Calcified Tissue International

https://doi.org/10.1007/s00223-019-00615-z

ORIGINAL RESEARCH

Pretreatment with Pamidronate Decreases Bone Formation

but Increases Callus Bone Volume in a Rat Closed Fracture Model
Alyson Morse1,2 · Michelle M. McDonald1,3 · Kathy Mikulec1 · Aaron Schindeler1,2 · Craig F. Munns2,4 ·
David G. Little1,2

Received: 18 June 2019 / Accepted: 5 August 2019

© Crown 2019

Abstract
Clinical concerns have been raised over prior exposure to bisphosphonates impairing fracture healing. To model this, groups
of male Wistar rats were assigned to saline control or treatment groups receiving 0.15 mg/kg (low dose), 0.5 mg/kg (medium
dose), and 5 mg/kg (high dose) Pamidronate (PAM) twice weekly for 4 weeks. At this point, closed fractures were made
using an Einhorn apparatus, and bisphosphonate dosing was continued until the experimental endpoint. Specimens were
analyzed at 2 and 6 weeks (N = 8 per group per time point). Twice weekly PAM dosing was found to have no effect on early
soft callus remodeling at 2 weeks post fracture. At this time point, the highest dose PAM group gave significant increases in
bone volume (+ 10%, p < 0.05), bone mineral content (+ 30%, p < 0.01), and bone mineral density (+ 10%, p < 0.01). This
PAM dosing regimen showed more substantive effects on hard callus at 6 weeks post fracture, with PAM treatment groups
showing + 46–79% increased bone volume. Dynamic bone labeling showed reduced calcein signal in the PAM-treated cal-
luses (38–63%, p < 0.01) and reduced MAR (32–49%, p < 0.01), suggesting a compensatory reduction in bone anabolism.
These data support the concept that bisphosphonates lead to profound decreases in bone turnover in fracture repair, however,
this does not affect soft callus remodeling.

Keywords  Bisphosphonate · Fracture · Pamidronate · Endochondral ossification · Osteogenesis imperfecta · Atypical

femoral fracture

Introduction and at higher doses to prevent secondary skeletal-related

Bisphosphonates (BPs) are a class of anti-resorptive bone
drugs that are widely used in the management of osteoporo-
sis and metabolic bone disorders in adults and children [1,
2]. In adult patients, they are widely used to reduce fracture
risk and associated morbidity and mortality for osteoporosis,
* David G. Little
david.little@health.nsw.gov.au
1
Orthopaedic Research & Biotechnology Unit, The Children’s
Hospital at Westmead, Locked Bag 4001, Westmead,
NSW 2145, Australia
2
Discipline of Child and Adolescent Health, Faculty
of Medicine and Health, University of Sydney, Sydney,
NSW, Australia
3
Bone Biology Division, The Garvan Institute for Medical
Research, Sydney, Australia
4
Department for Endocrinology and Diabetes, The Children’s
Hospital at Westmead, Westmead, NSW, Australia
events associated with metastatic cancers. In pediatrics, they
are routinely prescribed off-label to manage osteogenesis
imperfecta, idiopathic low bone density, hypercalcemia, and
heterotopic calcifications. Despite their widespread use in
these clinical settings, it is unclear how BPs may interact
with the process of fracture repair, and whether it fundamen-
tally differs from their usage in BP-naïve patients.
Fracture repair is a complex process responsible for facili-
tating the regeneration of bone without the development of
scar tissue. Endochondral bone repair has been historically
described as a sequence of events: inflammation, soft (car-
tilaginous) callus formation, replacement with hard (bony)
callus, and bone remodeling, but subsequent models of frac-
ture healing have focused on anabolic-catabolic processes
[3] or cellular drivers [4]. BPs are known to profoundly
interfere with the remodeling phase of fracture repair, inter-
fering with osteoclastic bone resorption [5].
Numerous preclinical studies featuring fracture models
concomitantly dosed with BPs have confirmed impaired

