BACKGROUND Better outcomes have been reported for two-stage total hip arthroplasty (THA) revision for infection. However, one-stage revision arthroplasty remains an attractive alternative option since it requires only one operation. A decision tree has been developed by the authors in order to determine which type of surgical procedure can be performed safely. The goal of this study was to assess this decision tree for THA replacement in the case of a peri-prosthetic infection. OBJECTIVE A one-stage procedure may be as successful as a two-stage procedure provided some criteria are fulfilled. METHODS A prospective study included 84 patients, all diagnosed with infected THA who had prosthesis replacement. A one-stage exchange was performed in 38 cases and a two-stage procedure in 46 cases. A two-stage procedure was decided in the case of important bone loss or unidentified germ. Postoperatively, patients received intravenous antibiotics (six weeks), then oral antibiotics (six weeks). The main evaluation criterion was the rate of infection eradication at 2 years minimal follow-up since surgery. If new infection was suspected, a hip aspiration was performed to determine whether it was non-eradication (same germ) or a new re-infection (other germ), which was not considered as a failure. RESULTS The initial infection was cured in 83 out of 84 patients (98.8%), 38 (100%) for the one-stage group and 45 (97.8%) for the two-stage group. Three patients were re-infected with different germs in the two-stage group. Eighty out of 84 (95.2%) patients were infection free, all patients (100%) of the one-stage group and 42 patients (91.3%) of two-stage group. CONCLUSIONS If some selection criteria were respected, a high success rate in THA replacement for infection may be achieved with a one-stage procedure. It permits to reduce the costs with no loss of chance for the patients. The decision tree was validated. METHODS Level III; prospective case control study.