 |
 |
 |
|
| home | CME | help | feedback | subscriptions | archive |
|
The Journal of Bone and Joint Surgery (American). 2007;89:2779-2786. doi:10.2106/JBJS.G.00043 © 2007 The Journal of Bone and Joint Surgery, Inc.
Clinical Performance of Highly Cross-Linked Polyethylenes in Total Hip Arthroplasty
1 Lexington Clinic, 1221 South Broadway, Lexington, KY 40504. E-mail address for C.A. Jacobs: cjaco{at}lexclin.com
Disclosure: The authors did not receive any outside funding or grants in support of their research for or preparation of this work. One or more of the authors, or a member of his or her immediate family, received, in any one year, payments or other benefits in excess of $10,000 or a commitment or agreement to provide such benefits from commercial entities (DePuy Orthopaedics and Biomet Orthopedics). Commercial entities (DePuy Orthopaedics, Biomet Orthopedics, and Smith and Nephew) paid or directed in any one year, or agreed to pay or direct, benefits in excess of $10,000 to a research fund, foundation, division, center, clinical practice, or other charitable or nonprofit organization with which one or more of the authors, or a member of his or her immediate family, is affiliated or associated.
Aseptic loosening secondary to wear-debris-induced osteolysis has been identified as the leading cause of late failure of total hip arthroplasty. Highly cross-linked polyethylene acetabular liners were developed as one approach to reducing this wear. Preclinical laboratory wear testing showed a number of cross-linked polyethylenes to have dramatically less wear than the polyethylene that had been in use for several decades. After the initial bedding-in phase (one to two years), the percent reductions in the wear rate, as indicated by the amount of penetration of the head into the socket evident on serial radiographs, have been comparable with what was predicted from preclinical hip-simulator testing of the highly cross-linked polyethylenes. To our knowledge, there have been no reports of clinically relevant osteolysis that was clearly attributable to wear of a highly cross-linked polyethylene acetabular liner. However, the clinical performance of these materials should be closely monitored with long-term follow-up.
Total hip arthroplasty with use of traditional metal-on-polyethylene bearing couples has been demonstrated to be an effective treatment to improve function and reduce pain in patients with severe osteoarthritis, rheumatoid arthritis, osteonecrosis, and other conditions affecting the hip1. Despite this overall clinical success, the clinical life span of total hip prostheses often has been limited by osteolysis and, in some cases, aseptic loosening brought on by the macrophage response to wear particles2. Osteolysis secondary to polyethylene wear has been described as one of the primary reasons for late revision of total hip arthroplasties3,4. It follows that improving the wear resistance of the polyethylene could contribute to greater durability of hip prostheses. As a result of the aging of the "baby boomer" population, the number of primary total hip arthroplasties performed annually in the United States has been projected to increase during the next fifteen years from 208,000 to more than 384,0005. Not surprisingly, the number of revision procedures also is projected to increase, from 36,000 to an estimated 67,000 annually by the year 20205. Furthermore, during the past decade, the indications for total hip arthroplasty have been expanded to include progressively younger patients. The increased activity level of these patients, combined with their increased life expectancy, has in turn amplified the need for greater implant durability. Since wear-related osteolysis was first identified as a cause of revision, other bearing couples have been developed in an attempt to decrease particle generation. In the 1990s, a number of highly cross-linked polyethylenes were developed as one approach to reducing wear of total hip prostheses and, potentially, to increasing implant longevity. Laboratory wear testing of the highly cross-linked polyethylenes in hip joint simulators demonstrated dramatically improved wear resistance, with many investigators reporting reductions in wear of >90% even under adverse conditions, such as with large-diameter femoral heads, with roughened femoral heads, and in the presence of third-body particles6-11. This review documents the midterm clinical performance, and summarizes factors potentially affecting the future performance, of total hip prostheses with highly cross-linked polyethylene bearings.
