Postoperative Infection Risk in Total Joint Arthroplasty After Perioperative IV Corticosteroid Administration: A Systematic Review and Meta-Analysis of Comparative Studies

BackgroundPerioperative corticosteroid administration is associated with reduced postoperative nausea, pain, and enhanced recovery after surgery. However, potential complications including wound and periprosthetic joint infections remain a concern for surgeons after total joint arthroplasty (TJA).MethodsA systematic review of the search databases PubMed, Google Scholar, and EMBASE was made in January 2021 to identify comparative studies evaluating infection risk after perioperative corticosteroid administration in TJA. PRISMA guidelines were used for this review. Meta-analysis was used to assess infection risk in accordance with joint and corticosteroid dosing regimen used.Results201 studies were returned after initial search strategy, with 29 included for review after application of inclusion and exclusion criteria. Studies were categorized as using low- or high-dose corticosteroid with single or repeat dosing regimens. Single low-dose corticosteroid administration was not associated with an increased risk of infection (P = .4; CI = 0.00-0.00). Single high-dose corticosteroid was not associated with an increased infection risk (P = .3; CI = 0.00-0.01) nor did repeat low-dose regimens result in increased risk of infection (P = .8; CI = ?0.02-0.02). Studies assessing repeat high-dosing regimens reported no increased infection, with small numbers of participants included. No significant risk difference in infection risk was noted in hip (P = .59; CI = ?0.03-0.02) or knee (P = .2; CI = 0.00-0.01) arthroplasty. Heterogeneity in patient profiles included in studies to date was noted.ConclusionUse of perioperative corticosteroid in TJA does not appear to be associated with increased risk of postoperative infection in patients with limited comorbidities. Further research is warranted to evaluate postoperative complications after TJA in these at-risk patient populations.