The Effect of Oxidative Aging on the Wear Performance of Highly Crosslinked Polyethylene Knee Inserts under Conditions of Severe Malalignment

Transcription

1 The Effect of Oxidative Aging on the Wear Performance of Highly Crosslinked Polyethylene Knee Inserts under Conditions of Severe Malalignment Juan C. Hermida, Andrew Fischler, Clifford W. Colwell Jr., and Darryl D. D Lima Shiley Center for Orthopedic Research and Education at Scripps Clinic, North Torrey Pines Road, Suite 140, La Jolla, California Received 13 September 2007; accepted 24 March 2008 Published online 3 June 2008 in Wiley InterScience ( DOI /jor ABSTRACT: First-generation crosslinked polyethylene developed for total hip arthroplasty has not gained wide acceptance for knee arthroplasty because of the increased potential for failure under high stresses and the increased risk for oxidative damage caused by free radicals. Sequentially crosslinked polyethylene (SQXL) is a second-generation crosslinked polyethylene that is reported to reduce the level of free radicals and preserve mechanical properties. Three groups of ultrahigh molecular weight polyethylene inserts were wear tested after artificial aging as per ASTM F2003: gamma-irradiated in air (GA-aged), sequentially crosslinked (SQXL-aged), and electron-beam irradiated and remelted (EBeam-aged). Inserts were tested in an AMTI knee wear simulator under malalignment conditions that were two standard deviations from the mean reported for computer-navigated and conventionally aligned techniques. GA-aged inserts delaminated by 500,000 cycles, and were severely damaged after 1 million cycles. None of the highly crosslinked inserts (SQXL-aged or EBeam-aged) delaminated or showed any signs of severe wear. Mean wear rate for GA-aged inserts was mg/million cycles. Mean wear rates for SQXL-aged and EBeam-aged inserts were significantly lower ( and , respectively). These results support the low levels of free radicals and preservation of mechanical properties reported in second-generation crosslinked ultrahigh molecular weight polyethylene. ß 2008 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. J Orthop Res 26: , 2008 Keywords: total knee arthroplasty; polyetheylene; crosslinked polyetheylene; wear; malalignment The primary causes for failure of total knee arthroplasty (TKA) are complications due to ultrahigh molecular weight polyethylene (UHMWPE) wear (25%) followed by loosening (24%), instability (21%), and infection (17.5%). 1 Crosslinking UHMWPE reduces wear rates when compared with conventional UHMWPE in the hip. 2 7 However, a major concern for the use of highly crosslinked UHMWPE in TKA is that the reduction in mechanical properties, decreased resistance to fatigue damage, and increased potential for oxidation might outweigh the benefits in wear when subjected to high contact stresses in the knee. Crosslinking of UHMWPE by radiation generates free radicals, which, when oxidized, cause chain-scission and embrittlement. In first-generation highly crosslinked polyethylenes, free radicals are quenched by thermal treatment at temperatures near or above the melting point. 8,9 Melting effectively quenches free radicals but retards the reformation of crystalline lamellae leading to a decrease in mechanical properties. 10,11 Annealing below the melting temperature may have less of an adverse effect on mechanical properties but is not considered as effective in quenching free radicals as remelting. 12,13 To address these issues, second-generation polyethylenes have been developed. Three methods to eliminate free radicals after gamma-irradiation without adverse changes to the mechanical properties are mechanical deformation, 14 vitamin-e treatment, 11 and sequential crosslinking. 15 Malalignment of TKA tibial components increases local stresses on the articular surfaces and at the bone implant interface, and can result in Correspondence to: Darryl D. D Lima (T: ; F: ; ddlima@scripps.edu) ß 2008 Orthopaedic Research Society. Published by Wiley Periodicals, Inc. increased wear and reduced implant survivorship. 