Reducing polyethylene wear could have a tremendous effect on the clinical success of total hip arthroplasty. Excessive polyethylene wear can lead to late failures as a result of several different mechanisms. The most common failure mechanisms that have been associated with polyethylene wear are osteolysis and aseptic loosening3,4, and Parvizi et al.12 recently reported that excessive wear contributed to the prevalence of late dislocations. Similarly, Clohisy et al.3 reported that aseptic loosening, dislocation, and osteolysis were the primary reasons for revision in 82% of their cases. With this in mind, it is easy to appreciate the clinical importance of improved polyethylene wear characteristics. Revision arthroplasty has not had the same level of clinical success as primary hip arthroplasty. Many factors contribute to the poorer outcomes following revision arthroplasty; these include increased age, infection, inadequate bone stock, and an increased number of comorbidities. The decreased wear rates associated with highly cross-linked polyethylene liners may improve the longevity of primary total hip prostheses by reducing the prevalence of late osteolysis and the resultant bone loss and implant loosening. This could help to reduce the need for revision hip arthroplasty.
The specific processes used to manufacture the highly cross-linked polyethylenes directly affect the amount of cross-linking and, therefore, the material characteristics of these bearings. Three of the most influential factors are the dose of irradiation, the type of post-irradiation thermal processing, and the type of end-point sterilization. So-called first-generation highly cross-linked polyethylenes have been categorized into four types on the basis of these manufacturing processes: cold-irradiated and annealed, cold-irradiated and subsequently melted, warm-irradiated and adiabatic melted, and warm-irradiated and subsequently melted (Fig. 1)13,14.
In the process of creating highly cross-linked polyethylene, ultra-high molecular weight polyethylene is irradiated in order to break the carbon-hydrogen chains within the polymer, and free radicals are created. In an oxygen-rich environment, free radicals facilitate oxidative degradation of the polymer; however, free radicals can also combine with adjacent molecules to form cross-links. As the dose of irradiation increases, so does the density of the cross-linking and, therefore, the resistance to wear15. However, despite increased resistance to wear, excessive cross-linking might not lead to a better overall clinical performance because, as the amount of cross-linking increases, there is a corresponding decrease in some mechanical properties, including ultimate tensile strength and resistance to fatigue crack propagation16. Preclinical testing suggested that the optimum irradiation dose to maximize wear resistance while maintaining tensile strength and resistance to fatigue crack propagation is from 5 to 10 Mrad (50 to 100 kGy)10,17,18. It is not clear whether the amount of cross-linking markedly affects the risk of accelerated polyethylene wear due to impingement of the femoral neck on the rim of the liner. For example, Holley et al.18 used a hip simulator to compare the wear behaviors of polyethylene liners that had been cross-linked with 2.8, 10, or 20 Mrad (28, 100, or 200 kGy) of gamma irradiation, with the added condition of neck-liner impingement. The cups irradiated with 10 Mrad demonstrated the lowest wear rates; however, as the test progressed, those cups apparently generated more small wear particles (approximately 0.2 µm) than did the cups irradiated with 2.8 Mrad. This could be a cause for concern since it has been suggested that particles in the size range of 0.1 to 0.5 µm are more biologically active than larger particles19,20. As mentioned previously, the radiation that is used to generate the cross-linking, whether gamma or electron-beam (e-beam), also generates uncombined free radicals. If these free radicals are allowed to remain in the polyethylene, and if oxygen diffuses into the polyethylene during shelf storage and/or clinical use of the implant, they can predispose the material to severe oxidative degradation. With the highly cross-linked polyethylenes that were introduced in the late 1990s, the concentration of these free radicals was reduced with use of either post-irradiation melting or annealing. Melting of polyethylene means heating it above the melt point (>135°C), so that the polyethylene changes from a partially crystalline state to a completely amorphous solid. Melting allows access to free radicals in the crystalline regions by unfolding the polymer chains. The cross-links that are formed act as molecular constraints during cooling, so that the final crystallinity is lower and the crystals are smaller than before cross-linking and melting. This decrease in crystallinity, in turn, leads to an additional reduction in certain mechanical properties, such as crack resistance and fracture toughness16,21,22. In contrast, annealing refers to heating the polyethylene to a temperature somewhat below the melting point. This avoids the reduction in crystallinity that occurs with melting. However, annealing is much less effective than melting in extinguishing the free radicals, and those that remain in the polyethylene can lead to long-term oxidative degradation of its mechanical properties15. As will be discussed, it is this inherent dissimilarity between melting and annealing that has led to the development of newer, so-called second-generation methods for removing