Cross-linking changes the wear and mechanical behavior of ultra-high
molecular weight polyethylene. Contemporary highly cross-linked ultra-high
molecular weight polyethylene has demonstrated markedly improved wear in vitro
and a reduction in static and fatigue strength compared with conventional
ultra-high molecular weight
polyethylene1-7.
Studies employing hip simulators have shown a reduction in the volume of
wear debris generated with cross-linked polyethylene in comparison with that
generated with conventional
polyethylene8.
However, the average size of wear particles also appeared to be reduced, which
means that the actual number of generated particles may not be
decreased8. The
osteolytic response to particulate debris is variable and may depend more on a
number of factors, including the number and the size of particles, than on the
total volume of debris
generated9,10.
The long-term clinical results of the use of currently available contemporary
highly cross-linked ultra-high molecular weight polyethylene components in
total hip arthroplasty are not yet known.
The following case demonstrates that early symptomatic osteolysis
associated with impingement can occur with the use of contemporary highly
cross-linked ultra-high molecular weight polyethylene components in total hip
arthroplasty. Our patient was informed that data concerning the case would be
submitted for publication.
Asixty-year-old woman underwent a right hybrid total hip arthroplasty,
without complications, for the treatment of osteoarthritis in July of 2000.
The components that were implanted included a Trilogy (Zimmer, Warsaw,
Indiana) 58-mm acetabular shell, a Longevity (Zimmer) 58/28-mm polyethylene
acetabular cup with a 10° offset lip, a Perfecta PDA (Wright Medical,
Arlington, Tennessee) 13.5-cm femoral stem, a 28-mm cobalt-chromium femoral
head (Wright Medical) with a +10.5-cm neck length, and a 13-mm distal spacer
(Wright Medical). The femoral component was a forged steel implant with a
surface-grit-blasted stem.
The hospital course was uncomplicated, and the patient was discharged to
home on the seventh postoperative day. She did well during the initial
postoperative period. Radiographs made at the twelve-month follow-up
evaluation showed a lucent area that measured 2 mm in width and 25 mm in
length on the medial aspect of the cement mantle, approximately 10 mm from the
distal end of the femoral component (Fig.
1). At twenty months after the operation, the patient reported
aching in the thigh and discomfort during activities that required twisting.
She also reported some minor difficulty in putting on socks and shoes.
Radiographs showed that the lucent area had increased to 5 mm in width and 44
mm in length on the medial distal aspect and 7 mm in width and 10 mm in length
on the lateral distal aspect of the cement mantle. At twenty-three months
after the operation, the patient had progressively increasing thigh pain.
Radiographs showed that the area of lucency had increased farther along the
entire length of the medial border of the stem
(Fig. 2). A bone scan showed
increased uptake about the femoral component, which was consistent with
loosening.
Approximately two years after the operation, the patient underwent a
revision right total hip arthroplasty. The intraoperative findings revealed
reactive fluid without evidence of infection and a loose femoral component at
the implant-cement interface, which was consistent with particulate
debris-induced osteolysis. Examination of the components during surgery
demonstrated impingement between the collar of the femoral head and the
anterior aspect of the acetabular component rim during hip flexion.
Intraoperative culture of tissue and fluid specimens was negative.
Retrieval Analysis
The retrieved acetabular liner was examined under light microscopy and
scanning electron microscopy. The implant was sputter-coated with
gold-palladium prior to examination under scanning electron microscopy.
Overall, the contemporary highly cross-linked ultra-high molecular weight
polyethylene insert showed no gross damage except for an area of deformation
on the rim that was consistent with impingement
(Figs. 3-A and 3-B). The area
of collar impingement showed normal morphologic surface features both on gross
visual examination and under scanning electron microscopic examination. Most
importantly, high-magnification scanning electron microscopic examination did
not demonstrate surface damage, such as cracking or pitting (Figs.
4-A and
4-B).
Tissue Analysis
Tissue specimens were obtained at the time of the revision surgery.
Specimens from the femoral side were obtained from the distal medial and
proximal lateral aspects of the femoral bone-cement interface. Additional
tissues were obtained from the area of the exposed femoral neck and were
labeled periarticular. The tip of the cement block was intact, and fixation
was solid. There was no evidence of osteolysis or loosening of the acetabular
component.
Specimens from each of the areas described were stained with hematoxylin
and eosin for examination under light microscopy and also with gold-palladium
in preparation for examination under scanning electron microscopy. Hematoxylin
and eosin staining of the retrieved periprosthetic tissue demonstrated
numerous birefringent particles (Fig.
5). The debris particles were of varied shapes, and most particles
appeared to be within the range of <1 µm to 5 µm
(Fig. 6).
Scanning electron microscopy of the retrieved periprosthetic tissue was
performed to characterize the debris in greater detail. The specimens were
initially placed in 10% neutral buffered formaldehyde solution. The specimens
were then processed for the scanning electron microscopic analysis with use of
a standard
protocol11.
Initially, they were fixed in 2% glutaraldehyde solution (containing 0.1-M
sodium cacodylate buffer, pH = 7.2) for one to two hours, rinsed in 0.1-M
sodium cacodylate buffer (pH = 7.2) for forty-five minutes, postfixed in 1%
osmium tetroxide solution (containing 0.1-M sodium cacodylate buffer, pH =
7.2) for one to two hours, and then rinsed again in 0.1-M sodium cacodylate
buffer (pH = 7.2) for fifteen minutes. They were next dehydrated in ethyl
alcohol in varying percentages for more than one hour. They were then
sputter-coated with 13 nm of gold-palladium to aid in surface conduction
during the scanning electron microscopic examination. Specimens were examined
with use of an Electroscan E3 environmental scanning electron microscope
(Royal Philips Electronics, Amsterdam, The Netherlands).
In vitro hip-simulator testing of contemporary highly cross-linked
ultra-high molecular weight polyethylene has demonstrated a substantial
reduction in volumetric
wear5,12
and a possible decrease in the number of wear particles in comparison with
conventional ultra-high molecular weight
polyethylene8,13.
The hip-simulator studies appear to provide compelling evidence that
contemporary cross-linked ultra-high molecular weight polyethylene can be
expected to increase the longevity of a total hip prosthesis. However, the
clinical response to particulate debris is quite variable and dependent on the
size, shape, and number of particles as well as on other
factors9,13,14.
The debris particles that were found in the periprosthetic tissues of our
patient were within the size range that is typically associated with
osteolysis15. Also,
the prosthesis failed clinically within two years after the total hip
arthroplasty, which demonstrates that osteolysis can develop rapidly in
association with components made of contemporary highly cross-linked
ultra-high molecular weight polyethylene.
Contemporary highly cross-linked ultra-high molecular weight polyethylene
has also been associated with a degradation of mechanical properties, which
may affect wear behavior related to fatigue and fracture
mechanisms16.
Early, clinically retrieved, electron-beam-irradiated, contemporary highly
cross-linked ultra-high molecular weight polyethylene inserts, which were
removed because of mechanical loosening of cementless components contaminated
with machining oil, demonstrated surface deformation and
cracking17,18.
Surface deformation has also been observed in ethylene-oxide-sterilized,
nonirradiated, retrieved ultra-high molecular weight polyethylene liners,
although the non-cross-linked surfaces demonstrated more abrasion and cold
flow at the bearing surface than did the clinically retrieved highly
cross-linked acetabular
liners18. The
decrease in ductility of contemporary ultra-high molecular weight polyethylene
caused by cross-linking may contribute to the different surface mechanical
behavior in vivo compared with that seen with the use of non-cross-linked
ultra-high molecular weight polyethylene.
Wear occurs primarily at the articulating surface of the implant but can
also occur as a result of impingement between the femoral neck and the rim of
the acetabular component. Impingement wear may be severe enough to cause
osteolysis13,14,19.
Deformation of the rim of the acetabular component due to impingement by the
prosthetic femoral neck is a common finding in retrieved total hip implants.
Yamaguchi et al. reported evidence of impingement in forty-one (39%) of 104
retrieved ultra-high molecular weight polyethylene acetabular
inserts13.
Impingement would be expected to occur over a relatively small area of the rim
of the acetabular component, which could be associated with relatively high
contact stress. If the ultra-high molecular weight polyethylene rim deforms
from impingement, the contact area may then increase, leading to decreased
local contact stresses. Ultra-high molecular weight polyethylene wear
increases as contact stress
increases20. In the
case of our patient, increased contact stress from contact of the femoral neck
against a relatively small area of the liner rim may have contributed to
increased wear. Metal-on-metal and ceramic-on-ceramic total hip prostheses are
usually designed to minimize impingement of the hard bearing surfaces through
the avoidance of the use of elevated rim liners. Conventional ultra-high
molecular weight polyethylene, which has greater ductility than contemporary
highly cross-linked ultra-high molecular weight polyethylene, may therefore
tolerate impingement better. However, contemporary highly cross-linked
ultra-high molecular weight polyethylene liners are available with elevated
rims, although, to our knowledge, quantitative in vitro measurements of
impingement wear of cross-linked and conventional polyethylene have not been
reported.
The severe local wear and osteolysis reported in this case report on a
patient treated with a contemporary highly cross-linked ultra-high molecular
weight polyethylene acetabular cup is a concern. The loosening in our patient
appears to have resulted from particle-induced bone resorption with subsequent
penetration, along the bone-cement interface, of joint fluid that contained
osteolytic enzymes. Schmalzried et al. described the "effective joint
space" as the periarticular tissues, which are invaded by joint fluid in
a patient who has a total hip prosthesis and
osteolysis21. The
implant-cement interface in our patient would be expected to be relatively
intact since the original surgery was performed less than two years
previously. However, the proximal aspect of the femur, which is exposed to
joint fluid, is more likely to undergo osteolysis, which can then extend
distally along the bone-cement interface.
During total hip arthroplasty, the range of motion of the trial components
until impingement should be assessed. If impingement occurs, it may be
necessary to alter the component position, avoid the use of a skirted femoral
head, reposition an elevated rim or a neutral liner, or increase the femoral
head size to avoid impingement, regardless of the material used for the
bearing surface. In vitro testing for component impingement may be a better
way of predicting the behavior of contemporary highly cross-linked ultra-high
molecular weight polyethylene in vivo.
Note: The authors thank Ron Wilson and Gordon Vrdoljak for
assistance with the scanning electron microscopy, and Ted Gallardo for his
diligent work in helping them to complete this project.