Traction radiographs versus CT in the evaluation of fracture morphology and consecutive treatment decisions in OTA/AO 43C3 fractures

• Published in: Injury 2019-02, Vol.50 (2), p.332-336
• Main Authors: Misir, Abdulhamit, Kizkapan, Turan Bilge, Yildiz, Kadir Ilker, Uzun, Erdal, Ozcamdalli, Mustafa
• Format: Article
• Language: English
• Subjects: Comminution zones; Decision Support Techniques; Fracture fragments; Fractures, Comminuted - diagnostic imaging; Humans; Interobserver reliability; Intra-Articular Fractures - diagnostic imaging; Observer Variation; OTA/AO 43C3; Preoperative Care; Radiographic Image Interpretation, Computer-Assisted; Radiography - instrumentation; Reproducibility of Results; Retrospective Studies; Sensitivity and Specificity; Surgical approach; Tibial Fractures - diagnostic imaging; Tomography, X-Ray Computed; Traction; Traction radiograph; Treatment recommendation
• ISSN: 0020-1383; 1879-0267
• DOI: 10.1016/j.injury.2018.11.023

•CT imaging is superior in the correct identification of fracture fragments and comminution zones in OTA/AO 43C3 fractures.•However, both assessment methods result in similar treatment and surgical approach recommendations.•Hence, traction radiographs may be a useful alternative to CT imaging for preoperative planning in settings where CT is not feasible. Standard radiographs are limited in the evaluation of fracture characteristics and preoperative planning of OTA/AO 43C3 fractures. Therefore, CT imaging is an accepted as a useful method. CT is however expensive and has high radiation, and traction radiographs could be an alternative. This study aimed to compare fracture fragment and comminution zone visualization between traction radiographs and CT and any potentially resulting differences in consecutive treatment and surgical approach recommendations. Twenty orthopaedic surgeons assessed traction radiographs and CT images of 12 OTA/AO 43C3 type fractures. Each observer was required to identify the anterolateral, posterolateral, and medial malleolus fragments and the lateral, central, and medial shoulder comminution zones. They then had to recommend treatment (nonoperative, ORIF, closed reduction and external fixation, percutaneous screw fixation, or primary tibiotalar arthrodesis) with the best surgical approach (medial, anterolateral, posterolateral, posteromedial, or combined). Intra- and interobserver reliability, correct identification of fracture fragments and comminution zones on both images, and consistency of treatment recommendations and surgical approaches were analyzed. The agreement of each observer’s assessment of the presence or absence of specific fracture fragments and comminution zones was substantially increased for CT as compared to traction radiographs, particularly for the posterolateral (p = 0.000) and anterolateral fragment (p = 0.000), and the lateral (p = 0.000), central (p = 0.000), and medial shoulder comminution zone (p = 0.000). The interobserver reliability when assessing the three fracture fragments and comminution zones on the traction radiographs was moderate, whereas it was substantial when assessing these characteristics on CT. The medial malleolus fragment was more often correctly identified on traction radiographs than CT images (p = 0.001). The ability to correctly identify lateral, central, and medial shoulder comminution zones was higher for CT than traction radiographs (p = 0.000). The treatment and surgical appr