Background: Radiographic assessment of the patella after total knee arthroplasty is typically performed with use of static, unloaded views that may not reproduce the in vivo patellofemoral kinematics. The purpose of the present study was to evaluate and validate the reliability and reproducibility of a weight-bearing radiographic assessment of the patellofemoral joint in patients who have undergone total knee arthroplasty.

Methods: Radiographs were made for 100 knees in sixty-nine patients who had undergone total knee arthroplasty. Radiographic assessment of the patellofemoral joint was performed with use of both the standard Merchant axial view and a modification of that view. The Merchant axial view was modified by positioning the standing patient in the semi-squatted position with the knees in 45° of flexion. The relationship between the x-ray source, the angle of incidence on the joint, and the cassette position was kept unchanged from the original view. The standing position and consequent muscle involvement were the only differences.

Results: Compared with the standard Merchant axial view, the weight-bearing axial view showed a number of patellofemoral tracking changes. Specifically, lateral tilt and subluxation of the patella were significantly reduced; the rate of exposed, uncovered patellar bone contact with the femoral trochlea was significantly increased; and radiographic evidence of maltracking was more closely correlated with clinical symptoms.

Conclusions: An axial weight-bearing radiographic view with the patient in the semi-squatting position was developed to reproduce patellofemoral joint loading. This view demonstrates that the position of the patella, as seen on the standard unloaded Merchant view, changes during squatting. Utilization of this axial weight-bearing view to evaluate total knee arthroplasty may provide additional information over standard radiographic views.

Level of Evidence: Prognostic Level I. See Instructions to Authors for a complete description of levels of evidence.