Commentary Internal malrotation of the prosthetic components has been the single most common surgical error associated with the advent of contemporary total knee arthroplasties. On the femoral side, malrotation was primarily the result of posterior-referencing instrument systems and symmetrical components. On the tibial side, the reasons for internal malrotation have been multiple: the 85% predominance of varus knee deformity, the overwhelming incidence of the medial approach, the interference of the tibial tuberosity and patellar tendon with the traditional tibial cutting guides, and the innate desire of the physician to cover the entire tibial surface with a prosthetic device that is rotated to cover the deeper posteromedial bone. This study clearly indicates the preference of the Berger or Insall techniques, which involve using the medial third of the tibial tuberosity as the ideal osseous anatomic reference1. This landmark is very reproducible, accessible, and reliable. It is, however, only an osseous landmark, which ignores the influence of the previously positioned femoral component and the impact of soft-tissue balancing on the entire total knee procedure. For a computer or robotic arthroplasty enthusiast, numerical choices such as this one are attractive. They do, however, only reflect averages and thus may be expected to produce average results. While this narrows the Gaussian distribution of error, it does not account for the wide variation of anatomic configurations seen in clinical practice. Also, the overwhelming majority of prosthetic components are symmetrical and the natural knee is asymmetrical. While we think in our minds that we are reproducing the natural knee, we are merely placing a symmetrical component in a neutral “sweet spot” on 3 axes with 6 degrees of freedom. The fairly recent advent of asymmetrical tibial components has radically reduced the incidence of internal tibial rotation, although they must still conform, to some degree, to the axes identified in this computed tomography (CT) study. A general principle of all arthroplasties is that one should take advantage of both osseous anatomic and ligamentous soft-tissue input at every step of the procedure. Since most knee arthroplasties are now performed with the femoral component being inserted first—hopefully rotated to be parallel to the transepicondylar axis—the input of the femoral component and the supporting ligaments have a greater influence on the kinematic outcome than any single average osseous anatomic landmark2. A traditional approach to this issue was to first apply the trial femoral component to the distal part of the femur, then use a “floating trial” that would, after putting the knee through a range of motion, define the preferred position of tibial rotation relative to the femur and the surrounding ligaments. A knee is not a hip, and performing these component positionings in isolation is folly. If a floating trial indicates that the tibia would favor a rotational position aligned with the medial third of the tibial tuberosity, and the osseous and soft-tissue markers agree, then there should be no issue. If the floating trial position does not align over the medial third of the tibial tuberosity, then one must not only search for the reason but probably reach an accommodation. In my experience, the osseous position would dominate in a varus knee because of the greater influence of the topography of the surfaces. In a valgus knee, the ligaments often have a higher level of influence and should be preferred. Berger’s choice of the axis, which was confirmed by this CT study, has stood the test of time, over a quarter of a century. Asymmetrical components have corrected the surgeon’s natural impulse for internal rotation. The enunciation and confirmation of the importance of these choices is the major contribution of this article.
Read full abstract