Low reproducibility of the intra-operative measurement of the transepicondylar axis during total knee replacement

Background The transepicondylar axis is often used for positioning of the femoral component in knee replacement. Methods We studied the reproducibility of the intra-operative palpation of the transepicondylar axis for rotational alignment of the femoral component in 20 total knee replacement (TKR) implantations with a non-image-based navigation system. 2 surgeons defined the transepicondylar axis 3 times each without changing the reference plane. Results The angle between the reference plane and the transepicondylar axis was measured by the navigation system. The mean intra-observer ranges of variation were 5degrees and 6degrees for both surgeons, with a maximum of 15degrees. The mean inter-observer range of variation was 9degrees, with a maximum of 15degrees. Variations occurred in either internal or external rotation. Intra-observer agreement was considered good for one surgeon and poor for the other. There was no agreement between the two observers. Interpretation To define the rotational alignment of the femoral component of a TKR according to the intra-operative palpation of the transepicondylar axis may not be as reproducible as expected. However, the exact effect of this variability on the outcome after TKR should be studied.