Abstract
Introduction
Studies have shown that increased implant conformity in total knee arthroplasty (TKA) has been linked to increased constraint and thus rotational torque at the bone/implant interface. Anterior stabilized (AS) tibial inserts were designed to compensate for excessive AP motion in less-conforming cruciate-retaining (CR) tibial inserts. However, increased constraint may affect implant loading. Therefore, the purpose of this study is to model rotational prosthesis constraint based on implant-specific data and to compare rotational torque and 3D contact forces in implants with CR-lipped and AS tibial inserts during normal gait.
Methods
A previously reported knee joint contact model was updated to include rotational torque due to prosthesis constraint (ASTM F1223(14)). Piecewise multiple linear regression with manually selected cutoff points was used to determine estimates of AP force, ML force, and rotation torque as functions of AP displacement, ML displacement, knee external rotation, respectively, and knee flexion angle from standard data. These functions were used to estimate total moment contribution of the prosthesis from measured knee displacement/rotation angles. Estimates were incorporated into the contact model equilibrium equations as needed by the model. As the model parametrically varies muscle activation coefficients to solve for the range of physiologically possible forces at each time point, the reported force/torque values are the mean across all solutions at each time point. Rotational torque and three dimensional contact forces were calculated for 14 informed-consented subjects, five with AS tibial inserts (1/4 m/f, 67±10 years, 29.2±4.4 BMI, 1/4 right/left) and nine with CR-lipped TKRs (2/7 m/f, 64±6 years, 30.6±5.8 BMI, 4/5 right/left). Rotational torque waveforms were compared using statistical nonparametric mapping; 3D contact forces were compared at mean timing of the flexion/extension moment peaks using independent samples t-tests.
Results
Waveform analysis of rotational torque found no significant differences between implant types. CR- lipped inserts showed an initial peak internal rotation torque during weight acceptance and continued with a pattern of internal rotation throughout stance. Peak torque for AS inserts also occurred during weight acceptance, but it varied between internal/external rotation torque. Additionally, after weight acceptance, AS subjects showed a pattern of external rotation torque. Mean axial force, medial-lateral shear force, and anterior-posterior shear force waveforms were similarly shaped between implant groups. Flexion and extension moment peaks occurred at 23% and 74% stance on average. There were no significant differences in three-dimensional knee joint contact forces between groups at either time point.
Discussion
There were different patterns of rotational torque between groups. Implants with lipped CR inserts tended to undergo internal rotation torques that peaked during weight acceptance. Torque seen in implants with AS inserts was also largest during weight acceptance, but greatly varied between internal and external rotation, before settling in a pattern of external rotation for the remainder of stance. This may be due to constraints added by AS insert geometry. In conclusion, a model of rotational torque due to implant constraints has been developed; increased implant constraint increased the external rotation torque experienced by the implant and may also affect shear forces at the implant surface.
For any figures or tables, please contact authors directly.
