BACKGROUNDPerineural inhibitors of tumor necrosis factor have recently generated intense interest as an alternative to epidural steroid injections for lumbosacral radiculopathy.OBJECTIVETo evaluate whether epidural steroids, etanercept, or saline better improves pain and function in adults with lumbosacral radiculopathy.DESIGNA multicenter, 3-group, randomized, placebo-controlled trial conducted from 2008 to 2011. Randomization was computer-generated and stratified by site. Pharmacists prepared the syringes. Patients, treating physicians, and nurses assessing outcomes were blinded to treatment assignment. (ClinicalTrials.gov registration number: NCT00733096)SETTINGMilitary and civilian treatment centers.PATIENTS84 adults with lumbosacral radiculopathy of less than 6 months' duration.INTERVENTION2 epidural injections of steroids, etanercept, or saline, mixed with bupivacaine and separated by 2 weeks.MEASUREMENTSThe primary outcome measure was leg pain 1 month after the second injection. All patients had 1-month follow-up visits; patients whose condition improved remained blinded for the 6-month study period.RESULTSThe group that received epidural steroids had greater reductions in the primary outcome measure than those who received saline (mean difference, -1.26 [95% CI, -2.79 to 0.27]; P = 0.11) or etanercept (mean difference, -1.01 [CI, -2.60 to 0.58]; P = 0.21). For back pain, smaller differences favoring steroids compared with saline (mean difference, -0.52 [CI, -1.85 to 0.81]; P = 0.44) and etanercept (mean difference, -0.92 [CI,-2.28 to 0.44]; P = 0.18) were observed. The largest differences were noted for functional capacity, in which etanercept fared worse than the other treatments: steroids vs. etanercept (mean difference, -16.16 [CI, -26.05 to -6.27]; P = 0.002), steroids vs. saline (mean difference, -5.87 [CI, -15.59 to 3.85]; P = 0.23), and etanercept vs. saline (mean difference, 10.29 [CI, 0.55 to 20.04]; P = 0.04). More patients treated with epidural steroids (75%) reported 50% or greater leg pain relief and a positive global perceived effect at 1 month than those who received saline (50%) or etanercept (42%) (P = 0.09).LIMITATIONShort-term follow-up, small sample size, and a possibly subtherapeutic dose of etanercept.CONCLUSIONEpidural steroid injections may provide modest short-term pain relief for some adults with lumbosacral radiculopathy, but larger studies with longer follow-up are needed to confirm their benefits.PRIMARY FUNDING SOURCEThe John P. Murtha Neuroscience and Pain Institute, International Spinal Intervention Society, and Center for Rehabilitation Sciences Research.