13
Vol.:(0123456789)

remodeling of the hard callus [6–8]. It was initially specu-
lated that BPs may impede chondroclast activity [9, 10], thus
negatively impacting on bony union. However, a seminal 2013
study revealed that BP treatment enhanced woven bone callus
size without impeding soft callus removal [6]. It was specu-
lated that soft callus breakdown was enabled by non-osteo-
clastic cells able to express matrix metalloproteases (MMPs).
Consistent with this concept, MMP inhibitor treatment hin-
dered cartilage resorption [6], a finding also supported by data
from MMP-9 and MMP-13 knockout models [11, 12]. To date,
preclinical models have not substantively addressed the effect
of chronic BP dosing on fracture healing. Prolonged high-dose
BP treatment was employed in a rat model of cyclic loading
induced stress fractures [13]. The group reported impaired
stress fracture remodeling at high-dose BPs, but even this
study featured concomitant treatment with risedronate during
cyclic loading rather than pretreatment.
The issue of chronic exposure is one with broad relevance
to the clinical use of BPs, as their effects on bone healing may
be divergent when comparing acute use during bone repair
with patients on long-term BP dosing schedules. Atypical fem-
oral fractures have been associated with long-term BP dosing
and represent spontaneous or low-energy fractures with poor
subsequent bone healing [14, 15]. In osteogenesis imperfecta,
it has been suggested that delayed union can occur in a subset
of children treated with Pamidronate (PAM) [16]. A recent
Cochrane review further highlights the need for additional
studies to examine the safety and efficacy of PAM in osteo-
genesis imperfecta [17]. Mechanistically, coupling between
bone catabolism and anabolism suggests that prolonged anti-
resorptive treatments can lead to a generalized depression of
bone formation necessary for bone healing. This effect and
its medical impact can be challenging to explore in a clinical
setting. Hence, we speculated that a rodent model of BP pre-
dosing could provide important insights into its bearing on the
fracture repair process.
Closed fractures were produced using an Einhorn appa-
ratus in rats pre-treated for 4 weeks with PAM administered
twice weekly at low (0.15 mg/kg), medium (0.5 mg/kg), and
high (5 mg/kg) doses. It was hypothesized that (1) no PAM
dose would impair of endochondral ossification, (2) hard
callus remodeling would be impaired by PAM in a dose-
dependent manner, and (3) pre-dosing with the higher doses
of PAM may result in a significant corresponding decrease
in new bone formation.
Methods
Animals and PAM Administration
Seventy male Wistar rats were housed from the age of
5 weeks and all procedures were approved by the Westmead
13
A. Morse et al.
Hospital Animal Ethics Committee with approval number
5024.06-07. Rats were randomly allocated into four treat-
ment groups and doses were administered subcutaneously
commencing at 5 weeks of age. Control animals received
twice weekly saline injections. Pamidronate disodium
(Novartis Pharma AG, Basil, Switzerland) was prepared
in sterile water. The three PAM treatment groups received
low, medium, and high doses of 0.15 mg/kg, 0.5 mg/kg,
and 5 mg/kg PAM twice weekly (biweekly/biw). Following
4 weeks of PAM treatment, fracture surgery was performed
when rats were 9 weeks of age.
Fracture Surgery, Monitoring, and Specimen
Collection
Using a combination of 75 mg/kg ketamine and 4 mg/kg
xylazine for anesthesia, a small incision was made ante-
rior to the patella of the right distal femur. The patellar was
moved slightly lateral to reveal the distal femoral condyles
and a 1.1-mm Kirschner wire was inserted into the medul-
lary canal from the femoral condyles until it slightly pro-
truded proximal to the greater trochanter. A model of the
Einhorn 3-point drop weight fracture apparatus was used to
create a clean transverse mid-diaphyseal closed femoral frac-
ture [18]. The MX20 digital Faxitron X-ray system (Faxitron
X-ray Corporation, Illinois, USA), was used to visualize all
fractures. Rats with fractures that were not mid-diaphyseal
or comminuted were euthanized and excluded (N = 1). For
pain relief buprenorphine (Temgesic, Reckitt Benckiser) was
administered as a subcutaneous dose of 0.05 mg/kg post-
operatively and daily monitoring of rats was carried out
throughout the course of the study. During the study, three
animals were culled due to worsening health and/or evidence
of infection and were excluded from subsequent analysis.
Calcein was administered at a dose of 10 mg/kg subcuta-
neously at 10 and 3 days prior to harvest to allow for analy-
sis of dynamic bone formation. Harvest time points were at
either 2 or 6 weeks after fracture surgery to examine soft and
hard callus remodeling stages, respectively. Left and right
femora and tibiae were harvested and fixed in 10% formalin
for 4 h at room temperature followed by 24 h at 4 °C, and
then stored in 70% ethanol at 4 °C.
Radiological Analysis
Longitudinal radiographs were performed using the digital
MX 20 Faxitron system on both right and left femora. The
fractured femora were graded as bridged or not united at
6 weeks by a blinded observer to assess radiological fracture
union. Femoral length was measured on all intact limbs.
