Investigation performed at the Anderson Orthopaedic Research Institute, Alexandria, Virginia

Transcription

1 464 COPYRIGHT 2003 BY THE JOURNAL OF BONE AND JOINT SURGERY, INCORPORATED Effect of Terminal Sterilization with Gas Plasma or Gamma Radiation on Wear of Polyethylene Liners BY ROBERT H. HOPPER JR., PHD, ANTHONY M. YOUNG, MS, KARL F. ORISHIMO, MS, AND C. ANDERSON ENGH JR., MD Investigation performed at the Anderson Orthopaedic Research Institute, Alexandria, Virginia Background: Although terminal sterilization with gamma radiation in air promotes cross-linking, which improves wear resistance, it also results in free radicals, which can oxidize and degrade the mechanical properties of polyethylene liners used for total hip arthroplasty. For this reason, non-cross-linked polyethylene components have also been sterilized with chemical surface treatments, such as gas plasma. In this study, we tested the hypothesis that conventional polyethylene liners cross-linked by sterilization with gamma radiation in air had better in vivo wear performance than non-cross-linked liners sterilized with gas plasma. Methods: We retrospectively reviewed the wear rates in a series of hips treated with a Duraloc 100 cup, a 28-mm femoral head, and an Enduron liner that had been sterilized with either gamma radiation (sixty-one hips followed for a mean of 5.2 years) or gas plasma (sixty-three hips followed for a mean of 3.9 years). The irradiated liners had been stored with access to ambient oxygen for an average of 1.0 year (range, 0.05 to 5.72 years) prior to implantation. Multiple linear regression was used to assess the effect of the sterilization method on the wear rate while accounting for the possible influence of other factors, including liner geometry, femoral head material, patient gender, cup abduction angle, and age at surgery. Results: The polyethylene liners that had been sterilized with gamma radiation in air had a significantly lower wear rate than did the gas-plasma-sterilized liners (0.097 compared with 0.19 mm/yr, p < 0.001). The sterilization method (p < 0.001) and age at surgery (p = 0.001) were the only factors that we analyzed that correlated with the wear rate. Conclusions: The in vivo wear of conventional polyethylene liners that had been sterilized with gamma radiation in air was, on the average, 50% less than that of non-cross-linked liners sterilized with gas plasma. Level of Evidence: Therapeutic study, Level III-2 (retrospective cohort study). See Intructions to Authors for a complete description of levels of evidence. With the development of highly cross-linked polyethylene components, implant manufacturers introduced material-processing treatments exclusively designed to promote polyethylene cross-linking. Prior to that time, the primary source of polyethylene cross-linking of many components appears to have been terminal sterilization with gamma radiation. (We will subsequently refer to polyethylene that was not treated to promote cross-linking prior to terminal sterilization as conventional. ) For many years, gamma radiation in air was widely used to sterilize conventional polyethylene components for total joint arthroplasty 1. During the 1990s, A commentary is available with the electronic versions of this article, on our web site ( and on our quarterly CD-ROM (call our subscription department, at , to order the CD-ROM). however, the orthopaedic industry became aware that sterilization with gamma radiation produces free radicals that can subsequently oxidize, reducing the strength and ductility of the polyethylene. In response to this possibility of oxidation, some manufacturers eliminated gamma radiation from their polyethylene processing protocol and terminally sterilized conventional polyethylene components with chemical surface treatments, such as gas plasma and ethylene oxide. Ironically, the same gamma-radiation sterilization process that rendered polyethylene vulnerable to oxidation also improved wear resistance by promoting polyethylene cross-linking 1. This study was performed to test the hypothesis that conventional polyethylene liners terminally sterilized with gamma radiation in air have better in vivo wear performance than do non-cross-linked liners sterilized with gas plasma. We antici-

