Effect of three-dimensional computed tomography reconstructions on reliability.

BACKGROUND: This study evaluated the impact of three-dimensional (3D) volume-rendering computed tomography (CT) reconstructions on the inter- and intraobserver reliability of six commonly used classification systems in the assessment of calcaneal fractures.

METHODS: Four independent observers with different levels of clinical training evaluated 64 fractures according to the classifications of the Orthopaedic Trauma Association (OTA), Essex-Lopresti, Sanders, Crosby, Zwipp, and Regazzoni, using two-dimensional (2D) CT scans with multiplanar reconstructions and 3D volume-rendering reconstructions.

RESULTS: Interobserver reliability was moderate for the OTA, Essex-Lopresti, Sanders, Crosby, and Regazzoni classifications with 2D CT scans and 3D CT reconstructions. The Zwipp classification was poor with 2D CT scans and improved to moderate with 3D reconstructions. Intraobserver reliability with 2D CT scans was good for the Essex-Lopresti classification and moderate for the OTA, Sanders, Crosby, Zwipp, and Regazzoni classifications. After the addition of 3D reconstructions, all classifications showed moderate intraobserver reliability.

CONCLUSION: According to the findings of this study, the additional use of 3D reconstructions is of minor value when used in conjunction with the classifications of the OTA, Sanders, Crosby, Regazzoni, and Essex-Lopresti. If calcaneal fractures are assessed with the Zwipp classification, 3D reconstructions could be used to achieve comparable reproducibility compared to other classifications.

CLINICAL RELEVANCE: 3D reconstructions may have other benefits not evaluated in the presented study and may give useful information not captured by current classification systems.