Do Open Fracture Classification Systems Predict Functional Outcomes in a Low-Income Country?: A Prospective Study of 287 People with Open Tibia Fractures in Malawi.

Background: Open fractures are common and severe injuries that are associated with poor functional outcomes and quality of life, and high societal costs. Several classifications systems have been developed to characterize these injuries, predict prognosis and plan treatment. We aimed to assess the agreement between open fracture classification and patient-reported function, fracture-related infection, and amputation.

Methods: In this prospective cohort study, we enrolled adults with open tibia fractures in 6 hospitals across Malawi. Radiographs and clinical photographs were classified according to the Gustilo-Anderson, Orthopaedic Trauma Society (OTS), Muller and Tscherne classification by an orthopaedic surgeon. Participants' function (using the Short Musculoskeletal Assessment Score), and risk of fracture-related infection and amputation were assessed by face-to-face interviews at 6 weeks, 3 months, 6 months, and 1 year postinjury. The Kendall rank correlation coefficient with 95% bootstrapped confidence intervals investigated correlation between fracture classifications and patient outcomes.

Results: Two hundred eighty-seven participants were recruited with 252 of 287 photographs (88%) and 274 of 287 radiographs (95%) available for review. The Kendall correlation with function score 1 year after injury was 0.34 for OTS classification, 0.18 for Gustilo, 0.17 for Tscherne, and -0.02 for Muller. For correlation with fracture-related-infection at 1 year, this was 0.34 for the Orthopaedic Trauma Society, 0.31 for Gustilo, 0.24 for Tscherne, -0.02 for Muller. For amputation, correlation was 0.39 for OTS, 0.24 for Gustilo, 0.24 for Tscherne, and 0.12 for Muller.

Discussions: Most open fracture classification systems had negligible or weak correlation with subsequent patient function, fracture-related infection, or amputation in Malawi. Treatment factors and other confounders may have a greater influence on outcomes, but classification systems must still account for this variability to remain useful. More research in low or middle income countries needs to be conducted to develop appropriate and relevant open fracture classifications to improve and standardize the management of open fractures.

Level of evidence: Level II. See Instructions for Authors for a complete description of levels of evidence.