Abstract
Introduction:
Varus alignment of the knee is common in patients undergoing unicondylar knee replacement. To measure the geometry and morphology of these knees is to know whether a single unicondylar knee implant design is suitable for all patients, i.e. for patients with varus deformity and those without. The aim of this study was to identify any significant differences between normal and varus knees that may influence unicondylar implant design for the latter group.
Methodology:
56 patients (31 varus, 25 normal) were evaluated through CT imaging. Images were segmented to create 3D models and aligned to a tri-spherical plane (centres of spheres fitted to the femoral head and the medial and lateral flexion facets). 30 key co-ordinates were recorded per specimen to define the important axes, angles and shapes (e.g. spheres to define flexion and extension facet surfaces) that describe the femoral condylar geometry using in-house software. The points were then projected in sagittal, coronal and transverse planes. Standardised distance and angular measurements were then carried out between the points and the differences between the morphology of normal and varus knee summarised. For the varus knee group, trends were investigated that could be related to the magnitude of varus deformity.
Results:
Several significant differences between normal and varus knees were found, but most of these were small differences unlikely to be clinically significant or have an influence on implant design. However, two strong trends were observed. Firstly, the version of the femoral neck was significantly less for patients with varus knees (mean difference 9°; p < 0.05).
The second trend was a significant difference in the sagittal morphology of the medial condyle. The kink angle, the angle formed by the intersection of the circles fitted to the flexion and extension facet surfaces, and their centres (Figure 1) was either absent or small in normal knees (mean 1°). An absent kink angle occurs when the circle defining the flexion facet surface lies within or makes a tangent to the circle defining the extension facet. However, for varus knees, the mean kink angle was 9°, with positive correlation with the angle of varus deformity (Figure 2).
Discussion:
Varus knees have a significantly larger kink angle than normal knees, influencing the relative positions of the flexion and extension facet spheres that define the medial condylar geometry, contributing to the commonly observed ‘flattening’ of the medial condyle in the sagittal plane. Varus knees are also associated with significantly less anteversion of the femoral neck. It has been shown that reduced femoral neck anteversion causes increased loading of the medial condyle [1], and our results support this finding. The data generated in this study will feed further biomechanical testing to investigate the influence of kink angle and femoral neck version on the kinematics and load distribution in the varus knee.
