Magnetic Resonance Imaging of Patellofemoral Kinematics with Weight-Bearing

Background: Previous studies of the patellofemoral joint have been limited by the use of invasive techniques, measurements under non-weight-bearing conditions, cadaveric specimens, or computerized models. It has been shown that soft tissue and bone can be accurately quantified with magnetic resonance imaging. The present study was designed to define the relationship between the patellofemoral contact area and patellofemoral kinematics in vivo.

Methods: Ten subjects with clinically normal knee joints were scanned with high-resolution magnetic resonance imaging while they pushed a constant weight (133 N) on the foot-plate of a custom-designed load-bearing apparatus. Images were obtained at five positions of flexion between –10° and 60°. Three-dimensional reconstructions were used to measure the patellofemoral cartilage contact area, patellar centroid, patellar medial and inferior translation, patellar medial and inferior tilt, and patellar varus-valgus rotation. All translation and area measurements were normalized on the basis of the interepicondylar distance. Random-effects models of quadratic regressions were used to evaluate the data.

Results: The mean contact area ranged from 126 mm2 in extension to 560 mm2 at 60° of flexion. The patella translated inferiorly to a maximum distance of 34 mm at 60° of flexion and translated medially to a maximum distance of 3.2 mm at 30° of flexion before returning to nearly 0 mm at 60° of flexion. The patella tilted inferiorly to a mean of nearly 35° at 60° of flexion and medially to a maximum of 4.2° at 30° of flexion. By 60° of flexion, the centroid of the contact area had shifted to an inferior and posterior maximum of 20 and 10 mm, respectively.

Conclusions: We found that lateral patellar subluxation and tilt occurred in these normal knees at full extension and the patella was reduced into the trochlear groove at 30° of flexion. Therefore, we believe that lateral patellar tilt and subluxation observed during arthroscopy of the extended knee may not represent a pathological condition.

Clinical Relevance: This study may be useful for refining arthroplasty design and surgical technique.