|
A Computational Tool for Predicting Wear in Total Knee Replacements
 HE DEVELOPMENT and validation of a computer aided engineering package that can predict the wear and damage to the
tibial bearing component in total knee replacements, will greatly enhance the development of more durable total knee
replacements and will contribute significantly to the development of other more durable prosthetic devices (in the U.S.
over ½ million joint replacement surgeries are performed annually with an average life of 10-15 years). In order to
realize this ambitious goal, significant advances in integrating fundamental aspects of tribological systems with
complex kinematics must be made.
This computational tool will be a great advance over the current state-of-the-art that
relies on lengthy and costly physical testing on one of the available knee simulators, which do not reliably reproduce
the damage that occurs in vivo. In the future, computer aided engineering techniques will make reliable wear predictions
quickly and economically. Ultimately, this will provide the opportunity to optimize designs for component life and
durability.
This project follows a fundamentally new approach for predicting wear in complex three-dimensional
components. We are building a computational tool to systematically make wear predictions from contact and kinematic data
at the surface; this approach is not directly tied to any particular technology. The approach is to discretize the
component geometry into a series of surface elements, and to apply constitute models developed from laboratory
experiments on an element-by-element basis to make predictions in wear and damage over the entire component surface.
This gives the system a tremendous amount of flexibility, and allows our research group as well as our collaborators to
investigate any three-dimensional components for which we have the contact and kinematic data, and the material models
for wear. The algorithms and the models for wear that we develop from our orthopaedic pin-on-disk tribometer are also
extendable to other applications. The research communities of computer aided engineering and tribology have been waiting
decades for the development of such a program.
In the future, with the use of computer aided engineering tools it will
be possible to optimize part designs for both performance and longevity prior to doing physical prototyping of the
design.
The current in vitro "gold standard" for evaluating the longevity of a particular total knee
replacement design is to run a 10 million cycle wear tests on one of the newest joint simulators, the AMTI (1997) or the
Stanmore (1998). Various types of knee simulators have been around for at least the last two decades, and none of them
has reproduced the wear and damage modes experience in vivo. The AMTI and the Stanmore simulators simultaneously test 4
or 6 specimens, and these experiments literally take months to complete (running continuously at 1 Hz just over 600
thousand cycles can be accumulated each week). These tests are also tremendously expensive with an average cost around
$40,000. The AMTI and Stanmore simulators run a "worst case" set of kinematics and loads created from data
collected by the gait laboratories. Because of the current speed of computer computations and the expense and relative
slowness of the existing experimental techniques, there is a tremendous opportunity for modeling to greatly accelerate
the evaluation, design, and development process of new implants.
|
