Zoledronic acid improves femoral head sphericity

in a rat model of perthes disease

David G. Little

a,b,*

, Michelle McDonald

a

, Ian T. Sharpe

a,c

, Rachel Peat

a

,

Paul Williams

a,d

, Tony McEvoy

a,e

a

Orthopaedic Research and Biotechnology, The Children’s Hospital at Westmead 2145, Sydney, NSW, Australia

b

Department of Paediatrics and Child Health, University of Sydney, NSW 2006, Australia

c

Royal Devon and Exeter Healthcare NHS Trust, Royal Devon and Exeter Hospital (Wonford), Bowmoor House,

Barrack Road, Exeter EX2 5DW, UK

d

Morriston Hospital, Morriston Hospital NHS Trust, Swansea SA6 6NL, UK

e

Royal Preston Hospital, Sharoe Green Lane North, Fulwood, Preston, Lancashire PR2 4QF, UK

Received 14 October 2004; accepted 19 November 2004

Abstract

We hypothesized that the bisphosphonate zoledronic acid (ZA) could improve femoral head sphericity in Perthes disease by

changing the balance between bone resorption and new bone formation. This study tests the effect of ZA in an established model
of Perthes disease, the spontaneously hypertensive rat (SHR).

One hundred and twenty 4-week old SHR rats were divided into three groups of 40: saline monthly, 0.015 mg/kg ZA weekly, or

0.05 mg/kg ZA monthly. At 15 weeks DXA measurements documented that femoral head BMD was increased by 18% in ZA weekly
and 21% in ZA monthly compared to controls (p < 0.01). Femoral head sphericity in animals with osteonecrosis was improved in
ZA-treatment groups (p < 0.01) as measured by epiphyseal quotient (EQ). The proportion of ‘‘flat’’ heads (EQ 6 0.40) was signif-
icantly reduced from 32% in saline-treated animals to 12% in weekly ZA and 3% in monthly ZA (p < 0.01). Histologically there was
a similar prevalence of osteonecrosis in all groups. The prevalence of ossification delay was significantly reduced by ZA treatment
(p < 0.01).

Zoledronic acid favorably altered femoral head shape in this spontaneous model of osteonecrosis in growing rats. Translation of

these results to Perthes disease could mean that deformity of the femoral head may be modified in children, perhaps reducing the
need for surgical intervention in childhood and adult life.

2005 Orthopaedic Research Society. Published by Elsevier Ltd. All rights reserved.

Keywords: Perthes disease; Bisphosphonates; Osteonecrosis; Ossification delay; Histomorphometry

Introduction

Perthes disease is a common idiopathic form of osteo-

necrosis in childhood, with an incidence of 8.5–21 per
100,000 children per annum

[8,13,19]

. The etiology of

Perthes disease remains unknown, but the consensus
of opinion suggests that a vascular insult results in a per-
iod of relative ischemia and subsequent osteonecrosis

[9]

. Rapid resorption of this necrotic bone causes weak-

ening of the growing femoral head, leading to deformity
and eventual collapse. Although differing processes may
occur in different parts of the same femoral head at any
one time, bone formation fails to keep pace with resorp-
tion, leading to femoral head deformation

[3]

. The long-

term prognosis in Perthes disease is directly referable to

0736-0266/$ - see front matter

2005 Orthopaedic Research Society. Published by Elsevier Ltd. All rights reserved.

doi:10.1016/j.orthres.2004.11.015

*

Corresponding author. Address: Orthopaedic Research and Bio-

technology, The ChildrenÕs Hospital at Westmead 2145, Sydney, NSW,
Australia. Tel.: [phone]; fax: [phone].

E-mail address:

[e-mail]

(D.G. Little).

Journal of Orthopaedic Research 23 (2005) 862–868

www.elsevier.com/locate/orthres

femoral head sphericity. It is well documented that
aspherical heads with incongruent joints commonly pro-
gress to degenerative osteoarthritis of the hip relatively
early in adult life

[29]

. Therefore, maintenance of femo-

ral head sphericity has become a main focus in the treat-
ment of Perthes disease.