22,23 Computer-assisted surgical navigation reduces implant malalignment; however, the relationship between malalignment and wear has not been adequately quantified. To achieve reasonable success in TKA, a highly crosslinked UHMWPE would have to survive stresses induced by malalignment and in vivo oxidation. This study was designed to measure wear and damage in artificially aged, sequentially crosslinked UHMWPE under severe malalignment conditions. Our hypothesis was that wear rates in inserts treated by sequential crosslinking would not be significantly affected by oxidative aging. MATERIALS AND METHODS Four groups of inserts belonging to two implant designs were tested (Table 1). GA-aged inserts were compression molded from GUR 1020 and gamma sterilized in air at 30 kgy. Sequentially crosslinked inserts (SQXL) were manufactured from compression molded GUR 1020 UHMWPE subjected to gamma irradiating to 30 kgy followed by annealing. This process was repeated two more times for a total dose of 90 kgy (X3 1, Stryker Orthopaedics, Mahwah, NJ). EBeam inserts (Prolong TM, Zimmer, Warsaw, IN) were compression molded from GUR 1050 subjected to 65 kgy of electron beam warm irradiation, remelted, and sterilized by gas plasma (as per manufacturer s data). Accelerated aging of inserts (GA-aged, SQXL-aged, EBeam-aged) was carried out in an oxygen bomb under five atmospheres at 708C for 14 days (as per ASTM F2003). 30 GA-aged and SQXL-aged inserts were manufactured in the Triathlon cruciate-retaining design. EBeam inserts were tested against corresponding NexGen 1 CR-Flex (Zimmer) femoral and tibial components. Data from a published report of computer-assisted surgical navigation versus conventional alignment techniques 31 were used to obtain alignment values for conventional intramedullary instrumentation and a surgical navigation 1585

2 1586 HERMIDA ET AL. Table 1. Wear Test Groups Group Design Implant Size Insert Thickness (mm) Total Radiation Dose (kgy) Artificial Aging Total Wear Cycles (Millions) GA-aged Triathlon CR Yes 1 SQXL Triathlon CR No 1 SQXL-aged Triathlon CR Yes 5 EBeam-aged NexGen CR-Flex Yes 5 GA-aged ¼ sterilization dose gamma irradiated in air and artificially aged; SQXL ¼ crosslinked by sequential gamma radiation; SQXL-aged ¼ crosslinked by sequential gamma radiation and artificially aged; EBeam-aged ¼ crosslinked by e-beam irradiation, remelted, and artificially aged. system. Postoperatively, the Perth CT protocol was used to measure alignment of the components relative to the anatomic axes as well as to each other. To simulate malalignment conditions, we chose varus and axial tibial rotation values that were 2 standard deviations from the mean for each alignment technique (conventional and navigation; Fig. 1 and Table 2). These conditions represent outliers; about 95% of a normally distributed population of TKA patients would be in better alignment than the conditions simulated in the present study. Six inserts from each group (N ¼ 3 per malalignment subgroup) were tested in a displacement-controlled knee wear simulator (AMTI, Watertown, MA). The ISO-recommended input waveforms (Fig. 1) 32 were used with the following modifications. The effect of varus malalignment was simulated by offsetting the axial load more medially on a tibial tray that was fully aligned with the femur in the coronal plane. The mediolateral distribution of vertical load for the appropriate varus angle was computed using the mechanical axis of the lower limb obtained from standing X-rays. 31 The malalignment of the tray in axial rotation was simulated by offsetting the tray relative to the femoral component in axial rotation. To simulate normal alignment, the ISO recommends offsetting the axial load medially to create a 60:40 distribution of medial to lateral forces. 32 The tibial tray was free to move in the mediolateral direction and free to tilt about the anteroposterior axis (pivot point 85 mm below the tray surface). Bovine serum (25%) supplemented with EDTA and sodium azide was used as a lubricant. Total protein concentration of the serum batch was 8.5 g/dl, with the albumin component being 3.2 g/dl (data provided by the supplier). Wear was measured by visual inspection and by monitoring weight loss at about 500,000 cycle intervals. To correct for fluid absorption, unloaded soak controls (N ¼ 3 per insert group) were used. The nonparametric Mann-Whitney test was used to test for significant differences in average wear rates among GA-aged, SQXL-aged, and EBeam-aged groups and between conventional and navigation alignment groups. RESULTS GA-aged inserts began delaminating under both malalignment conditions as early as 500,000 cycles. At the end of the 1-million-cycle interval, severe damage was visible on the medial compartments (Fig. 1), and wear testing was aborted. Mean wear rates in GA-aged inserts were higher in the conventional alignment Figure 1. ISO-recommended waveforms for displacement controlled knee wear simulation were used.

