Abstract
Background: Particle-induced osteolysis is currently
a major problem affecting the long-term survivorship of total joint
replacements. Alendronate is a third-generation bisphosphonate that blocks
osteoclastic bone resorption. The objective of this study was to
determine whether alendronate could prevent particle-induced osteolysis
or restore (reverse) bone loss in established osteolysis.
Methods: A rat model of particle-induced osteolysis
was used. A specially designed polyethylene implant was placed in
the proximal part of the right tibia of seventy-two animals. Following four
weeks of healing, the animals were randomized into control groups,
a prevention group, or a treatment group. In the prevention group,
animals received intra-articular injections of high-density polyethylene
particles (mean size, 2 m; all <10 m) at four, six, and
eight weeks postoperatively. Alendronate (0.01 mg/kg/day)
was administered concomitantly through an implantable pump from
the fourth week through the tenth week. In the treatment group,
animals were also exposed to polyethylene particles at four, six,
and eight weeks, to establish bone loss, but they received alendronate subsequently,
from the tenth week through the sixteenth week, to treat the bone
loss. Positive (particle-only) and negative (saline-solution-only)
control groups were assessed as well. Tissues were harvested at
ten weeks in the prevention group and at sixteen weeks in the treatment
group. Histological analyses and histomorphometric determinations
of the periprosthetic bone volume were carried out.
Results: Histological examination showed a rim of
new bone (neocortex) around the implant in the untreated and saline-solution-treated
control animals (no polyethylene particles). Treatment with saline
solution (no polyethylene particles) did not affect periprosthetic
bone. Animals exposed to polyethylene particles had bone loss. In
those that received alendronate, the bone loss was either prevented
or reversed, and the quantity of neocortical and trabecular bone
was increased compared with that of the controls. Alendronate effectively
preserved periprosthetic bone in both the prevention and treatment groups.
In the prevention arm, the mean periprosthetic bone volume of
the neocortex and the surrounding trabecular bone, as determined
with histomorphometry, was 21.5% %plusmn; 6.5% in
the saline-solution-treated controls (no particles), 13.1% %plusmn;
5.9% in the particle-treated animals, and 32.6% %plusmn;
6.4% in the alendronate-treated animals (p < 0.001).
In the treatment arm, the mean periprosthetic bone volume was 27.2% %plusmn;
5.6% in the saline-solution-treated controls, 17.7% %plusmn;
6.2% in the particle-treated animals, and 30.2% %plusmn;
5.9% in the alendronate-treated animals (p = 0.002).
Conclusions: In our model, the intra-articular injection
of polyethylene particles caused substantial bone loss around a
loaded implant. Alendronate effectively prevented and treated the particle-induced
periprosthetic bone loss.
Clinical Relevance: Alendronate may be useful in
preventing particle-induced osteolysis around total joint implants.
It may also elicit bone formation in established osteolytic lesions.
Currently, over 250,000 total joint replacements are performed
annually in the United States1.
While these procedures have revolutionized the treatment of arthritis,
the implants have finite life spans and some eventually fail. In
1994, for example, over 40,000 revision arthroplasties were performed
in the United States1. Revision
procedures not only are more challenging technically but also are
associated with higher morbidity and cost and with less predictable
long-term results1-4.
The leading cause for the late failure of joint replacements
is aseptic loosening5-8. Particulate
wear debris, particularly ultra-high molecular weight polyethylene
particles from the bearing surfaces, causes osteolysis, an intense
inflammatory foreign-body reaction that may ultimately result in
massive bone loss and implant loosening8-14.
Macrophages and foreign-body giant cells secrete potent mediators
of bone resorption that result in the loss of bone15-20. There are currently no proven
pharmacological measures for the prevention of osteolysis, and often
the only treatment option is revision surgery. Many patients, particularly
those who are poor operative risks, could benefit greatly from nonoperative
treatment alternatives.
Alendronate, a third-generation bisphosphonate, works by blocking
osteoclastic bone resorption and has been shown to prevent particle-induced
osteolysis21,22. We hypothesized
that alendronate might be useful not only in the prevention of particle-induced
osteolysis but also in its treatment. These hypotheses were tested
in a small-animal model of osteolysis23,
in which the capacities of alendronate to prevent particle-induced
osteolysis and to increase bone formation in established osteolysis
were examined.
Study Design
This study, which was reviewed and approved by the Institutional
Animal Care and Use Committee, was a randomized, prospective mixed-model
experiment (Fig. 1).
The Cambridge osteolysis model23,24,
a simple and reproducible animal model for particle-induced osteolysis,
was used. After a pre hoc power analysis was performed to determine
the sample size, seventy-two rats underwent a hemiarthroplasty of
the right knee with a specially fashioned polyethylene tibial implant.
Healing was allowed to occur for four weeks. Untreated control animals
(no saline solution or polyethylene particles) were killed at four,
ten, and sixteen weeks (Groups A, B, and C, respectively; Table I). Experimental
animals were then randomized to either a prevention arm or a treatment
arm (Table I).
In the prevention arm, there were three groups: (1) the saline-solution-treated
group (Group D), which was not exposed to particles and which underwent
intra-articular injections of saline solution (negative control
group); (2) the particle-treated group (Group E), which received
polyethylene particles by means of three intra-articular injections
(positive control group); and (3) the alendronate-treated group (Group
F), which was exposed to particles and concurrently received alendronate.
The animals in the prevention arm were killed at ten weeks. The
treatment arm also included three groups: the saline-solution-treated
group (Group G; negative control group); the particle-treated group
(Group H; positive control group); and the alendronate-treated group
(Group I), which was exposed to particles to produce bone loss and
then subsequently received alendronate. These animals were killed at
sixteen weeks. After the animals were killed, the specimens were
examined histologically and periprosthetic bone volume was determined
with histomorphometry.
Animals
Seventy-two adult male Sprague-Dawley rats weighing 300 g were
used. The rats were maintained on Laboratory Rodent Diet 5001 (PMI
Feeds, St. Louis, Missouri) and water ad libitum and were caged
individually. Unrestricted weight-bearing was allowed.
Surgical Procedure
Anesthesia was obtained with intraperitoneal ketamine (Ketaset;
100 mg/mL) at a dose of 80 mg/kg and with xylazine
(Rompun; 20 mg/mL) at a dose of 5 mg/kg. The right hindlimb
was clipped free of hair and scrubbed with Betadine (povidone-iodine)
soap and 70% isopropyl alcohol.
The right knee joint was approached through a medial parapatellar
arthrotomy and with lateral dislocation of the patella. A bone tunnel
1 cm in length was prepared in the proximal part of the tibia with
a custom-designed, handheld drill fitted with a 1.2-mm stainless-steel
drill bit. Specially fabricated ultra-high molecular weight polyethylene
implants (Fig. 2)
were fashioned (Dana Biomedical Center, The Hospital for Special
Surgery, New York, NY) and sterilized. The head of the implant was
2.5 mm in diameter and 1.5 mm in vertical height. The stem was 1.3
mm in diameter and 8.5 mm in length, and the total length of the
implant was 1 cm. In order to allow the implant head to lie flush
against the subchondral bone, a counterbore was used to create a
1.5-mm-diameter depression in the tibial plateau. Bone fragments
were flushed from the joint by lavage with saline solution, and
the implant was inserted in a press-fit fashion into the bone tunnel.
The patella was replaced in the trochlear grooves, and the incision was
closed in layers with simple interrupted sutures of monofilament
nylon (Ethilon; Ethicon, Somerville, New Jersey).
Perioperative antibiotics were administered. The rats were monitored
until they awoke and at least once daily thereafter by the veterinary
staff. They were allowed unrestricted activity.
Particle Injections
High-molecular-weight polyethylene particles (generously donated
by Mr. Neil Rushton, MD, FRCS, University of Cambridge, Cambridge,
United Kingdom) were used in the study. Particle-size distribution
was determined by a laser particle sizer. The mean particle size
was 2 m, and all particles were <10 m; previous work showed
that particles of this size produce osteolysis in this model23. The particles were sterilized prior
to use. A suspension of particles (300 particles/mL) was
prepared in a 1:50 solution of Sprague-Dawley rat serum and phosphate-buffered
saline solution. Group-E and F animals in the prevention arm and Group-H
and I animals in the treatment arm (table I) received an intra-articular injection
of 200 L of the particle suspension into the operatively treated
(right) knee at four, six, and eight weeks postoperatively (three
injections per animal). At the same time-points, the saline-solution-treated
control animals (Groups D and G) received intra-articular injections
of the saline-solution vehicle only. All injections were made with
a 25-gauge needle through the patellar tendon. Anesthesia was administered
to reduce the stress of handling during these procedures.
Alendronate
Alendronate (MK-0217; 4-amino-1-hydroxybutylidene bisphosphonate
sodium salt) is a bisphosphonate that was designed to inhibit bone
resorption. Alendronate (Merck, West Chester, Pennsylvania) binds
to apatite mineral and inhibits osteoclast-mediated bone resorption.
Group-F animals in the prevention arm received alendronate for six
weeks beginning four weeks postoperatively. Group-I animals in the treatment
arm also received alendronate for six weeks, but beginning ten weeks
postoperatively. The alendronate was administered systemically at
a dose of 0.01 mg/kg/day through an implanted
mini-osmotic pump (Alzet, Palo Alto, California). The dose of alendronate
(0.01 mg/kg/day) was based on effective dose ranges
determined during preclinical studies of alendronate in rats25,26 (personal communication; G. Seedor,
Department of Bone Biology and Osteoporosis Research, Merck Sharp
and Dohme Research Laboratories) and during previous studies at
our institution27,28.
The mini-osmotic pumps that were used to deliver the alendronate
were implanted subcutaneously posterior to the scapulae. A midscapular
incision was made, and subcutaneous tissues were spread to create
a pocket for the pump. The wound was closed with skin staples in
standard fashion. Three biweekly implantations of a subcutaneous
pump were required for each alendronate-treated animal. Procedures
were synchronized with particle injections when possible.
Method and Time of Death
The rats were killed with CO2 inhalation at appropriate end points,
as shown in Table I,
in compliance with the most recent recommendations of the American
Veterinary Medical Association. The end points were based on those
used in previously published reports by Howie et al.29 and Allen et al.23.
Histological Analysis
The right hindlimb was removed en bloc, and the soft tissues were
removed. The tibia was fixed in neutral buffered formaldehyde, dehydrated
through a graded series of alcohols, and embedded in methylmethacrylate.
Horizontal cross sections were cut, at a uniform depth of 2 mm distal
to the articular surface of the tibia, distal to the head of the
implant, around the proximal part of the stem. With use of a Reichert-Jung sliding
microtome (Cambridge Instruments, Buffalo, New York) and a tungsten
carbide knife, 58-m-thick sections of calcified tissue were collected.
The sections were stained with hematoxylin and eosin, Masson trichrome,
or von Kossa stain.
Two investigators blinded with regard to the study group examined
each specimen independently with light microscopy and recorded a
detailed description of the histological appearance of each. The
neocortex and the trabecular bone of the proximal metaphysis were
carefully examined for bone loss or evidence of resorption.
Histomorphometric Determination of Periprosthetic
Bone Volume
Histomorphometry was performed with use of a semiautomated image
analysis system. Von Kossa-stained sections were used to determine
the percentage of mineralized periprosthetic bone. Bone volume in
the periprosthetic tissues was determined as a percentage of total
tissue volume, as described below.
For each specimen, the volume of the neocortex and the trabecular
bone surrounding the neocortex (referred to in this article as the
periprosthetic bone volume) was determined from a defined perimeter,
of the same total area for all specimens, that included within it
the neocortex and the trabecular bone surrounding the neocortex.
The cortical bone of the tibia was excluded because of artifacts
from the sectioning process. Thus, the periprosthetic bone volume
was determined from the ratio of neocortex plus trabecular bone
area to total tissue area and is reported as a percentage.
Digitized images of the von Kossa-stained cross sections were used
because they provided excellent discrimination between mineralized
and unmineralized tissue. Data from the various groups were compared
and analyzed statistically.
Membrane Grade
A five-part quantitative histological grading scheme was developed
to classify the different types of interface membranes (Table II). Specimens
were classified accordingly, and comparisons were made across groups.
Membrane Thickness
On representative cross-sectional images, the thickness of each
periprosthetic membrane was measured directly at four different
standardized regions. The mean thickness was then calculated for
each specimen. Summary data were compiled and compared across groups.
Statistical Analysis
Outcome parameters included percent bone volume, membrane thickness,
and membrane grade. Parametric data (bone volume and membrane thickness)
were statistically analyzed with two-way analysis of variance with
post hoc Tukey multiple-comparisons tests. Nonparametric data (membrane
grade) were compared with use of the Kruskal-Wallis nonparametric analysis
of variance test with post hoc Dunn multiple-comparisons tests.
P values of <0.05 were considered significant.
Five animals died prematurely and were excluded from analysis.
Two other specimens were damaged during histological processing
and were also excluded. All groups had a minimum of six specimens
available for complete analysis (Table I).
Histological Findings
Untreated and Saline-Solution-Treated Controls
In the untreated controls (no polyethylene particles or saline solution;
Groups A, B, and C), a rim of bone (neocortex) and a fibrous membrane
formed around the stem of the implant by four weeks (Group A) and
remained present throughout the ten weeks (Group B) or sixteen weeks
(Group C) of the study. A similar neocortex and fibrous membrane
formed around the stem of the implant in the saline-solution-treated
animals (no polyethylene particles; Groups D and G) (Figs. 3-A and 3-B). The fibrous membranes
were thick and benign-appearing, and they did not infiltrate the
neocortical bone. The fibrous membranes also were relatively acellular,
were of variable thickness, and had linear arrays of fibroblasts
with abundant, well-organized fibrous connective tissue. Osteoclasts
and foreign-body giant cells were not seen. Intra-articular injections of
saline solution (no polyethylene particles) had no effect on the
histological findings; the untreated controls (Groups A, B, and
C) and the saline-solution-treated controls (Groups D and G) had
essentially identical histological appearances.
Particle-Treated Animals
In both arms of the study, periprosthetic bone loss occurred
in animals treated with polyethylene particles but not with alendronate
(Group E [prevention arm] and Group H [treatment arm])
(Table I). The
histological findings in these positive controls were characterized
by thinning of the neocortex and osseous trabeculae, neocortical
perforations, and an inflammatory response (Figs. 4-A and 4-B). The neocortex,
trabecular bone, and cortical bone were decreased compared with
those in the animals not treated with particles or saline solution
(Groups B and C) and those in the animals treated with saline solution
alone (groups D and G). In the particle-treated animals, osteoclasts
and foreign-body giant cells penetrated the neocortical bone, and
osteolysis occurred in the neocortex, trabecular bone, and cortical
bone.
The interface membranes of the particle-treated animals (Group
E [prevention arm] and Group H [treatment
arm]), were markedly different from those of the untreated
controls (Groups B and C) and the saline-solution-treated controls (Groups
D and G). Furthermore, membranes from the particle-treated animals
were variable in thickness but more cellular, with less fibrous
stroma. The membranes infiltrated and invaded the neocortices that
surrounded the implants. Osteoclasts and foreign-body giant cells
were seen in areas of neocortical perforation. The interfaces between
neocortical bone and membranes were irregular and marked by areas
of perforation. Polyethylene debris was localized to the membrane and
could be seen under polarized light.
The process of particle-induced osteolysis was progressive between
the tenth and sixteenth weeks (Group E compared with Group H). By
sixteen weeks, three of the eight Group-H implants were surrounded
by fluid and were no longer surrounded by a membrane, and four of
the eight were grossly loose at necropsy.
Alendronate-Treated Animals
In comparison with the untreated (Group-B and C), saline-olution-treated
(Group-D and G), and particle-treated (Group-E and H) animals, the
alendronate-treated animals (Groups F and I) had distinct differences
in both the peri-implant bone and the membrane. When administered
either for prevention or for treatment, alendronate resulted in
huge increases in trabecular bone volume compared with that seen
in either the untreated or the saline-solution-treated controls.
The neocortex was better preserved and there was more trabecular
bone around the implants of the alendronate-treated animals (Figs. 5-A, 5-B, 5-C, 6-A, 6-B, and 6-C). Alendronate,
however, did not completely block the osteolytic process, as evidence
of osteoclastic bone resorption was still present (Fig. 7).
The interface membranes in the alendronate-treated animals (Groups
F and I) were less invasive than those in the respective particle-treated
animals (Groups E and H). In the treatment arm, in which the alendronate
was administered after the particles, the differences in the membranes
between the particle-treated group (H) and the alendronate-treated
group (I) were not as striking as were the differences in the prevention arm
(Group E compared with Group F) (Figs. 4-A through 6-C). As was
the case in the prevention arm (Group F), in the treatment arm the
grades of the membranes from the alendronate-treated animals (Group
I) were intermediate between those of the saline-solution-treated
(Group-G) and particle-treated (Group-H) animals.
Ten-Week Prevention Arm Compared with Sixteen-Week Treatment
Arm
Bone quality appeared better in the alendronate-treated animals
in the ten-week prevention arm (Group F) than in the alendronate-treated
animals in the sixteen-week treatment arm (Group I). This was most
apparent on examination of the von Kossa-stained specimens, where
more mineralized bone was present not only in the neocortex but
also peripherally in the cancellous bone closer to the endosteal
surface. An interesting observation in the prevention arm was the
presence of more polyethylene debris within the interface membranes
of alendronate-treated animals (Group F) than in particle-treated animals
(Group E). Alendronate seemed to have limited particle migration
away from the implant. Under polarized light, we also observed more
polyethylene particles in the periprosthetic tissues and less centripetally
in the tibia in the prevention arm (Group F) than in the treatment
arm (Group I).
Histomorphometric Periprosthetic Bone Volume
The effects of alendronate on periprosthetic bone volume (the combined
volume of the neocortex and trabecular bone) are summarized in Table III. Because
the histological findings in the untreated and saline-solution-treated
animals were similar and because the results of our previous study
were similar15, only saline-solution-treated
specimens were analyzed with histomorphometry. In both the ten-week
prevention arm (Groups D, E, and F) and the sixteen-week treatment
arm (Groups G, H, and I), periprosthetic bone volume decreased after
the injection of polyethylene particles and increased after administration
of alendronate compared with those values in the saline-solution-treated
controls. These differences were all significant (p < 0.05).
The model effectively demonstrated substantial bone loss, with
significant differences between the periprosthetic bone volumes
in the saline-solution-treated negative controls and the particle-treated
positive controls. In the prevention arm Group D had significantly
more bone than Group E (p = 0.048), and in the treatment
arm Group G had significantly more bone than Group H (p = 0.01).
Alendronate treatment resulted in dramatic increases in periprosthetic
bone volume compared with that of the saline-solution-treated negative
controls (Table III).
There were highly significant differences between the particle-treated
animals and the alendronate-treated animals in both the ten-week
prevention arm (Group E compared with F) (p < 0.001) and
the sixteen-week treatment arm (Group H compared with I) (p = 0.002).
In the ten-week prevention arm, the alendronate-treated (Group-F)
animals had a highly significant increase (p = 0.005) in
bone volume when compared with the saline-solution-treated (Group-D)
animals. In the sixteen-week treatment arm, there was no significant
difference (p = 0.626) between the saline-solution-treated
controls (Group G) and the alendronate-treated animals (Group I),
although the same trend was noted.
No significant differences were found when equivalent groups from
the two arms of the study were compared (i.e., saline-solution-treated
compared with saline-solution-treated, particle-treated compared
with particle-treated, or alendronate-treated compared with alendronate-treated;
p > 0.05).
Membrane Grade
The membrane grades are summarized in Figure 8. They were
generally lowest in the saline-solution-treated controls (Groups
D and G) and highest in the particle-treated animals (Groups E and
H). In both arms of the study, the alendronate-treated animals (Groups
F and I) had lower membrane grades than the particle-treated animals
(Groups E and H). Of all particle-treated animals, only one, an
animal from the sixteen-week treatment arm (Group H), had a benign-appearing
(grade-1) membrane.
In the prevention arm, the average score for the alendronate-treated
animals (Group F) was 1.9 compared with 3.7 for the particle-treated
animals (Group E). Statistical analysis with use of the Kruskal-Wallis
nonparametric analysis of variance test revealed a p value of 0.0131,
which is significant. Variation across groups was significantly
greater than expected by chance. The Dunn multiple-comparisons test
also revealed p values of <0.05 for the differences between
the particle-treated (Group-E) and saline-solution-treated (Group-D)
controls and between the alendronate-treated (Group-F) and particle-treated
(Group-E) animals.
In the treatment arm, the average score was 2.3 for the alendronate-treated
animals (Group I) compared with 3.6 for the particle-treated animals
(Group H). Although the trend was the same as that found in the
prevention arm, Kruskal-Wallis nonparametric analysis of variance
revealed a p value of 0.0537, which is considered not significant.
The Dunn multiple-comparisons test did reveal p values of <0.05
for the difference between the particle-treated (Group-H) and saline-solution-treated
(Group-G) controls.
Membrane Thickness
The membrane-thickness data are shown in Figure 9. Membrane
thickness ranged from a mean of 252 m in the particle-treated animals
(Group E) of the prevention arm to 76 m in the particle-treated
animals (Group H) of the treatment arm. Comparisons across the three
groups within each arm of the study did not reveal significant differences.
However, comparison across the two arms of the study (Group E compared
with H) did show a significant decrease in membrane thickness in
the particle-treated animals over time (p = 0.009), which
most likely represents progressive osteolysis from a longer exposure
to particles with implant loosening and membrane destruction. The
membrane thicknesses of the two saline-solution-treated groups (Groups
D and G) were not significantly different from those of the two
alendronate-treated groups (Groups F and I) (p > 0.05).
The Cambridge osteolysis model used in this study is a small-animal
model of particle-induced osteolysis23,24.
It was developed to address some of the limitations inherent in
previous small-animal models29,30.
To simulate a cementless joint replacement, implants were inserted
into the proximal parts of rat tibiae and were countersunk so that
their heads rested flat against the subchondral bone, preventing
distal migration and ensuring a connection with the joint surface.
Thus, synovial fluid and wear debris had direct access to the bone-implant
interface. Schmalzried et al. found this connection to be an important
means of transit for wear debris in humans31.
The Cambridge model also provided weight-bearing effects by achieving
direct contact between the implant head and the femoral condyles.
After a period of healing, a new rim of bone, the neocortex, was
evident around all implants. We believe that the fibrous membrane
that formed around the implants in the untreated and saline-solution-treated
controls is indicative of loading and micromotion. The histological appearance
was similar to that of fibrous membranes that form around porous
ingrowth prostheses in humans.
Intra-articular injections of high-density polyethylene particles
simulated implant wear and initiated the process of osteolysis12,23. The model is designed so that
cyclical loading is transmitted to the bone-implant interface to
potentiate the spread of wear debris. At harvest, animals that had
received polyethylene particles displayed loss of bone and an intense
foreign-body inflammatory response that mimicked that seen around
loose implants in the clinical setting17,32,33.
We found the Cambridge osteolysis model to be a simple and reproducible
system in which to study the effects of particle-induced bone loss.
Alendronate, a third-generation bisphosphonate, is a potent inhibitor
of bone resorption25,34 and has
been shown to be effective in the treatment of several diseases
characterized by increased bone resorption6.
Alendronate has less effect on osteoblasts, and in turn bone formation,
than earlier-generation bisphosphonates do35.
The drug binds tightly to apatite and is subsequently released around
the osteoclasts, interfering with bone resorption and ruffled border
formation25,36,37. Although the
exact molecular mechanism of alendronate remains unclear, the overall
effect is the inhibition of osteoclastic bone resorption. As particle-induced
osteolysis is a problem of excessive bone resorption, we believe
that alendronate may prevent and possibly reverse this type of bone
loss.
Shanbhag et al. recently reported on the use of alendronate in the
treatment of wear-debris-mediated osteolysis in a cementless canine
total hip arthroplasty model22.
They found that, during the twenty-four-week study, treatment with
alendronate inhibited particle-induced osteolysis around the implants
but had no effect on macrophages or inflammatory mediators. Their
findings are consistent with the pharmacodynamics of alendronate,
which blocks resorption by inhibiting osteoclasts, and they support
our findings as well. In the present study, osteoclasts were present
in alendronate-treated animals. Because of their study design and
the model that they used, Shanbhag et al. could not answer questions
about the use of alendronate to reverse or prevent particle-induced
osteolysis.
The current study was undertaken to test the hypothesis that alendronate
could be used to prevent and treat particle-induced osteolysis.
Our histological data lend support to this hypothesis. A six-week
course of alendronate at a dose of 0.01 mg/kg/day
decreased particle-induced periprosthetic bone loss when administered
in preventative and therapeutic modes. Polyethylene particles reproducibly
caused a histological response that resulted in periprosthetic bone
loss and mimicked an aseptically loose prosthesis. The high-density polyethylene
particles used in our study were of the small size that seems to
be of greatest concern clinically12,15,38,
and the histological response was typical of that seen in association
with failed total joint replacements17.
Inflammatory responses at the interface membranes and periprosthetic
bone loss were caused by the polyethylene particles. Alendronate
prevented or at least retarded the process of particle-induced bone
loss. Bone mass in the proximal part of the tibia was increased
by the administration of alendronate in both arms of the study.
In the current study and in the study by Shanbhag et al.22, alendronate was shown to prevent
bone loss when administered concurrently with particles. Sabokbar
et al. recently showed that bisphosphonates are capable of inhibiting
particle-induced bone resorption in vitro21.
To our knowledge, however, the current study is the first to show
that alendronate has beneficial effects on periprosthetic bone when
administered therapeutically, after particle-induced bone loss has
occurred. Bone volume was increased significantly after the administration
of alendronate.
Membrane thickness appeared to be inversely proportional to membrane
grade, at least in the prevention arm. Although statistical analysis
was hampered by widespread scatter in the data, there was a strong
trend for the more benign-appearing membranes to be thicker and
the more inflamed membranes to be thinner. Similar findings were
reported in a previous experiment in which polymethylmethacrylate
pins were used in the Cambridge osteolysis model24.
Our explanation for this observation is that more aggressive membranes
cause more bone loss and more implant loosening, eventually resulting
in mechanical destruction of the membranes and therefore thinner
membranes. It is unclear whether alendronate protects the periprosthetic
membrane from mechanical destruction because of greater bone volume and
less loosening of the implants or whether alendronate has effects
on fibroblasts in the membrane.
The distribution of polyethylene debris within the proximal part
of the tibia varied across groups. Polyethylene debris was much
more confined to the interface membranes in the alendronate-treated
animals, particularly in the ten-week prevention arm. While the
mechanism whereby alendronate inhibits osteolysis is presumed to
be the inhibition of osteoclasts, a secondary effect of alendronate
may be the containment of the inciting particulate debris in the
interface membranes, preventing spread to adjacent bone. If the
neocortex that forms around the implant is preserved, there is no
channel for the migration of particles. Thus, periprosthetic bone
could be protected by alendronate in two ways: first, osteoclasts,
inhibited by alendronate, would not resorb bone, and second, polyethylene
debris, confined to the interface membrane, would not have access
to the remaining periprosthetic bone.
In our study, alendronate did not completely eliminate either the
presence or the activity of osteoclasts, as there was evidence of
both in some of the alendronate-treated animals. The data from the
current study and that by Shanbhag et al.22 show
that alendronate does decrease particle-induced bone loss, presumably
through reduced osteoclastic activity. In a recent study by Astrand
and Aspenberg, instability-induced bone loss was not affected by
alendronate39. At a dose of 0.063
mg/kg/day, alendronate was unable to inhibit instability-induced
bone loss, although it did affect bone-remodeling. We believe that
the data from these three studies support the contention that particle-induced
bone resorption and instability-induced bone resorption occur by different
mechanisms22-24,39.
In summary, particle-induced osteolysis is the major problem affecting
the long-term survival of total joint prostheses. The strategy outlined
in the current study is a biological approach that involves blocking
the final common pathway for particle-induced bone resorption. While
such a scheme shows great promise, many questions, such as long-term
effects of alendronate on periprosthetic bone, appropriate dosing
regimens, and relevant monitoring methods, remain.
Note: The authors are grateful to Steve Doty, PhD, and Margaret
Peterson, PhD, for their assistance with this project. They also
wish to acknowledge Peter Bullough, MD, and Adele Boskey, PhD, for
their critical reviews.
Wright TM, Goodman SB.Implant
wear: the future of total joint replacement. Rosemont,
IL: American Academy of Orthopaedic Surgeons; 1996. p 4
Bryan RS,Rand JA. Revision total knee arthroplasty. Clin Orthop,1982;170: 116-22. 170116
1982
[PubMed]
Insall JN,Dethmers DA. Revision of total knee arthroplasty. Clin Orthop,1982;170: 123-30. 170123
1982
[PubMed]
Total hip replacement. NIH Consens Statement,1994;12: 1-31. 121
1994
Harris WH. Osteolysis and particle disease in hip replacement. A
review. Acta Orthop Scand,1994;65: 113-23. 65113
1994
[PubMed]
Liberman UA, Weiss SR, Broll J, Minne HW, Quan H, Bell NH, Rodriguez-Portales J, Downs RW Jr, Dequeker J,Favus M. Effect of oral alendronate on bone mineral density and
the incidence of fractures in postmenopausal osteoporosis. The Alendronate
Phase III Osteoporosis Treatment Study Group. N Engl J Med,1995;333: 1437-43. 3331437
1995
[PubMed]
Maloney WJ,Smith RL. Instructional Course Lecture, American Academy of Orthopaedic
Surgeons. Periprosthetic osteolysis in total hip arthroplasty: the
role of particulate wear debris. J Bone Joint Surg Am,1995;77: 1448-61. 771448
1995
Murray DW,Rushton N. Macrophages stimulate bone resorption when they phagocytose
particles. J Bone Joint Surg Br,1990;72: 988-92. 72988
1990
[PubMed]
Goodman SB, Fornasier VL,Kei J. The effects of bulk versus particulate polymethylmethacrylate on
bone. Clin Orthop,1988;232: 255-62. 232255
1988
[PubMed]
Harris WH, Schiller AL, Scholler JM, Freiberg RA,Scott R. Extensive localized bone resorption in the femur following total
hip replacement. J Bone Joint Surg Am,1976;58: 612-8. 58612
1976
[PubMed]
Kreibich DN, Moran CG, Delves HT, Owen TD,Pinder IM. Systemic release of cobalt and chromium after uncemented total
hip replacement. J Bone Joint Surg Br,1996;78: 18-21. 7818
1996
[PubMed]
McKellop HA, Campbell P, Park SH, Schmalzried TP, Grigoris P, Amstutz HC,Sarmiento A. The origin of submicron polyethylene wear debris in total
hip arthroplasty. Clin Orthop,1995;311: 3-20. 3113
1995
[PubMed]
Salvati EA, Betts F,Doty SB. Particulate metallic debris in cemented total hip arthroplasty. Clin Orthop,1993;293: 160-73. 293160
1993
[PubMed]
Willert HG. Reactions of the articular capsule to wear products of
artificial joint prostheses. J Biomed Mater Res,1977;11: 157-64. 11157
1977
[PubMed]
Allen MJ, Myer BJ, Millett PJ,Rushton N. The effects of particulate cobalt, chromium and cobalt-chromium
alloy on human osteoblast-like cells in vitro. J Bone Joint Surg Br,1997;79: 475-82. 79475
1997
[PubMed]
Brien WW, Salvati EA, Betts F, Bullough P, Wright T, Rimnac C, Buly R,Garvin K. Metal levels in cemented total hip arthroplasty. A comparison of
well-fixed and loose implants. Clin Orthop,1992;276: 66-74. 27666
1992
[PubMed]
Chiba J, Rubash HE, Kim KJ,Iwaki Y. The characterization of cytokines in the interface tissue obtained
from failed cementless total hip arthroplasty with and without femoral
osteolysis. Clin Orthop,1994;300: 304-12. 300304
1994
[PubMed]
Kossovsky N, Liao K, Millett D, Feng D, Campbell PA, Amstutz HC, Finerman GA, Thomas BJ, Kilgus DJ, Cracchiolo A,et al. Periprosthetic chronic inflammation characterized through
the measurement of superoxide anion production by synovial-derived
macrophages. Clin Orthop,1991;263: 263-71. 263263
1991
[PubMed]
Millett PJ, Sabokbar A, Allen MJ, Myer B,Rushton N. Osteoblast activity around failed total hip replacements:
synovial fluid levels of alkaline phosphatase and osteocalcin. Hip Int,1995;5: 8-14. 58
1995
Ohlin A, Johnell O,Lerner UH. The pathogenesis of loosening of total hip arthroplasties.
The production of factors by periprosthetic tissues that stimulate in
vitro bone resorption. Clin Orthop,1990;253: 287-96. 253287
1990
[PubMed]
Sabokbar A, Fujikawa Y, Murray DW,Athanasou NA. Bisphosphonates in bone cement inhibit PMMA particle induced
bone resorption. Ann Rheum Dis,1998;57: 614-8. 57614
1998
[PubMed]
Shanbhag AS, Hasselman CT,Rubash HE. The John Charnley Award. Inhibition of wear debris mediated osteolysis
in a canine total hip arthroplasty model. Clin Orthop,1997;344: 33-43. 34433
1997
[PubMed]
Allen M, Brett F, Millett P,Rushton N. The effects of particulate polyethylene at a weight-bearing bone-implant
interface. A study in rats. J Bone Joint Surg Br,1996;78: 32-7. 7832
1996
[PubMed]
Allen MJ, Schoonmaker JE, Ayers DC. Pharmacological
inhibition of debris-induced osteolysis. Transactions of the 12th
Annual International Symposium for Technology in Arthroplasty; 1999
Sept 23-25; Chicago
Rodan GA. Bone mass homeostasis and bisphosphonate action. Bone,1997;20: 1-4. 201
1997
[PubMed]
Toolan BC, Shea M, Myers ER, Borchers RE, Seedor JG, Quartuccio H, Rodan G,Hayes WC. Effects of 4-amino-1-hydroxybutylidene bisphosphonate
on bone biomechanics in rats. J Bone Miner Res,1992;7: 1399-406. 71399
1992
[PubMed]
Camacho NP, Brayton CF, Buckmeyer J, Raggio CL, Root L, Toledano T,Boskey AL. Evaluation of alendronate for treatment in an animal model
of osteogenesis imperfecta. J Bone Miner Res,1997;12(Suppl): 389. 12(Suppl)389
1997
Camacho NP, Zraick V, Grillo D, Ilg WA, Toledano T, Brayton CF, Raggio CL, Root L,Boskey AL. Alendronate reduces fractures in a mouse model of osteogenesis
imperfecta. Proceedings of the American Society of Bone and Mineral
Research meeting.. J Bone Miner Res,1998;
1998
Howie DW, Vernon-Roberts B, Oakeshott R,Manthey B. A rat model of resorption of bone at cement-bone interface
in the presence of polyethylene wear particles. J Bone Joint Surg Am,1988;70: 257-63. 70257
1988
[PubMed]
Howie DW, Manthey B, Hay S,Vernon-Roberts B. The synovial response to intraarticular injection in rats
of polyethylene wear particles. Clin Orthop,1993;292: 352-7. 292352
1993
[PubMed]
Schmalzried TP, Jasty M,Harris WH. Periprosthetic bone loss in total hip arthroplasty. Polyethylene wear
debris and the concept of the effective joint space. J Bone Joint Surg Am,1992;74: 849-63. 74849
1992
[PubMed]
Horowitz SM, Doty SB, Lane JM,Burstein AH. Studies of the mechanism by which the mechanical failure
of polymethylmethacrylate leads to bone resorption. J Bone Joint Surg Am,1993;75: 802-13. 75802
1993
[PubMed]
Mirra JM, Amstutz HC, Matos M,Gold R. The pathology of the joint tissues and its clinical relevance
in prosthesis failure. Clin Orthop,1976;117: 221-40. 117221
1976
[PubMed]
Peter CP, Cook WO, Nunamaker DM, Provost MT, Seedor JG,Rodan GA. Effect of alendronate on fracture healing and bone remodeling in
dogs. J Orthop Res,1996;14: 74-9. 1474
1996
[PubMed]
Fleisch H. Bisphosphonates:
mechanisms of action and clinical use. In: Bilezikian JP, editor. Bone
biology. San Diego: Academic Press; 1996. p 1037
Rodan GA, Seedor JG,Balena R. Preclinical pharmacology of alendronate. Osteoporosis Int,1993;3(Suppl 3): 7-12. 3(Suppl 3)7
1993
Sato M, Grasser W, Endo N, Akins R, Simmons H, Thompson DD, Golub E,Rodan GA. Bisphosphonate action. Alendronate localization in rat
bone and effects on osteoclast ultrastructure. J Clin Invest,1991;88: 2095-105. 882095
1991
[PubMed]
Campbell P, Ma S, Yeom B, McKellop H, Schmalzried TP,Amstutz HC. Isolation of predominantly submicron-sized UHMWPE wear particles
from periprosthetic tissues. J Biomed Mater Res,1995;29: 127-31. 29127
1995
[PubMed]
Astrand J,Aspenberg P. Alendronate did not inhibit instability-induced bone resorption.
A study in rats. Acta Orthop Scand,1999;70: 67-70. 7067
1999
[PubMed]