2 465 pated that, in an analysis of total hip arthroplasties performed with a single cup design and head diameter and with polyethylene liners manufactured from ram-extruded bar stock, gas-plasma-sterilized liners would demonstrate significantly higher wear rates than would implants that were sterilized with gamma radiation in air. Materials and Methods or this Institutional Review Board-approved retrospective F study, we considered all hips treated with primary total hip arthroplasty at our institution between January 1994 and July While our standard practice during that time was to utilize a modular Duraloc acetabular cup (DePuy, a Johnson and Johnson Company, Warsaw, Indiana) with a 28-mm femoral head (DePuy), we transitioned from liners that had been sterilized with gamma radiation in air to those sterilized with gas plasma because of the inventory made available to us. The surgeons did not select polyethylene liners on the basis of the sterilization method. To limit the confounding influence of implantrelated factors, we restricted the study population to hips with a Duraloc 100 cup, a 28-mm femoral head, and an Enduron liner (DePuy) with a neutral or 4-mm-lateralized geometry. The study group was further restricted to hips with a liner sterilized with gamma radiation (at a nominal dose of 2.5 to 4.0 Mrad) or with gas plasma. At the time that the gamma-radiationsterilized liners were manufactured, a dose of 2.5 to 4.0 Mrad was DePuy s standard for terminal sterilization. Other inclusion criteria included radiographically stable components 2,3 ; a minimum of three years of radiographic follow-up; an immediate (nominal six-week) postoperative anteroposterior pelvic radiograph; and at least three follow-up radiographs, with the first made at least 0.8 year after the operation and with at least 0.5 year between all of the follow-up radiographs. For each hip that met the criteria for inclusion in the study, we recorded the age of the patient at the time of the operation, gender, preoperative diagnosis, activity level, and duration of radiographic follow-up (see Appendix). Activity level was graded by the physician on an ordinal scale as heavy, moderate, light, semi-sedentary, or sedentary. For each implant, we determined the femoral head material, cup abduction angle, polyethylene liner geometry, sterilization method, sterilization dosage, and duration of storage prior to implantation. The cup abduction angle was measured from the interteardrop line to the major axis of the cup face on an anteroposterior pelvic radiograph. We also calculated the total age of the polyethylene liner by adding the duration of shelf storage to the time that the liner had been in vivo. None of the liners were stored in oxygen-free packaging prior to implantation. No restriction was placed on the shelf life of the components that were included in the study. One hundred and twenty-four hips met the study inclusion criteria. A liner sterilized with gamma radiation in air was implanted in sixty-one hips (fifty-seven patients), whereas a liner sterilized with gas plasma was implanted in sixty-three hips (sixty-one patients). The mean dose of gamma radiation was 3.1 Mrad (range, 2.8 to 3.4 Mrad), and the mean shelf life of the liners sterilized with gamma radiation was 1.0 year Fig. 1 Head penetration data for a liner sterilized with gamma radiation in air demonstrates a wear rate of mm/yr based on the least-squares-regression (dashed) line. The mean wear rate for all sixty-one liners sterilized with gamma radiation in air was mm/yr. In comparison, the head penetration data for a gas-plasmasterilized liner demonstrates a wear rate of 0.18 mm/yr based on the least-squares-regression (solid) line. The mean wear rate for all sixty-three gas-plasma-sterilized liners was 0.19 mm/yr. (range, 0.05 to 5.72 years). There were no significant differences in the age at surgery, gender, preoperative diagnosis, or activity level between the sterilization groups (see Appendix). However, the gamma-radiation group included significantly more 4-mm-lateralized polyethylene liners than did the gasplasma group (85% compared with 67% of the group; p = 0.016, Pearson chi square) as well as significantly more ceramic femoral heads (25% compared with 10% of the group; p = 0.025, Pearson chi square). The average cup abduction angle was also found to be significantly greater in the gammaradiation group (40.4 compared with 36.3 ; p = 0.01, independent samples t test). Although the study population was not restricted to hips without a so-called skirted femoral head, only one of the 124 hips had such a skirted head. Two-dimensional penetration by the femoral head was determined on each pelvic anteroposterior follow-up radiograph, relative to that seen on the immediate postoperative radiograph, with validated computer-assisted techniques 