MATERIALS & METHOD

A geometrical model incorporated a hemi-pelvis and femur geometries of one individual with a THA (DePuy Pinnacle® acetabular cup with neutral and lipped liners; size 12 Corail® stem with 32mm diameter head) was created in SOLIDWORKS (Dassault Systèmes). Joint motions were taken from kinematic data of activities of daily living that were associated with dislocation of THA, such as stooping to pick an object off the floor and rolling over. The femoral component was positioned to conform within the geometry of the femur, and the acetabular component was orientated in a clinically acceptable position (45° inclination and 20° anteversion). Variation in orientation of the apex of the lip was investigated by rotating about the acetabular axes from the superior (0°) in increments of 45° (0°−315°), and compared to a neutral liner.

METHODS

A series of models of a 36mm BIOLOX^® Delta THR bearing (DePuy Synthes, Leeds, UK) were generated to match an experimental simulator study which included a mediolateral spring to cause lateral head separation due to a simulated mediolateral component misalignment of 4mm. A static, rigid, frictionless model was implemented in Python (PyEL, runtime: ∼1m), and results were compared against 1) a critically damped dynamic, rigid, FE model (runtime: ∼10h), 2) a critically damped dynamic, rigid, FE model with friction (µ = 0.05) (runtime: ∼10h), and 3) kinematic experimental test data from a hip simulator (ProSim EM13) under matching settings (runtime: ∼6h). Outputs recorded were the variation of mediolateral separation and force with time.

METHODS

Three-dimensional (3D) computed tomographic (CT) images of 56 subjects (28 males and 28 females) lying supine were obtained from a public image archive (Cancer Image Archive, website: www.cancerimagingarchive.net). 3D solid models of pelvis and spine were generated from the CT images. Two coordinate systems, pelvic and global body coordinate systems, were established. The pelvic coordinate system was established based on four bony landmarks on the pelvis: the bilateral anterior superior iliac spines (RASIS and LASIS) and the bilateral pubic tubercles (RPT and LPT). The global body coordinate system was generated based on the bony landmarks on the spine: the geometric centers of five lumbar vertebrae bodies and the most dorsal points of five corresponding spinous processes, as well as the anterior sacral promontory (Fig 1a and 1b). The acetabular rim plane was obtained by fitting a set of point along the acetabular rim to a plane using least squares method. The acetabular orientation was defined as the three coordinate components (x-, y- and z- components) of the unit normal vector of the acetabular rim plane in the two coordinate systems (Fig. 1c).

Methodology

The components of the THR were aligned and constrained as illustrated in Figure 1. CAD models of commercially available implant geometries were used (DePuy Synthes, Leeds, UK) modified for model simplicity by removing anti-rotation features.

The polyethylene cup liner was given elastic-plastic behaviour. An axial load following the Paul cycle pattern (5 repetitive cycles) with maximum of 3KN and swing phase load of 0.3KN, was applied through the cup holder. The effect of translational mismatch was implemented by using a spring element connected to the cup unit on the lateral side. The spring was compressed by a fixed amount to replicate a degree of medial-lateral mismatch of the components. The instantaneous resultant force vector dictated the dynamic sliding behaviour of the cup against the head. In this study, translational medial-lateral mismatch values of 1, 2, 3 and 4mm were used and the medial-lateral dynamic separation, contact pressure maps and plastic strain were recorded.