the free radicals from the highly cross-linked polyethylene in an effort to avoid reducing the crystallinity. The type of sterilization has been reported to directly affect the shelf life and wear characteristics of traditional polyethylene23-25. Polyethylene components that were sterilized with gamma radiation in air contained free radicals and therefore were susceptible to oxidation during shelf storage and during use in vivo. Sterilization without radiation—i.e., with use of ethylene oxide or gas plasma—avoided the generation of free radicals, but the lack of cross-linking resulted in less resistance to wear, both during laboratory testing26 and in clinical use24. In part, it was this inherent trade-off between oxidation resistance and wear resistance with the traditional polyethylenes that motivated the development of the cross-linked-remelted polyethylenes, in which the cross-linking is obtained by exposure to radiation. The resultant free radicals are eliminated by remelting, and the terminal sterilization is done with ethylene oxide or gas plasma to avoid reintroducing free radicals. The choice of gas plasma or ethylene oxide does not appear to have a measurable effect on the wear resistance. For example, Durasul and Longevity highly cross-linked polyethylenes (Zimmer, Warsaw, Indiana) are manufactured with use of very similar processes, with the exception that Longevity polyethylene is sterilized with gas plasma and Durasul polyethylene is sterilized with ethylene oxide. Despite the different types of sterilization, these two highly cross-linked polyethylenes have demonstrated very similar wear rates both in the laboratory and clinically27,28. In contrast, as with traditional polyethylene, terminal sterilization of a highly cross-linked polyethylene with gamma radiation generates free radicals in the polymer that are susceptible to oxidation. For example, in an examination of fourteen Crossfire acetabular cups (Stryker Orthopaedics, Mahwah, New Jersey) (which are annealed and terminally sterilized with gamma radiation) that were retrieved after a maximum of thirty-three months of clinical use, Wannomae et al.29 measured levels of oxidation that were as high as those observed in traditional gamma-sterilized-in-air cups after fifteen to twenty years of use. In contrast, Wannomae et al. found that the level of oxidation was negligible in twelve Longevity cups (which are remelted and terminally sterilized with gas plasma) after a comparable duration of use. In an analysis of twelve retrieved Crossfire liners, Kurtz et al.30 reported that oxidation was present in the unworn regions but not in the worn regions. Those authors suggested that the oxidation may not be clinically meaningful, as Crossfire acetabular liners have exhibited low wear in vivo, consistent with their high level of cross-linking31. Longer clinical follow-up will be needed to determine whether the oxidation of the non-bearing regions of Crossfire components might eventually impact their clinical performance.
Traditionally, patient factors such as age, gender, body mass index, and activity level have been related to increased polyethylene wear; however, neither early nor midterm clinical wear of highly cross-linked polyethylene liners appears to be measurably influenced by these factors28,31-33. The material properties and wear characteristics resulting from the specific processes used to induce cross-linking are perhaps more important than patient-related factors. Furthermore, it is important to note that no polyethylene-related complications were reported in four studies of midterm data (after a duration of follow-up of more than five years)31,32,34,35. However, the operations in these studies were all performed by highly experienced surgeons at high-volume centers. High surgeon and hospital volumes have been associated with a decreased risk of revision36,37—i.e., highly experienced surgeons tend to report very low revision rates, regardless of implant selection. Nevertheless, the potential for an increased risk of complications and/or failures of highly cross-linked polyethylene liners as a result of improper acetabular alignment and/or neck impingement serves as a reminder that improvements in polyethylene wear characteristics are not a substitute for sound surgical technique. Proper alignment of the acetabular component is essential for a satisfactory long-term clinical performance of any type of polyethylene. When a cup is inserted in excessive vertical alignment, the contact zone between the ball and cup may be near or at the upper rim (equator) of the cup. This may cause excessively high stresses in the polyethylene, in turn leading to rapid wear and/or fatigue fracture38. Theoretically, this problem might be exacerbated by use of some types of highly cross-linked polyethylene liners for two reasons. First, increased cross-linking and remelting both tend to reduce the fracture toughness of the polyethylene (relative to that of non-cross-linked polyethylene)16,21,22. Second, because of its greater resistance to wear, highly cross-linked polyethylene has been made available for use with larger-diameter balls, to decrease the risk of neck-socket impingement and dislocation. As a consequence, these liners are necessarily thinner, especially at the rim in the location of the locking mechanism, which also reduces the resistance to fatigue cracking39. Although large femoral heads have been associated with increased wear of traditional ultra-high molecular weight polyethylene acetabular liners40, this trend does not appear to carry over