Introduction: The Bartoníček/Rammelt classification is established for posterior malleolar fractures. It subdivides the fractures into 5 types and outlines treatment recommendations. This study aims to determine the intraobserver and interobserver reliability of the Bartoníček/Rammelt classification and investigates its applicability regarding treatment recommendations.
Materials and methods: Computed tomography (CT) scans of 80 ankle fractures with a posterior malleolar fracture were analyzed by four observers at two different time points 30 days apart (d1 and d2). Intrarater and interrater reliability was measured using kappa values. The corresponding surgery rates of the fracture subtypes were analyzed, and the surgery rates were correlated with fragment sizes and displacements.
Results: A moderate interobserver reliability for d1 0.41 (CI 0.35-0.47) and d2 0.42 (CI 0.36-0.48) was detected. Intraobserver reliability was documented as perfect, with a mean kappa of 0.83. Type II fractures were operated on in 50% of cases. In 50% of type II cases, a nonoperative treatment was chosen. Fragment size correlated strongly with the chosen therapy, and osteosynthesis was performed significantly more often when the fragment size exceeded 3 cm3 (P < .01).
Conclusions: The Bartoníček/Rammelt classification system showed moderate interobserver reliability and perfect to substantial intraobserver reliability. In clinical practice of this study cohort, the size of the posterior malleolar fragment rather than the dislocation and joint impaction seemed to have the decision to operate on type II or III fractures. Existing treatment recommendations based on the Bartoníček/Rammelt classification correspond to the therapy algorithm carried out in this cohort of patients.Levels of Evidence: Level III: Retrospective study.
Keywords: Bartoníček/Rammelt classification; CT-scan; interobserver reliability; intraobserver reliability; posterior malleolar fracture.