Failure Rates After Treatment of Periprosthetic Joint Infection Using Musculoskeletal Infection Society Outcome Reporting Tool, Delphi, and Clinical Outcome Definitions.

BACKGROUND: Reported success rates after treatment of periprosthetic joint infection (PJI) vary widely, in part because treatment failure is not defined uniformly. Modern consensus outcome frameworks such as the Musculoskeletal Infection Society Outcome Reporting Tool (MSIS ORT) and Delphi criteria may not fully capture clinically meaningful outcomes. This study compared failure rates after treatment of PJI using MSIS-ORT, Delphi, and a pragmatic clinical definition; assessed agreement between these definitions at the patient level; and examined the criteria responsible for discordant classifications.

METHODS: A retrospective cohort study was performed of 198 unique hip and knee PJI episodes treated at a single academic medical center from 2007 to 2024. The primary analytic cohort consisted of 184 episodes with either a documented failure event before 12 months or documented follow-up of at least 12 months. Failure rates were calculated using MSIS-ORT, Delphi, and pragmatic clinical definitions. Agreement between definitions was assessed using cross-tabulations, and discordant cases were reviewed to identify the criteria responsible for disagreement.

RESULTS: In the 184-episode primary analytic cohort, failure rates were 43.5% (80 of 184) by MSIS-ORT, 40.8% (75 of 184) by Delphi, and 47.3% (87 of 184) by the clinical definition. Among debridement, antibiotics, and implant retention (DAIR) procedures, failure rates were 44.8, 42.4, and 52.0%, respectively; among staged revisions, failure rates were 40.7, 37.3, and 37.3%, respectively. The MSIS ORT and Delphi showed high concordance, but 13 of 104 episodes (12.5%) classified as successful by MSIS-ORT were considered failures under the clinical definition.

CONCLUSIONS: Failure after treatment of PJI remained common regardless of the outcome framework applied. Although MSIS ORT and Delphi produced similar aggregate estimates, these frameworks were not fully interchangeable with a pragmatic clinical definition at the patient level. Definitional choice materially influences interpretation of PJI treatment outcomes.