3 MALALIGNMENT, OXIDATIVE AGING, AND CROSSLINKED POLYETHYLENE WEAR 1587 Table 2. Malalignment Conditions Conventional Navigation Tibiofemoral varus Mediolateral load distribution 80:20 70:30 Tibiofemoral Internal rotation group relative to the navigation alignment group, but the difference was not significant (Table 3). Wear rates in SQXL inserts were significantly lower relative to GAaged inserts (p < 0.001), and no delamination was observed. None of the SQXL-aged inserts or EBeam-aged inserts showed visible delamination or signs of severe wear over 5 million cycles of testing (Fig. 2). Both SQXL-aged and EBeam-aged inserts wore significantly less than the GA-aged inserts (Fig. 3, p < 0.001). Overall wear rates for the EBeam-aged group were higher than for the SQXLaged group ( vs , p < 0.01). A small decrease in wear rates was noted in the navigation alignment group relative to the conventional alignment group of EBeam-aged inserts (p ¼ 0.05), but no significant difference was noted between the SQXL-aged alignment groups. Wear rates were calculated after correcting for weight gain due to fluid absorption, which averaged 0.71 mg/million cycles in the unloaded soak controls. DISCUSSION We sought to determine whether second-generation highly crosslinked UHMWPE is resistant to oxidation and can withstand the stresses generated by malalignment in the knee. We also sought to measure any benefit in wear rates that might be attributed to improvement in malalignment outliers by using computer-assisted surgical navigation. We therefore tested polyethylene inserts radiated under three different conditions, subjected to accelerated aging, and subjected to two malalignment conditions. Gamma sterilization and subsequent oxidation and embrittlement of UHMWPE has been linked to severe damage and delamination of retrieved polyethylene inserts. 33 Similar delamination has been reproduced Table 3. Mean (SD) Wear Rates (mg/million Cycles) Conventional Navigation p-value GA-aged SQXL SQXL-aged EBeam-aged GA-aged, gamma-irradiated in air and artificially aged; SQXL, sequentially crosslinked; SQXL-aged, sequentially crosslinked and artificially aged; Ebeam-aged, crosslinked, remelted, and aged. in vitro in aged conventional UHMWPE inserts (exposed to sterilization doses of gamma radiation) under conditions simulating gait. 34 In that study, the ISOrecommended test waveforms were modified by increasing the axial load, AP translation, and IE rotation, but the components were not malaligned. Wear rates of conventional UHMWPE inserts that had been artificially aged were substantially higher than unaged inserts and delaminated occurred within 5 million cycles of testing. Delamination has also been produced in aged, conventional UHMWPE inserts using stair climbing kinematics under conditions simulating a tight posterior cruciate ligament. 35 In our study, we simulated the effects of malalignment. The delamination produced in aged gamma-irradiated UHMWPE in air within 500,000 cycles indicate that higher stresses were generated by malalignment. Knee simulator studies of unaged, highly crosslinked inserts showed little damage or delamination when tested under conditions that caused delamination in aged, conventional polyethylene inserts. 34,35 Similarly, in our study, both unaged and aged sequentially crosslinked inserts wore substantially less under the same conditions and did not delaminate. A small difference in mean wear rates occurred between aged and unaged sequentially crosslinked inserts, likely due to the fact that the two were tested in separate experiments for different durations (aged: 5 million cycles, unaged: 1 million cycles). This suggests that the sequential crosslinking process preserved the mechanical properties to an extent sufficient to sustain the stresses generated by malalignment conditions at least two standard deviations from the mean. Wear rates were low in both types of highly crosslinked inserts. Postirradiation thermal treatment above the melting point effectively quenches free radicals, and hence, reduces the risk for oxidative damage, 8,36 as supported by the low wear rates produced in vitro in aged, highly crosslinked components 34 In our study, highly crosslinked polyethylene that had been subsequently melted (EBeam) also demonstrated low wear despite being exposed to oxidative aging. Wear rates of EBeam-aged inserts were higher than SQXL-aged inserts. One reason could be the lower net radiation dose for EBeam (65 kgy) than for SQXL (90 kgy). The rationale for a lower dose is to preserve mechanical properties. Another reason for the differences in wear rates between designs could have been different contact geometries generating different stresses under the same kinematic inputs, which was a limitation of the study. Sequential crosslinking is an alternative method of quenching free radicals and better preserves mechanical properties. 15,37 Our findings of low wear in SQXL-aged inserts confirm these reports. Either method of removing free radicals appears effective in maintaining low wear and preventing oxidative damage under the stresses generated in a malaligned TKA. Using alignment outlier values from navigation data reduced wear in GA-aged and EBeam-aged inserts

