Medial Proximal Femoral Angle better than Neck-Shaft Angle? Influence of Rotation on the Anteroposterior Radiograph

Abstract

AbstractNeck-shaft angle (NSA) and medial proximal femoral angle (MPFA) describe the geometry of the proximal femur, but they are influenced by rotation. The MPFA is closer to the right angle and therefore assumed to be less influenced. This hypothesis was tested with a mathematical approach to compare the effects of rotation and their clinical relevance on both angles. We developed a three-dimensional (3D) geometrical model representing projectional radiography of the femoral axes under rotation and derived a mathematical equation. For validation purposes, we experimentally simulated projectional radiography and compared the results to those of the equation (five different inclinations of the femoral neck (IFN) each in 13 different rotations). The validated equation was used to determine the influence of rotation (–60° to +60°) on normal valued NSA and MPFA. The mean absolute difference between calculated and experimentally measured values was 0.6° (standard deviation [SD] 0.5; 0.0–2.2). The mean calculated difference between true and projected NSA was 7.1° (SD 7.1; 0.0–22.0), and for the MPFA, it was –1.9° (SD 2.2; –7.4 to 0.0). Clinically relevant deviations (≥5°) occurred at ≥30° external or ≥ 40° internal rotation for the NSA, for the MPFA at 60° external rotation. Mathematical modeling of radiographic projection of the IFN is possible to a high level of accuracy. Our results indicate that the NSA is significantly more susceptible to deviations caused by rotation than the MPFA. Clinicians must be aware of rotational influences in radiographic projection and possibly should use the MPFA instead of the NSA. The rotational influence during X-ray imaging must be controlled accurately, and femoral torsion should be quantified in computed tomography/magnetic resonance imaging (CT/MRI) scans if possible. 3-D bone models reconstructed from digital image data can be a future option for measuring true angles. Level of evidence was basic research.