This document specifies the relative angular movement between articulating patellofemoral joint components, the pattern of the applied force, speed and duration of testing, sample configuration and test environment to be used for the durability testing of total knee-joint prostheses in wear-testing machines with load control and displacement. Multi-user access to over 3, medical device standards, regulations, expert commentaries and other documents. Worldwide Standards We can source any standard from anywhere in the world. Learn more about the cookies we use and how to change your settings.
|Published (Last):||8 March 2013|
|PDF File Size:||20.99 Mb|
|ePub File Size:||2.51 Mb|
|Price:||Free* [*Free Regsitration Required]|
LTD, Beijing, China. ISO is the governing standard for wear testing of knee prostheses, but there is controversy over the correct direction of anterior-posterior AP displacement and loading and the correct direction of tibial rotation TR angles and torque. This study aimed to analyze how altering the direction of AP and TR affected wear on the tibial insert.
This study found that changing the input directions for AP and TR according to ISO —1 and —3 had an influence on the wear rate and wear contours on the tibial insert model. However, the extent of wear varies depending on the design features of the tibial insert and shape of the input curves. For displacement control according to ISO —3, changing the direction of AP displacement had a marked influence on the wear rate The clinical relevance of this study is that the results demonstrate that tibial wear is affected by the direction of application of AP and TR.
Incorrect application of the loading conditions during the design stage may lead to an ineffective preclinical evaluation and could subsequently influence implant longevity in clinical use. The success of total knee arthroplasty TKA for returning knee functionality has contributed to its widening application for treating diseases of the knee that have failed conservative treatments [ 1 ].
However, even with successive improvements in implant designs and materials, implant failure and patient dissatisfaction still persist [ 2 — 5 ]. Loosening of the implant is the most common reason for requiring a second TKA [ 6 — 8 ], which is reported to be linked to malalignment of the motion axis and the generation of wear particles which can induce osteolysis around the implant [ 9 — 10 ].
Therefore, in vitro wear testing is an important factor in the development of knee protheses and is a key requirement for regulatory clearance of such devices.
ISO is the most commonly used standard for evaluating the wear properties of knee implants [ 11 — 17 ]. There are two control modes for the wear test during simulated gait: displacement control according to ISO —3 [ 11 — 12 ], and load control according to ISO —1 [ 13 — 14 ].
The inputs for displacement control are anterior-posterior AP displacement and tibial rotation TR angle, and the inputs for load control are AP load and TR torque. The knee simulator used to perform the gait movements requires four inputs; flexion angle, axial load, AP displacement or AP load, TR angle or TR torque. The magnitude and direction of the flexion angle and axial load inputs are same across the ISO range of standards [ 11 — 14 ] Fig 1.
According to ISO [ 11 — 14 ], the TR angle is positive when the tibia rotates internally and AP displacement is considered to be positive when the tibia moves anteriorly. A precise and accurate standard for wear performance is crucial for the development of knee prostheses as it predicts the lifetime of the tibial bearing. ISO —3 is the primary standard for displacement-controlled simulations of wearing of knee prostheses.
Compared with the revision of ISO —3, the revision reversed the direction of AP displacement and TR angle but maintained the same magnitudes Fig 2. However, during normal human gait the tibia undergoes positive AP displacement and rotates through a positive TR angle [ 18 — 23 ]. ISO —1 is the primary standard for load-controlled simulations of wearing of knee prostheses. This is due to knee flexion and femoral rollback occurring simultaneously [ 18 — 23 ].
A number of studies [ 24 — 29 ] have investigated how varying the parameters outlined in ISO may impact the knee joint; differences in wear between load and displacement control [ 24 — 25 ], the effect of varying the amplitude of inputs for displacement control on wear [ 26 ], the effect of anterior-posterior and internal-external motion constraints on wear [ 27 ], the impact of different activities on the wear performance [ 28 ], comparison between electromechanically- and pneumatically-controlled knee simulators [ 29 ].
