[Epiphyseal distal torsion of the femur in osteoarthritic knees. A computed tomography study of 75 knees with medial arthrosis].

PURPOSE OF THE STUDY: We used computed tomography (CT) to measure external torsion of the distal femur preoperatively in patients with osteoarthritic knees presenting genu varum in order to assess the correlation between femoral distal external torsion (FDET) and radiological or epidemiological data obtained during the general preoperative work-up.

MATERIAL AND METHODS: Seventy-five knees were studied in 38 patients. Mean age was 70.4 years; there were 13 men and 25 women. The degree of external torsion of the distal femur was defined as the angle measured on the CT-scan between a line drawn tangent to the most posterior part of the condyles and the epidondylar line drawn from the lateral epicondyle to the most prominent point of the medial epicondyle. This angle was compared with the angle of the epiphyseal tibial varum measured on the anteroposterior radiograph and the overall deviation of the lower limb (HKA), and with the angle between the mechanical axis and the femoral shaft axis (HKS). We assessed the effect of gender and sex and looked for correlations between the FDET angle and measurements made on standard radiographs. Student's t test was used compare the FDET angle by sex and side. The alpha risk was set at 5%.

RESULTS: The FDET angle measured a mean 5.36 +/- 1.87 degrees (0-9 degrees ). We did not find any correlation between the FDET angle and tibial epiphyseal varum or HKS angles. A weak statistical correlation between the FDET and the HKA angles had no clinical value. Finally, the FDET angle was not correlated with sex or side.

DISCUSSION: Computed tomography provides reliable and reproducible measurements. Our work clearly demonstrated the wide interindividual variability of the FDET angle. We were unable to identify any measurement on standard radiographs allowing an indirect prediction of the FDET angle.

CONCLUSION: Our findings suggest that measuring the FDET angle on preoperative CT-scans in candidates for total hip arthroplasty can be useful for adapting the angle of rotation for the femoral component to each patient.