The posterolateral upslope of a low-conforming insert blocks the medial pivot during a deep knee bend in TKA: a comparative analysis of two implants with different insert conformities.

PURPOSE: Tibial insert conformity in total knee arthroplasty (TKA) is of interest due to the potential effect on tibiofemoral kinematics. This study determined differences in anterior-posterior movements of the femoral condyles, pivot locations, and internal tibial rotation in different arcs of flexion for two implants with different insert conformities in kinematically aligned TKA. METHODS: Twenty-five patients treated with a medial and lateral low-conforming, posterior cruciate ligament (PCL) retaining (LC CR) implant followed by a medial ball-in-socket and flat, lateral PCL sacrificing (B-in-S CS) implant in the contralateral knee underwent single-plane fluoroscopy during a deep knee bend. Analysis following 3D-to-2D image registration determined tibiofemoral kinematics and patients completed validated outcome scores for both knees. RESULTS: The mean follow-up of 1.6 ± 0.4 years for the knee with the B-in-S CS implant was shorter than the 2.7 ± 1.2 years for the LC CR implant. From 0º to 30º of flexion, a medial pivot occurred with the tibia rotating internally approximately 5º with both implants. From 30º to 90º, the pivot remained medial and internal rotation increased to 10º with the B-in-S CS implant. In contrast, neither femoral condyle moved more than 1 mm with the LC CR implant from 30º to 60º, but from 60º to 90º degrees, a lateral pivot occurred and internal rotation increased. Internal rotation of the tibia on the femur from 0° to maximum flexion occurred about a medial pivot similar to the native knee for the B-in-S CS implant and was 4.5° greater than that of the LC CR implant (10.4° vs 5.9°). There was no difference in the median patient-reported outcome scores between implant designs. CONCLUSIONS: Tibial insert conformity is a primary determinant of a medial or lateral pivot during a deep knee bend. One explanation for the transition from a medial to lateral pivot between 30º and 60º with the LC CR implant is the chock-block effect of the insert's posterolateral upslope which impedes posterior movement of the lateral femoral condyle. Because there is no posterolateral upslope in the insert of the B-in-S CS implant, the tibia pivots medially throughout flexion similar to the native knee. LEVEL OF EVIDENCE: Level III.

Circle-based model to estimate error in using the lowest points to indicate locations of contact developed by the femoral condyles on the tibial insert in total knee arthroplasty.

A common method used to study tibiofemoral joint biomechanics following total knee arthroplasty (TKA) is the lowest point method, which finds the lowest points of each femoral condyle in relation to the plane of the resected tibia. The objectives of this paper were twofold: 1) to use a circle-based model to demonstrate the large inherent error introduced when the lowest points are used to indicate anterior-posterior (AP) positions of contact by the femur on the tibial insert, 2) to use the circle-based model to estimate the magnitude of error. A circle-based model was created to simulate articular surfaces of the tibial insert and condyles of the femoral component and to demonstrate the error. Equations relating the error to radii of tibial and femoral articular surfaces were derived. The magnitude of the error was estimated for common low-conforming TKA components by determining radii using best-fit circles to approximate curvature of articular surfaces. Error in AP tibial insert contact locations is caused by the slope of the tibial articular surface and the magnitude increases with increasing slope and increasing radius of the femoral condyle. For radii approximating articular surfaces of common low-conforming components, relative errors range from 45% to 109%. The circle-based model effectively demonstrates the cause of the large error in using lowest points to indicate AP tibial insert contact locations and enables an estimate of relative error. Because relative error exceeds 45%, the lowest point method should not be used to indicate the AP tibial insert contact locations.