Pegg et al. (2015) described measurements of bearing and condyle movement fluoroscopically during active extension/flexion and during step-up after OUKA. These two exercises were chosen because the quadriceps force would be expected to be largest in extension during active extension/flexion and largest in flexion during step up.
Movements in activity were very different from those measured passively (Fig. 3.16). During active extension/flexion with the femur stationary and horizontal (Fig. 3.28), the bearing lay posterior to its position in passive extension/flexion up to about 30° and anterior to that position with further flexion. Figure A9 and the website’s Animation 9 show the model replaced knee during active extension/flexion. At full extension, the anteriorly directed model patellar tendon pulls the tibia anterior to the femur and stretches the ACL, allowing the model tibia to move anteriorly relative to the femur and the bearing to move backwards on the tibial plateau. As the knee flexes, the quadriceps and ACL forces diminish, reducing the difference between the active and passive position of the bearing. At larger flexion angles, the quadriceps force is further reduced and the patellar tendon is directed more posteriorly so that stretch of the PCL allows the bearing to move anterior to its position during passive movement, by about 4 mm in the fully flexed knee. The cross-over between these events occurred at about 30° in the patients (Fig. 3.28(a)) and at about 60°, according to the model calculations (Fig. 3.29(a)). The model assumed the presence only of agonistic quadriceps action and the differences between the bearing position during passive and active flexion/extension at larger flexion angles in the model was much smaller than in the patients. However, the patients could also have used antagonistic hamstrings action to control movements at the knee. In the flexed knee, the presence of hamstrings forces can greatly increase the level of force in the PCL (Toutoungi et al., 2000), resulting in the observed significant anterior movement of the condyle in flexion.
Figure 3.28 Bearing position plotted against flexion angle (a) during passive and active extension/flexion and (b) during active extension, active flexion and step-up.
Figure 3.29 (a) Calculated position of the model bearing on the tibia during passive and active flexion/extension. (b) Length of the most anterior fibre of the model ACL during passive and active flexion/extension. The reference length is the isometric length of the fibre during unloaded passive motion. (Reproduced from O’Connor J, Imran A. Bearing Movement after Oxford Unicompartmental Knee Arthroplasty: A Mathematical Model. Orthopedics|www.ORTHOSuperSite.com. 2007;30(5):42-45 (Supplement).)
The model calculations showed that the differences between the anterior translation of the tibia between passive motion and active extension are maximum at about 30° (Fig. 3.29(a)) where the extension of the ACL is also maximum (Fig. 3.29(b)).
There were significant differences between bearing position in the patients during active flexion/extension and during step up (Fig. 3.28(b)), particularly in mid-range. Whereas the bearing appeared to move forward and then backward as the knee was actively flexed, during step-up it appeared to remain essentially stationary on the tibia over most of the flexion range. The differences between the two activities doubtless arise from differing patterns of muscle activity and corresponding differences in the patterns of ligament stretch.
Gait analysis
Jefferson and Whittle (1989) assessed a group of medial OUKA patients in the gait laboratory. Seven parameters of their normal level walking gait (speed, cadence, stride length, sagittal plane and coronal plane angles, and sagittal plane and abduction moments) were compared with those of a group of age- and sex-matched volunteers with no locomotor problems. All seven parameters of the patients’ gait were restored to the normal range. This is probably the most convincing demonstration that the use of a ligament compatible unicompartmental prosthesis leads to a restoration of normal function.
Wiik et al. (2015) compared the downhill walking gait pattern in 19 UKA and 14 TKA patients who were well matched demographically and with high Oxford knee scores (OKS) for their operation type at a minimum 1 year after their operation. They also compared this data with 19 healthy young subjects used as controls. Downhill gait analysis was carried out on an instrumented treadmill that was ramped at the rear to produce a declination of 7°. All subjects after a period of habituation were tested for preferred and top downhill walking speed with associated ground reaction and temporo-spatial measurements. The UKA group walked downhill 15% faster than the TKA group (1.75 ± 0.14 vs 1.52 ± 0.13 m/s, p < 0.0001) despite having the same cadence (134.9 ± 8.0 vs 133.9 ± 9.6 steps/min). This 15% difference in speed appeared largely due to a 15% increase in stride length (173 ± 14 vs 150 ± 17 cm, p = 0.0007) and normal weight acceptance, both of which were similar to the controls.
Discussion
Since there is no single pattern of knee movement under load, it is difficult to ensure that comparisons before and after surgery are valid. The pattern of forces can only be adequately regulated to produce repeatable patterns of movement with very simple defined activities, such as straight leg raise, step-up, level walking etc.
The evidence of cadaver studies and the modelling is that the anatomical features of the joint (the shapes of its articular surfaces and the design of its ligaments) define the envelope within which the bones are moved by the very large extrinsic and intrinsic forces. These forces can either reinforce or reverse the pattern of contact in the unloaded joint, depending on how the ligaments are stretched in activity, implying that the detailed shapes of the articular surfaces allow rather than control the site of their contact areas. If the main function of the natural surface shapes is to maintain the ligaments at their appropriate tensions, any other pair of surfaces that can fulfil that one function might be able to restore normal movements if all the ligaments are intact.
The evidence of the PTA, bearing movement and gait laboratory studies is that a spherical femoral condyle articulating on a flat tibial plateau can replace the natural surfaces of the medial compartment. It is stressed that, in this regard, the meniscal bearing design of the OUKA implant is not different from fixed-bearing UKA implants, most of which employ a flat, or nearly flat, tibial plateau which allows the femoral condyle the same freedoms of anteroposterior translation as the OUKA enjoys. This is borne out by the studies of Parratte et al. (2012). We know of no evidence, and there is no theoretical reason, for the biomechanics and kinematics of the two designs of implant to differ as long as the tibial component of the fixed-bearing implant remains flat. However, if, because of the effects of wear, the flat form of the polyethylene becomes concave, the translational movements of the femoral condyle may become constrained (see Figs. 2.19 and 2.20). If stability can be restored to normal during OUKA with the Oxford instrumentation (Chapters 6 and 7 below), it is likely to remain so because of the minimal wear of the meniscal bearing.
Mathematical models
Explanations for many of the above observations have been deduced from analysis of the mathematical models of the knee. The two- and three-dimensional models are further discussed in the Appendix, and animations of the models are available at www.oxfordpartialknee.com.