Consensus on the definition and criteria for failure of surgical treatment in bacterial arthritis of a native joint: An international Delphi study

Purpose: The literature presents a wide range of success rates for a single surgical intervention of bacterial-septic-arthritis, and there is a lack of clear criteria for identifying treatment failure and making decisions about reintervention. This Delphi study aims to establish a consensus among an international panel of experts regarding the definition of treatment failure and the criteria for reintervention in case of bacterial arthritis. Methods: The conducting and reporting Delphi studies (CREDES) criteria were used. Data from a systematic review was provided as the basis for the study. A list of 100 potential experts were identified. The study was designed and conducted as follows: (I) identification and invitation of an expert panel, (II) informing the participating expert panel on the research question and subject, and (III) conducting two or three Delphi rounds to reach consensus on explicit research items. Potential criteria were rated on a five-point Likert scale. Results: Sixty orthopaedic experts from nine countries participated in this Delphi study, with 55 completing all three rounds. The mean experience as an orthopaedic surgeon was 15 years (SD ± 9). Strong (96%) consensus was reached on the definition of treatment failure: the persistence of physical signs of arthritis (e.g., pain and swelling) and/or systemic inflammation (e.g., fever and no improvement in CRP) despite surgical and antibiotic treatment. Furthermore, consensus (>80%) was reached on six criteria influencing the decision for reintervention; pain (81%), sepsis (98%), fever (88%), serum CRP (93%), blood culture (82%), and synovial fluid culture (84%). Conclusion: The definition of treatment failure for bacterial arthritis after a single surgical intervention was established through a three-round Delphi study. Additionally, consensus was reached on six criteria that are helpful for determining the need for reintervention. This definition and these criteria may help in the development of clinical guidelines, and will empower physicians to make more precise and consistent decisions regarding reintervention for patients, ultimately aiming to reduce over- and undertreatment and improve patient outcomes. Level of Evidence: Level V.