JOURNAL OF BONE AND MINERAL RESEARCH

Volume 20, Number 1, 2005

Published online on October 18, 2004; doi: 10.1359/JBMR.041010
© 2005 American Society for Bone and Mineral Research

Long-Term Disuse Osteoporosis Seems Less Sensitive to

Bisphosphonate Treatment Than Other Osteoporosis
Chao Yang Li, Christopher Price, Kemesha Delisser, Philip Nasser, Damien Laudier, Mariza Clement, Karl J Jepsen,
and Mitchell B Schaffler

ABSTRACT: We sought to determine whether risedronate can preserve cortical bone mass and mechanical
properties during long-term disuse in dogs, assessed by histomorphometry and biomechanics on metacarpal
diaphyses. Risedronate slowed cortical thinning and partially preserved mechanical properties, but it was
unable to suppress bone loss to the degree seen in other osteoporoses.
Introduction: Disuse induces dramatic bone loss resulting from greatly elevated osteoclastic resorption. Tar-
geting osteoclasts with antiresorptive agents, such as bisphosphonates, should be an effective countermeasure
for preventing disuse osteoporosis.
Materials and Methods: Single forelimbs from beagles (5–7 years old, n ⳱ 28) were immobilized (IM) for 12
months. Age-matched, non-IM dogs served as controls. One-half the animals received either risedronate (RIS,
1 mg/kg) or vehicle daily. Histomorphometry was performed on second metacarpal mid-diaphyses. Cortical
mechanical properties were determined by testing third metacarpal diaphyses in four-point bending.
Results: IM caused marked reduction in cortical area (−42%) and cortical thinning (−40%) through endo-
cortical resorption, extensive intracortical tunneling, and periosteal resorption; both bone resorption and
formation were significantly elevated over control levels on all envelopes. IM also decreased maximum load
and stiffness by ∼80% compared with controls. RIS reduced both periosteal bone loss and marrow cavity
expansion; however, cortical area remained significantly lower in RIS-treated IM animals than in untreated
non-IM controls (−16%). RIS also increased resorption indices in all envelopes compared with nontreated IM,
indicating that RIS suppressed osteoclast activity but not osteoclast recruitment. RIS did not affect bone
formation. RIS treatment conserved some whole bone mechanical properties, but they were still significantly
lower than in controls. There were no significant differences in tissue level material properties among the
groups.
Conclusion: RIS treatment reduces cortical bone loss at periosteal and endocortical surfaces caused by
long-term immobilization, thus partially conserving tissue mechanical properties. This modest effect contrasts
with more dramatic actions of the bisphosphonate in other osteoporoses. Our results suggest that risedronate
impairs osteoclastic function but cannot completely overcome the intense stimulus for osteoclast recruitment
during prolonged disuse.
J Bone Miner Res 2005;20:117–124. Published online on October 18, 2004; doi: 10.1359/JBMR.041010

Key words: disuse osteoporosis, bisphosphonates, bone histomorphometry, biomechanics, remodeling

INTRODUCTION hibit osteoclast activities.(12,13) They are extensively used in

prevention of bone loss from metabolic causes, such as

D ISUSE RESULTS IN dramatic bone loss, posing significant

problems for areas as diverse as spaceflight(1,2) and
rehabilitation medicine, where immobilization after spinal
cord injury(3–6) or hemiplegia after stroke(7–9) lead to rapid
bone loss and increased fracture risk.(10,11) Both programs
have identified effective suppression of the progressive
bone loss during disuse as the major countermeasure ob-
jective of both spaceflight and rehabilitation programs.
Bisphosphonates, the most effective antiresorptive
agents currently available, have been shown to strongly in-
The authors have no conflict of interest.
postmenopausal(14–18) and steroid-induced osteoporo-
sis,(19,20) as well as Paget’s bone disease.(21) Because bone
loss in larger mammals deprived of normal mechanical
loading results from extremely elevated osteoclastic resorp-
tion,(1–9) targeting osteoclasts with bisphosphonate should
be an effective countermeasure for preventing disuse osteo-
porosis. However, the efficacy of bisphosphonates in pre-
venting disuse-type bone loss is less clear. Recent studies
suggest that bisphosphonates may not work well in hypo-
dynamic situations. In patients with spinal cord injury(4–6)
and hemiplegia after a stroke,(7–9) treatment with high-dose
bisphosphonates attenuated BMD loss, but not to the de-

Leni and Peter W. May Department of Orthopaedics, Mount Sinai School of Medicine, New York, New York, USA.

