Cementless tibial trays commonly fail through failure of fixation due to excessive interface motion. However, the specific combination of axial and shear forces precipitating implant failure is unknown. This has led to generic loading profiles approximating walking to perform pre-clinical assessment of new designs, even though telemetric data demonstrates that much larger forces and moments are generated during other functional activities. This study was undertaken to test the hypotheses: (i) interface motion of cementless tibial trays varies as a function of specific activities, and (ii) the response of the cementless tibial interface to walking loading is not representative of other functional activities. Six fresh-frozen cadaveric tibias were tested using a custom designed functional activity simulator after implantation of a posterior stabilized total knee replacement (NexGen LPS, Zimmer, Warsaw IN). Activity scenarios were selected using force (Fx, Fy, Fz) and moment (Mx, My, Mz) data from patients with instrumented tibial trays (E-tibia) published by Bergmann et al. A pattern of black and white spray paint was applied to the surface of the specimen including the tibial tray and bone. Each specimen was preconditioned through application of a vertical load of 1050N for 500 cycles of flexion-extension from 5–100°. Following preconditioning, each tibia was loaded using e-tibia values of forces and moments for walking, stair-descent, and sit-to-stand activities. The differential motion of the tibial tray and the adjacent bony surface was monitored using digital image correlation (DIC) (resolution: 1–2 microns in plane; 3–4 microns out-of-plane). Four pairs of stereo-images of the tray and tibial bone were prepared at sites around the circumference of the construct in both the loaded and unloaded conditions: (i) before and after pre-conditioning and (ii) before and after the 6 functional loading profiles. The images were processed to provide circumferential measurements of interface motion during loading. Differences in micromotion and migration were evaluated statistically using step-wise multivariate regression.INTRODUCTION
MATERIALS and METHODS