Quantitative Computerized Tomography (QCT) was per-
formed on both right and left femora of all samples using a
Stratec xQCT research scanner (Norland Stratec, Durlacher,
Pretreatment with Pamidronate Decreases Bone Formation but Increases Callus Bone Volume in…
Germany). Right (fractured) femora underwent scans at a
voxel-size of 0.07 mm and a scan speed of 5 s with 91-mm-
thick slices taken with a slice separation of 1 mm through
the entire fracture callus. For analysis of the callus, cortical
bone was manually excluded and bone selected at a consist-
ent threshold for all specimens. Left (unfractured) femora
were scanned with the same voxel size and scan speed but
only two slices were measured, with one slice placed 8 mm
from the distal femoral boundary, measuring metaphyseal
bone, and the other placed in the mid-diaphyseal region.
Data from the diaphyseal slice of the left femora were multi-
plied by nine to extrapolate data for the comparison with the
contralateral limb. Data for Bone Mineral Content (BMC),
Bone Volume (BV), Bone Mineral Density (BMD), and
Moment of Inertia (MMI) were generated for each sample.
Cortical Thickness, Periosteal Circumference, and Endosteal
Circumference are reported for the diaphysis of intact legs.
Histology and Histomorphometry
Right and left femora were bisected into lateral and medial
halves. The lateral halves were decalcified along with the
left tibia in 0.39M EDTA in 1.25% formalin (pH 8.0) at
room temperature (RT) for 3 days, with further decalcifica-
tion in 0.34M EDTA (pH 8.0) for 2 weeks at RT, before
being processed and embedded in paraffin wax. The medial
halves of fractures and right tibia were processed through
graded acetones and embedded in methyl methacrylate resin
for un-decalcified histology. Sagittal sections of fractures
and coronal sections of tibia, 5 µm thick, were produced for
each sample on a Leica RM2155 semi-automated microtome
(Leica, Germany).
Right (Fractured) Femora Specimens
Decalcified callus sections were stained with Alcian Blue/
Picrosirius Red for cartilaginous and ossified callus mor-
phology, and adjacent sections were stained with Safranin O
and counterstained with 0.4% Light Green to analyze carti-
lage. Adjacent sections were also stained for tartrate-resist-
ant acid phosphatase (TRAP) expression, and counterstained
with Hematoxylin, to highlight and analyze osteoclasts.
Image capture and analysis were performed using BIO-
QUANT measure 32 Nova prime (Nashville, Texas) image
analysis system connected to a QICAM Fast 1394 color 12
bit camera (Quantitative Imaging Corporation, Canada).
Osteoclast number was counted within a controlled region
of interest in the callus adjacent to the fracture site in sam-
ples at 6 weeks post fracture only. The total bone surface
within this region was also measured, as well as the bone
surface in contact with osteoclasts, allowing calculation of
a percent osteoclast surface compared to total bone surface.
Resin-processed callus sections were Von Kossa stained
to allow analysis of trabecular and cortical bone in the cal-
lus. Bone volume within the primary trabecular regions was
then determined and BV/TV of the primary trabecular callus
was calculated. Adjacent sections were cleared for fluores-
cent analysis, which included quantification of the percent-
age fluorescent labeled bone in the total fracture callus. This
was determined using the total bone volume of each callus
as measured using adjacent Von Kossa-stained sections. The
distance between double labeling in the trabecular bone of
the callus was also measured in the 6W samples, generat-
ing data on mineral apposition rate (MAR). At 2 weeks, the
ossifying callus is not mature enough to show clear double
labels.
Left (Contralateral) Femora Specimens
Resin sections of the left femoral diaphyses were cleared
for analysis of MAR on both the periosteal and endosteal
surfaces of each sample. Mineralizing Surface was also
determined for both the periosteal and endosteal surfaces
after measuring the length of the single label surface (SLS),
double label surface (DLS), and bone surface. The formula
[(DLS + SLS/2)/bone surface] was used.
Statistics
For all data, means and standard deviations were calculated
and a one-way analysis of variance (ANOVA) applied with
post hoc t tests by the least squares difference method (SPSS
Inc Chicago, IL). The cutoff for statistical significance was
p < 0.05.
Results
Chronic PAM Dosing Does not Impair Early
Endochondral Bone Repair
At 2 weeks post fracture only high-dose PAM produced
minor but significant differences in radiographic callus size
compared to saline controls. In the high-dose PAM group,
QCT showed that callus BMC was increased by + 30%
(p < 0.01), callus bone volume by + 10% (p < 0.05), and
callus BMD by + 10% (p < 0.01) (Table 1). However, this
did not translate to a significant increase in callus moment
of inertia.
Histological analysis was performed on the 2  week
fracture specimens. The BV/TV of the ossified callus was
significantly increased in all of the PAM treatment groups
(Fig. 1a). PAM treatment was associated with no change
in the % of bone containing fluorescent label in the callus
(Fig. 1b); however, the average femur length was decreased
13
 A. Morse et al.