Current treatments aim to restore containment of the

femoral head, such that remodeling takes place in close
conformation with the acetabulum. Surgical contain-
ment procedures meet with limited or mixed success,
and the benefit of surgery remains controversial

[5,7,

24]

. Theoretically, it would be ideal if sphericity and

containment were never lost, or were at least minimized.
We hypothesized that by preventing or modifying
collapse of the necrotic bone with the use of bisphospho-
nate therapy, improved outcomes for children with Per-
thes disease could be achieved. In a study on traumatic
osteonecrosis in Wistar rats, we showed that zoledronic
acid (ZA, Novartis) significantly improved short-term
outcome

[18]

. In the current study we use a well-defined

model of spontaneous osteonecrosis, the spontaneously
hypertensive rat (SHR). Approximately 50% of SHR
rats suffer a spontaneous ischaemic insult and subse-
quent osteonecrosis of the femoral head epiphysis
around 6–9 weeks of age. As a result, considerable fem-
oral head flattening is seen by the age of 15 weeks

[10]

.

This insult occurs during the early to mid-stages of fem-
oral head development in these growing rats and there-
fore corresponds to a slightly earlier insult than that
seen in most Perthes patients.

Methods

Animal ethical approval was granted for this study (WAEC 103.06-

02). One hundred and twenty 4-week old rats were divided into three
groups of 40: 3 doses of saline monthly (saline), 3 doses of 0.05 mg/
kg ZA every four weeks (monthly ZA), or 10 doses of 0.015 mg/kg
ZA weekly (weekly ZA). The total dose of ZA at 15 weeks was
0.15 mg/kg in both ZA groups. Rats designated for undecalcified his-
tology (8 per group) were given subcutaneous injections of Calcein
10 mg/kg and Demeclocycline 30 mg/kg at 10 and 3 days prior to cull.

Animals were culled in a CO

2

chamber at 15 weeks. This end-point

age corresponding to that outlined by Hirano et al. as the time point of
considerable flattening and deformity seen in the SHR

[10]

. Both prox-

imal femora were harvested, radiographs were taken using a Faxitron
cabinet X-ray system (Hewlett Packard, McMinnville, OR). X-ray
images were digitized and enlarged 12X to be analyzed for sphericity
using Bioquant image analysis system (R & M Biometrics Inc., Nash-
ville, TN) connected to a digitizing tablet (GTCO CalComp, Inc.,
Columbia, MD). Sphericity measurements were derived using a mod-
ified epiphyseal quotient (EQ)

[20]

. The physis was horizontalized

and height at centre over width was recorded as EQ.

Femoral lengths were also measured on the digitized X-ray images.

Growth after the first ZA dose at 4 weeks was measured in the treated
animals from the mineralized lines that remained in the metaphyses
and their distance from the growth plate at 15 weeks.

Rat femora designated for decalcified histology (64 specimens per

group) were fixed in 4% paraformaldehyde for 24 h at 4

C, and then

decalcified in 0.5% paraformaldehyde/14.5% EDTA for approximately
6 weeks. The proximal femur of each sample was processed and embed-
ded in paraffin and 5 lm coronal sections through the center of the head
produced. Standard H&E stains were used for descriptive histology.

Safranin O, light green stain was employed for classification of remain-
ing cartilaginous tissue, Safranin O specifically binds cartilage matrix
staining it red, while bone tissue is stained green. Ossification of the
epiphysis proceeds from lateral to medial thus the percent of each epiph-
ysis that had ossified from the lateral side was calculated. Epiphyses
containing more than 50% cartilage were classified as showing signs
of delayed ossification (

Fig. 2

). The presence of osteonecrosis was deter-

mined by observational histology in the central slice, as seen by necrotic
osteocytes and empty lacunae. As the exact time at which the ischemic
insult occurred in each sample was unknown, many different stages of
osteonecrosis were noted within the population at the 15 week time
point. Thus, necrotic bone was described simply as the appearance of
numerous empty osteocyte lacunae, without any consideration to the
state of the surrounding marrow and accompanying reparative pro-
cesses. Statistical analysis of femoral head sphericity was reported for
only those femoral heads that exhibited signs of osteonecrosis.

Rat femora designated for undecalcified histology (16 specimens

per group) were stored at 4

C in 70% ethanol. Bone mineral density

(BMD) and bone mineral content (BMC) were measured with a
pDEXA Sabre scanner (Norland, Ft Atkinson, WI). The specimens
were then successively dehydrated in acetone and embedded in methyl-
methacrylate resin (Medim-Medizinische Diagnostik, Giessen, Ger-
many). Coronal sections (5 lm) were cut through the center of the
femoral head and fluorescent microscopy was used to determine bone
formation rates from the dual fluorochrome labels. Sections were
stained by von Kossa for histomorphometric data such as bone volume
(BV/TV), trabecular thickness (Tb.Th) and trabecular number (Tb.N).
Identification of osteonecrosis in the subset of animals undergoing
undecalcified histology could not be reliably performed, and this subset
of animals was not included in this outcome analysis. This subset was
included in the analysis of this purely mechanistic data, which reflects
the effect of ZA on SHR rats with and without osteonecrosis.

