Crosslinked Polyethylene Demonstrates Substantially Improved Performance at Minimum 10-Year Follow-up Compared to Conventional Polyethylene

This article refers to:

Cross-Linked Versus Conventional Polyethylene for Long-Term Clinical Outcomes After Total Hip Arthroplasty: A Systematic Review and Meta-Analysis

The introduction of ultra-high molecular weight polyethylene as a bearing surface for total hip arthroplasty (THA) by Dr. John Charnley during the 1960s was one of the most important innovations that revolutionized the field and ushered in the modern era of joint replacement. While Charley’s low-friction arthroplasty provided reliable pain relief and improved function, complications related to conventional polyethylene (CPE) bearing surface wear have been the most important concern at long-term follow-up. These complications include wear through the full thickness of the liner and osteolysis attributed to polyethylene debris particles. Liner wear can also contribute to late instability while osteolysis can lead to component loosening and periprosthetic fractures through expansile lesions.

Recognition of these complications prompted the orthopedic community to investigate different polyethylene formulations and processing techniques. While promising laboratory performance data often preceded the introduction of new materials, clinical performance proved to be marginally improved, at best, and sometimes worse than historically available materials. As investigators began to examine how particular factors influenced wear, the terminal sterilization method was found to be among the most important [1]. Although there is considerable variation among individual patients, CPE liners sterilized with gamma radiation dosages ranging from 2.5 to 4.0 Mrad have demonstrated mean wear rates in the neighborhood of 0.1 mm per year (mm/yr) [1,2]. In contrast, CPE liners terminally sterilized with non-crosslinking chemical surface treatments, such as ethylene oxide or gas plasma, usually have mean rates on the order of 0.2 mm/yr [1,3]. Since terminal sterilization using gamma radiation had demonstrated reduced wear compared to non-crosslinked polyethylene owing to radiation-induced crosslinking despite the presence of residual free radicals that could oxidize prior to implantation or in vivo, there was reason to be optimistic that higher crosslinking dosages coupled with additional processing to reduce or eliminate free radicals might further reduce wear rates. However, laboratory testing indicated that wear reductions tended to decrease with increasing crosslinking dosages and crosslinking was also accompanied by reductions in material properties including ultimate tensile strength and fracture toughness [4,5].

Because the optimal amount of crosslinking was unknown, different types of crosslinked polyethylene (XLPE) were developed that sought to balance the improved wear resistance associated with crosslinking against the reductions in mechanical properties that are inherent to the manufacturing process [4,5]. Among first-generation XLPE liners, crosslinking was typically achieved with radiation dosages ranging from 5 to 10 Mrad applied using gamma irradiation or an electron beam. Because the radiation also generates free radicals that can oxidize and degrade the mechanical properties of the polyethylene, heat treatments including annealing and remelting were typically used to reduce or eliminate these free radicals. Owing to reductions in mechanical properties and because first-generation formulations did not incorporate antioxidants, the introduction of XLPE has been accompanied by concerns regarding the potential for liner fracture, in vivo oxidation, and accelerated wear at long-term follow-up. Additionally, some investigators have questioned whether the smaller size of XLPE debris particles might contribute to an increased bioreactivity compared to CPE [6].

Even when substantial improvements occur, it can take many years to see clinical evidence of the changes. This is certainly true of XLPE. At 5-year follow-up, first-generation XLPE demonstrated substantially reduced radiographic wear rates compared to CPE liners without perceptible differences in osteolysis incidence or revision rates. This is consistent with observations that osteolysis associated with CPE typically becomes perceptible on radiographs obtained at 5 to 10-year follow-up and wear-related complications lead to increasing revision rates that are more apparent after 10 years. Based on these trends, decreases in the incidence of osteolysis with XLPE were not likely to become evident until later during the first decade of clinical service and reductions in revision rates would have been expected to become progressively more apparent during the second decade.

