Classification of bone defects: An extension of the orthopaedic trauma association open fracture classification

Objectives: To develop a post-traumatic bone defect classification scheme and complete a preliminary assessment of its reliability. Design: Retrospective classification. Setting: Tertiary referral trauma center. Patients/Participants: Twenty open fractures with bone loss. Intervention: Assignment of a bone defect classification grade. Main Outcome Measurements: Open fractures were classified based on orthogonal radiographs, assessing the extent and local geometry of bone loss, including D1-incomplete defects, D2- minor/subcritical (complete) defects (,2 cm), and D3-segmental/ critical-sized defects (≥2 cm). Incomplete defects (D1) include D1A -,25% cortical loss, D1B-25%-75% cortical loss, and D1C-. 75% cortical loss. Minor/subcritical (complete) defects (,2 cm) (D2) include D2A-2 oblique ends allowing for possible overlap, D2B-one end oblique/one end transverse, and D2C-2 transverse ends. Segmental/critical-sized Defects (≥2 cm) include D3A-moderate defects, 2 to ,4 cm; D3B-major defects, 4 to ,8 cm; and D3C-massive defects, ≥8 cm. Reliability was assessed among 3 independent observers using Fleiss' kappa tests. Results: Interobserver reliability demonstrated the classification scheme has very good agreement, k = 0.8371, P , 0.0005. Intraobserver reliability was excellent, k = 1.000 (standard error 0.1478-0.1634), P , 0.00001. Interobserver reliability for the distinction between categories alone (D1, D2, or D3) was also excellent, k = 1.000 (standard error 0.1421-0.1679), P , 0.00001. Conclusions: This classification scheme provides a robust guide to bone defect assessment that can potentially facilitate selection of the most appropriate treatment strategy to optimize clinical outcomes.