with higher doses of PAM (Fig. 1c). At the 2 week time in soft callus area (21–28%, N.S.) between the saline con-
point, cartilaginous tissue was present in the soft callus for trols and the 3 PAM treatment groups.
all groups (Fig. 1d), and quantification showed no difference
Chronic PAM Dosing Increases Callus Size at 6 Weeks
but Reduces the Formation of New Bone
Table 1  QCT parameters in fractured femora
2W right femora 6W right femora Radiographic analysis by an orthopedic surgeon blinded
Mean SD Mean SD to treatment group confirmed that all fractures were radio-
graphically united by 6 weeks. The effects of PAM on cal-
Callus BMC
lus size and density were examined by QCT (Table 1). By
 Saline 128.53 8.24 140.04 16.34
6 weeks post fracture, callus BV and BMD were signifi-
 PAM 0.15 mg/kg 146.35 25.43 217.61* 37.56
cantly increased with all levels of PAM treatment over saline
 PAM 0.5 mg/kg 144.54 21.35 231.24* 41.54
by 46–79% and 55–94%, respectively (p < 0.01). Calcula-
 PAM 5 mg/kg 166.68* 30.86 272.22* 47.13
tions of moment of inertia values similarly showed increases
Callus BMD
with high dose PAM over saline of + 127% (p < 0.01).
 Saline 717.9 22.2 917.4 105.9
Histological analysis of the 6-week specimens showed
 PAM 0.15 mg/kg 738.8 61.4 956.6 45.8
ossified callus BV/TV was increased with PAM treat-
 PAM 0.5 mg/kg 729.1 57.1 1030.9* 67.3
ment (Fig. 2a), similar to the 2-week samples. However,
 PAM 5 mg/kg 790.8* 51.9 985.3 47.0
by 6 weeks, the fluorescent label in the callus was signifi-
Callus BV
cantly reduced in all PAM treatment groups, indicating a
 Saline 182.26 11.02 156.01 31.64
compensatory decease in bone anabolism by this time point
 PAM 0.15 mg/kg 200.05 28.53 229.72* 45.42
(Fig. 2b). Despite suppressed bone formation, histological
 PAM 0.5 mg/kg 200.82 25.48 228.42* 55.38
staining with Alcian blue/Picrosirius red showed no evi-
 PAM 5 mg/kg 214.3 42.01 279.34* 54.04
dence of disrupted soft callus removal and retention of
MMI
cartilage (Fig. 2d). Nevertheless, hard callus remodeling
 Saline 102.1 15.8 78.8 33.5
was clearly disrupted by PAM, particularly in the PAM 5
 PAM 0.15 mg/kg 121.7 34.4 124.9 47.6
group with retention of a larger less remodeled hard callus.
 PAM 0.5 mg/kg 124.5 32.6 120.5 58.3
A trend was seen for decreased femur length in PAM-treated
 PAM 5 mg/kg 133.2 06.3 177.1* 62.0
groups, however, only PAM 0.5 showed a significant reduc-
*p < 0.05 tion (Fig. 2c).