As nicely demonstrated in the meta-analysis by Shi et al [7] based on minimum 10-year follow-up studies, both prospective, randomized clinical trials and retrospective cohort studies have demonstrated that XLPE substantially reduces the incidence of revision and osteolysis compared to CPE with risk reductions of ∼80%. While the similar performance of liners from several different manufactures indicates that it is reasonable to group first-generation XLPEs together for the purposes of a meta-analysis, considering whether there are any clinically important differences in the wear rates among these liners is also useful. Unfortunately, comparing wear rates among studies can be challenging because investigators often use different methods to evaluate wear [8] and Shi et al acknowledge this as a potential source of heterogeneity [7]. Because the XLPE liners included in their meta-analysis were manufactured using different radiation dosages and heat treatments, it also seems plausible that subtle differences in the wear rates could be contributing to some of the heterogeneity. When they reviewed the literature, Dumbleton et al. [9] found that osteolysis was rarely observed at a wear rate less than 0.1 mm/yr and suggested that a wear threshold of 0.05 mm/yr would eliminate osteolysis. Because almost all first-generation XLPEs have been reported to have mean wear rates less than 0.05 mm/year [2], there may not be meaningful differences in their clinical performance as quantified by osteolysis incidence and revision rates. This appears to be the case through 15-year follow-up but continued surveillance is warranted. It remains to be seen if sufficiently low wear rates will completely eliminate osteolysis or simply delay its onset until a critical cumulative wear threshold is reached.

Now that the orthopedic community has two decades of clinical experience with XPLE, the concerns about its performance have yet to be realized. Although liner fractures and other mechanical failures have been reported anecdotally [10], they appear to be rare [9]. When they do occur, these failures can usually be attributed to several factors that often include suboptimal cup position which can also contribute to the failure of CPE. Allaying concerns about increased bioreactivity, the review by Shi et al. found that the incidence of osteolysis was significantly lower in the XLPE group compared to the CPE group (5.4% vs. 31.5%, p < 0.001) [7].

Despite substantial reductions in wear, osteolysis and revision rates, the meta-analysis by Shi et al did not find significant differences in the Harris Hip or WOMAC scores [7]. Although the meta-analysis had limited power because fewer studies reported this type of data, the absence of a difference could also be attributed to the fact that wear and osteolysis are typically asymptomatic until they result in fracture, instability or component loosening.

While the meta-analysis by Shi et al reported lower revision rates for XLPE that were fairly uniform among the 14 studies the authors reviewed, there was some heterogeneity in osteolysis and wear rates [7]. In several instances, the authors were able to attribute the heterogeneity to a single study which underscores the importance of drawing conclusions based on a consensus derived from multiple peer-reviewed publications. Studies with atypical results should be closely scrutinized, particularly when they involve relatively small patient populations or rely on methods that are not widely used by other investigators.

When describing first-generation XLPE, some investigators have characterized polyethylene treated with 5 Mrad dosages as moderately crosslinked while polyethylenes exposed to 10 Mrad have been referred to as highly crosslinked. Although crosslinking is known to increase with radiation dosage, the marginal increases are reduced with higher dosages and subsequent heat treatments may also contribute to the cumulative amount of crosslinking. Based on the data currently available, distinctions between the incidence of osteolysis and revision rates for different first-generation XLPEs have not become apparent. In the future, it would be useful for investigators to report the crosslink density whenever possible to better characterize and more meaningfully compare XLPE materials.

With the success of first-generation XLPE, second-generation materials have introduced antioxidants and sequential crosslinking with the ambition of eliminating the potential for in vivo oxidation and maximizing mechanical properties. Long-term follow-up will be essential to demonstrate if second-generation XLPE has clinically perceptible benefits compared to first-generation materials.

Although medical innovations sometimes fail to meet expectations, the introduction of XLPE has substantially decreased wear rates and virtually eliminated osteolysis. Despite variations in manufacturing processes, first-generation XLPE liners appear to be performing similarly. Compared to CPE, reductions in revision rates are becoming more apparent during the second decade of clinical service. Through 15-year follow-up, mechanical failures remain rare and in vivo wear rates do not appear to be increasing with time despite the absence of antioxidants. For a patient who is ∼60 years of age at the time of their primary THA, it appears that XLPE has led to a hip replacement that is unlikely to fail due to wear-related complications during their lifetime. The clinical data now available indicates that XLPE represents one of the most important advances since Charnley revolutionized the field and should be regarded as the gold-standard for THA bearing surfaces.

Disclosure Statement

The author reports no conflict of interest.