Abstract
Background and purpose - Immediate postoperative pain management offered in knee arthroplasty is suboptimal in up to one-third of patients resulting in high opiate consumption and delayed discharge. In this meta-analysis we investigate the analgesic effect and safety of perioperative adjuvant corticosteroids in knee arthroplasty. Methods - Databases Medline, Embase, and Central were searched for randomized studies comparing the analgesic effect of adjuvant perioperative corticosteroids in knee arthroplasty. Our primary outcome was pain score at 24 hours postoperatively. Secondary outcomes included pain at 12, 48, and 72 hours, opiate consumption, postoperative nausea and vomiting, infection, and discharge time. Systemic (intravenous) and local (intra-articular) corticosteroids were analyzed separately. Results - 14 randomized controlled trials (1,396 knees) were included. Mean corticosteroid dosages were predominantly 50-75mg oral prednisolone equivalents for both systemic and local routes. Systemic corticosteroids demonstrated statistically significant and clinically modest reductions in pain at 12 hours by -1.1 points (95%CI -2.2 to 0.02), 24 hours by -1.3 points (CI -2.3 to -0.26) and 48 hours by -0.4 points (CI -0.67 to -0.04). Local corticosteroids did not reduce pain. Opiate consumption, postoperative nausea and vomiting, infection, or time till discharge were similar between groups. Interpretation - Corticosteroids modestly reduce pain postoperatively at 12 and 24 hours when used systemically without any increase in associated risks for dosages between 50 and 75 mg oral prednisolone equivalents.