Introduction

Multiple metrics are used to evaluate knee function after total knee arthroplasty (TKA). One is the contact force imbalance between the medial and lateral compartments of the tibiofemoral joint. A second metric is the laxities of the tibiofemoral joint. Measuring contact forces in vitro is relatively fast, however measuring laxities in vitro is relatively slow. It is unknown how long the laxities in a human cadaveric knee can be measured before the soft tissues of the knee significantly degrade. Accordingly, the first aim of this research was to determine how many days laxities can be measured in a human cadaveric knee specimen with a TKA before clinically important changes occur.

The goal of kinematically aligned total knee arthroplasty (KA TKA) is to restore native alignments of the limb, knee, and joint lines with the intent of restoring knee function to native without ligament release. During KA TKA there are two steps to set the alignment of the femoral component. First, the two distal condyles are resected, setting the varus-valgus (V-V), proximaldistal (P-D), and flexion-extension (F-E) degrees of freedom. Second, the two posterior condyles are resected, setting the internal-external (I-E) and anterior-posterior (A-P) degrees of freedom. The medial-lateral (M-L) position is set visually. The femoral component is kinematically aligned when the thicknesses of the distal and posterior resections of the femoral condyles are equal to the thicknesses of the corresponding portions of the femoral component after compensating for cartilage wear and kerf of the saw blade. However, the use of manual cutting guides and oscillating saws can lead to errors in making these resections. Errors in these resections can cause femoral component malalignment. Two important errors are V-V and I-E malalignment. These errors can lead to condylar lift-off, pain, tibial component loosening, early implant failure, and need for revision surgery.

Additionally, it is also unknown whether changes in the laxities are correlated to changes in contact forces or contact force imbalance. This correlation is of interest to guide surgeons in assessing the alignment of the femoral component by checking laxities to determine whether tibial contact forces are balanced without direct measurement of these forces.

Accordingly, the second aim of this research study was to determine:​ (1) whether 2° or 4° of V-V malalignment of the femoral component in KA TKA causes clinically important changes in contact force imbalance and laxities, (2) what degree of V-V malalignment causes a clinically important change in contact force imbalance of 67 N, and (3) whether each of eight changes in laxities was correlated to changes in medial contact force, lateral contact force, and contact force imbalance due to V-V malalignment. The third aim of this research study was to determine:​ (1) whether 2° or 4° of I-E malalignment of the femoral component in KA TKA causes clinically important changes in contact force imbalance and laxities, and (2) whether each of eight changes in laxities was correlated to changes in medial contact force, lateral contact force, and contact force imbalance due to I-E malalignment.

Aim 1

KA TKA was performed on three human cadaveric knee specimens. The laxities and neutral positions in four degrees of freedom were measured for seven consecutive days using a six degree-of-freedom load application system. The earliest statistically significant and clinically important changes were found at the end of day 4, when the varus laxity (3.3° ± 1.3°) was significantly greater than the varus laxity at the beginning of day 1 (1.9° ± 0.5°, p < 0.01).

Aim 2

KA TKA was performed on ten human cadaveric knee specimens. V-V malalignment was introduced using 3D printed femoral components which simulated errors in the joint line of 2° varus rotation, 4° varus rotation, 2° valgus rotation, and 4° valgus rotation. Contact forces were measured using a custom tibial force sensor. Laxities were measured using a six degree-offreedom load application system. A simple linear regression related the degree of V-V malalignment to the change in contact force imbalance. The correlation coefficient was computed for each combination of the eight changes in laxities and the changes in medial contact force, lateral contact force, and contact force imbalance.

Statistically significant and clinically important changes in contact force imbalance occurred at 0° flexion with 2° varus (103 N), 4° varus (211 N), 2° valgus (89 N), and 4° valgus (178 N) malaligned femoral components (p < 0.0010). A nearly perfect straight line (R2 > 0.99) relating the change in contact force imbalance at 0° flexion to the degree of V-V malalignment indicated that a V-V malalignment of 1.4° causes a change in contact force imbalance of 67 N. No clinically important changes in the laxities were found with the 2° or 4° malaligned femoral components. None of eight changes in laxities were strongly correlated (R2 > 0.5) with changes in medial contact force, lateral contact force, or contact force imbalance.

Aim 3

KA TKA was performed on ten human cadaveric knee specimens. I-E malalignment was introduced using 3D printed femoral components which simulated errors in the joint line of 2° internal rotation, 4° internal rotation, 2° external rotation, and 4° external rotation. Contact forces were measured using a custom tibial force sensor. Laxities were measured using a six degree-offreedom load application system. The correlation coefficient was computed for each combination of the eight changes in laxities and the changes in medial contact force, lateral contact force, and contact force imbalance.

Statistically significant and clinically important changes in contact force imbalance occurred with the 4° internal component at 45° flexion (83 N), 60° flexion (110 N), 75° flexion (111 N), and 90° flexion (96 N) (p < 0.0001). A statistically significant and clinically important change in the valgus laxity of 1.5° was found with 4° external component (p < 0.0001). None of eight changes in laxities were strongly correlated (r > 0.71 or r < -0.71) with changes in medial contact force, lateral contact force, or contact force imbalance.

Conclusion

V-V and I-E malalignments as small as 1.4° cause clinically important changes in contact force imbalance. Therefore, surgeons should strive to keep femoral condyle resection errors of under 1.4°. To do this, the resection thicknesses, after correcting for cartilage wear and kerf of the saw blade, should be within approximately 0.5 mm of the thicknesses of the condyles of the femoral component. Additionally, surgeons should not rely on adjusting the laxities to affect predictable changes in the contact forces.