Multisociety Letter to the Agency for Healthcare Research and Quality: Serious Methodological Flaws Plague Technology Assessment on Pain Management Injection Therapies for Low Back Pain.

Reviewer 1 Results I just attended (and spoke) at the FDA meeting on epidural steroid injections. There was nearly universal agreement that ESI were better than non-ESI through 6 weeks, which is different than this analysis. I spoke on effectiveness and showed a metaanalysis with 8 of the top studies (Tafazal, Karppinen, Cohen, Ghahreman etc.) that showed a 15% improvement at 4 weeks and a 10% at 6 weeks compared to the control (most of the time an epidural local anesthetic injection). Dr. John Farrar was there who did all of the research on "clinically meaningful benefit", serving on the panel. Basically, the difference compared to baseline in the ESI group was > 30%. The difference between the ESI and control group was > 10% through 6 weeks, which is more than the improvement for nearly all drugs used for neuropathic pain compared to placebo (pregabalin, gabapentin, duloxetine). We reviewed the materials from the FDA session (http://www.fda.gov/AdvisoryCommittees/CommitteesMeetingMaterials/Drugs/A nestheticAndAnalgesicDrugProductsAdvisoryCommittee/ucm425962.htm). The analysis described was based on a selected set of studies and was not systematic; we reviewed the references and found none that met inclusion criteria. We also found epidural steroid injections associated with a small improvement in pain at up to 4 weeks, so we do not think there is an inconsistency there. As described in the report, there were no differences between epidural steroid injections vs. placebo interventions in the proportion of patients experiencing a successful outcome as defined by the trials (most commonly >50% improvement in pain). We added a reference to the FDA materials to the Discussion.