Fig. 1  2W histology. Measurements of a BV/TV, b callus fluorescence, and c femur length were quantified from tissue sections. *p < 0.05,
**p < 0.01. d Histological images of closed fractures stained with Alcian blue/Picrosirius red

13
Pretreatment with Pamidronate Decreases Bone Formation but Increases Callus Bone Volume in…
Fig. 2  6W histology. Measurements of a BV/TV, b callus fluorescence, and c femur length were quantified from tissue sections. *p < 0.05,
**p < 0.01. d Histological images of closed fractures stained with Alcian blue/Picrosirius red
More detailed bone histomorphometry was performed
for further cellular mechanisms. PAM at 0.25, 0.5, and 5.0
reduced the callus MAR at 6 weeks by 32%, 38%, and 49%,
respectively (p < 0.01, Fig. 3a). Analysis of TRAP + osteo-
clast staining within an ROI of the callus showed increased
osteoclast number with PAM treatment (Fig. 3b), which has
been previously reported for BPs [19]. However, measures
that consider the osteoclast surface show a proportional
decrease in osteoclasts lining the bone relative to osteoclast
number (Figs. 3c, d).
The Effect of PAM Treatment on Intact Bones
In order to appreciate the effects of chronic BP treatment on
bone healing, the contralateral femora and the tibiae were
measured radiographically.
PAM treatment resulted in small changes in femur length
that were nonetheless statistically significant. At the study
endpoint (10 weeks of dosing), femur length was decreased
by an average of 2% with medium-dosed PAM (p < 0.05)
and 3% with high-dosed PAM (p < 0.01) compared to saline.
Growth plate height was assessed in all proximal tibiae and
revealed no changes with PAM treatment (data not shown).
Radiographic (QCT) analysis performed on the contralat-
eral femora at the study endpoint showed a significant + 9%
increase in BMC with high-dose PAM compared to saline
(p < 0.05) (Table 2). Bone volume was also increased by
+ 14% with low-dose PAM (p < 0.05) and 19% with high-
dose PAM compared to Saline (p < 0.01). Notably, cortical
bone BMD was significantly decreased 5–6% by 6 weeks
in all three PAM treatment groups compared to saline
(p < 0.01), but cortical thickness was significantly increased
by 16–22%. This increased cortical thickness reflected
decreases in endosteal circumference rather than increases
in periosteal circumference. However, a statistically signifi-
cant increase in periosteal circumference was still seen in the
high-dosed PAM group.
Discussion
This study examined a preclinical fracture model that fea-
tured bone pre-treated with increasing doses of PAM and
tested the effects on subsequent bone repair. Consistent with
prior reports looking at concomitant BP dosing [6–8, 20],
we saw no negative effect on bone union and soft callus
removal. While to date a significant research has focus has
been placed on the utility of BPs to augment bone healing,
the effect of chronic BP treatment on fracture repair remains
under-explored. In the context of the elderly osteoporotic
patient, there are concerns that long-term BP treatment can
lead to atypical femoral fractures, potentially through a
reduction in global bone anabolism and a capacity to repair
microdamage. However, there are little data to support or
reject extending these concerns to a pediatric setting, which
is relevant to patient populations at high risk of childhood
fracture who routinely receive off-label BPs.
The physiology of bone growth can be profoundly differ-
ent between children and adults. This may also lead to dif-
ferences in the localization and action of BPs as the bone to
13

