Component alignment and soft tissue constraints are key factors affecting function and implant survival after total knee replacement (TKR). Knee kinematics contribute to knee function whilst soft tissue constraints and component alignment impact polyethylene wear. This study experimentally investigated the effect of soft tissue constraints and component alignment on the kinematics and wear of a TKR.

A six station electromechanical ProSim knee simulator was used with the ISO 14243-1:2009 standard force control inputs; axial force, flexion-extension (FE), tibial rotation (TR) torque and anterior-posterior (AP) force. This allowed the kinematics to vary with the test conditions. The soft tissue constraints were simulated using virtual springs.

DePuy Sigma XLK fixed bearing TKRs were tested in 25% bovine serum (in 0.04% sodium azide) lubricant. The average output kinematics across 6 stations were found for each test and the peak values compared. The wear rates were calculated over 2 million cycles (MC), the serum was changed every 350,000 cycles and the tibial inserts weighed after every MC. A one way ANOVA and post hoc Tukey's test was used to compare the kinematics and wear with significance taken at p<0.05.

The kinematics and wear rates for three soft tissue conditions were established under ideal alignment (Table 1). The ISO standard springs for a cruciate substituting (CS) and a cruciate retaining (CR) prosthesis were used to represent a knee with a resected ACL and PCL and a knee with a resected ACL respectively. The third spring condition was based on clinical data to represent a “stiff” knee.

Three other alignment conditions were then assessed using “stiff” knee springs; 4° varus, 14° rotational mismatch and 10° posterior tibial slope. These alignments were chosen to represent the range found in clinical data.

Under ideal alignment the “stiff” knee springs had significantly lower peak AP and TR displacements (0.9mm, 2mm, 2mm and 3.6°, 7.1°, 7.8° for the “stiff”, CR and CS springs respectively) than the other springs (p<0.01). The “stiff” knee spring had a significantly lower wear rate than the CR spring; 1.58 ±1.20mm³/MC compared to 4.71±1.29 mm³/MC (p<0.01).

The varus and rotated components had significantly larger peak AP displacements of 2.56mm and 2.42mm respectively, than the ideal and tibial slope fixtures (1.97mm and 0.92mm respectively) (p<0.01). The rotated components had significantly higher internal rotation of 12.2° compared to 4.4°, 3.7° and 3.5° for the tibial slope, varus and ideal components respectively (p<0.01).

The ideal and varus components had significantly lower wear than the tibial slope and rotated components (1.58±1.20mm³/MC and 0.15±0.83mm³/MC compared to 8.24±7.72mm³/MC and 5.19±1.12mm³/MC respectively) (p<0.01). This may be due to increased AP and TR displacements with the rotated components and the increased anterior AP displacement with the tibial slope components, resulting in wear on the posterior edge of the tibial insert.

Soft tissue constraints and component alignment had a significant effect on the kinematics and wear. Experimental simulation should test a variety of soft tissue and alignment conditions to reflect the range observed clinically and determine causes for early failure.

For any figures or tables, please contact the authors directly.

Variation in soft tissue constraints influence the kinematics and wear of total knee replacements (TKRs). The aim of this study was to experimentally investigate the effect of variation in the soft tissue constraints on the output kinematics of a fixed bearing TKR with different insert geometries. The kinematics have been shown to affect the wear rate of TKRs; increased output displacements may result in an increased wear rate. The soft tissue constraints were simulated experimentally using virtual springs.

A new generation six station electromechanical ProSim knee simulator was used with the ISO 14243–1:2009 standard force control inputs; axial force, flexion-extension (FE), tibial rotation (TR) torque and anterior-posterior (AP) force. This allowed the kinematics to vary due to the test conditions. The ISO standard spring tensions of 44N/mm and 0.36Nm/° and gaps of 2.5mm and 6° were used for the AP and TR springs respectively.

Different combinations of the input profiles were run in order to test the effect of their absence. The spring gaps were varied between 0mm–3mm and 0°–6° and the tensions between 0N/mm–250N/mm and 0Nm/°–1Nm/° for the AP and TR respectively. Three tibial insert designs were tested; high conformity curved (CVD), partially lipped (PLI) and flat.

DePuy PFC Sigma fixed bearing components were tested in 25% bovine serum (in 0.04% sodium azide) lubricant. For each test 100 cycles were recorded on each station and then averaged. The CVD insert was used for all tests, the PLI insert was also used to test the effect of spring tension.

The TR and AP output displacement profiles were affected by the FE position along with the TR torque and AP force respectively. The absence of these inputs changed the shape of the output profiles significantly. The spring gaps affected the peak AP and TR displacements (6.4mm to 3.7mm and 8° to 5.8° for maximum and zero spring gaps respectively). The spring tensions had a higher effect on the peak AP than TR position due to the design of the CVD insert restricting the TR movement (8.3mm to 3.7mm and 8.8° to 7.4° for no springs and maximum tension respectively). The spring gaps and tensions affected the amplitudes of the output profiles not their shape.

The lower conformity inserts had a higher peak TR position (23° for the flat and 8.1° for the CVD insert) which occurred earlier in the cycle. The flat insert resulted in more anterior displacement, potentially due to the high conformity on the anterior side of the CVD and PLI inserts. The spring tension test had an increased effect on the PLI than the CVD insert. The PLI insert resulted in a higher change in displacements due to the spring tensions (10.4mm to 3.5mm and 13.6° to 8.8°).

Soft tissue constraints and insert design had a significant effect on the kinematic outputs. Spring tensions and gaps for experimental testing should be chosen to reflect those of a specific patient group.