Axial radiography of the distal femur to assess rotational alignment in total knee arthroplasty

A method for taking an axial radiograph of the distal femur was developed to see the epicondyles and posterior condyles of the femur. It was hypothesized that these radiographs would be acceptable for evaluating rotational alignment in total knee arthroplasty with comparable reproducibility and good correlation to the results obtained with computed tomography images. Radiographs were obtained of 50 knees in 32 patients having total knee arthroplasty. The radiographs were taken while the patients were with the knee in 90 degrees flexion. The angle between the clinical epicondylar axis and the posterior condylar axis (twist angle) was measured and compared with the results obtained by conventional computed tomography. The interobserver variation in the axial radiography was less than or comparable to the computed tomography method. The mean discrepancy between the two methods (standard deviation) was 0.5 degrees +/- 0.4 degrees (range, 0-1.9 degrees), and a strong correlation was observed. This plain radiography is acceptable for evaluation of femoral component rotation with comparable reproducibility and correlation to the results with computed tomography. It has several advantages regarding cost, radiation dose, and lack of scatter when used for postoperative assessment. Level of Evidence: Diagnostic study, Level III-1 (study of nonconsecutive patients-no consistently applied reference "gold" standard).