Fig. 3  6W callus histomor-
phometry. Measurements of a
MAR, b N.Oc, c Oc.S/N.Oc,
and d Oc.S/BS were quantified
from tissue sections. *p < 0.05,
**p < 0.01
which BPs bind becomes buried more rapidly in a pediatric
setting. Currently, there are no robust data to measure and
describe how BP embedding affects its long-term retention
in children. At a functional level, the long-term risks of BP
use on growth, modeling, microdamage repair, fracture risk,
and fracture healing are poorly described. It must neverthe-
less be acknowledged that the best clinical evidence to date
shows that the context of osteogenesis imperfecta BPs do
not significantly impair growth [21, 22], increase the risk of
atypical type fractures [23], or impair fracture repair [24].
Yet, the focus of previous studies testing extended BP
treatment has been in an adult osteoporosis setting. In ova-
riectomized rats treated with alendronate following femoral
shaft fractures, investigators reported increased callus size
and density, but also a decreased mineral apposition rate at
6 and 16 weeks post fracture [25]. A corresponding clinical
report indicated that osteoporotic fracture patients treated
with alendronate post-operatively showed no significant
worsening or improvement with treatment [26]. Other anti-
resorptive agents have shown similar outcomes in rodent
models, such as anti-RANKL antibody (denosumab), which
produces larger callus sizes after fracture [27].
Our model shows local reductions in callus MAR with
PAM treatment; however, small but significant reductions
in bone formation were also seen in contralateral intact
limbs with high-dose PAM (data not shown). The mecha-
nism underlying the corresponding decrease in bone forma-
tion with BPs remains unclear. There are currently multiple
13
A. Morse et al.
hypotheses for the coupling of osteoclastic and osteoblast
crosstalk [28], including paracrine signals from osteoclasts
as well as paracrine factors released from the bone by osteo-
clastic bone resorption. The observation of increased osteo-
clast numbers with nitrogen-containing BP treatment (simi-
lar to previous reports [19]) supports the latter model. The
progressive suppression of bone turnover by anti-resorptives
likely underlies conditions such as atypical femoral fractures
that are sporadically seen with long-term BP treatment. This
has been attributed to a decreased capacity to repair micro-
damage, leading to stress fractures and low-energy fractures
[13]. A recent study has suggested that the subsequently
impaired bone repair may be improved by PTH or teripara-
tide treatment [29]. Teriparatide has also been suggested to
improve cortical bone formation and turnover in patients
on long-term alendronate therapy in a clinical setting [30].
There are a number of caveats associated with the model
design and analysis that leave scope for future research.
Firstly, the length of pretreatment was not specifically
conceived to model the potentially years of BP treatment
that individuals with osteoporosis or OI may be exposed to
before developing a poorly healing fracture. Secondly, when
interpreting the data of this study and attempting to apply it
to clinical situations, it must be acknowledged that this study
did not feature models of osteoporosis or OI. Future studies
could attempt these experiments in ovariectomized rats or
rodent models of osteogenesis imperfecta. Thirdly, there are
practical limitations with the specific QCT calculations: BP
Pretreatment with Pamidronate Decreases Bone Formation but Increases Callus Bone Volume in…

