The aim of this study was to evaluate the efficacy and safety of rifampin for Staphylococcus aureus (SA) or coagulase negative staphylococci (CNS) prosthetic joint infection (PJI) treated with debridement and retention (D/R). We calculated the treatment failure cumulative incidence (TF) of a cohort of 101 patients with SA or CNS PJI treated with D/R and antimicrobial therapy. The effect of the use of a rifampin-based regimen was evaluated. Cox proportional hazards regression evaluated the association between treatment and time-to-TF controlling for the propensity to treat with rifampin and temporal confounders. Seven percent (1/14) of the prospective rifampin-treated patients, 32% (10/31) of the historical rifampin-treated patients and 38% (21/56) of the historical non-rifampin treated patients developed TF. After controlling for the propensity to treat with rifampin and American Society of Anesthesia scores, patients in the prospective cohort had a lower risk of TF compared to patients in the historical cohort not treated with rifampin (HR 0.11; 95%CI 0.01-0.84). None (0/14) of the patients in the prospective study developed hepatotoxicity. The outcome of staphylococcal PJI treated with D/R and rifampin-based regimens was better when compared with a historical cohort treated without rifampin.