Implant failure due to massive polyethylene (PE) wear and wear-associated aseptic loosening has been one of the main challenges concerning total knee replacements (TKRs) in the past decades [
Clinically, failure analysis of currently available TKR confirms a reduction in wear-related revisions [
Aetiology of instabilities is often multifactorial, but a relevant portion can be attributed to ligamentous insufficiency [
Wear testing should be carried out based on the clinical background of expected loading. Patient collectives designated for the use of a high-conforming knee design differ from patient collectives designated for the use of an unconstrained knee design. The use of a high-conforming knee design may be plausible if a ligamentous insufficiency already exists or is anticipated during one’s lifetime. This should be considered during wear testing.
The aim of this study is to analyze the effects of a ligamentous insufficiency on the stability and the wear behavior of a high-conforming knee design.
Two knee wear tests were performed on an AMTI knee simulator (Model KS2-6-1000, Advanced Mechanical Technology Inc., Watertown, MA, USA) using two different restraint characteristics. Restraint characteristics are defined by the restraint of the passive structures (ligaments, soft tissue, and capsule) which are based on
Anterior/posterior restraint in this study based on cadaveric studies [
Internal/external restraint in this study based on cadaveric studies [
Disregarding restraint characteristics, wear tests were run with force-controlled parameters according to ISO 14243-1:2009 with an extension/flexion of 0°–58°, a maximum axial load of 2600 N, anterior/posterior forces of −265 to 110 N, and internal/external torques of −1 to 6 Nm. Axial forces were transmitted with a 7% medial offset of the tibial plateau width in order to achieve physiologically higher forces on the medial plateau.
For wear testing, a deep-dished, ultracongruent (manufacturer specification), cruciate-substituting implant design (TC-Plus, Smith & Nephew, Baar, Switzerland) was used. PE-components were irradiated in an inert gas atmosphere (25–37 kGy). The inserts were presoaked in bovine serum prior to the simulation. Inserts were gravimetrically measured on a weekly basis until the incremental increase in weight was less than 10% of the total cumulative weight increase. In detail, components were presoaked for 105 days (stable conditions) and for 132 days (unstable conditions). Every wear test consisted of three specimens plus one axially loaded soak control. Tests were run for a total of 5 million cycles in diluted bovine serum (PAA Laboratories GmbH, Pasching, Austria) with a protein content of 20 g/L. The testing fluid (250 mL) was tempered to 37°C during the simulation. As additives, sodium azide (1.85 g/L) and ethylenediamine tetra-acetic acid (7.44 g/L) were used to prevent bacterial growth and to minimize calcium phosphate layers, respectively.
At intervals of 500,000 cycles, the wear testing was interrupted to replace the bovine serum and determine the PE wear mass. Components were cleaned and measured gravimetrically according to ISO 14243:2:2009. At the end of each test, wear particles were analyzed using acid digestion according to previously published methods [
Wear rates and kinematics were compared using Student’s
Wear particle characteristics are based on a high number of wear particles. Effect size was calculated according to Cohen [
Wear progression of both tests is shown in Figure
Wear progression for ligamentous-stable and ligamentous-unstable test conditions.
Considerably higher kinematics was observed for the unstable knee compared to the stable knee (Figures
Tibial anterior and posterior translation for ligamentous-stable and ligamentous-unstable test conditions (dashed line = standard deviation).
Tibial internal and external translation for ligamentous-stable and ligamentous-unstable test conditions (dashed line = standard deviation).
Wear areas on the PE for ligamentous-stable (a) and for ligamentous-unstable knee test conditions (b) (right knee in both figures).
Results of wear particle analysis are shown in Table
Results of wear particle analysis.
Unstable knee | Stable knee | Effect size | |
---|---|---|---|
Particles analysed | 2016 | 1510 | |
Estimated number of particles per 106 cycles | 1.09 |
0.80 |
2.23 |
Equivalent circle diameter | 0.263 ± 0.160 |
0.246 ± 0.162 |
0.11 |
Aspect ratio | 1.776 ± 0.584 | 1.700 ± 0.504 | 0.14 |
Roundness | 0.548 ± 0.151 | 0.577 ± 0.143 | 0.20 |
Form factor | 0.657 ± 0.137 | 0.687 ± 0.120 | 0.23 |
Example of analysed wear particles under ligamentous-stable (a) and for ligamentous-unstable knee test conditions (b). Relevant differences were observed in particular regarding the number of released wear particles.
In this study, the stabilization and wear behavior of a high-conforming knee design with two different ligament settings were simulated. Simulation of ligamentous-unstable TKR resulted in higher tibial posterior translation and higher tibial internal rotation.
It is known that anterior/posterior translations are mainly constrained by the cruciate ligaments [
ACL and PCL participate only to a small extent in rotational stabilization of the knee joint, whereas MCL is a main stabilizer for rotational movements [
Larger wear areas were observed on the lateral plateaus especially when testing unstable knee conditions. This may be related to the concept of wear simulation. Restraint during simulation is the sum of replicated passive structures, friction of the articulation, and restraint via implant design. Reducing the restraint of the passive structures during simulation will increase kinematics when no substituting via design or friction is occurring. Simulation is run with higher axial loading on the medial plateau. This results in smaller kinematics on the medial plateau (pivot point) and higher kinematics on the lateral plateau. This is a limitation of this study. During simulation only the restraint of the passive structures is replicated. However, ligamentously unstable conditions do alter not only restraint characteristics but also the mechanics (alignment and force transmission) of the joint, which has been neglected in this study.
Results showed that ligamentous-unstable TKR resulted in highly increased wear rates with an increased number of generated wear particles. The increased wear may be due to increased kinematics. Increased secondary movements, especially the cross-shear ratio [
Retrieval analysis of high-conforming TKR has been associated with an increased risk of wear-related failure [
Under standard laboratory test conditions, it remains unclear whether conformity of knee designs results in an increased or decreased wear behavior [
Typically, wear testing of TKR is carried out according to ISO standards. In these ISO standards, only the cruciate ligaments (ACL/PCL) are considered (sacrificed/retained). This seems to be appropriate as the ACL is typically sacrificed during TKR implantation and the absence or insufficiency of the PCL is seen commonly in clinical settings. However, deficient ligamentous conditions are clinically often related to traumatic and degenerative changes. Changes to isolated structures, as defined by ISO, are rare. They would mostly occur in several structures (e.g., capsule, cruciate, and collateral ligaments) to varying extents [
In this study, the unstable ligament model was chosen as the worst case scenario since (1) PCL is known for its restraining role in tibial posterior translation [
The tested high-conforming knee design resulted in increased tibial posterior translation and tibial internal rotations under ligamentous-unstable knee conditions. This can be related to insufficient stabilization via implant design. The tested design was not capable of compensating for the insufficient ligamentous stabilization.
The insufficient stabilization was accompanied by an increased wear related to higher kinematics. Increased wear rates and a higher number of wear particles of comparable size and morphology were observed under ligamentously unstable test conditions.
The authors declare that they have no conflict of interests regarding to the publication of this paper.