AMERICAN SOCIETY OF BIOMECHANICS
Presented at the Twenty-First Annual Meeting
of the American Society of Biomechanics
Clemson University, South Carolina
September 24-27, 1997
|
| AMERICAN SOCIETY OF BIOMECHANICS
Presented at the Twenty-First Annual Meeting |
This study examined three dimensional kinematics of cadaver knees implanted with fixed and mobile bearing total knee arthroplasties. The VICON motion analysis system was used in order to collect kinematic data before and after total knee arthroplasty. This information has important implications for total knee arthroplasty design and function.
Knee joint kinematics have been described through the use of the screw axis. 1, 2, 3, 4 Recently, investigators have studied the differences in fixed versus mobile bearing arthroplasties in terms of tibiofemoral translation and rotation. 5, 6 Stiehl et al. reported an anterior motion of the femur on the tibia during flexion also known as "paradoxical condylar translation" in meniscal bearing designs using video fluoroscopy.7, 8 However, there have been no reports in the literature describing total knee arthroplasty kinematics in terms of the screw axis. The purpose of this study was to examine the screw axis location in an intact knee compared to the same knee implanted with a fixed bearing design and an intact knee compared to a medial ball in socket mobile meniscal bearing design.
Two right-sided knee joint specimens were dissected to leave capsule and musculotendinous attachments. Marker carrying blocks were attached to the cortical bone of the femur and tibia. The knees were then scanned using computerized tomography (CT) in order to define marker and femoral contour geometry. The femur of each knee specimen was secured into a metal testing frame (0.75m x 0.75m x 0.88m) onto which three high speed VICON cameras were attached: one lateral, one anteriosuperior, and one anterioinferior to the specimen. A motor was attached to the quadriceps tendon to generate extension and flexion of the knee joint, and the medial and lateral hamstrings were loaded with symmetric and asymmetric loads during motion. Kinematic data were collected for 2 flexion and 2 extension trials under symmetric hamstring load, lateral hamstring load, and medial hamstring load.
Total knee arthroplasties were then implanted in each specimen using the measured resection technique. The first total knee was a cemented posterior cruciate ligament (PCL) substitute fixed bearing design, and the second a cemented PCL substitute medial ball in socket mobile bearing design. The femurs were then remounted into the testing frame and the collection process repeated for the gathering of kinematic data.
Using programs internal to the VICON system, the three-dimensional positions of each of the markers were calculated. VICON coordinates were transformed to CT coordinates by an orthogonal matrix obtained by nonlinear optimization in order to prevent distortion of CT rigid body shape. A method combining least squares and cubic splines created differentiable paths for each marker which allowed the calculation of velocity vectors. 9 The marker position vectors and velocity vectors in each frame were used to calculate the screw axes throughout the range of knee motion.
An outline of each intact distal femur was constructed by digitizing CT images. Piercing points of the screw axes were plotted across three sagittal femoral contours evenly spaced between the most lateral and most medial femur. A mean reference axis (MRA) was calculated by taking an average of all screw axes for each knee under each loading condition. Differences in the location of the MRA in the intact case compared to the total knee case were determined for each knee by calculating the distances between the MRAs across all planes. These results were then compared for each knee design. Paired t-tests were used to determine whether there was a significant difference in the MRA locations in the fixed versus mobile bearing knee (two-tailed t-test; p< .05).
The MRA showed a significantly greater displacement from the intact knee with the fixed bearing design compared to the mobile bearing design on the medial side of the joint under symmetric hamstring loading conditions (p = .033). The MRA showed a significantly greater displacement from the intact knee with the mobile bearing knee compared to the fixed design on the lateral side of the joint with medial hamstring loading conditions (p = .031). In the first case, the mean MRA displacement in the fixed design at the medial condyle was 2.01 cm, compared to 1.41 cm for the mobile bearing knee. In the second case, the MRA displacement in the mobile bearing design at the lateral condyle was 1.54 cm, compared to 1.08 cm for the fixed bearing knee. No significant differences in mean MRA displacements were found in either knee with lateral hamstring loading.
Analysis of the mean reference axis revealed a significant difference in displacement from the intact knee when a fixed knee design is used compared to a mobile meniscal bearing design in response to hamstring loading. With symmetric hamstring loading conditions, there was a greater change in the MRA displacement of the fixed design trials compared to the intact trials on the medial side of the joint. With medial hamstring loading, the greater change in the MRA occurred with the mobile bearing design. Unlike the fixed design, this change occurred on the lateral side of the joint. These findings suggest that a total knee fixed bearing design changes the kinematics of the medial knee under symmetric hamstring loading, whereas the mobile bearing design changes the kinematics of the lateral knee under medial hamstring loading. The mobile bearing knee used in this study was designed in an attempt to mirror "ball in socket" kinematics of the intact medial knee with a single axis of rotation. Thus the lesser displacement of the MRA on the medial side with the mobile bearing knee would be expected as the results of this study have demonstrated. The medial ball in socket mobile bearing design limits the "paradoxical translation" on the medial side of total knee arthroplasties as described by Stiehl, Markovich et al.7, 8
(1) Blacharski, P. et al. J. Biomech, 8, 375-384, 1975.
(2) Blankevoort, L. et al. J. Biomech, 21, 705-720, 1968.
(3) Van Dijk, R. et al. J. Biomech, 12, 727-731, 1979
(4) Mancinelli, C. Doctoral Dissertation, WVU, 1994.
(5) Nahass, B. et al. J. Biomech, 24, 1107-1117, 1991.
(6) Nilsson, K. et al. Ortho. Research Soc. 1997.
(7) Stiehl, J. et al. Ortho. Research Soc. 1997.
(8) Markovich, G. et al. Ortho. Research Soc. 1997.
(9) Simons, W. et al. J.Biomech. Eng,113, 348-351, 1991.