4 1588 HERMIDA ET AL. Figure 2. Inserts that were gammairradiated in air and artificially aged (A: GA-aged) delaminated and developed signs of severe wear by 1 million cycles. Shown here is a representative photograph of the conventional malalignment group. Inserts that were sequentially crosslinked and not aged (B: SQXL) did not show signs of delamination or severe wear by 1 million cycles. Inserts that that were sequentially crosslinked and artificially aged (C: SQXLaged) or inserts that were crosslinked by e-beam radiation and subsequent melting and were artificially aged (D: EBeam-aged) showed signs of burnishing, but did not delaminate or demonstrate evidence of severe wear up to 5 million cycles. relative to outliers obtained from conventional instrumentation alignment data, indicating that better alignment may be associated with a reduced risk for accelerated wear. However, no significant differences in wear rates between alignment groups were noted for the SQXL-aged inserts. Wear rates may have been too Figure 3. Inserts that were sequentially crosslinked and artificially aged (A: SQXL-aged) and inserts that were crosslinked by e- beam radiation and subsequent melting and were artificially aged (B: EBeam-aged) wore at substantially lower rates than inserts that were gamma irradiated in air and artificially aged (GA-aged). The navigation alignment group of the EBeam-aged inserts wore less than the conventional alignment group. low to be sensitive to the difference in loading conditions. This is an in vitro wear simulator, so the clinical risk cannot be quantified. Further, a direct linear correlation between malalignment and increased wear is unlikely, given the variable tolerance of different implant designs to malalignment. This study is only a wear simulation of gait kinematics under malalignment conditions and does not test wear performance under other activities. The components were artificially aged; artificial aging is an acceptable process of simulating shelf-oxidation of UHMWPE exposed to air. 30 Although differences between natural and artificial aging were reported, in general, artificial aging is more aggressive in oxidizing UHMWPE. 38 Only two values of malalignment were chosen based on clinically measured data. However, less severe malalignment would likely generate less wear. The load distribution used to reflect tibial component varus was a simplified representation of clinical varus, because we simulated altered loading rather than relative alignment. Another important surgical factor (not adequately addressed by current generation surgical navigation techniques) is laxity of the posterior cruciate ligament in cruciate-retaining designs. The resultant increased anteroposterior sliding with the potential for reverse rollback 39 can result in increased wear. However, the quantitative relationship of PCL laxity with the increase in sliding distance is highly variable. Additionally, our focus was the effect on altered alignment rather than increased sliding distance on aged crosslinked polyethylene. Wear was measured only up to 5 million cycles; however, wear rates would not likely change substantially beyond 5 million cycles. Wear rates were markedly low in both types of highly crosslinked polyethylene inserts under malaligned conditions, indicating low oxidative damage. These results are promising and support further testing including comparison with more widely used conventional

5 MALALIGNMENT, OXIDATIVE AGING, AND CROSSLINKED POLYETHYLENE WEAR 1589 polyethylene inserts. Finally, clinical trials are necessary before recommending widespread use of highly crosslinked polyethylene. ACKNOWLEDGMENTS We are grateful to Sandeep K. Chauhan, FRCS, for providing the component alignment data from his navigation study and for his advice. REFERENCES 1. Sharkey PF, Hozack WJ, Rothman RH, et al Insall Award paper. Why are total knee arthroplasties failing today? Clin Orthop Relat Res 404: McKellop HA, Shen FW, Campbell P, et al Effect of molecular weight, calcium stearate, and sterilization methods on the wear of ultra high molecular weight polyethylene acetabular cups in a hip joint simulator. J Orthop Res 17: D Lima DD, Hermida JC, Chen PC, et al Polyethylene cross-linking by two different methods reduces acetabular liner wear in a hip joint wear simulator. J Orthop Res 21: Dorr LD, Wan Z, Shahrdar C, et al Clinical performance of a Durasul highly cross-linked polyethylene acetabular liner for total hip arthroplasty at five years. J Bone Joint Surg Am 87: Heisel C, Silva M, dela Rosa MA, et al Short-term in vivo wear of cross-linked polyethylene. J Bone Joint Surg Am 86-A: Krushell RJ, Fingeroth RJ, Cushing MC Early femoral head penetration of a highly cross-linked polyethylene liner vs a conventional polyethylene liner: a case-controlled study. J. Arthroplasty 20: Manning DW, Chiang PP, Martell JM, et al In vivo comparative wear study of traditional and highly