Background:​ Kinematically aligned (KA) total knee arthroplasty (TKA) is a viable alternative to mechanically aligned (MA) TKA because randomized trials, meta-analyses, multicenter national, and matched cohort studies have shown that patients treated with KA TKA reported significantly better pain relief, function, flexion, and a more normal feeling knee than patients treated with MA TKA. KA TKA strives to co-align the axes and joint lines of the components with the three kinematic axes and joint lines of the native knee without placing restrictions on the preoperative deformity and postoperative correction, and without soft-tissue releases. Although KA results in biomechanical variables characterizing knee function in passive motion such as laxities and tibial contact forces which are not different generally from those of the native knee, the reported incidence of complications is infrequent, and implant survivorship was 97.5% at ten-year follow up, the in vivo kinematics of the tibiofemoral and patellofemoral joints are largely unknown.

Tibiofemoral and patellofemoral contact kinematics describe the relative motions and movement of the contact locations between the bones through flexion. Knowledge of contact kinematics provides an objective assessment of joint function and can be used to evaluate the performance of component designs and alignment philosophies in TKA and provide a means of comparison to the healthy native knee. In the tibiofemoral joint, anterior-posterior (AP) tibial contact locations of a TKA should not be identical to those of the native knee due to differences in curvature of the articular surfaces, but nevertheless differences should be limited. Further, contact on the posterior edge of the tibial insert should be avoided as this has been associated with uneven, accelerated wear of the insert. In the patellofemoral joint, complications such as anterior knee pain, subluxation, and extensor mechanism deficiency represent some of the primary sources of patient dissatisfaction and non-infectious indications of revision surgery following TKA. Biomechanical variables which provide objective measures of patellofemoral joint function are patellar flexion angles and proximal-distal patellar contact locations, which describe the rotation of the patella relative to the femur and location of contact on the patella by the femur, respectively, in the sagittal plane during flexion activities. The incidence of patellar loss of contact at full extension is another objective measure. Designing and surgically aligning TKA components to more closely restore tibiofemoral and patellofemoral kinematics to those of the native knee may prevent postoperative complications.

Hence, there were two objectives. The first objective determined differences in AP tibial contact locations of a KA TKA performed with anatomic (i.e. asymmetric) posterior cruciateretaining (CR) components from those of the native contralateral knee and also determined the incidence of posterior edge contact of the tibial insert during a deep knee bend and a step-up. The second objective determined whether the patellar flexion angle and proximal-distal patellar contact location were different from those of the native contralateral knee during a deep knee bend, and determined patellar loss of contact at full extension only for the KA TKA.

Methods:​ A review of one surgeon’s database of postoperative CT scans from November 2014 to April 2017 identified patients with a KA TKA performed with anatomic CR components and a native contralateral knee with no skeletal abnormalities in either limb except for the KA TKA. Patients were selected with no restriction on preoperative varus-valgus or flexion-contracture deformity. Twenty-five patients’ native and KA TKA knees were imaged with a fluoroscope at 15 frames per second while they performed a deep knee bend from full extension to maximum flexion, and again while they performed a step-up. For the deep knee bend both knees were imaged in an oblique view and in a lateral view to collect the images for the tibiofemoral and patellofemoral joint analyses, respectively. AP tibial contact locations and the occurrence of posterior edge contact of the tibial insert were determined with 3D model to 2D image registration for each patient. Patellar flexion angle, proximal-distal patellar contact location, and patellar loss of contact at full extension were determined with a series of measurements made directly on the 2D fluoroscopic images. Paired t-tests at each knee flexion angle determined the significance of the difference between the KA TKAs and the native knees.

Results:​ For the tibiofemoral joint, differences in mean AP tibial contact locations between the KA TKA and contralateral native knees were 2 – 8 mm in the medial and lateral compartments. The incidence of posterior edge contact of the tibial insert in the lateral compartment was 16% (4 of 25 patients). The incidence of posterior edge contact of the tibial insert in the medial compartment was 4% (1 of 25 patients).

For the patellofemoral joint, mean patellar flexion angles were not different between the KA TKAs and the native knees throughout the motion arc. The difference in the mean proximaldistal patellar contact locations was 4 mm when statistically significant. The incidence of patellar loss of contact in the KA TKAs at full extension was 8% (2 of 25 patients).

Discussion:​ For the tibiofemoral joint, KA TKA performed with anatomic CR components resulted in mean AP tibial contact locations which were relatively centered in the compartments and differed at most from those of the contralateral native knee by approximately 15% of the AP depth of a mid-sized tibial baseplate. The occurrence of posterior edge contact in some of the patients may be explained in part by the anatomical design of the tibial component which was internally rotated when aligned kinematically. However, all patients with posterior edge contact had high patient-reported outcome scores at the time of follow up.

For the patellofemoral joint, KA TKA performed with CR components restored patellar flexion angles to native and largely restored the proximal-distal patellar contact locations, which at most differed from native by approximately 10% of the mean proximal-distal patella length. In the KA TKA, patellar loss of contact was infrequent. These findings may explain the low incidence of patellofemoral complications in KA TKA.