Radiographic markers for measuring tibial rotation based on CT-reconstructed radiographs: an accuracy and feasibility study

Malreduction in the axial plane (malrotation) following tibial fracture surgery is often undiagnosed. A few clinical and radiographic methods have been proposed for measuring tibial rotation intraoperatively, yet have failed to match the accuracy of computed tomography (CT). The aim of this study was to develop radiographic tools for future intraoperative assessment of the tibial shaft rotation profile. The setting was a laboratory computerized analysis. Twenty lower limb CT scans were used to construct a three-dimensional (3D) model using AMIRA© software. A virtual 3D cylinder was implanted in the posterior condylar line and in the transmalleolar axis. The 3D models were used to simulate four standard knee and ankle plain radiographs. On each radiograph, four landmarks were depicted by two observers and their relation with the cylinder was measured and analyzed for accuracy and reproducibility. A cadaveric lower leg was implanted with two Kirschner wires. A CT scan was performed in addition to 2D fluoroscopy. The simulated radiographs and the fluoroscopy were compared for accuracy. Measurement of the landmarks showed reliability in most of the knee anteroposterior and ankle mortise radiographs (coefficients of variation