Can MRI accurately detect pilon articular malreduction? A quantitative comparison between CT and 3T MRI bone models

Background - Pilon fracture reduction is a challenging surgery. Radiographs are commonly used to assess the quality of reduction, but are limited in revealing the remaining bone incongruities. The study aimed to develop a method in quantifying articular malreductions using 3D computed tomography (CT) and magnetic resonance imaging (MRI) models. Methods - CT and MRI data were acquired using three pairs of human cadaveric ankle specimens. Common tibial pilon fractures were simulated by performing osteotomies to the ankle specimens. Five of the created fractures three AO type-B (43-B1), and two AO type-C (43-C1) fractures were then reduced and stabilised using titanium implants, then rescanned. All datasets were reconstructed into CT and MRI models, and were analysed in regards to intra-articular steps and gaps, surface deviations, malrotations and maltranslations of the bone fragments. Results - Initial results reveal that type B fracture CT and MRI models differed by ~0.2 (step), ~0.18 (surface deviations), ~0.56° (rotation) and ~0.4 mm (translation). Type C fracture MRI models showed metal artefacts extending to the articular surface, thus unsuitable for analysis. Type C fracture CT models differed from their CT and MRI contralateral models by ~0.15 (surface deviation), ~1.63° (rotation) and ~0.4 mm (translation). Conclusions - Type B fracture MRI models were comparable to CT and may potentially be used for the postoperative assessment of articular reduction on a case-to-case basis