• ABSTRACT
    • The aim of this study was to assess inter- and intraobserver agreement of the traditional systems (Ruedi-Allgower, AO [Arbeitsgemeinschaft für Osteosynthesefragen], and Topliss) and the newly proposed Leonetti classification system of pilon fractures. We studied all patients at our center who underwent pilon fracture surgery over a 2-year period: 68 patients (70 legs) were included. Four observers independently classified each pilon fracture according to the Ruedi-Allgower, AO, Topliss, and Leonetti systems by evaluating radiographs and computed tomography images on 2 occasions. The inter- and intraobserver agreements were calculated using the Fleiss kappa test. Interobserver reliability was good for AO types (A, B, and C) and Ruedi-Allgower (κ = 0.71 and 0.61, respectively), whereas the interobserver reliability was moderate for AO groups (A1, A2, A3, B1, B2, B3, C1, C2, and C3), Topliss families, Topliss subfamilies, Leonetti types, and Leonetti subtypes. Intraobserver reproducibility was excellent for the Ruedi-Allgower classification, AO types, and Topliss families and good for AO groups, Topliss subfamilies, and Leonetti types and subtypes. Ruedi-Allgower and AO classification systems are the most reliable among those currently used for pilon fractures, but with lower agreement at the AO group level. The use of Topliss and Leonetti classification systems is not recommended because of less favorable results.