•IN VIVO KINEMATICS OF MOBILE-BEARING KNEE ARTHROPLASTY IN DEEP KNEE BEND
+*Watanabe, T; *Yamazaki, T; *Sugamoto, K; *Tomita, T; *Yoshikawa, H
+*Osaka University Graduate School of Medicine, Suita, Osaka, JAPAN
INTRODUCTION
The aim of the current study was to analyze the kinematics during
deep knee bending motions of subjects with fully congruent designed
mobile-bearing total knee arthroplasties (TKA).
Fig.2: The AP translations of the medial and lateral femoral condyle
centers and the midpoint for them during deep knee bend in subjects
with DBK mobile-bearing TKA.
25
20
ext-60°flex
60-100°flex
100-120°flex
anterior (mm)
15
posterior
MATERIALS AND METHODS
Twelve subjects were implanted with Dual Bearing Knee (DBK, slot
type: Finsbury, UK) prostheses. This prosthesis has a mobile-bearing
insert that is fully congruent with the femoral component, which has
single-radius condyles, throughout flexion and allows axial rotation and
a 4-6 mm limited anterior/posterior (AP) translation on the polished
tibial tray. Posterior cruciate ligament (PCL) was retained and included
partial, subperiosteal release at its tibial insertion. All subjects
underwent a successful procedure resulting in over 100° of knee flexion.
Under fluoroscopic surveillance in the sagittal plane, each subject
was asked to do sequential deep knee bends under a loaded condition
from full extension to maximum flexion. The successive knee motion
was recorded as serial digital X-ray images (1024 x 1024 x 12
bits/pixels, 7.5 Hz serial spot images as a DICOM file) using a 12”
digital image intensifier system (C-vision PRO-T, Shimadzu, Japan) and
a 1.2-2.0 msec pulsed x-ray beam.
In vivo kinematic analyses of the knee prosthesis were analyzed using
a two-dimensional to three-dimensional (2D/3D) registration technique
[1] (Fig. 1). A computer assisted design (CAD) model of the femoral
and tibial components was used to reproduce the spatial posture of each
component from calibrated (including distortion correction) single view
fluoroscopic images. The root-mean-square errors of the estimated
relative pose between the two components were 0.5° rotation and 0.4
mm in-plane translation in our original validation. All rotations of
femoral components were expressed relative to individual femoral
component at 0° flexion in the tibial component coordinate system. To
evaluate the movement of medial and lateral femoral condyles, we used
the measured positions of the single-radius centers in femoral condyles
relative to the tibial tray.
10
5
0
-5
-10
-15
-20
-25
medial
lateral
Fig.3: Axial kinematic pathway of the bilateral femoral condyle centers
during knee flexion in subjects with DBK mobile-bearing TKA.
DISCUSSION
In the current study, the medial condyle exhibited greater anterior
translation, while the lateral condyle exhibited reduced posterior
translation compared with normal knees [2]. Increased anterior
translation of the medial condyle seems to result from reduced constraint
of mobile-bearing on the medial side. The lateral femoral condyle on the
mobile-bearing insert might be prevented from shifting backwards by
posterior lateral structures such as the popliteal tendon and posterior
capsule, contrasting with the lateral femoral condyle of normal knees,
which subluxates posteriorly from the tibial plateau in terminal flexion
[3]. Subjects with DBK mobile-bearing TKA in some degree reproduced
femoral external rotation during increasing knee flexion and bicondylar
posterior rollback during terminal flexion, due to surrounding soft tissue
structures. The geometry of replaced articular surfaces and mobility of
the mobile-bearing insert produced lateral-to-central pivoting motions
during the flexion cycle, a phenomenon not typically observed in normal
knees. Using the current technique, we characterized the unique
kinematics of fully congruent designed DBK mobile-bearing knee
prostheses.
REFERENCES
[1] Zuffi et al, IEEE Trans Med Imaging 18: 981-91, 1999
[2] Asano et al, Clin Orthop 388: 157-66, 2001
[3] Nakagawa et al, J Bone Joint Surg Br 82: 1199-200, 2000
ACKNOWLEDGEMENT
The authors would like to thank the Finsbury Orthopaedics Ltd.
(Surrey, UK) for providing the computer models for the prosthesis
components.
Fig. 1: 2D/3D registration overlaid upon sequential fluoroscopic images
during deep knee bend.
RESULTS
The average femoral component demonstrated 13.4° external axial
rotation for 0° to 120° flexion. On average, the medial condyle moved
anteriorly 6.2 mm for 0° to 100° flexion, then posteriorly 4.0 mm for
100° to 120° flexion (Fig. 2). On average, the lateral condyle moved
anteriorly 1.0 mm for 0° to 40° flexion, then posteriorly 8.7 mm for 40°
to 120° flexion (Fig. 2). The average subject experienced a lateral pivot
pattern from –5° to 60° flexion, a central pivot pattern from 60° to 100°
flexion, and a rollback pattern which bilateral condyles moved backward
from 100° to 120° of knee flexion (Fig. 3).
anterior (mm)
4
2
0
-2
-4
-6
-8
posterior
-10
-12
medial condyle center
-14
lateral condyle center
-16
midpoint of bilateral condyle centers
-18
-10
0
10
20
30
40
50
60
70
80
90
100
110
120
knee flexion angle (degree)
50th Annual Meeting of the Orthopaedic Research Society
Poster No: 1036