Landsforeningen for hjerte- og lungesyke

Superimposed three-dimensional (3D)-models obtained from CT-images have been used to evaluate displacement of femoral neck fractures, but this method assumes symmetrical anatomy of normal femurs. The present study aimed to compare the spatial orientation of the left and right proximal femur, thus establishing if 3D models can be used as a reference standard for the evaluation of fracture displacement. We generated 3D-CT-models of 20 patients with no skeletal pathology of the proximal femurs. Three observers independently determined the positions of the fovea and the femoral head, and a vector intersecting the centre points of the fovea and the femoral head defined the rotation. Differences in positions and rotations were determined by superimposing the 3D-CT-models of both femurs. The mean distance (95% CI) between positions of the left and right fovea was 3.1mm (2.7-3.4) and between the left and right femoral head 2.8mm (2.6-3.0). The minimal detectable change was 2.8 for the fovea and 2.3 for the femoral head, and the repeatability coefficients between 2.1-2.7 and 1.0-2.9, respectively. Mean difference in rotation of the femoral head was 6° (5.3-6.6) with a minimal detectable change of 8.8 and repeatability coefficients ranging from 5.8 to 10.0. Distances between the left and right femoral heads were larger than what could be explained by measurement error alone, suggesting that there may be minor side-to-side differences. However, these differences are small, and 3D-CT-models can be used as a reference standard to evaluate displacement of femoral neck fractures.