Cases of fractured mobile unicompartmental knee bearings have recently been reported. The purpose of this study was to understand the mechanics behind these fractures and to examine the influence of different design modifications. A parametric finite element model was used to examine the influence of different geometrical factors on the stresses within the bearing. Crack initiation occurred clinically in the centre of the bearing; this correlated with the position of the maximum von Mises stress. Tensile stresses, thought to propagate the fatigue crack, were maximal at the medial-lateral sides of the bearing, and the tensile vectors were normal to the fracture direction observed clinically. Fully congruent femoral articulation on the bearing, use of a thicker bearing size, and minimising wear of the component reduced the risk of fracture. For example, an unworn 6.5-mm-thick bearing (no clinical fractures reported) had 21.6% lower medial-lateral tensile stress compared to an unworn 3.5 mm bearing (five clinical fractures reported). In turn, an unworn 3.5 mm bearing had 34.3% lower tensile stress compared to a 3.5 mm bearing after 1.9 mm wear (average linear wear reported for clinically fractured bearings). The fracture risk was also reduced when the radio-opaque marker wire was positioned further from the centre of the bearing, and when marker balls were used instead of marker wires (19% reduction in tensile stress in some regions). These results indicate the importance of minimising component wear; the data also support the current component design which uses posterior marker balls instead of marker wires, and the continuing use of a congruous femoral component.
Keywords: Knee replacement; crack; finite element analysis; fracture mechanism; polyethylene; stress analysis.