Estimation, Search, and Planning (ESP) Research Group

Introduction Isolated lateral compartment osteoarthritis (OA) affects ~10% of arthritic knees. These patients are eligible for partial knee replacement surgery using the Oxford Domed Lateral (ODL) implant. However, dislocation of the mobile bearing is the largest problem, affecting around 1 to 6 % of implanted patients. While dislocations can occur anteriorly, posteriorly or medially, the most common type of dislocation is medial dislocation, where the bearing moves up onto the wall of the tibial component and becomes lodged. One method of preventing medial dislocations is to insert screws into the tibial eminence, with their heads positioned above the tibial wall to augment the tibial wall height. In this study, we computationally assessed the likelihood of mobile bearing dislocation for 825 physiologically relevant configurations of the ODL femoral and tibial components. We compared the original tibial component design with designs where the tibial wall height had been nominally increased by 1 mm, 2 mm, 3 mm or 4 mm.

Methods The original design was tested along with three new designs where the tibial component wall height was uniformly increased. However, increasing the tibial component wall height, required the use of a suitably thicker bearing to avoid collision between the metal wall and the femoral component. Table 1 shows the wall height increase and the associated thicker bearing that were used for testing. The Computer Aided Design (CAD) 3D models were inserted into the robotics dislocation analysis tool to computationally assess whether medial dislocations of the mobile bearing were possible. For each tibial component design, the relative distance between the femoral and tibial components was increased: vertically (0–8 mm) and mediolaterally (ML) (0–6 mm) in 0.25 mm increments (825 configurations in total), starting with the bearing flush against the tibial component wall. The dislocation analysis tool used the Rapidly-Exploring Random Trees (RRT) algorithm to define the boundary of bearing motion allowable between the femoral and tibial components. The RRT was allowed to run for a maxium of 405 seconds and 25 searches. A bounding box was used to confine the search to medial dislocations only. A medial dislocation occurred when the bearing entered the “goal area”, defined by the anterior and posterior edges of the tibial component and medially by the midpoint of the tibial wall. When a medial dislocation was detected the Vertical Distraction to Dislocation (VDD) was recorded i.e. the gap between the femoral and tibial components.

Results Section For the normal tibial component (without tibial component wall height increase), the minimum VDD was 4.5 mm when the mediolateral translation was 2 mm. Increasing the tibial wall height by 1 mm (4 mm bearing), 2 mm (5 mm bearing) or 3 mm (6 mm bearing) increased the VDD to 5.5 mm, 6.75 mm and 7.75 mm respectively. Implanting the bearing flush against the wall (ML translation = 0 mm) increased the VDD to 5.5 mm for the original component and 6.5 mm, 7.5 mm and greater than 8 mm, respectively for the new designs. Increasing the mediolateral distance reduced the VDD: for instance, when the ML translation was 6 mm, the VDDs were reduced to 3.5 mm, 3.5 mm, 4 mm and 5.5 mm, respectively.

Discussion Mobile bearing dislocations tend to occur in flexion as in this position, the lateral ligaments are lax, enabling lateral compartment distraction. By nominally increasing the wall height, we are required to accordingly increase the thickness of the bearing, to ensure that the minimum gap between the top of the tibial wall, and the spherical femoral component, is at least 2 mm apart. This study used the dislocation analysis tool to assess the potential beneficial effect of nominally increasing the tibial component wall height on reducing the risk of mobile bearing dislocation throughout the knee flexion cycle. Tokuhara et al. conducted an MRI study of the lateral side of the knee in flexion and showed that the lateral compartment distracts on average 6.7±1.9 mm following application of a varus stress to a knee joint in 90° of knee flexion. This average amount of distraction is easily enough for a medial dislocation. This explains why medial dislocations are the most common clinically, and why anterior or posterior dislocations are clinically acceptable. In our study, increasing the wall height by 2 mm should sufficiently reduce the risk of mobile bearing dislocation medially to match the clinically acceptable risk of anterior or posterior dislocations. Increasing the tibial wall height beyond 2 mm would surpass the maximal distraction possible in the knee, and introduce additional risk of metal on metal collision. As increases in ML translation reduced the VDD, to minimise the risk of mobile bearing dislocation, surgeons are recommended to implant the bearing as close as possible to the tibial wall without hitting it.

Significance/Clinical Relevance Our study assessed 825 physiologically relevant configurations of the ODL femoral and tibial components. Raising the tibial component wall height (and using a thicker bearing) and implanting the bearing as close as possible to the tibial wall significantly reduced the risk of mobile bearing dislocation medially.