to highly cross-linked polyethylene liners. In a hip-simulator study comparing polyethylene wear between nominally cross-linked polyethylene liners (irradiated with 2.9 Mrad [29 kGy]) and highly cross-linked polyethylene liners (irradiated with 10.5 Mrad [105 kGy] and annealed), Hermida et al. reported that the highly cross-linked polyethylene liners demonstrated a 90% reduction in wear when they were used with a 28-mm femoral head and an 85% reduction in wear when they were used with a 32-mm head41. Furthermore, the authors stated that increasing the head size did not significantly increase the wear of the highly cross-linked polyethylene liners. Similarly, in a study with a mean duration of follow-up of 3.3 years, Geller et al. reported that clinical wear rates of Durasul and Longevity highly cross-linked polyethylene liners did not differ among groups of patients with a 36, 38, or 40-mm femoral head42.
The amount of polyethylene wear during clinical use is typically estimated by measuring the distance that the femoral head penetrated into the polyethylene liner over time. Techniques for measuring femoral head penetration on serial radiographs fall into three categories: manual, computer-assisted, and radiostereometry analysis. Each technique has advantages and disadvantages, and there is no clear consensus in the literature regarding which of the three is the most suitable for assessing femoral head penetration into highly cross-linked polyethylene liners. We are not aware of any single study in which all three techniques have been directly compared, although manual techniques have been compared with computer-assisted techniques43,44 and computer-assisted techniques have been compared with radiostereometry analysis45. Some early manual and computer-assisted techniques for measurement of femoral head penetration resulted in relatively large errors in the estimations of polyethylene wear. In a hip-simulator study of polyethylene wear in which they used a phantom apparatus, Kang et al.43 evaluated the accuracy of wear measurements with use of the manual techniques described by Livermore et al.46 and Dorr and Wan47 and the computerized technique described by Devane et al.48. A modified version of the technique described by Dorr and Wan appeared to be the more accurate of the two manual techniques, with the 0.17-mm mean error with this technique being closer to the 0.14-mm mean error of the computerized technique described by Devane et al. than to the 0.21-mm mean error of the manual technique described by Livermore et al. In another laboratory study, Ebramzadeh et al.44 reported that the median error of the method described by Livermore et al. was 0.1 mm, which was similar to the median errors of the two computer-assisted techniques described by Devane et al. and Martell et al.49. The computer-assisted method described by Martell et al.49 presently is the most commonly used technique for assessing femoral head penetration into highly cross-linked polyethylene liners. This method utilizes edge-detection software that first identifies the margins of the femoral head and acetabular cup and then calculates the center of each component. Because variabilities in patient position, exposure dose, and other factors can distort the ellipse created by the mouth of the acetabular component, clinical radiographs can be rejected by the edge-detection software. Consequently, the method of Martell et al. is not purely automated, as observers may be required either to verify the ellipse identified by the software or to manually identify the outer rim of the acetabular cup. Several authors have reported a relatively high prevalence of negative wear values with use of this technique; these may be due to a combination of the small magnitude of wear of highly cross-linked polyethylene, inconsistent radiographic quality due to underexposure and/or patient positioning, and human error in detecting the rim of the acetabular component31,45,50. However, since the measurement error can be in the positive or the negative direction, it is unclear if negative values lead to an underestimation of the amount of polyethylene wear. For example, despite noting that two of three observers occasionally reported negative femoral head penetration values, Bragdon et al.45 found that wear rates estimated with the use of the method of Martell et al. were significantly larger (p = 0.0001) than those determined with radiostereometry analysis. Radiostereometry analysis is considered to be the most accurate method for measuring femoral head penetration. This technique demonstrates the relative change of position over time between the femoral head and multiple radiolucent beads embedded in the polyethylene liner. The mean accuracy has been reported to range from 0.033 to 0.036 mm in the medial direction and from 0.022 to 0.023 mm in the superior direction51,52. However, use of this technique in prospective clinical trials involving large numbers of patients can be both time and cost prohibitive. Also, while radiostereometry analysis may be the most accurate method for detecting migration of the head into the articulating surface of the acetabular liner, it cannot detect backside wear45. Retrieval studies have demonstrated backside wear in 16%53 and 27%54 of modular liners made of traditional ultra-high molecular weight polyethylene, and the resultant polyethylene debris could contribute to osteolysis. We are not aware of any published studies addressing the amount of backside wear occurring in acetabular cups made of highly cross-linked polyethylene.