117

J0403169R1 117 124

118
gree seen in other types of osteoporoses.(14–20) Moreover,
the bisphosphonate dosages used in these studies were
much higher than those used in postmenopausal osteopo-
rosis, suggesting that skeletal sites susceptible to bone loss
after spinal cord injury and stroke may be less sensitive to
bisphosphonate treatment than more “metabolic” bone.
Whether these situations are unique to spinal cord injury
and stroke, or reflect a more global difference in the way
that disuse osteoporosis responds to bisphosphonate, is un-
known. In this study, we sought to test the hypothesis that
a new generation pyraminyl bisphosphonate (risedronate)
would preserve cortical bone mass and mechanical proper-
ties during long-term disuse.
MATERIALS AND METHODS
Disuse osteoporosis was induced in the right forelimbs of
female retired breeder beagle dogs (5–7 years old, n ⳱ 28;
Marshall Farms). The canine skeleton was examined be-
cause of its high degree of similarity to human bone in
overall remodeling physiology. Limbs were immobilized
(IM) for 12 months, using a custom formed thermoplastic
splint (AquaPlast; Smith & Nephew, Memphis, TN, USA)
in which the elbow was flexed at 90°, and the carpal joint
volar was flexed slightly as described by Li et al.(22) Syn-
thetic cast padding was used under the splint. The splint was
designed with velcro straps to allow easy removal of the
splint for skin inspection, cleaning, and change of padding.
Age-matched, non-IM dogs served as controls (Con). One-
half the animals from each group received risedronate so-
dium (RIS; Proctor & Gamble Pharmaceuticals, Cincinnati,
OH, USA) dissolved in sterile water, at a dose of 1 mg/kg
orally daily (Con + RIS and IM + RIS). Risedronate treat-
ment was started 2 days before start of immobilization. The
remaining dogs received sterile water vehicle (Con and
IM). Risedronate was chosen for use in this experiment
because it has the shortest functional life in vivo among the
new generation of bisphosphonates.(23) One of the interests
that drove this study was the concept that bisphosphonates
may play a role in preventing bone loss in long-term space-
flight, reasoning that bisphosphonates that would clear
most rapidly from the bone once an astronaut returned to
the earth and stopped taking the drug would be advanta-
geous. The dose of RIS in this study is about 7–10 times
higher than the clinical dose used in the treatment of post-
menopausal osteoporosis.(14–16) RIS at one-half of the cur-
rent dose has been shown to inhibit cortical and cancellous
bone activation frequencies by 50–90% in dogs.(24,25) The
high dose of RIS used in this study was chosen to maximize
the osteoclastic inhibition with bisphosphonate treatment.
Because the drug is poorly absorbed from the gastrointes-
tinal tract, animals received no food or drink for at least 2
h before and after dosing. The absorption characteristics of
RIS in dogs are similar to those in humans (Proctor &
Gamble, unpublished data, 2003). Animals were group
housed in one of two large rooms (∼47 m2 each). Animal
health was recorded daily, and body weight was monitored
weekly. For IM animals, the splints were removed two to
three times per week to assess skin health. Animals were
conditioned to handling before the start of the experiment
LI ET AL.
so the splints could be removed and replaced without
chemical restraint or sedation. Animals remained healthy
and tolerated the IM procedure well throughout the study
period. No significant change in feeding behavior or
weights loss (<15% of starting body weight) was seen in
RIS-treated animals, suggesting good gastrointestinal toler-
ance of the bisphosphonate. All animals received double
bone fluorochrome labels at the end of the experiment.
Calcein (10 mg/kg, IV; Sigma) was given following an in-
jection schedule of 2–10–2–5 (2 days on, 10 days off, 2 days
on, and 5 days off), after which the animals were killed by
overdose of pentobarbital. All procedures were approved
by Institutional Animal Care and Use Committees of
Mount Sinai School of Medicine and VA Medical Center,
Bronx, NY.
Long bones were harvested and cleaned of soft tissues.
Bones used for histology were fixed in 10% neutral buff-
ered formalin, whereas those used for biomechanical stud-
ies were wrapped in saline-soaked gauze and frozen at
−40°C until testing. This study was performed on the sec-
ond and third metacarpal bones because of their high de-
gree of anatomical similarity to each other.
Cortical bone histomorphometry and geometry
Second metacarpals from right forelimbs were cut into
thirds using a low-speed diamond saw, and the pieces were
immersion fixed for 48 h. Mid-diaphyseal segments were
stained with Villanueva Mineralized Bone Stain (Arizona
Histology and Histomorphometry Services, Phoenix, AZ,
USA) for 10 days, dehydrated with ethylene glycol mono-
ethyl ether (Fisher Chemicals, Fair Lawn, NJ, USA),
cleared in methyl salicylate (J. T. Baker, Phillipsburg, NJ,
USA) and embedded with methyl methacrylate with 15%
dibutyl phthalate (Fisher Scientific). Undecalcified cross-
sections (150 ␮m thickness) were cut using a Leica 1600
Sawing Microtome (Leica Instruments, Nussloch, Germany),
polished to 30 ␮m thickness, and coverslipped with nonfluo-
rescing mounting medium for histomorphometric analysis.
Cortical bone changes were assessed using brightfield
and fluorescent microscopy. Histomorphometry was per-
formed using an OsteoMeasure system (Osteometrics, At-
lanta, GA, USA) following standard measures described by
Parfitt et al.(26) The structural (static) measures that were
quantified include total subperiosteal area (T.Ar; mm2),
marrow area (Ma.Ar; mm2), cortical area (Ct.Ar ⳱ T.Ar −
Ma.Ar; mm2), porosity (diameter ⱖ 30 ␮m) number (Po.N/
Ct.Ar; #/mm2) and area (Po.Ar/Ct.Ar; %), net cortical area
(Net.Ct.Ar ⳱ Ct.Ar − Po.Ar; mm2), and average cortical
width (Ave.Ct.Wi; ␮m). Cellular-based (dynamic) indices
included periosteal bone formation (P-sL.Pm/P.Pm; %) and
resorption (P-Er.Pm/P.Pm; %), endosteal bone formation
(E-sL.Pm/E.Pm; %) and resorption (E-Er.Pm/E.Pm; %),
osteonal resorption space number (RON/Ct.Ar; #/mm2),
forming osteon number (FON/Ct.Ar ⳱ [single-labeled os-
teon number + double-labeled osteon number]/Ct.Ar;
#/mm2), and newly formed osteon number (NON/Ct.Ar;
identified by the diffuse bone stain uptake visible under
fluorescence microscopy; #/mm2). All measurements were
made by a single observer who was blinded to the specimen
identity.
RISEDRONATE ON LONG-TERM DISUSE OSTEOPOROSIS
FIG. 1. Photomicrographs of metacarpal midshaft cross sections showing (A) control + vehicle, (B) control + RIS, (C) IM + vehicle,
and (D) IM + RIS. (C) IM bone showed a smaller subperiosteal area, larger marrow cavity, thinner cortex, and elevated porosity
compared with (A) control bone. (D) RIS-treated IM bone showed evidence of significant bone loss, but to a marked lesser degree than
IM alone. Bar ⳱ 500 ␮m.
Cross-sectional geometry measures (bone area, diam-
eters, and polar moment of inertia) were determined from
␮CT scans of the mid-diaphysis of the third metacarpal
bones. Bones were scanned using a GE MS8X ␮CT Scan-
ner (GE Medical Systems, London, Ontario, Canada) op-
erated at a 13-␮m isotropic voxel resolution. Bone tissue
was segment from non-bone tissue using the thresholding
algorithm provided by the ␮CT manufacturer. Cross-
sectional properties were determined from ∼200 ␮CT slices
(∼2.6 mm region) of the metacarpal centered about the
mid-diaphysis using a custom MATLAB analysis program
(Mathworks, Nattik, MA, USA). The mean polar moment
of inertia (Jo) of these slices was used in the calculation of
tissue properties.
Cortical bone mechanical properties
Bone mechanical properties were evaluated from the
third metacarpal bones after ␮CT scans. Cortical bone
structural-mechanical integrity was determined by loading
the diaphysis to failure in four-point bending. Bones were
positioned in the test apparatus with the dorsal side in com-
pression and the volar side in tension. The midshaft was
centered over two cylindrical supports positioned 21 mm
(L) apart with upper loading points spaced 7 mm apart (a).
Bones were loaded to failure at a displacement rate of 0.1
mm/s. Tests were performed using a servohydraulic testing
system (model 8872; Instron, Canton, MA, USA) equipped
with a 2000N load cell. The displacements were measured
from the system LVDT. Structural properties of whole
bones were determined from load-displacement curves as
follows: maximum load (the ultimate force that the speci-
men sustained), stiffness (the slope of linear portion of the
load-deformation curve), work-to-failure (the area under
the load-deformation curve before failure), and postyield
deflection (PYD, as a measure of brittleness). Yield was
defined as a 10% reduction in the secant stiffness (load
range normalized for deflection range) relative to the initial
tangent stiffness. Maximum stress (␴) and tissue modulus
(E) were calculated by the formulae as described by Turner
and Burr(27): ␴ ⳱ 1/4 maximum load × a × ⌽xx/Ixx; and E ⳱
−1/12 × stiffness × a2 × (4a − 3L)/Ixx, where ⌽xx and Ixx are
the diameter of diaphyses and rectangular moment of iner-
tia in the dorsal-volar direction, respectively.
Statistical analysis
Results are expressed as mean ± SD. Differences among
groups were tested using ANOVA with Fisher’s protected
least significant difference (PLSD) for posthoc testing. Sig-
nificance is reported at p < 0.05. Comparison of changes in
119
bone mass and mechanical properties versus polar moment
of inertia was performed using regression analyses. Statis-
tical analyses were performed using the StatView program
(version 5.0.1; Abacus, Mountview, CA, USA).
RESULTS
Cortical bone histomorphometry and geometry
Effects of immobilization: All comparisons of results are
expressed relative to non-IM controls. As expected, 12
months of IM resulted in a marked reduction of net cortical
bone area (Net.Ct.Ar, −42%) and dramatic cortical thin-
ning (Ave.Ct.Wi, −40%) (Fig. 1; Table 1). These occurred
through endocortical loss (Ma.Ar, +51%; Fig. 1; Table 1),
extensive intracortical tunneling (Po.Ar/Ct.Ar, +563%;
Figs. 2 and 3; Table 1), and periosteal bone loss (T.Ar,
−26%; Fig. 1; Table 1). Loss of periosteal bone resulted in
a marked decline of ∼70% from control value in the polar
moment of inertia (Table 1).
In IM animals, resorption surface on both periosteal and
endocortical envelopes was significantly elevated (P-Er.Pm/
P.Pm, +95%; E-Er.Pm/E.Pm, +216%; Fig. 2; Table 1) over
control levels. Surprisingly, there was an increase of bone
formation on both periosteal and endocortical surfaces as
well (Table 1). Within the intracortical envelope, resorption
space number was greatly elevated (Figs. 2 and 3; Table 1).
There was also extensive infilling of osteons, with the num-
ber of forming osteons (p < 0.006) and newly formed os-
teons significantly increased (p < 0.0001) in IM bones com-
pared with control bones (Fig. 3; Table 1). Overall, these
data indicate that cortical bone loss in long-term disuse is a
typical “high bone turnover” process occurring on all cor-
tical bone envelopes.
Effects of risedronate treatment: RIS treatment signifi-
cantly reduced, but did not completely prevent cortical
bone loss in IM animals. Overall bone loss (Net.Ct.Ar) in
RIS-treated IM animals was reduced nearly 60% relative to
nontreated IM animals, representing the reduction of 16%
(p < 0.003) and 42% (p < 0.0001) from the control level,
respectively. RIS treatment in IM significantly reduced
both periosteal bone loss and marrow cavity expansion.
Changes of Jo and in overall bone area in RIS-treated IM
were similar (Fig. 1; Table 1).
Surprisingly, indices of bone resorption (intracortical re-
sorption space number, eroded surface at periosteal and
endocortical surfaces) were elevated in RIS-treated IM
bone, more so in fact than in IM alone (Figs. 2 and 3; Table
1). Bone formation indices (labeled periosteal and endo-
cortical surfaces, forming and newly form osteons) were at
120 LI ET AL.

