Structural analysis of total knee replacements has led to improved material behavior and implant fixation. As implants have become more durable, knee replacement has been made available to younger patients. These patients often make greater functional demands that are not consistently met by current implant designs. Because functional performance is determined by the interplay of implant geometry and soft tissue mechanics, new biomechanical simulations that incorporate models of muscles and ligaments are needed to assist in the design of new knee replacements.

The goals of this dissertation were (1) to develop novel computer simulations in which implant motions are determined by both implant geometry and soft tissue forces;​ and (2) to demonstrate the utility of these simulations by addressing relevant questions related to implant motions and soft tissue function. Two simulations were developed to meet these goals. The first, a kinematic simulation, generated knee implant motions on the basis of implant geometry;​ these motions were subsequently applied to a musculoskeletal model of the knee joint to study how implant design and surgical errors affect ligament lengths. The second, a dynamic simulation, computed three-dimensional implant motions during a stepup task by integrating dynamic equations of motion subject to simulated muscle, ligament, and contact forces. Validation of simulations was accomplished through comparisons with experimental data collected from isolated implants, cadavers, and subjects with and without knee replacements.

The kinematic simulation demonstrated that a common surgical variation, posterior tilting of the tibial component in posterior cruciate-substituting knee replacement, alters the tibiofemoral interaction intended by its designers. The kinematic simulation also predicted that anterior placement of the femoral component may compromise mediolateral stability by inducing collateral ligament laxity. The dynamic simulation demonstrated that tibial conformity reduces the sliding motion between the tibial and femoral condyles, which has been shown to damage tibial inserts. Similar variation in the sagittal plane geometry of the femoral condyles had little effect on sliding motions. These studies clearly demonstrate the utility of using kinematic and dynamic musculoskeletal models to design and evaluate knee replacements.