Influence of lower limb rotation in navigated alignment analysis: implications for high tibial osteotomies

Inaccurate coronal plane realignment is a common problem after high tibial osteotomy. It has been shown that lower limb rotation has an effect on the two-dimensional measurement of lower limb alignment. Although alignment errors are known to occur due to limb rotation, the magnitude of this effect is unknown. Navigation systems allow for the measurements of coronal plane alignment and dynamically rotational and sagittal plane. Our study evaluated the effect of rotational leg movements on coronal plane alignment as determined by image-free navigation. We hypothesized that a linear relationship exists between rotation and angular measurements. Eight cadavers were used, while three test conditions of the complete lower limbs were established: (1) solid knee arthrodesis, (2) provisional knee arthrodesis and (3) unconstrained knee conditions. Navigated measurements of coronal and sagittal lower limb axis were done initially without knee flexion for defined internal/external rotations of 5 degrees, 10 degrees and maximal values for all test series. Repeated test for the unconstrained knee included stepwise knee flexion of 5 degrees, 10 degrees and 20 degrees. Statistical analysis comparing the test conditions 1, 2 and 3 and comparison between flexion movements of the unconstrained knee were done. Results revealed no significant differences between the different rotations of test condition 1 (mean 0.34 degrees, SD 0.23, range, 0 degrees-0.8 degrees). Condition 2 similarly did not result in significant deviations (mean 0.51 degrees, SD 0.24, range 0.1 degrees-0.9 degrees). Measurement deviations ranging from 0.4 degrees to 4.3 degrees were found for condition 3, the unconstrained knee. However, no statistically different testings from the arthrodesed knee were found (P=0.099-0.410). Knee flexion from 5 degrees, 10 degrees or 20 degrees, showed significant deviations (P<0.05) for all rotations at all degrees of flexion. Rotation and flexion of 5 degrees led to significant alignment errors of 3.4 degrees and 2.8 degrees, respectively, for internal and external rotations. Measurement failures due to the rotational movements of 1 degree-4 degrees might add to additional sources of errors causing relevant under- or over-corrections of the mechanical leg axis. Discrepancies of the axis due to rotational movements as well as flexion of the knee joint can be avoided and corrected immediately with the help of navigation.