Although the apparent wear rate of the highly cross-linked polyethylenes during the first few years of clinical use has been lower than that of traditional polyethylenes, the amount of reduction has been smaller than the percent reductions measured with prior hip-simulator wear testing. As shown in Table I, the percent reductions in the rate of femoral head penetration have ranged from 23%27 to 95%35, depending on which traditional polyethylene was used as a control. This difference is most likely due to the fact that the majority of the penetration of the ball into the cup that occurs during the first six months of use is due to creep deformation of the polyethylene rather than to wearing away of material27. Since the rate of creep is not markedly affected by the amount of cross-linking, the total penetration during the first one to two years of use tends to be comparable between the two types of polyethylene, even if one is wearing substantially less than the other.
Because of this, in midterm studies of the highly cross-linked polyethylenes (with mean durations of follow-up of approximately five years), investigators typically have excluded the penetration data from the early, bedding-in period (i.e., when creep is substantial) in order to obtain a more accurate measure of the actual steady-state wear rates. Consequently, the percent reduction in wear due to elevated cross-linking has tended to be greater in the midterm studies. Using a manual measurement technique, Dorr et al.32 found that the wear rate of thirty-seven Durasul acetabular liners (cross-linked with 9.5 Mrad [95 kGy]) at five years postoperatively was 55% lower than that of thirty-seven polyethylene liners that had been compression molded, packaged in an oxygen-free environment, and then gamma-sterilized (i.e., cross-linked) with the standard range of 2.5 to 4 Mrad (25 to 40 kGy). Using the computer-assisted method of Martell et al.49, D'Antonio et al.31 found, after a mean duration of follow-up of 4.9 years, a 60% reduction in the wear of fifty-six Crossfire liners (cross-linked with 10.5 Mrad [105 kGy]) compared with the wear of fifty-three polyethylene liners that had been gamma-sterilized and thereby cross-linked with 2.5 Mrad (25 kGy) in a nitrogen environment and vacuum packaged. Furthermore, D'Antonio et al. stated that there were fewer osteolytic lesions and no revisions in the Crossfire group. Very favorable midterm results (at 4.1 to 7.2 years) also were reported by Engh et al.35, who performed a randomized, prospective study comparing the clinical performance of seventy-six Marathon liners (DePuy Orthopaedics, Warsaw, Indiana), cross-linked with 5 Mrad (50 kGy), with that of ninety traditional ultra-high molecular weight polyethylene liners that had been sterilized with gas plasma and therefore were not cross-linked. Use of the method of Martell et al.49 showed the mean linear wear rate of the Marathon liners to be 95% lower than that of the non-cross-linked polyethylene liners at a mean of 5.5 ± 0.6 years postoperatively. As indicated above, when the percent reductions in wear are compared among clinical studies of different highly cross-linked polyethylenes, the type of polyethylene used for the control group must be taken into account. This is due to the fact that traditional polyethylene cups that are gamma-sterilized with the allowable range of 2.5 to 4 Mrad (25 to 40 kGy) have a corresponding amount of cross-linking that, on the basis of hip-simulator studies, would result in wear rates that were about 50% lower (for cups sterilized with 2.5 Mrad) to 75% lower (for those sterilized with 4 Mrad) than those of non-cross-linked polyethylene cups (Marathon cups sterilized with ethylene oxide or gas plasma)10. Thus, if a particular highly cross-linked polyethylene is compared with a non-cross-linked traditional polyethylene in a clinical study, it will show a greater percent reduction in wear than it would if it is compared with a gamma-sterilized (i.e., moderately cross-linked) traditional polyethylene. This is illustrated by comparing the 95% reduction in wear relative to the wear of gas-plasma-sterilized (i.e., non-cross-linked) liners reported by Engh et al.35 with the 