A matter of definition : variation in treatment success of fracture-related infections.

AIMS: The establishment of the fracture-related infection (FRI) consensus definition and the FRI classification system provide a foundation for research and the development of clinical recommendations. To date, however, the evaluation of treatment success is still impeded by a lack of clear and standardized outcome definitions. We hypothesized that applying different outcome measurements to the same cohort of FRI patients results in significantly different treatment success rates.

METHODS: Adult patients undergoing curative surgery for FRI at the Center of Musculoskeletal Infections, University Hospital Basel, between 1 January 2020 and 31 December 2023 were included in this prospective cohort study. Treatment success was evaluated by review of patient charts; after one year of follow-up according to seven predefined outcome definitions (S1 to S7): no infection persistence at one year (S1); no recurrence within one year (with the same (S2) or a new (S3) pathogen); bony union (S4); no infection recurrence, or persistence and bony union (S5); no unplanned revision surgery (S6); and bony union and no infection recurrence, or persistence and no unplanned revision surgery (S7).

RESULTS: In total, 106 patients with 112 confirmed FRIs were included. At one year, treatment success varied significantly (p < 0.001) between the different outcome definitions: for S1 the success rate was 95.5% (n = 107), for S2 89.3% (n = 100), for S3 92.9% (n = 104), and for S6 60.7% (n = 68). In cases with healed fractures (n = 28) success rate was 100% for S1 to 3 definitions and 92.9% for S6 (n = 26). In cases with non-healed fractures (n = 84) the success rate decreased significantly from 94.0% (n = 79) for the S1 definition to 39.3% for S7 (n = 33; p < 0.001), with 50% of these patients (n = 42) undergoing an unplanned revision surgery. In unhealed fractures a recurrent infection (p < 0.013) and the need for an unplanned revision surgery (p < 0.001) were significantly more frequent than in healed fractures.

CONCLUSION: At the one-year follow-up, treatment success differed significantly within a single cohort depending on the outcome metric used to define success.