Table 2  QCT parameters in intact femora did lead to decreased bone turnover resulting in significant
2W left femora 6W left femora
increases in callus size and predicted strength. The effects
of PAM were dose-dependent, although all doses did have
Mean SD Mean SD
effects on new bone formation as well as resorption. These
BMC data support further investigation into the long-term dosing
 Saline 44.35 2.87 53.97 2.37 of BP, with a particular focus on their effects on bone turno-
 PAM 0.15 mg/kg 46.22 1.42 57.57 3.95 ver and in clinically relevant models, such as murine models
 PAM 0.5 mg/kg 45.85 2.15 55.96 3.76 of osteogenesis imperfecta.
 PAM 5 mg/kg 45.39 1.91 59.04 5.64
BMD
 Saline 1163.7 15.1 1227.0 10.6 Compliance with Ethical Standards 
 PAM 0.15 mg/kg 1158.9 18.8 1163.3 32.3
 PAM 0.5 mg/kg 1153.4 23.4 1168.6 16.8 Conflict of interest Dr. Little and Dr Schindeler report grants and
non-financial support from Novartis Pharma, grants from N8 Medical,
 PAM 5 mg/kg 1133.8 13.0 1151.6 29.1 grants from Celgene, grants and nonfinancial support from Amgen Inc.
Cortical BV outside the submitted work.
 Saline 38.07 2.33 43.95 1.84
 PAM 0.15 mg/kg 39.83 1.29 49.91 4.57 Human and Animal Rights and Informed Consent  All procedures
performed in studies involving human participants were in accord-
 PAM 0.5 mg/kg 39.81 2.57 48.35 3.75 ance with the ethical standards of the institutional research committee
 PAM 5 mg/kg 40.13 1.87 52.35 6.56 (Westmead Hospital Animal Ethics Committee, #5024.06-07) and with
Periosteal Cir. the 1964 Helsinki declaration and its later amendments or comparable
 Saline 11.01 0.28 11.76 0.30 ethical standards.
 PAM 0.15 mg/kg 11.19 0.26 12.10 0.52
 PAM 0.5 mg/kg 11.03 0.42 11.90 0.35
 PAM 5 mg/kg 11.14 0.24 12.30* 0.61
Endosteal Cir.
References
 Saline 4.96 0.20 5.23 0.44
1. Maraka S, Kennel KA (2015) Bisphosphonates for the prevention
 PAM 0.15 mg/kg 4.88 0.33 4.39* 0.74 and treatment of osteoporosis. BMJ 351:h3783
 PAM 0.5 mg/kg 4.56* 0.36 4.33* 0.46 2. Baroncelli GI, Bertelloni S (2014) The use of bisphosphonates in
 PAM 5 mg/kg 4.74 0.41 4.31* 0.67 pediatrics. Horm Res Paediatr 82(5):290–302
3. Little DG, Ramachandran M, Schindeler A (2007) The anabolic
Cortical thickness
and catabolic responses in bone repair. J Bone Joint Surg Br
 Saline 0.96 0.05 1.04 0.05 89(4):425–433
 PAM 0.15 mg/kg 1.00 0.04 1.23* 0.12 4. Schindeler A, McDonald MM, Bokko P, Little DG (2008) Bone
 PAM 0.5 mg/kg 1.03 0.03 1.21* 0.10 remodeling during fracture repair: the cellular picture. Sem Cell
Dev Biol 19(5):459–466
 PAM 5 mg/kg 1.02* 0.05 1.27* 0.16
5. Chambers TJ, Fuller K (2011) How are osteoclasts induced to
MMI resorb bone? Ann N Y Acad Sci 1240:1–6
 Saline 17.6 1.8 22.3 2.2 6. McDonald MM, Morse A, Mikulec K, Peacock L, Baldock PA,
 PAM 0.15 mg/kg 18.9 1.5 25.7 3.7 Kostenuik PJ, Little DG (2013) Matrix metalloproteinase-driven
endochondral fracture union proceeds independently of osteoclast
 PAM 0.5 mg/kg 18.2 2.7 24.1 2.7
activity. J Bone Miner Res 28(7):1550–1560
 PAM 5 mg/kg 18.7 1.5 28.2* 5.7 7. Hao Y, Wang X, Wang L, Lu Y, Mao Z, Ge S, Dai K (2015)
Zoledronic acid suppresses callus remodeling but enhances callus
*p < 0.05 strength in an osteoporotic rat model of fracture healing. Bone
81:702–711
8. Menzdorf L, Weuster M, Klüter T, Brüggemann S, Behrendt P,
treatment may lead to subtle changes in cortical porosity that Fitchen-Oestern S, Varoga D, Seekamp A, Purcz N, Glueer CC,
may alter the thresholds used to designate “bone,” which Pufe T, Lippross S (2016) Local pamidronate influences fracture
would affect measures of bone volume and BMD, although healing in a rodent femur fracture model: an experimental study.
BMC Musculoskelet Disord 17:255
not BMC. Additionally, MMI calculations in the high-dose
9. Takuma S (1962) Electron microscopy of cartilage resorption by
BP group be affected by the shortened bone length, but the chondroclasts. J Dent Res 41:883–889
substantive changes in moment of inertia are best explained 10. Knowles HJ, Moskovsky L, Thompson MS, Grunhen J, Cheng
by the substantive changes in cortical thickness rather than X, Kashima TG, Athanasou NA (2012) Chondroclasts are mature
osteoclasts which are capable of cartilage matrix resorption. Vir-
the relatively subtle changes in length.
chows Arch 461(2):205–210
In conclusion, this study confirms many of our original 11. Colnot C, Thompson Z, Miclau T, Werb Z, Helms JA (2003)
hypotheses. PAM pretreatment, even at doses far higher than Altered fracture repair in the absence of MMP9. Development
in clinical use, did not impair endochondral ossification but 130(17):4123–4133

