Abstract
INTRODUCTION
Accurate knowledge of knee joint kinematics following total knee arthroplasty (TKA) is critical for evaluating the functional performance of specific implant designs. Biplane fluoroscopy is currently the most accurate method for measuring 3D knee joint kinematics in vivo during daily activities such as walking. However, the relatively small imaging field of these systems has limited measurement of knee kinematics to only a portion of the gait cycle. We developed a mobile biplane X-ray (MoBiX) fluoroscopy system that enables concurrent tracking and imaging of the knee joint for multiple cycles of overground gait. The primary aim of the present study was to measure 6-degree-of-freedom (6-DOF) knee joint kinematics for one complete cycle of overground walking. A secondary aim was to quantify the position of the knee joint centre of rotation (COR) in the transverse plane during TKA gait.
METHODS
Ten unilateral posterior-stabilised TKA patients (5 females, 5 males) were recruited to the study. Each subject walked over ground at their self-selected speed (0.93±0.12 m/s). The MoBiX imaging system tracked and recorded biplane X-ray images of the knee, from which tibiofemoral kinematics were calculated using an image processing and pose-estimation pipeline created in MATLAB. Mean 6-DOF tibiofemoral joint kinematics were plotted against the mean knee flexion angle for one complete cycle of overground walking. The joint COR in the transverse plane was calculated as the least squares intersection of the femoral flexion axis projected onto the tibial tray during the stance and swing phases. The femoral and tibial axes and 6-DOF kinematics were defined in accordance with the convention defined by Grood and Suntay in 1983.
RESULTS AND DISCUSSION
The offset in secondary joint motions at a given flexion angle was greater at larger knee flexion angles than at smaller flexion angles for abduction, anterior drawer, and lateral shift, whereas the opposite was true for external rotation. Significant variability was observed between subjects for the COR. The mean COR was on the lateral side during stance, consistent with results reported in the literature for the intact knee. Interestingly, the mean COR was on the medial side during swing.
CONCLUSIONS
Our results suggest that secondary joint motions in the TKA knee, specifically, external rotation, abduction, anterior drawer and lateral shift, are determined not only by implant geometry and ligament anatomy but also by external loading, and are therefore task-dependent. The mean COR in the transverse plane shifted from the lateral to the medial side of the knee as the leg transitioned from stance to swing. Mobile dynamic X-ray imaging is a valuable tool for evaluating the functional performance of knee implants during locomotion over ground.