TABLE 1. STRUCTURAL AND REMODELING INDICES IN RIGHT METACARPAL MID-DIAPHYSES

Groups
Region Parameters Con Con + RIS IM IM + RIS
2 ‡
Cortex T.Ar (mm ) 16.32 ± 1.40 16.19 ± 1.45 12.07 ± 2.14 14.64 ± 1.64§
Ma.Ar (mm2) 2.60 ± 0.73 2.36 ± 0.82 3.92 ± 1.9* 2.78 ± 0.80
Net.Cr.Ar (mm2) 13.67 ± 1.04 13.74 ± 0.96 7.91 ± 1.5‡ 11.50 ± 1.36†††
Ave.Ct.Wi (␮m) 1301.4 ± 77.9 1336.5 ± 130.0 776.1 ± 201.1‡ 1137.1 ± 112.2*††
Jo (mm2) 33.10 ± 4.17 35.17 ± 5.22 9.78 ± 0.84‡ 23.59 ± 6.87†††
Po.N/Ct.Ar (#/mm2) 2.60 ± 0.37 3.26 ± 1.17 6.05 ± 2.49† 9.11 ± 3.17‡§
Po.Ar/Ct.Ar (%) 0.41 ± 0.15 0.66 ± 0.53 2.72 ± 1.87* 2.96 ± 3.45*
RON/Ct.Ar (#/mm2) 1.69 ± 0.40 2.01 ± 1.06 3.37 ± 0.88* 5.90 ± 2.45‡¶
FON/Ct.Ar (#/mm2) 0.27 ± 0.22 0.33 ± 0.49 1.24 ± 0.97† 0.57 ± 0.45§
NON/Ct.Ar (#/mm2) 0.00 ± 0.00 0.16 ± 0.39 3.82 ± 1.78† 1.13 ± 1.15††
Periosteal surface P-Er.Pm/P.Pm (%) 20.51 ± 5.40 21.16 ± 6.58 39.96 ± 17.62† 59.82 ± 14.76‡¶
P-sL.Pm/P.Pm (%) 0.30 ± 0.53 0.37 ± 0.67 11.39 ± 11.63† 1.70 ± 3.98¶
Endosteal surface E-Er.Pm/E/Pm (%) 3.54 ± 2.28 6.53 ± 6.26 13.21 ± 12.22* 24.23 ± 6.79‡§
E-sL.Pm/E.Pm (%) 2.18 ± 5.77 0.11 ± 0.28 15.46 ± 12.04† 0.00 ± 0.00**

Data are expressed as mean ± SD.

