In this study of components retrieved at the time of revision surgery, we matched 44 XLPE to 44 conventional PE inserts from four manufacturers; the matching approach considered implant design (exact match), insert size (exact match), and length of implantation (matched 6 months). Surface damage on the articular surfaces was subjectively graded and digitally mapped to determine the percent damaged area of each damage mode. Three-dimensional changes that had occurred as a result of implantation were determined by comparing laser scans of the retrieved inserts with size-matched pristine inserts.
We found no difference in surface damage between matched XLPE and conventional PE inserts of the same designs. However, increased dimensional changes in TKAs with XLPE may reflect larger contact areas and potentially explain improved performance of XLPE in published simulator studies.
pe design 8 serial crack for 12
Highly crosslinked ultrahigh-molecular-weight polyethylene (XLPE) has demonstrated superior wear performance compared with conventional polyethylene (PE) in THA with followup studies beyond 10 years showing few or no patients with osteolysis [5, 6, 10, 21, 24, 26, 32]. XLPE was more recently introduced into TKA after in vitro knee simulator findings of lower wear rates compared with those found with conventional PE [12, 22, 33, 34]. Thus far, however, the clinical benefit of XLPE remains unclear. For example, the Kaiser Implant Registry [16] with 62,177 primary TKAs at followups of 1 to 5 years showed no difference in revision surgery between XLPE and conventional PE bearing surfaces. In contrast, the 2014 Annual Report of the Australian Orthopaedic Association National Joint Replacement Registry [1] found primary TKAs with XLPE had a lower rate of revision than those with conventional PE, although this result was confounded by the much smaller amount of XLPE in use and variations in implant design.
Additional concerns exist for the use of XLPE in TKA. The larger stresses and more complex stress distributions in tibial TKA PE components compared with PE acetabular components in THA [3, 4] must be considered in light of the alterations in material properties caused by the crosslinking process [2, 29]. For example, the reduced fracture toughness in XLPE is a cause for concern in tibial post fractures in posterior-stabilized TKAs [30]. Evaluation of surface damage and oxidative properties of retrieved XLPE TKA tibial components from revision surgeries showed little improvement with the use of XLPE [20, 27]. The first retrieval analysis of XLPE tibial inserts used an optical grading method to analyze surface damage and found no differences between eight XLPE tibial inserts and 71 conventional PE inserts [23]. However, a major limitation of these retrieval studies is the lack of controlled conditions across study groups. Implant design and length of implantation are important variables that may bias a comparison of the benefit of XLPE and conventional PE tibial components based on materials alone. Also, none of the investigations on XLPE damage were done in three dimensions, and differences in damage patterns between materials were not examined.
Therefore, we designed a matched-pair analysis of components retrieved at revision surgery and used the techniques of damage scoring, damage mapping, and three-dimensional (3-D) laser scanning to ask the following questions: (1) is XLPE more damage-resistant than conventional PE inserts in TKA based on damage score and percent area affected; and (2) does XLPE differ in extent and location of dimensional changes as measured with 3-D laser scanning compared with conventional PE?
For 29 pairs (four pairs of DePuy, 10 pairs of Smith & Nephew, and 15 pairs of Zimmer TKAs), pristine, never-implanted TKA conventional PE tibial inserts were available in our laboratory to match the designs and sizes of the retrievals. Funding for the study did not permit the purchase of matching pristine inserts for the additional 15 pairs.
The locations of the dimensional changes on the articular surfaces were consistent within designs; however, the location of dimensional changes varied among designs: retrieved Zimmer posterior-stabilized inserts had negative deviations centrally located on the plateaus, Smith & Nephew posterior-stabilized inserts had negative deviations located on the perimeter of the plateaus, and DePuy posterior-stabilized inserts had negative deviations on the entire surface of the plateaus (Fig. 6).
Limitations in our study must be considered. Inherent limitations of any retrieval study are that such studies are retrospective for which certain variables such as reason for revision cannot be controlled and that the cohort includes implants that by definition have failed and thus may not reflect well-functioning devices. However, we were able to use a large series of retrievals to match the manufacturer, implant design, and LOI, and in the current study, the differences between materials in revision indications did not reach statistical significance (Table 3). Another limitation is the short LOI (an average of 15 months) for the studied inserts. Future retrieval studies will need to see whether the no-difference findings in our report remain as such or whether indeed differences in surface damage between XLPE and conventional PE develop over time. A third limitation was the scanning of pristine inserts to determine the deviations in the retrieved inserts. We assumed the pristine conventional PE inserts had the same original surface geometries that our retrieved XLPE and conventional PE implants had before implantation. However, the manufacturing tolerances of the pristine and retrieved implants could have been large enough to influence our results. However, this limitation was unavoidable, because the original dimensions of the retrieved implants were unavailable. As a final limitation, the size-matched pristine implants in our study were conventional PE, and we only had 29 of 44 pairs available for analysis. However, based on our 3-D scanning data, similar colorimetric maps were seen on the articular surfaces in both XLPE and conventional PE inserts. Therefore, we believe it acceptable to use conventional PE inserts in investigating the dimensional changes of the same design XLPE implants.
Approaches other than laser scanning exist to assess PE wear. For example, knee simulator studies use gravimetric methods to measure the amount of PE worn from the articulating surfaces [12, 22, 33, 34]. Of course this approach cannot be applied to retrieved implants because the original weights of the implants are unknown and the implants may show added weight through embedded third-body particles. Clinically, wear can be determined from serial radiographs by measuring the penetrating depth of the opposing metallic implant into the polyethylene surface. Although not as widely adapted for knee arthroplasty as for hip arthroplasty, the technique has been used to measure wear rates as a function of polyethylene sterilization techniques [7] and to predict TKA failure [9]. Of course the technique relies on the availability of high-quality, serial radiographs, which were unavailable for our study.
To our knowledge, this is the first matched-pair study investigating the differences between XLPE and conventional PE inserts. This large short-term revision retrieval study demonstrated only minor differences in accumulated surface damage and dimensional changes between matched groups of highly crosslinked and conventional PE TKA inserts of the same designs. The lack of meaningful differences between the two PE materials suggests caution in adopting a new, more expensive bearing material over another material that has a long track record of excellent behavior. Long-term clinical and longer-term retrieval studies will be necessary to elucidate any clinical advantages of using XLPE tibial components. 2ff7e9595c
Comments