13

12. Behonick DJ, Xing Z, Lieu S et al (2007) Role of matrix metal-
loproteinase 13 in both endochondral and intramembranous ossi-
fication during skeletal regeneration. PLoS ONE 2(11):e1150
13. Kidd LJ, Cowling NR, Wu AC, Kelly WL, Forwood MR (2011)
Bisphosphonate treatment delays stress fracture remodeling in the
rat ulna. J Orthop Res 29(12):1827–1833
14. Shane E, Burr D, Ebeling PR, Abrahamsen B, Adler RA, Brown
TD et al (2010) Atypical subtrochanteric and diaphyseal femoral
fractures: report of a task force of the American Society for Bone
and Mineral research. J Bone Miner Res 25:2267–2294
15. Shane E, Burr D, Abrahamsen B, Adler RA, Brown TD, Cheung
AM et al (2014) Atypical subtrochanteric and diaphyseal femoral
fractures: second report of a task force of the American Society
for Bone and Mineral research. J Bone Miner Res 29:1–23
16. Alharbi M, Pinto G, Finidori G, Souberbielle JC, Guillou F,
Gaubicher S, Le Merrer M, Polak M (2009) Pamidronate treat-
ment of children with moderate-to-severe osteogenesis imperfecta:
a note of caution. Horm Res 71(1):38–44
17. Dwan K, Phillipi CA, Steiner RD, Basel D (2016) Bisphosphonate
therapy for osteogenesis imperfecta. Cochrane Database Syst Rev.
https​://doi.org/10.1002/14651​858.CD005​088.pub4
18. Bonnarens F, Einhorn TA (1984) Production of a standard closed
fracture in laboratory animal bone. J Orthop Res 2(1):97–101
19. Schindeler A, Ramachandran M, Godfrey C, Morse A, McDonald
M, Mikulec K, Little DG (2008) Modeling bone morphogenetic
protein and bisphosphonate combination therapy in wild-type and
Nf1 haploinsufficient mice. J Orthop Res 26(1):65–74
20. Amanat N, McDonald M, Godfrey C, Bilston L, Little D (2007)
Optimal timing of a single dose of zoledronic acid to increase
strength in rat fracture repair. J Bone Miner Res 22(6):867–876
21. Munns CF, Rauch F, Travers R, Glorieux FH (2005) Effects of
intravenous pamidronate treatment in infants with osteogenesis
imperfecta: clinical and histomorphometric outcome. J Bone
Miner Res 20(7):1235–1243
22. Biggin A, Munns CF (2014) Osteogenesis imperfecta: diagnosis
and treatment. Curr Osteoporos Rep 12(3):279–288
23. Trejo P, Fassier F, Glorieux FH, Rauch F (2017) diaphyseal
femur fractures in osteogenesis imperfecta: characteristics and
13
A. Morse et al.
relationship with bisphosphonate treatment. J Bone Miner Res
32(5):1034–1039
24. Munns CF, Rauch F, Zeitlin L, Fassier F, Glorieux FH (2004)
Delayed osteotomy but not fracture healing in pediatric osteogen-
esis imperfecta patients receiving pamidronate. J Bone Miner Res
19(11):1779–1786
25. Fu LJ, Tang TT, Hao YQ, Dai KR (2013) Long-term effects of
alendronate on fracture healing and bone remodeling of femoral
shaft in ovariectomized rats. Acta Pharmacol Sin 34(3):387–392
26. Seo JB, Yoo JS, Ryu JW, Yu KW (2016) Influence of early bis-
phosphonate administration for fracture healing in patients with
osteoporotic proximal humerus fractures. Clin Orthop Surg
8(4):437–443
27. Gerstenfeld LC, Sacks DJ, Pelis M, Mason ZD, Graves DT, Bar-
rero M, Ominsky MS, Kostenuik PJ, Morgan EF, Einhorn TA
(2009) Comparison of effects of the bisphosphonate alendronate
versus the RANKL inhibitor denosumab on murine fracture heal-
ing. J Bone Miner Res 24(2):196–208
28. Sims NA, Martin TJ (2015) Coupling signals between the osteo-
clast and osteoblast: how are messages transmitted between these
temporary visitors to the bone surface? Front Endocrinol (Laus-
anne) 6:41
29. Bakr MM, Kelly WL, Brunt AR, Paterson BC, Massa HM, Mor-
rison NA, Forwood MR (2019) Single injection of PTH improves
osteoclastic parameters of remodeling at a stress fracture site in
rats. J Orthop Res 37(5):1172–1182
30. Ma YL, Zeng QQ, Chiang AY, Burr D, Li J, Dobnig H, Fahrleit-
ner-Pammer A, Michalská D, Marin F, Pavo I, Stepan JJ (2014)
Effects of teriparatide on cortical histomorphometric variables in
postmenopausal women with or without prior alendronate treat-
ment. Bone 59:139–147
Publisher’s Note Springer Nature remains neutral with regard to
jurisdictional claims in published maps and institutional affiliations.