Abstract
Introduction:
Kinematic studies are used to evaluate function and efficacy of various implant designs. Given the large variation between subjects, matched pairs are ideal when comparing competing designs. It is logical to deduce that both limbs in a subject will behave identically during a given motion [1], barring unilateral underlying pathology, thus allowing for the most direct comparison of two designs. It is our goal to determine if this is a valid assumption by assessing whether or not there are significant differences present in the kinematics of left and right knees from the same subject. Gait studies have compared pre-and postoperative implantation kinematics for various pathologies like ACL rupture [2] and osteoarthritis [3, 4]. We designed a study to assess squatting in cadaver specimens.
Methods:
Sixteen matched pairs of fresh-frozen cadavers, (Eleven males, five females; aged 71 years [± 10 yrs]) were tested. Each knee, intact, was tested by mounting it on a dynamic, quadriceps-driven, closed-kinetic-chain Oxford knee rig (OKR), which simulated a deep knee bend from full extension to 120° flexion. We chose femoral rollback, tibiofemoral external rotation, tibial adduction, patellofemoral tilt and shift as our outcomes, which were recorded using an active infrared tracking system.
Results:
The maximum difference over 120° of flexion between the mean values of tibial adduction, tibiofemoral external rotation and patellofemoral tilt were 1.48 (Fig. 1), 0.94 and 1.64 degrees, respectively. The maximum differences between means for femoral rollback and patellofemoral lateral shift were 1.19 (Fig. 2) and 0.55 millimeters, respectively. Using the left knees as a baseline, the cumulative kinematic difference of each of our five outcomes was also calculated by summing the area under the curve of the absolute difference between right and left knees. The summed differences for each outcome were then broken down into four flexion ranges (Fig. 3).
Discussion:
Being able to utilize direct comparisons and control variables during the design and testing of an implant is an important aspect in comparing one iteration or feature to another. Using paired knees from the same subject can be done to remove a level of variability and to provide a more one-to-one comparison. In our study we found that while the kinematics remain very similar, there is a trend of divergence between the knees as they go into deeper flexion. This trend was observed in each of our five outcomes, though no statistical significance could be determined. We also observed that the standard error of the kinematic means was equal to, or greater than the difference between means, indicating that subject to subject variation was greater than left to right variation. This finding potentially gives further weight to the validity of bilateral comparisons.
