Experimental Design
The study was approved by the Institutional Animal Care and Use Committee.
Twenty-four male Yorkshire piglets, four to five weeks of age and weighing 6
to 8 kg, were used. Ischemic necrosis of the right femoral head was surgically
induced in all piglets by placing a ligature tightly around the right femoral
neck, as described
previously22. The
wound was closed in layers, and no immobilization was used
postoperatively.
The animals were divided into three groups according to whether they
received saline solution, prophylactic treatment, or post-ischemia treatment.
The contralateral, untreated femoral heads from the animals that had received
saline solution served as the normal control group. Eight animals were
assigned to each group. In the prophylactic treatment group, ibandronate was
administered to the animals seven, five, and two days before the operation and
then, beginning one week after the operation, three times per week for six
weeks. Although this protocol is not clinically realistic, except for patients
with a high risk for the development of ischemic necrosis, the group was
included to investigate the optimal efficacy of ibandronate. Administering
ibandronate before the disruption of blood flow to the femoral head ensured
that the drug would reach the target tissue. The total cumulative dose
administered to each animal in this group was 1.5 mg/kg, of which a half was
administered prior to the operation.
The animals in the post-ischemia treatment group received ibandronate three
times per week, starting one week after the operation, for six weeks. Each
dose was 44.4 µg/kg, and the total cumulative dose given to each animal was
0.8 mg/kg.
The third group of animals was given injections of saline solution three
times per week, starting one week after the operation, for six weeks.
All of the injections of ibandronate and saline solution were administered
subcutaneously. Ibandronate was prepared by dissolving ibandronic acid
monosodium monohydrate (Roche Diagnostics, Mannheim, Germany) in saline
solution and titrating the pH to 7.4 prior to the injection. The doses are
expressed as free acid equivalents of ibandronate.
All animals were killed eight weeks after the operation. This time-point
was chosen because severe deformity of the femoral head due to the
predominance of osteoclastic bone resorption is observed by this
time19,20.
Radiography
The entire femora were retrieved from the animals and were radiographed
with use of a Faxitron x-ray machine (model 43855A; Hewlett-Packard,
McMinnville, Oregon). The film was placed directly under the specimen to
maintain a constant distance from the x-ray source and to minimize any effects
of magnification. An architectural ruler was used to measure the maximum
height and diameter of the femoral head on each anteroposterior radiograph.
The epiphyseal quotient was calculated by dividing the maximum height of the
osseous epiphysis by the maximum
diameter23.
The length of the untreated femur was measured from the tip of the femoral
head to the tip of the distal femoral condyle on anteroposterior
radiographs.
Histological Procedure
The femoral heads were bisected at the central load-bearing region with a
precision diamond saw. The halves were sectioned into 4-mm-thick slices to
facilitate tissue fixation. The sections were radiographed and were fixed in
10% neutral buffered formalin under vacuum. After three days, half of the
sections from each head were transferred to EDTA solution for decalcification
and paraffin embedding. The remaining sections were placed in a series of
graded ethanol solutions for embedding in methylmethacrylate. The paraffin
blocks were cut into 5-µm sections and stained with hematoxylin and eosin
or stained for tartrate-resistant alkaline phosphatase to visualize
osteoclasts with use of a standard azo dye technique at pH 5.0. The plastic
blocks were sectioned to a 4-µm-thickness on a sledge microtome and were
stained with 5% silver nitrate with the von Kossa method and with McNeal
tetrachrome stain.
Histomorphometry
The sections stained with the von Kossa method and with McNeal tetrachrome
reaction were used to determine the histomorphometric indices of bone
resorption and formation adopted by the American Society for Bone and Mineral
Research24. A
representative section from the central region of the femoral head from each
animal was used. This region was chosen because it is the region of maximum
femoral head height and diameter and thus corresponds to the outline of the
femoral head on a radiograph. Contiguous images of entire femoral head
sections were captured with use of a Nikon E800 microscope (Nikon USA,
Melville, New York) equipped with an automated stage system and a
high-resolution digital camera. The contiguous images were tiled together with
use of Image-Pro Plus software (Media Cybernetics, Silver Spring, Maryland)
prior to the image analysis.
To determine the trabecular bone volume, tissue volume, amount of
trabecular separation, trabecular thickness, and trabecular number according
to the parallel plate model, the digital images were obtained at 20×
magnification and the tissue area, bone area, and trabecular perimeter were
measured. The derived indices were calculated according to the guidelines
approved by the American Society for Bone and Mineral
Research24.
Measurements were performed 1 mm from the chondro-osseous junction to exclude
the primary spongiosa from the analysis. The area of tissue measured in each
femoral head ranged from 50 to 60 mm2.
To determine the osteoclast number, percentage of osteoblast surface, and
percentage of osteoclast surface in the areas of repair, images were obtained
at 200× magnification and analyzed with use of Bioquant image analysis
software (Bioquant Image Analysis, Nashville, Tennessee). In the analysis of
the osteoblast surface, only the trabecular surface with a layer of osteoid
and cuboid-shaped osteoblasts was measured.
Statistical Analysis
One-way analysis of variance was used to determine the overall difference
among groups. If the difference was significant at p < 0.05, post hoc
Fisher protected least-significant-difference testing was performed to assess
the significance of the differences among groups. All data are presented as
the mean and standard deviation.
Ischemic necrosis did not develop in one animal in the group treated with
saline solution and in one in the prophylactic treatment group, and they were
excluded from the study. The rest of the animals showed clear evidence of
ischemic necrosis; i.e., growth arrest of the osseous epiphysis, necrotic
changes in the marrow space, and empty lacunae in the trabecular bone. All of
the animals that had received ibandronate demonstrated inhibition of bone
resorption, seen as radiodense bands near the growth plates, indicating
pharmacological activity by ibandronate on the growing
skeleton25.
A moderate-to-severe deformity of the femoral head was observed in the
group treated with saline solution (Figs.
1-A and
1-B), which had the lowest mean
epiphyseal quotient (Table I).
Compared with that group, the groups treated with ibandronate had
significantly better preservation of the femoral heads, with the mean
epiphyseal quotient being 60% greater in the group that had received
prophylactic treatment (p < 0.001) and 40% greater in the group that had
received post-ischemia treatment (p = 0.02).
Histomorphometric assessment revealed a severe loss of trabecular bone
volume (bone volume divided by tissue volume) in the group treated with saline
solution, which had the lowest mean trabecular bone volume (Figs.
2,
3, and
4); the mean trabecular bone
volume was 100% greater in the prophylactic treatment group (p = 0.002) and
93% greater in the post-ischemia treatment group (p = 0.003).
The group treated with saline solution had the lowest mean trabecular
number and the highest mean amount of trabecular separation, indicating a
substantial loss of the trabecular framework. Compared with that group, the
prophylactic (p < 0.0001) and post-ischemia (p = 0.007) treatment groups
had a significantly higher trabecular number and a significantly lower amount
of trabecular separation (p < 0.01), indicating better preservation of the
trabecular framework. No significant difference in the trabecular thickness
was observed between the saline-solution and ibandronate treatment groups.
The percentage of osteoclast surface did not differ significantly between
the saline-solution and ibandronate treatment groups. However, the total
number of osteoclasts per tissue area was significantly higher in the group
treated with saline solution than in those treated with ibandronate. This was
due to the increased number of osteoclasts found in the areas where trabecular
bone had been resorbed. These osteoclasts were unattached to the trabeculae
since the trabeculae in the region had been completely resorbed. No
significant difference in the percentage of osteoblast surface was found
between the saline-solution and ibandronate treatment groups.
The mean femoral length on the untreated side was 14.1 ± 0.8, 12.4
± 1.7, and 13.3 ± 0.3 cm in the saline-solution, prophylactic,
and post-ischemia treatment groups, respectively, at eight weeks. On the basis
of measurements of six normal piglets, the mean baseline femoral length at the
beginning of the study was considered to be 9.2 ± 0.6 cm. Thus,
compared with the baseline, the femoral length increased by 4.9, 3.2, and 4.1
cm in the saline-solution, prophylactic, and post-ischemia treatment groups,
respectively, during the eight-week study period. The femora were 1.7 cm (p
< 0.00001) and 0.8 cm (p = 0.01) shorter in the prophylactic and
post-ischemia treatment groups, respectively, compared with the femora of the
animals that had received saline solution. The prophylactic treatment group
had a significantly shorter mean femoral length compared with the
post-ischemia treatment group (p = 0.006).
Femoral head deformity is the most serious consequence of ischemic necrosis
of the immature femoral head as it can severely compromise the longevity of
the hip
joint12-14.
It is important to recognize that the repair process following ischemic
necrosis of the femoral head contributes to the development of femoral head
deformity19,20.
Production of a net bone loss in the early phase of the repair compromises the
structural integrity of the femoral head, and deformity occurs as the femoral
head is loaded. Once the femoral head is deformed, it is likely to
subsequently reossify and heal in the deformed shape.
A safe, simple medical treatment that can preserve the spherical shape of
the femoral head following ischemic necrosis would decrease the morbidity
associated with this condition by improving the longevity of the affected hip
and by decreasing the need for operative treatment. On the basis of the
resorptive changes observed on the radiographs of patients with
Legg-Calvé-Perthes disease in the fragmentation
stage26 and on the
basis of previous studies that have shown the predominance of bone resorption
during the early phase of repair in a piglet
model19,20,
we investigated the effectiveness of a potent antiresorptive agent
(ibandronate) for preserving the trabecular framework of the femoral head and
preventing femoral head deformity during repair. We found that, in this model,
the administration of ibandronate altered the repair process and decreased
bone loss. More importantly, the antiresorptive treatment helped to preserve
the structure of the femoral head within the time-course of the study. The
epiphyseal quotient and the trabecular bone parameters were significantly
better in the animals treated with ibandronate than they were in the animals
that had received saline solution. These early results are encouraging since
the piglet model is a model of severe femoral head deformity. In addition to
the repair process that favors the development of deformity, the piglets bore
weight ad libitum and they tripled their weight during the study period.
Following ischemic necrosis, the ideal repair process is one that preserves
the trabecular framework of the osseous epiphysis while stimulating new
appositional bone formation on the preserved trabecular scaffold, thus
maintaining the spherical shape of the femoral head. Studies of rabbit models
of ischemic necrosis have shown that this type of repair process is possible
naturally in a small animal
model27,28.
Our study showed that it is possible to alter the repair process following
ischemic necrosis in a piglet model by using the antiresorptive agent
ibandronate. A longer-term study is needed to determine whether appositional
new bone formation will occur on the preserved trabecular framework. If
ibandronate proves beneficial in such a study, one possible application of the
drug is to use it in the fragmentation stage of Legg-Calvé-Perthes
disease, when bone resorption predominates. Åstrand and
Aspenberg29 as well
as Little et al.30
reported preservation of the trabecular network and new appositional bone
formation on the necrotic bone following the use of zoledronate and
alendronate, respectively, in rats. It remains to be seen whether such a
repair process is possible in a large animal model, such as the piglet, and in
humans.
The optimal time at which to initiate ibandronate treatment following
ischemic necrosis of the femoral head is unknown. Our recent experiments using
14C-ibandronate to study the distribution of the radiolabeled drug
following parenteral administration revealed that the ibandronate is not
delivered into the infarcted femoral head until revascularization of the head
has been initiated, approximately three weeks after the induction of the
ischemia. This finding underscores the importance of carrying out additional
investigations to determine the pharmacokinetics of ibandronate as it applies
to the infarcted femoral head undergoing revascularization and repair. Such
investigations will allow optimization of the drug dose and dosing schedule,
including the ideal time at which to initiate treatment. This would reduce
unnecessary administration of the drug when the target tissue is not
accessible.
In the present study, we used relatively high doses of ibandronate because
of uncertainty regarding the effective dose required to prevent bone
resorption in a rapidly growing animal model and uncertainty regarding the
delivery of ibandronate into the ischemic femoral head. Since the delivery of
the drug is hindered by ischemia and since a rapidly growing animal may have
faster elimination of a drug bound on bone surface, we used a 44.4-µg/kg
dose in the post-ischemia treatment group. This schedule would correspond to a
daily dose of 14.3 µg/kg if a continuous daily dosing schedule was used.
That dosage is greater than the therapeutic dosage normally used in
ovariectomized adult
animals31,32
(1 to 10 µg/kg/day) and in patients with postmenopausal osteoporosis or
metastatic bone
disease33,34.
It is likely that the relatively high doses used in this study affected the
long-bone growth in the rapidly growing animals. In adult
ovariohysterectomized dogs, ibandronate at dosages of 10 µg/kg/day has been
shown to inhibit bone turnover, and thus bone formation, so that the measured
parameters were below control
values32.
Our data suggest that the growth inhibition may depend on the duration and
dose of the ibandronate treatment since the prophylactic treatment group,
which had longer exposure to the drug and a greater cumulative dose, had
greater growth inhibition than did the post-ischemia treatment group.
Decreased long-bone growth has been reported in studies in which other
bisphosphonates were used in growing animal
models35-38.
Alendronate and clodronate have been shown to decrease long-bone growth in
growing mice and rats, respectively, only when given in high
doses35,37.
It is important to point out that the piglets grew rapidly during the study
period, with a 53% increase in femoral length within the eight-week period.
The rapid growth may have accentuated the inhibitory effect of the high doses
of ibandronate on long-bone growth, which may not have been observed in an
animal with slower growth. Since children with Legg-Calvé-Perthes
disease do not grow rapidly during the early stage of the disease, and because
experience with treating children with bisphosphonates is limited, the
clinical relevance of the growth reduction found in the animal studies remains
unknown. Pamidronate and alendronate did not alter the growth rates of
children with osteogenesis imperfecta and osteoporosis secondary to diffuse
connective-tissue diseases,
respectively39,40.
It remains to be seen whether intermittent administration of ibandronate
during the fragmentation stage of Legg-Calvé-Perthes disease will
produce a clinically relevant decrease in long-bone growth.
In conclusion, ibandronate preserved the trabecular framework and prevented
femoral head deformity during the early phase of repair following ischemic
necrosis in a piglet model. Given the protective effects of ibandronate on the
infarcted femoral head during the early phase of repair, when osteoclastic
bone resorption predominates, additional preclinical studies are needed to
determine the optimal dose and delivery of the drug to prevent the deformity
while minimizing its effect on long-bone growth. Preclinical studies are also
necessary to determine the long-term effectiveness of ibandronate in
preventing the deformity. ?