73% reduction in the wear of Marathon liners relative to wear of gamma-sterilized (i.e., moderately cross-linked) liners reported by Bitsch et al.34. In several independent reviews of the literature, it was found that osteolysis is rare in patients in whom the polyethylene cup is wearing at a rate of less than about 0.1 mm/yr, but osteolysis becomes much more frequent and extensive as the wear rate increases substantially above this "threshold" value55-58. In three studies with a mean duration of follow-up of about five years or longer, the mean rates of wear of Crossfire, Durasul, and Marathon highly cross-linked polyethylenes (Table I) all were well below 0.1 mm/yr31,32,35. On the other hand, the osteolysis threshold of 0.1 mm/yr was established for hips with traditional polyethylene liners—i.e., those that either were not cross-linked or were moderately cross-linked during gamma sterilization. Some investigators have reported that the mean particle size is smaller with highly cross-linked polyethylene and that, in equivalent volumes, smaller particles tend to be more likely to cause osteolysis19,20,59. If that is correct, these factors could lead to the osteolysis threshold being somewhat lower for highly cross-linked polyethylenes. We are aware of only one published case report of clinically relevant osteolysis in a hip with a highly cross-linked polyethylene liner (a Longevity liner)60. However, the hip in question also had a forged-steel surface-grit-blasted femoral component that, at revision, was found to be loose at the stem-cement interface. Since the osteolysis in this hip occurred endosteally around the loosened stem, with no acetabular lysis, it is very possible that the lesions were primarily due to debris produced at the stem-cement interface rather than from the cross-linked liner. Continued close monitoring of patients with highly cross-linked polyethylene components is essential to determine if the improved wear resistance that has been observed in the midterm, as summarized here, will translate into a substantial reduction in the prevalence and severity of osteolysis after long-term follow-up.
In order to maximize wear resistance without the reduced mechanical properties associated with post-irradiation melting, second-generation highly cross-linked polyethylenes are being developed with use of alternative methods to extinguish free radicals. With one method, polyethylene that has been cross-linked with radiation is soaked in vitamin E to extinguish the residual free radicals61. Another method involves applying the cross-linking radiation in three doses, with annealing after each dose to reduce the free radicals62. As with the first-generation highly cross-linked polyethylenes, it will require years of close clinical follow-up to determine the safety and effectiveness of these recently developed cross-linking methods for reducing wear and/or osteolysis. In conclusion, polyethylene wear of total hip prostheses is affected by multiple factors, including the type of resin used, the specific steps in the manufacturing process, the diameter of the femoral head, the alignment of the acetabular and femoral components, the weight and activity level of the patient, and other factors. At the time of midterm follow-up, first-generation highly cross-linked polyethylene liners have appeared to have decreased wear rates during clinical use and corresponding reductions in the prevalence and severity of osteolysis, which may reduce the prevalence of implant loosening and the need for revision. Despite improved wear characteristics, first-generation highly cross-linked polyethylene may be more susceptible to fatigue fracture in the presence of improper acetabular alignment and/or neck-liner impingement, particularly with the thinner liners that are used with larger-diameter balls. This highlights the need for sound surgical technique. Finally, surgeons should use caution when considering the use of second-generation cross-linked polyethylenes, pending the availability of clinical data on their safety and efficacy.
  CiteULike  Connotea  Del.icio.us  Technorati What's this?
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Top
Abstract
Introduction