4-6. For each hip, the wear vector magnitudes were plotted against the follow-up time and these data were used to calculate a wear rate on the basis of a least-squares linear regression 7. The slope of the linear regression represented the wear rate of the polyethylene (Fig. 1). Additional details regarding the linearity of the head penetration data and how we determined that the calculated head penetration rates represented wear rather than creep are presented in the Appendix. To control for the possible influence of variables other than the sterilization method on the wear rate, a multiple linear regression analysis was employed. To quantify the effect of implant usage in our analysis, we sought a continuous variable

3 466 that would represent activity level. We found a significant correlation between our ordinal physician-assessed activity-level score and the age at surgery (rho = 0.43, p < 0.001, Spearman correlation). Consequently, we used the age at surgery as a proxy for activity level. The liner geometry, head material, sterilization method, and cup abduction angle were included as independent variables in the regression analysis because of the significant differences in these parameters between the sterilization groups (see Appendix). Gender was also included on the basis of data in the literature 8,9. Dichotomous categorical variables were incorporated in the multiple linear regression analysis with use of effects coding 10. Because we suspected that the effects of the sterilization method, liner geometry, head material, gender, and cup abduction angle might be influenced by implant usage, we incorporated interaction terms between these variables and the age at surgery in the regression analysis. Independent variables with a significant predictive value were identified with use of a stepwise method of variable entry for the multiple linear regression analysis with an inclusion F probability of 0.05 and an exclusion value of The effect of the sterilization method was isolated from other independent model variables that were found to be significant with use of the best-fit regression model that accounted for the largest portion of the variance in the wear rates. To ensure that storage duration did not influence the wear rate of the liners Fig. 2 The wear rates for all 124 hips included in the study population demonstrate that the gas-plasma-sterilized liners had a higher mean wear rate than the liners sterilized with gamma radiation in air. The best-fit regression equation was given by wear rate = *(sterilization method) *(age at surgery in years), where sterilization method = 1 for gas plasma and 1 for gamma radiation in air. With use of only these terms, the regression equation accounted for 20% (r = 0.44, r 2 = 0.20, p < 0.001) of the variance in the wear rate. The dashed line depicts the model for the liners sterilized with gamma radiation based on this best-fit regression equation. The solid line corresponds to the model for the gas-plasmasterilized liners based on the same regression equation. sterilized with gamma radiation, the relationship between wear rate and storage time was evaluated. Relationships between variables were evaluated with use of the Pearson correlation for continuous data and the Spearman correlation for ordinal data. Differences between the two groups were assessed with an independent samples t test when the outcome measure was continuous and with a Mann- Whitney U test when the outcome measure was ordinal. Categorical data were compared with use of a Pearson chi-square test. Continuous data were compared with a specified value with use of a one-sample t test. Statistical analysis was performed with SPSS software (version 11.0; SPSS, Chicago, Illinois). A p value of <0.05 was considered significant. Results ithout accounting for the effect of any of the potential covariates, the mean wear rate was mm/yr for the W gamma-radiation-sterilized polyethylene and 0.19 mm/yr for the gas-plasma-sterilized components. Multiple linear regression analysis revealed only two significant independent variables: sterilization method (p < 0.001) and age at surgery (p = 0.001). With use of the overall population mean (61.0 years) for the age at surgery to compensate for the small difference in age between the groups, the best-fit multiple linear regression equation (Fig. 2) yielded an average wear rate of mm/yr for the gamma-radiation-sterilized polyethylene and an average wear rate of 0.19 mm/yr for the gas-plasma-sterilized polyethylene. With use of either the raw wear-rate data or the best-fit regression equation evaluated at the mean population age, the in vivo wear rate of the polyethylene liners sterilized with gas plasma was twice that of the liners sterilized with gamma radiation. Although there were significant differences between the sterilization groups with respect to the geometry of the polyethylene liner (p = 0.016, Pearson chi square), the femoral head material (p = 0.025, Pearson chi square), and the cup abduction angle (p = 0.01, independent samples t test), none of these factors were significant in the context of the multiple linear regression analysis (p = 0.80 for liner geometry, p = 0.18 for head material, and p = 0.48 for cup abduction angle). The effect of gender was also insignificant (p = 0.21). Because none of the interaction terms employed in the analysis were significant (p > 0.23), the slopes of the best-fit regression lines for the gamma-radiation-sterilized and gas-plasma-sterilized components were identical (Fig. 2). There was no relationship between shelf life and wear rate in either the gamma-radiation group (r = 0.17, p = 0.19, Pearson correlation) or the gas-plasma group (r = 0.07, p = 0.61, Pearson correlation). There was also no relationship between the total age of the liner and the wear rate in either the gammaradiation group (r = 0.21, p = 0.10, Pearson correlation) or the gas-plasma group (r = 0.15, p = 0.24, Pearson correlation). Discussion he sterilization method was found to be the most signifi- factor affecting the in vivo wear rate of polyethylene Tcant liners in this analysis. However, even though the sterilization

4 467 method and the age at surgery were the only significant factors that were identified in this study, wear is a multifactorial process 11 and substantial variations in wear rates can be anticipated within a patient population with the same type of implant. Several studies support the notion that terminal sterilization with gamma radiation in air substantially reduces polyethylene wear rates compared with those following chemical surface treatments, such as gas plasma and ethylene oxide, that do not induce cross-linking. In a study in which they employed radiostereometry to measure three-dimensional head penetration two years after hip arthroplasty, Digas et al. 12 reported a mean head penetration of 0.41 mm in a group of seventy-six ethylene-oxide-sterilized cups compared with 0.27 mm in a group of eighty-six cups sterilized with gamma radiation. Although several different cup designs were included in that study, this difference in head penetration was significant (p < ). In a retrospective study of 355 hips in which four different designs of cementless metal-backed modular cups had been implanted, Sychterz et al. 13 found that 118 Arthropor cups (Joint Medical Products, Stamford, Connecticut) sterilized with ethylene oxide had an average wear rate of 0.17 mm/ yr compared with an average rate of mm/yr for 237 ACS Triloc+ (DePuy), Harris-Galante (Zimmer, Warsaw, Indiana), and Duraloc (DePuy) cups sterilized with gamma radiation in air. This difference was significant (p < 0.01), and the wear rates closely approximate those for the gas-plasma and gamma-radiation-sterilized components in the current study. In a hip-simulator study, McKellop et al. 14 reported that, compared with ethylene oxide sterilization, 4.5 Mrad of gamma radiation resulted in a 54% to 64% reduction in wear. We did not find a relationship between shelf life and wear rate in our group of gamma-radiation-sterilized components (r = 0.17, p = 0.19, Pearson correlation). That group also demonstrated a significantly lower head-penetration (wear) rate despite a significantly longer shelf life and duration of follow-up. However, it should be noted that only eighteen (30%) of the gamma-radiation-sterilized liners were stored for longer than one year prior to implantation 15,16. Although polyethylene oxidation may continue in vivo, after sixteen years of follow-up Orishimo et al. 17 did not find evidence of substantially increased rates of wear of liners sterilized with gamma radiation in air. Similarly, Oonishi et al. 18 reported no in vivo deterioration of the chemical and mechanical properties of two gamma-radiationsterilized polyethylene liners retrieved at fifteen and sixteen years postoperatively. To control for the possible influence of variables other than the sterilization method, we used a multiple linear regression analysis with stepwise variable entry so that insignificant factors could be excluded from the final regression equation. The absence of a significant relationship between the head material and the wear rate (p = 0.18) is consistent with the findings of other in vivo studies 19,20. Other investigators have also reported that the cup abduction angle and the patient s gender are not significantly correlated with polyethylene wear rates. Liner geometry (neutral or lateralized by 4 mm) did not have a significant influence on wear rate (p = 0.80) in the context of our multiple linear regression analysis. While we would have preferred to have used a quantitative measure of activity level, had one been available, the literature indicates that age and activity level are correlated 26,27. In the present study, we also found a significant correlation between our ordinal physician-assessed activity-level score and the age at surgery (rho = 0.43, p < 0.001, Spearman correlation). We consequently elected to use age as a proxy for activity level in our analysis. A significant relationship between the age at surgery and wear is consistent with the findings of other investigators 24,28. We should emphasize that the results of our study are based on conventional polyethylene that was not processed to enhance cross-linking prior to terminal sterilization. Thus, we discourage readers from attempting to extrapolate our results to the new generation of highly cross-linked 14,29 polyethylene liners. Although some of the contemporary manufacturing processes employ cross-linking radiation doses that approximate historical sterilization doses, the literature indicates that a substantial amount of polyethylene cross-linking can be achieved during the subsequent heat treatment used to eliminate free radicals 14. Additional work is necessary to quantify the in vivo performance of the new generation of highly cross-linked polyethylene components. Clinical Relevance he long-term clinical ramifications of these findings remain Tto be defined, although several studies have demonstrated an increased disposition for osteolysis with higher wear rates 30,31. While osteolysis is probably a multifactorial process, in this study only 11% (seven) of the sixty-one components that had been sterilized with gamma radiation in air had a wear rate in excess of 0.2 mm/yr compared with 43% (twenty-seven) of the sixty-three gas-plasma-sterilized liners. As a consequence of these findings, we continue to monitor all patients prospectively for polyethylene wear and clinical evidence of wear-related complications. Furthermore, we believe that, when reporting wear performance, authors of outcome studies should explicitly identify the method of polyethylene sterilization. Appendix A table presenting the demographic characteristics of the study population and a detailed explanation of the authors method of calculating head penetration rates are available with the electronic versions of this article, on our web site at (go to the article citation and click on Supplementary Material ) and on our quarterly CD- ROM (call our subscription department, at , to order the CD-ROM). Robert H. Hopper Jr., PhD Anthony M. Young, MS Karl F. Orishimo, MS C. Anderson Engh Jr., MD

5 468 Anderson Orthopaedic Research Institute, P.O. Box 7088, Alexandria, VA address for R.H. Hopper: In support of their research, one or more of the authors received grants or outside funding from Inova Health Care Services, Virginia. In addition, one of the authors (C.A.E. Jr.) maintains a consulting relationship with DePuy, a Johnson and Johnson Company. No commercial entity paid or directed, or agreed to pay or direct, any benefits to any research fund, foundation, educational institution, or other charitable or nonprofit organization with which the authors are affiliated or associated. References 1. Bargmann LS, Bargmann BC, Collier JP, Currier BH, Mayor MB. Current sterilization and packaging methods for polyethylene. Clin Orthop. 1999;369: Engh CA, Massin P, Suthers KE. Roentgenographic assessment of the biologic fixation of porous-surfaced femoral components. Clin Orthop. 1990; 257: Engh CA Jr, Culpepper WJ 2nd, Engh CA. Long-term results of use of the anatomic medullary locking prosthesis in total hip arthroplasty. J Bone Joint Surg Am. 1997;79: Martell JM, Berdia S. Determination of polyethylene wear in total hip replacements with use of digital radiographs. J Bone Joint Surg Am. 1997;79: Devane PA, Bourne RB, Rorabeck CH, Hardie RM, Horne JG. Measurement of polyethylene wear in metal-backed acetabular cups. I. Three-dimensional technique. Clin Orthop. 1995;319: Sychterz CJ, Young AM, Engh CA. Effect of radiographic quality on computerassisted head penetration measurements. Clin Orthop. 2001;386: Sychterz CJ, Engh CA Jr, Yang A, Engh CA. Analysis of temporal wear patterns of porous-coated acetabular components: distinguishing between true wear and so-called bedding-in. J Bone Joint Surg Am. 1999;81: Griffith MJ, Seidenstein MK, Williams D, Charnley J. Socket wear in Charnley low friction arthroplasty of the hip. Clin Orthop. 1978;137: Joshi A, Ilchmann T, Markovic L. Socket wear in bilateral simultaneous total hip arthroplasty. J Arthroplasty. 2001;16: Pedhazur EJ. Multiple regression in behavioral research: explanation and prediction. 3rd ed. Philadelphia: Harcourt Brace; Schmalzried TP, Dorey FJ, McKellop H. The multifactorial nature of polyethylene wear in vivo. J Bone Joint Surg Am. 1998;80: Digas G, Kärrholm J, Nivbrant B, Röhrl S, Thanner J. Increased femoral head penetration using polyethylene sterilised with ethylene oxide. Read at the European Federation of National Associations of Orthopaedics and Traumatology June 1-7; Rhodes, Greece. 13. Sychterz CJ, Engh CA Jr, Yang AM, Engh CA Sr. Polyethylene wear data on porous coated acetabular components: temporal analysis of head penetration patterns to separate true wear from bedding-in. Read at the Annual Meeting of the American Academy of Orthopaedic Surgeons; 1999 Feb 4-8; Anaheim, CA. 14. McKellop H, Shen FW, Lu B, Campbell P, Salovey R. Development of an extremely wear-resistant ultra high molecular weight polyethylene for total hip replacements. J Orthop Res. 1999;17: Collier JP, Sutula LC, Currier BH, Currier JH, Wooding RE, Williams IR, Farber KB, Mayor MB. Overview of polyethylene as a bearing material: comparison of sterilization methods. Clin Orthop. 1996;333: Currier BH, Currier JH, Collier JP, Mayor MB, Scott RD. Shelf life and in vivo duration. Impacts on performance of tibial bearings. Clin Orthop. 1997;342: Orishimo KF, Hopper RH Jr, Engh CA. Long-term effects of polyethylene sterilization by gamma irradiation on in vivo wear performance. Read at the Annual Fall Meeting of the American Society of Hip and Knee Surgeons; 2001 Nov 9-11; Dallas, TX. 18. Oonishi H, Kadoya Y, Masuda S. Gamma-irradiated cross-linked polyethylene in total hip replacements analysis of retrieved sockets after long-term implantation. J Biomed Mater Res. 2001;58: Devane PA, Horne JG. Assessment of polyethylene wear in total hip replacement. Clin Orthop. 1999;369: Sychterz CJ, Engh CA Jr, Young AM, Hopper RH Jr, Engh CA. Comparison of in vivo wear between polyethylene liners articulating with ceramic and cobaltchrome femoral heads. J Bone Joint Surg Br. 2000;82: Del Schutte H Jr, Lipman AJ, Bannar SM, Livermore JT, Ilstrup D, Morrey BF. Effects of acetabular abduction on cup wear rates in total hip arthroplasty. J Arthroplasty. 1998;13: Devane PA, Robinson EJ, Bourne RB, Rorabeck CH, Nayak NN, Horne JG. Measurement of polyethylene wear in acetabular components inserted with and without cement. A randomized trial. J Bone Joint Surg Am. 1997;79: Woolson ST, Murphy MG. Wear of the polyethylene of Harris-Galante acetabular components inserted without cement. J Bone Joint Surg Am. 1995;77: Devane PA, Horne JG, Martin K, Coldham G, Krause B. Three-dimensional polyethylene wear of a press-fit titanium prosthesis. Factors influencing generation of polyethylene debris. J Arthroplasty. 1997;12: Feller JA, Kay PR, Hodgkinson JP, Wroblewski BM. Activity and socket wear in the Charnley low-friction arthroplasty. J Arthroplasty. 1994;9: Schmalzried TP, Szuszczewicz ES, Northfield MR, Akizuki KH, Frankel RE, Belcher G, Amstutz HC. Quantitative assessment of walking activity after total hip or knee replacement. J Bone Joint Surg Am. 1998;80: McClung CD, Zahiri CA, Higa JK, Amstutz HC, Schmalzried TP. Relationship between body mass index and activity in hip or knee arthroplasty patients. JOrthop Res. 2000;18: Gómez-Barrena E, Li S, Furman BS, Masri BA, Wright TM, Salvati EA. Role of polyethylene oxidation and consolidation defects in cup performance. Clin Orthop. 1998;352: Muratoglu OK, Bragdon CR, O Connor DO, Jasty M, Harris WH. A novel method of cross-linking ultra-high-molecular-weight polyethylene to improve wear, reduce oxidation, and retain mechanical properties. J Arthroplasty. 2001;16: Dowd JE, Sychterz CJ, Young AM, Engh CA. Characterization of long-term femoral-head-penetration rates. Association with and prediction of osteolysis. J Bone Joint Surg Am. 2000;82: Sochart DH. Relationship of acetabular wear to osteolysis and loosening in total hip arthroplasty. Clin Orthop. 1999;363: