This is an enhanced PDF from The Journal of Bone and Joint Surgery

The PDF of the article you requested follows this cover page.

Wear Rate of Highly Cross-Linked Polyethylene in Total Hip

Arthroplasty. A Randomized Controlled Trial
Richard W. McCalden, Steven J. MacDonald, Cecil H. Rorabeck, Robert B. Bourne, David G. Chess and Kory
D. Charron
J Bone Joint Surg Am. 2009;91:773-782. doi:10.2106/JBJS.H.00244

This information is current as of July 20, 2010

Supplementary Material

http://www.ejbjs.org/cgi/content/full/91/4/773/DC1

Reprints and Permissions

Click here to order reprints or request permission to use material from this
article, or locate the article citation on jbjs.org and click on the [Reprints and
Permissions] link.

Publisher Information

The Journal of Bone and Joint Surgery

20 Pickering Street, Needham, MA 02492-3157
www.jbjs.org

773
C OPYRIGHT 2009

BY

T HE J OURNAL

OF

B ONE

AND J OINT

S URGERY, I NCORPORATED

Wear Rate of Highly Cross-Linked Polyethylene

in Total Hip Arthroplasty
A Randomized Controlled Trial
By Richard W. McCalden, MD, MPhil, FRCS(C), Steven J. MacDonald, MD, FRCS(C), Cecil H. Rorabeck, MD, FRCS(C),
Robert B. Bourne, MD, FRCS(C), David G. Chess, MD, FRCS(C), and Kory D. Charron, Dipl.MET
Investigation performed at University Hospital, University of Western Ontario, London, Ontario, Canada

Background: Highly cross-linked polyethylene was introduced for clinical use in total hip arthroplasty with the expectation
that it would exhibit less wear when compared with conventional polyethylene. The purpose of this study was to report the
clinical and radiographic results, after a minimum of five years of follow-up, of a randomized, blinded, controlled trial
comparing a conventional polyethylene with a first-generation highly cross-linked polyethylene.
Methods: One hundred patients were enrolled in a prospective, randomized controlled study comparing highly crosslinked and conventional polyethylene acetabular liners in total hip arthroplasty. Fifty patients were in each group. At the
time of follow-up, clinical outcomes were assessed and steady-state femoral head penetration rates (after bedding-in) for
each patient were calculated with use of a validated radiographic technique. In addition, a statistical comparison of
polyethylene wear between groups was performed with use of generalized estimating equations.
Results: At a mean of 6.8 years postoperatively, there were no differences between the two polyethylene groups with regard to
the Harris hip score, Western Ontario and McMaster Universities Osteoarthritis Index (WOMAC), or Short Form-12 (SF-12) score.
The mean femoral head penetration rate in the first through fifth years was found to be significantly lower in the group treated with
the highly cross-linked polyethylene (0.003 mm/yr [95% confidence interval, 0.027]) than it was in the group treated with
conventional polyethylene (0.051 mm/yr [95% confidence interval, 0.022]) (p = 0.006). Men treated with a conventional
polyethylene liner had a significantly higher (p 0.012) femoral head penetration rate (0.081 mm/yr [95% confidence interval,
0.065]) than both men and women with a highly cross-linked liner (20.013 mm/yr [95% confidence interval, 0.074] and
0.009 mm/yr [95% confidence interval, 0.028], respectively). The general estimating equations demonstrated that the group
with a highly cross-linked polyethylene liner had a significantly lower femoral head penetration rate than the group with a
conventional polyethylene liner (p = 0.025), and a significantly higher femoral head penetration rate was demonstrated in men
with a conventional polyethylene liner when compared with both men and women with a highly cross-linked liner (p = 0.003).
Conclusions: At a minimum of five years postoperatively, the steady-state femoral head penetration rate associated
with this first-generation highly cross-linked polyethylene liner was significantly lower than that associated with a conventional polyethylene liner. Long-term follow-up is required to demonstrate the clinical benefit of this new material.
Level of Evidence: Therapeutic Level I. See Instructions to Authors for a complete description of levels of evidence.

lthough total hip replacements have been extremely

successful, their survivorship has been limited by aseptic
loosening and osteolysis secondary to wear and particu-

late polyethylene debris1-3. In response to the problem of polyethylene wear debris, efforts have been made to alter ultra-high
molecular weight polyethylene to improve its wear performance.

Disclosure: In support of their research for or preparation of this work, one or more of the authors received, in any one year, outside funding or grants in
excess of $10,000 from Zimmer, Inc., Warsaw, Indiana. In addition, 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 a commercial entity (Zimmer, Inc.).
Also, a commercial entity (Zimmer, Inc.) 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.
A commentary is available with the electronic versions of this article, on our web site (www.jbjs.org) and on our quarterly CD-ROM/DVD (call our
subscription department, at 781-449-9780, to order the CD-ROM or DVD).

J Bone Joint Surg Am. 2009;91:773-82

doi:10.2106/JBJS.H.00244

774
T H E J O U R N A L O F B O N E & J O I N T S U R G E RY J B J S . O R G
V O L U M E 9 1-A N U M B E R 4 A P R I L 2 009
d

W E A R R AT E O F H I G H LY C R O S S -L I N K E D P O LY E T H Y L E N E
T O TA L H I P A R T H R O P L A S T Y

IN

Fig. 1-A

Figs. 1-A and 1-B Individual patient regression lines from one through five years postoperatively as well as the average group regression lines
(thicker lines) for conventional polyethylene (CPE) and highly cross-linked polyethylene (XLP). The equation on each graph represents the average
regression line for the polyethylene group. The first variable (slope) represents the average head penetration rate in millimeters per year. Note the larger
variance of the regression lines on the graph for the conventional polyethylene and the clustering of the regression lines toward the horizontal on
the graph for the highly cross-linked polyethylene. Also of note is the greater prevalence of negative penetration rates on the graph for the highly
cross-linked polyethylene. The wear rates differed significantly between the polyethylene groups (p = 0.006).

In the late 1990s, techniques were developed to create a more

highly cross-linked polyethylene, and these resulted in substantial in vitro improvements in wear properties4-7. Soon
thereafter, as a result of this laboratory evidence, highly crosslinked polyethylene was widely adopted despite very little clinical
or in vivo radiographic evidence supporting its use.
In 1999, our institution began a prospective, randomized,
blinded, controlled study to compare the clinical results and
wear performance of highly cross-linked polyethylene acetabular liners with those of conventional polyethylene acetabular
liners in total hip arthroplasty. We hypothesized that there
would be no difference in clinical outcome scores between
the cohorts and that a significant reduction in femoral head
penetration would be seen in association with the highly
cross-linked polyethylene. In addition, we sought to examine
these patients at the time of midterm (five to ten-year) and longterm (more than ten-year) follow-up to demonstrate any advantages or potentially deleterious effects of highly cross-linked
polyethylene.
The purpose of this study was to evaluate the clinical,
radiographic, and wear-performance results at the time of mid-

term follow-up (at a minimum of five years) in a randomized,

blinded, controlled trial comparing conventional with highly
cross-linked polyethylene. In particular, we report the steadystate femoral head penetration rates (after bedding-in) in these
two groups of patients.
Materials and Methods
etween November 1999 and October 2001, 100 consecutive
patients were enrolled, after providing written consent, in
a randomized controlled trial that had been approved by the
hospital ethics board. To be included in the study, a patient had
to have degenerative arthritis of one hip requiring total hip
arthroplasty, a designation of A or B according to the Charnley
hip classification8, and an age between forty and seventy-nine
years. Exclusion criteria were preexisting bone disease (such as
severe osteoporosis or osteomalacia), systemic conditions affecting bone density (such as inflammatory arthritis or renal
disease), and a contralateral revision or poorly functioning total
hip replacement. Patient demographics and follow-up data are
summarized in Table I. There were no apparent differences
between the groups with regard to those parameters.

775
T H E J O U R N A L O F B O N E & J O I N T S U R G E RY J B J S . O R G
V O L U M E 9 1-A N U M B E R 4 A P R I L 2 009
d

W E A R R AT E O F H I G H LY C R O S S -L I N K E D P O LY E T H Y L E N E
T O TA L H I P A R T H R O P L A S T Y

IN

Fig. 1-B

All patients received the identical hybrid total hip replacement consisting of a cemented collared femoral stem
(VerSys; Zimmer, Warsaw, Indiana) and cementless trispiked
acetabular component (Trilogy; Zimmer). At the time of surgery,
the patients were randomized to receive either a conventional
polyethylene liner (Trilogy; Zimmer) or a highly cross-linked
polyethylene liner (Longevity; Zimmer). We defined conventional polyethylene as a polyethylene for which no formal
cross-linking or free-radical-reduction process had been used,
although there may be some degree of cross-linking in the
material as a result of the sterilization process. The conventional
polyethylene liners used in this study were made of calciumstearate-free GUR 1050 resin machined from compressionmolded sheet polyethylene. The final implant was then sterilized
with gamma radiation (25 kGy) in an inert nitrogen environment. To manufacture the highly cross-linked polyethylene liners, calcium-stearate-free GUR 1050 resin was also utilized to
create compression-molded sheets, which were then machined
into the final implant geometry. Prior to machining, the
compression-molded sheet was subjected to a process consisting of e-beam irradiation of 10 Mrad (100 kGy) to achieve the
desired level of cross-linking with melt-annealing for free-radical
reduction. The final liners were then sterilized with use of a gasplasma technique. All liners had a 28-mm inner diameter and
a 10 lip. The acetabular components had an outer diameter
ranging from 48 to 58 mm. The surgery was performed, through
a modified lateral approach, at a teaching institution by five

experienced hip surgeons, each of whom had performed more

than 100 total hip replacements per year.
The patients were followed preoperatively and postoperatively by a blinded study nurse who measured clinical performance with a number of different validated clinical outcome
TABLE I Summary of Demographics for Both Study Groups*
Conventional
Polyethylene
No. of cases
Male
Female
Duration of
follow-up (yr)

50
14 (28%)
36 (72%)
6.64 (5.73-7.55)

Highly Cross-Linked
Polyethylene
50
17 (34%)
33 (66%)
7.04 (5.62-7.55)

Age at surgery (yr)

72.58 (56-79)

72.31 (56-79)

Body mass
2
index (kg/m )

29.71 (18-48)

29.7 (22-39)

Deaths

2 (4%)

7 (14%)

No. lost to follow-up

1 at 3 yr

1 at 3 yr

*No significant difference between groups was observed for any

demographic variable. The values are given as the number of
patients with the percentage in parentheses. The values are
given as the mean with the range in parentheses.

776
T H E J O U R N A L O F B O N E & J O I N T S U R G E RY J B J S . O R G
V O L U M E 9 1-A N U M B E R 4 A P R I L 2 009
d

W E A R R AT E O F H I G H LY C R O S S -L I N K E D P O LY E T H Y L E N E
T O TA L H I P A R T H R O P L A S T Y

IN

TABLE II Comparison of Clinical Outcomes Between Polyethylene Groups*

Preop.

1 Year

5 Year

Harris hip score

Highly cross-linked
Conventional

35.64 12.97
38.96 11.35

85.07 10.32
83.40 13.10

86.03 13.08
83.11 15.41

WOMAC
Highly cross-linked
Conventional

38.26 14.86
41.29 14.75

83.04 17.19
81.61 17.65

78.00 19.42
78.12 18.20

SF-12 mental component score

Highly cross-linked
Conventional

53.38 10.73
54.40 11.70

55.79 7.38
56.01 8.55

55.24 8.01
53.36 10.13

SF-12 physical component score

Highly cross-linked
Conventional

25.70 8.21
26.10 6.41

42.20 11.37
40.86 11.11

37.24 12.16
40.00 11.78

*All one-year and five-year outcome values, except for the SF-12 mental component scores, were significantly different from the preoperative
values (all p < 0.0001). The values are given, in points, as the mean and standard deviation.

tools, including the Harris hip score9, Western Ontario and

McMaster Universities Osteoarthritis Index (WOMAC)10, and
Short Form-12 (SF-12)11. The patients were seen at six weeks,
at three months, and then at yearly intervals. Radiographs were
made at six weeks postoperatively and yearly thereafter. Two
patients, one from each group, were lost to follow-up at approximately three years postoperatively. Both patients had
been placed in a nursing home and their families refused to
have them return for follow-up. At the time of writing, nine
study patients had died of causes unrelated to the index procedure. One revision, due to late deep infection, was done at
5.5 years postoperatively. There were no other reoperations,

complications, or failures. Therefore, the study outcomes were

known for ninety-eight of the original 100 patients.
Polyethylene wear rates were measured on anteroposterior pelvic radiographs at six weeks postoperatively and at
yearly intervals with use of a validated computer-assisted edgedetection method (Hip Analysis Suite, version 8.0.3.0; University of Chicago, Chicago, Illinois)12. Radiographic analysis was
performed by a single individual with considerable expertise
with the technique who was blinded to the type of polyethylene
that had been used for each patient. Postoperative radiographs
were excluded from the analysis if they did not meet the strict
criteria previously reported in the literature12-15. Specifically,

TABLE III One-Year Through Five-Year and Two-Year Through Five-Year Penetration Rates in Polyethylene and Polyethylene/Sex Groups
Based on Individual Patients Penetration Rates
1-Yr to 5-Yr Penetration Rate (mm/yr)
Mean
Polyethylene group
Conventional*
Highly cross-linked*
Polyethylene/sex group
Conventional/male
Conventional/female
Highly cross-linked/male
Highly cross-linked/female

2-Yr to 5-Yr Penetration Rate (mm/yr)

95% Confidence Interval

Mean

95% Confidence Interval

0.051
0.003

0.029-0.073
20.024-0.030

0.057
20.029

0.028-0.085
20.060-0.003

0.081
0.041
20.013
0.009

0.016-0.146
0.019-0.063
20.087-0.062
20.019-0.037

0.085
0.049
20.037
20.026

20.003-0.173
0.018-0.079
20.133-0.059
20.058-0.007

*The femoral head penetration rates differed significantly between the conventional and highly cross-linked polyethylene groups (p = 0.006).
The femoral head penetration rate in the men treated with conventional polyethylene was significantly higher than that in the men or
women treated with highly cross-linked polyethylene (p = 0.006 and 0.012, respectively). The femoral head penetration rate in the women
treated with conventional polyethylene was not significantly different from that in the women treated with highly cross-linked polyethylene
(p = 0.11).

777
T H E J O U R N A L O F B O N E & J O I N T S U R G E RY J B J S . O R G
V O L U M E 9 1-A N U M B E R 4 A P R I L 2 009
d

W E A R R AT E O F H I G H LY C R O S S -L I N K E D P O LY E T H Y L E N E
T O TA L H I P A R T H R O P L A S T Y

IN

Fig. 2

Box plots of penetration rates by polyethylene group. The top and bottom of each box represent the 25th and 75th percentiles (lower and upper quartiles),
and the solid line in the box represents the median value (50th percentile). The dashed line represents the mean value. The upper and lower whiskers
represent the maximum and minimum values of the data set that are not outliers or extreme values. The circles represent outliers (values between 1.5 and
three times the interquartile range), and the asterisks represent extreme values (values more than three times the interquartile range). The penetration rates
differed significantly between the polyethylene groups (p = 0.006). CPE = conventional polyethylene, and XLPE = highly cross-linked polyethylene.

radiographs on which the metallic shell obscured the femoral

head such that a clear delineation between the femoral head
and the acetabular cup was not visible could not be used for the
computer analysis16. A six-week postoperative radiograph as
well as one made at a minimum of two years postoperatively
were deemed necessary for inclusion of a patient into the study
in order to properly calculate a regression line representing the
penetration rate.
We collected 94% (520) of the 556 possible radiographs
(i.e., excluding those not obtained because of the patients death
or loss to follow-up). Twenty-six additional radiographs were
eliminated from the analysis because of poor radiographic
quality or failure to meet the selection criteria; this left 494
radiographs to be analyzed with the software. Evaluation with
Hip Analysis Suite12,17 resulted in elimination of additional ra-

diographs from the final analysis, so that 78% (385) of the 494
analyzed radiographs were included in the final analysis. Every
patient included in the analysis had a minimum of three postoperative radiographs available. The average completeness of the
radiographic collection (including the postoperative radiograph
and those made at each of the five years of follow-up) per patient
included in the final analysis was 81% (range, 67% to 100%).
A linear regression analysis was performed with use of
the linear penetration measurements in each hip over time to
yield a specific steady-state penetration rate for each hip18-20.
The steady-state penetration rate for each polyethylene group
(highly cross-linked and conventional polyethylene) was calculated as the average of the individual patients penetration
rates within that group. Analysis was performed by using the
six-week postoperative radiograph as the baseline and fitting

778
T H E J O U R N A L O F B O N E & J O I N T S U R G E RY J B J S . O R G
V O L U M E 9 1-A N U M B E R 4 A P R I L 2 009
d

W E A R R AT E O F H I G H LY C R O S S -L I N K E D P O LY E T H Y L E N E
T O TA L H I P A R T H R O P L A S T Y

IN

Fig. 3

Box plots of penetration rates by polyethylene/sex group. The top and bottom of each box represent the 25th and 75th percentiles (lower and upper
quartiles), and the solid line in the box represents the median value (50th percentile). The dashed line represents the mean value. The upper and lower
whiskers represent the maximum and minimum values of the data set that are not outliers or extreme values. The circles represent outliers (values
between 1.5 and three times the interquartile range), and the asterisk represents an extreme value (value more than three times the interquartile
range). The penetration rate was significantly higher in the men treated with conventional polyethylene (M-CPE) than it was in either the men treated
with highly cross-linked polyethylene (M-XLPE) (p = 0.006) or the women treated with highly cross-linked polyethylene (F-XLPE) (p = 0.012). Interestingly, the penetration rate in the women treated with highly cross-linked polyethylene was not significantly different from that in the women treated
with conventional polyethylene (F-CPE) (p = 0.11). The penetration rates were not found to differ significantly between the men and women treated with
conventional polyethylene (p = 0.154).

regression lines through the one through five-year follow-up

points (the steady-state wear rate beginning at one year) and
through the two through five-year follow-up points (the steadystate wear rate beginning at two years). In addition, an analysis
based on the sex of the patient was performed to determine if
there were any sex-related differences between the cohorts.
We attempted to identify femoral and acetabular osteolytic lesions by comparing the most recent follow-up anteroposterior pelvic radiograph with the perioperative views. The
location of any femoral osteolytic lesion was classified ac-

cording to the zones of Gruen et al.21 and the location of any

osteolytic lesion in the pelvis was classified according to the
zones defined by DeLee and Charnley 22. Osteolytic lesions were
noted to be present or absent, but their size was not measured.
Statistical Methods
The sample size was determined on the basis of an overall alpha
error (two-sided) of 0.05 and a statistical power of 80% (beta
error = 0.20). As our primary outcome measure was polyethylene wear, we assumed a conservative decrease of 20% in the

779
T H E J O U R N A L O F B O N E & J O I N T S U R G E RY J B J S . O R G
V O L U M E 9 1-A N U M B E R 4 A P R I L 2 009
d

femoral head penetration rate in the group with the highly

cross-linked polyethylene compared with that in the control
group. This gave us a sample size of thirty-six patients per group,
which was increased to fifty patients to allow for withdrawal and
losses that would occur over the period of follow-up.
The demographic variables and clinical outcome measures
were all continuous variables, and the statistical comparison of
the polyethylene groups was carried out with the one-sample
Kolmogorov-Smirnov test for normal distribution and subsequently with the independent-samples t test. Comparisons
between preoperative and latest outcome measures for each
group were performed with the paired-samples t test. A p value
of 0.05 was considered to be significant for all statistical tests.
To compare the steady-state femoral head penetration
rate between groups, the linear regressions of the patients within
each polyethylene group or each group categorized according to
both the type of polyethylene and the sex of the patient
(polyethylene/sex group) were tested with use of the onesample Kolmogorov-Smirnov test for normal distribution.
Subsequently, the comparison of the polyethylene groups was
carried out with the independent-samples t test and the comparison of the polyethylene/sex groups was carried out with
one-way analysis of variance with least significant difference as
the post hoc method. In addition, we employed general linear
regression with the generalized estimating equation approach
to better analyze longitudinal data that consist of repeated
observations23.
Source of Funding
Financial support for this study was provided by Zimmer, Inc.
(Warsaw, Indiana). The funding was used to support the salary
of a research nurse who enrolled patients and gathered preoperative and postoperative clinical outcome data. In addition,
the funding supported the salary of a research technician who
performed the radiographic image analysis.
Results
significant clinical improvement was found in both groups
when preoperative outcome scores were compared with
postoperative scores (Table II). However, there was no difference in any of the clinical outcome scores between the groups,
which had virtually identical Harris hip, WOMAC, and SF-12
scores (Table II).
On radiographic analysis, there was a significant difference in the steady-state femoral head penetration rate between
the group treated with the highly cross-linked polyethylene and
that treated with the conventional polyethylene (p = 0.006)
(Figs. 1 and 2 and Table III). The steady-state penetration rate
was found to be extremely low in the highly cross-linked
polyethylene group on the basis of both one through five-year
and two through five-year baseline linear regression data
(mean, 0.003 mm/yr [95% confidence interval, 0.027] and
20.029 mm/yr [95% confidence interval, 0.032], respectively) (Table III). The sex-based analysis, with use of both one
through five-year and two through five-year baseline regression data, showed a higher femoral head penetration rate in the

W E A R R AT E O F H I G H LY C R O S S -L I N K E D P O LY E T H Y L E N E
T O TA L H I P A R T H R O P L A S T Y

IN

men treated with the conventional polyethylene (mean, 0.081

mm/yr [95% confidence interval, 0.065]) compared with
both the men and the women treated with the highly crosslinked polyethylene. A summary of these sex-related results is
provided in Table III and Figure 3. Box plots of head penetration data, based on the polyethylene type and the sex of the
patient, are provided to illustrate any differences in the distribution and the number of outliers within the groups (Figs. 2
and 3). There was no evidence of osteolysis on radiographic
follow-up.
The generalized estimating equations revealed a significant interaction between the effects of the polyethylene type and
the sex of the patient on femoral head penetration. Generalized
estimating equations demonstrated that cross-linked polyethylene had a significantly lower wear rate than conventional
polyethylene (p = 0.025). The effect of time (a reflection of
the wear rate) was significant. Thus, at the initial (one-year)
follow-up evaluation, the polyethylene groups were not significantly different (p = 0.219), but there was a highly significant
difference between the groups at the later follow-up points (p =
0.009).
The generalized estimating equation analysis of the effect
of the sex of the patient on femoral head penetration rates
revealed a significant difference among the four polyethylene
groups (men and women treated with conventional polyethylene and men and women treated with highly cross-linked
polyethylene). In particular, a significant difference was found
between men and women treated with conventional polyethylene (p = 0.014) but not between men and women treated
with cross-linked polyethylene (p = 0.604). In addition, the
polyethylene type had a strong effect in the male cohort but
not in the female cohortthat is, the femoral head penetration rate in the male group treated with the highly crosslinked polyethylene was significantly lower than that in the
male group treated with the conventional polyethylene (p =
0.003), but no significant difference was found between the
female group treated with the conventional polyethylene and the
female group treated with the highly cross-linked polyethylene
(p = 0.091).
Discussion
his randomized controlled trial demonstrated that, after a
minimum of five years of follow-up, there were no clinical
differences between patients treated with an acetabular liner
made of conventional polyethylene and those treated with a
cross-linked polyethylene liner. This finding was not unexpected
as it is likely that the clinical benefits of highly cross-linked
polyethylene, if they exist, would not be apparent until ten years
postoperatively and beyond. Our findings are similar to those of
other studies comparing the early and midterm clinical results
between patients treated with highly cross-linked polyethylene
and those treated with conventional polyethylene18,24-29. We did
not find any significant difference between cohorts with regard
to radiographic changes or the presence of osteolysis. Once
again, it is unlikely that any difference, if one exists, would be
seen at the time of midterm follow-up, especially since both

780
T H E J O U R N A L O F B O N E & J O I N T S U R G E RY J B J S . O R G
V O L U M E 9 1-A N U M B E R 4 A P R I L 2 009
d

cohorts had average wear rates that were well below the socalled osteolysis threshold of 0.1 mm/yr as proposed by
Dumbleton et al.30. Interestingly, Leung et al.31 recently demonstrated that patients treated with a highly cross-linked
polyethylene acetabular liner had a decreased incidence and
volume of pelvic osteolysis, as shown by computed tomography, compared with patients treated with a conventional liner.
In our clinical trial, steady-state femoral head penetration rates were significantly lower in the group treated with the
highly cross-linked polyethylene than they were in the group
treated with the conventional polyethylene. The steady-state
head penetration rate in the highly cross-linked polyethylene
group was 0.003 mm/yr (95% confidence interval, 20.024 to
0.030), representing a reduction in head penetration of >95%
as compared with the conventional polyethylene group. This
very low wear rate is likely a result of the substantial number of
negative wear rates seen in the highly cross-linked polyethylene
group, as indicated by the negative lower bound of the 95%
confidence interval. The low wear rate and the prevalence of
negative wear rates compare well with the findings in a number
of recent series in which the steady-state wear of highly crosslinked polyethylene was examined13,14,18,24,32-34. It is important to
note that while the present study showed a trend suggesting
decreased wear of highly cross-linked polyethylene at two to
three years postoperatively, it was only after five years of followup, when steady-state rates could be calculated, that we were
able to detect a difference between the cohorts at a significance
level below 0.05. This finding is consistent with the fact that a
number of earlier investigators who reported on total head
penetration, and therefore did not take into account the bedding-in process, generally observed little improvement with
highly cross-linked polyethylene as compared with conventional polyethylene25-28,35-38. Our study supports the concept
that more than five years in vivo is required to show the true
difference between the mean wear rates of these materials39.
We believe that the use of generalized estimating equations is the most appropriate method of analysis for this type of
study 23. We contend that it may be inappropriate to represent
the wear rate by creating a regression line for the entire polyethylene group (containing all data points at various time intervals for all patients), as was done in previous studies13,14,16,17,40,
because all data points within the group are not independent of
one another23,41-44. Similarly, although the method of averaging
individual patient regressions (i.e., the regression lines representing the wear rates for each patient)18,24 is a reasonable way
to provide a visual comparison between groups, the method
may be problematic statistically as it does not account for each
patients five data points (i.e., one for each year of follow-up)
but instead applies statistical testing of the interpolated line of
best fit41,43. In comparison, statistical analysis with generalized
estimating equations accounts for the potential dependence
between the related observations23,42,44. This issue is applicable to our data since repeated measurements were performed
for each patient at various time intervals. A major strength
of generalized estimating equations as compared with other
repeated-measures approaches, such as analysis of variance, is

W E A R R AT E O F H I G H LY C R O S S -L I N K E D P O LY E T H Y L E N E
T O TA L H I P A R T H R O P L A S T Y

IN

that the generalized estimating equation model allows inclusion of several covariates (in our case, sex, polyethylene group,
and time) whereas analysis of variance analysis allows only one
explanatory variable. Therefore, the generalized estimating
equation approach was used to determine if there were statistical differences between polyethylene groups and between
polyethylene/sex groups.
To our knowledge, we are the first to examine the effect of
the sex of the patient on the steady-state rate of femoral head
penetration into highly cross-linked polyethylene. All analysis
methods demonstrated significantly higher wear rates in male
patients who had conventional polyethylene than in either male
or female patients treated with highly cross-linked polyethylene.
These findings are in agreement with the previous observations
by Schmalzried et al.45, who demonstrated that polyethylene
wear was related to use and was higher in male patients. While
our data support this concept, with the relatively small numbers and the limitations of the edge-detection technique,
one might still question the validity of this finding. A larger
study with utilization of more precise tools, such as radiostereometric analysis, is probably required to confirm the
effect of the patients sex on wear of these new highly crosslinked polymers.
While this prospective, randomized, controlled trial provides Level-I evidence to support the concept that first-generation
highly cross-linked polyethylene is associated with decreased
femoral head penetration at the time of midterm follow-up, the
study had several potential limitations. First, the numbers of
patients were not large, and, although they were sufficient to
demonstrate a clear difference in head penetration rates, the
study was likely underpowered to show more subtle differences
in clinical performance or the prevalence of osteolysis. Also, the
addition of more sophisticated tools, such as computed tomography, is probably required to accurately assess osteolysis.
Of note, the decision to use a cemented stem for the femoral
reconstruction led to a tendency for older patients (those with
Dorr type-B or C femoral bone46) to be enrolled in the study,
which limited our follow-up (nine deaths occurred prior to the
five-year follow-up point). In addition, we had no detailed
activity scores with which to demonstrate the potential lower
activity of these elderly patients, which may in turn have influenced their wear rates45.
Our decision to use edge-detection techniques as our
primary wear-measurement tool is potentially problematic.
There remains little doubt that the steady-state wear rate of
highly cross-linked polyethylene is below the limit of detection
of this technique. This is evidenced by the substantial number
of patients who demonstrated negative wear values. This phenomenon of negative wear has been noted by several authors
and there is no clear consensus on how to deal with it13,14,18,39.
Lastly, because of image-quality issues, only approximately
78% of all available radiographs could be used for the analysis.
While this is potentially worrisome, it is typical of the inclusion
and exclusion rates in other investigations involving use of this
technique12-15. Future work involving the prospective evaluation of these low-wear materials probably requires the use of

781
T H E J O U R N A L O F B O N E & J O I N T S U R G E RY J B J S . O R G
V O L U M E 9 1-A N U M B E R 4 A P R I L 2 009
d

radiostereometric analysis, the accuracy and precision of which

exceed those of conventional edge-detection techniques32,39.
In summary, this prospective, randomized, controlled
trial, which had a 98% rate of clinical follow-up and was performed at a single center, demonstrated, after the longest average duration of follow-up reported in this type of study in the
literature, a clear decrease in femoral head penetration with the
use of this first-generation highly cross-linked polyethylene.
Our data confirm our initial hypothesis and strongly support
the earlier laboratory and in vitro work4-7 that predicted a
marked improvement in wear with this highly cross-linked
polymer. Longer follow-up is required to determine if this will
translate into improved clinical performance and longevity of
these implants. To confound the issue further, because of concerns about ongoing oxidation47-50 and/or mechanical changes51-54,
several manufacturers have now produced second-generation
highly cross-linked polyethylene, which is already in clinical
use55,56. These new materials should undergo the same rigorous
in vivo testing to confirm their benefit. n

W E A R R AT E O F H I G H LY C R O S S -L I N K E D P O LY E T H Y L E N E
T O TA L H I P A R T H R O P L A S T Y

IN

NOTE: The authors thank Dr. John Martell for his help and guidance in the analysis of the head
penetration data and Dr. Jan Brandt for his guidance regarding the interpretation of the individual
patient regression analyses and correlated statistics. They also acknowledge the Biostatistics
Consulting Unit at the University of Western Ontario for conducting the advanced statistical
analyses utilized in this study.

Richard W. McCalden, MD, MPhil, FRCS(C)

Steven J. MacDonald, MD, FRCS(C)
Cecil H. Rorabeck, MD, FRCS(C)
Robert B. Bourne, MD, FRCS(C)
Kory D. Charron, Dipl.MET
Division of Orthopaedic Surgery, London Health
Sciences Centre, University Campus, 339 Windermere Road,
London, ON N6A 5A5, Canada. E-mail address for R.W. McCalden:
richard.mccalden@lhsc.on.ca
David G. Chess, MD, FRCS(C)
The Hand and Upper Limb Centre,
St. Josephs Health Centre, 268 Grosvenor Street, Suite D2-167, London,
ON N6A 4L6, Canada

References
1. Harris WH. The problem is osteolysis. Clin Orthop Relat Res. 1995;311:46-53.
2. Sochart DH. Relationship of acetabular wear to osteolysis and loosening in total
hip arthroplasty. Clin Orthop Relat Res. 1999;363:135-50.
3. Willert HG, Bertram H, Buchhorn GH. Osteolysis in alloarthroplasty of the hip.
The role of ultra-high molecular weight polyethylene wear particles. Clin Orthop
Relat Res. 1990;258:95-107.
4. McKellop H, Shen FW, DiMaio W, Lancaster JG. Wear of gamma-crosslinked
polyethylene acetabular cups against roughened femoral balls. Clin Orthop Relat
Res. 1999;369:73-82.

15. Bragdon CR, Kwon YM, Geller JA, Greene ME, Freiberg AA, Harris WH, Malchau
H. Minimum 6-year followup of highly cross-linked polyethylene in THA. Clin Orthop
Relat Res. 2007;465:122-7.
16. Bragdon CR, Martell JM, Greene ME, Estok DM 2nd, Thanner J, Karrholm J,
Harris WH, Malchau H. Comparison of femoral head penetration using RSA and the
Martell method. Clin Orthop Relat Res. 2006;448:52-7.
17. Martell JM, Berkson E, Berger R, Jacobs J. Comparison of two and threedimensional computerized polyethylene wear analysis after total hip arthroplasty.
J Bone Joint Surg Am. 2003;85:1111-7.

5. 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;17:157-67.

18. Engh CA Jr, Stepniewski AS, Ginn SD, Beykirch SE, Sychterz-Terefenko CJ,
Hopper RH Jr, Engh CA. A randomized prospective evaluation of outcomes after
total hip arthroplasty using cross-linked Marathon and non-cross-linked Enduron
polyethylene liners. J Arthroplasty. 2006;21(6 Suppl 2):17-25.

6. McKellop H, Shen FW, Lu B, Campbell P, Salovey R. Effect of sterilization

method and other modifications on the wear resistance of acetabular cups made of
ultra-high molecular weight polyethylene. A hip-simulator study. J Bone Joint Surg
Am. 2000;82:1708-25.

19. Sychterz CJ, Engh CA Jr, Shah N, Engh CA Sr. Radiographic evaluation of
penetration by the femoral head into the polyethylene liner over time. J Bone Joint
Surg Am. 1997;79:1040-6.

7. Muratoglu OK, Bragdon CR, OConnor 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:
149-60.
8. Charnley J. The long-term results of low-friction arthroplasty of the hip performed
as a primary intervention. J Bone Joint Surg Br. 1972;54:61-76.
9. Harris WH. Traumatic arthritis of the hip after dislocation and acetabular
fractures: treatment by mold arthroplasty. An end-result study using a new method
of result evaluation. J Bone Joint Surg Am. 1969;51:737-55.
10. Bellamy N, Buchanan WW, Goldsmith CH, Campbell J, Stitt LW. Validation
study of WOMAC: a health status instrument for measuring clinically important
patient relevant outcomes to antirheumatic drug therapy in patients with
osteoarthritis of the hip or knee. J Rheumatol. 1988;15:1833-40.
11. Ware JE, Kosinski M, Keller SD. SF-12: how to score the SF-12 physical and
mental health summary scales. 3rd ed. Lincoln, RI: Quality Metric; 1998.
12. Martell JM, Berdia S. Determination of polyethylene wear in total hip
replacements with use of digital radiographs. J Bone Joint Surg Am.
1997;79:1635-41.
13. Bragdon CR, Barrett S, Martell JM, Greene ME, Malchau H, Harris WH.
Steady-state penetration rates of electron beam-irradiated, highly cross-linked
polyethylene at an average 45-month follow-up. J Arthroplasty. 2006;21:
935-43.
14. Manning DW, Chiang PP, Martell JM, Galante JO, Harris WH. In vivo
comparative wear study of traditional and highly cross-linked polyethylene in total
hip arthroplasty. J Arthroplasty. 2005;20:880-6.

20. Clarke IC, Johnson S, Phipatanakul W, Good V. Effects of hip-loading input on

simulated wear of Al2O3-PTFE materials. In: Rigney DA, Bayer RG. WOM: 13th International Conference on Wear of Materials. Oxford: Elsevier Science; 2001. p 159-66.
21. Gruen TA, McNeice GM, Amstutz HC. Modes of failure of cemented
stem-type femoral components: a radiographic analysis of loosening. Clin Orthop
Relat Res. 1979;141:17-27.
22. DeLee JG, Charnley J. Radiological demarcation of cemented sockets in total
hip replacement. Clin Orthop Relat Res. 1976;121:20-32.
23. Zeger SL, Liang KY. Longitudinal data analysis for discrete and continuous
outcomes. Biometrics. 1986;42:121-30.
24. DAntonio JA, Manley MT, Capello WN, Bierbaum BE, Ramakrishnan R,
Naughton M, Sutton K. Five-year experience with Crossfire highly cross-linked
polyethylene. Clin Orthop Relat Res. 2005;441:143-50.
25. Digas G, Karrholm J, Thanner J, Malchau H, Herberts P. Highly cross-linked
polyethylene in cemented THA: randomized study of 61 hips. Clin Orthop Relat Res.
2003;417:126-38.
26. Digas G, Karrholm J, Thanner J, Malchau H, Herberts P. Highly cross-linked
polyethylene in total hip arthroplasty: randomized evaluation of penetration rate in
cemented and uncemented sockets using radiostereometric analysis. Clin Orthop
Relat Res. 2004;429:6-16.
27. Dorr LD, Wan Z, Shahrdar C, Sirianni L, Boutary M, Yun A. Clinical performance
of a Durasul highly cross-linked polyethylene acetabular liner for total hip
arthroplasty at five years. J Bone Joint Surg Am. 2005;87:1816-21.
28. Triclot P, Grosjean G, El Masri F, Courpied JP, Hamadouche M. A comparison
of the penetration rate of two polyethylene acetabular liners of different levels of

782
T H E J O U R N A L O F B O N E & J O I N T S U R G E RY J B J S . O R G
V O L U M E 9 1-A N U M B E R 4 A P R I L 2 009
d

W E A R R AT E O F H I G H LY C R O S S -L I N K E D P O LY E T H Y L E N E
T O TA L H I P A R T H R O P L A S T Y

IN

cross-linking. A prospective randomised trial. J Bone Joint Surg Br. 2007;89:

1439-45.

43. Zar JH. Biostatistical analysis. 4th ed. Upper Saddle River, NJ: Prentice Hall;
1999. Simple linear regression; p 324-59.

29. Jacobs CA, Christensen CP, Greenwald AS, McKellop H. Clinical performance
of highly cross-linked polyethylenes in total hip arthroplasty. J Bone Joint Surg Am.
2007;89:2779-86. Erratum in: J Bone Joint Surg Am. 2008;90:628.

44. Diggle P, Heagarty P, Liang KY, Zeger SL. Analysis of longitudinal data. New
York: Oxford University Press; 1994.

30. Dumbleton JH, Manley MT, Edidin AA. A literature review of the association
between wear rate and osteolysis in total hip arthroplasty. J Arthroplasty.
2002;17:649-61.
31. Leung SB, Egawa H, Stepniewski A, Beykirch S, Engh CA Jr, Engh CA Sr.
Incidence and volume of pelvic osteolysis at early follow-up with highly cross-linked
and noncross-linked polyethylene. J Arthroplasty. 2007;22(6 Suppl 2):134-9.
32. Bragdon CR, Greene ME, Freiberg AA, Harris WH, Malchau H. Radiostereometric analysis comparison of wear of highly cross-linked polyethylene
against 36- vs 28-mm femoral heads. J Arthroplasty. 2007;22(6 Suppl 2):125-9.

45. Schmalzried TP, Shepherd EF, Dorey FJ, Jackson WO, dela Rosa M, Favae F,
McKellop HA, McClung CD, Martell J, Moreland JR, Amstutz HC. Wear is a function
of use, not time. Clin Orthop Relat Res. 2000;381:36-46.
46. Dorr LD, Absatz M, Gruen TA, Saberi MT, Doerzbacher JF. Anatomic Porous
Replacement hip arthroplasty: first 100 consecutive cases. Semin Arthroplasty.
1990;1:77-86.
47. Muratoglu OK, Wannomae K, Christensen S, Rubash HE, Harris WH. Ex vivo
wear of conventional and cross-linked polyethylene acetabular liners. Clin Orthop
Relat Res. 2005;438:158-64.

33. Geller JA, Malchau H, Bragdon C, Greene M, Harris WH, Freiberg AA. Large
diameter femoral heads on highly cross-linked polyethylene: minimum 3-year results. Clin Orthop Relat Res. 2006;447:53-9.

48. Wannomae KK, Bhattacharyya S, Freiberg A, Estok D, Harris WH, Muratoglu O.

In vivo oxidation of retrieved cross-linked ultra-high-molecular-weight polyethylene
acetabular components with residual free radicals. J Arthroplasty. 2006;21:
1005-11.

34. Glyn-Jones S, McLardy-Smith P, Gill HS, Murray DW. The creep and wear of
highly cross-linked polyethylene: a three-year randomised, controlled trial using
radiostereometric analysis. J Bone Joint Surg Br. 2008;90:556-61.

49. Wannomae KK, Christensen SD, Freiberg AA, Bhattacharyya S, Harris WH,
Muratoglu OK. The effect of real-time aging on the oxidation and wear of highly
cross-linked UHMWPE acetabular liners. Biomaterials. 2006;27:1980-7.

35. Heisel C, Silva M, dela Rosa MA, Schmalzried TP. Short-term in vivo wear of
cross-linked polyethylene. J Bone Joint Surg Am. 2004;86:748-51.

50. Willie BM, Bloebaum RD, Ashrafi S, Dearden C, Steffensen T, Hofmann AA.
Oxidative degradation in highly cross-linked and conventional polyethylene after
2 years of real-time shelf aging. Biomaterials. 2006;27:2275-84.

36. Hopper RH Jr, Young AM, Orishimo KF, McAuley JP. Correlation between early
and late wear rates in total hip arthroplasty with application to the performance of
Marathon cross-linked polyethylene liners. J Arthroplasty. 2003;18(7 Suppl 1):60-7.
37. Krushell RJ, Fingeroth RJ, Cushing MC. Early femoral head penetration of a
highly cross-linked polyethylene liner vs a conventional polyethylene liner: a casecontrolled study. J Arthroplasty. 2005;20(7 Suppl 3):73-6.
38. Rohrl S, Nivbrant B, Mingguo L, Hewitt B. In vivo wear and migration of highly
cross-linked polyethylene cups: a radiostereometry analysis study. J Arthroplasty.
2005;20:409-13.
39. McCalden RW, Naudie DD, Yuan X, Bourne RB. Radiographic methods for the
assessment of polyethylene wear after total hip arthroplasty. J Bone Joint Surg Am.
2005;87:2323-34.
40. Martell JM, Verner JJ, Incavo SJ. Clinical performance of a highly cross-linked
polyethylene at two years in total hip arthroplasty: a randomized prospective trial.
J Arthroplasty. 2003;18(7 Suppl 1):55-9.
41. Kleinbaum DG, Kupper LL, Muller KE, Nizam A, editors. Applied regression
analysis and other multivariable methods. Pacific Grove, CA: Duxbury Press; 1998.
Straight-line regression analysis; p 39-87.
42. Kleinbaum DG, Kupper LL, Muller KE, Nizam A, editors. Applied regression
analysis and other multivariable methods. Pacific Grove, CA: Duxbury Press; 1998.
Analysis of repeated measures data; p 589-638.

51. Baker DA, Hastings RS, Pruitt L. Study of fatigue resistance of chemical
and radiation crosslinked medical grade ultrahigh molecular weight polyethylene.
J Biomed Mater Res. 1999;46:573-81.
52. Furmanski J, Gupta S, Chawan A, Kohm A, Lannutti J, Jewett B, Pruitt LA, Ries
MD. Aspherical femoral head with highly cross-linked ultra-high molecular weight
polyethylene surface cracking. A case report. J Bone Joint Surg Am. 2007;89:
2266-70.
53. Ries MD, Pruitt L. Effect of cross-linking on the microstructure and mechanical
properties of ultra-high molecular weight polyethylene. Clin Orthop Relat Res.
2005;440:149-56.
54. Tower SS, Currier JH, Currier BH, Lyford KA, Van Citters DW, Mayor MB. Rim
cracking of the cross-linked Longevity polyethylene acetabular liner after total hip
arthroplasty. J Bone Joint Surg Am. 2007;89:2212-7.
55. Dumbleton JH, DAntonio JA, Manley MT, Capello WN, Wang A. The basis for
a second-generation highly cross-linked UHMWPE. Clin Orthop Relat Res.
2006;453:265-71.
56. Oral E, Christensen SD, Malhi AS, Wannomae KK, Muratoglu OK. Wear
resistance and mechanical properties of highly cross-linked, ultrahighmolecular weight polyethylene doped with vitamin E. J Arthroplasty. 2006;21:
580-91.