Con, control group; Con + RIS, control dogs treated with risedronate at 1 mg/kg per day; IM, immobilization group; IM + RIS, IM dogs treated with
risedronate at 1 mg/kg per day.
* p < 0.05; †p < 0.01; ‡p < 0.0001 versus control.
§
p < 0.05; ¶p < 0.01; **p < 0.001; ††p < 0.0001 versus nontreated IM.

FIG. 2. Photomicrographs from metacarpal midshafts showing (A) control + vehicle, (B) control + RIS, (C) IM + vehicle, and (D)
IM + RIS. (C) IM bone showed greatly activated bone remodeling, with eroded surface evident on both periosteal (arrow) and
endosteal surface (arrowhead) and osteonal canal (thin arrow) compared with (A) the control. (D) RIS-treated IM bone showed thicker
cortex with much greater numbers of porosities compared with (C) nontreated IM bone. Bar ⳱ 400 ␮m.

FIG. 3. Fluorescent photomicrographs of metacarpal midshafts showing (A) control + vehicle, (B) control + RIS, (C) IM + vehicle,
and (D) IM + RIS. (C) IM bone showed that both bone resorption (skinny arrow) and formation (arrow: green color, calcein-labeled
surface; red color, osteoid surface; arrowhead, newly formed osteon) were greatly elevated compared with (A) control bone.
(D) RIS-treated IM bone showed that bone resorption sites were greatly elevated but bone formation was at the control level. Bar ⳱
100 ␮m.

control levels in the RIS-treated IM group (Figs. 2 and 3;
Table 1), indicating that the high dose of risedronate used
in this study did not inhibit osteoblastic activity. Together,
these data suggest that bisphosphonate treatment did not
prevent the activation of osteoclasts during disuse, but it did
blunt the activity of osteoclasts so that bone loss was slowed
after the bisphosphonate treatment.
Cortical bone mechanical properties
Maximum load, stiffness, and work to failure in IM bone
decreased by ∼80% compared with control levels (Table 2).
Loss of these mechanical properties was almost double the
loss of cortical bone mass (Net.Ct.Ar, −42%; Table 1) but
was closely related to the decrease in the average diaphy-
seal polar moment of inertia (Table 3). RIS treatment in IM
conserved some of the mechanical properties, but maxi-
mum load and stiffness were still significantly lower than
that in controls (−29% and −24%, respectively; Table 2).
Postyield displacement was unchanged in IM- and RIS-
treated bones, indicating no change in bone brittleness with
either long-term IM or bisphosphonate treatment. Maxi-
mum stress in IM bone was reduced by 45% compared with
the control level, but in RIS-treated IM, was nearly identi-
cal to the control level. Tissue modulus was unaffected by
either IM or RIS treatment (Table 2). The relationships
between declines in Jo and loss of bone mass and strength

Fig 3 live 4/C

RISEDRONATE ON LONG-TERM DISUSE OSTEOPOROSIS 121

TABLE 2. MECHANICAL PROPERTIES OF FOUR-POINT BENDING TEST IN RIGHT METACARPAL MID-DIAPHYSES

Groups
Parameters Con Con + RIS IM IM + RIS
‡
Maximum load (N) 624.7 ± 124.9 678.4 ± 117.2 125.6 ± 46.0 442.6 ± 125.0†**
Stiffness (N/mm) 932.4 ± 190.7 1063.0 ± 142.8 205.2 ± 47.7‡ 708.4 ± 186.7***
Work to failure (Nmm) 1695.1 ± 481.9 1743.9 ± 431.0 315.2 ± 262.3‡ 1100.9 ± 346.9*§
Postyield displacement (mm) 0.95 ± 0.25 0.94 ± 0.16 0.77 ± 0.58 0.86 ± 0.18
Maximum stress (Mpa) 356.3 ± 29.7 342.1 ± 40.4 197.6 ± 76.7‡ 304.7 ± 26.6¶
Tissue modulus (GPa) 11.25 ± 1.49 11.23 ± 2.03 10.74 ± 2.82 12.62 ± 1.95

Data are expressed as mean ± SD.

Con, control group; Con + RIS, control dogs treated with risedronate at 1 mg/kg per day; IM, immobilization group; IM + RIS, IM dogs treated with
risedronate at 1 mg/kg per day.
* p < 0.05; †p < 0.01; ‡p < 0.0001 versus control.
§
p < 0.01; ¶p < 0.001; **p < 0.0001 versus nontreated IM.

TABLE 3. REGRESSION ANALYSIS OF POLAR MOMENT OF
INERTIAL VS. BONE MASS AND MECHANICAL PROPERTIES IN
RIGHT METACARPAL MID-DIAPHYSES
y x Equation R2
Jo Net.Ct.Ar y ⳱ −18.94 + 3.80x 0.828
Jo Maximum load y ⳱ 4.61 + 0.04x 0.951
Jo Stiffness y ⳱ 3.82 + 0.03x 0.945
(maximum load and stiffness) for all groups were essentially
identical, showing that diaphyseal strength was closely cor-
related with cortical bone mass and diaphyseal polar mo-
ment of inertia (Table 3).
DISCUSSION
This study shows that RIS treatment attenuated cortical
bone loss in long-term disuse osteoporosis but not to the
degree expected from studies of bisphosphonates in other
type of osteoporoses. Dramatic bone loss in disuse is caused
by extremely elevated bone resorption. Thus, targeting os-
teoclasts to reduce bone loss would be expected to be an
effective strategy in disuse. Bisphosphonates are the most
potent antiresorptive agents currently available. They have
been shown to stop bone loss after menopause(14–18) and in
glucocorticoid treatment.(19,20) So effective are the bisphos-
phonates in preventing bone loss in these metabolic osteo-
poroses that increases of BMD are seen from a combina-
tion of infilling of the remodeling space, increased BMC,
and perhaps uncoupled bone formation.(14,17) Thus, it has
been shown that alendronate at 5–10 mg daily effectively
prevented bone loss in postmenopausal osteoporosis(17,18)
and in patients on steroid therapy.(19) Risedronate at 5 mg
daily prevented bone loss and led to increases of BMD in
both early postmenopausal women(15) and those with es-
tablished osteoporosis.(16)
In contrast, bisphosphonates seem to be much less effec-
tive in preventing mechanically driven bone loss. Pamidro-
nate at 30 mg per every 4 weeks (the similar dose for post-
menopausal osteoporosis) for 6 months did not improve
BMD in the paretic side of complete spinal cord injury
patients.(4) Etidronate at high dose did not decrease the
rate of bone loss in complete spinal cord injury patients(6)
and in hemiplegia patients after a stroke.(8) Sniger and
Garshick recently reported that high-dose alendronate
p Value treatment in spinal cord injury patient attenuated BMD loss
<0.0001
but not to the degree seen in other osteoporoses.(5) Given
<0.0001 that the bisphosphonate dosages used in these studies(5,6,8)
<0.0001 were typically much higher than those used in postmeno-
pausal osteoporosis, bone loss after spinal cord injury and
stroke may be less sensitive to bisphosphonate treatment
than more metabolically driven bone loss. Whether these
situations are unique to spinal cord injury and stroke, or
reflect a more global difference in the way that disuse os-
teoporosis responds to bisphosphonate, is unclear. In view
of these studies, these data suggest that bone loss that fol-
lows unloading, that is, disuse osteoporosis, is less sensitive
than other, more metabolically driven bone loss processes
to suppression of resorption with bisphosphonates.
Disuse in adult dogs leads to a very rapidly evolving,
dramatic, and severe bone loss.(28–31) In this study, cortical
bone loss in canine metacarpal bones after 1 year of IM was
on the order of 40%, similar to the 50% bone loss reported
in dog metacarpals by Jaworski and Uhthoff and Uhthoff et
al. in their 1+ year long-term studies of cortical bone
loss.(28,29) Thus, the cortical bone loss rates over 1 year of
immobilization in adult dogs are in the range of 3–4% per
month. Because cortical bone loss (∼5.8 mm2) occurred
principally at the periosteal (∼4 mm2), followed by endos-
teal surface (∼1.3 mm2) and intracortical envelope (∼0.5
mm2) in IM animals, this in agreement with the finding of
Jaworski and Uhthoff in the young dogs.(28) Jaworski and
Uhthoff and Uhthoff et al. also found that bone loss in
disuse is not uniform. They showed that bone loss was more
pronounced in more distally located bones, with the meta-
carpals on the high end of bone loss range for this
model.(28,29) Nevertheless, profound bone loss occurs with
disuse at all forelimb bones with this immobilization model.
In this study, we examined only the metacarpals to allow us
to correlate bone histomorphometric and biomechanical
changes in two anatomically comparable bones within the
same animal. Moreover, the cortical bone loss rates in these
immobilized limbs are comparable to loss rates in long
bones after spinal cord injury and in bed rest. Wilmet et al.
found that BMD loss occurred at ∼2% per month at cortical
122
bone sites (4% in cancellous bone) in paralyzed legs of
spinal cord injury patients within the first year after IM.(32)
Other studies of spinal cord injury reported similar monthly
BMD loss rates, typically ranging 3–6% per month.(33)
Minaire et al. reported that cortical bone of iliac crests in
spinal cord injury patients thinned by almost 50% over 10
months (5% per month).(34) Bone loss in bed rest study has
been reported in the range of 2–4% per month at some
sites.(35) In contrast to disuse-type bone loss, BMD decline
in postmenopausal osteoporosis is on the order of 0.5–1.5%
per year,(36) but may reach as high as 3% per year in the
most rapid bone loss period after menopause.(37) Thus,
these data indicate that bone loss caused by unloading is
much greater and faster than that in postmenopausal osteo-
porosis.
Our understanding of the bone cellular components af-
fected by disuse is still far from complete. Whereas there is
an agreement that disuse in the adult skeleton activates
high levels of bone resorption, the precise osteoclast kinet-
ics in disuse is not well understood. It has been argued that
the unusually large resorption spaces seen in immobilized
bone are the result of extremely aggressive osteoclasts (i.e.,
increased cellular activity).(38) However, recent studies sug-
gest that early disuse in dog cortical bone is characterized
by very high activation of normal-sized resorption sites, in-
dicating normal cellular activity and a high birthrate of new
resorption foci. These foci later coalesce into large resorp-
tion spaces eventually join with the marrow cavity.(39) Stud-
ies on spinal cord injury(6) or paraplegic patients(40,41)
showed that increase of bone resorption in iliac crest bone
occurs in conjunction with decrease of bone formation.
Similarly, in healthy subjects after 12 weeks of bed rest,
osteoclast number and eroded surface increased, but osteo-
blast surface and mineral apposition rate decreased.(35)
Study of humans during spaceflight showed that bone re-
sorption markers tend to increase, whereas formation
markers tend to decrease.(2) These studies have led to the
idea that disuse uncouples of resorption from formation
during bone remodeling. However, several lines of evi-
dence suggest that this is not the case in larger mammals.
This study on dogs showed that both bone formation and
resorption markedly increased in disuse after 1 year of IM.
A number of others studies in dogs and primates showed
similar results. Lane et al.(30) and Grynpas et al.(31) deter-
mined from the level of newly created bone mineral that
significant new bone must form during disuse. Study of
monkeys showed that long-term immobilization signifi-
cantly increased level of immature collagen cross-links in
bone, which results from new bone formation.(42) Schaffler
and Pan(43) found that bone formation resumes on previ-
ously resorbed bone surfaces during long-term disuse, but
the onset of this formation is delayed compared with nor-
mal bone remodeling. Together, these data suggest that the
bone loss pattern in disuse in dogs is a typical “high bone
turnover” process, similar to postmenopausal osteoporosis,
even though the former is a remodeling-based osteoporosis
in which there may be a temporal delay in the onset of bone
formation processes.
It has been widely accepted that bisphosphonates inhibit
both osteoclast recruitment and activity.(12,13) However, in
LI ET AL.
this study, we found that the resorption indices on all en-
velopes of immobilized cortical bone under RIS treatment
at the end of 1 year were highly elevated and in fact greater
than that in nontreated IM animals. RIS-treated IM bone
had more eroded surface at the periosteal and endocortical
envelopes and more numerous intracortical resorption
spaces than IM bones. However, RIS treatment during
long-term IM suppressed bone loss, suggesting that the ef-
fectiveness of osteoclastic bone removal is diminished by
bisphosphonate treatment. RIS suppression of bone loss
appeared to differ in each bone envelope, ranging from
29% at the endocortical surface, 21% at periosteal, and
eventually unchanged at the intracortical envelope. Such
differences could reflect site-specific cellular responsiveness
to bisphosphonate treatment or difference in drug variabil-
ity. However, because of variability particularly in endocor-
tical area measurement in IM animals, it is not clear of the
difference is significant. While seemingly paradoxical, these
observations are consistent with the previous data from the
iliac crest in bed rest and spinal cord injury. Etidronate
treatment of bed rest subjects for 120 days led to increases
in eroded surface in the iliac crest, but decreases in osteo-
clast number.(44) In paraplegic patients treated with high-
dose tiludronate for 3 months, Chappard et al. found that
the eroded surface was at the same level as in placebo pa-
tients, whereas osteoclast number was reduced to normal
level.(9) Minaire et al.(41) reported similar results in spinal
cord injury patients treated with clodronate. The reasons
for this high level of eroded surface and diminished osteo-
clast number in immobilization patients treated with bis-
phosphonate are unclear. In this study, the dose of RIS used
was quite high. High doses of bisphosphonates can suppress
bone formation(45); however, in our study, bone formation
indices in the RIS-treated IM group remained at control
levels, indicating that the high dose of RIS we used in this
study did not inhibit osteoblastic activity. Thus, the high
levels of eroded surface and intracortical resorption spaces
in RIS-treated IM animals was not a result of failure to
form bone. Alternatively, the effect of bisphosphonates on
osteoclasts in long-term immobilization may differ from
other circumstances.
These data, as well as those from previous studies of
bisphosphonates in bed rest and paralysis, suggest a sce-
nario wherein bisphosphonate treatment may not prevent
the activation of osteoclasts during long-term disuse. How-
ever, once recruited, bisphosphonates seem to be able to
diminish the osteoclastic activity such that bone loss is re-
duced after the treatment. Why the recruitment of osteo-
clast in unloading would be especially high is not known,
but disuse can cause a number of physiological changes that
lead to increased bone resorption (e.g., osteocyte injury,
hypoxia and apoptosis in bone, changes in limb blood flow,
venous stasis, tissue perfusion, and changes in local tissue
pH).(46,47) Interestingly, recent studies indicate that an el-
evation of resorption indices also occurs with very-long-
term bisphosphonate treatment of postmenopausal osteo-
porosis. With both long-term RIS(14) and alendronate(17)
treatment, bone resorption indices (eroded surface and os-
teoclast number) were not suppressed by the treatment,
despite long-term cessation of bone loss. Taken together,
RISEDRONATE ON LONG-TERM DISUSE OSTEOPOROSIS
these results suggest that the effects of bisphosphonates on
osteoclast recruitment versus those on cell activity (vigor
and lifespan) can be uncoupled in vivo. Thus, osteoclast
recruitment will occur in the presence of bisphosphonates if
the resorption signal is very intense (as in immobilization)
or perhaps, less intense signal of sufficient duration (as in
post menopausal bone loss). However, once recruited, the
effectiveness of these cells in removing bone is diminished
by the presence of bisphosphonate.
A previous study found that pamidronate treatment in
short-term disuse (12 weeks) in dogs preserved diaphyseal
mechanical properties.(31) This differs from the results of
this study, in which we found that RIS treatment during a
1-year immobilization conserved a significant amount of
diaphyseal strength and stiffness, but the RIS-treated bones
were still much weaker (∼30% lower) than normal bones.
Given that studies from the spinal cord injury or stroke
patients showed that bone loss was characteristically sus-
tained for all bisphosphonates examined, it seems likely
that the differences in biomechanical results between this
long-term study (12 months) and that of Grynpas et al. (3
months)(31) reflect the difference in experimental duration
rather than a fundamental difference in the action of dif-
ferent bisphosphonates.
Long-duration treatment with bisphosphonates is associ-
ated with altered material properties in bone. Treatment
with either alendronate or risedronate for 1 year in young
adult dogs(24,25) has recently shown that bisphosphonate
treatment results in a significant increase in bone brittleness
and a decrease in the energy needed to fracture a bone;
these properties were attributed to both bone microdamage
accumulation and increased matrix mineralization resulting
from inhibition of bone turnover. In this study, we did not
find any change in bone brittleness or work-to-fracture. The
possible reason for this discrepancy could be attributed to
the animal age and the different bone sites studied. Bone
turnover rates are high in young adult dogs at the vertebral
cancellous bone (BFR/BV, ∼80 ␮m3/␮m2/yr) and the rib
cortical bone (BFR/BV, ∼20%/yr in intracortical site; BFR/
BS on periosteal and endocortical sites, ∼11–15 ␮m3/␮m2/
yr)(24,25) compared with 5- to 7-year-old dogs, where bone
turnover in diaphyses is very low (BFR/B.Ar in cancellous
bone, ∼2%/yr [data not shown], BFR/BS cannot be deter-
mined for periosteal and endocortical envelopes in control
bones from this study because of lack of double labeling).
Thus, it seems reasonable to expect that remodeling sup-
pression in a high turnover bone will show a greater relative
effect than a similar degree of suppression in a low turnover
bone. This latter possibility may have important implica-
tions for understanding how remodeling suppression differ-
entially affects bone sites, and in turn, bone fragility.
In conclusion, the results of this study in combination
with data from a range of other bisphosphonate studies in
paralysis patients suggest that bisphosphonates cannot fully
overcome the stimulus to recruit osteoclast during long du-
ration of immobilization. However, once osteoclasts are re-
cruited, their activity seems to be diminished by bisphos-
phonates, and they remove less bone. Thus, RIS treatment
can attenuate cortical bone loss in long-term disuse osteo-
porosis, but can not inhibit bone loss to the degree expected
123
based on the action of bisphosphonate in other, more
“metabolic” types of osteoporosis. Nevertheless, the results
of this study show that RIS treatment can reduce cortical
bone loss by more than one-half and was particularly effec-
tive in reducing the periosteal and endocortical bone loss
that can disproportionately weaken long bones diaphyses.
ACKNOWLEDGMENTS
This study was supported by the National Space Bio-
medical Research Institute (BL00203) and NIH AR41210
and AR48699. The authors thank Dr Robert Majeska for
the English editing, Gregory Bouyer for expert technical
assistance, and Richard Mann, DVM, from Bronx VA
Medical Center for the veterinary guidance. The risedro-
nate used in this study was generously provided by Procter
& Gamble Pharmaceuticals; we also thank Roger Phipps,
PhD, and Mark Lundy, PhD, from Proctor & Gamble for
helpful comments.
REFERENCES
1. Collet P, Uebelhart D, Vico L, Moro L, Hartmann D, Roth M,
Alexandre C 1997 Effects of 1- and 6-month spaceflight on
bone mass and biochemistry in two humans. Bone 20:547–551.
2. Caillot-Augusseau A, Lafage-Proust MH, Soler C, Pernod J,
Dubois F, Alexandre C 1998 Bone formation and resorption
biological markers in cosmonauts during and after a 180-day
space flight (Euromir 95). Clin Chem 44:578–585.
3. Roberts D, Lee W, Cuneo RC, Wittmann J, Ward G, Flatman
R, McWhinney B, Hickman PE 1998 Longitudinal study of
bone turnover after acute spinal cord injury. J Clin Endocrinol
Metab 83:415–422.
4. Nance PW, Schryvers O, Leslie W, Ludwig S, Krahn J, Uebel-
hart D 1999 Intravenous pamidronate attenuates bone density
loss after acute spinal cord injury. Arch Phys Med Rehabil
80:243–251.
5. Sniger W, Garshick E 2002 Alendronate increases bone den-
sity in chronic spinal cord injury: A case report. Arch Phys Med
Rehabil 83:139–140.
6. Pearson EG, Nance PW, Leslie WD, Ludwig S 1997 Cyclical
etidronate: Its effect on bone density in patients with acute
spinal cord injury. Arch Phys Med Rehabil 78:269–272.
7. Minaire P, Berard E, Meunier PJ, Edouard C, Goedert G,
Pilonchery G 1981 Effects of disodium dichloromethylene di-
phosphonate on bone loss in paraplegic patients. J Clin Invest
68:1086–1092.
8. Sato Y, Asoh T, Kaji M, Oizumi K 2000 Beneficial effect of
intermittent cyclical etidronate therapy in hemiplegic patients
following an acute stroke. J Bone Miner Res 15:2487–2494.
9. Chappard D, Minaire P, Privat C, Berard E, Mendoza-
Sarmiento J, Tournebise H, Basle MF, Audran M, Rebel A,
Picot C, Gaud C 1995 Effects of tiludronate on bone loss in
paraplegic patients. J Bone Miner Res 10:112–118.
10. Kanis J, Oden A, Johnell O 2001 Acute and long-term increase
in fracture risk after hospitalization for stroke. Stroke 32:702–
706.
11. Ramnemark A, Nyberg L, Borssén B, Olsson T, Gustafson Y
1998 Fracture after stroke. Osteoporos Int 8:92–95.
12. Rodan G 1998 Mechanisms of action of bisphosphonates.
Annu Rev Pharmacol Toxicol 38:375–388.
13. Russell RGG, Rogers MJ 1999 Bisphosphonates: From the
laboratory to the clinic and back again. Bone 25:97–106.
14. Eriksen EF, Melsen F, Sod E, Barton I, Chines A 2002 Effects
of long-term risedronate on bone quality and bone turnover in
women with postmenopausal osteoporosis. Bone 31:620–625.
15. Mortensen L, Charles P, Bekker P, Digennaro J, Johnston CC
1998 Risedronate increases bone mass in an early postmeno-
124
pausal population: Two years of treatment plus one year of
follow-up. J Clin Endocrinol Metab 83:396–402.
16. Granney A, Tugwell P, Adachi J, Weaver B, Zytaruk N, Pa-
paioannou A, Robinson V, Shea B, Wells G, Guyatt G 2002
III. Meta-analysis of risedronate for the treatment of post-
menopausal osteoporosis. Endocr Rev 23:517–523.
17. Chavassieux PM, Arlot ME, Reda C, Wei L, Yates AJ, Meu-
nier PJ 1997 Histomorphometric assessment of the long-term
effects of alendronate on bone quality and remodeling in pa-
tients with osteoporosis. J Clin Invest 100:1475–1480.
18. Bone HG, Downs RW, Tucci JR, Harris ST, Weinstein RS,
Licata AA, McClung MR, Kimmel DB, Gertz BJ, Hale E,
Polvino WJ 1997 Dose-response relationships for alendronate
treatment in osteoporotic elderly women. J Clin Endocrinol
Metab 82:265–274.
19. Lau EMC, Woo J, Chan YH, Li M 2001 Alendronate for the
prevention of bone loss in patients on inhaled steroid therapy.
Bone 29:506–510.
20. Adachi JD, Bensen WG, Brown J, Hanley D, Hodsman A,
Josse R, Kendler DL, Lentle B, Olszynski W, Ste-Marie LG,
Tenenhouse A, Chines AA 1997 Intermittent cyclical etidro-
nate therapy to prevent corticosteroid-induced osteoporosis.
N Engl J Med 337:382–387.
21. Brown JP, Chines AA, Myers WR, Eusebio RA, Ritter-
Hrncirik C, Hayes CW 2000 Improvement of Pagetic bone
lesions with risedronate treatment: A radiologic study. Bone
26:263–267.
22. Li CY, Laudier D, Schaffler MB 2003 Remobilization restores
cancellous bone mass but not microarchiteture after long-term
disuse in older adult dogs. Trans Orthop Res Soc 28:1052.
23. Christiansen C, Phipps R, Burgio D, Sun L, Russell D, Keck B,
Kuzmak B, Lindsay R 2003 Comparison of risedronate and
alendronate pharmacokinetics at clinical doses. Osteoporos Int
14(Suppl 7):S38.
24. Mashiba T, Turner CH, Hirano T, Forwood MR, Johnston CC,
Burr DB 2001 Effects of suppressed bone turnover by bisphos-
phonates on microdamage accumulation and biomechanical
properties in clinically relevant skeletal sites in beagles. Bone
28:524–531.
25. Mashiba T, Hirano T, Turner CH, Forwood MR, Johnston CC,
Burr DB 2000 Suppressed bone turnover by bisphosphonates
increases microdamage accumulation and reduces some bio-
mechanical properties in dog rib. J Bone Miner Res 15:613–
620.
26. Parfitt AM, Glorieux FH, Kanis JA, Malluche H, Meunier PJ,
Ott SM, Recker RR 1987 Bone histomorphometry nomencla-
ture, symbols and units: Report of the ASBMR Histomor-
phometry Nomenclature Committee. J Bone Miner Res 2:595–
610.
27. Turner CH, Burr DB 1993 Basic biomechanical measurements
of bone: A tutorial. Bone 14:595–608.
28. Jaworski ZFG, Uhthoff HK 1986 Reversibility of nontraumatic
disuse osteoporosis during its active phase. Bone 7:431–439.
29. Uhthoff HK, Sélaky G, Jaworski ZFG 1985 Effect of long-term
nontraumatic immobilization on metaphyseal spongiosa in
young adult and old beagle dogs. Clin Orthop 182:278–283.
30. Lane NE, Kaneps AJ, Stover SM, Modin G, Kimmel DB 1996
Bone mineral density and turnover following forelimb immo-
bilization and recovery in young adult dogs. Calcif Tissue Int
59:401–406.
31. Grynpas MD, Kasra M, Renlund R, Pritzker KPH 1995 The
effect of pamidronate in a new model of immobilization in the
dog. Bone 17(Suppl 4):S225–S232.
32. Wilmet E, Ismail AA, Heilporn A, Welraeds D, Bergmann B
1995 Longitudinal study of the bone mineral content and of
soft tissue composition after spinal cord section. Paraplegia
33:674–677.
33. Biering-Sårensen F, Bohr HH, Schaadr OP 1990 Longitudinal
LI ET AL.
study of bone mineral content in the lumbar spine, the forearm
and the lower extremities after spinal cord injury. Eur J Clin
Invest 20:330–335.
34. Minaire P, Meunier P, Edouard C, Bernard J, Courpron P,
Bourret J 1974 Quantitative histological data on disuse osteo-
porosis, comparison with biological data. Calcif Tissue Res
17:57–73.
35. Zerwekh JE, Ruml LA, Gottschalk F, Pak CYC 1998 The
effects of twelve weeks of bed rest on bone histology, bio-
chemical markers of bone turnover, and calcium homeostasis
in eleven normal subjects. J Bone Miner Res 13:1594–1601.
36. Jones G, Nguyen T, Sambrook P, Kelly PJ, Eisman JA 1994
Progressive loss of bone in the femoral neck in elderly people:
Longitudinal findings from the Dubbo osteoporosis epidemi-
ology study. BMJ 309:691–695.
37. Okano H, Mizunuma H, Soda M, Kagami I, Miyamoto S, Oh-
sawa M, Ibuki Y, Shiraki M, Suzuki T, Shibata H 1998 The
long-term effect of menopause on postmenopausal bone loss in
Japanese women: Results from a prospective study. J Bone
Miner Res 13:303–309.
38. Young DR, Niklowitz WJ, Brown RJ, Jee WSS 1986 Immobi-
lization-associated osteoporosis in primates. Bone 7:109–117.
39. Schaffler MB, Li XJ 1990 Immobilization induced bone loss:
Quantitative histological studies of cortical bone resorption.
Trans Orthop Res Soc 15:187.
40. Chantraine A, Nusgens B, Lapiere CM 1986 Bone remodeling
during the development of osteoporosis in paraplegia. Calcif
Tissue Int 38:323–327.
41. Minaire P, Depassio J, Berard E, Meunier PJ, Edouard C,
Pilonchery G, Goedert G 1987 Effects of clodronate on immo-
bilization bone loss. Bone 8(Suppl 1):S63–S68.
42. Mechanic GL, Young DR, Banes AJ, Yamauchi M 1986 Non-
mineralized and mineralized bone collagen in bone of immo-
bilized monkeys. Calcif Tissue Int 39:63–68.
43. Schaffler MB, Pan HQ 1993 Bone cell kinetics in the adult
canine skeleton during immobilization. Trans Orthop Res Soc
18:633.
44. Chappard D, Alexandre C, Palle S, Vico L, Morukov BV,
Rodionova P, Minaire P, Riffat G 1989 Effects of a bisphos-
phonate (1-hydroxy ethylidene-1, 1 bisphosphonic acid) on os-
teoclast number during prolonged bed rest in healthy humans).
Metabolism 38:822–825.
45. Li C, Mori S, Li J, Kaji Y, Akiyama T, Kawanishi J, Norimatsu
H 2001 Long-term effect of incadronate disodium (YM-175)
on fracture healing of femoral shaft in growing rats. J Bone
Miner Res 16:429–436.
46. Gross TS, Damji AA, Judex S, Bray RC, Zernicke RF 1999
Bone hyperemia precedes disuse-induced intracortical bone
resorption. J Appl Physiol 86:230–235.
47. Epstein S, Inzerillo AM, Caminis J, Zaidi M 2003 Disorders
associated with acute rapid and severe bone lose. J Bone Miner
Res 18:2083–2094.
Address reprint requests to:
Mitchell B Schaffler, PhD
Leni and Peter W. May Department of Orthopaedics
The Mount Sinai School of Medicine
One Gustave L. Levy Place, Box 1188
New York, NY 10029-6574, USA
Received in original form March 19, 2004; revised form June 21,
2004; accepted August 11, 2004.