IN VIVO STUDY OF THE PATELLAR KINETICS OF TOTAL HIP ARTHROPLASTY: DYNAMIC COMPUTED TOMOGRAPHY IN 0° TO 90° FLEXION

Abstract

Purpose: The position of the patella after implantation of a total knee arthroplasty is generally determined by static measurements on the femoropatellar 30° flexion view or on a computed tomography (CT) scan in full extension. We studied the kinetics of the patellar implant between 0° and 90° on dynamic CT scans to determine the influence of torsion of the femoral and tibial components on the patellar course.

Material and methods: Twenty patients with titanium total knee arthroplasties implanted in 1991 and 1992 underwent a dynamic CT study preoperatively and during the year following prosthesis implantation. An Imatron machine was used to obtain dynamic slices during knee flexion from 0° to 90°. The patient was installed in the prone position. Slice thickness was 8 mm for images centred on the lower end of the femur. Ten 50 ms images were obtained during flexion from 0° to 90°. The technique used preoperatively and postoperatively enabled study of transversal translation and tilt of the patella.

Results: The orientation of the patellar transversal bony axis remained closely parallel to the flexion axis of the knee (i.e. the epicondylar axis) both preoperatively and postoperatively, irrespective of the orientation of the femoral and tibial components. Between 0° and 90° flexion, the prosthetic patella exhibited a translation movement laterally to medially during the first degrees of flexion then medially to laterally during the last degrees of flexion. The transversal displacement of the patellar insert was less pronounced when the femoral component was placed in external rotation from the epicondylar axis. For femoral implants in internal rotation (mean 5°), the translation of the patellar insert was a mean 1.5 mm between extension and flexion. This translation was only a mean 1 cm for implants in external rotation (mean 5°). Contact between the patellar prosthetic component and the femoral prosthetic component were more harmonious when the femoral implant was placed in external rotation. Torsion of the tibial implant did not appear to have an impact on the transversal course of the patella in this series.

Discussion, conclusion: Rotation of the femoral component influenced the relationship between the patella and the femoral component in flexion-extension movements. However, torsion of the femoral component had little effect on the position of the patella itself. The patellar remained schematically parallel to the epicondylar axis, maintaining an orientation close to that observed preoperatively. The torsion of the femoral implant is the element that modifies its position under the patella and by consequence the relations between the patella and the femoral trochleae. These modifications are more pronounced when the knee is in extension than when the knee is in 90° flexion, excepting for the lift-off phenomenon observed at 90° flexion which is related to the trapezoidal femorotibial resection spaces.