Abstract
Introduction
Differing descriptions of patellar motion relative to the femur have resulted from many in-vitro and in-vivo studies. The aim of this study was to examine the tracking behaviour of the patella. We hypothesized that patellar kinematics would correlate to the trochlear geometry.
Method
Seven normal fresh-frozen knees were CT scanned and their kinematics with quadriceps loading was measured by an optical tracker system and calculated in relation to the previously-established femoral axes. CT scans were used to reliably define frames of reference for the femur, tibia and the patella. A novel trochlear axis was defined, between the centres of best-fit medial and lateral trochlear articular surfaces spheres.
Results
The path of the centre of the patella was circular and uniplanar (RMS error 0.3mm) above 16°±3° knee flexion. The distal end of the median ridge of the patella entered the groove at 6° knee flexion, and the midpoint at 22°. This circle was aligned 6°±2° (mean± SD) from the femoral anatomical axis, 91°±3° from the epicondylar axis, and 88°±3° from the trochlear axis, in the coronal plane. In the transverse plane it was 91°±3° and 88°±3° from the epicondylar and trochlear axes. Manipulation of the data using a transformation matrix to the previously reported anatomic axis of the femur showed an initial medial and then lateral translation.
Discussion
The motion path of the patella is dominantly circular in normal knees, when observed using a trochlea specific frame of reference. There is lateral translation in the last 16 degrees of extension, but in the rest of the range of motion, the patella motion path is on a single circular path with an RMS of less than 0.5mm. The two plane motion paths previously described can be replicated by a geometric transformation. Patella kinematics and patho-kinematics are more easily understood using this method.