cross-linked polyethylene in total hip arthroplasty. J Arthroplasty 20: Muratoglu OK, Bragdon CR, O Connor DO, et al A novel method of cross-linking ultra-high-molecular-weight polyethylene to improve wear, reduce oxidation, and retain mechanical properties. J Arthroplasty 16: Wang A, Zeng H, Yau S-S, et al Wear, oxidation and mechanical properties of a sequentially irradiated and annealed UHMWPE in total joint replacement. J Phys D: Appl Phys 39: Pruitt LA Deformation, yielding, fracture and fatigue behavior of conventional and highly cross-linked ultra high molecular weight polyethylene. Biomaterials 26: Oral E, Wannomae KK, Hawkins N, et al Alphatocopherol-doped irradiated UHMWPE for high fatigue resistance and low wear. Biomaterials 25: Collier JP, Currier BH, Kennedy FE, et al Comparison of cross-linked polyethylene materials for orthopaedic applications. Clin Orthop Relat Res 414: Gencur SJ, Rimnac CM, Kurtz SM Fatigue crack propagation resistance of virgin and highly crosslinked, thermally treated ultra-high molecular weight polyethylene. Biomaterials 27: Kurtz SM, Mazzucco D, Rimnac CM, et al Anisotropy and oxidative resistance of highly crosslinked UHMWPE after deformation processing by solid-state ram extrusion. Biomaterials 27: Dumbleton JH, D Antonio JA, Manley MT, et al The basis for a second-generation highly cross-linked UHMWPE. Clin Orthop Relat Res 453: D Lima DD, Chen PC, Colwell CW Jr Polyethylene contact stresses, articular congruity and knee alignment. Clin Orthop Relat Res 392: Windsor RE, Scuderi GR, Moran MC, et al Mechanisms of failure of the femoral and tibial components in total knee arthroplasty. Clin Orthop Relat Res 248: Feng EL, Stulberg SD, Wixson RL Progressive subluxation and polyethylene wear in total knee replacements with flat articular surfaces. Clin Orthop Relat Res 299: D Lima DD, Hermida JC, Chen PC, et al Polyethylene wear and variations in knee kinematics. Clin Orthop Relat Res 392: Hernigou P, Deschamps G influences wear in the knee after medial unicompartmental arthroplasty. Clin Orthop Relat Res 423: Dorr LD, Boiardo RA Technical considerations in total knee arthroplasty. Clin Orthop Relat Res 205: Matsuda S, Miura H, Nagamine R, et al Changes in knee alignment after total knee arthroplasty. J Arthroplasty 14: Berend ME, Ritter MA, Meding JB, et al Tibial component failure mechanisms in total knee arthroplasty. Clin Orthop Relat Res 428: Nabeyama R, Matsuda S, Miura H, et al The accuracy of image-guided knee replacement based on computed tomography. J Bone Joint Surg Br 86: Sparmann M, Wolke B, Czupalla H, et al Positioning;1; of total knee arthroplasty with and without navigation support. A prospective, randomised study. J Bone Joint Surg Br 85: Jenny JY, Boeri C Computer-assisted implantation of total knee prostheses: a case control comparative study with classical instrumentation. Comput Aided Surg 6: Matsumoto T, Tsumura N, Kurosaka M, et al Prosthetic alignment and sizing in computer-assisted total knee arthroplasty. Int Orthop 28: Haaker RG, Stockheim M, Kamp M, et al Computerassisted navigation increases precision of component placement in total knee arthroplasty. Clin Orthop Relat Res 433: David Stulberg S How accurate is current TKR instrumentation? Clin Orthop Relat Res 416: Historical standard F Standard guide for accelerated aging of ultra-high molecular weight polyethylene. West Conshohocken, PA: ASTM International. 31. Chauhan SK, Scott RG, Breidahl W, et al Computerassisted knee arthroplasty versus a conventional jig-based technique. A randomised, prospective trial. J Bone Joint Surg Br 86: International Standards Organization International Standard, ISO ; Implants for surgery wear of total knee-joint prostheses, Part 3: loading and desplacement parameters for wear-testing machine with displacement control and corresponding environmental conditions for test. Geneva, Switzerland: International Standards Organization. 33. Collier JP, Sperling DK, Currier JH, et al Impact of gamma sterilization on clinical performance of polyethylene in the knee. J Arthroplasty 11: Muratoglu OK, Bragdon CR, Jasty M, et al Kneesimulator testing of conventional and cross-linked polyethylene tibial inserts. J Arthroplasty 19: Muratoglu OK, Bragdon CR, O Connor DO, et al Aggressive wear testing of a cross-linked polyethylene in total knee arthroplasty. Clin Orthop Relat Res 404:

6 1590 HERMIDA ET AL. 36. McKellop H, Shen FW, Lu B, et al Development of an extremely wear-resistant ultra high molecular weight polyethylene for total hip replacements. J Orthop Res 17: Yau SS, Wang A, Essner A, et al Sequential irradiation and annealing of highly crosslinked polyethylenes: resist oxidation without sacrificing physical/mechanical properties. Trans 51st Orthop Res Soc Mazzucco DC, Dumbleton J, Kurtz SM Can accelerated aqueous aging simulate in vivo oxidation of gamma-sterilized UHMWPE? J Biomed Mater Res B Appl Biomater 79: Dennis DA, Komistek RD, Colwell CW Jr, et al In vivo anteroposterior femorotibial translation of total knee arthroplasty: a multicenter analysis. Clin Orthop Relat Res 356: