Measurements of surface and subsurface damage in retrieved polyethylene tibial inserts of a...

1050-6934/12 $35.00 © 2012 by Begell House, Inc. 21

Journal of Long-Term Effects of Medical Implants, 22(1): 21–31 (2012)

Measurements of Surface and Subsurface Damage in Retrieved Polyethylene Tibial

Inserts of a Contemporary DesignMatthew G. Teeter1,*, Jan-M. Brandt2, Douglas D.R. Naudie1, Eric R. Bohm2,

Richard W. McCalden1, & David W. Holdsworth3

1Division of Orthopaedic Surgery, London Health Sciences Centre, London, Ontario, N6A 5A5, Canada; 2Concordia Joint Replacement Group, Concordia Hip and Knee Institute, Winnipeg, MB,

R2K 2M9, Canada; 3Imaging Research Laboratories, Robarts Research Institute, London, ON, N6A 5K8, Canada

*Address all correspondence to: Matthew G. Teeter; Division of Orthopaedic Surgery, London Health Sciences Centre, University Campus, 339 Windemere Road, London, Ontario, Canada N6A 5A5; Tel.: 519-685-8500; Fax: 519-931-5713; [email protected]

ABSTRACT: In the present study, surface and subsurface damage due to wear and creep in retrieved tibial inserts from the Genesis II total knee replacement (Smith & Nephew, Memphis, TN) are quantified. The utility of a number of recently validated micro-computed tomography (micro-CT) techniques for use in retrieval studies are also demonstrated. Sixteen inserts retrieved from patients after an implantation time from 0.5 to 86 months were examined. The inserts were scanned using micro-CT, and the three-dimensional surface deviations (corresponding to wear and creep) between the retrieved inserts and a reference geometry were determined. The subsurface of the inserts was also examined. Deviations within damage features were measured, and a surface deviation rate (mm/year) was calculated from the length of implantation. No subsurface fatigue damage was found. The mean deviation within the most damaged regions of the articular surface was 0.115 ± 0.064 mm medially and 0.099 ± 0.061 mm laterally (p = 0.20). The mean articular deviation rate for inserts in vivo for more than 1 year was 0.049 mm/year and was reduced to 0.026 mm/year in inserts implanted for more than 4 years. Wear and creep of the Genesis II PE insert was comparable to reported values in other total knee replacements.

KEY WORDS: total knee replacement, retrieval study, micro-computed tomography

I. INTRODUCTION

A number of studies have been published detailing the use of micro-computed tomography (micro-CT) for quantifying volumetric changes due to both wear and creep in polyethylene (PE) components from joint arthroplasty.1–7 Micro-CT enables the construction of three-dimensional maps quantifying surface deviation across the entire component geometry, in addition to measurements of volumetric changes. Surface deviation is calculated as the three-dimensional point-to-point difference between a reference geometry and the worn PE insert. For studies of components retrieved from patients during revision surgery, the sur-face deviations are measured between the geometry of a new, never implanted component

Teeter et al.

Journal of Long-Term Effects of Medical Implants

22

and a retrieved component, while in wear simulator studies the component is scanned before testing for comparison to the post-test geometry. Deviations due to wear and creep have been quantified in re-trieval studies of polyethylene acetabular liners1,3,6 and spinal discs,5 as well as in wear simulator stud-ies of tibial inserts7 and spinal discs.2 Additionally, micro-CT can be used to non-invasively examine the subsurface of polyethylene components for cracks due to subsurface fatigue and delamination.8

Recently, Heyse et al. published studies of surface damage in retrieved Genesis II PE tibial inserts (Smith & Nephew, Memphis, TN) that had been articulating against oxidized zirconium and co-balt–chrome femoral components.9,10 They observed evidence of heavy damage on only one of the PE tibial inserts, with the majority of PE tibial inserts receiving low scores in the semi-quantitative scoring system that was used to describe the surface dam-age. This semi-quantitative scoring system, as first described by Hood et al., has been widely used in re-trieval studies and enables researchers to determine the extent of damage on the component surface in addition to differentiating between various damage features, such as burnishing, pitting, abrasion, or de-

lamination.11,12 Unfortunately, this semi-quantitative scoring system does not provide a quantitative mea-surement of wear and creep that would be suitable for calculating a deviation rate, while micro-CT has been shown to be particularly useful for this purpose. Although a number of studies have been published demonstrating patient outcomes for the Genesis II implant,13–16 the deviation rate occurring for the PE tibial insert in vivo due to both wear and creep has not been reported for retrieved components.

The present study serves two purposes. First, surface and subsurface damage due to wear and creep in retrieved PE tibial inserts of the Genesis II total knee replacement are quantified and reported for the first time. Second, the utility of a number of recently validated micro-CT techniques are demon-strated for the first time in a retrieval study of total knee replacement components.

II. MATERIALS AND METHODS

A. PE Tibial Inserts

Sixteen PE tibial inserts (Genesis II, Smith & Nephew, Memphis, TN) retrieved between 1999

TABLE 1: Patient information for the retrieved components. Insert corresponds to the deviation maps in Figure 2.

Insert Sex BMI Side Time Implanted Reason for RevisionA F 28.5 Right 0.5 months InfectionB F 39.8 Right 4 months InfectionC F 39.5 Left 8 months MalpositionD F 40.8 Right 13 months InfectionE F 40.8 Right 14 months InfectionF F 29.6 Right 14 months InfectionG F 39.1 Right 17 months FractureH M 37.0 Left 18 months Infection

I F 38.4 Right 22 monthsArthrofibrosis and aseptic loosening of the patella component

J F 28.1 Left 22 months InfectionK F 30.7 Right 46 months Pain and effusionL F 33.2 Right 52 months InfectionM M 36.3 Left 53 months InfectionN F 25.9 Right 58 months InfectionO F 30.7 Right 78 months InfectionP F 24.1 Left 86 months Pain and decreased range of motion

Volume 22, Number 1, 2012

23Damage in Retrieved Tibial Inserts

inserts were brought into the software utility and co-aligned using an iterative closest points algorithm, set to converge when the root mean square aver-age distance dropped below 0.1 µm for the 1000 sample points. All differences between the two co-aligned surfaces (i.e. the three-dimensional surface deviations) were calculated continuously across the entire three-dimensional surfaces of the loaded PE tibial inserts. The software utility then generated two outputs: a new geometry created by averaging the PE tibial insert geometries loaded into the soft-ware, and maps of the three-dimensional surface deviations between the PE tibial insert geometries.

The scanned volumes of the five new, never-implanted PE tibial inserts were averaged to con-struct an idealized reference geometry to represent the unworn state of the retrieved PE tibial inserts.4 Subsequently, the 16 retrieved PE tibial inserts were each co-aligned to the reference geometry of the averaged new, never-implanted geometry, the three-dimensional surface deviations between the two geometries were calculated, and a map with these deviations was generated. In addition, aver-aging of PE tibial inserts surfaces for all left-sided inserts and all right-sided PE tibial inserts was per-formed. Surface deviation maps between the aver-aged PE tibial inserts and the reference geometry were then constructed.

Each deviation map was visualized in Para-View (Kitware Inc., Clifton Park, NY). Patterns of deviations on the articular and backside surfaces of the PE inserts were examined simultaneously under careful macroscopic inspection and using the generated deviation maps of the PE tibial insert surfaces. This procedure ensured that all patterns observed on the deviation maps were true represen-tations of the PE tibial insert surface and enabled a basic classification of the damage features to be performed. The observed deviations were clas-sified into three categories (Figure 1): category 1 deviations were referred to as diffuse regions and accounted for damage features such as burnishing and striations;19 category 2 deviations were com-prised of more focally defined regions than the dif-fuse regions and included damage features such as macroscopic pitting and abrasions (both two-body grooving and three-body indentations); category 3 deviations were comprised of damage features that

and 2010 at two institutions were examined. All PE tibial inserts were of posterior-stabilized (PS) de-sign, size 3/4 and 11 mm in nominal thickness. The PE tibial inserts were all uniformly machined from ram-extruded bar stock PE and were subsequently sterilized in an ethylene oxide environment.17 This size and thickness was selected due to the availabil-ity of five new, never implanted, PE tibial inserts of the same type (used in a previous study) for use as an unworn reference geometry. The majority of patients were female (14 of 16), with a mean BMI of 33.9 (Table 1). Most components were implanted on the right side (11 of 16) and were primarily re-vised for infection (11 of 16). The mean duration of implantation was 32 months.

B. Micro-CT Scanning and Image Reconstruction

The retrieved and new, never implanted PE tibial inserts were scanned with micro-CT in a previ-ously described manner.3,4,6,7 Each PE tibial insert was scanned with a dedicated laboratory micro-CT scanner (eXplore Vision 120, GE Healthcare, London, ON) at 50 µm isotropic voxel spacing. All images were acquired over 1200 views, with 10 frames averaged per view at an exposure time of 16 ms per frame. The x-ray tube voltage was 90 kVp with a current of 40 mA. Image reconstruction was performed at the full isotropic resolution.

The reconstructed images were analyzed with MicroView (GE Healthcare, London, ON). The PE tibial inserts were segmented from the images based on an automatic threshold determined by the soft-ware using the Otsu method.18 Isosurface rendering was performed to generate a three-dimensional sur-face using the determined threshold, at the highest quality without decimation. The PE tibial insert volume was recorded and the geometry was saved in STL format.

C. Measurement of Surface Deviations

A previously developed custom software utility was used to iteratively co-align the insert geom-etries, calculate surface deviations, and, for some instances, average together multiple PE insert ge-ometries.4 Briefly, data from two or more PE tibial

Teeter et al.


24

by the author. Six probe points, each with a radius of 1 mm, were placed within the categorized regions at the location of the greatest deviation (as deter-mined visually from the deviation maps). The mean deviation between the six points was recorded; this ensured that no local minima or maxima were used for the analysis.

D. Measurement of Subsurface Deviations

Using the MicroView software, the reconstructed scan images were examined for the presence of subsurface cracks using a previously described method.8 A threshold (to separate the insert from the surroundings) was determined automatically for each reconstructed scan by the software. Fram-ing was adjusted to aid visualization. For each in-sert, projections were visualized for each slice in the X, Y, and Z planes. The presence of subsurface cracks and their location were noted. A digital line tool within the software was used to measure the minimum and maximum crack widths, as well as distance from the articular or backside surface.

E. Statistical Analysis

The mean medial and lateral diffuse region devia-tions (category 1) were compared with a two-tailed paired t-test to assess the significance of side on wear. The overall mean articular deviation was obtained by averaging the medial and lateral de-viations. A mean deviation rate (mm/year) was then calculated for each insert by dividing the mean ar-ticular deviation by the length of implantation. De-viation rates were divided into less than 1 year of implantation, 1–4 years of implantation, and more than 4 years of implantation. Finally, the mean ar-ticular deviations were compared to the deviations from the pits and scratches using a Mann-Whitney test. A p value of less than 0.05 was considered sta-tistically significant.

III. RESULTS

The general pattern of articular surface deviations varied among the PE tibial inserts (Figure 2), al-though similar rectangular patches were noted among five of the right-sided PE tibial inserts (Fig-

FIGURE 1: The types of surface damage quantified with micro-CT. A: Category 1 deviations, referred to as diffuse regions, included burnishing and striations. B: Category 2 deviations were more focally defined regions than the diffuse regions and included macroscopic pitting and abrasions. C: Category 3 deviations were damage features that occurred during removal of the PE tibial insert from the tibial tray during revision surgery, such as osteotome marks.

occurred during removal of the PE tibial insert from the tibial tray during revision surgery. Deviations from categories 1 and 2 were examined using the software’s probe location tool, which displays the exact quantity of deviation within a region specified



ure 2E, F, I, K, and L). The majority of deviations were classified as diffuse regions from burnishing and striations (category 1), although focal damage features such as pits (category 2) were also preva-lent. Backside damage not attributable to the re-trieval process consisted of slight burnishing that was observed macroscopically on eight inserts (in-serts D, E, G, H, J, K, L, and O). These correspond-ed to a minimal change in the appearance of the PE tibial insert surface that was not reliably quanti-fied using micro-CT (Figure 3). We therefore did not include backside wear in the overall analysis. However, macroscopic agreement was observed between the insert articular surfaces and micro-CT derived deviation maps.

Within the regions of the greatest diffuse de-viations on the articular surface (corresponding primarily to burnishing damage and striations

and representative of overall three-dimensional change of wear and creep), the mean deviation was 0.115 ± 0.064 mm medially and 0.099 ± 0.061 mm laterally. No statistically significant difference was found between the medial and lateral sides (p = 0.20). The greatest deviations (0.311 mm medi-ally and 297 mm laterally) were found within the diffusely deviated regions of a PE tibial insert that had been in vivo for 86 months (Insert P). The mean deviation within the various pits was 0.078 ± 0.052 mm. There was no statistically sig-nificant difference between the deviations in the diffuse regions and the deviations within the pits (p = 0.32). The mean articular deviation rate for PE tibial inserts in vivo for more than 1 year was 0.049 mm/year, decreasing to 0.026 mm/year for PE tibial inserts implanted for at least 4 years. The deviation rate was 0.422 mm/year for those PE tibial inserts

FIGURE 2: Articular surface deviation maps for the retrieved inserts. Deviation scale is in µm. Surface deviation is defined as the difference (including wear and creep) between the retrieved insert geometry and the never-implanted reference insert geometry. Inserts with starred labels (C, H, J, M, and P) are left-sided. All other inserts are right-sided.

Teeter et al.


26

FIGURE 4: Averaged deviation maps for all left-sided inserts (A,C) and all right-sided inserts (B,D) AB are maps of the mean deviations between inserts, and CD are maps of the standard deviations between inserts. Deviation scales are in µm.

FIGURE 3: Backside surface of one retrieved insert (A) with corresponding surface deviation map (B). Some backside damage, mainly burnishing, (left arrow in A and B) was noted on the inserts. More common was damage from the retrieval process (right arrow in A and B).



were quantified within the surface deviation maps (Figure 2), the markings were not included when probing the surface deviations within the diffuse and pitted regions. Therefore, retrieval damage was deemed unlikely to considerably interfere with the calculation of the deviation rate.

IV. DISCUSSION

Overall, the pattern of deviations varied between the PE tibial inserts, although a small visual simi-larity appeared between five of them (inserts E, F, I, K and L). This similarity consisted of localized rectangular patches, more posterior and deeper me-dially than laterally (Figure 2E, F, I, K and L). Aside from diffuse deviation patterns from burnishing, pitting was also seen on 10 of 16 PE tibial inserts. The articular surface deviation rate was found to be 0.049 mm/year for those PE tibial inserts in vivo for more than 1 year, decreasing to 0.026 mm/year for PE tibial inserts in vivo for more than 4 years. The deviation rate dropped rapidly from time of im-plantation to 1 year, with most of those deviations possibly attributable to an initial creep response.20 Minimal backside damage was macroscopically ob-served, and the damage that did exist was below a level reliably quantifiable using the micro-CT tech-nique. This may be regarded as a limitation of the utilized micro-CT technique with different, new PE tibial inserts as the unworn reference geometries, as it may not be possible to detect very small (less than 0.01 mm) three-dimensional deviations. With a small sample size extending to 86 months in vivo, it was not possible to determine whether the rela-tive lack of backside damage was a design benefit of the Genesis II or simply that the retrieved PE tibial inserts were not implanted for a period long enough for such damage to become apparent. How-

in vivo for less than 1 year (n = 3). The averaged deviation maps reproduced the

majority of deviations found on the individual PE tibial inserts (Figure 4). Major deviations (such as large, diffuse patches from burnishing, or category 1) tended to be maintained in the averaging, while smaller deviations (such as pits and retrieval dam-age, or categories 2 and 3) were not. Many of these features were apparent on the maps of the standard deviations, which were calculated as a measure of the confidence in the mean deviations (Figure 4C and 4D). Overall, the deviations were mostly toward the posterior of the PE tibial insert, on the medial and lateral articular surfaces. The measure-ments of the mean deviations within the regions of greatest diffuse deviation for the averaged maps were within one standard deviation (1 SD) of the mean value calculated by averaging the measured deviations from each of the individual PE tibial in-serts used to construct the averaged deviation maps (Table 2).

No subsurface damage was found in any of the retrievals. No cracks were visible beneath any of the surfaces, nor were there any changes to the polyethylene density. In two cases (Inserts I and L) cracks were found running from the articular surface into the subsurface of the polyethylene, but they were attributable to retrieval damage (Figure 5A). Retrieval damage (category 3) was apparent on the majority of the PE tibial inserts, in some cas-es resulting in surface deviations exceeding 1 mm (Figure 5B–E). The majority of retrieval damage consisted of osteotome marks around the locking mechanism at the PE tibial insert midline and on the backside. In two cases, however, damage from the retrieval process was found on the articular sur-face, including an osteotome mark and deformation from a pliers-like tool. Although these deviations

TABLE 2: Deviations measured in the averaged maps (in µm) compared to the mean deviations calculated from the individual inserts used in constructing the averaged maps.

Inserts Medial Lateral

Average MapMean of Individual

InsertsAverage Map

Mean of Individual Inserts

Left Side 79 125 ± 124 71 106 ± 108

Right Side 99 110 ± 59 91 95 ± 68

Teeter et al.


28

ever, Azzam and Roy et al.21 found significantly lower backside damage in retrieved PE tibial inserts that had a peripheral locking mechanism and were sterilized in an ethylene oxide environment.

The averaging of the PE tibial insert geometries allowed observation and quantification of trends in deviation patterns. Agreement (within 1 SD) was found between the deviations measured in the averaged maps and the average of the individual deviations (Table 2), suggesting that the averag-ing method can reasonably reproduce similarities between the individual PE tibial inserts. Beyond examining retrieved PE tibial inserts, this averag-ing method could be particularly useful for wear simulator studies, which may have a more con-sistent pattern of damage on the PE tibial inserts, due to the contact mechanics of wear simulators.22 Retrieval damage, along with other features such as pits, was mostly removed in the averaging of the deviations. It is important to note retrieval damage when using micro-CT, as the technique calculates deviations across the entire three-dimensional ge-

ometry. Micro-CT analysis allows damage from the retrieval process to be separated from damage caused by in vivo function when calculating a sur-face deviation rate. However, particularly severe retrieval damage could conceivably interfere with the co-alignment of the new (never implanted) and retrieved component geometries.

The results of this study may be compared in a limited extent to the retrieval studies of Genesis II inserts by Heyse et al.9,10 Some potential for dif-ferences exist; Heyse and Davis et al.9 examined PE tibial inserts that articulated against both cobalt chromium alloy and oxidized zirconium femoral components using a damage scoring system, while all of the femoral components for this study were manufactured from cobalt–chromium alloy, and three-dimensional surface deviations were mea-sured. Heyse and Davis et al.9 noted generally low damage scores overall, with most damage features being burnishing, scratching, and pitting, without any delamination. This was consistent with this study, assuming a linear relationship between

FIGURE 5: Examples of retrieval damage seen in the subsurface (A) and on the micro-CT generated surface (B-D) of the inserts. Osteotome marks (arrows) were visible in the subsurface examination of one insert (A), and corresponded to deviations on the articular (B) and backside (C) deviation maps of the insert.



the three-dimensional surface deviations and the semi-quantitative damage scores.9 A low damage score for the Genesis II PE tibial inserts could be related to the use of ethylene oxide for steriliza-tion, among other factors. PE tibial inserts steril-ized with ethylene oxide have been found to have statistically significantly lower damage scores than those sterilized with gamma radiation in air or nitrogen.17,21 While Heyse and Davis et al. found greater damage laterally than medially,9 we found no statistically significant difference between the medial and lateral sides with respect to mean sur-face deviation. The surface deviation rate (due to wear and creep) for the Genesis II found in this study is on the low end of wear rates reported in the literature for other fixed and mobile total knee replacements, which range from 0.046 to 0.350 mm/year.23 These wear rates were obtained from radiographs, coordinate measuring machines, and toolmaker’s dial gauges, and thus may not be en-tirely comparable to the measurements obtained with micro-CT. The relatively low surface devia-tion rate found in this study stands in agreement with the good clinical outcomes (greater than 95% survivorship at 10 years) found in previous studies of this implant and thus may indicate low wear.14,15

Certain limitations were associated with this study. The sample size was small, as a single PE tibial insert design (PS), size (3-4) and thickness (11 mm) was studied to match a pre-existing un-worn reference geometry available in our labora-tory. However, we believe the results serve as a good demonstration of the utility of micro-CT for retrieval studies, which was a secondary goal of the study. The majority of retrievals (10 of 16) were in vivo for less than 2 years, but they did include five PE tibial inserts implanted for more than 4 years. The mean time in vivo was 32 months, 16 months longer than in a previous retrieval study of this im-plant.9 As with all retrieval studies, a bias exists be-cause only failed PE tibial inserts were examined, which may not represent the volumetric changes of well-functioning implants. The geometries of five new, never-implanted PE tibial inserts were aver-aged to construct the reference geometry used to represent the unworn state of the retrieved PE tibial inserts. As this is not the “true” original geometry of each retrieved PE tibial insert, this may introduce

a small error into the measured surface deviations. However, the method of averaging multiple new, never-implanted PE tibial inserts to construct a ref-erence geometry has been shown to be superior to the more common method of using a single never-implanted PE tibial insert, providing a representa-tion of the “true” PE tibial insert geometry to within 10 µm.4 Minimal artifact was apparent that could be due to using an improper reference geometry. Addi-tionally, the measured surface deviations included both wear and creep. As creep has been suggested to predominate over wear at shorter implant dura-tions,20 the surface deviations in the three PE tibial inserts (Inserts A, B and C) in vivo for less than 1 year can possibly be attributed to creep, and may provide some indication of the relative amount of wear versus creep in the remaining PE tibial inserts. There was a decrease in deviation rate after 1 year, further decreasing as time went on, consistent with a creep response.22 Finally, micro-CT alone cannot necessarily distinguish between the various types of damage features. Micro-CT may therefore best be used in combination with some type of visual inspection (as in this study) or scoring system to maximize insight into the wear process in retrieval studies.

V. CONCLUSION

The present study provides insight into the over-all utility of micro-CT for quantifying surface and subsurface deviations in retrieved PE tibial inserts as well as reporting the first measurement of in vivo wear and creep of the Genesis II implant. Micro-CT was found to be useful for studying the retrieved PE tibial inserts, producing data including three-dimensional surface deviation maps of the PE tibial insert surfaces, quantification of both large diffuse damage features such as burnishing and small focal damage features such as pits, averaging of PE tibial insert surface deviations, examination for potential subsurface fatigue damage, and examinations of damage that occurred during the retrieval process. In particular, the ability to construct an averaged map of surface deviations could be useful in larger-scale studies, as it would allow grouping by implant size, implantation time, and other variables. Ad-ditionally, as a non-invasive technique, the use of

Teeter et al.


30

micro-CT does not preclude any additional types of analysis such as scanning electron microcopy and detailed surface profilometry.

In conclusion, this study of retrieved PE tibial inserts revealed mean articular surface deviations (due to both wear and creep) of less than 0.115 mm. The deviation rate was 0.049 mm/year for PE tibial inserts in vivo for more than 1 year, decreasing to 0.026 mm/year for PE tibial inserts in vivo for more than 4 years. Further retrieval studies of other sizes and models of the Genesis II will be required to provide a more complete understanding of wear in this implant. Micro-CT was found to be useful in the present study of the retrieved PE tibial inserts, providing observations such as averaged surface deviation maps and evaluation of subsurface dam-age that are not widely available or possible with other wear assessment techniques.

ACKNOWLEDGMENTS

The authors wish to thank Kory Charron, Sean O’Brien, Nick Paterson, Martin Petrak, and Lynd-say Somerville for their assistance. This study was funded by a grant from the Canadian Institutes of Health Research (#MOP-89852). Smith & Neph-ew Inc. (Memphis, TN) provided the new, never implanted PE tibial inserts for this study. MGT is supported by a Frederick Banting and Charles Best Canada Graduate Scholarship and the Joint Motion Program – A CIHR Training Program in Muscu-loskeletal Health Research and Leadership. DWH holds the Dr. Sandy Kirkley Chair for Musculoskel-etal Research.

REFERENCES

1. Bowden AE, Kurtz SM, Edidin AA. Validation of a micro-CT technique for measuring volumetric wear in retrieved acetabular liners. J Biomed Mater Res B Appl Biomater. 2005;75(1):205–9.

2. Vicars R, Fisher J, Hall RM. The accuracy and precision of a micro computer tomography volumetric measurement technique for the analysis of in-vitro tested total disc replacements. Proc Inst Mech Eng H. 2009;223(3):383–8.

3. Teeter MG, Naudie DDR, Charron KD, Holdsworth

DW. Three-dimensional surface deviation maps for analysis of retrieved polyethylene acetabular liners using micro-computed tomography. J Arthroplasty. 2010;25(2):330–2.

4. Teeter MG, Naudie DDR, Milner JS, Holdsworth DW. Determination of reference geometry for polyethylene tibial insert wear analysis. J Arthroplasty. 2010;26(3):497–503.

5. Kurtz SM, Patwardhan A, MacDonald D, Ciccarelli L, van Ooij A, Lorenz M, Zindrick M, O’Leary P, Isaza J, Ross R. What is the correlation of in vivo wear and damage patterns with in vitro TDR motion response? Spine (Phila Pa 1976). 2008;33(5):481–9.

6. Teeter MG, Naudie DDR, Charron KD, Holdsworth DW. Highly cross-linked polyethylene acetabular liners retrieved four to five years after revision surgery: A report of two cases. J Mech Behav Biomed Mater. 2010;3(6):464–9.

7. Teeter MG, Naudie DDR, McErlain DD, Brandt J-M, Yuan X, MacDonald SJ, Holdsworth, DW. In vitro quantification of wear in tibial inserts using microcomputed tomography. Clin Orthop Rel Res. 2010;469(1):107–12.

8. Teeter MG, Yuan X, Naudie DD, Holdsworth DW. Technique to quantify subsurface cracks in retrieved polyethylene components using micro-CT. J Long Term Eff Med Implants. 2010;20(1):27–34.

9. Heyse TJ, Davis J, Haas SB, Chen DX, Wright TM, Laskin RS. Retrieval analysis of femoral zirconium components in total knee arthroplasty preliminary results. J Arthroplasty. 2011;26(3):445–50.

10. Heyse TJ, Chen DX, Kelly N, Boettner F, Wright TM, Haas SB. Matched-pair total knee arthroplasty retrieval analysis: Oxidized zirconium vs. CoCrMo. Knee. 2011 Dec;18(6):448–52.

11. Hood RW, Wright TM, Burstein AH. Retrieval analysis of total knee prostheses: a method and its application to 48 total condylar prostheses. J Biomed Mater Res. 1983;17(5):829–42.

12. McKellop HA. The lexicon of polyethylene wear in artificial joints. Biomaterials. 2007;28(34):5049–57.

13. Laskin RS, Davis J. Total knee replacement using the Genesis II prosthesis: a 5-year follow up study of the first 100 consecutive cases. Knee.



2005;12(3):163–7. 14. Bourne RB, Laskin RS, Guerin JS. Ten-year results

of the first 100 Genesis II total knee replacement procedures. Orthopedics. 2007;30(8 Suppl):83–5.

15. Bourne RB, McCalden RW, MacDonald SJ, Mokete L, Guerin J. Influence of patient factors on TKA outcomes at 5 to 11 years followup. Clin Orthop Relat Res. 2007;464:27–31.

16. Harato K, Bourne RB, Victor J, Snyder M, Hart J, Ries MD. Midterm comparison of posterior cruciate-retaining versus -substituting total knee arthroplasty using the Genesis II prosthesis. A multicenter prospective randomized clinical trial. Knee. 2008;15(3):217–21.

17. White SE, Paxson RD, Tanner MG, Whiteside LA. Effects of sterilization on wear in total knee arthroplasty. Clin Orthop Relat Res. 1996(331):164–71.

18. Teeter MG, Naudie DDR, McErlain DD, Brandt J-M, Yuan X, MacDonald SJ, Holdsworth, DW. In vitro quantification of wear in tibial inserts using microcomputed tomography. Clin Orthop Rel Res. 2011 Jan;469(1):107–12.

19. Wimmer MA, Andriacchi TP, Natarajan RN, Loos J,

Karlhuber M, Petermann J, Schneider E, Rosenberg, AG. A striated pattern of wear in ultrahigh-molecular-weight polyethylene components of Miller-Galante total knee arthroplasty. J Arthroplasty. 1998;13(1):8–16.

20. Muratoglu OK, Perinchief RS, Bragdon CR, O’Connor DO, Konrad R, Harris WH. Metrology to quantify wear and creep of polyethylene tibial knee inserts. Clin Orthop Relat Res. 2003(410):155–64.

21. Azzam MG, Roy ME, Whiteside LA. Second-generation locking mechanisms and ethylene oxide sterilization reduce tibial insert backside damage in total knee arthroplasty. J Arthroplasty. 2010;In press.

22. Harman MK, DesJardins J, Benson L, Banks SA, LaBerge M, Hodge WA. Comparison of polyethylene tibial insert damage from in vivo function and in vitro wear simulation. J Orthop Res. 2009;27(4):540–8.

23. Kop A, Swarts E. Quantification of polyethylene degradation in mobile bearing knees: a retrieval analysis of the anterior-posterior-glide (APG) and rotating platform (RP) low contact stress (LCS) knee. Acta Orthop. 2007;78(3):364–70.

Measurements of surface and subsurface damage in retrieved polyethylene tibial inserts of a...

Documents

Transcript of Measurements of surface and subsurface damage in retrieved polyethylene tibial inserts of a...