To the best of our knowledge, no previous studies considered the influence of altering the direction of AP and TR on wear rates.
A misjudgment or incorrect evaluation in this area may lead to an ineffective preclinical evaluation and subsequently influence the longevity of the tibial insert in clinical use. A retrieved knee prosthesis PFC, Depuy Synthes of the right knee was used both to construct the finite element model and for experimental work Fig 4A. An examination of the surfaces of the femoral component and tibia insert found only minor burnishing, abrasion, and scratching, but no obvious cold flow, pitting, embedded metal, delamination or wear through [ 30 ].
In the absence of any marked deformities, the minor surface abrasion was deemed acceptable as it would not have any noticeable influence on knee kinematics. Inconsistencies on the articular surface between the 3D model and the physical PFC implant were controlled to be less than 0.
A Retrieved tibial insert from PFC implant; B Variations in the articular surface between the 3D model and the retrieved insert were within 0. The two models were identical except the load control model had additional nonlinear connector constraints for limiting AP motion and tibial rotation TR according to ISO —1 [ 12 , 14 ].
The tibial insert was modeled as deformable elastic polyethylene Gur with approximately 81, C3D10M elements of size 1 mm for the articular surface and 1.
The femoral component was modeled as a rigid body. A coefficient of friction of 0. The boundary conditions were set according to ISO standards [ 11 — 14 ]. The loading points on the tibia and femur was offset to the medial side by a distance of 0.
The mechanical axes described by Grood and Sunday were used [ 32 ]. Input curves were applied to the tibia and only a flexion angle was applied to the femoral side. For the displacement controlled models Fig 5A , the input curves were AP displacement, tibial rotation TR angle, flexion angle and compression loading.
For the load controlled models Fig 5B , the input curves were AP load, TR torque, flexion angle and compression loading. Nonlinear constraints in AP and TR directions were imposed on the load controlled model. Along the AP direction, the constraint stiffness was set as 9. The tibial insert was free to move in the medial-lateral, anterior-posterior, superior-inferior and valgus-varus directions, as well being free to rotate within the tibial component, but was constrained in flexion.
In contrast, the femoral component was only permitted to move in flexion. The flexion axis was defined as a line connecting the medial and lateral centers of the posterior femoral radius arc. The wear coefficient K was 2. According to the calculated wear depth, each node on the surface was moved in the direction normal to the articular surface. An adaptive remeshing procedure was employed to simulate the progression of surface wear.
It was calculated for 5 million cycles according to ISO [ 11 — 14 ]. The volume of wear was calculated by the difference between the initial volume of the tibial insert and its final volume. The tibial insert surface was updated every , cycles, which has been shown to only have a difference of between 2. The contact pressure was validated using a Tekscan pressure distribution measuring system Tekscan Inc, America S1 File.
The following sections detail the validation of the sliding distance and contact areas using wear contours and feedback curves. The calculation process was developed and validated for use in previous studies on TKA in our laboratory [ 34 — 35 ]. As shown in Fig 6A , an electromechanically-controlled knee simulator, Prosim Prosim, Simulation Solutions , was used to perform the gait movements according to the suggested modifications to ISO —3 with a positive AP displacement tibia moves anteriorly and positive TR angle tibia rotates internally.
ABS plastic was used to construct the jigs Fig 6B. Before the test, the medial and lateral articular surfaces were uniformly coated with small dots using a permanent marker, which was selected as an easy to apply and non-water soluble coating material Fig 6C.
A simple template with uniformly spaced holes was used to ensure consistency with positioning of the holes. The loading points on the tibia and femur were offset towards the medial side by a distance of 0. The flexion axis was the same as the FEA model detailed above.
