Abstract
Introduction
Highly crosslinked ultrahigh-molecular-weight polyethylene (XLPE) reduces wear and osteolysis in total hip arthroplasty, but it is unclear if XLPE will provide the same clinical benefit in total knee arthroplasty (TKA). Adhesive and abrasive wear generally dominate in polyethylene acetabular components, whereas fatigue wear is an important wear mechanism in polyethylene TKA tibial inserts. The wear resistance of XLPE depends on the crosslink density of the material, which may decrease during in vivo mechanical loading, leading to more wear and increased oxidation. To examine this possibility, we measured crosslink density and oxidation levels in loaded and unloaded locations of retrieved tibial inserts to evaluate the short-term performance of XLPE material in TKA.
Materials and Methods
Forty retrieved XLPE tibial inserts (23 remelted, 17 annealed) retrieved after a mean time of 18 ± 14 months were visibly inspected to identify loaded (burnished) and unloaded (unburnished) locations on the plateaus of each insert using a previously published damage mapping method. For each insert, four cubes (3 mm3) were cut from loaded and unloaded surface and subsurface locations (Fig. 1). Swell ratio testing was done according to ASTM F2214 to calculate crosslink density of the cubes. With a microtome, 200 μm sections were taken adjacent to the cubes and oxidation was assessed with Fourier transform infrared spectroscopy following ASTM F2102 (Fig. 2). Surface oxidation was measured in the sections adjacent the surface cubes and subsurface oxidation was measured in sections adjacent to the subsurface cubes. The effects of location (surface vs. subsurface in the loaded and unloaded regions) and thermal treatment (annealed vs. remelted) on crosslink density and oxidation were assessed with repeated measures generalized estimating equations (GEEs), with the implant treated as the repeated factor. Results are presented as means and 95% confidence intervals and the level of significance was α=0.05.
Results
Crosslink density was associated with location within the polyethylene tibial inserts (p<0.001), while oxidation was associated with both location (p<0.001) and heat treatment (p=0.003). The loaded surface (location 1 in Fig. 1) had 13% lower crosslink density than all other locations (p<0.001 for each), and greater oxidation than all other locations (Fig. 3). Specifically, oxidation of the loaded surface was 0.29[0.17,0.40] greater (two times greater) than that of the unloaded surface (p < 0.001), whereas subsurface areas of loaded and unloaded regions differed by only 0.03[0.00,0.07] (p<0.022). Additionally, surface oxidation was over 7-fold greater than subsurface oxidation in the loaded region (difference: 0.56[0.44,0.68], p<0.001). Annealed XLPE had 2-fold greater oxidation than remelted XLPE (difference 0.159, 95% CI = 0.045, 0.126), and this was independent of location within the inserts.
Conclusions
In vivo loading of XLPE decreased the crosslink density and increased the oxidation in areas that underwent wear and deformation at the articular surface of TKA inserts. Nonetheless, in these short term retrievals, no clinical complications were attributed to the change in material properties. However, if crosslink density continues to decrease with load over time, XLPE may not provide a clinical advantage over conventional polyethylene in TKA.
