Extended Oral Antibiotics After Primary Total Hip and Knee Arthroplasty in an Ambulatory Surgery Center: A Bayesian Threshold and Cost-Effectiveness Analysis.

Background: Given evidence for extended oral antibiotic prophylaxis (EOA), we compared EOA with no EOA after primary total hip and knee arthroplasty in an ambulatory surgery center (ASC) using Bayesian and probabilistic cost modeling. Methods: This retrospective study included same-day discharge THA/TKA cases at a single ASC. The primary outcome was 90-day periprosthetic joint infection (PJI). Bayesian logistic regression estimated posterior odds ratios (ORs) and probabilities of benefit. Probabilistic cost-effectiveness analysis incorporated antibiotic and debridement, antibiotics, and implant retention costs. Results: Among 7723 TKAs (EOA n = 2337; no-EOA n = 5386) and 5872 THAs (EOA n = 1670; no-EOA n = 4202), 90-day PJI rates were 0.13% versus 0.26% in TKA and 0.48% versus 0.40% in THA. Median posterior ORs were 0.58 for TKA (84.3% probability of benefit) and 1.24 for THA (30.7%). In TKA, the absolute risk reduction was 1.13 per 1000 cases, yielding a number needed to treat of 885;clinically favorable values would require baseline PJI rates of 0.8%-1.2%. With costs of $15 for EOA and $30,000 for PJI, EOA was cost-effective at $50,000 per PJI averted and cost-saving when PJI costs exceeded $13,000 in TKA, but not THA. Patient-reported outcome improvements were similar. Conclusions: In this low-risk, same-day discharge ASC cohort, EOA showed a modest probablitly of benefit and favorable economic performance in TKA. These findings support selective, risk-based use consistent with antibiotic stewardship.