In vitro determination of the passive knee flexion axis: Effects of axis alignment on coupled tibiofemoral motions.

The natural passive flexion axis of human cadaveric knees was determined using a technique that minimized coupled tibiofemoral motions (translations and rotations), and the kinematic effects of mal-positioned flexion axes were determined. The femur was clamped in an apparatus that allowed unconstrained tibial motions as the knee was flexed from 0° to 90°. To establish the natural flexion axis, the femur's position was adjusted such that coupled tibiofemoral motions were minimized. Tests were repeated, first with the femur rotated internally and externally from its original position, and again after positioning the femur to flex the knee about the transepicondylar axis. Compared to the transepicondylar axis, flexion about the natural axis significantly reduced mean tibial translation by 66.4% (p < 0.01) and varus-valgus rotation by 70.1% (p <0.01). Mean varus-valgus rotation increased by 3.4° (factor of 4) when the femur was rotated 3° internally or externally from the optimum position. Differences in condylar location coordinates between the transepicondylar and natural flexion axes most likely indistinguishable clinically. Knee flexion about an axis that minimizes coupled tibiofemoral motions could be important for placement and orientation of a femoral total knee component and for specimen alignment during biomechanical knee testing.