatient reimbursement for one ESI procedure "event" to be $637, based on 2010 Medicare reimbursement amounts ($505 technical and $132 professional payments).

CONCLUSION: This systematic review of ESI for treating lumbar spinal stenosis found a limited amount of data that suggest that ESI is effective in some patients for improving select short-term outcomes, but results differed depending on study design, outcome measures used, and comparison groups evaluated. Overall, there are relatively few comparative clinical or economic studies for ESI procedures for lumbar spinal stenosis in adults, which indicated a need for additional evidence.