Statistics and power

For numeric data such as bone area, BMD and BV/TV, means and

standard deviations were calculated and one-way ANOVA applied,
with post-hoc t-tests by the least squares difference method (SPSS
Inc., Chicago, IL). For analysis of proportions, such as the prevalence
of osteonecrosis between groups, a proportional test (chi square) was
applied. Significance was set at the 0.05 level.

A power analysis based on EQ showed that power of 0.9 could be

achieved with a difference in EQ of 0.03 with 80 femoral heads. The
final number of 40 animals per group (80 femoral heads) was chosen
as an adequate sample.

Results

Mineralization

There was a range of femoral head deformity from

spherical to flat throughout the study population. Many
of the samples showed radiographic signs of ossification
delay. Observational analyses of the radiographs indi-
cated increases in the amount of epiphyseal ossification
in ZA-treated animals. This was confirmed in the 16
femoral heads in each group examined by DXA, with in-
creases in femoral head BMD measuring 18% in the
weekly ZA group and 21% in the monthly ZA group
over saline (p < 0.01).

Osteonecrosis

Osteonecrosis was observed in 53% of saline femoral

heads, 65% of weekly and 52% of monthly ZA femoral
heads. This was not significantly different between groups.

D.G. Little et al. / Journal of Orthopaedic Research 23 (2005) 862–868

863

Sphericity

Epiphyseal quotient (EQ) was significantly increased

in ZA-treated animals which showed evidence of an
osteonecrotic insult, with the mean EQ rising from
0.44 in the saline group to 0.47 in weekly ZA
(p < 0.05) and 0.49 in monthly ZA groups (p < 0.01).
The proportion of saline femoral heads in this cohort
that had an EQ less than or equal to 0.40 was 32%,
whereas this was reduced to 12% for weekly ZA
(p < 0.05) and 3% for monthly ZA groups (p < 0.01,

Fig. 1

). These EQ data were also significantly improved

for the ZA groups over saline for the entire cohort of
animals, ie those with and without osteonecrosis (data
not shown).

Ossification delay

Ossification delay is typical in SHR rats (and present

to some extent in Perthes disease). Areas of hypertrophic
chondrocytes were seen, surrounded by proteoglycan-
containing (Safranin O positive) matrix that was partly
mineralized on von Kossa sections. Invasion of blood
vessels and subsequent endochondral ossification was
seen to proceed from lateral to medial in the epiphyses
in these animals (

Fig. 2

A and B). There was a significant

difference in the incidence of ossification delay at 15
weeks between treatment groups in the femoral heads
affected by osteonecrosis. While 53% of the saline femo-
ral heads exhibited ossification delay, this was rare in
ZA-treated animals, with only 10% of weekly ZA and
21% of monthly ZA-treated animals exhibiting such
delay (p < 0.01) (

Table 1

). There was also a significant

decrease in the incidence of ossification delay in the en-
tire cohort of animals with ZA treatment (data not
shown).

In the subset of femoral heads with both osteonecro-

sis and ossification delay, there was a marked difference
in EQ with treatment. Saline-treated animals in this sub-
set had a mean EQ of only 0.40, while this was increased
to 0.49 (p < 0.01) in weekly ZA and 0.49 (p < 0.01) in
monthly ZA groups.

Histomorphometry

Static histomorphometric variables for all specimens

analyzed by undecalcified histology are listed in

Table

2

. There was a significant increase in BV/TV in ZA-trea-

ted animals based on increased trabecular number,
although the trabeculae were significantly thinner in
the ZA-treated groups when compared to saline (

Fig.

3

B, D and F). Bone formation rate (BFR) was signifi-

cantly reduced compared to saline in both ZA-treated
groups (post-hoc p < 0.01).

Growth disturbance

Femoral length was reduced 2.6% in the weekly and

2.3% in the monthly ZA group (p < 0.01) as compared
to saline. However, the actual amount of growth at
the distal femur from weeks 4–15 was 14.2 mm (SD
0.9) in weekly and 15.0 mm (SD 1.0) in monthly ZA
groups. This 5.4% reduction in longitudinal growth in
weekly versus monthly was significant (p < 0.01).

