Antioxidant-Loaded Highly Cross-Linked Polyethylene in Total Knee Arthroplasty: The Need for Radiostereometric Randomized Controlled Trials

Prentice et al. have published a well-performed registry study on the association between antioxidants in highly cross-linked polyethylene and the risk of revision in patients who underwent primary total knee arthroplasty (TKA) with patellar resurfacing for osteoarthritis. They observed a lower revision risk in the group with antioxidants compared with the group without antioxidants, especially regarding revision due to polyethylene wear. Although this study is undoubtedly timely and relevant to the ongoing debate about implant materials, there are some methodological concerns that need to be addressed for a more robust interpretation of the findings. For this purpose, it could be useful to consider how closely the study by Prentice et al. replicates the conditions and rigor of a well-conducted randomized controlled trial (RCT)1. In this light, confounding is an important concern, as the groups were not comparable at baseline on important prognostic factors, as would be the case for a well-conducted RCT. For instance, there was an unequal distribution of annual utilization between the groups: polyethylene with antioxidants was used mostly from 2017 onwards, whereas polyethylene without antioxidants was mostly used before 2017. It is reassuring to see that there was no significant interaction between the polyethylene group and year on the risk of revision, but this does not completely rule out confounding. The difference in when surgeries were performed could create a time-related bias. Over the years, outcomes for TKA have generally improved because of advances in surgical techniques and implant designs. If this is not taken into account, by adjusting for time or analyzing results by specific time periods, there is a risk of wrongly crediting antioxidant polyethylene for these better outcomes, when they might actually just reflect overall progress in the field. Additionally, the analysis does not fully account for potential confounding by implant. The polyethylene with antioxidants group is not identical to the polyethylene without antioxidants group in terms of implant models according to a table in the Appendix. For instance, the Zimmer-Biomet Vanguard TKA was only used in the antioxidants group, whereas Zimmer-Biomet Durasul, Zimmer-Biomet Natural Knee, Zimmer-Biomet NexGen, and Zimmer-Biomet Prolong were only used in the group without antioxidants. A possible consequence could be that any observed benefit could be due to the implant differences rather than the antioxidants. Therefore, it is important to also consider the results from other studies. Other studies have failed to find an effect of antioxidants in polyethylene on the risk of revision or wear. An analysis of the American Joint Replacement Registry on approximately 340,000 primary TKAs found no evidence of reduced risk of revision for polyethylene with antioxidants compared with polyethylene without antioxidants2. There have also been retrieval studies showing no difference in wear between retrievals of TKA implants with polyethylene with antioxidants compared with polyethylene without antioxidants3. Given the small observed effect size, the methodological concerns outlined, and the results from other studies, the results found by Prentice et al. should be interpreted with some caution. The authors themselves suggest that studies with more rigorous designs are necessary to draw definitive conclusions. A relatively small RCT of ≤100 patients utilizing radiostereometric analysis to assess wear and implant migration, with a 1-year follow-up, could provide clearer evidence for the potential benefits of antioxidants in polyethylene4. For now, it may be prudent to follow the advice from Kendall et al.2 to caution against widespread adoption of highly cross-linked polyethylene with antioxidants in the absence of proven superiority over other types of polyethylene with proven track records.