Background: Distal radial fracture characterization with standard radiographs has been consistently poor, leading to a widespread use of computed tomography (CT) with this injury. Traction radiographs have recently been shown to improve interobserver reliability for fracture characterization and treatment. Our goal was to compare five traction radiographs and CT images to evaluate the interobserver and intraobserver reliability of individual fracture fragment identification, the correct identification of fracture fragments on CT imaging compared with traction images, and the consistency of treatment recommendations.
Methods: Eleven observers were asked to evaluate two blinded presentations of either traction images or CT images displaying seventeen different intra-articular distal radial fractures. Each observer was then asked to identify the presence or absence of six specific fracture fragments (radial column, dorsal wall, dorsal ulnar corner, volar ulnar corner, volar rim, and central impaction) and recommend treatment (nonoperative, open reduction and internal fixation, and external fixation or distraction plating). Analyses were conducted to evaluate the interobserver reliability of traction images and CT images for fracture fragment identification, the intraobserver variability of fracture fragment identification, the correct fracture fragment identification with traction radiographs compared with a gold standard CT scan, and the consistency in treatment selection.
Results: Interobserver reliability for traction images and CT images were both fair to poor. Intraobserver variability for fragment identification was similar for each fragment, without significance. Treatment recommendations based on traction radiographs agreed in 80.9% of the cases for open reduction and internal fixation and in 67.9% for external fixation compared with CT images.
Conclusions: Traction radiographic images are a suitable alternative to CT imaging for identifying and assessing distal radial fractures.
Level of Evidence: Diagnostic Level II. See Instructions for Authors for a complete description of levels of evidence.
1St. Luke’s University Hospital, PPHP-2, 801 Ostrum Street, Bethlehem, PA 18015. E-mail address for K.S. Matullo: Kristofer.Matullo@sluhn.org