Fig. 1. Representative radiographs of femoral heads with an EQ of 0.5
(A) and 0.3 (B). Arrow shows region of flat head that is considerably
deformed. The mean EQ was higher, and the proportion of severely
flattened femoral heads was lower with ZA treatment.

Fig. 2. (A) Ossification delay with large areas of Safranin O staining (red*) cartilage still present in the medial half of the femoral head. The femoral
head in (B) is fully ossified (green tissue). Small arrow in A points to the vascular front of the ossification. Large arrow in B shows complete
ossification of epiphysis. Safranin O, light green (

·2.5).

864

D.G. Little et al. / Journal of Orthopaedic Research 23 (2005) 862–868

Table 1
Proportion of fully ossified femoral heads

a

Saline

Weekly

Monthly

n

%

n

%

n

%

Ossified

16

47

37

90

*

26

79

*

Not ossified

18

53

4

10

*

7

21

*

Total

34

100

41

100

33

100

*

p < 0.01 v saline.

a

Decalcified specimens with osteonecrosis only.

Table 2
Histomorphometry

a

Saline

Weekly

Monthly

Mean

SD

Mean

SD

Mean

SD

Bone volume (BV/TV) (%)

62.8

8.5

69.2

*

8.6

69.9

*

8.4

Trabecular number (n/mm)

9.0

2.1

12.3

*

1.7

13.4

*

2.7

Trabecular thickness (lm)

72.6

16.2

55.2

*

20.0

57.8

*

14.2

Bone formation rate (lm

2

/lm/day)

0.47

0.28

0.20

*

0.13

0.25

*

0.15

*

p < 0.05 v saline.

a

Undecalcified specimens only n = 16 in each group.

Fig. 3. Representative radiographs (A,C,E) and von Kossa sections magnification

·2.5 (B,D,F) from saline (A,B), weekly ZA (C,D) and monthly ZA

(E,F) groups. Increased apparent density in ZA-treated groups on radiographs was due to increased trabecular number. Trabecular thickness was
higher in the saline group.

D.G. Little et al. / Journal of Orthopaedic Research 23 (2005) 862–868

865

Discussion

Zoledronic acid treatment improves sphericity in SHR
rats

We have demonstrated a significant effect of ZA on

sphericity of the SHR rat femoral head in this study,
with both an increase in mean EQ and a decrease in
the number of heads defined as aspherical (EQ 6 0.4).
As hypothesized, enhanced mineralization was demon-
strated in SHR rats treated with ZA. We also found
increased bone volume (BV/TV) and an increase in
trabecular number (

Fig. 3

;

Table 2

). These outcome

and mechanistic data are in agreement with previous
animal studies on the effect of nitrogen-containing bis-
phosphonates in models of osteonecrosis.

In a rat bone chamber study, Astrand and Aspenberg

showed that high dose alendronate therapy was able to
preserve necrotic bone volume as a scaffold for new bone
ingrowth

[1]

. Total bone volume was doubled in the high

dose alendronate group compared to low dose and saline
animals. In a study by our group, Wistar rats with trau-
matic osteonecrosis treated with ZA improved femoral
head shape and showed significant increase in trabecular
number over saline animals

[18]

. In a large animal piglet

model, indentation studies showed a reduction in stiff-
ness of the infarcted femoral head by 52% at 2 weeks
and 75% by 4 weeks, leading to considerable deformity
by 8 weeks

[23]

. In the same model, Kim et al.

[15,16]

have shown positive effects on femoral head shape and
retention of trabecular architecture with ibandronate.
Trabecular number was again increased in this study.

It is of interest that in the current study, trabecular

thickness was larger in saline-treated controls, this
may suggest hypertrophy of those remaining trabeculae
which are not resorbed. Grossly thickened trabeculae
have also been noted in human pathological specimens
of Perthes disease

[4]

. It is particularly noteworthy that,

for the same amount of bone, maintaining trabecular
number by thinning results in better strength retention
in trabecular bone than does losing trabeculae but
retaining trabecular thickness

[28]

. It is extremely likely

that the retention of trabecular number significantly
contributed to the improved retention of overall femoral
head shape.

Effect of zoledronic acid on ossification delay in
SHR rats

An unexpected outcome of this study was the increase

in the occurrence of complete epiphyseal ossification in
the femoral heads of the ZA-treated rats (

Fig. 2

A and

B;

Table 1

). This was an impressive outcome, with the

53% incidence of ossification delay in femoral heads of
saline-treated animals reduced to 10% in weekly and
21% in monthly ZA-treated animals. This indicates that

in this model, osteoclast function may not be absolutely
required in the process of endochondral ossification of
the epiphysis. Further work is needed to fully eluci-
date the mechanism behind the accelerated epiphyseal
ossification, but it has already been established that
osteoclasts (and chondroclasts) are only required for
remodeling primary bone trabeculae in endochondral
ossification, not for the removal of chondrocytes and
unmineralized chondral matrix

[6,25,26]

.

Limitations of the SHR model

We chose the spontaneously hypertensive rat model

for this investigation as it more closely resembles Perthes
disease than surgically induced models. Extensive stud-
ies have been carried out on the occurrence of spontane-
ous osteonecrosis in the SHR rat. It is of note that
weight bearing is a prerequisite for osteonecrosis in this
model as osteonecrosis is not observed if the limb is
denervated, or if amputation is carried out through the
knee

[12]

. Around 50% of male SHR rats develop avas-

cular changes at around 6–9 weeks of age leading to
maximal femoral head deformity by 15 weeks

[10]

. A

limitation of this model is the difficulty in determining
which femoral heads had suffered an ischemic insult.
At the 15 week end time point in this study we were able
to histologically detect the occurrence of osteonecrosis
in the samples. However this excludes those samples that
had completely recovered from the insult by this time
point.

It has been documented that SHR rats have a carti-

lage disorder that results in a delay in ossification of
the femoral head epiphysis

[14]

. Thus, at the age of 6–

9 weeks the underdeveloped head cannot withstand nor-
mal weight bearing, resulting in occlusion of the lateral
epiphyseal vessels (LEVÕs) and subsequent ischemia and
osteonecrosis. Resorption of the necrotic bone in the lat-
eral epiphysis ensues, resulting in reduced mechanical
properties and deformation, thus further vascular occlu-
sion. Epiphyseal ossification is therefore further delayed.

We speculate that ZA treatment stabilized the lateral

epiphysis by suppressing the resorption of necrotic bone.
The un-resorbed trabeculae may have provided the
mechanical strength required by the weight bearing lat-
eral epiphysis to resist further deformation. ZA treat-
ment may have thus prevented the weight-bearing
induced compression of the LEVÕs known to be the
pathogenesis of osteonecrosis in this rat model

[11,21]

.

Therefore, the vascular endothelial cells were unhin-
dered in their vascular invasion of the cartilaginous
epiphysis, thus subsequent endochondral ossification
of the femoral heads was able to proceed

[17]

. Further

experiments and analysis will be employed to evaluate
this hypothesis. As ZA was administered from week 4,
the drug may have had a prophylactic effect. We showed
that prophylaxis was more effective than treatment in

866

D.G. Little et al. / Journal of Orthopaedic Research 23 (2005) 862–868

our previous study on traumatic osteonecrosis

[18]

,

assessment of further dosing time-points might also be
relevant.

Similar processes may or may not be occurring in

Perthes disease, and this could explain why some clini-
cal specimens display multiple infarctions. Whether the
ossification delay demonstrated by delayed bone age in
Perthes disease

[27]

is also relevant to the delay in reo-

ssification after ischemia is unknown. In some animals
in this study with ossification delay, the presence of
hypertrophic cartilage in the femoral head was due to
primary ossification delay, rather than fibrocartilage re-
pair tissue as seen in Perthes disease after bone resorp-
tion. However, in Perthes disease, deformity occurs
when there is a mixture of necrotic bone and cartilage
in the femoral head. The SHR rat can still be viewed
as a valid model of this particular circumstance. This
mixture of cartilage and necrotic bone would be the
‘‘softest’’ group of femoral heads, and it is relevant that
in this subset, ZA treatment increased EQ from 0.40 to
0.49.

We adjusted for the various limitations by incorpo-

rating a large number of animals in the study. It remains
very important to study spontaneously induced osteone-
crosis in a growing animal, as this has far more rele-
vance to Perthes disease than traumatic models. The
fact that we have similar results and mechanistic data
(increased BV/TV, increased trabecular number) in a
traumatic osteonecrosis model in rats gives us encour-
agement that ZA treatment may be relevant to both cir-
cumstances

[18]

.

Potential negative effects of bisphosphonates

Bisphosphonate therapy in children has one poten-

tially worrying negative effect—a decrease in longitu-
dinal growth. In this study a 2.3–2.6% decrease in
femoral length with ZA treatment was documented.
The animals approximately double the size of the femur
during the experimental period, so the growth distur-
bance can be approximated at around 5% inhibition.
We also noted a further 5% decrease in longitudinal
growth in weekly over monthly dosing. Close monitor-
ing of longitudinal growth should be performed in all
children on bisphosphonates to provide further defini-
tive information on this topic, however clinical studies
in children with osteogenesis imperfecta, osteoporosis
and fibrous dysplasia show no detectable growth distur-
bance

[2,22,31]

.

As bisphosphonates avidly bind bone mineral, they

have a very long half-life in the skeleton. The effect on
bone metabolism is to reduce bone turnover, but when
treatment is ceased in adults, slow reversal of the effects
are seen

[30]

. Further safety data in children will be

needed before the application of this potential therapy
in Perthes disease.

Summary

This and other studies suggest that a treatment ap-

proach utilizing bisphosphonates could prove a valuable
adjunct in childhood osteonecrosis. As the safety of po-
tent bisphosphonates becomes more established, it is
possible that these compounds could be used to improve
outcome and decrease the number of surgical proce-
dures in Perthes disease. Further experimentation in this
and other models of Perthes disease and eventual clinical
trials will be needed to evaluate the efficacy and safety of
potent bisphosphonate therapy in preserving femoral
head shape while repair occurs.

Acknowledgement

Funding for this project was received from The Na-

tional Health and Medical Research Council of Austra-
lia. Dr Ian Sharpe and Dr Paul Williams received salary
support as Ingham Fellows funded by Ingham Enter-
prises Pty Ltd. Dr Tony McEvoy received salary
support as the Smith & Nephew Fellow funded by
Smith and Nephew Orthopaedics. Zoledronic acid was
donated by Novartis Pharmaceuticals (Australia). Dr
David Little has received patent licensing funding and
other research grants from Novartis Pharma AG. We
would like to thank Lyndon Mason for his assistance
with the femoral length measurements.

References

[1] Astrand J, Aspenberg P. Systemic alendronate prevents resorption

of necrotic bone during revascularization. A bone chamber study
in rats. BMC Musculoskelet Disord 2002;3:19–23.

[2] Brumsen C, Hamdy NA, Papapoulos SE. Long-term effects of

bisphosphonates on the growing skeleton. Studies of young
patients with severe osteoporosis. Medicine 1997;76:266–83.

[3] Catterall A, Pringle J, Byers PD, Fulford GE, Kemp HB. PerthesÕ

disease: is the epiphysial infarction complete? J Bone Joint Surg Br
1982;64:276–81.

[4] Catterall A, Pringle J, Byers PD, Fulford GE, Kemp HB, Dolman

CL, et al. A review of the morphology of PerthesÕ disease. J Bone
Joint Surg Br 1982;64:269–75.

[5] Daly K, Bruce C, Catterall A. Lateral shelf acetabuloplasty in

PerthesÕ disease. A review of the end of growth. J Bone Joint Surg
Br 1999;81:380–4.

[6] Deckers MM, Van Beek ER, Van Der Pluijm G, Wetterwald A,

Van Der Wee-Pals L, Cecchini MG, et al. Dissociation of
angiogenesis and osteoclastogenesis during endochondral bone
formation in neonatal mice. J Bone Miner Res 2002;17:998–
1007.

[7] Friedlander JK, Weiner DS. Radiographic results of proximal

femoral varus osteotomy in Legg–Calve–Perthes disease. J Pediatr
Orthop 2000;20:566–71.

[8] Hall AJ, Barker DJ, Dangerfield PH, Taylor JF. PerthesÕ disease

of the hip in Liverpool. Br Med J(Clin Res Ed) 1983;10:1757–9.

[9] Herring JA. Management of PerthesÕ disease. J Pediatr Orthop

1996;16:1–2.

D.G. Little et al. / Journal of Orthopaedic Research 23 (2005) 862–868

867

[10] Hirano T, Iwasaki K, Yamane Y. Osteonecrosis of the femoral

head of growing, spontaneously hypertensive rats. Acta Orthop
Scand 1988;59:530–5.

