Abstract
Introduction:
Following total knee arthroplasty, patients often complain of an unnatural feeling in their knee joint, which in turn limits their activities [Noble et al, CORR 2006]. To develop an implant design that recreates the motion of the natural knee, both the functional kinematics as well as the laxity of the joint need to be understood. In vitro testing that accurately quantifies the functional kinematics and laxity of the knee joint can facilitate development of implant designs that are more likely to result in a natural feeling, reconstructed knee. The objective of this study is to demonstrate that robotic in vitro testing can produce clinically relevant functional kinematics and joint laxities.
Methods:
All testing was performed using a KUKA (KUKA Robotics, Augsburg, Germany) 6 degree of freedom robotic arm and a six degree of freedom load cell (ATI Industrial Automation, Apex, North Carolina, USA), attached to the arm (Figure 1).
FUNCTIONAL KINEMATICS: Eight cadaveric specimens implanted with contemporary cruciate retaining implants were used for this evaluation. The functional activity, lunge, was simulated using kinematic control for flexion/extension and force-torque control for the other degrees of freedom. The inputs for the force-torque control were obtained from e-tibia data from live patients during the lunge activity [Varadarajan et al, J Biomech 2008]. At a given flexion angle, the robot moved in force-torque control to obtain the desired values within given tolerances (± 2.5N & ± 0.1 Nm). When these tolerances were met the position of femur with respect to the tibia was recorded and the knee flexed to the next level. The lunge simulation began at full extension and ended at 120 degrees of knee flexion, through 1 degree increments. The kinematic data from the contemporary CR implants were compared to in vivo kinematics of patients that were implanted with the same knee replacements performing a lunge activity [Varadarajan et al, Med Eng Phys 2009].
JOINT LAXITY: Eight native, unimplanted knees were used for this evaluation. Joint laxity of the knee joint was evaluated at 0, 30, 60, 90, and 120 degrees of knee flexion by applying various loads to the tibia and quantifying the resulting motion of the tibia. The resulting laxities were compared to various knee laxity studies in the literature.
Results:
The in vitro functional kinematics correlated well with the in vivo results. Femoral external rotation and tibial varus angulation were found not be statistically different between the in vitro and in vivo results (Figure 2). The laxities measurements correlated well with reported values in the literature.
Discussion:
In vitro robotic evaluations allow for a better understanding of the motion at the knee joint by simulating clinically relevant functional kinematics as well as quantifying joint laxities in the same testing system. Both of these metrics are needed to understand how the knee moves and should be used to evaluate the performance of new knee designs (Figure 3).