The frequency was set at 1 Hz and the test was performed in the air without a medium. The wear contours on the tibial insert produced by the FEA model Fig 7A were very similar with the wear patterns produced by the knee simulator Fig 7B. The flexion angle, tibial rotation angle, AP displacement, and axial force were recorded from both the FEA model and experimental setup and then compared with the expected inputs from the modified ISO —3 requirements Fig 8. Fig 8 demonstrates the remarkable similarity between the three sets of data, which confirms the validity of the FEA models introduced in this study.
For the displacement control models, as shown in Fig 9 , the graph of AP displacement plotted according to ISO — is an inverse of the graph plotted according to the modified ISO —3. The AP reaction loads are also acting in the opposite direction, but the plot is not symmetrical. A similar tendency can be seen for tibial rotation whereby the graph of TR angle plotted according to ISO — is an inverse of the graph plotted according to the modified ISO —3.
The corresponding TR torque also acts in the opposite direction, although is not symmetrical. For the load control models, as shown in Fig 9 , the graph of AP load plotted according to ISO — is an inverse of the graph plotted according to the modified ISO —1.
The wear rate, volumetric wear and maximum wear depth calculated from the FEA models are detailed in Table 2. For load control, reversing the direction of AP load ISO only increased the wear rate by 6.
Fig 10 shows the wear contours for all models. The wear contours from the modified ISO —3 and modified ISO —1 standards are more consistent in comparison to the other models both central and slightly posterior on the tibial insert , demonstrating a more natural knee motion [ 18 — 23 ].
In the displacement control models Fig 10A—10C it can be seen that the wear contours for the modified ISO —3 model are central and slightly posterior on the tibial insert, whereas wearing on the ISO — model occurs in a more anterior position.
This is due to the reversed input direction for AP displacement. For load control Fig 10D and 10E , the wear contours were positioned more anteriorly or posteriorly depending on the direction of the input for AP load. However, the level of wear varies depending on the design features of the tibial insert and the details of the input curves.
For the displacement control models, reversing the direction of AP displacement ISO — increased the rate of surface wear by Anterior positioning of the knee contact points results in a greater AP load Fig 9 and higher wear rates.
Reversing the direction of TR angle ISO — does not have such a marked influence on wear, which may due to the symmetrical design of the tibial insert whereby the lateral side is the same as the medial side. ISO — If the insert was not symmetrical, but instead was an anatomical or medial pivot design, reversing the direction of TR angle may have a marked influence on wear.
For the load control models, reversing the AP load also increased the wear rate but to a lesser degree than the displacement controlled models. This may be due to the AP load input curves Fig 3 having both positive values and negative values.
Even when assigned opposing directions for AP loads according to ISO — and modified ISO —1, there was less of a difference in the mean AP loads for the load control models, and thus less of a difference in wear In contrast, AP displacement Fig 2 is either always positive modified ISO —3 or always negative ISO — during the motion cycles, so changing the direction of AP displacement had a marked influence on wear. The load control method modified ISO —1 yielded Therefore, for the load control models, the AP load and TR torque inputs resulted in a greater range of AP displacement and TR angles than in the displacement control models Fig 9.
The direction of AP and TR also had a marked influence on the wear contours. In the displacement control models, reversing the direction of AP displacement according to ISO —3 impacted the loading patterns and magnitude, resulting in a marked difference in wear contours.
The same phenomenon was observed for the load control models ISO —1. Overall, the wear contours from the modified ISO —3 and modified ISO —1 were found to be more consistent than the other models.
This could be explained by most of the AP displacements and TR angles being in a positive direction Fig 9. The FEA model used in this study was fully validated prior to commencing any wear simulations, as highlighted in the following five points. Previous studies on wearing of knee implants [ 24 — 25 ] failed to consider variations in the direction of AP and TR, which has been shown in this study to be a critical factor in determining knee kinematics and the level of wear on the tibial insert.
Similarly, Schwenke T et al. There are a number of limitations to this study that should be noted.
ISO 14243-1:2009/Amd 1:2020
Knee Wear Testing - Test Methods
BS ISO 14243-5:2019