[11] Hirano T, Majima R, Yoshida G, Iwasaki K. Characteristics of

blood vessels feeding the femoral head liable to osteonecrosis in
spontaneously hypertensive rats. Calcif Tissue Int 1996;58:201–5.

[12] Iwasaki K, Hirano T, Sagara K, Nishimura Y. Idiopathic necrosis

of the femoral epiphyseal nucleus in rats. Clin Orthop 1992;277:
31–40.

[13] Kealey WD, Moore AJ, Cook S, Cosgrove AP. Deprivation,

urbanisation and PerthesÕ disease in Northern Ireland. J Bone
Joint Surg Br 2000;82:167–71.

[14] Kikkawa M, Imai S, Hukuda S. Altered post-natal expression of

insulin-like growth factor-I (IGF-I) and type X collagen preceding
the PerthesÕ disease like lesion of a rat model. J Bone Miner Res
2000;15:111–9.

[15] Kim HKW, Pringle D, Hunter D, Pritchard W, Bauss F.

Ibandronate decreases femoral head flattening in a piglet model
of ischemic necrosis of femoral head. J Bone Miner Res 2002;
17:S475.

[16] Kim HKW, Randall TS, Bian H, Jenkins J, Garces A, Bauss F.

Ibandronate prevents femoral head deformity following ischemic
necrosis of the capital femoral epiphysis in immature pigs. JBJS,
in press.

[17] Lewinson D, Silbermann M. Chondroclasts and endothelial cells

collaborate in the process of cartilage resorption. Anat Rec 1992;
233:504–14.

[18] Little DG, Peat RA, McEvoy A, Williams PR, Smith EJ, Baldock

PA. Zoledronic acid treatment results in retention of femoral head
structure after traumatic osteonecrosis in young Wistar rats.
J Bone Miner Res 2003;18:2016–22.

[19] Moberg A, Rehnberg L. Incidence of PerthesÕ disease in Uppsala,

Sweden. Acta Orthop Scand 1992;63:157–8.

[20] Mose K. Methods of measuring in Legg–Calve–Perthes disease

with special regard to the prognosis. Clin Orthop 1980;150:103–9.

[21] Oda J, Hirano T, Iwasaki K, Majima R. Vascular occlusion and

cartilage disorders in osteonecrosis of the femoral head in rats. Int
Orthop 1996;20:185–9.

[22] Plotkin H, Rauch F, Zeitlin L, Munns C, Travers R, Glorieux

FH. Effect of pamidronate treatment in children with polyostotic
fibrous dysplasia of bone. J Clin Endocrinol Metab 2003;88:
4569–75.

[23] Pringle D, Koob TJ, Kim HK. Indentation properties of growing

femoral head following ischemic necrosis. J Orthop Res 2004;22:
122–30.

[24] Robinson Jr HJ, Putter H, Sigmond MB, OÕConnor S, Murray

KR. Innominate osteotomy in Perthes disease. J Pediatr Orthop
1988;8:426–35.

[25] Schenk RK, Wiener J, Spiro D. Fine structural aspects of vascular

invasion of the tibial epiphyseal plate of growing rats. Acta Anat
(Basel) 1968;69:1–17.

[26] Schenk RK, Spiro D, Wiener J. Cartilage resorption in the tibial

epiphyseal plate of growing rats. J Cell Biol 1967;34:275–91.

[27] Shapiro F. Legg–Calve–Perthes disease: a study of lower extre-

mity length discrepancies and skeletal maturation. Acta Orthop
Scand 1982;53:437–44.

[28] Silva MJ, Gibson LJ. Modeling the mechanical behavior of

vertebral trabecular bone: effects of age-related changes in
microstructure. Bone 1997;21:191–9.

[29] Stulberg SD, Cooperman DR, Wallensten R. The natural history

of Legg–Calve–Perthes disease. J Bone Joint Surg Am 1981;63:
1095–108.

[30] The Alendronate Phase III Osteoporosis Treatment Study Group.

Ten yearsÕ experience with alendronate for osteoporosis in post-
menopausal women. Obstet Gynecol Surv 2004;59:597–8.

[31] Zeitlin L, Rauch F, Plotkin H, Glorieux FH. Height and

weight development during four years of therapy with cyclical
intravenous pamidronate in children and adolescents with
osteogenesis imperfecta types I, III, and IV. Pediatrics 2003;111:
1030–6.

868

D.G. Little et al. / Journal of Orthopaedic Research 23 (2005) 862–868